From 8d7dc9283f399e1fda4e48a1c453f689326d9396 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 14 Apr 2015 19:33:59 -0700 Subject: rcu: Control grace-period delays directly from value In a misguided attempt to avoid an #ifdef, the use of the gp_init_delay module parameter was conditioned on the corresponding RCU_TORTURE_TEST_SLOW_INIT Kconfig variable, using IS_ENABLED() at the point of use in the code. This meant that the compiler always saw the delay, which meant that RCU_TORTURE_TEST_SLOW_INIT_DELAY had to be unconditionally defined. This in turn caused "make oldconfig" to ask pointless questions about the value of RCU_TORTURE_TEST_SLOW_INIT_DELAY in cases where it was not even used. This commit avoids these pointless questions by defining gp_init_delay under #ifdef. In one branch, gp_init_delay is initialized to RCU_TORTURE_TEST_SLOW_INIT_DELAY and is also a module parameter (thus allowing boot-time modification), and in the other branch gp_init_delay is a const variable initialized by default to zero. This approach also simplifies the code at the delay point by eliminating the IS_DEFINED(). Because gp_init_delay is constant zero in the no-delay case intended for production use, the "gp_init_delay > 0" check causes the delay to become dead code, as desired in this case. In addition, this commit replaces magic constant "10" with the preprocessor variable PER_RCU_NODE_PERIOD, which controls the number of grace periods that are allowed to elapse at full speed before a delay is inserted. Reported-by: Linus Torvalds Signed-off-by: Paul E. McKenney diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 233165d..8cf7304 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -162,11 +162,14 @@ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp); static int kthread_prio = CONFIG_RCU_KTHREAD_PRIO; module_param(kthread_prio, int, 0644); -/* Delay in jiffies for grace-period initialization delays. */ -static int gp_init_delay = IS_ENABLED(CONFIG_RCU_TORTURE_TEST_SLOW_INIT) - ? CONFIG_RCU_TORTURE_TEST_SLOW_INIT_DELAY - : 0; +/* Delay in jiffies for grace-period initialization delays, debug only. */ +#ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT +static int gp_init_delay = CONFIG_RCU_TORTURE_TEST_SLOW_INIT_DELAY; module_param(gp_init_delay, int, 0644); +#else /* #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT */ +static const int gp_init_delay; +#endif /* #else #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT */ +#define PER_RCU_NODE_PERIOD 10 /* Number of grace periods between delays. */ /* * Track the rcutorture test sequence number and the update version @@ -1843,9 +1846,8 @@ static int rcu_gp_init(struct rcu_state *rsp) raw_spin_unlock_irq(&rnp->lock); cond_resched_rcu_qs(); ACCESS_ONCE(rsp->gp_activity) = jiffies; - if (IS_ENABLED(CONFIG_RCU_TORTURE_TEST_SLOW_INIT) && - gp_init_delay > 0 && - !(rsp->gpnum % (rcu_num_nodes * 10))) + if (gp_init_delay > 0 && + !(rsp->gpnum % (rcu_num_nodes * PER_RCU_NODE_PERIOD))) schedule_timeout_uninterruptible(gp_init_delay); } diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 1ad74c0..5f5ff7d 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1268,6 +1268,7 @@ config RCU_TORTURE_TEST_SLOW_INIT_DELAY int "How much to slow down RCU grace-period initialization" range 0 5 default 3 + depends on RCU_TORTURE_TEST_SLOW_INIT help This option specifies the number of jiffies to wait between each rcu_node structure initialization. -- cgit v0.10.2 From 96f05be37f4ece1dffba92d4ae079a486a4cf6df Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 13 Apr 2015 14:23:29 +0200 Subject: ASoC: qcom: Return an error for invalid PCM trigger command MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix a compile warning sound/soc/qcom/lpass-cpu.c: In function ‘lpass_cpu_daiops_trigger’: sound/soc/qcom/lpass-cpu.c:224:2: warning: ‘ret’ may be used uninitialized in this function [-Wmaybe-uninitialized] return ret; ^ Although switch () lists the most of existing cases, it's still better to cover the rest as an error properly. Signed-off-by: Takashi Iwai Acked-by: Kenneth Westfield Signed-off-by: Mark Brown diff --git a/sound/soc/qcom/lpass-cpu.c b/sound/soc/qcom/lpass-cpu.c index 6698d05..dc790ab 100644 --- a/sound/soc/qcom/lpass-cpu.c +++ b/sound/soc/qcom/lpass-cpu.c @@ -194,7 +194,7 @@ static int lpass_cpu_daiops_trigger(struct snd_pcm_substream *substream, int cmd, struct snd_soc_dai *dai) { struct lpass_data *drvdata = snd_soc_dai_get_drvdata(dai); - int ret; + int ret = -EINVAL; switch (cmd) { case SNDRV_PCM_TRIGGER_START: -- cgit v0.10.2 From d1acba2fdebb449bad01e7cf77a9f9730dfaca6e Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 13 Apr 2015 06:00:45 +0000 Subject: ASoC: rsnd: set dmaen->chan = NULL when error case rsnd_dmaen_quit() is assuming dmaen->chan is NULL if it failed to get DMAEngine channel. but, current dmaen->chan might have error value when error case (this driver is checking it by IS_ERR_OR_NULL()) This patch makes sure dmaen->chan is NULL when error case. Otherwise, it will contact to unknown address in rsnd_dmaen_quit() Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown diff --git a/sound/soc/sh/rcar/dma.c b/sound/soc/sh/rcar/dma.c index ac3756f..144308f 100644 --- a/sound/soc/sh/rcar/dma.c +++ b/sound/soc/sh/rcar/dma.c @@ -156,6 +156,7 @@ static int rsnd_dmaen_init(struct rsnd_priv *priv, struct rsnd_dma *dma, int id, (void *)id); } if (IS_ERR_OR_NULL(dmaen->chan)) { + dmaen->chan = NULL; dev_err(dev, "can't get dma channel\n"); goto rsnd_dma_channel_err; } -- cgit v0.10.2 From ac98b4c015b50b1e452f8d55b612320be7f80825 Mon Sep 17 00:00:00 2001 From: Jin Yao Date: Mon, 13 Apr 2015 14:20:54 +0800 Subject: ASoC: Intel: Remove invalid kfree of devm allocated data kbuild robot reports following warning: "sound/soc/intel/haswell/sst-haswell-ipc.c:2204:1-6: WARNING: invalid free of devm_ allocated data" As julia explains to me, the memory allocated with devm_kalloc is freed automatically on failure of a probe function. So this kfree should be removed otherwise the double free will be got in error handler path. Signed-off-by: Jin Yao Signed-off-by: Mark Brown diff --git a/sound/soc/intel/haswell/sst-haswell-ipc.c b/sound/soc/intel/haswell/sst-haswell-ipc.c index 344a1e9..324eceb 100644 --- a/sound/soc/intel/haswell/sst-haswell-ipc.c +++ b/sound/soc/intel/haswell/sst-haswell-ipc.c @@ -2201,7 +2201,6 @@ dma_err: dsp_new_err: sst_ipc_fini(ipc); ipc_init_err: - kfree(hsw); return ret; } EXPORT_SYMBOL_GPL(sst_hsw_dsp_init); -- cgit v0.10.2 From 0ea001d3b43cc9d387c093ae205c4228cd88a886 Mon Sep 17 00:00:00 2001 From: Chris Zhong Date: Wed, 15 Apr 2015 13:57:11 +0800 Subject: ARM: rockchip: disable dapswjdp during suspend Reset dapswjdp is controlled by JTAG_TRSTN, if the iomux of this pin is not "jtag_trstn". the AP would think this pin is always high, so it can not reset before resume. When system resume, but the dapswjdp is not in a default state, it may Access some illegal address, it cause system crash during resume. Let's disable this jtag function by clear the dapdeviceen bit, it prohibit the dapswjdp to access memory and registers. This bit would be enable in MASKROM, so we need clear it in suspend everytime. Signed-off-by: Chris Zhong Reviewed-by: Doug Anderson Signed-off-by: Heiko Stuebner diff --git a/arch/arm/mach-rockchip/pm.c b/arch/arm/mach-rockchip/pm.c index b07d886..b0dcbe2 100644 --- a/arch/arm/mach-rockchip/pm.c +++ b/arch/arm/mach-rockchip/pm.c @@ -83,6 +83,13 @@ static void rk3288_slp_mode_set(int level) SGRF_PCLK_WDT_GATE | SGRF_FAST_BOOT_EN | SGRF_PCLK_WDT_GATE_WRITE | SGRF_FAST_BOOT_EN_WRITE); + /* + * The dapswjdp can not auto reset before resume, that cause it may + * access some illegal address during resume. Let's disable it before + * suspend, and the MASKROM will enable it back. + */ + regmap_write(sgrf_regmap, RK3288_SGRF_CPU_CON0, SGRF_DAPDEVICEEN_WRITE); + /* booting address of resuming system is from this register value */ regmap_write(sgrf_regmap, RK3288_SGRF_FAST_BOOT_ADDR, rk3288_bootram_phy); diff --git a/arch/arm/mach-rockchip/pm.h b/arch/arm/mach-rockchip/pm.h index 03ff31d..3e8d39c 100644 --- a/arch/arm/mach-rockchip/pm.h +++ b/arch/arm/mach-rockchip/pm.h @@ -55,6 +55,10 @@ static inline void rockchip_suspend_init(void) #define SGRF_FAST_BOOT_EN BIT(8) #define SGRF_FAST_BOOT_EN_WRITE BIT(24) +#define RK3288_SGRF_CPU_CON0 (0x40) +#define SGRF_DAPDEVICEEN BIT(0) +#define SGRF_DAPDEVICEEN_WRITE BIT(16) + #define RK3288_CRU_MODE_CON 0x50 #define RK3288_CRU_SEL0_CON 0x60 #define RK3288_CRU_SEL1_CON 0x64 -- cgit v0.10.2 From b403125d3bbf8046c1186e1a49cb17bb5551db14 Mon Sep 17 00:00:00 2001 From: Chris Zhong Date: Sun, 22 Mar 2015 00:04:51 +0800 Subject: ARM: rockchip: fix undefined instruction of reset_ctrl_regs Sometimes the debug module may not work well after resume, since it has not been correctly reset when wakeup from suspend. That cause system crash during reusme, and a 'undefined instruction' is displayed on the console. Set the GRF_FORCE_JTAG bit of RK3288_GRF_SOC_CON0 can ensure that debug modul is reset. And we can change the value of RK3288_GRF_SOC_CON0 back when system resume. Signed-off-by: Chris Zhong Tested-by: Caesar Wang Reviewed-by: Douglas Anderson According to discussions, there does not seem a better solution available. Please also see the potential security implication described in the comment inline in the code. Signed-off-by: Heiko Stuebner diff --git a/arch/arm/mach-rockchip/pm.c b/arch/arm/mach-rockchip/pm.c index b0dcbe2..22812fe 100644 --- a/arch/arm/mach-rockchip/pm.c +++ b/arch/arm/mach-rockchip/pm.c @@ -44,9 +44,11 @@ static void __iomem *rk3288_bootram_base; static phys_addr_t rk3288_bootram_phy; static struct regmap *pmu_regmap; +static struct regmap *grf_regmap; static struct regmap *sgrf_regmap; static u32 rk3288_pmu_pwr_mode_con; +static u32 rk3288_grf_soc_con0; static u32 rk3288_sgrf_soc_con0; static inline u32 rk3288_l2_config(void) @@ -70,12 +72,26 @@ static void rk3288_slp_mode_set(int level) { u32 mode_set, mode_set1; + regmap_read(grf_regmap, RK3288_GRF_SOC_CON0, &rk3288_grf_soc_con0); + regmap_read(sgrf_regmap, RK3288_SGRF_SOC_CON0, &rk3288_sgrf_soc_con0); regmap_read(pmu_regmap, RK3288_PMU_PWRMODE_CON, &rk3288_pmu_pwr_mode_con); /* + * We need set this bit GRF_FORCE_JTAG here, for the debug module, + * otherwise, it may become inaccessible after resume. + * This creates a potential security issue, as the sdmmc pins may + * accept jtag data for a short time during resume if no card is + * inserted. + * But this is of course also true for the regular boot, before we + * turn of the jtag/sdmmc autodetect. + */ + regmap_write(grf_regmap, RK3288_GRF_SOC_CON0, GRF_FORCE_JTAG | + GRF_FORCE_JTAG_WRITE); + + /* * SGRF_FAST_BOOT_EN - system to boot from FAST_BOOT_ADDR * PCLK_WDT_GATE - disable WDT during suspend. */ @@ -135,6 +151,9 @@ static void rk3288_slp_mode_set_resume(void) regmap_write(sgrf_regmap, RK3288_SGRF_SOC_CON0, rk3288_sgrf_soc_con0 | SGRF_PCLK_WDT_GATE_WRITE | SGRF_FAST_BOOT_EN_WRITE); + + regmap_write(grf_regmap, RK3288_GRF_SOC_CON0, rk3288_grf_soc_con0 | + GRF_FORCE_JTAG_WRITE); } static int rockchip_lpmode_enter(unsigned long arg) @@ -193,6 +212,13 @@ static int rk3288_suspend_init(struct device_node *np) return PTR_ERR(pmu_regmap); } + grf_regmap = syscon_regmap_lookup_by_compatible( + "rockchip,rk3288-grf"); + if (IS_ERR(grf_regmap)) { + pr_err("%s: could not find grf regmap\n", __func__); + return PTR_ERR(pmu_regmap); + } + sram_np = of_find_compatible_node(NULL, NULL, "rockchip,rk3288-pmu-sram"); if (!sram_np) { diff --git a/arch/arm/mach-rockchip/pm.h b/arch/arm/mach-rockchip/pm.h index 3e8d39c..f8a747b 100644 --- a/arch/arm/mach-rockchip/pm.h +++ b/arch/arm/mach-rockchip/pm.h @@ -48,6 +48,10 @@ static inline void rockchip_suspend_init(void) #define RK3288_PMU_WAKEUP_RST_CLR_CNT 0x44 #define RK3288_PMU_PWRMODE_CON1 0x90 +#define RK3288_GRF_SOC_CON0 0x244 +#define GRF_FORCE_JTAG BIT(12) +#define GRF_FORCE_JTAG_WRITE BIT(28) + #define RK3288_SGRF_SOC_CON0 (0x0000) #define RK3288_SGRF_FAST_BOOT_ADDR (0x0120) #define SGRF_PCLK_WDT_GATE BIT(6) -- cgit v0.10.2 From 2a9fe3ca84afff6259820c4f62e579f41476becc Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Tue, 20 Jan 2015 23:47:30 +0100 Subject: rockchip: make sure timer7 is enabled on rk3288 platforms timer7 supplies the architected timer and thus as has to run when the system clocksource and clockevents drivers are registered. While it should be the responsibility of the bootloader to do this, and there exists a fix in a community u-boot, all u-boot based systems that actually shipped have the mentioned issue. Therefore to not require every developer to update their u-boot, add a snippet for this, enabling the timer early in the kernel. Signed-off-by: Heiko Stuebner diff --git a/arch/arm/mach-rockchip/rockchip.c b/arch/arm/mach-rockchip/rockchip.c index d360ec0..b6cf3b4 100644 --- a/arch/arm/mach-rockchip/rockchip.c +++ b/arch/arm/mach-rockchip/rockchip.c @@ -30,11 +30,30 @@ #include "pm.h" #define RK3288_GRF_SOC_CON0 0x244 +#define RK3288_TIMER6_7_PHYS 0xff810000 static void __init rockchip_timer_init(void) { if (of_machine_is_compatible("rockchip,rk3288")) { struct regmap *grf; + void __iomem *reg_base; + + /* + * Most/all uboot versions for rk3288 don't enable timer7 + * which is needed for the architected timer to work. + * So make sure it is running during early boot. + */ + reg_base = ioremap(RK3288_TIMER6_7_PHYS, SZ_16K); + if (reg_base) { + writel(0, reg_base + 0x30); + writel(0xffffffff, reg_base + 0x20); + writel(0xffffffff, reg_base + 0x24); + writel(1, reg_base + 0x30); + dsb(); + iounmap(reg_base); + } else { + pr_err("rockchip: could not map timer7 registers\n"); + } /* * Disable auto jtag/sdmmc switching that causes issues -- cgit v0.10.2 From 28ecc0b658e2ac882faa80e7ff1d72d144299bd0 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Wed, 15 Apr 2015 00:08:15 -0300 Subject: ASoC: fsl_ssi: Fix platform_get_irq() error handling We should check whether platform_get_irq() returns a negative number and propagate the error in this case. Signed-off-by: Fabio Estevam Signed-off-by: Mark Brown diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index e8bb8ee..0d48804 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -1357,7 +1357,7 @@ static int fsl_ssi_probe(struct platform_device *pdev) } ssi_private->irq = platform_get_irq(pdev, 0); - if (!ssi_private->irq) { + if (ssi_private->irq < 0) { dev_err(&pdev->dev, "no irq for node %s\n", pdev->name); return ssi_private->irq; } -- cgit v0.10.2 From 427ced4b203dfea4f08b1298cd1f88e19039fca4 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Thu, 16 Apr 2015 20:17:46 +0800 Subject: ASoC: tfa9879: Fix return value check in tfa9879_i2c_probe() In case of error, the function devm_kzalloc() returns NULL not ERR_PTR(). The IS_ERR() test in the return value check should be replaced with NULL test. Signed-off-by: Wei Yongjun Acked-by: Peter Rosin Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/codecs/tfa9879.c b/sound/soc/codecs/tfa9879.c index 16f1b71..aab0af6 100644 --- a/sound/soc/codecs/tfa9879.c +++ b/sound/soc/codecs/tfa9879.c @@ -280,8 +280,8 @@ static int tfa9879_i2c_probe(struct i2c_client *i2c, int i; tfa9879 = devm_kzalloc(&i2c->dev, sizeof(*tfa9879), GFP_KERNEL); - if (IS_ERR(tfa9879)) - return PTR_ERR(tfa9879); + if (!tfa9879) + return -ENOMEM; i2c_set_clientdata(i2c, tfa9879); -- cgit v0.10.2 From c479163a1b6ab424786fbcd9225b4e3c1c58eb0b Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Thu, 16 Apr 2015 20:18:02 +0800 Subject: ASoC: samsung: s3c24xx-i2s: Fix return value check in s3c24xx_iis_dev_probe() In case of error, the function devm_ioremap_resource() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Signed-off-by: Wei Yongjun Reviewed-by: Krzysztof Kozlowski Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/samsung/s3c24xx-i2s.c b/sound/soc/samsung/s3c24xx-i2s.c index 326d3c3..5bf7236 100644 --- a/sound/soc/samsung/s3c24xx-i2s.c +++ b/sound/soc/samsung/s3c24xx-i2s.c @@ -461,8 +461,8 @@ static int s3c24xx_iis_dev_probe(struct platform_device *pdev) return -ENOENT; } s3c24xx_i2s.regs = devm_ioremap_resource(&pdev->dev, res); - if (s3c24xx_i2s.regs == NULL) - return -ENXIO; + if (IS_ERR(s3c24xx_i2s.regs)) + return PTR_ERR(s3c24xx_i2s.regs); s3c24xx_i2s_pcm_stereo_out.dma_addr = res->start + S3C2410_IISFIFO; s3c24xx_i2s_pcm_stereo_in.dma_addr = res->start + S3C2410_IISFIFO; -- cgit v0.10.2 From d09a6b4a1412149c133a58b53f50e9c05a95e834 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Thu, 16 Apr 2015 21:46:29 +0800 Subject: ASoC: Intel: sst_byt: remove kfree for memory allocated with devm_kzalloc It's not necessary to free memory allocated with devm_kzalloc and using kfree leads to a double free. Signed-off-by: Wei Yongjun Acked-by: Jarkko Nikula Signed-off-by: Mark Brown diff --git a/sound/soc/intel/baytrail/sst-baytrail-ipc.c b/sound/soc/intel/baytrail/sst-baytrail-ipc.c index 1efb33b..a839dbf 100644 --- a/sound/soc/intel/baytrail/sst-baytrail-ipc.c +++ b/sound/soc/intel/baytrail/sst-baytrail-ipc.c @@ -759,7 +759,6 @@ fw_err: dsp_new_err: sst_ipc_fini(ipc); ipc_init_err: - kfree(byt); return err; } -- cgit v0.10.2 From c57dcb566d3d866a302a1da2e06344bec31d5bcd Mon Sep 17 00:00:00 2001 From: Ross Lagerwall Date: Thu, 2 Apr 2015 08:39:00 +0100 Subject: efivarfs: Ensure VariableName is NUL-terminated Some buggy firmware implementations update VariableNameSize on success such that it does not include the final NUL character which results in garbage in the efivarfs name entries. Use kzalloc on the efivar_entry (as is done in efivars.c) to ensure that the name is always NUL-terminated. The buggy firmware is: BIOS Information Vendor: Intel Corp. Version: S1200RP.86B.02.02.0005.102320140911 Release Date: 10/23/2014 BIOS Revision: 4.6 System Information Manufacturer: Intel Corporation Product Name: S1200RP_SE Signed-off-by: Ross Lagerwall Acked-by: Matthew Garrett Cc: Jeremy Kerr Cc: Signed-off-by: Matt Fleming diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c index ddbce42..acf9a67 100644 --- a/fs/efivarfs/super.c +++ b/fs/efivarfs/super.c @@ -121,7 +121,7 @@ static int efivarfs_callback(efi_char16_t *name16, efi_guid_t vendor, int len, i; int err = -ENOMEM; - entry = kmalloc(sizeof(*entry), GFP_KERNEL); + entry = kzalloc(sizeof(*entry), GFP_KERNEL); if (!entry) return err; -- cgit v0.10.2 From 98b228f55014870092c15d7d168fecac69f2f12a Mon Sep 17 00:00:00 2001 From: Roy Franz Date: Wed, 15 Apr 2015 16:32:24 -0700 Subject: x86/efi: Store upper bits of command line buffer address in ext_cmd_line_ptr Until now, the EFI stub was only setting the 32 bit cmd_line_ptr in the setup_header structure, so on 64 bit platforms this could be truncated. This patch adds setting the upper bits of the buffer address in ext_cmd_line_ptr. This case was likely never hit, as the allocation for this buffer is done at the lowest available address. Only x86_64 kernels have this problem, as the 1-1 mapping mandated by EFI ensures that all memory is 32 bit addressable on 32 bit platforms. The EFI stub does not support mixed mode, so the 32 bit kernel on 64 bit firmware case does not need to be handled. Signed-off-by: Roy Franz Cc: Signed-off-by: Matt Fleming diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c index 92b9a5f..5999980 100644 --- a/arch/x86/boot/compressed/eboot.c +++ b/arch/x86/boot/compressed/eboot.c @@ -1110,6 +1110,8 @@ struct boot_params *make_boot_params(struct efi_config *c) if (!cmdline_ptr) goto fail; hdr->cmd_line_ptr = (unsigned long)cmdline_ptr; + /* Fill in upper bits of command line address, NOP on 32 bit */ + boot_params->ext_cmd_line_ptr = (u64)(unsigned long)cmdline_ptr >> 32; hdr->ramdisk_image = 0; hdr->ramdisk_size = 0; -- cgit v0.10.2 From aae013d646aae8787b76255d1a1cadd7f64a47dd Mon Sep 17 00:00:00 2001 From: Jie Yang Date: Sun, 19 Apr 2015 09:45:48 +0800 Subject: ASoC: add static inline funcs to fix a compiling issue When CONFIG_PM_SLEEP is not selected, calling funcs snd_soc_suspend and _resume will generate a compiling issue. Here add static inline stub functions to fix it. Signed-off-by: Jie Yang Signed-off-by: Mark Brown diff --git a/include/sound/soc.h b/include/sound/soc.h index 0d1ade1..cd8a125 100644 --- a/include/sound/soc.h +++ b/include/sound/soc.h @@ -387,8 +387,20 @@ int snd_soc_codec_set_pll(struct snd_soc_codec *codec, int pll_id, int source, int snd_soc_register_card(struct snd_soc_card *card); int snd_soc_unregister_card(struct snd_soc_card *card); int devm_snd_soc_register_card(struct device *dev, struct snd_soc_card *card); +#ifdef CONFIG_PM_SLEEP int snd_soc_suspend(struct device *dev); int snd_soc_resume(struct device *dev); +#else +static inline int snd_soc_suspend(struct device *dev) +{ + return 0; +} + +static inline int snd_soc_resume(struct device *dev) +{ + return 0; +} +#endif int snd_soc_poweroff(struct device *dev); int snd_soc_register_platform(struct device *dev, const struct snd_soc_platform_driver *platform_drv); -- cgit v0.10.2 From 625e03461ba51592ac785214b2254af7bd630077 Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Mon, 20 Apr 2015 00:59:50 +0200 Subject: MAINTAINERS: add entry for Rockchip drm drivers Mark Yao looks after the Rockchip drm drivers and should thus also get patches touching these. Signed-off-by: Heiko Stuebner diff --git a/MAINTAINERS b/MAINTAINERS index 6520042..76dfb2b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3398,6 +3398,13 @@ F: drivers/gpu/drm/rcar-du/ F: drivers/gpu/drm/shmobile/ F: include/linux/platform_data/shmob_drm.h +DRM DRIVERS FOR ROCKCHIP +M: Mark Yao +L: dri-devel@lists.freedesktop.org +S: Maintained +F: drivers/gpu/drm/rockchip/ +F: Documentation/devicetree/bindings/video/rockchip* + DSBR100 USB FM RADIO DRIVER M: Alexey Klimov L: linux-media@vger.kernel.org -- cgit v0.10.2 From 3ea68922fc4148abc97557df43d4ba9a136b1c8d Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Mon, 20 Apr 2015 01:00:53 +0200 Subject: drm/rockchip: fix error check when getting irq platform_get_irq() can return negative error values and we already test for these. Therefore the variable holding this value should be signed to not loose possible error values. Reported-by: David Binderman Signed-off-by: Heiko Stuebner Reviewed-By: Daniel Kurtz diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c index ccb0ce0..4557f33 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c @@ -1409,7 +1409,7 @@ static int vop_bind(struct device *dev, struct device *master, void *data) struct vop *vop; struct resource *res; size_t alloc_size; - int ret; + int ret, irq; of_id = of_match_device(vop_driver_dt_match, dev); vop_data = of_id->data; @@ -1445,11 +1445,12 @@ static int vop_bind(struct device *dev, struct device *master, void *data) return ret; } - vop->irq = platform_get_irq(pdev, 0); - if (vop->irq < 0) { + irq = platform_get_irq(pdev, 0); + if (irq < 0) { dev_err(dev, "cannot find irq for vop\n"); - return vop->irq; + return irq; } + vop->irq = (unsigned int)irq; spin_lock_init(&vop->reg_lock); spin_lock_init(&vop->irq_lock); -- cgit v0.10.2 From 94d4b4765b7ddb8478b0d57663cf7a08e2263bbf Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 23 Nov 2012 19:19:07 +0100 Subject: x86/mm: Clean up types in xlate_dev_mem_ptr() Pavel Machek reported the following compiler warning on x86/32 CONFIG_HIGHMEM64G=y builds: arch/x86/mm/ioremap.c:344:10: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] Clean up the types in this function by using a single natural type for internal calculations (unsigned long), to make it more apparent what's happening, and also to remove fragile casts. Reported-by: Pavel Machek Cc: jgross@suse.com Cc: roland@purestorage.com Link: http://lkml.kernel.org/r/20150416080440.GA507@amd Signed-off-by: Ingo Molnar diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index fdf617c..4bf037b 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -332,18 +332,20 @@ EXPORT_SYMBOL(iounmap); */ void *xlate_dev_mem_ptr(phys_addr_t phys) { - void *addr; - unsigned long start = phys & PAGE_MASK; + unsigned long start = phys & PAGE_MASK; + unsigned long offset = phys & ~PAGE_MASK; + unsigned long vaddr; /* If page is RAM, we can use __va. Otherwise ioremap and unmap. */ if (page_is_ram(start >> PAGE_SHIFT)) return __va(phys); - addr = (void __force *)ioremap_cache(start, PAGE_SIZE); - if (addr) - addr = (void *)((unsigned long)addr | (phys & ~PAGE_MASK)); + vaddr = (unsigned long)ioremap_cache(start, PAGE_SIZE); + /* Only add the offset on success and return NULL if the ioremap() failed: */ + if (vaddr) + vaddr += offset; - return addr; + return (void *)vaddr; } void unxlate_dev_mem_ptr(phys_addr_t phys, void *addr) -- cgit v0.10.2 From 3b6e042188994466ec257b71296b5f85b894dcd9 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Tue, 21 Apr 2015 17:26:23 +0200 Subject: perf/x86/intel: Add cpu_(prepare|starting|dying) for core_pmu The core_pmu does not define cpu_* callbacks, which handles allocation of 'struct cpu_hw_events::shared_regs' data, initialization of debug store and PMU_FL_EXCL_CNTRS counters. While this probably won't happen on bare metal, virtual CPU can define x86_pmu.extra_regs together with PMU version 1 and thus be using core_pmu -> using shared_regs data without it being allocated. That could could leave to following panic: BUG: unable to handle kernel NULL pointer dereference at (null) IP: [] _spin_lock_irqsave+0x1f/0x40 SNIP [] __intel_shared_reg_get_constraints+0x69/0x1e0 [] intel_get_event_constraints+0x9b/0x180 [] x86_schedule_events+0x75/0x1d0 [] ? check_preempt_curr+0x7c/0x90 [] ? try_to_wake_up+0x24e/0x3e0 [] ? default_wake_function+0x12/0x20 [] ? autoremove_wake_function+0x16/0x40 [] ? __wake_up_common+0x59/0x90 [] ? __d_lookup+0xa7/0x150 [] ? do_lookup+0x9f/0x230 [] ? dput+0x9a/0x150 [] ? path_to_nameidata+0x25/0x60 [] ? __link_path_walk+0x7da/0x1000 [] ? x86_pmu_add+0xb9/0x170 [] x86_pmu_commit_txn+0x67/0xc0 [] ? mntput_no_expire+0x30/0x110 [] ? path_put+0x31/0x40 [] ? current_fs_time+0x27/0x30 [] ? mem_cgroup_get_reclaim_stat_from_page+0x20/0x70 [] group_sched_in+0x13a/0x170 [] ? sched_clock+0x9/0x10 [] ctx_sched_in+0x2e8/0x330 [] perf_event_sched_in+0x6b/0xb0 [] perf_event_context_sched_in+0x76/0xc0 [] perf_event_comm+0x1bb/0x2e0 [] set_task_comm+0x69/0x80 [] setup_new_exec+0xe1/0x2e0 [] load_elf_binary+0x3ce/0x1ab0 Adding cpu_(prepare|starting|dying) for core_pmu to have shared_regs data allocated for core_pmu. AFAICS there's no harm to initialize debug store and PMU_FL_EXCL_CNTRS either for core_pmu. Signed-off-by: Jiri Olsa Acked-by: Peter Zijlstra (Intel) Cc: Arnaldo Carvalho de Melo Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/20150421152623.GC13169@krava.redhat.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 219d3fb..960e85d 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -2533,34 +2533,6 @@ ssize_t intel_event_sysfs_show(char *page, u64 config) return x86_event_sysfs_show(page, config, event); } -static __initconst const struct x86_pmu core_pmu = { - .name = "core", - .handle_irq = x86_pmu_handle_irq, - .disable_all = x86_pmu_disable_all, - .enable_all = core_pmu_enable_all, - .enable = core_pmu_enable_event, - .disable = x86_pmu_disable_event, - .hw_config = x86_pmu_hw_config, - .schedule_events = x86_schedule_events, - .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, - .perfctr = MSR_ARCH_PERFMON_PERFCTR0, - .event_map = intel_pmu_event_map, - .max_events = ARRAY_SIZE(intel_perfmon_event_map), - .apic = 1, - /* - * Intel PMCs cannot be accessed sanely above 32 bit width, - * so we install an artificial 1<<31 period regardless of - * the generic event period: - */ - .max_period = (1ULL << 31) - 1, - .get_event_constraints = intel_get_event_constraints, - .put_event_constraints = intel_put_event_constraints, - .event_constraints = intel_core_event_constraints, - .guest_get_msrs = core_guest_get_msrs, - .format_attrs = intel_arch_formats_attr, - .events_sysfs_show = intel_event_sysfs_show, -}; - struct intel_shared_regs *allocate_shared_regs(int cpu) { struct intel_shared_regs *regs; @@ -2743,6 +2715,44 @@ static struct attribute *intel_arch3_formats_attr[] = { NULL, }; +static __initconst const struct x86_pmu core_pmu = { + .name = "core", + .handle_irq = x86_pmu_handle_irq, + .disable_all = x86_pmu_disable_all, + .enable_all = core_pmu_enable_all, + .enable = core_pmu_enable_event, + .disable = x86_pmu_disable_event, + .hw_config = x86_pmu_hw_config, + .schedule_events = x86_schedule_events, + .eventsel = MSR_ARCH_PERFMON_EVENTSEL0, + .perfctr = MSR_ARCH_PERFMON_PERFCTR0, + .event_map = intel_pmu_event_map, + .max_events = ARRAY_SIZE(intel_perfmon_event_map), + .apic = 1, + /* + * Intel PMCs cannot be accessed sanely above 32-bit width, + * so we install an artificial 1<<31 period regardless of + * the generic event period: + */ + .max_period = (1ULL<<31) - 1, + .get_event_constraints = intel_get_event_constraints, + .put_event_constraints = intel_put_event_constraints, + .event_constraints = intel_core_event_constraints, + .guest_get_msrs = core_guest_get_msrs, + .format_attrs = intel_arch_formats_attr, + .events_sysfs_show = intel_event_sysfs_show, + + /* + * Virtual (or funny metal) CPU can define x86_pmu.extra_regs + * together with PMU version 1 and thus be using core_pmu with + * shared_regs. We need following callbacks here to allocate + * it properly. + */ + .cpu_prepare = intel_pmu_cpu_prepare, + .cpu_starting = intel_pmu_cpu_starting, + .cpu_dying = intel_pmu_cpu_dying, +}; + static __initconst const struct x86_pmu intel_pmu = { .name = "Intel", .handle_irq = intel_pmu_handle_irq, -- cgit v0.10.2 From 80bcffb376a6890dd7452b12c1ba032f8f24fef6 Mon Sep 17 00:00:00 2001 From: Sonny Rao Date: Mon, 20 Apr 2015 15:34:07 -0700 Subject: perf/x86/intel/uncore: Add support for Intel Haswell ULT (lower power Mobile Processor) IMC uncore PMUs This uncore is the same as the Haswell desktop part but uses a different PCI ID. Signed-off-by: Sonny Rao Cc: Arnaldo Carvalho de Melo Cc: Bjorn Helgaas Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1429569247-16697-1-git-send-email-sonnyrao@chromium.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c b/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c index 3001015..ca75e70 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c +++ b/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c @@ -1,6 +1,9 @@ /* Nehalem/SandBridge/Haswell uncore support */ #include "perf_event_intel_uncore.h" +/* Uncore IMC PCI Id */ +#define PCI_DEVICE_ID_INTEL_HSW_U_IMC 0x0a04 + /* SNB event control */ #define SNB_UNC_CTL_EV_SEL_MASK 0x000000ff #define SNB_UNC_CTL_UMASK_MASK 0x0000ff00 @@ -472,6 +475,10 @@ static const struct pci_device_id hsw_uncore_pci_ids[] = { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_HSW_IMC), .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0), }, + { /* IMC */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_HSW_U_IMC), + .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0), + }, { /* end: all zeroes */ }, }; @@ -502,6 +509,7 @@ static const struct imc_uncore_pci_dev desktop_imc_pci_ids[] = { IMC_DEV(IVB_IMC, &ivb_uncore_pci_driver), /* 3rd Gen Core processor */ IMC_DEV(IVB_E3_IMC, &ivb_uncore_pci_driver), /* Xeon E3-1200 v2/3rd Gen Core processor */ IMC_DEV(HSW_IMC, &hsw_uncore_pci_driver), /* 4th Gen Core Processor */ + IMC_DEV(HSW_U_IMC, &hsw_uncore_pci_driver), /* 4th Gen Core ULT Mobile Processor */ { /* end marker */ } }; -- cgit v0.10.2 From 0140e6141e4f1d4b15fb469e6912b0e71b7d1cc2 Mon Sep 17 00:00:00 2001 From: Sonny Rao Date: Tue, 21 Apr 2015 12:33:11 -0700 Subject: perf/x86/intel/uncore: Move PCI IDs for IMC to uncore driver This keeps all the related PCI IDs together in the driver where they are used. Signed-off-by: Sonny Rao Acked-by: Bjorn Helgaas Cc: Arnaldo Carvalho de Melo Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1429644791-25724-1-git-send-email-sonnyrao@chromium.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c b/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c index ca75e70..4562e9e 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c +++ b/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c @@ -1,7 +1,11 @@ /* Nehalem/SandBridge/Haswell uncore support */ #include "perf_event_intel_uncore.h" -/* Uncore IMC PCI Id */ +/* Uncore IMC PCI IDs */ +#define PCI_DEVICE_ID_INTEL_SNB_IMC 0x0100 +#define PCI_DEVICE_ID_INTEL_IVB_IMC 0x0154 +#define PCI_DEVICE_ID_INTEL_IVB_E3_IMC 0x0150 +#define PCI_DEVICE_ID_INTEL_HSW_IMC 0x0c00 #define PCI_DEVICE_ID_INTEL_HSW_U_IMC 0x0a04 /* SNB event control */ diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index e63c02a..a593858 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2539,10 +2539,6 @@ #define PCI_VENDOR_ID_INTEL 0x8086 #define PCI_DEVICE_ID_INTEL_EESSC 0x0008 -#define PCI_DEVICE_ID_INTEL_SNB_IMC 0x0100 -#define PCI_DEVICE_ID_INTEL_IVB_IMC 0x0154 -#define PCI_DEVICE_ID_INTEL_IVB_E3_IMC 0x0150 -#define PCI_DEVICE_ID_INTEL_HSW_IMC 0x0c00 #define PCI_DEVICE_ID_INTEL_PXHD_0 0x0320 #define PCI_DEVICE_ID_INTEL_PXHD_1 0x0321 #define PCI_DEVICE_ID_INTEL_PXH_0 0x0329 -- cgit v0.10.2 From d83769a580f1132ac26439f50068a29b02be535e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 21 Apr 2015 18:32:24 -0700 Subject: tcp: fix possible deadlock in tcp_send_fin() Using sk_stream_alloc_skb() in tcp_send_fin() is dangerous in case a huge process is killed by OOM, and tcp_mem[2] is hit. To be able to free memory we need to make progress, so this patch allows FIN packets to not care about tcp_mem[2], if skb allocation succeeded. In a follow-up patch, we might abort tcp_send_fin() infinite loop in case TIF_MEMDIE is set on this thread, as memory allocator did its best getting extra memory already. This patch reverts d22e15371811 ("tcp: fix tcp fin memory accounting") Fixes: d22e15371811 ("tcp: fix tcp fin memory accounting") Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 8c8d7e0..2ade67b 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2812,6 +2812,21 @@ begin_fwd: } } +/* We allow to exceed memory limits for FIN packets to expedite + * connection tear down and (memory) recovery. + * Otherwise tcp_send_fin() could loop forever. + */ +static void sk_forced_wmem_schedule(struct sock *sk, int size) +{ + int amt, status; + + if (size <= sk->sk_forward_alloc) + return; + amt = sk_mem_pages(size); + sk->sk_forward_alloc += amt * SK_MEM_QUANTUM; + sk_memory_allocated_add(sk, amt, &status); +} + /* Send a fin. The caller locks the socket for us. This cannot be * allowed to fail queueing a FIN frame under any circumstances. */ @@ -2834,11 +2849,14 @@ void tcp_send_fin(struct sock *sk) } else { /* Socket is locked, keep trying until memory is available. */ for (;;) { - skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation); + skb = alloc_skb_fclone(MAX_TCP_HEADER, + sk->sk_allocation); if (skb) break; yield(); } + skb_reserve(skb, MAX_TCP_HEADER); + sk_forced_wmem_schedule(sk, skb->truesize); /* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */ tcp_init_nondata_skb(skb, tp->write_seq, TCPHDR_ACK | TCPHDR_FIN); -- cgit v0.10.2 From e2307ed6cbe71c74e291681aaa7e92ab98bc3177 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Wed, 22 Apr 2015 09:41:45 +0200 Subject: rhashtable: Schedule async resize when sync realloc fails When rhashtable_insert_rehash() fails with ENOMEM, this indicates that we can't allocate the necessary memory in the current context but the limits as set by the user would still allow to grow. Thus attempt an async resize in the background where we can allocate using GFP_KERNEL which is more likely to succeed. The insertion itself will still fail to indicate pressure. This fixes a bug where the table would never continue growing once the utilization is above 100%. Fixes: ccd57b1bd324 ("rhashtable: Add immediate rehash during insertion") Signed-off-by: Thomas Graf Acked-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/lib/rhashtable.c b/lib/rhashtable.c index 4898442..f648cfd 100644 --- a/lib/rhashtable.c +++ b/lib/rhashtable.c @@ -410,8 +410,13 @@ int rhashtable_insert_rehash(struct rhashtable *ht) return -EBUSY; new_tbl = bucket_table_alloc(ht, size, GFP_ATOMIC); - if (new_tbl == NULL) + if (new_tbl == NULL) { + /* Schedule async resize/rehash to try allocation + * non-atomic context. + */ + schedule_work(&ht->run_work); return -ENOMEM; + } err = rhashtable_rehash_attach(ht, tbl, new_tbl); if (err) { -- cgit v0.10.2 From a87b9ebf1709687ff213091d0fdb4254b1564803 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Wed, 22 Apr 2015 09:41:46 +0200 Subject: rhashtable: Do not schedule more than one rehash if we can't grow further The current code currently only stops inserting rehashes into the chain when no resizes are currently scheduled. As long as resizes are scheduled and while inserting above the utilization watermark, more and more rehashes will be scheduled. This lead to a perfect DoS storm with thousands of rehashes scheduled which lead to thousands of spinlocks to be taken sequentially. Instead, only allow either a series of resizes or a single rehash. Drop any further rehashes and return -EBUSY. Fixes: ccd57b1bd324 ("rhashtable: Add immediate rehash during insertion") Signed-off-by: Thomas Graf Acked-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/lib/rhashtable.c b/lib/rhashtable.c index f648cfd..b28df40 100644 --- a/lib/rhashtable.c +++ b/lib/rhashtable.c @@ -405,8 +405,8 @@ int rhashtable_insert_rehash(struct rhashtable *ht) if (rht_grow_above_75(ht, tbl)) size *= 2; - /* More than two rehashes (not resizes) detected. */ - else if (WARN_ON(old_tbl != tbl && old_tbl->size == size)) + /* Do not schedule more than one rehash */ + else if (old_tbl != tbl) return -EBUSY; new_tbl = bucket_table_alloc(ht, size, GFP_ATOMIC); -- cgit v0.10.2 From 909d9faae2a447110aa061070145297fffe129cb Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Wed, 22 Apr 2015 12:47:32 +0300 Subject: bnx2x: Prevent inner-reload while VFs exist On some feature changes, driver employes an inner-reload flow where it resets the function and re-configures it with the new required set of parameters. Such a flow proves fatal to any VF since those were not intended to be used while HW is being reset underneath, causing them [at best] to lose all connectivity. This changes driver behavior to fail all configuration changes [e.g., mtu change] requested of the driver in case VFs are active. Signed-off-by: Yuval Mintz Signed-off-by: Ariel Elior Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 2f63467..6f7dc81 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -4809,6 +4809,23 @@ netdev_features_t bnx2x_fix_features(struct net_device *dev, { struct bnx2x *bp = netdev_priv(dev); + if (pci_num_vf(bp->pdev)) { + netdev_features_t changed = dev->features ^ features; + + /* Revert the requested changes in features if they + * would require internal reload of PF in bnx2x_set_features(). + */ + if (!(features & NETIF_F_RXCSUM) && !bp->disable_tpa) { + features &= ~NETIF_F_RXCSUM; + features |= dev->features & NETIF_F_RXCSUM; + } + + if (changed & NETIF_F_LOOPBACK) { + features &= ~NETIF_F_LOOPBACK; + features |= dev->features & NETIF_F_LOOPBACK; + } + } + /* TPA requires Rx CSUM offloading */ if (!(features & NETIF_F_RXCSUM)) { features &= ~NETIF_F_LRO; @@ -4839,15 +4856,18 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features) else flags &= ~GRO_ENABLE_FLAG; - if (features & NETIF_F_LOOPBACK) { - if (bp->link_params.loopback_mode != LOOPBACK_BMAC) { - bp->link_params.loopback_mode = LOOPBACK_BMAC; - bnx2x_reload = true; - } - } else { - if (bp->link_params.loopback_mode != LOOPBACK_NONE) { - bp->link_params.loopback_mode = LOOPBACK_NONE; - bnx2x_reload = true; + /* VFs or non SRIOV PFs should be able to change loopback feature */ + if (!pci_num_vf(bp->pdev)) { + if (features & NETIF_F_LOOPBACK) { + if (bp->link_params.loopback_mode != LOOPBACK_BMAC) { + bp->link_params.loopback_mode = LOOPBACK_BMAC; + bnx2x_reload = true; + } + } else { + if (bp->link_params.loopback_mode != LOOPBACK_NONE) { + bp->link_params.loopback_mode = LOOPBACK_NONE; + bnx2x_reload = true; + } } } @@ -4931,6 +4951,11 @@ int bnx2x_resume(struct pci_dev *pdev) } bp = netdev_priv(dev); + if (pci_num_vf(bp->pdev)) { + DP(BNX2X_MSG_IOV, "VFs are enabled, can not change MTU\n"); + return -EPERM; + } + if (bp->recovery_state != BNX2X_RECOVERY_DONE) { BNX2X_ERR("Handling parity error recovery. Try again later\n"); return -EAGAIN; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c index e3d853c..48ed005 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c @@ -1843,6 +1843,12 @@ static int bnx2x_set_ringparam(struct net_device *dev, "set ring params command parameters: rx_pending = %d, tx_pending = %d\n", ering->rx_pending, ering->tx_pending); + if (pci_num_vf(bp->pdev)) { + DP(BNX2X_MSG_IOV, + "VFs are enabled, can not change ring parameters\n"); + return -EPERM; + } + if (bp->recovery_state != BNX2X_RECOVERY_DONE) { DP(BNX2X_MSG_ETHTOOL, "Handling parity error recovery. Try again later\n"); @@ -2899,6 +2905,12 @@ static void bnx2x_self_test(struct net_device *dev, u8 is_serdes, link_up; int rc, cnt = 0; + if (pci_num_vf(bp->pdev)) { + DP(BNX2X_MSG_IOV, + "VFs are enabled, can not perform self test\n"); + return; + } + if (bp->recovery_state != BNX2X_RECOVERY_DONE) { netdev_err(bp->dev, "Handling parity error recovery. Try again later\n"); @@ -3468,6 +3480,11 @@ static int bnx2x_set_channels(struct net_device *dev, channels->rx_count, channels->tx_count, channels->other_count, channels->combined_count); + if (pci_num_vf(bp->pdev)) { + DP(BNX2X_MSG_IOV, "VFs are enabled, can not set channels\n"); + return -EPERM; + } + /* We don't support separate rx / tx channels. * We don't allow setting 'other' channels. */ -- cgit v0.10.2 From 03c57747a7020a28a200e7e920fb48ecdc9b0fb8 Mon Sep 17 00:00:00 2001 From: Robert Shearman Date: Wed, 22 Apr 2015 11:14:37 +0100 Subject: mpls: Per-device MPLS state Add per-device MPLS state to supported interfaces. Use the presence of this state in mpls_route_add to determine that this is a supported interface. Use the presence of mpls_dev to drop packets that arrived on an unsupported interface - previously they were allowed through. Cc: "Eric W. Biederman" Signed-off-by: Robert Shearman Reviewed-by: "Eric W. Biederman" Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index bcbde79..dae106a 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -60,6 +60,7 @@ struct phy_device; struct wireless_dev; /* 802.15.4 specific */ struct wpan_dev; +struct mpls_dev; void netdev_set_default_ethtool_ops(struct net_device *dev, const struct ethtool_ops *ops); @@ -1627,6 +1628,9 @@ struct net_device { void *ax25_ptr; struct wireless_dev *ieee80211_ptr; struct wpan_dev *ieee802154_ptr; +#if IS_ENABLED(CONFIG_MPLS_ROUTING) + struct mpls_dev __rcu *mpls_ptr; +#endif /* * Cache lines mostly used on receive path (including eth_type_trans()) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index db8a2ea..ad45017 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -53,6 +53,11 @@ static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned index) return rt; } +static inline struct mpls_dev *mpls_dev_get(const struct net_device *dev) +{ + return rcu_dereference_rtnl(dev->mpls_ptr); +} + static bool mpls_output_possible(const struct net_device *dev) { return dev && (dev->flags & IFF_UP) && netif_carrier_ok(dev); @@ -136,6 +141,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, struct mpls_route *rt; struct mpls_entry_decoded dec; struct net_device *out_dev; + struct mpls_dev *mdev; unsigned int hh_len; unsigned int new_header_size; unsigned int mtu; @@ -143,6 +149,10 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, /* Careful this entire function runs inside of an rcu critical section */ + mdev = mpls_dev_get(dev); + if (!mdev) + goto drop; + if (skb->pkt_type != PACKET_HOST) goto drop; @@ -352,9 +362,9 @@ static int mpls_route_add(struct mpls_route_config *cfg) if (!dev) goto errout; - /* For now just support ethernet devices */ + /* Ensure this is a supported device */ err = -EINVAL; - if ((dev->type != ARPHRD_ETHER) && (dev->type != ARPHRD_LOOPBACK)) + if (!mpls_dev_get(dev)) goto errout; err = -EINVAL; @@ -428,10 +438,27 @@ errout: return err; } +static struct mpls_dev *mpls_add_dev(struct net_device *dev) +{ + struct mpls_dev *mdev; + int err = -ENOMEM; + + ASSERT_RTNL(); + + mdev = kzalloc(sizeof(*mdev), GFP_KERNEL); + if (!mdev) + return ERR_PTR(err); + + rcu_assign_pointer(dev->mpls_ptr, mdev); + + return mdev; +} + static void mpls_ifdown(struct net_device *dev) { struct mpls_route __rcu **platform_label; struct net *net = dev_net(dev); + struct mpls_dev *mdev; unsigned index; platform_label = rtnl_dereference(net->mpls.platform_label); @@ -443,14 +470,33 @@ static void mpls_ifdown(struct net_device *dev) continue; rt->rt_dev = NULL; } + + mdev = mpls_dev_get(dev); + if (!mdev) + return; + + RCU_INIT_POINTER(dev->mpls_ptr, NULL); + + kfree(mdev); } static int mpls_dev_notify(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct mpls_dev *mdev; switch(event) { + case NETDEV_REGISTER: + /* For now just support ethernet devices */ + if ((dev->type == ARPHRD_ETHER) || + (dev->type == ARPHRD_LOOPBACK)) { + mdev = mpls_add_dev(dev); + if (IS_ERR(mdev)) + return notifier_from_errno(PTR_ERR(mdev)); + } + break; + case NETDEV_UNREGISTER: mpls_ifdown(dev); break; diff --git a/net/mpls/internal.h b/net/mpls/internal.h index fb6de92..8090cb3 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -22,6 +22,9 @@ struct mpls_entry_decoded { u8 bos; }; +struct mpls_dev { +}; + struct sk_buff; static inline struct mpls_shim_hdr *mpls_hdr(const struct sk_buff *skb) -- cgit v0.10.2 From 37bde79979c3862c79294c62ddcef7efc477e4bf Mon Sep 17 00:00:00 2001 From: Robert Shearman Date: Wed, 22 Apr 2015 11:14:38 +0100 Subject: mpls: Per-device enabling of packet input An MPLS network is a single trust domain where the edges must be in control of what labels make their way into the core. The simplest way of ensuring this is for the edge device to always impose the labels, and not allow forward labeled traffic from untrusted neighbours. This is achieved by allowing a per-device configuration of whether MPLS traffic input from that interface should be processed or not. To be secure by default, the default state is changed to MPLS being disabled on all interfaces unless explicitly enabled and no global option is provided to change the default. Whilst this differs from other protocols (e.g. IPv6), network operators are used to explicitly enabling MPLS forwarding on interfaces, and with the number of links to the MPLS core typically fairly low this doesn't present too much of a burden on operators. Cc: "Eric W. Biederman" Signed-off-by: Robert Shearman Reviewed-by: "Eric W. Biederman" Signed-off-by: David S. Miller diff --git a/Documentation/networking/mpls-sysctl.txt b/Documentation/networking/mpls-sysctl.txt index 639ddf0..9ed15f8 100644 --- a/Documentation/networking/mpls-sysctl.txt +++ b/Documentation/networking/mpls-sysctl.txt @@ -18,3 +18,12 @@ platform_labels - INTEGER Possible values: 0 - 1048575 Default: 0 + +conf//input - BOOL + Control whether packets can be input on this interface. + + If disabled, packets will be discarded without further + processing. + + 0 - disabled (default) + not 0 - enabled diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index ad45017..9fdd94c 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -150,7 +150,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, /* Careful this entire function runs inside of an rcu critical section */ mdev = mpls_dev_get(dev); - if (!mdev) + if (!mdev || !mdev->input_enabled) goto drop; if (skb->pkt_type != PACKET_HOST) @@ -438,6 +438,60 @@ errout: return err; } +#define MPLS_PERDEV_SYSCTL_OFFSET(field) \ + (&((struct mpls_dev *)0)->field) + +static const struct ctl_table mpls_dev_table[] = { + { + .procname = "input", + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + .data = MPLS_PERDEV_SYSCTL_OFFSET(input_enabled), + }, + { } +}; + +static int mpls_dev_sysctl_register(struct net_device *dev, + struct mpls_dev *mdev) +{ + char path[sizeof("net/mpls/conf/") + IFNAMSIZ]; + struct ctl_table *table; + int i; + + table = kmemdup(&mpls_dev_table, sizeof(mpls_dev_table), GFP_KERNEL); + if (!table) + goto out; + + /* Table data contains only offsets relative to the base of + * the mdev at this point, so make them absolute. + */ + for (i = 0; i < ARRAY_SIZE(mpls_dev_table); i++) + table[i].data = (char *)mdev + (uintptr_t)table[i].data; + + snprintf(path, sizeof(path), "net/mpls/conf/%s", dev->name); + + mdev->sysctl = register_net_sysctl(dev_net(dev), path, table); + if (!mdev->sysctl) + goto free; + + return 0; + +free: + kfree(table); +out: + return -ENOBUFS; +} + +static void mpls_dev_sysctl_unregister(struct mpls_dev *mdev) +{ + struct ctl_table *table; + + table = mdev->sysctl->ctl_table_arg; + unregister_net_sysctl_table(mdev->sysctl); + kfree(table); +} + static struct mpls_dev *mpls_add_dev(struct net_device *dev) { struct mpls_dev *mdev; @@ -449,9 +503,17 @@ static struct mpls_dev *mpls_add_dev(struct net_device *dev) if (!mdev) return ERR_PTR(err); + err = mpls_dev_sysctl_register(dev, mdev); + if (err) + goto free; + rcu_assign_pointer(dev->mpls_ptr, mdev); return mdev; + +free: + kfree(mdev); + return ERR_PTR(err); } static void mpls_ifdown(struct net_device *dev) @@ -475,6 +537,8 @@ static void mpls_ifdown(struct net_device *dev) if (!mdev) return; + mpls_dev_sysctl_unregister(mdev); + RCU_INIT_POINTER(dev->mpls_ptr, NULL); kfree(mdev); @@ -958,7 +1022,7 @@ static int mpls_platform_labels(struct ctl_table *table, int write, return ret; } -static struct ctl_table mpls_table[] = { +static const struct ctl_table mpls_table[] = { { .procname = "platform_labels", .data = NULL, diff --git a/net/mpls/internal.h b/net/mpls/internal.h index 8090cb3..693877d 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -23,6 +23,9 @@ struct mpls_entry_decoded { }; struct mpls_dev { + int input_enabled; + + struct ctl_table_header *sysctl; }; struct sk_buff; -- cgit v0.10.2 From 5a9ab0176198d91dfc153f5e6c5fdc5afa613607 Mon Sep 17 00:00:00 2001 From: Robert Shearman Date: Wed, 22 Apr 2015 11:14:39 +0100 Subject: mpls: Prevent use of implicit NULL label as outgoing label The reserved implicit-NULL label isn't allowed to appear in the label stack for packets, so make it an error for the control plane to specify it as an outgoing label. Suggested-by: "Eric W. Biederman" Signed-off-by: Robert Shearman Reviewed-by: "Eric W. Biederman" Signed-off-by: David S. Miller diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 9fdd94c..954810c 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -646,6 +646,15 @@ int nla_get_labels(const struct nlattr *nla, if ((dec.bos != bos) || dec.ttl || dec.tc) return -EINVAL; + switch (dec.label) { + case LABEL_IMPLICIT_NULL: + /* RFC3032: This is a label that an LSR may + * assign and distribute, but which never + * actually appears in the encapsulation. + */ + return -EINVAL; + } + label[i] = dec.label; } *labels = nla_labels; -- cgit v0.10.2 From 26349c71b4323e62f8de0fe45f2f06c4df535b9b Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 22 Apr 2015 16:56:16 +0200 Subject: ip6_gre: use netdev_alloc_pcpu_stats() The code there just open-codes the same, so use the provided macro instead. Signed-off-by: Johannes Berg Signed-off-by: David S. Miller diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index b5e6cc1..a38d3ac 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -1246,7 +1246,6 @@ static void ip6gre_tunnel_setup(struct net_device *dev) static int ip6gre_tunnel_init(struct net_device *dev) { struct ip6_tnl *tunnel; - int i; tunnel = netdev_priv(dev); @@ -1260,16 +1259,10 @@ static int ip6gre_tunnel_init(struct net_device *dev) if (ipv6_addr_any(&tunnel->parms.raddr)) dev->header_ops = &ip6gre_header_ops; - dev->tstats = alloc_percpu(struct pcpu_sw_netstats); + dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); if (!dev->tstats) return -ENOMEM; - for_each_possible_cpu(i) { - struct pcpu_sw_netstats *ip6gre_tunnel_stats; - ip6gre_tunnel_stats = per_cpu_ptr(dev->tstats, i); - u64_stats_init(&ip6gre_tunnel_stats->syncp); - } - return 0; } -- cgit v0.10.2 From a2d97723cb3a7741af81868427b36bba274b681b Mon Sep 17 00:00:00 2001 From: Charles Keepax Date: Wed, 22 Apr 2015 13:58:47 +0100 Subject: ASoC: dapm: Enable autodisable on SOC_DAPM_SINGLE_TLV_AUTODISABLE Correct small copy and paste error where autodisable was not being enabled for the SOC_DAPM_SINGLE_TLV_AUTODISABLE control. Signed-off-by: Charles Keepax Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/include/sound/soc-dapm.h b/include/sound/soc-dapm.h index 8d7416e..1535589 100644 --- a/include/sound/soc-dapm.h +++ b/include/sound/soc-dapm.h @@ -287,7 +287,7 @@ struct device; .access = SNDRV_CTL_ELEM_ACCESS_TLV_READ | SNDRV_CTL_ELEM_ACCESS_READWRITE,\ .tlv.p = (tlv_array), \ .get = snd_soc_dapm_get_volsw, .put = snd_soc_dapm_put_volsw, \ - .private_value = SOC_SINGLE_VALUE(reg, shift, max, invert, 0) } + .private_value = SOC_SINGLE_VALUE(reg, shift, max, invert, 1) } #define SOC_DAPM_SINGLE_TLV_VIRT(xname, max, tlv_array) \ SOC_DAPM_SINGLE(xname, SND_SOC_NOPM, 0, max, 0, tlv_array) #define SOC_DAPM_ENUM(xname, xenum) \ -- cgit v0.10.2 From 79930f5892e134c6da1254389577fffb8bd72c66 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 22 Apr 2015 07:33:36 -0700 Subject: net: do not deplete pfmemalloc reserve build_skb() should look at the page pfmemalloc status. If set, this means page allocator allocated this page in the expectation it would help to free other pages. Networking stack can do that only if skb->pfmemalloc is also set. Also, we must refrain using high order pages from the pfmemalloc reserve, so __page_frag_refill() must also use __GFP_NOMEMALLOC for them. Under memory pressure, using order-0 pages is probably the best strategy. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d1967da..456ead5 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -311,7 +311,11 @@ struct sk_buff *build_skb(void *data, unsigned int frag_size) memset(skb, 0, offsetof(struct sk_buff, tail)); skb->truesize = SKB_TRUESIZE(size); - skb->head_frag = frag_size != 0; + if (frag_size) { + skb->head_frag = 1; + if (virt_to_head_page(data)->pfmemalloc) + skb->pfmemalloc = 1; + } atomic_set(&skb->users, 1); skb->head = data; skb->data = data; @@ -348,7 +352,8 @@ static struct page *__page_frag_refill(struct netdev_alloc_cache *nc, gfp_t gfp = gfp_mask; if (order) { - gfp_mask |= __GFP_COMP | __GFP_NOWARN | __GFP_NORETRY; + gfp_mask |= __GFP_COMP | __GFP_NOWARN | __GFP_NORETRY | + __GFP_NOMEMALLOC; page = alloc_pages_node(NUMA_NO_NODE, gfp_mask, order); nc->frag.size = PAGE_SIZE << (page ? order : 0); } -- cgit v0.10.2 From 21d3515ce7179523a2941cc015960dd788290e33 Mon Sep 17 00:00:00 2001 From: Ben Shelton Date: Wed, 22 Apr 2015 17:28:54 -0500 Subject: net/macb: Factor out one-time assignment from loop In 02c958dd3 (net/macb: add TX multiqueue support for gem), the initialization of tx_head and tx_tail in macb_init_rings() was moved inside the loop that iterates over each element in the ring. Since tx_head and tx_tail only need to be assigned once, move them back out of the loop. Signed-off-by: Ben Shelton Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 9f53872..665c290 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -1473,9 +1473,9 @@ static void macb_init_rings(struct macb *bp) for (i = 0; i < TX_RING_SIZE; i++) { bp->queues[0].tx_ring[i].addr = 0; bp->queues[0].tx_ring[i].ctrl = MACB_BIT(TX_USED); - bp->queues[0].tx_head = 0; - bp->queues[0].tx_tail = 0; } + bp->queues[0].tx_head = 0; + bp->queues[0].tx_tail = 0; bp->queues[0].tx_ring[TX_RING_SIZE - 1].ctrl |= MACB_BIT(TX_WRAP); bp->rx_tail = 0; -- cgit v0.10.2 From 608404290e2d9d1756db4013c4ee12fa7617dad9 Mon Sep 17 00:00:00 2001 From: Li RongQing Date: Wed, 22 Apr 2015 15:49:10 +0800 Subject: vxlan: remove the unnecessary codes The return value of vxlan_fdb_replace always is greater than or equal to 0 Signed-off-by: Li RongQing Signed-off-by: David S. Miller diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 154116a..27a5f95 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -730,12 +730,8 @@ static int vxlan_fdb_create(struct vxlan_dev *vxlan, /* Only change unicasts */ if (!(is_multicast_ether_addr(f->eth_addr) || is_zero_ether_addr(f->eth_addr))) { - int rc = vxlan_fdb_replace(f, ip, port, vni, + notify |= vxlan_fdb_replace(f, ip, port, vni, ifindex); - - if (rc < 0) - return rc; - notify |= rc; } else return -EOPNOTSUPP; } -- cgit v0.10.2 From 8faf141a9903477910387af062ece04ea7d730ed Mon Sep 17 00:00:00 2001 From: Vinod Koul Date: Thu, 23 Apr 2015 14:38:13 +0530 Subject: ASoC: Intel: fix the makefile for atom code The tom code should be using SND_SST_MFLD_PLATFORM and not the baytrail one. So fix it now Signed-off-by: Vinod Koul Signed-off-by: Mark Brown diff --git a/sound/soc/intel/Makefile b/sound/soc/intel/Makefile index cd9aee9..3853ec2 100644 --- a/sound/soc/intel/Makefile +++ b/sound/soc/intel/Makefile @@ -4,7 +4,7 @@ obj-$(CONFIG_SND_SOC_INTEL_SST) += common/ # Platform Support obj-$(CONFIG_SND_SOC_INTEL_HASWELL) += haswell/ obj-$(CONFIG_SND_SOC_INTEL_BAYTRAIL) += baytrail/ -obj-$(CONFIG_SND_SOC_INTEL_BAYTRAIL) += atom/ +obj-$(CONFIG_SND_SST_MFLD_PLATFORM) += atom/ # Machine support obj-$(CONFIG_SND_SOC_INTEL_SST) += boards/ -- cgit v0.10.2 From 7e01b5acd88b3f3108d8c4ce44e3205d67437202 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 16 Apr 2015 14:47:33 +0200 Subject: kexec: allocate the kexec control page with KEXEC_CONTROL_MEMORY_GFP Introduce KEXEC_CONTROL_MEMORY_GFP to allow the architecture code to override the gfp flags of the allocation for the kexec control page. The loop in kimage_alloc_normal_control_pages allocates pages with GFP_KERNEL until a page is found that happens to have an address smaller than the KEXEC_CONTROL_MEMORY_LIMIT. On systems with a large memory size but a small KEXEC_CONTROL_MEMORY_LIMIT the loop will keep allocating memory until the oom killer steps in. Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/include/asm/kexec.h b/arch/s390/include/asm/kexec.h index 694bcd6..2f924bc 100644 --- a/arch/s390/include/asm/kexec.h +++ b/arch/s390/include/asm/kexec.h @@ -26,6 +26,9 @@ /* Not more than 2GB */ #define KEXEC_CONTROL_MEMORY_LIMIT (1UL<<31) +/* Allocate control page with GFP_DMA */ +#define KEXEC_CONTROL_MEMORY_GFP GFP_DMA + /* Maximum address we can use for the crash control pages */ #define KEXEC_CRASH_CONTROL_MEMORY_LIMIT (-1UL) diff --git a/include/linux/kexec.h b/include/linux/kexec.h index e60a745..e804306 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -40,6 +40,10 @@ #error KEXEC_CONTROL_MEMORY_LIMIT not defined #endif +#ifndef KEXEC_CONTROL_MEMORY_GFP +#define KEXEC_CONTROL_MEMORY_GFP GFP_KERNEL +#endif + #ifndef KEXEC_CONTROL_PAGE_SIZE #error KEXEC_CONTROL_PAGE_SIZE not defined #endif diff --git a/kernel/kexec.c b/kernel/kexec.c index 38c25b1..7a36fdc 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -707,7 +707,7 @@ static struct page *kimage_alloc_normal_control_pages(struct kimage *image, do { unsigned long pfn, epfn, addr, eaddr; - pages = kimage_alloc_pages(GFP_KERNEL, order); + pages = kimage_alloc_pages(KEXEC_CONTROL_MEMORY_GFP, order); if (!pages) break; pfn = page_to_pfn(pages); -- cgit v0.10.2 From 0b46e0a3ec0d7a04af6a091354f1b5e1b952d70a Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 15 Apr 2015 13:23:26 +0200 Subject: s390/kvm: remove delayed reallocation of page tables for KVM Replacing a 2K page table with a 4K page table while a VMA is active for the affected memory region is fundamentally broken. Rip out the page table reallocation code and replace it with a simple system control 'vm.allocate_pgste'. If the system control is set the page tables for all processes are allocated as full 4K pages, even for processes that do not need it. Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h index a5e6562..d29ad95 100644 --- a/arch/s390/include/asm/mmu.h +++ b/arch/s390/include/asm/mmu.h @@ -14,7 +14,9 @@ typedef struct { unsigned long asce_bits; unsigned long asce_limit; unsigned long vdso_base; - /* The mmu context has extended page tables. */ + /* The mmu context allocates 4K page tables. */ + unsigned int alloc_pgste:1; + /* The mmu context uses extended page tables. */ unsigned int has_pgste:1; /* The mmu context uses storage keys. */ unsigned int use_skey:1; diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index d25d9ff..fb1b93e 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -20,8 +20,11 @@ static inline int init_new_context(struct task_struct *tsk, mm->context.flush_mm = 0; mm->context.asce_bits = _ASCE_TABLE_LENGTH | _ASCE_USER_BITS; mm->context.asce_bits |= _ASCE_TYPE_REGION3; +#ifdef CONFIG_PGSTE + mm->context.alloc_pgste = page_table_allocate_pgste; mm->context.has_pgste = 0; mm->context.use_skey = 0; +#endif mm->context.asce_limit = STACK_TOP_MAX; crst_table_init((unsigned long *) mm->pgd, pgd_entry_type(mm)); return 0; diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h index 51e7fb6..7b7858f 100644 --- a/arch/s390/include/asm/pgalloc.h +++ b/arch/s390/include/asm/pgalloc.h @@ -21,6 +21,7 @@ void crst_table_free(struct mm_struct *, unsigned long *); unsigned long *page_table_alloc(struct mm_struct *); void page_table_free(struct mm_struct *, unsigned long *); void page_table_free_rcu(struct mmu_gather *, unsigned long *, unsigned long); +extern int page_table_allocate_pgste; int set_guest_storage_key(struct mm_struct *mm, unsigned long addr, unsigned long key, bool nq); diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 989cfae..1fba639 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -423,6 +423,15 @@ static inline int mm_has_pgste(struct mm_struct *mm) return 0; } +static inline int mm_alloc_pgste(struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + if (unlikely(mm->context.alloc_pgste)) + return 1; +#endif + return 0; +} + /* * In the case that a guest uses storage keys * faults should no longer be backed by zero pages diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index 33f5894..b33f661 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -920,6 +921,40 @@ unsigned long get_guest_storage_key(struct mm_struct *mm, unsigned long addr) } EXPORT_SYMBOL(get_guest_storage_key); +static int page_table_allocate_pgste_min = 0; +static int page_table_allocate_pgste_max = 1; +int page_table_allocate_pgste = 0; +EXPORT_SYMBOL(page_table_allocate_pgste); + +static struct ctl_table page_table_sysctl[] = { + { + .procname = "allocate_pgste", + .data = &page_table_allocate_pgste, + .maxlen = sizeof(int), + .mode = S_IRUGO | S_IWUSR, + .proc_handler = proc_dointvec, + .extra1 = &page_table_allocate_pgste_min, + .extra2 = &page_table_allocate_pgste_max, + }, + { } +}; + +static struct ctl_table page_table_sysctl_dir[] = { + { + .procname = "vm", + .maxlen = 0, + .mode = 0555, + .child = page_table_sysctl, + }, + { } +}; + +static int __init page_table_register_sysctl(void) +{ + return register_sysctl_table(page_table_sysctl_dir) ? 0 : -ENOMEM; +} +__initcall(page_table_register_sysctl); + #else /* CONFIG_PGSTE */ static inline int page_table_with_pgste(struct page *page) @@ -963,7 +998,7 @@ unsigned long *page_table_alloc(struct mm_struct *mm) struct page *uninitialized_var(page); unsigned int mask, bit; - if (mm_has_pgste(mm)) + if (mm_alloc_pgste(mm)) return page_table_alloc_pgste(mm); /* Allocate fragments of a 4K page as 1K/2K page table */ spin_lock_bh(&mm->context.list_lock); @@ -1165,116 +1200,25 @@ static inline void thp_split_mm(struct mm_struct *mm) } #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ -static unsigned long page_table_realloc_pmd(struct mmu_gather *tlb, - struct mm_struct *mm, pud_t *pud, - unsigned long addr, unsigned long end) -{ - unsigned long next, *table, *new; - struct page *page; - spinlock_t *ptl; - pmd_t *pmd; - - pmd = pmd_offset(pud, addr); - do { - next = pmd_addr_end(addr, end); -again: - if (pmd_none_or_clear_bad(pmd)) - continue; - table = (unsigned long *) pmd_deref(*pmd); - page = pfn_to_page(__pa(table) >> PAGE_SHIFT); - if (page_table_with_pgste(page)) - continue; - /* Allocate new page table with pgstes */ - new = page_table_alloc_pgste(mm); - if (!new) - return -ENOMEM; - - ptl = pmd_lock(mm, pmd); - if (likely((unsigned long *) pmd_deref(*pmd) == table)) { - /* Nuke pmd entry pointing to the "short" page table */ - pmdp_flush_lazy(mm, addr, pmd); - pmd_clear(pmd); - /* Copy ptes from old table to new table */ - memcpy(new, table, PAGE_SIZE/2); - clear_table(table, _PAGE_INVALID, PAGE_SIZE/2); - /* Establish new table */ - pmd_populate(mm, pmd, (pte_t *) new); - /* Free old table with rcu, there might be a walker! */ - page_table_free_rcu(tlb, table, addr); - new = NULL; - } - spin_unlock(ptl); - if (new) { - page_table_free_pgste(new); - goto again; - } - } while (pmd++, addr = next, addr != end); - - return addr; -} - -static unsigned long page_table_realloc_pud(struct mmu_gather *tlb, - struct mm_struct *mm, pgd_t *pgd, - unsigned long addr, unsigned long end) -{ - unsigned long next; - pud_t *pud; - - pud = pud_offset(pgd, addr); - do { - next = pud_addr_end(addr, end); - if (pud_none_or_clear_bad(pud)) - continue; - next = page_table_realloc_pmd(tlb, mm, pud, addr, next); - if (unlikely(IS_ERR_VALUE(next))) - return next; - } while (pud++, addr = next, addr != end); - - return addr; -} - -static unsigned long page_table_realloc(struct mmu_gather *tlb, struct mm_struct *mm, - unsigned long addr, unsigned long end) -{ - unsigned long next; - pgd_t *pgd; - - pgd = pgd_offset(mm, addr); - do { - next = pgd_addr_end(addr, end); - if (pgd_none_or_clear_bad(pgd)) - continue; - next = page_table_realloc_pud(tlb, mm, pgd, addr, next); - if (unlikely(IS_ERR_VALUE(next))) - return next; - } while (pgd++, addr = next, addr != end); - - return 0; -} - /* * switch on pgstes for its userspace process (for kvm) */ int s390_enable_sie(void) { - struct task_struct *tsk = current; - struct mm_struct *mm = tsk->mm; - struct mmu_gather tlb; + struct mm_struct *mm = current->mm; /* Do we have pgstes? if yes, we are done */ - if (mm_has_pgste(tsk->mm)) + if (mm_has_pgste(mm)) return 0; - + /* Fail if the page tables are 2K */ + if (!mm_alloc_pgste(mm)) + return -EINVAL; down_write(&mm->mmap_sem); + mm->context.has_pgste = 1; /* split thp mappings and disable thp for future mappings */ thp_split_mm(mm); - /* Reallocate the page tables with pgstes */ - tlb_gather_mmu(&tlb, mm, 0, TASK_SIZE); - if (!page_table_realloc(&tlb, mm, 0, TASK_SIZE)) - mm->context.has_pgste = 1; - tlb_finish_mmu(&tlb, 0, TASK_SIZE); up_write(&mm->mmap_sem); - return mm->context.has_pgste ? 0 : -ENOMEM; + return 0; } EXPORT_SYMBOL_GPL(s390_enable_sie); -- cgit v0.10.2 From 0a7c501e6759db49d9dffb10ed62142d705e3f90 Mon Sep 17 00:00:00 2001 From: Christophe Jaillet Date: Mon, 20 Apr 2015 10:26:52 +0200 Subject: s390/3215: free memory in error path If one memory allocation fails, there is a memory leak. Signed-off-by: Christophe Jaillet Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/char/con3215.c b/drivers/s390/char/con3215.c index c43aca6..0fc3fe5 100644 --- a/drivers/s390/char/con3215.c +++ b/drivers/s390/char/con3215.c @@ -667,6 +667,8 @@ static struct raw3215_info *raw3215_alloc_info(void) info->buffer = kzalloc(RAW3215_BUFFER_SIZE, GFP_KERNEL | GFP_DMA); info->inbuf = kzalloc(RAW3215_INBUF_SIZE, GFP_KERNEL | GFP_DMA); if (!info->buffer || !info->inbuf) { + kfree(info->inbuf); + kfree(info->buffer); kfree(info); return NULL; } -- cgit v0.10.2 From 77a87f0cb1a57237860754525d4e8cb2789e6e12 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Tue, 21 Apr 2015 16:50:08 +0200 Subject: s390/bpf: add dependency to z196 features The new ebpf code uses e.g. the laal instruction which is part of the interlocked-access facility 1 which again was introduced with z196. So we must make sure the ebpf code generator depends on MARCH_Z196_FEATURES. Signed-off-by: Heiko Carstens diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 8e58c61..b06dc38 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -115,7 +115,7 @@ config S390 select HAVE_ARCH_SECCOMP_FILTER select HAVE_ARCH_TRACEHOOK select HAVE_ARCH_TRANSPARENT_HUGEPAGE - select HAVE_BPF_JIT if PACK_STACK && HAVE_MARCH_Z9_109_FEATURES + select HAVE_BPF_JIT if PACK_STACK && HAVE_MARCH_Z196_FEATURES select HAVE_CMPXCHG_DOUBLE select HAVE_CMPXCHG_LOCAL select HAVE_DEBUG_KMEMLEAK -- cgit v0.10.2 From b7d14f3a92223c3f5e52e9f20c74cb96dc130e87 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 22 Apr 2015 10:26:20 +0200 Subject: s390/mm: correct transfer of dirty & young bits in __pmd_to_pte The dirty & young bit from the pmd is not copied correctly to the pseudo pte in __pmd_to_pte. In fact it is not copied at all, the bits get lost. As the old style huge page currently does not need the dirty & young information this has no effect, but may be needed in the future. Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index 210ffed..c67e8bf 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -70,8 +70,8 @@ static inline pte_t __pmd_to_pte(pmd_t pmd) pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_WRITE) << 4; pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_INVALID) << 5; pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_PROTECT); - pmd_val(pmd) |= (pte_val(pte) & _PAGE_DIRTY) << 10; - pmd_val(pmd) |= (pte_val(pte) & _PAGE_YOUNG) << 10; + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY) >> 10; + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_YOUNG) >> 10; } else pte_val(pte) = _PAGE_INVALID; return pte; -- cgit v0.10.2 From a1c843b82541fdd4c4644607c942dabc7c7e6f6c Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 22 Apr 2015 13:55:59 +0200 Subject: s390/mm: change swap pte encoding and pgtable cleanup After the file ptes have been removed the bit combination used to encode non-linear mappings can be reused for the swap ptes. This frees up a precious pte software bit. Reflect the change in the swap encoding in the comments and do some cleanup while we are at it. Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 1fba639..fc64239 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -12,12 +12,9 @@ #define _ASM_S390_PGTABLE_H /* - * The Linux memory management assumes a three-level page table setup. For - * s390 31 bit we "fold" the mid level into the top-level page table, so - * that we physically have the same two-level page table as the s390 mmu - * expects in 31 bit mode. For s390 64 bit we use three of the five levels - * the hardware provides (region first and region second tables are not - * used). + * The Linux memory management assumes a three-level page table setup. + * For s390 64 bit we use up to four of the five levels the hardware + * provides (region first tables are not used). * * The "pgd_xxx()" functions are trivial for a folded two-level * setup: the pgd is never bad, and a pmd always exists (as it's folded @@ -101,8 +98,8 @@ extern unsigned long zero_page_mask; #ifndef __ASSEMBLY__ /* - * The vmalloc and module area will always be on the topmost area of the kernel - * mapping. We reserve 96MB (31bit) / 128GB (64bit) for vmalloc and modules. + * The vmalloc and module area will always be on the topmost area of the + * kernel mapping. We reserve 128GB (64bit) for vmalloc and modules. * On 64 bit kernels we have a 2GB area at the top of the vmalloc area where * modules will reside. That makes sure that inter module branches always * happen without trampolines and in addition the placement within a 2GB frame @@ -131,38 +128,6 @@ static inline int is_module_addr(void *addr) } /* - * A 31 bit pagetable entry of S390 has following format: - * | PFRA | | OS | - * 0 0IP0 - * 00000000001111111111222222222233 - * 01234567890123456789012345678901 - * - * I Page-Invalid Bit: Page is not available for address-translation - * P Page-Protection Bit: Store access not possible for page - * - * A 31 bit segmenttable entry of S390 has following format: - * | P-table origin | |PTL - * 0 IC - * 00000000001111111111222222222233 - * 01234567890123456789012345678901 - * - * I Segment-Invalid Bit: Segment is not available for address-translation - * C Common-Segment Bit: Segment is not private (PoP 3-30) - * PTL Page-Table-Length: Page-table length (PTL+1*16 entries -> up to 256) - * - * The 31 bit segmenttable origin of S390 has following format: - * - * |S-table origin | | STL | - * X **GPS - * 00000000001111111111222222222233 - * 01234567890123456789012345678901 - * - * X Space-Switch event: - * G Segment-Invalid Bit: * - * P Private-Space Bit: Segment is not private (PoP 3-30) - * S Storage-Alteration: - * STL Segment-Table-Length: Segment-table length (STL+1*16 entries -> up to 2048) - * * A 64 bit pagetable entry of S390 has following format: * | PFRA |0IPC| OS | * 0000000000111111111122222222223333333333444444444455555555556666 @@ -220,7 +185,6 @@ static inline int is_module_addr(void *addr) /* Software bits in the page table entry */ #define _PAGE_PRESENT 0x001 /* SW pte present bit */ -#define _PAGE_TYPE 0x002 /* SW pte type bit */ #define _PAGE_YOUNG 0x004 /* SW pte young bit */ #define _PAGE_DIRTY 0x008 /* SW pte dirty bit */ #define _PAGE_READ 0x010 /* SW pte read bit */ @@ -240,31 +204,34 @@ static inline int is_module_addr(void *addr) * table lock held. * * The following table gives the different possible bit combinations for - * the pte hardware and software bits in the last 12 bits of a pte: + * the pte hardware and software bits in the last 12 bits of a pte + * (. unassigned bit, x don't care, t swap type): * * 842100000000 * 000084210000 * 000000008421 - * .IR...wrdytp - * empty .10...000000 - * swap .10...xxxx10 - * file .11...xxxxx0 - * prot-none, clean, old .11...000001 - * prot-none, clean, young .11...000101 - * prot-none, dirty, old .10...001001 - * prot-none, dirty, young .10...001101 - * read-only, clean, old .11...010001 - * read-only, clean, young .01...010101 - * read-only, dirty, old .11...011001 - * read-only, dirty, young .01...011101 - * read-write, clean, old .11...110001 - * read-write, clean, young .01...110101 - * read-write, dirty, old .10...111001 - * read-write, dirty, young .00...111101 + * .IR.uswrdy.p + * empty .10.00000000 + * swap .11..ttttt.0 + * prot-none, clean, old .11.xx0000.1 + * prot-none, clean, young .11.xx0001.1 + * prot-none, dirty, old .10.xx0010.1 + * prot-none, dirty, young .10.xx0011.1 + * read-only, clean, old .11.xx0100.1 + * read-only, clean, young .01.xx0101.1 + * read-only, dirty, old .11.xx0110.1 + * read-only, dirty, young .01.xx0111.1 + * read-write, clean, old .11.xx1100.1 + * read-write, clean, young .01.xx1101.1 + * read-write, dirty, old .10.xx1110.1 + * read-write, dirty, young .00.xx1111.1 + * HW-bits: R read-only, I invalid + * SW-bits: p present, y young, d dirty, r read, w write, s special, + * u unused, l large * - * pte_present is true for the bit pattern .xx...xxxxx1, (pte & 0x001) == 0x001 - * pte_none is true for the bit pattern .10...xxxx00, (pte & 0x603) == 0x400 - * pte_swap is true for the bit pattern .10...xxxx10, (pte & 0x603) == 0x402 + * pte_none is true for the bit pattern .10.00000000, pte == 0x400 + * pte_swap is true for the bit pattern .11..ooooo.0, (pte & 0x201) == 0x200 + * pte_present is true for the bit pattern .xx.xxxxxx.1, (pte & 0x001) == 0x001 */ /* Bits in the segment/region table address-space-control-element */ @@ -335,6 +302,8 @@ static inline int is_module_addr(void *addr) * read-write, dirty, young 11..0...0...11 * The segment table origin is used to distinguish empty (origin==0) from * read-write, old segment table entries (origin!=0) + * HW-bits: R read-only, I invalid + * SW-bits: y young, d dirty, r read, w write */ #define _SEGMENT_ENTRY_SPLIT_BIT 11 /* THP splitting bit number */ @@ -591,10 +560,9 @@ static inline int pte_none(pte_t pte) static inline int pte_swap(pte_t pte) { - /* Bit pattern: (pte & 0x603) == 0x402 */ - return (pte_val(pte) & (_PAGE_INVALID | _PAGE_PROTECT | - _PAGE_TYPE | _PAGE_PRESENT)) - == (_PAGE_INVALID | _PAGE_TYPE); + /* Bit pattern: (pte & 0x201) == 0x200 */ + return (pte_val(pte) & (_PAGE_PROTECT | _PAGE_PRESENT)) + == _PAGE_PROTECT; } static inline int pte_special(pte_t pte) @@ -1595,51 +1563,51 @@ static inline int has_transparent_hugepage(void) #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ /* - * 31 bit swap entry format: - * A page-table entry has some bits we have to treat in a special way. - * Bits 0, 20 and bit 23 have to be zero, otherwise an specification - * exception will occur instead of a page translation exception. The - * specifiation exception has the bad habit not to store necessary - * information in the lowcore. - * Bits 21, 22, 30 and 31 are used to indicate the page type. - * A swap pte is indicated by bit pattern (pte & 0x603) == 0x402 - * This leaves the bits 1-19 and bits 24-29 to store type and offset. - * We use the 5 bits from 25-29 for the type and the 20 bits from 1-19 - * plus 24 for the offset. - * 0| offset |0110|o|type |00| - * 0 0000000001111111111 2222 2 22222 33 - * 0 1234567890123456789 0123 4 56789 01 - * * 64 bit swap entry format: * A page-table entry has some bits we have to treat in a special way. * Bits 52 and bit 55 have to be zero, otherwise an specification * exception will occur instead of a page translation exception. The * specifiation exception has the bad habit not to store necessary * information in the lowcore. - * Bits 53, 54, 62 and 63 are used to indicate the page type. - * A swap pte is indicated by bit pattern (pte & 0x603) == 0x402 - * This leaves the bits 0-51 and bits 56-61 to store type and offset. - * We use the 5 bits from 57-61 for the type and the 53 bits from 0-51 - * plus 56 for the offset. - * | offset |0110|o|type |00| - * 0000000000111111111122222222223333333333444444444455 5555 5 55566 66 - * 0123456789012345678901234567890123456789012345678901 2345 6 78901 23 + * Bits 54 and 63 are used to indicate the page type. + * A swap pte is indicated by bit pattern (pte & 0x201) == 0x200 + * This leaves the bits 0-51 and bits 56-62 to store type and offset. + * We use the 5 bits from 57-61 for the type and the 52 bits from 0-51 + * for the offset. + * | offset |01100|type |00| + * |0000000000111111111122222222223333333333444444444455|55555|55566|66| + * |0123456789012345678901234567890123456789012345678901|23456|78901|23| */ -#define __SWP_OFFSET_MASK (~0UL >> 11) +#define __SWP_OFFSET_MASK ((1UL << 52) - 1) +#define __SWP_OFFSET_SHIFT 12 +#define __SWP_TYPE_MASK ((1UL << 5) - 1) +#define __SWP_TYPE_SHIFT 2 static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset) { pte_t pte; - offset &= __SWP_OFFSET_MASK; - pte_val(pte) = _PAGE_INVALID | _PAGE_TYPE | ((type & 0x1f) << 2) | - ((offset & 1UL) << 7) | ((offset & ~1UL) << 11); + + pte_val(pte) = _PAGE_INVALID | _PAGE_PROTECT; + pte_val(pte) |= (offset & __SWP_OFFSET_MASK) << __SWP_OFFSET_SHIFT; + pte_val(pte) |= (type & __SWP_TYPE_MASK) << __SWP_TYPE_SHIFT; return pte; } -#define __swp_type(entry) (((entry).val >> 2) & 0x1f) -#define __swp_offset(entry) (((entry).val >> 11) | (((entry).val >> 7) & 1)) -#define __swp_entry(type,offset) ((swp_entry_t) { pte_val(mk_swap_pte((type),(offset))) }) +static inline unsigned long __swp_type(swp_entry_t entry) +{ + return (entry.val >> __SWP_TYPE_SHIFT) & __SWP_TYPE_MASK; +} + +static inline unsigned long __swp_offset(swp_entry_t entry) +{ + return (entry.val >> __SWP_OFFSET_SHIFT) & __SWP_OFFSET_MASK; +} + +static inline swp_entry_t __swp_entry(unsigned long type, unsigned long offset) +{ + return (swp_entry_t) { pte_val(mk_swap_pte(type, offset)) }; +} #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index c67e8bf..e617e74 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -14,20 +14,23 @@ static inline pmd_t __pte_to_pmd(pte_t pte) /* * Convert encoding pte bits pmd bits - * .IR...wrdytp dy..R...I...wr - * empty .10...000000 -> 00..0...1...00 - * prot-none, clean, old .11...000001 -> 00..1...1...00 - * prot-none, clean, young .11...000101 -> 01..1...1...00 - * prot-none, dirty, old .10...001001 -> 10..1...1...00 - * prot-none, dirty, young .10...001101 -> 11..1...1...00 - * read-only, clean, old .11...010001 -> 00..1...1...01 - * read-only, clean, young .01...010101 -> 01..1...0...01 - * read-only, dirty, old .11...011001 -> 10..1...1...01 - * read-only, dirty, young .01...011101 -> 11..1...0...01 - * read-write, clean, old .11...110001 -> 00..0...1...11 - * read-write, clean, young .01...110101 -> 01..0...0...11 - * read-write, dirty, old .10...111001 -> 10..0...1...11 - * read-write, dirty, young .00...111101 -> 11..0...0...11 + * lIR.uswrdy.p dy..R...I...wr + * empty 010.000000.0 -> 00..0...1...00 + * prot-none, clean, old 111.000000.1 -> 00..1...1...00 + * prot-none, clean, young 111.000001.1 -> 01..1...1...00 + * prot-none, dirty, old 111.000010.1 -> 10..1...1...00 + * prot-none, dirty, young 111.000011.1 -> 11..1...1...00 + * read-only, clean, old 111.000100.1 -> 00..1...1...01 + * read-only, clean, young 101.000101.1 -> 01..1...0...01 + * read-only, dirty, old 111.000110.1 -> 10..1...1...01 + * read-only, dirty, young 101.000111.1 -> 11..1...0...01 + * read-write, clean, old 111.001100.1 -> 00..1...1...11 + * read-write, clean, young 101.001101.1 -> 01..1...0...11 + * read-write, dirty, old 110.001110.1 -> 10..0...1...11 + * read-write, dirty, young 100.001111.1 -> 11..0...0...11 + * HW-bits: R read-only, I invalid + * SW-bits: p present, y young, d dirty, r read, w write, s special, + * u unused, l large */ if (pte_present(pte)) { pmd_val(pmd) = pte_val(pte) & PAGE_MASK; @@ -48,20 +51,23 @@ static inline pte_t __pmd_to_pte(pmd_t pmd) /* * Convert encoding pmd bits pte bits - * dy..R...I...wr .IR...wrdytp - * empty 00..0...1...00 -> .10...001100 - * prot-none, clean, old 00..0...1...00 -> .10...000001 - * prot-none, clean, young 01..0...1...00 -> .10...000101 - * prot-none, dirty, old 10..0...1...00 -> .10...001001 - * prot-none, dirty, young 11..0...1...00 -> .10...001101 - * read-only, clean, old 00..1...1...01 -> .11...010001 - * read-only, clean, young 01..1...1...01 -> .11...010101 - * read-only, dirty, old 10..1...1...01 -> .11...011001 - * read-only, dirty, young 11..1...1...01 -> .11...011101 - * read-write, clean, old 00..0...1...11 -> .10...110001 - * read-write, clean, young 01..0...1...11 -> .10...110101 - * read-write, dirty, old 10..0...1...11 -> .10...111001 - * read-write, dirty, young 11..0...1...11 -> .10...111101 + * dy..R...I...wr lIR.uswrdy.p + * empty 00..0...1...00 -> 010.000000.0 + * prot-none, clean, old 00..1...1...00 -> 111.000000.1 + * prot-none, clean, young 01..1...1...00 -> 111.000001.1 + * prot-none, dirty, old 10..1...1...00 -> 111.000010.1 + * prot-none, dirty, young 11..1...1...00 -> 111.000011.1 + * read-only, clean, old 00..1...1...01 -> 111.000100.1 + * read-only, clean, young 01..1...0...01 -> 101.000101.1 + * read-only, dirty, old 10..1...1...01 -> 111.000110.1 + * read-only, dirty, young 11..1...0...01 -> 101.000111.1 + * read-write, clean, old 00..1...1...11 -> 111.001100.1 + * read-write, clean, young 01..1...0...11 -> 101.001101.1 + * read-write, dirty, old 10..0...1...11 -> 110.001110.1 + * read-write, dirty, young 11..0...0...11 -> 100.001111.1 + * HW-bits: R read-only, I invalid + * SW-bits: p present, y young, d dirty, r read, w write, s special, + * u unused, l large */ if (pmd_present(pmd)) { pte_val(pte) = pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN_LARGE; -- cgit v0.10.2 From 57127645d79d2e83e801f141f7d03f64accf28aa Mon Sep 17 00:00:00 2001 From: Harald Freudenberger Date: Mon, 16 Mar 2015 14:52:52 +0100 Subject: s390/zcrypt: Introduce new SHA-512 based Pseudo Random Generator. Rework of the prandom device with introduction of a new SHA-512 based NIST SP 800-90 conform deterministic random bit generator. Signed-off-by: Harald Freudenberger Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/crypto/crypt_s390.h b/arch/s390/crypto/crypt_s390.h index ba3b2ae..d9c4c31 100644 --- a/arch/s390/crypto/crypt_s390.h +++ b/arch/s390/crypto/crypt_s390.h @@ -3,9 +3,10 @@ * * Support for s390 cryptographic instructions. * - * Copyright IBM Corp. 2003, 2007 + * Copyright IBM Corp. 2003, 2015 * Author(s): Thomas Spatzier * Jan Glauber (jan.glauber@de.ibm.com) + * Harald Freudenberger (freude@de.ibm.com) * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the Free @@ -28,15 +29,17 @@ #define CRYPT_S390_MSA 0x1 #define CRYPT_S390_MSA3 0x2 #define CRYPT_S390_MSA4 0x4 +#define CRYPT_S390_MSA5 0x8 /* s390 cryptographic operations */ enum crypt_s390_operations { - CRYPT_S390_KM = 0x0100, - CRYPT_S390_KMC = 0x0200, - CRYPT_S390_KIMD = 0x0300, - CRYPT_S390_KLMD = 0x0400, - CRYPT_S390_KMAC = 0x0500, - CRYPT_S390_KMCTR = 0x0600 + CRYPT_S390_KM = 0x0100, + CRYPT_S390_KMC = 0x0200, + CRYPT_S390_KIMD = 0x0300, + CRYPT_S390_KLMD = 0x0400, + CRYPT_S390_KMAC = 0x0500, + CRYPT_S390_KMCTR = 0x0600, + CRYPT_S390_PPNO = 0x0700 }; /* @@ -138,6 +141,16 @@ enum crypt_s390_kmac_func { KMAC_TDEA_192 = CRYPT_S390_KMAC | 3 }; +/* + * function codes for PPNO (PERFORM PSEUDORANDOM NUMBER + * OPERATION) instruction + */ +enum crypt_s390_ppno_func { + PPNO_QUERY = CRYPT_S390_PPNO | 0, + PPNO_SHA512_DRNG_GEN = CRYPT_S390_PPNO | 3, + PPNO_SHA512_DRNG_SEED = CRYPT_S390_PPNO | 0x83 +}; + /** * crypt_s390_km: * @func: the function code passed to KM; see crypt_s390_km_func @@ -162,11 +175,11 @@ static inline int crypt_s390_km(long func, void *param, int ret; asm volatile( - "0: .insn rre,0xb92e0000,%3,%1 \n" /* KM opcode */ - "1: brc 1,0b \n" /* handle partial completion */ + "0: .insn rre,0xb92e0000,%3,%1\n" /* KM opcode */ + "1: brc 1,0b\n" /* handle partial completion */ " la %0,0\n" "2:\n" - EX_TABLE(0b,2b) EX_TABLE(1b,2b) + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) : "=d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest) : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); if (ret < 0) @@ -198,11 +211,11 @@ static inline int crypt_s390_kmc(long func, void *param, int ret; asm volatile( - "0: .insn rre,0xb92f0000,%3,%1 \n" /* KMC opcode */ - "1: brc 1,0b \n" /* handle partial completion */ + "0: .insn rre,0xb92f0000,%3,%1\n" /* KMC opcode */ + "1: brc 1,0b\n" /* handle partial completion */ " la %0,0\n" "2:\n" - EX_TABLE(0b,2b) EX_TABLE(1b,2b) + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) : "=d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest) : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); if (ret < 0) @@ -233,11 +246,11 @@ static inline int crypt_s390_kimd(long func, void *param, int ret; asm volatile( - "0: .insn rre,0xb93e0000,%1,%1 \n" /* KIMD opcode */ - "1: brc 1,0b \n" /* handle partial completion */ + "0: .insn rre,0xb93e0000,%1,%1\n" /* KIMD opcode */ + "1: brc 1,0b\n" /* handle partial completion */ " la %0,0\n" "2:\n" - EX_TABLE(0b,2b) EX_TABLE(1b,2b) + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) : "=d" (ret), "+a" (__src), "+d" (__src_len) : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); if (ret < 0) @@ -267,11 +280,11 @@ static inline int crypt_s390_klmd(long func, void *param, int ret; asm volatile( - "0: .insn rre,0xb93f0000,%1,%1 \n" /* KLMD opcode */ - "1: brc 1,0b \n" /* handle partial completion */ + "0: .insn rre,0xb93f0000,%1,%1\n" /* KLMD opcode */ + "1: brc 1,0b\n" /* handle partial completion */ " la %0,0\n" "2:\n" - EX_TABLE(0b,2b) EX_TABLE(1b,2b) + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) : "=d" (ret), "+a" (__src), "+d" (__src_len) : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); if (ret < 0) @@ -302,11 +315,11 @@ static inline int crypt_s390_kmac(long func, void *param, int ret; asm volatile( - "0: .insn rre,0xb91e0000,%1,%1 \n" /* KLAC opcode */ - "1: brc 1,0b \n" /* handle partial completion */ + "0: .insn rre,0xb91e0000,%1,%1\n" /* KLAC opcode */ + "1: brc 1,0b\n" /* handle partial completion */ " la %0,0\n" "2:\n" - EX_TABLE(0b,2b) EX_TABLE(1b,2b) + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) : "=d" (ret), "+a" (__src), "+d" (__src_len) : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); if (ret < 0) @@ -340,11 +353,11 @@ static inline int crypt_s390_kmctr(long func, void *param, u8 *dest, int ret = -1; asm volatile( - "0: .insn rrf,0xb92d0000,%3,%1,%4,0 \n" /* KMCTR opcode */ - "1: brc 1,0b \n" /* handle partial completion */ + "0: .insn rrf,0xb92d0000,%3,%1,%4,0\n" /* KMCTR opcode */ + "1: brc 1,0b\n" /* handle partial completion */ " la %0,0\n" "2:\n" - EX_TABLE(0b,2b) EX_TABLE(1b,2b) + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) : "+d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest), "+a" (__ctr) : "d" (__func), "a" (__param) : "cc", "memory"); @@ -354,6 +367,47 @@ static inline int crypt_s390_kmctr(long func, void *param, u8 *dest, } /** + * crypt_s390_ppno: + * @func: the function code passed to PPNO; see crypt_s390_ppno_func + * @param: address of parameter block; see POP for details on each func + * @dest: address of destination memory area + * @dest_len: size of destination memory area in bytes + * @seed: address of seed data + * @seed_len: size of seed data in bytes + * + * Executes the PPNO (PERFORM PSEUDORANDOM NUMBER OPERATION) + * operation of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of random + * bytes stored in dest buffer for generate function + */ +static inline int crypt_s390_ppno(long func, void *param, + u8 *dest, long dest_len, + const u8 *seed, long seed_len) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; /* param block (240 bytes) */ + register u8 *__dest asm("2") = dest; /* buf for recv random bytes */ + register long __dest_len asm("3") = dest_len; /* requested random bytes */ + register const u8 *__seed asm("4") = seed; /* buf with seed data */ + register long __seed_len asm("5") = seed_len; /* bytes in seed buf */ + int ret = -1; + + asm volatile ( + "0: .insn rre,0xb93c0000,%1,%5\n" /* PPNO opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "+d" (ret), "+a"(__dest), "+d"(__dest_len) + : "d"(__func), "a"(__param), "a"(__seed), "d"(__seed_len) + : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? dest_len - __dest_len : 0; +} + +/** * crypt_s390_func_available: * @func: the function code of the specific function; 0 if op in general * @@ -373,6 +427,9 @@ static inline int crypt_s390_func_available(int func, return 0; if (facility_mask & CRYPT_S390_MSA4 && !test_facility(77)) return 0; + if (facility_mask & CRYPT_S390_MSA5 && !test_facility(57)) + return 0; + switch (func & CRYPT_S390_OP_MASK) { case CRYPT_S390_KM: ret = crypt_s390_km(KM_QUERY, &status, NULL, NULL, 0); @@ -390,8 +447,12 @@ static inline int crypt_s390_func_available(int func, ret = crypt_s390_kmac(KMAC_QUERY, &status, NULL, 0); break; case CRYPT_S390_KMCTR: - ret = crypt_s390_kmctr(KMCTR_QUERY, &status, NULL, NULL, 0, - NULL); + ret = crypt_s390_kmctr(KMCTR_QUERY, &status, + NULL, NULL, 0, NULL); + break; + case CRYPT_S390_PPNO: + ret = crypt_s390_ppno(PPNO_QUERY, &status, + NULL, 0, NULL, 0); break; default: return 0; @@ -419,15 +480,14 @@ static inline int crypt_s390_pcc(long func, void *param) int ret = -1; asm volatile( - "0: .insn rre,0xb92c0000,0,0 \n" /* PCC opcode */ - "1: brc 1,0b \n" /* handle partial completion */ + "0: .insn rre,0xb92c0000,0,0\n" /* PCC opcode */ + "1: brc 1,0b\n" /* handle partial completion */ " la %0,0\n" "2:\n" - EX_TABLE(0b,2b) EX_TABLE(1b,2b) + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) : "+d" (ret) : "d" (__func), "a" (__param) : "cc", "memory"); return ret; } - #endif /* _CRYPTO_ARCH_S390_CRYPT_S390_H */ diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c index 94a35a4..1f374b3 100644 --- a/arch/s390/crypto/prng.c +++ b/arch/s390/crypto/prng.c @@ -1,106 +1,529 @@ /* - * Copyright IBM Corp. 2006, 2007 + * Copyright IBM Corp. 2006, 2015 * Author(s): Jan Glauber + * Harald Freudenberger * Driver for the s390 pseudo random number generator */ + +#define KMSG_COMPONENT "prng" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include +#include #include #include +#include #include #include #include +#include #include #include #include #include +#include #include "crypt_s390.h" MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Jan Glauber "); +MODULE_AUTHOR("IBM Corporation"); MODULE_DESCRIPTION("s390 PRNG interface"); -static int prng_chunk_size = 256; -module_param(prng_chunk_size, int, S_IRUSR | S_IRGRP | S_IROTH); + +#define PRNG_MODE_AUTO 0 +#define PRNG_MODE_TDES 1 +#define PRNG_MODE_SHA512 2 + +static unsigned int prng_mode = PRNG_MODE_AUTO; +module_param_named(mode, prng_mode, int, 0); +MODULE_PARM_DESC(prng_mode, "PRNG mode: 0 - auto, 1 - TDES, 2 - SHA512"); + + +#define PRNG_CHUNKSIZE_TDES_MIN 8 +#define PRNG_CHUNKSIZE_TDES_MAX (64*1024) +#define PRNG_CHUNKSIZE_SHA512_MIN 64 +#define PRNG_CHUNKSIZE_SHA512_MAX (64*1024) + +static unsigned int prng_chunk_size = 256; +module_param_named(chunksize, prng_chunk_size, int, 0); MODULE_PARM_DESC(prng_chunk_size, "PRNG read chunk size in bytes"); -static int prng_entropy_limit = 4096; -module_param(prng_entropy_limit, int, S_IRUSR | S_IRGRP | S_IROTH | S_IWUSR); -MODULE_PARM_DESC(prng_entropy_limit, - "PRNG add entropy after that much bytes were produced"); + +#define PRNG_RESEED_LIMIT_TDES 4096 +#define PRNG_RESEED_LIMIT_TDES_LOWER 4096 +#define PRNG_RESEED_LIMIT_SHA512 100000 +#define PRNG_RESEED_LIMIT_SHA512_LOWER 10000 + +static unsigned int prng_reseed_limit; +module_param_named(reseed_limit, prng_reseed_limit, int, 0); +MODULE_PARM_DESC(prng_reseed_limit, "PRNG reseed limit"); + /* * Any one who considers arithmetical methods of producing random digits is, * of course, in a state of sin. -- John von Neumann */ -struct s390_prng_data { - unsigned long count; /* how many bytes were produced */ - char *buf; +static int prng_errorflag; + +#define PRNG_GEN_ENTROPY_FAILED 1 +#define PRNG_SELFTEST_FAILED 2 +#define PRNG_INSTANTIATE_FAILED 3 +#define PRNG_SEED_FAILED 4 +#define PRNG_RESEED_FAILED 5 +#define PRNG_GEN_FAILED 6 + +struct prng_ws_s { + u8 parm_block[32]; + u32 reseed_counter; + u64 byte_counter; }; -static struct s390_prng_data *p; +struct ppno_ws_s { + u32 res; + u32 reseed_counter; + u64 stream_bytes; + u8 V[112]; + u8 C[112]; +}; -/* copied from libica, use a non-zero initial parameter block */ -static unsigned char parm_block[32] = { -0x0F,0x2B,0x8E,0x63,0x8C,0x8E,0xD2,0x52,0x64,0xB7,0xA0,0x7B,0x75,0x28,0xB8,0xF4, -0x75,0x5F,0xD2,0xA6,0x8D,0x97,0x11,0xFF,0x49,0xD8,0x23,0xF3,0x7E,0x21,0xEC,0xA0, +struct prng_data_s { + struct mutex mutex; + union { + struct prng_ws_s prngws; + struct ppno_ws_s ppnows; + }; + u8 *buf; + u32 rest; + u8 *prev; }; -static int prng_open(struct inode *inode, struct file *file) +static struct prng_data_s *prng_data; + +/* initial parameter block for tdes mode, copied from libica */ +static const u8 initial_parm_block[32] __initconst = { + 0x0F, 0x2B, 0x8E, 0x63, 0x8C, 0x8E, 0xD2, 0x52, + 0x64, 0xB7, 0xA0, 0x7B, 0x75, 0x28, 0xB8, 0xF4, + 0x75, 0x5F, 0xD2, 0xA6, 0x8D, 0x97, 0x11, 0xFF, + 0x49, 0xD8, 0x23, 0xF3, 0x7E, 0x21, 0xEC, 0xA0 }; + + +/*** helper functions ***/ + +static int generate_entropy(u8 *ebuf, size_t nbytes) { - return nonseekable_open(inode, file); + int n, ret = 0; + u8 *pg, *h, hash[32]; + + pg = (u8 *) __get_free_page(GFP_KERNEL); + if (!pg) { + prng_errorflag = PRNG_GEN_ENTROPY_FAILED; + return -ENOMEM; + } + + while (nbytes) { + /* fill page with urandom bytes */ + get_random_bytes(pg, PAGE_SIZE); + /* exor page with stckf values */ + for (n = 0; n < sizeof(PAGE_SIZE/sizeof(u64)); n++) { + u64 *p = ((u64 *)pg) + n; + *p ^= get_tod_clock_fast(); + } + n = (nbytes < sizeof(hash)) ? nbytes : sizeof(hash); + if (n < sizeof(hash)) + h = hash; + else + h = ebuf; + /* generate sha256 from this page */ + if (crypt_s390_kimd(KIMD_SHA_256, h, + pg, PAGE_SIZE) != PAGE_SIZE) { + prng_errorflag = PRNG_GEN_ENTROPY_FAILED; + ret = -EIO; + goto out; + } + if (n < sizeof(hash)) + memcpy(ebuf, hash, n); + ret += n; + ebuf += n; + nbytes -= n; + } + +out: + free_page((unsigned long)pg); + return ret; } -static void prng_add_entropy(void) + +/*** tdes functions ***/ + +static void prng_tdes_add_entropy(void) { __u64 entropy[4]; unsigned int i; int ret; for (i = 0; i < 16; i++) { - ret = crypt_s390_kmc(KMC_PRNG, parm_block, (char *)entropy, - (char *)entropy, sizeof(entropy)); + ret = crypt_s390_kmc(KMC_PRNG, prng_data->prngws.parm_block, + (char *)entropy, (char *)entropy, + sizeof(entropy)); BUG_ON(ret < 0 || ret != sizeof(entropy)); - memcpy(parm_block, entropy, sizeof(entropy)); + memcpy(prng_data->prngws.parm_block, entropy, sizeof(entropy)); } } -static void prng_seed(int nbytes) + +static void prng_tdes_seed(int nbytes) { char buf[16]; int i = 0; - BUG_ON(nbytes > 16); + BUG_ON(nbytes > sizeof(buf)); + get_random_bytes(buf, nbytes); /* Add the entropy */ while (nbytes >= 8) { - *((__u64 *)parm_block) ^= *((__u64 *)(buf+i)); - prng_add_entropy(); + *((__u64 *)prng_data->prngws.parm_block) ^= *((__u64 *)(buf+i)); + prng_tdes_add_entropy(); i += 8; nbytes -= 8; } - prng_add_entropy(); + prng_tdes_add_entropy(); + prng_data->prngws.reseed_counter = 0; +} + + +static int __init prng_tdes_instantiate(void) +{ + int datalen; + + pr_debug("prng runs in TDES mode with " + "chunksize=%d and reseed_limit=%u\n", + prng_chunk_size, prng_reseed_limit); + + /* memory allocation, prng_data struct init, mutex init */ + datalen = sizeof(struct prng_data_s) + prng_chunk_size; + prng_data = kzalloc(datalen, GFP_KERNEL); + if (!prng_data) { + prng_errorflag = PRNG_INSTANTIATE_FAILED; + return -ENOMEM; + } + mutex_init(&prng_data->mutex); + prng_data->buf = ((u8 *)prng_data) + sizeof(struct prng_data_s); + memcpy(prng_data->prngws.parm_block, initial_parm_block, 32); + + /* initialize the PRNG, add 128 bits of entropy */ + prng_tdes_seed(16); + + return 0; } -static ssize_t prng_read(struct file *file, char __user *ubuf, size_t nbytes, - loff_t *ppos) + +static void prng_tdes_deinstantiate(void) +{ + pr_debug("The prng module stopped " + "after running in triple DES mode\n"); + kzfree(prng_data); +} + + +/*** sha512 functions ***/ + +static int __init prng_sha512_selftest(void) { - int chunk, n; + /* NIST DRBG testvector for Hash Drbg, Sha-512, Count #0 */ + static const u8 seed[] __initconst = { + 0x6b, 0x50, 0xa7, 0xd8, 0xf8, 0xa5, 0x5d, 0x7a, + 0x3d, 0xf8, 0xbb, 0x40, 0xbc, 0xc3, 0xb7, 0x22, + 0xd8, 0x70, 0x8d, 0xe6, 0x7f, 0xda, 0x01, 0x0b, + 0x03, 0xc4, 0xc8, 0x4d, 0x72, 0x09, 0x6f, 0x8c, + 0x3e, 0xc6, 0x49, 0xcc, 0x62, 0x56, 0xd9, 0xfa, + 0x31, 0xdb, 0x7a, 0x29, 0x04, 0xaa, 0xf0, 0x25 }; + static const u8 V0[] __initconst = { + 0x00, 0xad, 0xe3, 0x6f, 0x9a, 0x01, 0xc7, 0x76, + 0x61, 0x34, 0x35, 0xf5, 0x4e, 0x24, 0x74, 0x22, + 0x21, 0x9a, 0x29, 0x89, 0xc7, 0x93, 0x2e, 0x60, + 0x1e, 0xe8, 0x14, 0x24, 0x8d, 0xd5, 0x03, 0xf1, + 0x65, 0x5d, 0x08, 0x22, 0x72, 0xd5, 0xad, 0x95, + 0xe1, 0x23, 0x1e, 0x8a, 0xa7, 0x13, 0xd9, 0x2b, + 0x5e, 0xbc, 0xbb, 0x80, 0xab, 0x8d, 0xe5, 0x79, + 0xab, 0x5b, 0x47, 0x4e, 0xdd, 0xee, 0x6b, 0x03, + 0x8f, 0x0f, 0x5c, 0x5e, 0xa9, 0x1a, 0x83, 0xdd, + 0xd3, 0x88, 0xb2, 0x75, 0x4b, 0xce, 0x83, 0x36, + 0x57, 0x4b, 0xf1, 0x5c, 0xca, 0x7e, 0x09, 0xc0, + 0xd3, 0x89, 0xc6, 0xe0, 0xda, 0xc4, 0x81, 0x7e, + 0x5b, 0xf9, 0xe1, 0x01, 0xc1, 0x92, 0x05, 0xea, + 0xf5, 0x2f, 0xc6, 0xc6, 0xc7, 0x8f, 0xbc, 0xf4 }; + static const u8 C0[] __initconst = { + 0x00, 0xf4, 0xa3, 0xe5, 0xa0, 0x72, 0x63, 0x95, + 0xc6, 0x4f, 0x48, 0xd0, 0x8b, 0x5b, 0x5f, 0x8e, + 0x6b, 0x96, 0x1f, 0x16, 0xed, 0xbc, 0x66, 0x94, + 0x45, 0x31, 0xd7, 0x47, 0x73, 0x22, 0xa5, 0x86, + 0xce, 0xc0, 0x4c, 0xac, 0x63, 0xb8, 0x39, 0x50, + 0xbf, 0xe6, 0x59, 0x6c, 0x38, 0x58, 0x99, 0x1f, + 0x27, 0xa7, 0x9d, 0x71, 0x2a, 0xb3, 0x7b, 0xf9, + 0xfb, 0x17, 0x86, 0xaa, 0x99, 0x81, 0xaa, 0x43, + 0xe4, 0x37, 0xd3, 0x1e, 0x6e, 0xe5, 0xe6, 0xee, + 0xc2, 0xed, 0x95, 0x4f, 0x53, 0x0e, 0x46, 0x8a, + 0xcc, 0x45, 0xa5, 0xdb, 0x69, 0x0d, 0x81, 0xc9, + 0x32, 0x92, 0xbc, 0x8f, 0x33, 0xe6, 0xf6, 0x09, + 0x7c, 0x8e, 0x05, 0x19, 0x0d, 0xf1, 0xb6, 0xcc, + 0xf3, 0x02, 0x21, 0x90, 0x25, 0xec, 0xed, 0x0e }; + static const u8 random[] __initconst = { + 0x95, 0xb7, 0xf1, 0x7e, 0x98, 0x02, 0xd3, 0x57, + 0x73, 0x92, 0xc6, 0xa9, 0xc0, 0x80, 0x83, 0xb6, + 0x7d, 0xd1, 0x29, 0x22, 0x65, 0xb5, 0xf4, 0x2d, + 0x23, 0x7f, 0x1c, 0x55, 0xbb, 0x9b, 0x10, 0xbf, + 0xcf, 0xd8, 0x2c, 0x77, 0xa3, 0x78, 0xb8, 0x26, + 0x6a, 0x00, 0x99, 0x14, 0x3b, 0x3c, 0x2d, 0x64, + 0x61, 0x1e, 0xee, 0xb6, 0x9a, 0xcd, 0xc0, 0x55, + 0x95, 0x7c, 0x13, 0x9e, 0x8b, 0x19, 0x0c, 0x7a, + 0x06, 0x95, 0x5f, 0x2c, 0x79, 0x7c, 0x27, 0x78, + 0xde, 0x94, 0x03, 0x96, 0xa5, 0x01, 0xf4, 0x0e, + 0x91, 0x39, 0x6a, 0xcf, 0x8d, 0x7e, 0x45, 0xeb, + 0xdb, 0xb5, 0x3b, 0xbf, 0x8c, 0x97, 0x52, 0x30, + 0xd2, 0xf0, 0xff, 0x91, 0x06, 0xc7, 0x61, 0x19, + 0xae, 0x49, 0x8e, 0x7f, 0xbc, 0x03, 0xd9, 0x0f, + 0x8e, 0x4c, 0x51, 0x62, 0x7a, 0xed, 0x5c, 0x8d, + 0x42, 0x63, 0xd5, 0xd2, 0xb9, 0x78, 0x87, 0x3a, + 0x0d, 0xe5, 0x96, 0xee, 0x6d, 0xc7, 0xf7, 0xc2, + 0x9e, 0x37, 0xee, 0xe8, 0xb3, 0x4c, 0x90, 0xdd, + 0x1c, 0xf6, 0xa9, 0xdd, 0xb2, 0x2b, 0x4c, 0xbd, + 0x08, 0x6b, 0x14, 0xb3, 0x5d, 0xe9, 0x3d, 0xa2, + 0xd5, 0xcb, 0x18, 0x06, 0x69, 0x8c, 0xbd, 0x7b, + 0xbb, 0x67, 0xbf, 0xe3, 0xd3, 0x1f, 0xd2, 0xd1, + 0xdb, 0xd2, 0xa1, 0xe0, 0x58, 0xa3, 0xeb, 0x99, + 0xd7, 0xe5, 0x1f, 0x1a, 0x93, 0x8e, 0xed, 0x5e, + 0x1c, 0x1d, 0xe2, 0x3a, 0x6b, 0x43, 0x45, 0xd3, + 0x19, 0x14, 0x09, 0xf9, 0x2f, 0x39, 0xb3, 0x67, + 0x0d, 0x8d, 0xbf, 0xb6, 0x35, 0xd8, 0xe6, 0xa3, + 0x69, 0x32, 0xd8, 0x10, 0x33, 0xd1, 0x44, 0x8d, + 0x63, 0xb4, 0x03, 0xdd, 0xf8, 0x8e, 0x12, 0x1b, + 0x6e, 0x81, 0x9a, 0xc3, 0x81, 0x22, 0x6c, 0x13, + 0x21, 0xe4, 0xb0, 0x86, 0x44, 0xf6, 0x72, 0x7c, + 0x36, 0x8c, 0x5a, 0x9f, 0x7a, 0x4b, 0x3e, 0xe2 }; + int ret = 0; - int tmp; + u8 buf[sizeof(random)]; + struct ppno_ws_s ws; + + memset(&ws, 0, sizeof(ws)); + + /* initial seed */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_SEED, + &ws, NULL, 0, + seed, sizeof(seed)); + if (ret < 0) { + pr_err("The prng self test seed operation for the " + "SHA-512 mode failed with rc=%d\n", ret); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + /* check working states V and C */ + if (memcmp(ws.V, V0, sizeof(V0)) != 0 + || memcmp(ws.C, C0, sizeof(C0)) != 0) { + pr_err("The prng self test state test " + "for the SHA-512 mode failed\n"); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + /* generate random bytes */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &ws, buf, sizeof(buf), + NULL, 0); + if (ret < 0) { + pr_err("The prng self test generate operation for " + "the SHA-512 mode failed with rc=%d\n", ret); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &ws, buf, sizeof(buf), + NULL, 0); + if (ret < 0) { + pr_err("The prng self test generate operation for " + "the SHA-512 mode failed with rc=%d\n", ret); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + /* check against expected data */ + if (memcmp(buf, random, sizeof(random)) != 0) { + pr_err("The prng self test data test " + "for the SHA-512 mode failed\n"); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + return 0; +} + + +static int __init prng_sha512_instantiate(void) +{ + int ret, datalen; + u8 seed[64]; + + pr_debug("prng runs in SHA-512 mode " + "with chunksize=%d and reseed_limit=%u\n", + prng_chunk_size, prng_reseed_limit); + + /* memory allocation, prng_data struct init, mutex init */ + datalen = sizeof(struct prng_data_s) + prng_chunk_size; + if (fips_enabled) + datalen += prng_chunk_size; + prng_data = kzalloc(datalen, GFP_KERNEL); + if (!prng_data) { + prng_errorflag = PRNG_INSTANTIATE_FAILED; + return -ENOMEM; + } + mutex_init(&prng_data->mutex); + prng_data->buf = ((u8 *)prng_data) + sizeof(struct prng_data_s); + + /* selftest */ + ret = prng_sha512_selftest(); + if (ret) + goto outfree; + + /* generate initial seed bytestring, first 48 bytes of entropy */ + ret = generate_entropy(seed, 48); + if (ret != 48) + goto outfree; + /* followed by 16 bytes of unique nonce */ + get_tod_clock_ext(seed + 48); + + /* initial seed of the ppno drng */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_SEED, + &prng_data->ppnows, NULL, 0, + seed, sizeof(seed)); + if (ret < 0) { + prng_errorflag = PRNG_SEED_FAILED; + ret = -EIO; + goto outfree; + } + + /* if fips mode is enabled, generate a first block of random + bytes for the FIPS 140-2 Conditional Self Test */ + if (fips_enabled) { + prng_data->prev = prng_data->buf + prng_chunk_size; + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &prng_data->ppnows, + prng_data->prev, + prng_chunk_size, + NULL, 0); + if (ret < 0 || ret != prng_chunk_size) { + prng_errorflag = PRNG_GEN_FAILED; + ret = -EIO; + goto outfree; + } + } + + return 0; + +outfree: + kfree(prng_data); + return ret; +} + + +static void prng_sha512_deinstantiate(void) +{ + pr_debug("The prng module stopped after running in SHA-512 mode\n"); + kzfree(prng_data); +} + + +static int prng_sha512_reseed(void) +{ + int ret; + u8 seed[32]; + + /* generate 32 bytes of fresh entropy */ + ret = generate_entropy(seed, sizeof(seed)); + if (ret != sizeof(seed)) + return ret; + + /* do a reseed of the ppno drng with this bytestring */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_SEED, + &prng_data->ppnows, NULL, 0, + seed, sizeof(seed)); + if (ret) { + prng_errorflag = PRNG_RESEED_FAILED; + return -EIO; + } + + return 0; +} + + +static int prng_sha512_generate(u8 *buf, size_t nbytes) +{ + int ret; + + /* reseed needed ? */ + if (prng_data->ppnows.reseed_counter > prng_reseed_limit) { + ret = prng_sha512_reseed(); + if (ret) + return ret; + } + + /* PPNO generate */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &prng_data->ppnows, buf, nbytes, + NULL, 0); + if (ret < 0 || ret != nbytes) { + prng_errorflag = PRNG_GEN_FAILED; + return -EIO; + } + + /* FIPS 140-2 Conditional Self Test */ + if (fips_enabled) { + if (!memcmp(prng_data->prev, buf, nbytes)) { + prng_errorflag = PRNG_GEN_FAILED; + return -EILSEQ; + } + memcpy(prng_data->prev, buf, nbytes); + } + + return ret; +} + + +/*** file io functions ***/ + +static int prng_open(struct inode *inode, struct file *file) +{ + return nonseekable_open(inode, file); +} + + +static ssize_t prng_tdes_read(struct file *file, char __user *ubuf, + size_t nbytes, loff_t *ppos) +{ + int chunk, n, tmp, ret = 0; + + /* lock prng_data struct */ + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; - /* nbytes can be arbitrary length, we split it into chunks */ while (nbytes) { - /* same as in extract_entropy_user in random.c */ if (need_resched()) { if (signal_pending(current)) { if (ret == 0) ret = -ERESTARTSYS; break; } + /* give mutex free before calling schedule() */ + mutex_unlock(&prng_data->mutex); schedule(); + /* occopy mutex again */ + if (mutex_lock_interruptible(&prng_data->mutex)) { + if (ret == 0) + ret = -ERESTARTSYS; + return ret; + } } /* @@ -112,12 +535,11 @@ static ssize_t prng_read(struct file *file, char __user *ubuf, size_t nbytes, /* PRNG only likes multiples of 8 bytes */ n = (chunk + 7) & -8; - if (p->count > prng_entropy_limit) - prng_seed(8); + if (prng_data->prngws.reseed_counter > prng_reseed_limit) + prng_tdes_seed(8); /* if the CPU supports PRNG stckf is present too */ - asm volatile(".insn s,0xb27c0000,%0" - : "=m" (*((unsigned long long *)p->buf)) : : "cc"); + *((unsigned long long *)prng_data->buf) = get_tod_clock_fast(); /* * Beside the STCKF the input for the TDES-EDE is the output @@ -132,35 +554,259 @@ static ssize_t prng_read(struct file *file, char __user *ubuf, size_t nbytes, * Note: you can still get strict X9.17 conformity by setting * prng_chunk_size to 8 bytes. */ - tmp = crypt_s390_kmc(KMC_PRNG, parm_block, p->buf, p->buf, n); - BUG_ON((tmp < 0) || (tmp != n)); + tmp = crypt_s390_kmc(KMC_PRNG, prng_data->prngws.parm_block, + prng_data->buf, prng_data->buf, n); + if (tmp < 0 || tmp != n) { + ret = -EIO; + break; + } - p->count += n; + prng_data->prngws.byte_counter += n; + prng_data->prngws.reseed_counter += n; - if (copy_to_user(ubuf, p->buf, chunk)) + if (copy_to_user(ubuf, prng_data->buf, chunk)) return -EFAULT; nbytes -= chunk; ret += chunk; ubuf += chunk; } + + /* unlock prng_data struct */ + mutex_unlock(&prng_data->mutex); + return ret; } -static const struct file_operations prng_fops = { + +static ssize_t prng_sha512_read(struct file *file, char __user *ubuf, + size_t nbytes, loff_t *ppos) +{ + int n, ret = 0; + u8 *p; + + /* if errorflag is set do nothing and return 'broken pipe' */ + if (prng_errorflag) + return -EPIPE; + + /* lock prng_data struct */ + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; + + while (nbytes) { + if (need_resched()) { + if (signal_pending(current)) { + if (ret == 0) + ret = -ERESTARTSYS; + break; + } + /* give mutex free before calling schedule() */ + mutex_unlock(&prng_data->mutex); + schedule(); + /* occopy mutex again */ + if (mutex_lock_interruptible(&prng_data->mutex)) { + if (ret == 0) + ret = -ERESTARTSYS; + return ret; + } + } + if (prng_data->rest) { + /* push left over random bytes from the previous read */ + p = prng_data->buf + prng_chunk_size - prng_data->rest; + n = (nbytes < prng_data->rest) ? + nbytes : prng_data->rest; + prng_data->rest -= n; + } else { + /* generate one chunk of random bytes into read buf */ + p = prng_data->buf; + n = prng_sha512_generate(p, prng_chunk_size); + if (n < 0) { + ret = n; + break; + } + if (nbytes < prng_chunk_size) { + n = nbytes; + prng_data->rest = prng_chunk_size - n; + } else { + n = prng_chunk_size; + prng_data->rest = 0; + } + } + if (copy_to_user(ubuf, p, n)) { + ret = -EFAULT; + break; + } + ubuf += n; + nbytes -= n; + ret += n; + } + + /* unlock prng_data struct */ + mutex_unlock(&prng_data->mutex); + + return ret; +} + + +/*** sysfs stuff ***/ + +static const struct file_operations prng_sha512_fops = { + .owner = THIS_MODULE, + .open = &prng_open, + .release = NULL, + .read = &prng_sha512_read, + .llseek = noop_llseek, +}; +static const struct file_operations prng_tdes_fops = { .owner = THIS_MODULE, .open = &prng_open, .release = NULL, - .read = &prng_read, + .read = &prng_tdes_read, .llseek = noop_llseek, }; -static struct miscdevice prng_dev = { +static struct miscdevice prng_sha512_dev = { + .name = "prandom", + .minor = MISC_DYNAMIC_MINOR, + .fops = &prng_sha512_fops, +}; +static struct miscdevice prng_tdes_dev = { .name = "prandom", .minor = MISC_DYNAMIC_MINOR, - .fops = &prng_fops, + .fops = &prng_tdes_fops, }; + +/* chunksize attribute (ro) */ +static ssize_t prng_chunksize_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%u\n", prng_chunk_size); +} +static DEVICE_ATTR(chunksize, 0444, prng_chunksize_show, NULL); + +/* counter attribute (ro) */ +static ssize_t prng_counter_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + u64 counter; + + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; + if (prng_mode == PRNG_MODE_SHA512) + counter = prng_data->ppnows.stream_bytes; + else + counter = prng_data->prngws.byte_counter; + mutex_unlock(&prng_data->mutex); + + return snprintf(buf, PAGE_SIZE, "%llu\n", counter); +} +static DEVICE_ATTR(byte_counter, 0444, prng_counter_show, NULL); + +/* errorflag attribute (ro) */ +static ssize_t prng_errorflag_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%d\n", prng_errorflag); +} +static DEVICE_ATTR(errorflag, 0444, prng_errorflag_show, NULL); + +/* mode attribute (ro) */ +static ssize_t prng_mode_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (prng_mode == PRNG_MODE_TDES) + return snprintf(buf, PAGE_SIZE, "TDES\n"); + else + return snprintf(buf, PAGE_SIZE, "SHA512\n"); +} +static DEVICE_ATTR(mode, 0444, prng_mode_show, NULL); + +/* reseed attribute (w) */ +static ssize_t prng_reseed_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; + prng_sha512_reseed(); + mutex_unlock(&prng_data->mutex); + + return count; +} +static DEVICE_ATTR(reseed, 0200, NULL, prng_reseed_store); + +/* reseed limit attribute (rw) */ +static ssize_t prng_reseed_limit_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%u\n", prng_reseed_limit); +} +static ssize_t prng_reseed_limit_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + unsigned limit; + + if (sscanf(buf, "%u\n", &limit) != 1) + return -EINVAL; + + if (prng_mode == PRNG_MODE_SHA512) { + if (limit < PRNG_RESEED_LIMIT_SHA512_LOWER) + return -EINVAL; + } else { + if (limit < PRNG_RESEED_LIMIT_TDES_LOWER) + return -EINVAL; + } + + prng_reseed_limit = limit; + + return count; +} +static DEVICE_ATTR(reseed_limit, 0644, + prng_reseed_limit_show, prng_reseed_limit_store); + +/* strength attribute (ro) */ +static ssize_t prng_strength_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "256\n"); +} +static DEVICE_ATTR(strength, 0444, prng_strength_show, NULL); + +static struct attribute *prng_sha512_dev_attrs[] = { + &dev_attr_errorflag.attr, + &dev_attr_chunksize.attr, + &dev_attr_byte_counter.attr, + &dev_attr_mode.attr, + &dev_attr_reseed.attr, + &dev_attr_reseed_limit.attr, + &dev_attr_strength.attr, + NULL +}; +static struct attribute *prng_tdes_dev_attrs[] = { + &dev_attr_chunksize.attr, + &dev_attr_byte_counter.attr, + &dev_attr_mode.attr, + NULL +}; + +static struct attribute_group prng_sha512_dev_attr_group = { + .attrs = prng_sha512_dev_attrs +}; +static struct attribute_group prng_tdes_dev_attr_group = { + .attrs = prng_tdes_dev_attrs +}; + + +/*** module init and exit ***/ + static int __init prng_init(void) { int ret; @@ -169,43 +815,105 @@ static int __init prng_init(void) if (!crypt_s390_func_available(KMC_PRNG, CRYPT_S390_MSA)) return -EOPNOTSUPP; - if (prng_chunk_size < 8) - return -EINVAL; + /* choose prng mode */ + if (prng_mode != PRNG_MODE_TDES) { + /* check for MSA5 support for PPNO operations */ + if (!crypt_s390_func_available(PPNO_SHA512_DRNG_GEN, + CRYPT_S390_MSA5)) { + if (prng_mode == PRNG_MODE_SHA512) { + pr_err("The prng module cannot " + "start in SHA-512 mode\n"); + return -EOPNOTSUPP; + } + prng_mode = PRNG_MODE_TDES; + } else + prng_mode = PRNG_MODE_SHA512; + } - p = kmalloc(sizeof(struct s390_prng_data), GFP_KERNEL); - if (!p) - return -ENOMEM; - p->count = 0; + if (prng_mode == PRNG_MODE_SHA512) { - p->buf = kmalloc(prng_chunk_size, GFP_KERNEL); - if (!p->buf) { - ret = -ENOMEM; - goto out_free; - } + /* SHA512 mode */ - /* initialize the PRNG, add 128 bits of entropy */ - prng_seed(16); + if (prng_chunk_size < PRNG_CHUNKSIZE_SHA512_MIN + || prng_chunk_size > PRNG_CHUNKSIZE_SHA512_MAX) + return -EINVAL; + prng_chunk_size = (prng_chunk_size + 0x3f) & ~0x3f; - ret = misc_register(&prng_dev); - if (ret) - goto out_buf; - return 0; + if (prng_reseed_limit == 0) + prng_reseed_limit = PRNG_RESEED_LIMIT_SHA512; + else if (prng_reseed_limit < PRNG_RESEED_LIMIT_SHA512_LOWER) + return -EINVAL; + + ret = prng_sha512_instantiate(); + if (ret) + goto out; + + ret = misc_register(&prng_sha512_dev); + if (ret) { + prng_sha512_deinstantiate(); + goto out; + } + ret = sysfs_create_group(&prng_sha512_dev.this_device->kobj, + &prng_sha512_dev_attr_group); + if (ret) { + misc_deregister(&prng_sha512_dev); + prng_sha512_deinstantiate(); + goto out; + } -out_buf: - kfree(p->buf); -out_free: - kfree(p); + } else { + + /* TDES mode */ + + if (prng_chunk_size < PRNG_CHUNKSIZE_TDES_MIN + || prng_chunk_size > PRNG_CHUNKSIZE_TDES_MAX) + return -EINVAL; + prng_chunk_size = (prng_chunk_size + 0x07) & ~0x07; + + if (prng_reseed_limit == 0) + prng_reseed_limit = PRNG_RESEED_LIMIT_TDES; + else if (prng_reseed_limit < PRNG_RESEED_LIMIT_TDES_LOWER) + return -EINVAL; + + ret = prng_tdes_instantiate(); + if (ret) + goto out; + + ret = misc_register(&prng_tdes_dev); + if (ret) { + prng_tdes_deinstantiate(); + goto out; + } + ret = sysfs_create_group(&prng_tdes_dev.this_device->kobj, + &prng_tdes_dev_attr_group); + if (ret) { + misc_deregister(&prng_tdes_dev); + prng_tdes_deinstantiate(); + goto out; + } + + } + +out: return ret; } + static void __exit prng_exit(void) { - /* wipe me */ - kzfree(p->buf); - kfree(p); - - misc_deregister(&prng_dev); + if (prng_mode == PRNG_MODE_SHA512) { + sysfs_remove_group(&prng_sha512_dev.this_device->kobj, + &prng_sha512_dev_attr_group); + misc_deregister(&prng_sha512_dev); + prng_sha512_deinstantiate(); + } else { + sysfs_remove_group(&prng_tdes_dev.this_device->kobj, + &prng_tdes_dev_attr_group); + misc_deregister(&prng_tdes_dev); + prng_tdes_deinstantiate(); + } } + module_init(prng_init); module_exit(prng_exit); -- cgit v0.10.2 From ec65aafb9e3f316ff9167289e288856a7d528773 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 23 Apr 2015 12:06:30 +0200 Subject: netdev_alloc_pcpu_stats: use less common iterator variable With the CPU iteration variable called 'i', it's relatively easy to have variable shadowing which sparse will warn about. Avoid that by renaming the variable to __cpu which is less likely to be used in the surrounding context. Signed-off-by: Johannes Berg Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index dae106a..dbad4d7 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2025,10 +2025,10 @@ struct pcpu_sw_netstats { ({ \ typeof(type) __percpu *pcpu_stats = alloc_percpu(type); \ if (pcpu_stats) { \ - int i; \ - for_each_possible_cpu(i) { \ + int __cpu; \ + for_each_possible_cpu(__cpu) { \ typeof(type) *stat; \ - stat = per_cpu_ptr(pcpu_stats, i); \ + stat = per_cpu_ptr(pcpu_stats, __cpu); \ u64_stats_init(&stat->syncp); \ } \ } \ -- cgit v0.10.2 From 4fce14820c1b0a3fd399719f970e0c3ae40dd270 Mon Sep 17 00:00:00 2001 From: David Gibson Date: Thu, 23 Apr 2015 14:43:05 +1000 Subject: ibmveth: Fix off-by-one error in ibmveth_change_mtu() AFAIK the PAPR document which defines the virtual device interface used by the ibmveth driver doesn't specify a specific maximum MTU. So, in the ibmveth driver, the maximum allowed MTU is determined by the maximum allocated buffer size of 64k (corresponding to one page in the common case) minus the per-buffer overhead IBMVETH_BUFF_OH (which has value 22 for 14 bytes of ethernet header, plus 8 bytes for an opaque handle). This suggests a maximum allowable MTU of 65514 bytes, but in fact the driver only permits a maximum MTU of 65513. This is because there is a < instead of an <= in ibmveth_change_mtu(), which only permits an MTU which is strictly smaller than the buffer size, rather than allowing the buffer to be completely filled. This patch fixes the buglet. Signed-off-by: David Gibson Acked-by: Thomas Falcon Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index cd7675a..1813476 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -1238,7 +1238,7 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu) return -EINVAL; for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++) - if (new_mtu_oh < adapter->rx_buff_pool[i].buff_size) + if (new_mtu_oh <= adapter->rx_buff_pool[i].buff_size) break; if (i == IBMVETH_NUM_BUFF_POOLS) @@ -1257,7 +1257,7 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu) for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++) { adapter->rx_buff_pool[i].active = 1; - if (new_mtu_oh < adapter->rx_buff_pool[i].buff_size) { + if (new_mtu_oh <= adapter->rx_buff_pool[i].buff_size) { dev->mtu = new_mtu; vio_cmo_set_dev_desired(viodev, ibmveth_get_desired_dma -- cgit v0.10.2 From def81f69bfbd70a3278a7592a4ab8717300cbac1 Mon Sep 17 00:00:00 2001 From: Ying Xue Date: Thu, 23 Apr 2015 09:37:38 -0400 Subject: tipc: fix topology server broken issue When a new topology server is launched in a new namespace, its listening socket is inserted into the "init ns" namespace's socket hash table rather than the one owned by the new namespace. Although the socket's namespace is forcedly changed to the new namespace later, the socket is still stored in the socket hash table of "init ns" namespace. When a client created in the new namespace connects its own topology server, the connection is failed as its server's socket could not be found from its own namespace's socket table. If __sock_create() instead of original sock_create_kern() is used to create the server's socket through specifying an expected namesapce, the socket will be inserted into the specified namespace's socket table, thereby avoiding to the topology server broken issue. Fixes: 76100a8a64bc ("tipc: fix netns refcnt leak") Reported-by: Erik Hugne Signed-off-by: Ying Xue Signed-off-by: Jon Maloy Signed-off-by: David S. Miller diff --git a/net/tipc/server.c b/net/tipc/server.c index ab6183c..77ff03e 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -102,7 +102,7 @@ static void tipc_conn_kref_release(struct kref *kref) } saddr->scope = -TIPC_NODE_SCOPE; kernel_bind(sock, (struct sockaddr *)saddr, sizeof(*saddr)); - sk_release_kernel(sk); + sock_release(sock); con->sock = NULL; } @@ -321,12 +321,9 @@ static struct socket *tipc_create_listen_sock(struct tipc_conn *con) struct socket *sock = NULL; int ret; - ret = sock_create_kern(AF_TIPC, SOCK_SEQPACKET, 0, &sock); + ret = __sock_create(s->net, AF_TIPC, SOCK_SEQPACKET, 0, &sock, 1); if (ret < 0) return NULL; - - sk_change_net(sock->sk, s->net); - ret = kernel_setsockopt(sock, SOL_TIPC, TIPC_IMPORTANCE, (char *)&s->imp, sizeof(s->imp)); if (ret < 0) @@ -376,7 +373,7 @@ static struct socket *tipc_create_listen_sock(struct tipc_conn *con) create_err: kernel_sock_shutdown(sock, SHUT_RDWR); - sk_release_kernel(sock->sk); + sock_release(sock); return NULL; } -- cgit v0.10.2 From 9871b27f6705fc6e0ba633b136369a289b2bfb99 Mon Sep 17 00:00:00 2001 From: Erik Hugne Date: Thu, 23 Apr 2015 09:37:39 -0400 Subject: tipc: fix random link reset problem In the function tipc_sk_rcv(), the stack variable 'err' is only initialized to TIPC_ERR_NO_PORT for the first iteration over the link input queue. If a chain of messages are received from a link, failure to lookup the socket for any but the first message will cause the message to bounce back out on a random link. We fix this by properly initializing err. Signed-off-by: Erik Hugne Signed-off-by: Jon Maloy Signed-off-by: David S. Miller diff --git a/net/tipc/socket.c b/net/tipc/socket.c index ee90d74..9074b5c 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1764,13 +1764,14 @@ static int tipc_sk_enqueue(struct sk_buff_head *inputq, struct sock *sk, int tipc_sk_rcv(struct net *net, struct sk_buff_head *inputq) { u32 dnode, dport = 0; - int err = -TIPC_ERR_NO_PORT; + int err; struct sk_buff *skb; struct tipc_sock *tsk; struct tipc_net *tn; struct sock *sk; while (skb_queue_len(inputq)) { + err = -TIPC_ERR_NO_PORT; skb = NULL; dport = tipc_skb_peek_port(inputq, dport); tsk = tipc_sk_lookup(net, dport); -- cgit v0.10.2 From 73a317377303b5ec14d4703d73ba87efffbb779d Mon Sep 17 00:00:00 2001 From: Erik Hugne Date: Thu, 23 Apr 2015 09:37:40 -0400 Subject: tipc: fix node refcount issue When link statistics is dumped over netlink, we iterate over the list of peer nodes and append each links statistics to the netlink msg. In the case where the dump is resumed after filling up a nlmsg, the node refcnt is decremented without having been incremented previously which may cause the node reference to be freed. When this happens, the following info/stacktrace will be generated, followed by a crash or undefined behavior. We fix this by removing the erroneous call to tipc_node_put inside the loop that iterates over nodes. [ 384.312303] INFO: trying to register non-static key. [ 384.313110] the code is fine but needs lockdep annotation. [ 384.313290] turning off the locking correctness validator. [ 384.313290] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.0.0+ #13 [ 384.313290] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 384.313290] ffff88003c6d0290 ffff88003cc03ca8 ffffffff8170adf1 0000000000000007 [ 384.313290] ffffffff82728730 ffff88003cc03d38 ffffffff810a6a6d 00000000001d7200 [ 384.313290] ffff88003c6d0ab0 ffff88003cc03ce8 0000000000000285 0000000000000001 [ 384.313290] Call Trace: [ 384.313290] [] dump_stack+0x4c/0x65 [ 384.313290] [] __lock_acquire+0xf3d/0xf50 [ 384.313290] [] lock_acquire+0xd5/0x290 [ 384.313290] [] ? link_timeout+0x1c/0x170 [tipc] [ 384.313290] [] ? link_state_event+0x4e0/0x4e0 [tipc] [ 384.313290] [] _raw_spin_lock_bh+0x40/0x80 [ 384.313290] [] ? link_timeout+0x1c/0x170 [tipc] [ 384.313290] [] link_timeout+0x1c/0x170 [tipc] [ 384.313290] [] call_timer_fn+0xb8/0x490 [ 384.313290] [] ? process_timeout+0x10/0x10 [ 384.313290] [] run_timer_softirq+0x21c/0x420 [ 384.313290] [] ? link_state_event+0x4e0/0x4e0 [tipc] [ 384.313290] [] __do_softirq+0xf4/0x630 [ 384.313290] [] irq_exit+0x5d/0x60 [ 384.313290] [] smp_apic_timer_interrupt+0x41/0x50 [ 384.313290] [] apic_timer_interrupt+0x70/0x80 [ 384.313290] [] ? default_idle+0x20/0x210 [ 384.313290] [] ? default_idle+0x1e/0x210 [ 384.313290] [] arch_cpu_idle+0xa/0x10 [ 384.313290] [] cpu_startup_entry+0x2c3/0x530 [ 384.313290] [] ? clockevents_register_device+0x113/0x200 [ 384.313290] [] start_secondary+0x13f/0x170 Fixes: 8a0f6ebe8494 ("tipc: involve reference counter for node structure") Signed-off-by: Erik Hugne Signed-off-by: Jon Maloy Signed-off-by: David S. Miller diff --git a/net/tipc/link.c b/net/tipc/link.c index a6b30df..57be6e6 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -2143,7 +2143,6 @@ int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb) err = __tipc_nl_add_node_links(net, &msg, node, &prev_link); tipc_node_unlock(node); - tipc_node_put(node); if (err) goto out; -- cgit v0.10.2 From 9283b42e46c2646dff1bec47e2dd683add7f9972 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Thu, 23 Apr 2015 10:24:45 -0600 Subject: NVMe: Fix VPD B0 max sectors translation Use the namespace's block format for reporting the max transfer length. Max unmap count is left as-is since NVMe doesn't provide a max, so the value the driver provided the block layer is valid for any format. Reported-by: Martin K. Petersen Signed-off-by: Keith Busch Signed-off-by: Jens Axboe diff --git a/drivers/block/nvme-scsi.c b/drivers/block/nvme-scsi.c index 6b736b0..88f13c5 100644 --- a/drivers/block/nvme-scsi.c +++ b/drivers/block/nvme-scsi.c @@ -944,7 +944,8 @@ static int nvme_trans_ext_inq_page(struct nvme_ns *ns, struct sg_io_hdr *hdr, static int nvme_trans_bdev_limits_page(struct nvme_ns *ns, struct sg_io_hdr *hdr, u8 *inq_response, int alloc_len) { - __be32 max_sectors = cpu_to_be32(queue_max_hw_sectors(ns->queue)); + __be32 max_sectors = cpu_to_be32( + nvme_block_nr(ns, queue_max_hw_sectors(ns->queue))); __be32 max_discard = cpu_to_be32(ns->queue->limits.max_discard_sectors); __be32 discard_desc_count = cpu_to_be32(0x100); -- cgit v0.10.2 From f054b56c951bf1731ba7314a4c7f1cc0b2977cc9 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Tue, 21 Apr 2015 10:00:19 +0800 Subject: blk-mq: fix race between timeout and CPU hotplug Firstly during CPU hotplug, even queue is freezed, timeout handler still may come and access hctx->tags, which may cause use after free, so this patch deactivates timeout handler inside CPU hotplug notifier. Secondly, tags can be shared by more than one queues, so we have to check if the hctx has been unmapped, otherwise still use-after-free on tags can be triggered. Cc: Reported-by: Dongsu Park Tested-by: Dongsu Park Signed-off-by: Ming Lei Signed-off-by: Jens Axboe diff --git a/block/blk-mq.c b/block/blk-mq.c index ade8a2d..1fccb98 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -677,8 +677,11 @@ static void blk_mq_rq_timer(unsigned long priv) data.next = blk_rq_timeout(round_jiffies_up(data.next)); mod_timer(&q->timeout, data.next); } else { - queue_for_each_hw_ctx(q, hctx, i) - blk_mq_tag_idle(hctx); + queue_for_each_hw_ctx(q, hctx, i) { + /* the hctx may be unmapped, so check it here */ + if (blk_mq_hw_queue_mapped(hctx)) + blk_mq_tag_idle(hctx); + } } } @@ -2090,9 +2093,16 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb, */ list_for_each_entry(q, &all_q_list, all_q_node) blk_mq_freeze_queue_start(q); - list_for_each_entry(q, &all_q_list, all_q_node) + list_for_each_entry(q, &all_q_list, all_q_node) { blk_mq_freeze_queue_wait(q); + /* + * timeout handler can't touch hw queue during the + * reinitialization + */ + del_timer_sync(&q->timeout); + } + list_for_each_entry(q, &all_q_list, all_q_node) blk_mq_queue_reinit(q); -- cgit v0.10.2 From 2a34c0872adf252f23a6fef2d051a169ac796cef Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Tue, 21 Apr 2015 10:00:20 +0800 Subject: blk-mq: fix CPU hotplug handling hctx->tags has to be set as NULL in case that it is to be unmapped no matter if set->tags[hctx->queue_num] is NULL or not in blk_mq_map_swqueue() because shared tags can be freed already from another request queue. The same situation has to be considered during handling CPU online too. Unmapped hw queue can be remapped after CPU topo is changed, so we need to allocate tags for the hw queue in blk_mq_map_swqueue(). Then tags allocation for hw queue can be removed in hctx cpu online notifier, and it is reasonable to do that after mapping is updated. Cc: Reported-by: Dongsu Park Tested-by: Dongsu Park Signed-off-by: Ming Lei Signed-off-by: Jens Axboe diff --git a/block/blk-mq.c b/block/blk-mq.c index 1fccb98..76f460e 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -1574,22 +1574,6 @@ static int blk_mq_hctx_cpu_offline(struct blk_mq_hw_ctx *hctx, int cpu) return NOTIFY_OK; } -static int blk_mq_hctx_cpu_online(struct blk_mq_hw_ctx *hctx, int cpu) -{ - struct request_queue *q = hctx->queue; - struct blk_mq_tag_set *set = q->tag_set; - - if (set->tags[hctx->queue_num]) - return NOTIFY_OK; - - set->tags[hctx->queue_num] = blk_mq_init_rq_map(set, hctx->queue_num); - if (!set->tags[hctx->queue_num]) - return NOTIFY_STOP; - - hctx->tags = set->tags[hctx->queue_num]; - return NOTIFY_OK; -} - static int blk_mq_hctx_notify(void *data, unsigned long action, unsigned int cpu) { @@ -1597,8 +1581,11 @@ static int blk_mq_hctx_notify(void *data, unsigned long action, if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) return blk_mq_hctx_cpu_offline(hctx, cpu); - else if (action == CPU_ONLINE || action == CPU_ONLINE_FROZEN) - return blk_mq_hctx_cpu_online(hctx, cpu); + + /* + * In case of CPU online, tags may be reallocated + * in blk_mq_map_swqueue() after mapping is updated. + */ return NOTIFY_OK; } @@ -1778,6 +1765,7 @@ static void blk_mq_map_swqueue(struct request_queue *q) unsigned int i; struct blk_mq_hw_ctx *hctx; struct blk_mq_ctx *ctx; + struct blk_mq_tag_set *set = q->tag_set; queue_for_each_hw_ctx(q, hctx, i) { cpumask_clear(hctx->cpumask); @@ -1806,16 +1794,20 @@ static void blk_mq_map_swqueue(struct request_queue *q) * disable it and free the request entries. */ if (!hctx->nr_ctx) { - struct blk_mq_tag_set *set = q->tag_set; - if (set->tags[i]) { blk_mq_free_rq_map(set, set->tags[i], i); set->tags[i] = NULL; - hctx->tags = NULL; } + hctx->tags = NULL; continue; } + /* unmapped hw queue can be remapped after CPU topo changed */ + if (!set->tags[i]) + set->tags[i] = blk_mq_init_rq_map(set, i); + hctx->tags = set->tags[i]; + WARN_ON(!hctx->tags); + /* * Set the map size to the number of mapped software queues. * This is more accurate and more efficient than looping -- cgit v0.10.2 From 464d1387acb94dc43ba772b35242345e3d2ead1b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 21 Apr 2015 16:49:13 -0400 Subject: writeback: use |1 instead of +1 to protect against div by zero mm/page-writeback.c has several places where 1 is added to the divisor to prevent division by zero exceptions; however, if the original divisor is equivalent to -1, adding 1 leads to division by zero. There are three places where +1 is used for this purpose - one in pos_ratio_polynom() and two in bdi_position_ratio(). The second one in bdi_position_ratio() actually triggered div-by-zero oops on a machine running a 3.10 kernel. The divisor is x_intercept - bdi_setpoint + 1 == span + 1 span is confirmed to be (u32)-1. It isn't clear how it ended up that but it could be from write bandwidth calculation underflow fixed by c72efb658f7c ("writeback: fix possible underflow in write bandwidth calculation"). At any rate, +1 isn't a proper protection against div-by-zero. This patch converts all +1 protections to |1. Note that bdi_update_dirty_ratelimit() was already using |1 before this patch. Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org Reviewed-by: Jan Kara Signed-off-by: Jens Axboe diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 5daf556..eb59f7e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -580,7 +580,7 @@ static long long pos_ratio_polynom(unsigned long setpoint, long x; x = div64_s64(((s64)setpoint - (s64)dirty) << RATELIMIT_CALC_SHIFT, - limit - setpoint + 1); + (limit - setpoint) | 1); pos_ratio = x; pos_ratio = pos_ratio * x >> RATELIMIT_CALC_SHIFT; pos_ratio = pos_ratio * x >> RATELIMIT_CALC_SHIFT; @@ -807,7 +807,7 @@ static unsigned long bdi_position_ratio(struct backing_dev_info *bdi, * scale global setpoint to bdi's: * bdi_setpoint = setpoint * bdi_thresh / thresh */ - x = div_u64((u64)bdi_thresh << 16, thresh + 1); + x = div_u64((u64)bdi_thresh << 16, thresh | 1); bdi_setpoint = setpoint * (u64)x >> 16; /* * Use span=(8*write_bw) in single bdi case as indicated by @@ -822,7 +822,7 @@ static unsigned long bdi_position_ratio(struct backing_dev_info *bdi, if (bdi_dirty < x_intercept - span / 4) { pos_ratio = div64_u64(pos_ratio * (x_intercept - bdi_dirty), - x_intercept - bdi_setpoint + 1); + (x_intercept - bdi_setpoint) | 1); } else pos_ratio /= 4; -- cgit v0.10.2 From 8406a4d56ea94d1d91f62cab3bed15399bac73cb Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Thu, 23 Apr 2015 10:47:44 -0600 Subject: elevator: fix double release of elevator module Our issue is descripted in below call path: ->elevator_init ->elevator_init_fn ->{cfq,deadline,noop}_init_queue ->elevator_alloc ->kzalloc_node fail to call kzalloc_node and then put module in elevator_alloc; fail to call elevator_init_fn and then put module again in elevator_init. Remove elevator_put invoking in error path of elevator_alloc to avoid double release issue. Signed-off-by: Chao Yu Reviewed-by: Jeff Moyer Signed-off-by: Jens Axboe diff --git a/block/elevator.c b/block/elevator.c index 59794d0..8985038 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -157,7 +157,7 @@ struct elevator_queue *elevator_alloc(struct request_queue *q, eq = kzalloc_node(sizeof(*eq), GFP_KERNEL, q->node); if (unlikely(!eq)) - goto err; + return NULL; eq->type = e; kobject_init(&eq->kobj, &elv_ktype); @@ -165,10 +165,6 @@ struct elevator_queue *elevator_alloc(struct request_queue *q, hash_init(eq->hash); return eq; -err: - kfree(eq); - elevator_put(e); - return NULL; } EXPORT_SYMBOL(elevator_alloc); -- cgit v0.10.2 From d1ab39f17f8653021620d0355ee1cd24d7442a4f Mon Sep 17 00:00:00 2001 From: Jason Eastman Date: Wed, 22 Apr 2015 00:56:42 -0600 Subject: net: unix: garbage: fixed several comment and whitespace style issues fixed several comment and whitespace style issues Signed-off-by: Jason Eastman Signed-off-by: David S. Miller diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 99f7012..a73a226 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -95,39 +95,36 @@ static DECLARE_WAIT_QUEUE_HEAD(unix_gc_wait); unsigned int unix_tot_inflight; - struct sock *unix_get_socket(struct file *filp) { struct sock *u_sock = NULL; struct inode *inode = file_inode(filp); - /* - * Socket ? - */ + /* Socket ? */ if (S_ISSOCK(inode->i_mode) && !(filp->f_mode & FMODE_PATH)) { struct socket *sock = SOCKET_I(inode); struct sock *s = sock->sk; - /* - * PF_UNIX ? - */ + /* PF_UNIX ? */ if (s && sock->ops && sock->ops->family == PF_UNIX) u_sock = s; } return u_sock; } -/* - * Keep the number of times in flight count for the file - * descriptor if it is for an AF_UNIX socket. +/* Keep the number of times in flight count for the file + * descriptor if it is for an AF_UNIX socket. */ void unix_inflight(struct file *fp) { struct sock *s = unix_get_socket(fp); + if (s) { struct unix_sock *u = unix_sk(s); + spin_lock(&unix_gc_lock); + if (atomic_long_inc_return(&u->inflight) == 1) { BUG_ON(!list_empty(&u->link)); list_add_tail(&u->link, &gc_inflight_list); @@ -142,10 +139,13 @@ void unix_inflight(struct file *fp) void unix_notinflight(struct file *fp) { struct sock *s = unix_get_socket(fp); + if (s) { struct unix_sock *u = unix_sk(s); + spin_lock(&unix_gc_lock); BUG_ON(list_empty(&u->link)); + if (atomic_long_dec_and_test(&u->inflight)) list_del_init(&u->link); unix_tot_inflight--; @@ -161,32 +161,27 @@ static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *), spin_lock(&x->sk_receive_queue.lock); skb_queue_walk_safe(&x->sk_receive_queue, skb, next) { - /* - * Do we have file descriptors ? - */ + /* Do we have file descriptors ? */ if (UNIXCB(skb).fp) { bool hit = false; - /* - * Process the descriptors of this socket - */ + /* Process the descriptors of this socket */ int nfd = UNIXCB(skb).fp->count; struct file **fp = UNIXCB(skb).fp->fp; + while (nfd--) { - /* - * Get the socket the fd matches - * if it indeed does so - */ + /* Get the socket the fd matches if it indeed does so */ struct sock *sk = unix_get_socket(*fp++); + if (sk) { struct unix_sock *u = unix_sk(sk); - /* - * Ignore non-candidates, they could + /* Ignore non-candidates, they could * have been added to the queues after * starting the garbage collection */ if (test_bit(UNIX_GC_CANDIDATE, &u->gc_flags)) { hit = true; + func(u); } } @@ -203,24 +198,22 @@ static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *), static void scan_children(struct sock *x, void (*func)(struct unix_sock *), struct sk_buff_head *hitlist) { - if (x->sk_state != TCP_LISTEN) + if (x->sk_state != TCP_LISTEN) { scan_inflight(x, func, hitlist); - else { + } else { struct sk_buff *skb; struct sk_buff *next; struct unix_sock *u; LIST_HEAD(embryos); - /* - * For a listening socket collect the queued embryos + /* For a listening socket collect the queued embryos * and perform a scan on them as well. */ spin_lock(&x->sk_receive_queue.lock); skb_queue_walk_safe(&x->sk_receive_queue, skb, next) { u = unix_sk(skb->sk); - /* - * An embryo cannot be in-flight, so it's safe + /* An embryo cannot be in-flight, so it's safe * to use the list link. */ BUG_ON(!list_empty(&u->link)); @@ -249,8 +242,7 @@ static void inc_inflight(struct unix_sock *usk) static void inc_inflight_move_tail(struct unix_sock *u) { atomic_long_inc(&u->inflight); - /* - * If this still might be part of a cycle, move it to the end + /* If this still might be part of a cycle, move it to the end * of the list, so that it's checked even if it was already * passed over */ @@ -263,8 +255,7 @@ static bool gc_in_progress; void wait_for_unix_gc(void) { - /* - * If number of inflight sockets is insane, + /* If number of inflight sockets is insane, * force a garbage collect right now. */ if (unix_tot_inflight > UNIX_INFLIGHT_TRIGGER_GC && !gc_in_progress) @@ -288,8 +279,7 @@ void unix_gc(void) goto out; gc_in_progress = true; - /* - * First, select candidates for garbage collection. Only + /* First, select candidates for garbage collection. Only * in-flight sockets are considered, and from those only ones * which don't have any external reference. * @@ -320,15 +310,13 @@ void unix_gc(void) } } - /* - * Now remove all internal in-flight reference to children of + /* Now remove all internal in-flight reference to children of * the candidates. */ list_for_each_entry(u, &gc_candidates, link) scan_children(&u->sk, dec_inflight, NULL); - /* - * Restore the references for children of all candidates, + /* Restore the references for children of all candidates, * which have remaining references. Do this recursively, so * only those remain, which form cyclic references. * @@ -350,8 +338,7 @@ void unix_gc(void) } list_del(&cursor); - /* - * not_cycle_list contains those sockets which do not make up a + /* not_cycle_list contains those sockets which do not make up a * cycle. Restore these to the inflight list. */ while (!list_empty(¬_cycle_list)) { @@ -360,8 +347,7 @@ void unix_gc(void) list_move_tail(&u->link, &gc_inflight_list); } - /* - * Now gc_candidates contains only garbage. Restore original + /* Now gc_candidates contains only garbage. Restore original * inflight counters for these as well, and remove the skbuffs * which are creating the cycle(s). */ -- cgit v0.10.2 From 61e77d297dd1fc739828027a2790ee5365de50d2 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 23 Apr 2015 19:59:33 +0200 Subject: ethernet: amd: AMD_XGBE should depend on HAS_DMA If NO_DMA=y: drivers/built-in.o: In function `xgbe_probe': xgbe-main.c:(.text+0x2def0a): undefined reference to `dma_set_mask' xgbe-main.c:(.text+0x2def20): undefined reference to `dma_supported' drivers/built-in.o: In function `xgbe_rx_poll': xgbe-drv.c:(.text+0x2e0320): undefined reference to `dma_sync_single_for_cpu' xgbe-drv.c:(.text+0x2e035e): undefined reference to `dma_sync_single_for_cpu' drivers/built-in.o: In function `xgbe_unmap_rdata': xgbe-desc.c:(.text+0x2e5fe4): undefined reference to `dma_unmap_page' xgbe-desc.c:(.text+0x2e5ffa): undefined reference to `dma_unmap_single' xgbe-desc.c:(.text+0x2e604a): undefined reference to `dma_unmap_page' xgbe-desc.c:(.text+0x2e6084): undefined reference to `dma_unmap_page' drivers/built-in.o: In function `xgbe_alloc_pages': xgbe-desc.c:(.text+0x2e6156): undefined reference to `dma_map_page' xgbe-desc.c:(.text+0x2e6164): undefined reference to `dma_mapping_error' drivers/built-in.o: In function `xgbe_free_ring': xgbe-desc.c:(.text+0x2e63d4): undefined reference to `dma_unmap_page' xgbe-desc.c:(.text+0x2e640e): undefined reference to `dma_unmap_page' xgbe-desc.c:(.text+0x2e644a): undefined reference to `dma_free_coherent' drivers/built-in.o: In function `xgbe_init_ring': xgbe-desc.c:(.text+0x2e64d4): undefined reference to `dma_alloc_coherent' drivers/built-in.o: In function `xgbe_map_tx_skb': xgbe-desc.c:(.text+0x2e6628): undefined reference to `dma_map_single' xgbe-desc.c:(.text+0x2e6638): undefined reference to `dma_mapping_error' xgbe-desc.c:(.text+0x2e66b2): undefined reference to `dma_map_single' xgbe-desc.c:(.text+0x2e66c2): undefined reference to `dma_mapping_error' xgbe-desc.c:(.text+0x2e6762): undefined reference to `dma_map_page' xgbe-desc.c:(.text+0x2e6772): undefined reference to `dma_mapping_error' Signed-off-by: Geert Uytterhoeven Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig index c638c85..089c269 100644 --- a/drivers/net/ethernet/amd/Kconfig +++ b/drivers/net/ethernet/amd/Kconfig @@ -179,7 +179,7 @@ config SUNLANCE config AMD_XGBE tristate "AMD 10GbE Ethernet driver" - depends on (OF_NET || ACPI) && HAS_IOMEM + depends on (OF_NET || ACPI) && HAS_IOMEM && HAS_DMA select PHYLIB select AMD_XGBE_PHY select BITREVERSE -- cgit v0.10.2 From 0357cc1def050c4d6107835ed6b79122243bd240 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 23 Apr 2015 19:59:34 +0200 Subject: ethernet: arc: ARC_EMAC and EMAC_ROCKCHIP should depend on HAS_DMA If NO_DMA=y: drivers/built-in.o: In function `arc_emac_tx_clean': emac_main.c:(.text+0x2decde): undefined reference to `dma_unmap_single' drivers/built-in.o: In function `arc_emac_rx': emac_main.c:(.text+0x2dee1c): undefined reference to `dma_unmap_single' emac_main.c:(.text+0x2dee72): undefined reference to `dma_map_single' emac_main.c:(.text+0x2dee7e): undefined reference to `dma_mapping_error' drivers/built-in.o: In function `arc_emac_probe': (.text+0x2df2ee): undefined reference to `dmam_alloc_coherent' drivers/built-in.o: In function `arc_emac_open': emac_main.c:(.text+0x2df6d8): undefined reference to `dma_map_single' emac_main.c:(.text+0x2df6e4): undefined reference to `dma_mapping_error' drivers/built-in.o: In function `arc_emac_tx': emac_main.c:(.text+0x2df9e4): undefined reference to `dma_map_single' emac_main.c:(.text+0x2df9f0): undefined reference to `dma_mapping_error' Signed-off-by: Geert Uytterhoeven Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/arc/Kconfig b/drivers/net/ethernet/arc/Kconfig index 8e262e2..dea29ee 100644 --- a/drivers/net/ethernet/arc/Kconfig +++ b/drivers/net/ethernet/arc/Kconfig @@ -25,8 +25,7 @@ config ARC_EMAC_CORE config ARC_EMAC tristate "ARC EMAC support" select ARC_EMAC_CORE - depends on OF_IRQ - depends on OF_NET + depends on OF_IRQ && OF_NET && HAS_DMA ---help--- On some legacy ARC (Synopsys) FPGA boards such as ARCAngel4/ML50x non-standard on-chip ethernet device ARC EMAC 10/100 is used. @@ -35,7 +34,7 @@ config ARC_EMAC config EMAC_ROCKCHIP tristate "Rockchip EMAC support" select ARC_EMAC_CORE - depends on OF_IRQ && OF_NET && REGULATOR + depends on OF_IRQ && OF_NET && REGULATOR && HAS_DMA ---help--- Support for Rockchip RK3066/RK3188 EMAC ethernet controllers. This selects Rockchip SoC glue layer support for the -- cgit v0.10.2 From 2fb42aab48762fb1d5e21fbb40acad0f2650b9b4 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 23 Apr 2015 20:04:51 +0200 Subject: can: CAN_GRCAN should depend on HAS_DMA If NO_DMA=y: drivers/built-in.o: In function `grcan_free_dma_buffers': grcan.c:(.text+0x2d7716): undefined reference to `dma_free_coherent' drivers/built-in.o: In function `grcan_allocate_dma_buffers': grcan.c:(.text+0x2d779c): undefined reference to `dma_alloc_coherent' Signed-off-by: Geert Uytterhoeven Signed-off-by: David S. Miller diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig index 58808f6..e8c96b8 100644 --- a/drivers/net/can/Kconfig +++ b/drivers/net/can/Kconfig @@ -112,7 +112,7 @@ config PCH_CAN config CAN_GRCAN tristate "Aeroflex Gaisler GRCAN and GRHCAN CAN devices" - depends on OF + depends on OF && HAS_DMA ---help--- Say Y here if you want to use Aeroflex Gaisler GRCAN or GRHCAN. Note that the driver supports little endian, even though little -- cgit v0.10.2 From e4b6c30375e83b92d9c3e9b9d853417e8cc74006 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Tue, 21 Apr 2015 13:09:45 -0700 Subject: ethernet: myri10ge: use arch_phys_wc_add() This driver already uses ioremap_wc() on the same range so when write-combining is available that will be used instead. Cc: Hyong-Youb Kim Cc: Andy Lutomirski Cc: Suresh Siddha Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Juergen Gross Cc: Daniel Vetter Cc: netdev@vger.kernel.org Cc: Juergen Gross Cc: Daniel Vetter Cc: Andy Lutomirski Cc: Dave Airlie Cc: Antonino Daplas Cc: Jean-Christophe Plagniol-Villard Cc: Tomi Valkeinen Cc: linux-kernel@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Luis R. Rodriguez Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c index 1412f5a..2bae502 100644 --- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c +++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c @@ -69,11 +69,7 @@ #include #include #include -#include #include -#ifdef CONFIG_MTRR -#include -#endif #include #include "myri10ge_mcp.h" @@ -242,8 +238,7 @@ struct myri10ge_priv { unsigned int rdma_tags_available; int intr_coal_delay; __be32 __iomem *intr_coal_delay_ptr; - int mtrr; - int wc_enabled; + int wc_cookie; int down_cnt; wait_queue_head_t down_wq; struct work_struct watchdog_work; @@ -1905,7 +1900,7 @@ static const char myri10ge_gstrings_main_stats[][ETH_GSTRING_LEN] = { "tx_aborted_errors", "tx_carrier_errors", "tx_fifo_errors", "tx_heartbeat_errors", "tx_window_errors", /* device-specific stats */ - "tx_boundary", "WC", "irq", "MSI", "MSIX", + "tx_boundary", "irq", "MSI", "MSIX", "read_dma_bw_MBs", "write_dma_bw_MBs", "read_write_dma_bw_MBs", "serial_number", "watchdog_resets", #ifdef CONFIG_MYRI10GE_DCA @@ -1984,7 +1979,6 @@ myri10ge_get_ethtool_stats(struct net_device *netdev, data[i] = ((u64 *)&link_stats)[i]; data[i++] = (unsigned int)mgp->tx_boundary; - data[i++] = (unsigned int)mgp->wc_enabled; data[i++] = (unsigned int)mgp->pdev->irq; data[i++] = (unsigned int)mgp->msi_enabled; data[i++] = (unsigned int)mgp->msix_enabled; @@ -4040,14 +4034,7 @@ static int myri10ge_probe(struct pci_dev *pdev, const struct pci_device_id *ent) mgp->board_span = pci_resource_len(pdev, 0); mgp->iomem_base = pci_resource_start(pdev, 0); - mgp->mtrr = -1; - mgp->wc_enabled = 0; -#ifdef CONFIG_MTRR - mgp->mtrr = mtrr_add(mgp->iomem_base, mgp->board_span, - MTRR_TYPE_WRCOMB, 1); - if (mgp->mtrr >= 0) - mgp->wc_enabled = 1; -#endif + mgp->wc_cookie = arch_phys_wc_add(mgp->iomem_base, mgp->board_span); mgp->sram = ioremap_wc(mgp->iomem_base, mgp->board_span); if (mgp->sram == NULL) { dev_err(&pdev->dev, "ioremap failed for %ld bytes at 0x%lx\n", @@ -4146,14 +4133,14 @@ static int myri10ge_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto abort_with_state; } if (mgp->msix_enabled) - dev_info(dev, "%d MSI-X IRQs, tx bndry %d, fw %s, WC %s\n", + dev_info(dev, "%d MSI-X IRQs, tx bndry %d, fw %s, MTRR %s, WC Enabled\n", mgp->num_slices, mgp->tx_boundary, mgp->fw_name, - (mgp->wc_enabled ? "Enabled" : "Disabled")); + (mgp->wc_cookie > 0 ? "Enabled" : "Disabled")); else - dev_info(dev, "%s IRQ %d, tx bndry %d, fw %s, WC %s\n", + dev_info(dev, "%s IRQ %d, tx bndry %d, fw %s, MTRR %s, WC Enabled\n", mgp->msi_enabled ? "MSI" : "xPIC", pdev->irq, mgp->tx_boundary, mgp->fw_name, - (mgp->wc_enabled ? "Enabled" : "Disabled")); + (mgp->wc_cookie > 0 ? "Enabled" : "Disabled")); board_number++; return 0; @@ -4175,10 +4162,7 @@ abort_with_ioremap: iounmap(mgp->sram); abort_with_mtrr: -#ifdef CONFIG_MTRR - if (mgp->mtrr >= 0) - mtrr_del(mgp->mtrr, mgp->iomem_base, mgp->board_span); -#endif + arch_phys_wc_del(mgp->wc_cookie); dma_free_coherent(&pdev->dev, sizeof(*mgp->cmd), mgp->cmd, mgp->cmd_bus); @@ -4220,11 +4204,7 @@ static void myri10ge_remove(struct pci_dev *pdev) pci_restore_state(pdev); iounmap(mgp->sram); - -#ifdef CONFIG_MTRR - if (mgp->mtrr >= 0) - mtrr_del(mgp->mtrr, mgp->iomem_base, mgp->board_span); -#endif + arch_phys_wc_del(mgp->wc_cookie); myri10ge_free_slices(mgp); kfree(mgp->msix_vectors); dma_free_coherent(&pdev->dev, sizeof(*mgp->cmd), -- cgit v0.10.2 From cb24d01d217497fb32467de22d773655f47d3896 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 22 Apr 2015 10:04:23 -0300 Subject: perf trace: Enable events when doing system wide tracing and starting a workload commit f7aa222ff397 Author: Arnaldo Carvalho de Melo Date: Tue Feb 3 13:25:39 2015 -0300 perf trace: No need to enable evsels for workload started from perf The assumption was that whenever a workload is specified, the attr.enable_on_exec evsel flag would be set, but that is not happening when perf_record_opts.system_wide is set, for instance That resulted in both perf_evlist__enable() and attr.enable_on_exec being not called/set, which made the events to remain disabled while the workload runs, producing no output. Fix it, by calling perf_evlist__enable() in the 'trace' tool when forking and not targetting a workload started from trace v2: Test against !target__none(), as suggested by Namhyung Kim, that is what is used in perf_evsel__config() when deciding if the attr.enable_on_exec flag to be set. More work is needed to cover other cases such as opts->initial_delay. Acked-by: Jiri Olsa Cc: Adrian Hunter Cc: Borislav Petkov Cc: David Ahern Cc: Don Zickus Cc: Frederic Weisbecker Cc: Namhyung Kim Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-27z7169pvfxgj8upic636syv@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index e124741..8842218 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -2241,10 +2241,11 @@ static int trace__run(struct trace *trace, int argc, const char **argv) if (err < 0) goto out_error_mmap; + if (!target__none(&trace->opts.target)) + perf_evlist__enable(evlist); + if (forks) perf_evlist__start_workload(evlist); - else - perf_evlist__enable(evlist); trace->multiple_threads = evlist->threads->map[0] == -1 || evlist->threads->nr > 1 || -- cgit v0.10.2 From 02ac5421ddc634767c732f9b6a10a395a9ecfc4f Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 22 Apr 2015 11:11:57 -0300 Subject: perf trace: Disable events and drain events when forked workload ends We were not checking in the inner event processing loop if the forked workload had finished, which, on a busy system, may make it take a long time trying to drain events, entering a seemingly neverending loop, waiting for the system to get idle enough to make it drain the buffers. Fix it by disabling the events when 'done' is true, in the inner loop, to start draining what is in the buffers. Now: [root@ssdandy ~]# time trace --filter-pids 14003 -a sleep 1 | tail 996.748 ( 0.002 ms): sh/30296 rt_sigprocmask(how: SETMASK, nset: 0x7ffc83418160, sigsetsize: 8) = 0 996.751 ( 0.002 ms): sh/30296 rt_sigprocmask(how: BLOCK, nset: 0x7ffc834181f0, oset: 0x7ffc83418270, sigsetsize: 8) = 0 996.755 ( 0.002 ms): sh/30296 rt_sigaction(sig: INT, act: 0x7ffc83417f50, oact: 0x7ffc83417ff0, sigsetsize: 8) = 0 1004.543 ( 0.362 ms): tail/30198 ... [continued]: read()) = 4096 1004.548 ( 7.791 ms): sh/30296 wait4(upid: -1, stat_addr: 0x7ffc834181a0) ... 1004.975 ( 0.427 ms): tail/30198 read(buf: 0x7633f0, count: 8192) = 4096 1005.390 ( 0.410 ms): tail/30198 read(buf: 0x765410, count: 8192) = 4096 1005.743 ( 0.348 ms): tail/30198 read(buf: 0x7633f0, count: 8192) = 4096 1006.197 ( 0.449 ms): tail/30198 read(buf: 0x765410, count: 8192) = 4096 1006.492 ( 0.290 ms): tail/30198 read(buf: 0x7633f0, count: 8192) = 4096 real 0m1.219s user 0m0.704s sys 0m0.331s [root@ssdandy ~]# Reported-by: Michael Petlan Suggested-by: Jiri Olsa Acked-by: Jiri Olsa Cc: Adrian Hunter Cc: Borislav Petkov Cc: David Ahern Cc: Don Zickus Cc: Frederic Weisbecker Cc: Namhyung Kim Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-p6kpn1b26qcbe47pufpw0tex@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index 8842218..e122970 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -2273,6 +2273,11 @@ next_event: if (interrupted) goto out_disable; + + if (done && !draining) { + perf_evlist__disable(evlist); + draining = true; + } } } -- cgit v0.10.2 From 6145c259cd454bcb7a1288f7bbb7b4fbc18175dd Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 23 Apr 2015 14:40:37 +0100 Subject: perf kmem: Consistently use PRIu64 for printing u64 values MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Building the perf tool for 32-bit ARM results in the following build error due to a combination of an incorrect conversion specifier and compiling with -Werror: builtin-kmem.c: In function ‘print_page_summary’: builtin-kmem.c:644:9: error: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 3 has type ‘u64’ [-Werror=format=] nr_alloc_freed, (total_alloc_freed_bytes) / 1024); ^ builtin-kmem.c:647:9: error: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 3 has type ‘u64’ [-Werror=format=] (total_page_alloc_bytes - total_alloc_freed_bytes) / 1024); ^ cc1: all warnings being treated as errors This patch fixes the problem by consistently using PRIu64 for printing out u64 values. Signed-off-by: Will Deacon Cc: David Ahern Cc: Jiri Olsa Cc: Joonsoo Kim Cc: Minchan Kim Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1429796437-1790-1-git-send-email-will.deacon@arm.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index 63ea013..a1915b4 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -640,9 +640,9 @@ static void print_page_summary(void) nr_page_frees, total_page_free_bytes / 1024); printf("\n"); - printf("%-30s: %'16lu [ %'16"PRIu64" KB ]\n", "Total alloc+freed requests", + printf("%-30s: %'16"PRIu64" [ %'16"PRIu64" KB ]\n", "Total alloc+freed requests", nr_alloc_freed, (total_alloc_freed_bytes) / 1024); - printf("%-30s: %'16lu [ %'16"PRIu64" KB ]\n", "Total alloc-only requests", + printf("%-30s: %'16"PRIu64" [ %'16"PRIu64" KB ]\n", "Total alloc-only requests", nr_page_allocs - nr_alloc_freed, (total_page_alloc_bytes - total_alloc_freed_bytes) / 1024); printf("%-30s: %'16lu [ %'16"PRIu64" KB ]\n", "Total free-only requests", -- cgit v0.10.2 From de28c15daf60e9625bece22f13a091fac8d05f1d Mon Sep 17 00:00:00 2001 From: Bobby Powers Date: Tue, 21 Apr 2015 19:19:41 -0400 Subject: tools lib api: Undefine _FORTIFY_SOURCE before setting it Some toolchains (like Hardened Gentoo) define _FORTIFY_SOURCE in the built-in, default args. This causes perf builds to fail with: :0:0: error: "_FORTIFY_SOURCE" redefined [-Werror] : note: this is the location of the previous definition cc1: all warnings being treated as errors To avoid this, undefine _FORTIFY_SOURCE before (possibly re-)defining it in tools/lib/api. v2 applies cleanly on top of already pulled kbuild changes for 4.1-rc1. Signed-off-by: Bobby Powers Acked-by: Jiri Olsa Cc: Dirk Gouders Cc: Michal Marek Cc: Paul Mackerras Cc: Peter Zijlstra Cc: linux-kbuild@vger.kernel.org Link: http://lkml.kernel.org/r/1429658381-3039-1-git-send-email-bobbypowers@gmail.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/lib/api/Makefile b/tools/lib/api/Makefile index d8fe29f..8bd9606 100644 --- a/tools/lib/api/Makefile +++ b/tools/lib/api/Makefile @@ -16,7 +16,7 @@ MAKEFLAGS += --no-print-directory LIBFILE = $(OUTPUT)libapi.a CFLAGS := $(EXTRA_WARNINGS) $(EXTRA_CFLAGS) -CFLAGS += -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -D_FORTIFY_SOURCE=2 -fPIC +CFLAGS += -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 -fPIC CFLAGS += -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 RM = rm -f -- cgit v0.10.2 From 74d6ea52aeef0236242221c6eff6d892565c5a92 Mon Sep 17 00:00:00 2001 From: Bard Liao Date: Fri, 24 Apr 2015 15:19:29 +0800 Subject: ASoC: rt5677: add register patch for PLL The PLL output will be unstable in some cases. We can fix it by setting some registers. Signed-off-by: Bard Liao Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/codecs/rt5677.c b/sound/soc/codecs/rt5677.c index 5d0bb87..c6d4e8f 100644 --- a/sound/soc/codecs/rt5677.c +++ b/sound/soc/codecs/rt5677.c @@ -62,6 +62,9 @@ static const struct reg_default init_list[] = { {RT5677_PR_BASE + 0x1e, 0x0000}, {RT5677_PR_BASE + 0x12, 0x0eaa}, {RT5677_PR_BASE + 0x14, 0x018a}, + {RT5677_PR_BASE + 0x15, 0x0490}, + {RT5677_PR_BASE + 0x38, 0x0f71}, + {RT5677_PR_BASE + 0x39, 0x0f71}, }; #define RT5677_INIT_REG_LEN ARRAY_SIZE(init_list) -- cgit v0.10.2 From 3168c201f7ca333d12f80f8d98bbbe3a33746f8b Mon Sep 17 00:00:00 2001 From: "Fang, Yang A" Date: Thu, 23 Apr 2015 16:35:17 -0700 Subject: ASoC: rt5645: Add ACPI match ID This patch adds the ACPI match ID for rt5645/5650 codec Signed-off-by: Fang, Yang A Signed-off-by: Mark Brown diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index c9a4c5b..e16724a 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -2618,6 +2619,15 @@ static const struct i2c_device_id rt5645_i2c_id[] = { }; MODULE_DEVICE_TABLE(i2c, rt5645_i2c_id); +#ifdef CONFIG_ACPI +static struct acpi_device_id rt5645_acpi_match[] = { + { "10EC5645", 0 }, + { "10EC5650", 0 }, + {}, +}; +MODULE_DEVICE_TABLE(acpi, rt5645_acpi_match); +#endif + static int rt5645_i2c_probe(struct i2c_client *i2c, const struct i2c_device_id *id) { @@ -2834,6 +2844,7 @@ static struct i2c_driver rt5645_i2c_driver = { .driver = { .name = "rt5645", .owner = THIS_MODULE, + .acpi_match_table = ACPI_PTR(rt5645_acpi_match), }, .probe = rt5645_i2c_probe, .remove = rt5645_i2c_remove, -- cgit v0.10.2 From 845704a535e9b3c76448f52af1b70e4422ea03fd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 23 Apr 2015 10:42:39 -0700 Subject: tcp: avoid looping in tcp_send_fin() Presence of an unbound loop in tcp_send_fin() had always been hard to explain when analyzing crash dumps involving gigantic dying processes with millions of sockets. Lets try a different strategy : In case of memory pressure, try to add the FIN flag to last packet in write queue, even if packet was already sent. TCP stack will be able to deliver this FIN after a timeout event. Note that this FIN being delivered by a retransmit, it also carries a Push flag given our current implementation. By checking sk_under_memory_pressure(), we anticipate that cooking many FIN packets might deplete tcp memory. In the case we could not allocate a packet, even with __GFP_WAIT allocation, then not sending a FIN seems quite reasonable if it allows to get rid of this socket, free memory, and not block the process from eventually doing other useful work. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 2ade67b..a369e8a 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2814,7 +2814,8 @@ begin_fwd: /* We allow to exceed memory limits for FIN packets to expedite * connection tear down and (memory) recovery. - * Otherwise tcp_send_fin() could loop forever. + * Otherwise tcp_send_fin() could be tempted to either delay FIN + * or even be forced to close flow without any FIN. */ static void sk_forced_wmem_schedule(struct sock *sk, int size) { @@ -2827,33 +2828,40 @@ static void sk_forced_wmem_schedule(struct sock *sk, int size) sk_memory_allocated_add(sk, amt, &status); } -/* Send a fin. The caller locks the socket for us. This cannot be - * allowed to fail queueing a FIN frame under any circumstances. +/* Send a FIN. The caller locks the socket for us. + * We should try to send a FIN packet really hard, but eventually give up. */ void tcp_send_fin(struct sock *sk) { + struct sk_buff *skb, *tskb = tcp_write_queue_tail(sk); struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *skb = tcp_write_queue_tail(sk); - int mss_now; - /* Optimization, tack on the FIN if we have a queue of - * unsent frames. But be careful about outgoing SACKS - * and IP options. + /* Optimization, tack on the FIN if we have one skb in write queue and + * this skb was not yet sent, or we are under memory pressure. + * Note: in the latter case, FIN packet will be sent after a timeout, + * as TCP stack thinks it has already been transmitted. */ - mss_now = tcp_current_mss(sk); - - if (tcp_send_head(sk)) { - TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_FIN; - TCP_SKB_CB(skb)->end_seq++; + if (tskb && (tcp_send_head(sk) || sk_under_memory_pressure(sk))) { +coalesce: + TCP_SKB_CB(tskb)->tcp_flags |= TCPHDR_FIN; + TCP_SKB_CB(tskb)->end_seq++; tp->write_seq++; + if (!tcp_send_head(sk)) { + /* This means tskb was already sent. + * Pretend we included the FIN on previous transmit. + * We need to set tp->snd_nxt to the value it would have + * if FIN had been sent. This is because retransmit path + * does not change tp->snd_nxt. + */ + tp->snd_nxt++; + return; + } } else { - /* Socket is locked, keep trying until memory is available. */ - for (;;) { - skb = alloc_skb_fclone(MAX_TCP_HEADER, - sk->sk_allocation); - if (skb) - break; - yield(); + skb = alloc_skb_fclone(MAX_TCP_HEADER, sk->sk_allocation); + if (unlikely(!skb)) { + if (tskb) + goto coalesce; + return; } skb_reserve(skb, MAX_TCP_HEADER); sk_forced_wmem_schedule(sk, skb->truesize); @@ -2862,7 +2870,7 @@ void tcp_send_fin(struct sock *sk) TCPHDR_ACK | TCPHDR_FIN); tcp_queue_skb(sk, skb); } - __tcp_push_pending_frames(sk, mss_now, TCP_NAGLE_OFF); + __tcp_push_pending_frames(sk, tcp_current_mss(sk), TCP_NAGLE_OFF); } /* We get here when a process closes a file descriptor (either due to -- cgit v0.10.2 From e580267df97eda407c525dbaee5430e0d51a0edb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Thu, 23 Apr 2015 20:56:29 +0200 Subject: bgmac: fix requests for extra polling calls from NAPI MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After d75b1ade567f ("net: less interrupt masking in NAPI") polling function has to return whole budget when it wants NAPI to call it again. Signed-off-by: Rafał Miłecki Cc: Felix Fietkau Fixes: eb64e2923a886 ("bgmac: leave interrupts disabled as long as there is work to do") Acked-by: Felix Fietkau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index de77d3a..21e3c38 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -1260,7 +1260,7 @@ static int bgmac_poll(struct napi_struct *napi, int weight) /* Poll again if more events arrived in the meantime */ if (bgmac_read(bgmac, BGMAC_INT_STATUS) & (BGMAC_IS_TX0 | BGMAC_IS_RX)) - return handled; + return weight; if (handled < weight) { napi_complete(napi); -- cgit v0.10.2 From 1d8dc3d3c8f1d8ee1da9d54c5d7c8694419ade42 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 23 Apr 2015 16:38:43 +0200 Subject: rhashtable: don't attempt to grow when at max_size The conversion of mac80211's station table to rhashtable had a bug that I found by accident in code review, that hadn't been found as rhashtable apparently managed to have a maximum hash chain length of one (!) in all our testing. In order to test the bug and verify the fix I set my rhashtable's max_size very low (4) in order to force getting hash collisions. At that point, rhashtable WARNed in rhashtable_insert_rehash() but didn't actually reject the hash table insertion. This caused it to lose insertions - my master list of stations would have 9 entries, but the rhashtable only had 5. This may warrant a deeper look, but that WARN_ON() just shouldn't happen. Fix this by not returning true from rht_grow_above_100() when the rhashtable's max_size has been reached - in this case the user is explicitly configuring it to be at most that big, so even if it's now above 100% it shouldn't attempt to resize. This fixes the "lost insertion" issue and consequently allows my code to display its error (and verify my fix for it.) Signed-off-by: Johannes Berg Acked-by: Thomas Graf Signed-off-by: David S. Miller diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h index e23d242..dbcbcc5 100644 --- a/include/linux/rhashtable.h +++ b/include/linux/rhashtable.h @@ -282,7 +282,8 @@ static inline bool rht_shrink_below_30(const struct rhashtable *ht, static inline bool rht_grow_above_100(const struct rhashtable *ht, const struct bucket_table *tbl) { - return atomic_read(&ht->nelems) > tbl->size; + return atomic_read(&ht->nelems) > tbl->size && + (!ht->p.max_size || tbl->size < ht->p.max_size); } /* The bucket lock is selected based on the hash and protects mutations -- cgit v0.10.2 From b357a364c57c940ddb932224542494363df37378 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 23 Apr 2015 18:03:44 -0700 Subject: inet: fix possible panic in reqsk_queue_unlink() [ 3897.923145] BUG: unable to handle kernel NULL pointer dereference at 0000000000000080 [ 3897.931025] IP: [] reqsk_timer_handler+0x1a6/0x243 There is a race when reqsk_timer_handler() and tcp_check_req() call inet_csk_reqsk_queue_unlink() on the same req at the same time. Before commit fa76ce7328b2 ("inet: get rid of central tcp/dccp listener timer"), listener spinlock was held and race could not happen. To solve this bug, we change reqsk_queue_unlink() to not assume req must be found, and we return a status, to conditionally release a refcount on the request sock. This also means tcp_check_req() in non fastopen case might or not consume req refcount, so tcp_v6_hnd_req() & tcp_v4_hnd_req() have to properly handle this. (Same remark for dccp_check_req() and its callers) inet_csk_reqsk_queue_drop() is now too big to be inlined, as it is called 4 times in tcp and 3 times in dccp. Fixes: fa76ce7328b2 ("inet: get rid of central tcp/dccp listener timer") Signed-off-by: Eric Dumazet Reported-by: Yuchung Cheng Signed-off-by: David S. Miller diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 7b5887c..48a81582 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -279,12 +279,6 @@ static inline void inet_csk_reqsk_queue_add(struct sock *sk, void inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req, unsigned long timeout); -static inline void inet_csk_reqsk_queue_removed(struct sock *sk, - struct request_sock *req) -{ - reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req); -} - static inline void inet_csk_reqsk_queue_added(struct sock *sk, const unsigned long timeout) { @@ -306,19 +300,7 @@ static inline int inet_csk_reqsk_queue_is_full(const struct sock *sk) return reqsk_queue_is_full(&inet_csk(sk)->icsk_accept_queue); } -static inline void inet_csk_reqsk_queue_unlink(struct sock *sk, - struct request_sock *req) -{ - reqsk_queue_unlink(&inet_csk(sk)->icsk_accept_queue, req); -} - -static inline void inet_csk_reqsk_queue_drop(struct sock *sk, - struct request_sock *req) -{ - inet_csk_reqsk_queue_unlink(sk, req); - inet_csk_reqsk_queue_removed(sk, req); - reqsk_put(req); -} +void inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req); void inet_csk_destroy_sock(struct sock *sk); void inet_csk_prepare_forced_close(struct sock *sk); diff --git a/include/net/request_sock.h b/include/net/request_sock.h index fe41f3c..9f4265c 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -212,24 +212,6 @@ static inline int reqsk_queue_empty(struct request_sock_queue *queue) return queue->rskq_accept_head == NULL; } -static inline void reqsk_queue_unlink(struct request_sock_queue *queue, - struct request_sock *req) -{ - struct listen_sock *lopt = queue->listen_opt; - struct request_sock **prev; - - spin_lock(&queue->syn_wait_lock); - - prev = &lopt->syn_table[req->rsk_hash]; - while (*prev != req) - prev = &(*prev)->dl_next; - *prev = req->dl_next; - - spin_unlock(&queue->syn_wait_lock); - if (del_timer(&req->rsk_timer)) - reqsk_put(req); -} - static inline void reqsk_queue_add(struct request_sock_queue *queue, struct request_sock *req, struct sock *parent, diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 2b4f21d..ccf4c56 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -453,7 +453,8 @@ static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) iph->saddr, iph->daddr); if (req) { nsk = dccp_check_req(sk, skb, req); - reqsk_put(req); + if (!nsk) + reqsk_put(req); return nsk; } nsk = inet_lookup_established(sock_net(sk), &dccp_hashinfo, diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 9d05510..5165571 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -301,7 +301,8 @@ static struct sock *dccp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) &iph->daddr, inet6_iif(skb)); if (req) { nsk = dccp_check_req(sk, skb, req); - reqsk_put(req); + if (!nsk) + reqsk_put(req); return nsk; } nsk = __inet6_lookup_established(sock_net(sk), &dccp_hashinfo, diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 5f56666..30addee 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -186,8 +186,7 @@ struct sock *dccp_check_req(struct sock *sk, struct sk_buff *skb, if (child == NULL) goto listen_overflow; - inet_csk_reqsk_queue_unlink(sk, req); - inet_csk_reqsk_queue_removed(sk, req); + inet_csk_reqsk_queue_drop(sk, req); inet_csk_reqsk_queue_add(sk, req, child); out: return child; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 5c3dd62..8976ca4 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -564,6 +564,40 @@ int inet_rtx_syn_ack(struct sock *parent, struct request_sock *req) } EXPORT_SYMBOL(inet_rtx_syn_ack); +/* return true if req was found in the syn_table[] */ +static bool reqsk_queue_unlink(struct request_sock_queue *queue, + struct request_sock *req) +{ + struct listen_sock *lopt = queue->listen_opt; + struct request_sock **prev; + bool found = false; + + spin_lock(&queue->syn_wait_lock); + + for (prev = &lopt->syn_table[req->rsk_hash]; *prev != NULL; + prev = &(*prev)->dl_next) { + if (*prev == req) { + *prev = req->dl_next; + found = true; + break; + } + } + + spin_unlock(&queue->syn_wait_lock); + if (del_timer(&req->rsk_timer)) + reqsk_put(req); + return found; +} + +void inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) +{ + if (reqsk_queue_unlink(&inet_csk(sk)->icsk_accept_queue, req)) { + reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req); + reqsk_put(req); + } +} +EXPORT_SYMBOL(inet_csk_reqsk_queue_drop); + static void reqsk_timer_handler(unsigned long data) { struct request_sock *req = (struct request_sock *)data; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 3571f2b..fc1c658 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1348,7 +1348,8 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) req = inet_csk_search_req(sk, th->source, iph->saddr, iph->daddr); if (req) { nsk = tcp_check_req(sk, skb, req, false); - reqsk_put(req); + if (!nsk) + reqsk_put(req); return nsk; } diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 63d6311..e5d7649 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -755,10 +755,11 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, if (!child) goto listen_overflow; - inet_csk_reqsk_queue_unlink(sk, req); - inet_csk_reqsk_queue_removed(sk, req); - + inet_csk_reqsk_queue_drop(sk, req); inet_csk_reqsk_queue_add(sk, req, child); + /* Warning: caller must not call reqsk_put(req); + * child stole last reference on it. + */ return child; listen_overflow: diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index ad51df8..b6575d6 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -946,7 +946,8 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk, struct sk_buff *skb) &ipv6_hdr(skb)->daddr, tcp_v6_iif(skb)); if (req) { nsk = tcp_check_req(sk, skb, req, false); - reqsk_put(req); + if (!nsk) + reqsk_put(req); return nsk; } nsk = __inet6_lookup_established(sock_net(sk), &tcp_hashinfo, -- cgit v0.10.2 From 4ad1f4300e3bddf63109aa63cfb2d37e8585ecc7 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Tue, 14 Apr 2015 13:49:33 -0400 Subject: perf kmem: Fix compiles on RHEL6/OL6 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 0d68bc92c48 breaks compiles on RHEL6/OL6: cc1: warnings being treated as errors builtin-kmem.c: In function ‘search_page_alloc_stat’: builtin-kmem.c:322: error: declaration of ‘stat’ shadows a global declaration node = &parent->rb_left; /usr/include/sys/stat.h:455: error: shadowed declaration is here builtin-kmem.c: In function ‘perf_evsel__process_page_alloc_event’: builtin-kmem.c:378: error: declaration of ‘stat’ shadows a global declaration /usr/include/sys/stat.h:455: error: shadowed declaration is here builtin-kmem.c: In function ‘perf_evsel__process_page_free_event’: builtin-kmem.c:431: error: declaration of ‘stat’ shadows a global declaration /usr/include/sys/stat.h:455: error: shadowed declaration is here Rename local variable to pstat to avoid the name conflict. Signed-off-by: David Ahern Link: http://lkml.kernel.org/r/1429033773-31383-1-git-send-email-david.ahern@oracle.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index a1915b4..1634186 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -319,7 +319,7 @@ static int page_stat_cmp(struct page_stat *a, struct page_stat *b) return 0; } -static struct page_stat *search_page_alloc_stat(struct page_stat *stat, bool create) +static struct page_stat *search_page_alloc_stat(struct page_stat *pstat, bool create) { struct rb_node **node = &page_alloc_tree.rb_node; struct rb_node *parent = NULL; @@ -331,7 +331,7 @@ static struct page_stat *search_page_alloc_stat(struct page_stat *stat, bool cre parent = *node; data = rb_entry(*node, struct page_stat, node); - cmp = page_stat_cmp(data, stat); + cmp = page_stat_cmp(data, pstat); if (cmp < 0) node = &parent->rb_left; else if (cmp > 0) @@ -345,10 +345,10 @@ static struct page_stat *search_page_alloc_stat(struct page_stat *stat, bool cre data = zalloc(sizeof(*data)); if (data != NULL) { - data->page = stat->page; - data->order = stat->order; - data->gfp_flags = stat->gfp_flags; - data->migrate_type = stat->migrate_type; + data->page = pstat->page; + data->order = pstat->order; + data->gfp_flags = pstat->gfp_flags; + data->migrate_type = pstat->migrate_type; rb_link_node(&data->node, parent, node); rb_insert_color(&data->node, &page_alloc_tree); @@ -375,7 +375,7 @@ static int perf_evsel__process_page_alloc_event(struct perf_evsel *evsel, unsigned int migrate_type = perf_evsel__intval(evsel, sample, "migratetype"); u64 bytes = kmem_page_size << order; - struct page_stat *stat; + struct page_stat *pstat; struct page_stat this = { .order = order, .gfp_flags = gfp_flags, @@ -401,21 +401,21 @@ static int perf_evsel__process_page_alloc_event(struct perf_evsel *evsel, * This is to find the current page (with correct gfp flags and * migrate type) at free event. */ - stat = search_page(page, true); - if (stat == NULL) + pstat = search_page(page, true); + if (pstat == NULL) return -ENOMEM; - stat->order = order; - stat->gfp_flags = gfp_flags; - stat->migrate_type = migrate_type; + pstat->order = order; + pstat->gfp_flags = gfp_flags; + pstat->migrate_type = migrate_type; this.page = page; - stat = search_page_alloc_stat(&this, true); - if (stat == NULL) + pstat = search_page_alloc_stat(&this, true); + if (pstat == NULL) return -ENOMEM; - stat->nr_alloc++; - stat->alloc_bytes += bytes; + pstat->nr_alloc++; + pstat->alloc_bytes += bytes; order_stats[order][migrate_type]++; @@ -428,7 +428,7 @@ static int perf_evsel__process_page_free_event(struct perf_evsel *evsel, u64 page; unsigned int order = perf_evsel__intval(evsel, sample, "order"); u64 bytes = kmem_page_size << order; - struct page_stat *stat; + struct page_stat *pstat; struct page_stat this = { .order = order, }; @@ -441,8 +441,8 @@ static int perf_evsel__process_page_free_event(struct perf_evsel *evsel, nr_page_frees++; total_page_free_bytes += bytes; - stat = search_page(page, false); - if (stat == NULL) { + pstat = search_page(page, false); + if (pstat == NULL) { pr_debug2("missing free at page %"PRIx64" (order: %d)\n", page, order); @@ -453,18 +453,18 @@ static int perf_evsel__process_page_free_event(struct perf_evsel *evsel, } this.page = page; - this.gfp_flags = stat->gfp_flags; - this.migrate_type = stat->migrate_type; + this.gfp_flags = pstat->gfp_flags; + this.migrate_type = pstat->migrate_type; - rb_erase(&stat->node, &page_tree); - free(stat); + rb_erase(&pstat->node, &page_tree); + free(pstat); - stat = search_page_alloc_stat(&this, false); - if (stat == NULL) + pstat = search_page_alloc_stat(&this, false); + if (pstat == NULL) return -ENOENT; - stat->nr_free++; - stat->free_bytes += bytes; + pstat->nr_free++; + pstat->free_bytes += bytes; return 0; } -- cgit v0.10.2 From 410ceb8f2f1d4edeb02d229ef192e76602005b8b Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Fri, 24 Apr 2015 10:45:16 +0900 Subject: tools lib traceevent: Fix build failure on 32-bit arch In my i386 build, it failed like this: CC event-parse.o event-parse.c: In function 'print_str_arg': event-parse.c:3868:5: warning: format '%lu' expects argument of type 'long unsigned int', but argument 3 has type 'uint64_t' [-Wformat] Signed-off-by: Namhyung Kim Acked-by: Javi Merino Link: http://lkml.kernel.org/r/20150424020218.GF1905@sejong Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c index 12a7e2a..aa21bd5 100644 --- a/tools/lib/traceevent/event-parse.c +++ b/tools/lib/traceevent/event-parse.c @@ -3865,7 +3865,7 @@ static void print_str_arg(struct trace_seq *s, void *data, int size, } else if (el_size == 4) { trace_seq_printf(s, "%u", *(uint32_t *)num); } else if (el_size == 8) { - trace_seq_printf(s, "%lu", *(uint64_t *)num); + trace_seq_printf(s, "%"PRIu64, *(uint64_t *)num); } else { trace_seq_printf(s, "BAD SIZE:%d 0x%x", el_size, *(uint8_t *)num); -- cgit v0.10.2 From 2d6c9091ab7630dfcf34417c6683ce4764d7d40a Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Wed, 22 Apr 2015 13:06:54 -0400 Subject: net: mdio-gpio: support access that may sleep Some systems using mdio-gpio may use gpio on message based busses, which require sleeping (e.g. gpio from an I2C I/O expander). Since this driver does not use IRQ handler, it is safe to use the _cansleep suffixed gpio accessors. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/phy/mdio-gpio.c b/drivers/net/phy/mdio-gpio.c index 49ce7ec..c9cb486c 100644 --- a/drivers/net/phy/mdio-gpio.c +++ b/drivers/net/phy/mdio-gpio.c @@ -80,7 +80,8 @@ static void mdio_dir(struct mdiobb_ctrl *ctrl, int dir) * assume the pin serves as pull-up. If direction is * output, the default value is high. */ - gpio_set_value(bitbang->mdo, 1 ^ bitbang->mdo_active_low); + gpio_set_value_cansleep(bitbang->mdo, + 1 ^ bitbang->mdo_active_low); return; } @@ -96,7 +97,8 @@ static int mdio_get(struct mdiobb_ctrl *ctrl) struct mdio_gpio_info *bitbang = container_of(ctrl, struct mdio_gpio_info, ctrl); - return gpio_get_value(bitbang->mdio) ^ bitbang->mdio_active_low; + return gpio_get_value_cansleep(bitbang->mdio) ^ + bitbang->mdio_active_low; } static void mdio_set(struct mdiobb_ctrl *ctrl, int what) @@ -105,9 +107,11 @@ static void mdio_set(struct mdiobb_ctrl *ctrl, int what) container_of(ctrl, struct mdio_gpio_info, ctrl); if (bitbang->mdo) - gpio_set_value(bitbang->mdo, what ^ bitbang->mdo_active_low); + gpio_set_value_cansleep(bitbang->mdo, + what ^ bitbang->mdo_active_low); else - gpio_set_value(bitbang->mdio, what ^ bitbang->mdio_active_low); + gpio_set_value_cansleep(bitbang->mdio, + what ^ bitbang->mdio_active_low); } static void mdc_set(struct mdiobb_ctrl *ctrl, int what) @@ -115,7 +119,7 @@ static void mdc_set(struct mdiobb_ctrl *ctrl, int what) struct mdio_gpio_info *bitbang = container_of(ctrl, struct mdio_gpio_info, ctrl); - gpio_set_value(bitbang->mdc, what ^ bitbang->mdc_active_low); + gpio_set_value_cansleep(bitbang->mdc, what ^ bitbang->mdc_active_low); } static struct mdiobb_ops mdio_gpio_ops = { -- cgit v0.10.2 From 4c4ed0748f82e26d8b884383e6737cf5a861ed6f Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 14 Apr 2015 16:44:14 +0200 Subject: netfilter: nf_tables: fix wrong length for jump/goto verdicts NFT_JUMP/GOTO erronously sets length to sizeof(void *). We then allocate insufficient memory when such element is added to a vmap. Suggested-by: Patrick McHardy Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 78af83b..ad9d11f 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -4340,7 +4340,6 @@ static int nft_verdict_init(const struct nft_ctx *ctx, struct nft_data *data, case NFT_CONTINUE: case NFT_BREAK: case NFT_RETURN: - desc->len = sizeof(data->verdict); break; case NFT_JUMP: case NFT_GOTO: @@ -4355,10 +4354,10 @@ static int nft_verdict_init(const struct nft_ctx *ctx, struct nft_data *data, chain->use++; data->verdict.chain = chain; - desc->len = sizeof(data); break; } + desc->len = sizeof(data->verdict); desc->type = NFT_DATA_VERDICT; return 0; } -- cgit v0.10.2 From 547c4b547e07dcc60874b6ef6252dd49ff74aec1 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 20 Apr 2015 12:35:47 +0200 Subject: netfilter: bridge: fix NULL deref in physin/out ifindex helpers Might not have an outdev yet. We'll oops when iface goes down while skbs are still nfqueue'd: RIP: 0010:[] [] dev_cmp+0x4f/0x80 nfqnl_rcv_dev_event+0xe2/0x150 notifier_call_chain+0x53/0xa0 Fixes: c737b7c4510026 ("netfilter: bridge: add helpers for fetching physin/outdev") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h index ab8f76d..f2fdb5a 100644 --- a/include/linux/netfilter_bridge.h +++ b/include/linux/netfilter_bridge.h @@ -39,12 +39,24 @@ static inline void br_drop_fake_rtable(struct sk_buff *skb) static inline int nf_bridge_get_physinif(const struct sk_buff *skb) { - return skb->nf_bridge ? skb->nf_bridge->physindev->ifindex : 0; + struct nf_bridge_info *nf_bridge; + + if (skb->nf_bridge == NULL) + return 0; + + nf_bridge = skb->nf_bridge; + return nf_bridge->physindev ? nf_bridge->physindev->ifindex : 0; } static inline int nf_bridge_get_physoutif(const struct sk_buff *skb) { - return skb->nf_bridge ? skb->nf_bridge->physoutdev->ifindex : 0; + struct nf_bridge_info *nf_bridge; + + if (skb->nf_bridge == NULL) + return 0; + + nf_bridge = skb->nf_bridge; + return nf_bridge->physoutdev ? nf_bridge->physoutdev->ifindex : 0; } static inline struct net_device * -- cgit v0.10.2 From 10a50f1ab5f06c9a3ee5ece3ec52e607ed53c79f Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Wed, 15 Apr 2015 11:14:11 +0300 Subject: genirq: Set IRQCHIP_SKIP_SET_WAKE flag for dummy_irq_chip Without this system suspend is broken on systems that have drivers calling enable/disable_irq_wake() for interrupts based off the dummy irq hook. (e.g. drivers/gpio/gpio-pcf857x.c) Signed-off-by: Roger Quadros Cc: Cc: Cc: Cc: Gregory Clement Link: http://lkml.kernel.org/r/552E1DD3.4040106@ti.com Signed-off-by: Thomas Gleixner diff --git a/kernel/irq/dummychip.c b/kernel/irq/dummychip.c index 988dc58..2feb6fe 100644 --- a/kernel/irq/dummychip.c +++ b/kernel/irq/dummychip.c @@ -57,5 +57,6 @@ struct irq_chip dummy_irq_chip = { .irq_ack = noop, .irq_mask = noop, .irq_unmask = noop, + .flags = IRQCHIP_SKIP_SET_WAKE, }; EXPORT_SYMBOL_GPL(dummy_irq_chip); -- cgit v0.10.2 From 1dcc73d7bb0429994c54d33b40c5fb82b741a791 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Wed, 22 Apr 2015 18:20:04 +0100 Subject: irqchip: gic: Drop support for gic_arch_extn Now that the users of gic_arch_extn have been fixed, drop the "feature" for good. This leads to the removal of some now useless locking. Signed-off-by: Marc Zyngier Cc: linux-arm-kernel@lists.infradead.org Cc: Jason Cooper Signed-off-by: Thomas Gleixner diff --git a/drivers/irqchip/irq-gic.c b/drivers/irqchip/irq-gic.c index a6ce347..dd98914 100644 --- a/drivers/irqchip/irq-gic.c +++ b/drivers/irqchip/irq-gic.c @@ -80,19 +80,6 @@ static DEFINE_RAW_SPINLOCK(irq_controller_lock); #define NR_GIC_CPU_IF 8 static u8 gic_cpu_map[NR_GIC_CPU_IF] __read_mostly; -/* - * Supported arch specific GIC irq extension. - * Default make them NULL. - */ -struct irq_chip gic_arch_extn = { - .irq_eoi = NULL, - .irq_mask = NULL, - .irq_unmask = NULL, - .irq_retrigger = NULL, - .irq_set_type = NULL, - .irq_set_wake = NULL, -}; - #ifndef MAX_GIC_NR #define MAX_GIC_NR 1 #endif @@ -165,34 +152,16 @@ static int gic_peek_irq(struct irq_data *d, u32 offset) static void gic_mask_irq(struct irq_data *d) { - unsigned long flags; - - raw_spin_lock_irqsave(&irq_controller_lock, flags); gic_poke_irq(d, GIC_DIST_ENABLE_CLEAR); - if (gic_arch_extn.irq_mask) - gic_arch_extn.irq_mask(d); - raw_spin_unlock_irqrestore(&irq_controller_lock, flags); } static void gic_unmask_irq(struct irq_data *d) { - unsigned long flags; - - raw_spin_lock_irqsave(&irq_controller_lock, flags); - if (gic_arch_extn.irq_unmask) - gic_arch_extn.irq_unmask(d); gic_poke_irq(d, GIC_DIST_ENABLE_SET); - raw_spin_unlock_irqrestore(&irq_controller_lock, flags); } static void gic_eoi_irq(struct irq_data *d) { - if (gic_arch_extn.irq_eoi) { - raw_spin_lock(&irq_controller_lock); - gic_arch_extn.irq_eoi(d); - raw_spin_unlock(&irq_controller_lock); - } - writel_relaxed(gic_irq(d), gic_cpu_base(d) + GIC_CPU_EOI); } @@ -249,8 +218,6 @@ static int gic_set_type(struct irq_data *d, unsigned int type) { void __iomem *base = gic_dist_base(d); unsigned int gicirq = gic_irq(d); - unsigned long flags; - int ret; /* Interrupt configuration for SGIs can't be changed */ if (gicirq < 16) @@ -261,25 +228,7 @@ static int gic_set_type(struct irq_data *d, unsigned int type) type != IRQ_TYPE_EDGE_RISING) return -EINVAL; - raw_spin_lock_irqsave(&irq_controller_lock, flags); - - if (gic_arch_extn.irq_set_type) - gic_arch_extn.irq_set_type(d, type); - - ret = gic_configure_irq(gicirq, type, base, NULL); - - raw_spin_unlock_irqrestore(&irq_controller_lock, flags); - - return ret; -} - -static int gic_retrigger(struct irq_data *d) -{ - if (gic_arch_extn.irq_retrigger) - return gic_arch_extn.irq_retrigger(d); - - /* the genirq layer expects 0 if we can't retrigger in hardware */ - return 0; + return gic_configure_irq(gicirq, type, base, NULL); } #ifdef CONFIG_SMP @@ -310,21 +259,6 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *mask_val, } #endif -#ifdef CONFIG_PM -static int gic_set_wake(struct irq_data *d, unsigned int on) -{ - int ret = -ENXIO; - - if (gic_arch_extn.irq_set_wake) - ret = gic_arch_extn.irq_set_wake(d, on); - - return ret; -} - -#else -#define gic_set_wake NULL -#endif - static void __exception_irq_entry gic_handle_irq(struct pt_regs *regs) { u32 irqstat, irqnr; @@ -383,11 +317,9 @@ static struct irq_chip gic_chip = { .irq_unmask = gic_unmask_irq, .irq_eoi = gic_eoi_irq, .irq_set_type = gic_set_type, - .irq_retrigger = gic_retrigger, #ifdef CONFIG_SMP .irq_set_affinity = gic_set_affinity, #endif - .irq_set_wake = gic_set_wake, .irq_get_irqchip_state = gic_irq_get_irqchip_state, .irq_set_irqchip_state = gic_irq_set_irqchip_state, }; @@ -1053,7 +985,6 @@ void __init gic_init_bases(unsigned int gic_nr, int irq_start, set_handle_irq(gic_handle_irq); } - gic_chip.flags |= gic_arch_extn.flags; gic_dist_init(gic); gic_cpu_init(gic); gic_pm_init(gic); diff --git a/include/linux/irqchip/arm-gic.h b/include/linux/irqchip/arm-gic.h index 36ec4ae..9de976b 100644 --- a/include/linux/irqchip/arm-gic.h +++ b/include/linux/irqchip/arm-gic.h @@ -95,8 +95,6 @@ struct device_node; -extern struct irq_chip gic_arch_extn; - void gic_set_irqchip_flags(unsigned long flags); void gic_init_bases(unsigned int, int, void __iomem *, void __iomem *, u32 offset, struct device_node *); -- cgit v0.10.2 From 149aabcc44e3e2c1f8fe4f0832be53d2db55b598 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Fri, 10 Apr 2015 12:56:41 +0530 Subject: clockevents: Shutdown detached clockevent device A clockevent device is marked DETACHED when it is replaced by another clockevent device. The device is shutdown properly for drivers that implement legacy ->set_mode() callback, as we call ->set_mode() for CLOCK_EVT_MODE_UNUSED as well. But for the new per-state callback interface, we skip shutting down the device, as we thought its an internal state change. That wasn't correct. The effect is that the device is left programmed in oneshot or periodic mode. Fall-back to 'case CLOCK_EVT_STATE_SHUTDOWN', to shutdown the device. Fixes: bd624d75db21 "clockevents: Introduce mode specific callbacks" Reported-by: Daniel Lezcano Signed-off-by: Viresh Kumar Cc: linaro-kernel@lists.linaro.org Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/eef0a91c51b74d4e52c8e5a95eca27b5a0563f07.1428650683.git.viresh.kumar@linaro.org Signed-off-by: Thomas Gleixner diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 25d942d..629be4e 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -117,11 +117,7 @@ static int __clockevents_set_state(struct clock_event_device *dev, /* Transition with new state-specific callbacks */ switch (state) { case CLOCK_EVT_STATE_DETACHED: - /* - * This is an internal state, which is guaranteed to go from - * SHUTDOWN to DETACHED. No driver interaction required. - */ - return 0; + /* The clockevent device is getting replaced. Shut it down. */ case CLOCK_EVT_STATE_SHUTDOWN: return dev->set_state_shutdown(dev); -- cgit v0.10.2 From dfc8f370316b31b9ca9ce96e50f1c438d9410b4f Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Fri, 24 Apr 2015 15:22:23 +1000 Subject: net/tg3: Release IRQs on permanent error When having permanent EEH error, the PCI device will be removed from the system. For this case, we shouldn't set pcierr_recovery to true wrongly, which blocks the driver to release the allocated interrupts and their handlers. Eventually, we can't disable MSI or MSIx successfully because of the MSI or MSIx interrupts still have associated interrupt actions, which is turned into following stack dump. Oops: Exception in kernel mode, sig: 5 [#1] : [c0000000003b76a8] .free_msi_irqs+0x80/0x1a0 (unreliable) [c00000000039f388] .pci_remove_bus_device+0x98/0x110 [c0000000000790f4] .pcibios_remove_pci_devices+0x9c/0x128 [c000000000077b98] .handle_eeh_events+0x2d8/0x4b0 [c0000000000782d0] .eeh_event_handler+0x130/0x1c0 [c000000000022bd4] .kernel_thread+0x54/0x70 Signed-off-by: Gavin Shan Acked-by: Prashant Sreedharan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 1270b18..069952f 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -18129,7 +18129,9 @@ static pci_ers_result_t tg3_io_error_detected(struct pci_dev *pdev, rtnl_lock(); - tp->pcierr_recovery = true; + /* We needn't recover from permanent error */ + if (state == pci_channel_io_frozen) + tp->pcierr_recovery = true; /* We probably don't have netdev yet */ if (!netdev || !netif_running(netdev)) -- cgit v0.10.2 From 3051f39253680624504021fd3b29fc621a4fd6b6 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 24 Apr 2015 15:52:32 +1000 Subject: ehea: Fix memory hook reference counting crashes The recent commit to only register the EHEA memory hotplug hooks on adapter probe has a few problems. Firstly the reference counting is wrong for multiple adapters, in that the hooks are registered multiple times. Secondly the check in the tear down path is backward. Finally the error path doesn't decrement the count. The multiple registration of the hooks is the biggest problem, as it leads to oopses when the system is rebooted, and/or errors during memory hotplug, eg: $ ./mem-on-off-test.sh -r 2 ... ehea: memory is going offline ehea: LPAR memory changed - re-initializing driver ehea: re-initializing driver complete ehea: memory is going offline ehea: LPAR memory changed - re-initializing driver ehea: opcode=26c ret=fffffffffffffffc arg1=8000000003000003 arg2=0 arg3=700000060000d600 arg4=3fded0000 arg5=200 arg6=0 arg7=0 ehea: register_rpage_mr failed ehea: registering mr failed ehea: register MR failed - driver inoperable! ehea: memory is going offline Fixes: aa183323312d ("ehea: Register memory hotplug, reboot and crash hooks on adapter probe") Signed-off-by: Michael Ellerman Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c index 291c870..2a0dc12 100644 --- a/drivers/net/ethernet/ibm/ehea/ehea_main.c +++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c @@ -3347,7 +3347,7 @@ static int ehea_register_memory_hooks(void) { int ret = 0; - if (atomic_inc_and_test(&ehea_memory_hooks_registered)) + if (atomic_inc_return(&ehea_memory_hooks_registered) > 1) return 0; ret = ehea_create_busmap(); @@ -3381,12 +3381,14 @@ out3: out2: unregister_reboot_notifier(&ehea_reboot_nb); out: + atomic_dec(&ehea_memory_hooks_registered); return ret; } static void ehea_unregister_memory_hooks(void) { - if (atomic_read(&ehea_memory_hooks_registered)) + /* Only remove the hooks if we've registered them */ + if (atomic_read(&ehea_memory_hooks_registered) == 0) return; unregister_reboot_notifier(&ehea_reboot_nb); -- cgit v0.10.2 From 20d96964b51844d4f3bea0249f62db7073c13c30 Mon Sep 17 00:00:00 2001 From: Chee Nouk Phoon Date: Fri, 24 Apr 2015 02:32:07 -0500 Subject: net: eth: altera: Resolve false errors from MSGDMA to TSE This patch resolves false errors from MSGDMA in TX mSGDMA MM to ST mode, and is a continuation of the patch recently submitted by Andrea Oetken. The MSGDMA had a logic bug that masked detection of this issue prior to Quartus 14.1/Build 164. When the MSGDMA logic bug was addressed in Quartus 14.1/Build 164, the driver problem was exposed. The problem is corrected by making sure MSGDMA_DESC_CTL_TR_ERR_IRQ is not set for any of the transmit DMA descriptors, and only used for receive descriptors. Fixes: 71cd26e altera tse: Error-Bit on tx-avalon-stream always set. Signed-off-by: Chee Nouk Phoon Signed-off-by: Vince Bridgers a Cc: Andreas Oetken Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/altera/altera_msgdmahw.h b/drivers/net/ethernet/altera/altera_msgdmahw.h index eba070f..89cd11d 100644 --- a/drivers/net/ethernet/altera/altera_msgdmahw.h +++ b/drivers/net/ethernet/altera/altera_msgdmahw.h @@ -58,15 +58,12 @@ struct msgdma_extended_desc { /* Tx buffer control flags */ #define MSGDMA_DESC_CTL_TX_FIRST (MSGDMA_DESC_CTL_GEN_SOP | \ - MSGDMA_DESC_CTL_TR_ERR_IRQ | \ MSGDMA_DESC_CTL_GO) -#define MSGDMA_DESC_CTL_TX_MIDDLE (MSGDMA_DESC_CTL_TR_ERR_IRQ | \ - MSGDMA_DESC_CTL_GO) +#define MSGDMA_DESC_CTL_TX_MIDDLE (MSGDMA_DESC_CTL_GO) #define MSGDMA_DESC_CTL_TX_LAST (MSGDMA_DESC_CTL_GEN_EOP | \ MSGDMA_DESC_CTL_TR_COMP_IRQ | \ - MSGDMA_DESC_CTL_TR_ERR_IRQ | \ MSGDMA_DESC_CTL_GO) #define MSGDMA_DESC_CTL_TX_SINGLE (MSGDMA_DESC_CTL_GEN_SOP | \ -- cgit v0.10.2 From 2ea2f62c8bda242433809c7f4e9eae1c52c40bbe Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 24 Apr 2015 16:05:01 -0700 Subject: net: fix crash in build_skb() When I added pfmemalloc support in build_skb(), I forgot netlink was using build_skb() with a vmalloc() area. In this patch I introduce __build_skb() for netlink use, and build_skb() is a wrapper handling both skb->head_frag and skb->pfmemalloc This means netlink no longer has to hack skb->head_frag [ 1567.700067] kernel BUG at arch/x86/mm/physaddr.c:26! [ 1567.700067] invalid opcode: 0000 [#1] PREEMPT SMP KASAN [ 1567.700067] Dumping ftrace buffer: [ 1567.700067] (ftrace buffer empty) [ 1567.700067] Modules linked in: [ 1567.700067] CPU: 9 PID: 16186 Comm: trinity-c182 Not tainted 4.0.0-next-20150424-sasha-00037-g4796e21 #2167 [ 1567.700067] task: ffff880127efb000 ti: ffff880246770000 task.ti: ffff880246770000 [ 1567.700067] RIP: __phys_addr (arch/x86/mm/physaddr.c:26 (discriminator 3)) [ 1567.700067] RSP: 0018:ffff8802467779d8 EFLAGS: 00010202 [ 1567.700067] RAX: 000041000ed8e000 RBX: ffffc9008ed8e000 RCX: 000000000000002c [ 1567.700067] RDX: 0000000000000004 RSI: 0000000000000000 RDI: ffffffffb3fd6049 [ 1567.700067] RBP: ffff8802467779f8 R08: 0000000000000019 R09: ffff8801d0168000 [ 1567.700067] R10: ffff8801d01680c7 R11: ffffed003a02d019 R12: ffffc9000ed8e000 [ 1567.700067] R13: 0000000000000f40 R14: 0000000000001180 R15: ffffc9000ed8e000 [ 1567.700067] FS: 00007f2a7da3f700(0000) GS:ffff8801d1000000(0000) knlGS:0000000000000000 [ 1567.700067] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1567.700067] CR2: 0000000000738308 CR3: 000000022e329000 CR4: 00000000000007e0 [ 1567.700067] Stack: [ 1567.700067] ffffc9000ed8e000 ffff8801d0168000 ffffc9000ed8e000 ffff8801d0168000 [ 1567.700067] ffff880246777a28 ffffffffad7c0a21 0000000000001080 ffff880246777c08 [ 1567.700067] ffff88060d302e68 ffff880246777b58 ffff880246777b88 ffffffffad9a6821 [ 1567.700067] Call Trace: [ 1567.700067] build_skb (include/linux/mm.h:508 net/core/skbuff.c:316) [ 1567.700067] netlink_sendmsg (net/netlink/af_netlink.c:1633 net/netlink/af_netlink.c:2329) [ 1567.774369] ? sched_clock_cpu (kernel/sched/clock.c:311) [ 1567.774369] ? netlink_unicast (net/netlink/af_netlink.c:2273) [ 1567.774369] ? netlink_unicast (net/netlink/af_netlink.c:2273) [ 1567.774369] sock_sendmsg (net/socket.c:614 net/socket.c:623) [ 1567.774369] sock_write_iter (net/socket.c:823) [ 1567.774369] ? sock_sendmsg (net/socket.c:806) [ 1567.774369] __vfs_write (fs/read_write.c:479 fs/read_write.c:491) [ 1567.774369] ? get_lock_stats (kernel/locking/lockdep.c:249) [ 1567.774369] ? default_llseek (fs/read_write.c:487) [ 1567.774369] ? vtime_account_user (kernel/sched/cputime.c:701) [ 1567.774369] ? rw_verify_area (fs/read_write.c:406 (discriminator 4)) [ 1567.774369] vfs_write (fs/read_write.c:539) [ 1567.774369] SyS_write (fs/read_write.c:586 fs/read_write.c:577) [ 1567.774369] ? SyS_read (fs/read_write.c:577) [ 1567.774369] ? __this_cpu_preempt_check (lib/smp_processor_id.c:63) [ 1567.774369] ? trace_hardirqs_on_caller (kernel/locking/lockdep.c:2594 kernel/locking/lockdep.c:2636) [ 1567.774369] ? trace_hardirqs_on_thunk (arch/x86/lib/thunk_64.S:42) [ 1567.774369] system_call_fastpath (arch/x86/kernel/entry_64.S:261) Fixes: 79930f5892e ("net: do not deplete pfmemalloc reserve") Signed-off-by: Eric Dumazet Reported-by: Sasha Levin Signed-off-by: David S. Miller diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 06793b5..66e374d 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -773,6 +773,7 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, struct sk_buff *__alloc_skb(unsigned int size, gfp_t priority, int flags, int node); +struct sk_buff *__build_skb(void *data, unsigned int frag_size); struct sk_buff *build_skb(void *data, unsigned int frag_size); static inline struct sk_buff *alloc_skb(unsigned int size, gfp_t priority) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 456ead5..3cfff2a 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -280,13 +280,14 @@ nodata: EXPORT_SYMBOL(__alloc_skb); /** - * build_skb - build a network buffer + * __build_skb - build a network buffer * @data: data buffer provided by caller - * @frag_size: size of fragment, or 0 if head was kmalloced + * @frag_size: size of data, or 0 if head was kmalloced * * Allocate a new &sk_buff. Caller provides space holding head and * skb_shared_info. @data must have been allocated by kmalloc() only if - * @frag_size is 0, otherwise data should come from the page allocator. + * @frag_size is 0, otherwise data should come from the page allocator + * or vmalloc() * The return is the new skb buffer. * On a failure the return is %NULL, and @data is not freed. * Notes : @@ -297,7 +298,7 @@ EXPORT_SYMBOL(__alloc_skb); * before giving packet to stack. * RX rings only contains data buffers, not full skbs. */ -struct sk_buff *build_skb(void *data, unsigned int frag_size) +struct sk_buff *__build_skb(void *data, unsigned int frag_size) { struct skb_shared_info *shinfo; struct sk_buff *skb; @@ -311,11 +312,6 @@ struct sk_buff *build_skb(void *data, unsigned int frag_size) memset(skb, 0, offsetof(struct sk_buff, tail)); skb->truesize = SKB_TRUESIZE(size); - if (frag_size) { - skb->head_frag = 1; - if (virt_to_head_page(data)->pfmemalloc) - skb->pfmemalloc = 1; - } atomic_set(&skb->users, 1); skb->head = data; skb->data = data; @@ -332,6 +328,23 @@ struct sk_buff *build_skb(void *data, unsigned int frag_size) return skb; } + +/* build_skb() is wrapper over __build_skb(), that specifically + * takes care of skb->head and skb->pfmemalloc + * This means that if @frag_size is not zero, then @data must be backed + * by a page fragment, not kmalloc() or vmalloc() + */ +struct sk_buff *build_skb(void *data, unsigned int frag_size) +{ + struct sk_buff *skb = __build_skb(data, frag_size); + + if (skb && frag_size) { + skb->head_frag = 1; + if (virt_to_head_page(data)->pfmemalloc) + skb->pfmemalloc = 1; + } + return skb; +} EXPORT_SYMBOL(build_skb); struct netdev_alloc_cache { diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 19909d0..ec4adbd 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1629,13 +1629,11 @@ static struct sk_buff *netlink_alloc_large_skb(unsigned int size, if (data == NULL) return NULL; - skb = build_skb(data, size); + skb = __build_skb(data, size); if (skb == NULL) vfree(data); - else { - skb->head_frag = 0; + else skb->destructor = netlink_skb_destructor; - } return skb; } -- cgit v0.10.2 From b58d1a9ef92031a6fc2f418c01abafb4458ba009 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Sat, 25 Apr 2015 18:29:16 +0100 Subject: Btrfs: fix race between start dirty bg cache writeout and bg deletion While running xfstests I ran into the following: [20892.242791] ------------[ cut here ]------------ [20892.243776] WARNING: CPU: 0 PID: 13299 at fs/btrfs/super.c:260 __btrfs_abort_transaction+0x52/0x114 [btrfs]() [20892.245874] BTRFS: Transaction aborted (error -2) [20892.247329] Modules linked in: btrfs dm_snapshot dm_bufio dm_flakey dm_mod crc32c_generic xor raid6_pq nfsd auth_rpcgss oid_registry nfs_acl nfs lockd grace fscache sunrpc loop fuse$ [20892.258488] CPU: 0 PID: 13299 Comm: fsstress Tainted: G W 4.0.0-rc5-btrfs-next-9+ #2 [20892.262011] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014 [20892.264738] 0000000000000009 ffff880427f8bc18 ffffffff8142fa46 ffffffff8108b6a2 [20892.266244] ffff880427f8bc68 ffff880427f8bc58 ffffffff81045ea5 ffff880427f8bc48 [20892.267761] ffffffffa0509a6d 00000000fffffffe ffff8803545d6f40 ffffffffa05a15a0 [20892.269378] Call Trace: [20892.269915] [] dump_stack+0x4f/0x7b [20892.271097] [] ? console_unlock+0x361/0x3ad [20892.272173] [] warn_slowpath_common+0xa1/0xbb [20892.273386] [] ? __btrfs_abort_transaction+0x52/0x114 [btrfs] [20892.274857] [] warn_slowpath_fmt+0x46/0x48 [20892.275851] [] __btrfs_abort_transaction+0x52/0x114 [btrfs] [20892.277341] [] write_one_cache_group+0x68/0xaf [btrfs] [20892.278628] [] btrfs_start_dirty_block_groups+0x18d/0x29b [btrfs] [20892.280191] [] btrfs_commit_transaction+0x130/0x9c9 [btrfs] [20892.281781] [] ? trace_hardirqs_on+0xd/0xf [20892.282873] [] btrfs_sync_file+0x313/0x387 [btrfs] [20892.284111] [] vfs_fsync_range+0x95/0xa4 [20892.285203] [] ? time_hardirqs_on+0x15/0x28 [20892.286290] [] ? trace_hardirqs_on_thunk+0x3a/0x3f [20892.287469] [] vfs_fsync+0x1c/0x1e [20892.288412] [] do_fsync+0x34/0x4e [20892.289348] [] SyS_fsync+0x10/0x14 [20892.290255] [] system_call_fastpath+0x12/0x17 [20892.291316] ---[ end trace 597f77e664245373 ]--- [20892.293955] BTRFS: error (device sdg) in write_one_cache_group:3184: errno=-2 No such entry [20892.297390] BTRFS info (device sdg): forced readonly This happens because in btrfs_start_dirty_block_groups() we splice the transaction's list of dirty block groups into a local list and then we keep extracting the first element of the list without holding the cache_write_mutex mutex. This means that before we acquire that mutex the first block group on the list might be removed by a conurrent task running btrfs_remove_block_group(). So make sure we extract the first element (and test the list emptyness) while holding that mutex. Fixes: 1bbc621ef284 ("Btrfs: allow block group cache writeout outside critical section in commit") Signed-off-by: Filipe Manana Signed-off-by: Chris Mason diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 1eef4ee..fc0db98 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3408,17 +3408,14 @@ int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans, int loops = 0; spin_lock(&cur_trans->dirty_bgs_lock); - if (!list_empty(&cur_trans->dirty_bgs)) { - list_splice_init(&cur_trans->dirty_bgs, &dirty); + if (list_empty(&cur_trans->dirty_bgs)) { + spin_unlock(&cur_trans->dirty_bgs_lock); + return 0; } + list_splice_init(&cur_trans->dirty_bgs, &dirty); spin_unlock(&cur_trans->dirty_bgs_lock); again: - if (list_empty(&dirty)) { - btrfs_free_path(path); - return 0; - } - /* * make sure all the block groups on our dirty list actually * exist @@ -3431,18 +3428,16 @@ again: return -ENOMEM; } + /* + * cache_write_mutex is here only to save us from balance or automatic + * removal of empty block groups deleting this block group while we are + * writing out the cache + */ + mutex_lock(&trans->transaction->cache_write_mutex); while (!list_empty(&dirty)) { cache = list_first_entry(&dirty, struct btrfs_block_group_cache, dirty_list); - - /* - * cache_write_mutex is here only to save us from balance - * deleting this block group while we are writing out the - * cache - */ - mutex_lock(&trans->transaction->cache_write_mutex); - /* * this can happen if something re-dirties a block * group that is already under IO. Just wait for it to @@ -3495,7 +3490,6 @@ again: } if (!ret) ret = write_one_cache_group(trans, root, path, cache); - mutex_unlock(&trans->transaction->cache_write_mutex); /* if its not on the io list, we need to put the block group */ if (should_put) @@ -3503,7 +3497,16 @@ again: if (ret) break; + + /* + * Avoid blocking other tasks for too long. It might even save + * us from writing caches for block groups that are going to be + * removed. + */ + mutex_unlock(&trans->transaction->cache_write_mutex); + mutex_lock(&trans->transaction->cache_write_mutex); } + mutex_unlock(&trans->transaction->cache_write_mutex); /* * go through delayed refs for all the stuff we've just kicked off @@ -3514,8 +3517,15 @@ again: loops++; spin_lock(&cur_trans->dirty_bgs_lock); list_splice_init(&cur_trans->dirty_bgs, &dirty); + /* + * dirty_bgs_lock protects us from concurrent block group + * deletes too (not just cache_write_mutex). + */ + if (!list_empty(&dirty)) { + spin_unlock(&cur_trans->dirty_bgs_lock); + goto again; + } spin_unlock(&cur_trans->dirty_bgs_lock); - goto again; } btrfs_free_path(path); -- cgit v0.10.2 From 24b89d08ef63f9c62e55593e27a55e57b0062046 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Sat, 25 Apr 2015 18:31:05 +0100 Subject: Btrfs: fix deadlock when starting writeback of bg caches While starting the writes of the dirty block group caches, if we don't find a block group item in the extent tree we were leaving without releasing our path, running delayed references and then looping again to process any new dirty block groups. However this second iteration of the loop could cause a deadlock because it tries to lock some other extent tree node/leaf which another task already locked and it's blocked because it's waiting for a lock on some node/leaf that is in our path that was not released before. We could also deadlock when running the delayed references - as we could end up trying to lock the same nodes/leafs that we have in our local path (with a different lock type). Got into such case when running xfstests: [20892.242791] ------------[ cut here ]------------ [20892.243776] WARNING: CPU: 0 PID: 13299 at fs/btrfs/super.c:260 __btrfs_abort_transaction+0x52/0x114 [btrfs]() [20892.245874] BTRFS: Transaction aborted (error -2) (...) [20892.269378] Call Trace: [20892.269915] [] dump_stack+0x4f/0x7b [20892.271097] [] ? console_unlock+0x361/0x3ad [20892.272173] [] warn_slowpath_common+0xa1/0xbb [20892.273386] [] ? __btrfs_abort_transaction+0x52/0x114 [btrfs] [20892.274857] [] warn_slowpath_fmt+0x46/0x48 [20892.275851] [] __btrfs_abort_transaction+0x52/0x114 [btrfs] [20892.277341] [] write_one_cache_group+0x68/0xaf [btrfs] [20892.278628] [] btrfs_start_dirty_block_groups+0x18d/0x29b [btrfs] [20892.280191] [] btrfs_commit_transaction+0x130/0x9c9 [btrfs] (...) [20892.291316] ---[ end trace 597f77e664245373 ]--- [20892.293955] BTRFS: error (device sdg) in write_one_cache_group:3184: errno=-2 No such entry [20892.297390] BTRFS info (device sdg): forced readonly [20892.298222] ------------[ cut here ]------------ [20892.299190] WARNING: CPU: 0 PID: 13299 at fs/btrfs/ctree.c:2683 btrfs_search_slot+0x7e/0x7d2 [btrfs]() (...) [20892.326253] Call Trace: [20892.326904] [] dump_stack+0x4f/0x7b [20892.329503] [] ? console_unlock+0x361/0x3ad [20892.330815] [] warn_slowpath_common+0xa1/0xbb [20892.332556] [] ? btrfs_search_slot+0x7e/0x7d2 [btrfs] [20892.333955] [] warn_slowpath_null+0x1a/0x1c [20892.335562] [] btrfs_search_slot+0x7e/0x7d2 [btrfs] [20892.336849] [] ? arch_local_irq_save+0x9/0xc [20892.338222] [] ? cache_save_setup+0x43/0x2a5 [btrfs] [20892.339823] [] ? cache_save_setup+0x57/0x2a5 [btrfs] [20892.341275] [] ? _raw_spin_unlock+0x32/0x46 [20892.342810] [] write_one_cache_group+0x3f/0xaf [btrfs] [20892.344184] [] btrfs_start_dirty_block_groups+0x18d/0x29b [btrfs] [20892.347162] [] btrfs_commit_transaction+0x130/0x9c9 [btrfs] (...) [20892.361015] ---[ end trace 597f77e664245374 ]--- [21120.688097] INFO: task kworker/u8:17:29854 blocked for more than 120 seconds. [21120.689881] Tainted: G W 4.0.0-rc5-btrfs-next-9+ #2 [21120.691384] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. (...) [21120.703696] Call Trace: [21120.704310] [] schedule+0x74/0x83 [21120.705490] [] btrfs_tree_lock+0xd7/0x236 [btrfs] [21120.706757] [] ? signal_pending_state+0x31/0x31 [21120.708156] [] lock_extent_buffer_for_io+0x3e/0x194 [btrfs] [21120.709892] [] ? btree_write_cache_pages+0x273/0x385 [btrfs] [21120.711605] [] btree_write_cache_pages+0x32f/0x385 [btrfs] [21120.723440] [] btree_writepages+0x23/0x5c [btrfs] [21120.724943] [] do_writepages+0x23/0x2c [21120.726008] [] __writeback_single_inode+0x73/0x2fa [21120.727230] [] ? writeback_sb_inodes+0xe5/0x38b [21120.728526] [] ? writeback_sb_inodes+0x196/0x38b [21120.729701] [] writeback_sb_inodes+0x205/0x38b (...) [21120.747853] INFO: task btrfs:13282 blocked for more than 120 seconds. [21120.749459] Tainted: G W 4.0.0-rc5-btrfs-next-9+ #2 [21120.751137] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. (...) [21120.768457] Call Trace: [21120.769039] [] schedule+0x74/0x83 [21120.770107] [] btrfs_commit_transaction+0x315/0x9c9 [btrfs] [21120.771558] [] ? signal_pending_state+0x31/0x31 [21120.773659] [] prepare_to_relocate+0xcb/0xd2 [btrfs] [21120.776257] [] relocate_block_group+0x44/0x4a9 [btrfs] [21120.777755] [] ? btrfs_relocate_block_group+0x161/0x288 [btrfs] [21120.779459] [] btrfs_relocate_block_group+0x169/0x288 [btrfs] [21120.781153] [] btrfs_relocate_chunk.isra.29+0x3e/0xa7 [btrfs] [21120.783918] [] btrfs_balance+0xaa4/0xc52 [btrfs] [21120.785436] [] ? cpu_cache_get.isra.39+0xe/0x1f [21120.786434] [] btrfs_ioctl_balance+0x23f/0x2b0 [btrfs] (...) [21120.889251] INFO: task fsstress:13288 blocked for more than 120 seconds. [21120.890526] Tainted: G W 4.0.0-rc5-btrfs-next-9+ #2 [21120.891773] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. (...) [21120.899960] Call Trace: [21120.900743] [] schedule+0x74/0x83 [21120.903004] [] btrfs_tree_lock+0xd7/0x236 [btrfs] [21120.904383] [] ? signal_pending_state+0x31/0x31 [21120.905608] [] btrfs_search_slot+0x766/0x7d2 [btrfs] [21120.906812] [] ? virt_to_head_page+0x9/0x2c [21120.907874] [] ? cache_alloc_debugcheck_after.isra.42+0x16c/0x1cb [21120.909551] [] btrfs_insert_empty_items+0x5d/0xa8 [btrfs] [21120.910914] [] btrfs_insert_item+0x5a/0xa5 [btrfs] [21120.912181] [] ? btrfs_create_pending_block_groups+0x96/0x130 [btrfs] [21120.913784] [] btrfs_create_pending_block_groups+0xaf/0x130 [btrfs] [21120.915374] [] __btrfs_end_transaction+0x84/0x366 [btrfs] [21120.916735] [] btrfs_end_transaction+0x10/0x12 [btrfs] [21120.917996] [] btrfs_check_data_free_space+0x11f/0x27c [btrfs] [21120.919478] [] btrfs_delalloc_reserve_space+0x1e/0x51 [btrfs] [21120.921226] [] btrfs_truncate_page+0x85/0x2c4 [btrfs] [21120.923121] [] btrfs_cont_expand+0x41/0x3ef [btrfs] [21120.924449] [] ? btrfs_file_write_iter+0x19a/0x431 [btrfs] [21120.926602] [] ? arch_local_irq_save+0x9/0xc [21120.927769] [] ? btrfs_file_write_iter+0x19a/0x431 [btrfs] [21120.929324] [] ? btrfs_file_write_iter+0x1a9/0x431 [btrfs] [21120.930723] [] btrfs_file_write_iter+0x1e2/0x431 [btrfs] [21120.931897] [] ? get_parent_ip+0xe/0x3e [21120.934446] [] new_sync_write+0x7c/0xa0 [21120.935528] [] vfs_write+0xb2/0x117 (...) Fixes: 1bbc621ef284 ("Btrfs: allow block group cache writeout outside critical section in commit") Signed-off-by: Filipe Manana Signed-off-by: Chris Mason diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index fc0db98..eeb429c 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3178,8 +3178,8 @@ static int write_one_cache_group(struct btrfs_trans_handle *trans, bi = btrfs_item_ptr_offset(leaf, path->slots[0]); write_extent_buffer(leaf, &cache->item, bi, sizeof(cache->item)); btrfs_mark_buffer_dirty(leaf); - btrfs_release_path(path); fail: + btrfs_release_path(path); if (ret) btrfs_abort_transaction(trans, root, ret); return ret; -- cgit v0.10.2 From e4c88f007be78d38eaef316c599a1ee2f0272c15 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Sat, 18 Apr 2015 05:22:48 -0700 Subject: Btrfs: don't check for delalloc_bytes in cache_save_setup Now that we're doing free space cache writeback outside the critical section in the commit, there is a bigger window for delalloc_bytes to be added after a cache has been written. find_free_extent may do this without putting the block group back into the dirty list, and also without a transaction running. Checking for delalloc_bytes in cache_save_setup means we might leave the cache marked as written without invalidating it. Consistency checks during mount will toss the cache, but it's better to get rid of the check in cache_save_setup and let it get invalidated by the checks already done during cache write out. Signed-off-by: Chris Mason diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index eeb429c..6f420c8 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3305,8 +3305,7 @@ again: spin_lock(&block_group->lock); if (block_group->cached != BTRFS_CACHE_FINISHED || - !btrfs_test_opt(root, SPACE_CACHE) || - block_group->delalloc_bytes) { + !btrfs_test_opt(root, SPACE_CACHE)) { /* * don't bother trying to write stuff out _if_ * a) we're not cached, -- cgit v0.10.2 From 1b9845081633072c18f30d8cfd09c267adf0b109 Mon Sep 17 00:00:00 2001 From: Forrest Liu Date: Mon, 9 Feb 2015 17:30:47 +0800 Subject: Btrfs: fix find_free_dev_extent() malfunction in case device tree has hole If device tree has hole, find_free_dev_extent() cannot find available address properly. The problem can be reproduce by following script. mntpath=/btrfs loopdev=/dev/loop0 filepath=/home/forrest/image umount $mntpath losetup -d $loopdev truncate --size 100g $filepath losetup $loopdev $filepath mkfs.btrfs -f $loopdev mount $loopdev $mntpath # make device tree with one big hole for i in `seq 1 1 100`; do fallocate -l 1g $mntpath/$i done sync for i in `seq 1 1 95`; do rm $mntpath/$i done sync # wait cleaner thread remove unused block group sleep 300 fallocate -l 1g $mntpath/aaa # failed to allocate new chunk fallocate -l 1g $mntpath/bbb Above script will make device tree with one big hole, and can only allocate just one chunk in a transaction, so failed to allocate new chunk for $mntpath/bbb item 8 key (1 DEV_EXTENT 2185232384) itemoff 15859 itemsize 48 dev extent chunk_tree 3 chunk objectid 256 chunk offset 106292051968 length 1073741824 item 9 key (1 DEV_EXTENT 104190705664) itemoff 15811 itemsize 48 dev extent chunk_tree 3 chunk objectid 256 chunk offset 103108575232 length 1073741824 Signed-off-by: Forrest Liu Reviewed-by: Liu Bo Signed-off-by: Chris Mason diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 8bcd2a0..96aebf3 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1058,6 +1058,7 @@ static int contains_pending_extent(struct btrfs_trans_handle *trans, struct extent_map *em; struct list_head *search_list = &trans->transaction->pending_chunks; int ret = 0; + u64 physical_start = *start; again: list_for_each_entry(em, search_list, list) { @@ -1068,9 +1069,9 @@ again: for (i = 0; i < map->num_stripes; i++) { if (map->stripes[i].dev != device) continue; - if (map->stripes[i].physical >= *start + len || + if (map->stripes[i].physical >= physical_start + len || map->stripes[i].physical + em->orig_block_len <= - *start) + physical_start) continue; *start = map->stripes[i].physical + em->orig_block_len; @@ -1193,8 +1194,14 @@ again: */ if (contains_pending_extent(trans, device, &search_start, - hole_size)) - hole_size = 0; + hole_size)) { + if (key.offset >= search_start) { + hole_size = key.offset - search_start; + } else { + WARN_ON_ONCE(1); + hole_size = 0; + } + } if (hole_size > max_hole_size) { max_hole_start = search_start; -- cgit v0.10.2 From 67b7859e9bfa0dcee5a8256932e393e98739be1b Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Tue, 24 Feb 2015 02:47:04 -0800 Subject: btrfs: handle ENOMEM in btrfs_alloc_tree_block This is one of the first places to give out when memory is tight. Handle it properly rather than with a BUG_ON. Also fix the comment about the return value, which is an ERR_PTR, not NULL, on error. Signed-off-by: Omar Sandoval Reviewed-by: David Sterba Signed-off-by: Chris Mason diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 6f420c8..0ec8e22 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -7546,7 +7546,7 @@ static void unuse_block_rsv(struct btrfs_fs_info *fs_info, * returns the key for the extent through ins, and a tree buffer for * the first block of the extent through buf. * - * returns the tree buffer or NULL. + * returns the tree buffer or an ERR_PTR on error. */ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root, @@ -7557,6 +7557,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, struct btrfs_key ins; struct btrfs_block_rsv *block_rsv; struct extent_buffer *buf; + struct btrfs_delayed_extent_op *extent_op; u64 flags = 0; int ret; u32 blocksize = root->nodesize; @@ -7577,13 +7578,14 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, ret = btrfs_reserve_extent(root, blocksize, blocksize, empty_size, hint, &ins, 0, 0); - if (ret) { - unuse_block_rsv(root->fs_info, block_rsv, blocksize); - return ERR_PTR(ret); - } + if (ret) + goto out_unuse; buf = btrfs_init_new_buffer(trans, root, ins.objectid, level); - BUG_ON(IS_ERR(buf)); /* -ENOMEM */ + if (IS_ERR(buf)) { + ret = PTR_ERR(buf); + goto out_free_reserved; + } if (root_objectid == BTRFS_TREE_RELOC_OBJECTID) { if (parent == 0) @@ -7593,9 +7595,11 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, BUG_ON(parent > 0); if (root_objectid != BTRFS_TREE_LOG_OBJECTID) { - struct btrfs_delayed_extent_op *extent_op; extent_op = btrfs_alloc_delayed_extent_op(); - BUG_ON(!extent_op); /* -ENOMEM */ + if (!extent_op) { + ret = -ENOMEM; + goto out_free_buf; + } if (key) memcpy(&extent_op->key, key, sizeof(extent_op->key)); else @@ -7610,13 +7614,24 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, extent_op->level = level; ret = btrfs_add_delayed_tree_ref(root->fs_info, trans, - ins.objectid, - ins.offset, parent, root_objectid, - level, BTRFS_ADD_DELAYED_EXTENT, - extent_op, 0); - BUG_ON(ret); /* -ENOMEM */ + ins.objectid, ins.offset, + parent, root_objectid, level, + BTRFS_ADD_DELAYED_EXTENT, + extent_op, 0); + if (ret) + goto out_free_delayed; } return buf; + +out_free_delayed: + btrfs_free_delayed_extent_op(extent_op); +out_free_buf: + free_extent_buffer(buf); +out_free_reserved: + btrfs_free_reserved_extent(root, ins.objectid, ins.offset, 0); +out_unuse: + unuse_block_rsv(root->fs_info, block_rsv, blocksize); + return ERR_PTR(ret); } struct walk_control { -- cgit v0.10.2 From 5ca64f45e92dc52bd3bc1ad93f4f9e5a57955f28 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Tue, 24 Feb 2015 02:47:05 -0800 Subject: btrfs: fix race on ENOMEM in alloc_extent_buffer Consider the following interleaving of overlapping calls to alloc_extent_buffer: Call 1: - Successfully allocates a few pages with find_or_create_page - find_or_create_page fails, goto free_eb - Unlocks the allocated pages Call 2: - Calls find_or_create_page and gets a page in call 1's extent_buffer - Finds that the page is already associated with an extent_buffer - Grabs a reference to the half-written extent_buffer and calls mark_extent_buffer_accessed on it mark_extent_buffer_accessed will then try to call mark_page_accessed on a null page and panic. The fix is to decrement the reference count on the half-written extent_buffer before unlocking the pages so call 2 won't use it. We should also set exists = NULL in the case that we don't use exists to avoid accidentally returning a freed extent_buffer in an error case. Signed-off-by: Omar Sandoval Reviewed-by: David Sterba Reviewed-by: Liu Bo Signed-off-by: Chris Mason diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 782f3bc..ea100eb 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4870,6 +4870,7 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info, mark_extent_buffer_accessed(exists, p); goto free_eb; } + exists = NULL; /* * Do this so attach doesn't complain and we need to @@ -4933,12 +4934,12 @@ again: return eb; free_eb: + WARN_ON(!atomic_dec_and_test(&eb->refs)); for (i = 0; i < num_pages; i++) { if (eb->pages[i]) unlock_page(eb->pages[i]); } - WARN_ON(!atomic_dec_and_test(&eb->refs)); btrfs_release_extent_buffer(eb); return exists; } -- cgit v0.10.2 From b86054540e4ad7d8df67f3a80658d037a2ce1c02 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Tue, 24 Feb 2015 02:47:06 -0800 Subject: btrfs: check io_ctl_prepare_pages return in __btrfs_write_out_cache If io_ctl_prepare_pages fails, the pages in io_ctl.pages are not valid. When we try to access them later, things will blow up in various ways. Also fix the comment about the return value, which is an errno on error, not -1, and update the cases where it was not. Reviewed-by: Liu Bo Signed-off-by: Omar Sandoval Signed-off-by: Chris Mason diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 81fa75a..41c510b 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -1218,7 +1218,7 @@ out: * * This function writes out a free space cache struct to disk for quick recovery * on mount. This will return 0 if it was successfull in writing the cache out, - * and -1 if it was not. + * or an errno if it was not. */ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, struct btrfs_free_space_ctl *ctl, @@ -1235,12 +1235,12 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, int must_iput = 0; if (!i_size_read(inode)) - return -1; + return -EIO; WARN_ON(io_ctl->pages); ret = io_ctl_init(io_ctl, inode, root, 1); if (ret) - return -1; + return ret; if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA)) { down_write(&block_group->data_rwsem); @@ -1258,7 +1258,9 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, } /* Lock all pages first so we can lock the extent safely. */ - io_ctl_prepare_pages(io_ctl, inode, 0); + ret = io_ctl_prepare_pages(io_ctl, inode, 0); + if (ret) + goto out; lock_extent_bits(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1, 0, &cached_state); -- cgit v0.10.2 From 909e26dce3f7600f5e293ac0522c28790a0c8c9c Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Fri, 10 Apr 2015 14:20:40 -0700 Subject: btrfs: unlock i_mutex after attempting to delete subvolume during send Whenever the check for a send in progress introduced in commit 521e0546c970 (btrfs: protect snapshots from deleting during send) is hit, we return without unlocking inode->i_mutex. This is easy to see with lockdep enabled: [ +0.000059] ================================================ [ +0.000028] [ BUG: lock held when returning to user space! ] [ +0.000029] 4.0.0-rc5-00096-g3c435c1 #93 Not tainted [ +0.000026] ------------------------------------------------ [ +0.000029] btrfs/211 is leaving the kernel with locks still held! [ +0.000029] 1 lock held by btrfs/211: [ +0.000023] #0: (&type->i_mutex_dir_key){+.+.+.}, at: [] btrfs_ioctl_snap_destroy+0x2df/0x7a0 Make sure we unlock it in the error path. Reviewed-by: Filipe Manana Reviewed-by: David Sterba Cc: stable@vger.kernel.org Signed-off-by: Omar Sandoval Signed-off-by: Chris Mason diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index ca5d968..cdcd98c 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2410,7 +2410,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, "Attempt to delete subvolume %llu during send", dest->root_key.objectid); err = -EPERM; - goto out_dput; + goto out_unlock_inode; } d_invalidate(dentry); @@ -2505,6 +2505,7 @@ out_up_write: root_flags & ~BTRFS_ROOT_SUBVOL_DEAD); spin_unlock(&dest->root_item_lock); } +out_unlock_inode: mutex_unlock(&inode->i_mutex); if (!err) { shrink_dcache_sb(root->fs_info->sb); -- cgit v0.10.2 From 6e17d30bfaf43e04d991392d8484f1c556810c33 Mon Sep 17 00:00:00 2001 From: Yang Dongsheng Date: Thu, 9 Apr 2015 12:08:43 +0800 Subject: Btrfs: fill ->last_trans for delayed inode in btrfs_fill_inode. We need to fill inode when we found a node for it in delayed_nodes_tree. But we did not fill the ->last_trans currently, it will cause the test of xfstest/generic/311 fail. Scenario of the 311 is shown as below: Problem: (1). test_fd = open(fname, O_RDWR|O_DIRECT) (2). pwrite(test_fd, buf, 4096, 0) (3). close(test_fd) (4). drop_all_caches() <-------- "echo 3 > /proc/sys/vm/drop_caches" (5). test_fd = open(fname, O_RDWR|O_DIRECT) (6). fsync(test_fd); <-------- we did not get the correct log entry for the file Reason: When we re-open this file in (5), we would find a node in delayed_nodes_tree and fill the inode we are lookup with the information. But the ->last_trans is not filled, then the fsync() will check the ->last_trans and found it's 0 then say this inode is already in our tree which is commited, not recording the extents for it. Fix: This patch fill the ->last_trans properly and set the runtime_flags if needed in this situation. Then we can get the log entries we expected after (6) and generic/311 passed. Signed-off-by: Dongsheng Yang Reviewed-by: Miao Xie Signed-off-by: Chris Mason diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index cde698a..a2ae427 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1802,6 +1802,8 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev) set_nlink(inode, btrfs_stack_inode_nlink(inode_item)); inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item)); BTRFS_I(inode)->generation = btrfs_stack_inode_generation(inode_item); + BTRFS_I(inode)->last_trans = btrfs_stack_inode_transid(inode_item); + inode->i_version = btrfs_stack_inode_sequence(inode_item); inode->i_rdev = 0; *rdev = btrfs_stack_inode_rdev(inode_item); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 27b59b83..0020b56 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3633,25 +3633,28 @@ static void btrfs_read_locked_inode(struct inode *inode) BTRFS_I(inode)->generation = btrfs_inode_generation(leaf, inode_item); BTRFS_I(inode)->last_trans = btrfs_inode_transid(leaf, inode_item); + inode->i_version = btrfs_inode_sequence(leaf, inode_item); + inode->i_generation = BTRFS_I(inode)->generation; + inode->i_rdev = 0; + rdev = btrfs_inode_rdev(leaf, inode_item); + + BTRFS_I(inode)->index_cnt = (u64)-1; + BTRFS_I(inode)->flags = btrfs_inode_flags(leaf, inode_item); + +cache_index: /* * If we were modified in the current generation and evicted from memory * and then re-read we need to do a full sync since we don't have any * idea about which extents were modified before we were evicted from * cache. + * + * This is required for both inode re-read from disk and delayed inode + * in delayed_nodes_tree. */ if (BTRFS_I(inode)->last_trans == root->fs_info->generation) set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); - inode->i_version = btrfs_inode_sequence(leaf, inode_item); - inode->i_generation = BTRFS_I(inode)->generation; - inode->i_rdev = 0; - rdev = btrfs_inode_rdev(leaf, inode_item); - - BTRFS_I(inode)->index_cnt = (u64)-1; - BTRFS_I(inode)->flags = btrfs_inode_flags(leaf, inode_item); - -cache_index: path->slots[0]++; if (inode->i_nlink != 1 || path->slots[0] >= btrfs_header_nritems(leaf)) -- cgit v0.10.2 From 0e03fd3e335d272bee88fe733d5fd13f5c5b7140 Mon Sep 17 00:00:00 2001 From: Alexey Khoroshilov Date: Sat, 25 Apr 2015 04:07:03 +0300 Subject: pxa168: fix double deallocation of managed resources Commit 43d3ddf87a57 ("net: pxa168_eth: add device tree support") starts to use managed resources by adding devm_clk_get() and devm_ioremap_resource(), but it leaves explicit iounmap() and clock_put() in pxa168_eth_remove() and in failure handling code of pxa168_eth_probe(). As a result double free can happen. The patch removes explicit resource deallocation. Also it converts clk_disable() to clk_disable_unprepare() to make it symmetrical with clk_prepare_enable(). Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c index af829c5..7ace07d 100644 --- a/drivers/net/ethernet/marvell/pxa168_eth.c +++ b/drivers/net/ethernet/marvell/pxa168_eth.c @@ -1508,7 +1508,8 @@ static int pxa168_eth_probe(struct platform_device *pdev) np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0); if (!np) { dev_err(&pdev->dev, "missing phy-handle\n"); - return -EINVAL; + err = -EINVAL; + goto err_netdev; } of_property_read_u32(np, "reg", &pep->phy_addr); pep->phy_intf = of_get_phy_mode(pdev->dev.of_node); @@ -1526,7 +1527,7 @@ static int pxa168_eth_probe(struct platform_device *pdev) pep->smi_bus = mdiobus_alloc(); if (pep->smi_bus == NULL) { err = -ENOMEM; - goto err_base; + goto err_netdev; } pep->smi_bus->priv = pep; pep->smi_bus->name = "pxa168_eth smi"; @@ -1551,13 +1552,10 @@ err_mdiobus: mdiobus_unregister(pep->smi_bus); err_free_mdio: mdiobus_free(pep->smi_bus); -err_base: - iounmap(pep->base); err_netdev: free_netdev(dev); err_clk: - clk_disable(clk); - clk_put(clk); + clk_disable_unprepare(clk); return err; } @@ -1574,13 +1572,9 @@ static int pxa168_eth_remove(struct platform_device *pdev) if (pep->phy) phy_disconnect(pep->phy); if (pep->clk) { - clk_disable(pep->clk); - clk_put(pep->clk); - pep->clk = NULL; + clk_disable_unprepare(pep->clk); } - iounmap(pep->base); - pep->base = NULL; mdiobus_unregister(pep->smi_bus); mdiobus_free(pep->smi_bus); unregister_netdev(dev); -- cgit v0.10.2 From 7cdbc6f74f8e6c06304b69b4e944fbd669581c7c Mon Sep 17 00:00:00 2001 From: Andreas Oetken Date: Sat, 25 Apr 2015 18:07:52 +0200 Subject: altera tse: add support for fixed-links. Add support for fixed-links in configurations without PHY. (e.g. connection to a switch, SGMII point to point, SFPs) Check: Documentation/devicetree/bindings/net/fixed-link.txt. Signed-off-by: Andreas Oetken Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 90a7630..0533c05 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -777,6 +777,8 @@ static int init_phy(struct net_device *dev) struct altera_tse_private *priv = netdev_priv(dev); struct phy_device *phydev; struct device_node *phynode; + bool fixed_link = false; + int rc = 0; /* Avoid init phy in case of no phy present */ if (!priv->phy_iface) @@ -789,13 +791,32 @@ static int init_phy(struct net_device *dev) phynode = of_parse_phandle(priv->device->of_node, "phy-handle", 0); if (!phynode) { - netdev_dbg(dev, "no phy-handle found\n"); - if (!priv->mdio) { - netdev_err(dev, - "No phy-handle nor local mdio specified\n"); - return -ENODEV; + /* check if a fixed-link is defined in device-tree */ + if (of_phy_is_fixed_link(priv->device->of_node)) { + rc = of_phy_register_fixed_link(priv->device->of_node); + if (rc < 0) { + netdev_err(dev, "cannot register fixed PHY\n"); + return rc; + } + + /* In the case of a fixed PHY, the DT node associated + * to the PHY is the Ethernet MAC DT node. + */ + phynode = of_node_get(priv->device->of_node); + fixed_link = true; + + netdev_dbg(dev, "fixed-link detected\n"); + phydev = of_phy_connect(dev, phynode, + &altera_tse_adjust_link, + 0, priv->phy_iface); + } else { + netdev_dbg(dev, "no phy-handle found\n"); + if (!priv->mdio) { + netdev_err(dev, "No phy-handle nor local mdio specified\n"); + return -ENODEV; + } + phydev = connect_local_phy(dev); } - phydev = connect_local_phy(dev); } else { netdev_dbg(dev, "phy-handle found\n"); phydev = of_phy_connect(dev, phynode, @@ -819,10 +840,10 @@ static int init_phy(struct net_device *dev) /* Broken HW is sometimes missing the pull-up resistor on the * MDIO line, which results in reads to non-existent devices returning * 0 rather than 0xffff. Catch this here and treat 0 as a non-existent - * device as well. + * device as well. If a fixed-link is used the phy_id is always 0. * Note: phydev->phy_id is the result of reading the UID PHY registers. */ - if (phydev->phy_id == 0) { + if ((phydev->phy_id == 0) && !fixed_link) { netdev_err(dev, "Bad PHY UID 0x%08x\n", phydev->phy_id); phy_disconnect(phydev); return -ENODEV; -- cgit v0.10.2 From a31196b07f8034eba6a3487a1ad1bb5ec5cd58a5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 25 Apr 2015 09:35:24 -0700 Subject: net: rfs: fix crash in get_rps_cpus() Commit 567e4b79731c ("net: rfs: add hash collision detection") had one mistake : RPS_NO_CPU is no longer the marker for invalid cpu in set_rps_cpu() and get_rps_cpu(), as @next_cpu was the result of an AND with rps_cpu_mask This bug showed up on a host with 72 cpus : next_cpu was 0x7f, and the code was trying to access percpu data of an non existent cpu. In a follow up patch, we might get rid of compares against nr_cpu_ids, if we init the tables with 0. This is silly to test for a very unlikely condition that exists only shortly after table initialization, as we got rid of rps_reset_sock_flow() and similar functions that were writing this RPS_NO_CPU magic value at flow dismantle : When table is old enough, it never contains this value anymore. Fixes: 567e4b79731c ("net: rfs: add hash collision detection") Signed-off-by: Eric Dumazet Cc: Tom Herbert Cc: Ben Hutchings Signed-off-by: David S. Miller diff --git a/Documentation/networking/scaling.txt b/Documentation/networking/scaling.txt index cbfac09..59f4db2 100644 --- a/Documentation/networking/scaling.txt +++ b/Documentation/networking/scaling.txt @@ -282,7 +282,7 @@ following is true: - The current CPU's queue head counter >= the recorded tail counter value in rps_dev_flow[i] -- The current CPU is unset (equal to RPS_NO_CPU) +- The current CPU is unset (>= nr_cpu_ids) - The current CPU is offline After this check, the packet is sent to the (possibly updated) current diff --git a/net/core/dev.c b/net/core/dev.c index 1796cef5..c7ba038 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3079,7 +3079,7 @@ static struct rps_dev_flow * set_rps_cpu(struct net_device *dev, struct sk_buff *skb, struct rps_dev_flow *rflow, u16 next_cpu) { - if (next_cpu != RPS_NO_CPU) { + if (next_cpu < nr_cpu_ids) { #ifdef CONFIG_RFS_ACCEL struct netdev_rx_queue *rxqueue; struct rps_dev_flow_table *flow_table; @@ -3184,7 +3184,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, * If the desired CPU (where last recvmsg was done) is * different from current CPU (one in the rx-queue flow * table entry), switch if one of the following holds: - * - Current CPU is unset (equal to RPS_NO_CPU). + * - Current CPU is unset (>= nr_cpu_ids). * - Current CPU is offline. * - The current CPU's queue tail has advanced beyond the * last packet that was enqueued using this table entry. @@ -3192,14 +3192,14 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, * have been dequeued, thus preserving in order delivery. */ if (unlikely(tcpu != next_cpu) && - (tcpu == RPS_NO_CPU || !cpu_online(tcpu) || + (tcpu >= nr_cpu_ids || !cpu_online(tcpu) || ((int)(per_cpu(softnet_data, tcpu).input_queue_head - rflow->last_qtail)) >= 0)) { tcpu = next_cpu; rflow = set_rps_cpu(dev, skb, rflow, next_cpu); } - if (tcpu != RPS_NO_CPU && cpu_online(tcpu)) { + if (tcpu < nr_cpu_ids && cpu_online(tcpu)) { *rflowp = rflow; cpu = tcpu; goto done; @@ -3240,14 +3240,14 @@ bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, struct rps_dev_flow_table *flow_table; struct rps_dev_flow *rflow; bool expire = true; - int cpu; + unsigned int cpu; rcu_read_lock(); flow_table = rcu_dereference(rxqueue->rps_flow_table); if (flow_table && flow_id <= flow_table->mask) { rflow = &flow_table->flows[flow_id]; cpu = ACCESS_ONCE(rflow->cpu); - if (rflow->filter == filter_id && cpu != RPS_NO_CPU && + if (rflow->filter == filter_id && cpu < nr_cpu_ids && ((int)(per_cpu(softnet_data, cpu).input_queue_head - rflow->last_qtail) < (int)(10 * flow_table->mask))) -- cgit v0.10.2 From 73b5a6f2a7a1cb78ccdec3900afc8657e11bc6bf Mon Sep 17 00:00:00 2001 From: Matan Barak Date: Sun, 26 Apr 2015 15:55:57 +0300 Subject: net/bonding: Make DRV macros private The bonding modules currently defines four macros with general names that pollute the global namespace: DRV_VERSION DRV_RELDATE DRV_NAME DRV_DESCRIPTION Fixing that by defining a private bonding_priv.h header files which includes those defines. Signed-off-by: Matan Barak Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 78dde56..3a10551 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -82,6 +82,8 @@ #include #include +#include "bonding_priv.h" + /*---------------------------- Module parameters ----------------------------*/ /* monitor all links that often (in milliseconds). <=0 disables monitoring */ diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c index 62694cf..b20b35ac 100644 --- a/drivers/net/bonding/bond_procfs.c +++ b/drivers/net/bonding/bond_procfs.c @@ -4,6 +4,7 @@ #include #include +#include "bonding_priv.h" static void *bond_info_seq_start(struct seq_file *seq, loff_t *pos) __acquires(RCU) diff --git a/drivers/net/bonding/bonding_priv.h b/drivers/net/bonding/bonding_priv.h new file mode 100644 index 0000000..5a4d81a --- /dev/null +++ b/drivers/net/bonding/bonding_priv.h @@ -0,0 +1,25 @@ +/* + * Bond several ethernet interfaces into a Cisco, running 'Etherchannel'. + * + * Portions are (c) Copyright 1995 Simon "Guru Aleph-Null" Janes + * NCM: Network and Communications Management, Inc. + * + * BUT, I'm the one who modified it for ethernet, so: + * (c) Copyright 1999, Thomas Davis, tadavis@lbl.gov + * + * This software may be used and distributed according to the terms + * of the GNU Public License, incorporated herein by reference. + * + */ + +#ifndef _BONDING_PRIV_H +#define _BONDING_PRIV_H + +#define DRV_VERSION "3.7.1" +#define DRV_RELDATE "April 27, 2011" +#define DRV_NAME "bonding" +#define DRV_DESCRIPTION "Ethernet Channel Bonding Driver" + +#define bond_version DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n" + +#endif diff --git a/include/net/bonding.h b/include/net/bonding.h index fda6fee..78ed135 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -30,13 +30,6 @@ #include #include -#define DRV_VERSION "3.7.1" -#define DRV_RELDATE "April 27, 2011" -#define DRV_NAME "bonding" -#define DRV_DESCRIPTION "Ethernet Channel Bonding Driver" - -#define bond_version DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n" - #define BOND_MAX_ARP_TARGETS 16 #define BOND_DEFAULT_MIIMON 100 -- cgit v0.10.2 From 8ad31bf45db1c4a8f400cbc10400ce9fc0bed227 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 26 Apr 2015 11:53:54 +0200 Subject: dmaengine: usb-dmac: Protect PM-only functions to kill warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If CONFIG_PM=n: drivers/dma/sh/usb-dmac.c:677: warning: ‘usb_dmac_runtime_suspend’ defined but not used drivers/dma/sh/usb-dmac.c:688: warning: ‘usb_dmac_runtime_resume’ defined but not used Protect the unused functions by #ifdef CONFIG_PM to fix this. Signed-off-by: Geert Uytterhoeven Signed-off-by: Vinod Koul diff --git a/drivers/dma/sh/usb-dmac.c b/drivers/dma/sh/usb-dmac.c index f705798..ebd8a5f 100644 --- a/drivers/dma/sh/usb-dmac.c +++ b/drivers/dma/sh/usb-dmac.c @@ -673,6 +673,7 @@ static struct dma_chan *usb_dmac_of_xlate(struct of_phandle_args *dma_spec, * Power management */ +#ifdef CONFIG_PM static int usb_dmac_runtime_suspend(struct device *dev) { struct usb_dmac *dmac = dev_get_drvdata(dev); @@ -690,6 +691,7 @@ static int usb_dmac_runtime_resume(struct device *dev) return usb_dmac_init(dmac); } +#endif /* CONFIG_PM */ static const struct dev_pm_ops usb_dmac_pm = { SET_RUNTIME_PM_OPS(usb_dmac_runtime_suspend, usb_dmac_runtime_resume, -- cgit v0.10.2 From 801661467fd50832191c81665d061ffabcc1c5de Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 26 Apr 2015 12:52:02 +0200 Subject: dmaengine: xgene: Set hardware dependency The xgene-dma driver is only useful on X-Gene SoC. Signed-off-by: Jean Delvare Cc: Rameshwar Prasad Sahu Cc: Loc Ho Cc: Vinod Koul Signed-off-by: Vinod Koul diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig index fd7ac13..bda2cb0 100644 --- a/drivers/dma/Kconfig +++ b/drivers/dma/Kconfig @@ -437,6 +437,7 @@ config IMG_MDC_DMA config XGENE_DMA tristate "APM X-Gene DMA support" + depends on ARCH_XGENE || COMPILE_TEST select DMA_ENGINE select DMA_ENGINE_RAID select ASYNC_TX_ENABLE_CHANNEL_SWITCH -- cgit v0.10.2 From 325301892a2d348323e09598ae108ba26889f7f9 Mon Sep 17 00:00:00 2001 From: Sylvain Rochet Date: Sun, 26 Apr 2015 20:40:52 +0200 Subject: ppp: mppe: sanity error path rework We are going to need sanity error path a little further, rework to be able to use the sanity error path anywhere in decompressor. Signed-off-by: Sylvain Rochet Signed-off-by: David S. Miller diff --git a/drivers/net/ppp/ppp_mppe.c b/drivers/net/ppp/ppp_mppe.c index 911b216..692ee0f 100644 --- a/drivers/net/ppp/ppp_mppe.c +++ b/drivers/net/ppp/ppp_mppe.c @@ -478,7 +478,6 @@ mppe_decompress(void *arg, unsigned char *ibuf, int isize, unsigned char *obuf, struct blkcipher_desc desc = { .tfm = state->arc4 }; unsigned ccount; int flushed = MPPE_BITS(ibuf) & MPPE_BIT_FLUSHED; - int sanity = 0; struct scatterlist sg_in[1], sg_out[1]; if (isize <= PPP_HDRLEN + MPPE_OVHD) { @@ -514,31 +513,19 @@ mppe_decompress(void *arg, unsigned char *ibuf, int isize, unsigned char *obuf, "mppe_decompress[%d]: ENCRYPTED bit not set!\n", state->unit); state->sanity_errors += 100; - sanity = 1; + goto sanity_error; } if (!state->stateful && !flushed) { printk(KERN_DEBUG "mppe_decompress[%d]: FLUSHED bit not set in " "stateless mode!\n", state->unit); state->sanity_errors += 100; - sanity = 1; + goto sanity_error; } if (state->stateful && ((ccount & 0xff) == 0xff) && !flushed) { printk(KERN_DEBUG "mppe_decompress[%d]: FLUSHED bit not set on " "flag packet!\n", state->unit); state->sanity_errors += 100; - sanity = 1; - } - - if (sanity) { - if (state->sanity_errors < SANITY_MAX) - return DECOMP_ERROR; - else - /* - * Take LCP down if the peer is sending too many bogons. - * We don't want to do this for a single or just a few - * instances since it could just be due to packet corruption. - */ - return DECOMP_FATALERROR; + goto sanity_error; } /* @@ -649,6 +636,16 @@ mppe_decompress(void *arg, unsigned char *ibuf, int isize, unsigned char *obuf, state->sanity_errors >>= 1; return osize; + +sanity_error: + if (state->sanity_errors < SANITY_MAX) + return DECOMP_ERROR; + else + /* Take LCP down if the peer is sending too many bogons. + * We don't want to do this for a single or just a few + * instances since it could just be due to packet corruption. + */ + return DECOMP_FATALERROR; } /* -- cgit v0.10.2 From 03654763148f9a3878b8b70c30d1ffce2fca3dff Mon Sep 17 00:00:00 2001 From: Sylvain Rochet Date: Sun, 26 Apr 2015 20:40:53 +0200 Subject: ppp: mppe: discard late packet in stateless mode When PPP is used over a link which does not guarantee packet ordering, we might get late MPPE packets. This is a problem because MPPE must be kept synchronized and the current implementation does not drop them and rekey 4095 times instead of 0, which is wrong. In order to prevent rekeying about a whole count space times (~ 4095 times), drop packets which are not within the forward 4096/2 window and increase sanity error counter. Signed-off-by: Sylvain Rochet Signed-off-by: David S. Miller diff --git a/drivers/net/ppp/ppp_mppe.c b/drivers/net/ppp/ppp_mppe.c index 692ee0f..05005c6 100644 --- a/drivers/net/ppp/ppp_mppe.c +++ b/drivers/net/ppp/ppp_mppe.c @@ -533,6 +533,13 @@ mppe_decompress(void *arg, unsigned char *ibuf, int isize, unsigned char *obuf, */ if (!state->stateful) { + /* Discard late packet */ + if ((ccount - state->ccount) % MPPE_CCOUNT_SPACE + > MPPE_CCOUNT_SPACE / 2) { + state->sanity_errors++; + goto sanity_error; + } + /* RFC 3078, sec 8.1. Rekey for every packet. */ while (state->ccount != ccount) { mppe_rekey(state, 0); -- cgit v0.10.2 From 41b4b3bc7905615f7617c88bf9cf4b3b04bda78f Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 17 Feb 2015 16:31:39 +0100 Subject: drivers: sh: Disable PM runtime for multi-platform sh73a0 with genpd If the default PM domain using PM_CLK is used for PM runtime, the real PM domain(s) cannot be registered from DT later. Hence do not enable it when running a multi-platform kernel with genpd support on an sh73a0. The R-Mobile PM domain driver will take care of PM runtime management of the module clocks. The default PM domain is still needed for: - platforms without genpd support, - the legacy (non-DT) case, where genpd may take over later, except for the C5 "always on" PM domain. Signed-off-by: Geert Uytterhoeven Signed-off-by: Simon Horman diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c index cd4c293..6b16a67 100644 --- a/drivers/sh/pm_runtime.c +++ b/drivers/sh/pm_runtime.c @@ -83,6 +83,7 @@ static int __init sh_pm_runtime_init(void) !of_machine_is_compatible("renesas,r8a73a4") && #ifndef CONFIG_PM_GENERIC_DOMAINS_OF !of_machine_is_compatible("renesas,r8a7740") && + !of_machine_is_compatible("renesas,sh73a0") && #endif !of_machine_is_compatible("renesas,r8a7778") && !of_machine_is_compatible("renesas,r8a7779") && @@ -91,8 +92,7 @@ static int __init sh_pm_runtime_init(void) !of_machine_is_compatible("renesas,r8a7792") && !of_machine_is_compatible("renesas,r8a7793") && !of_machine_is_compatible("renesas,r8a7794") && - !of_machine_is_compatible("renesas,sh7372") && - !of_machine_is_compatible("renesas,sh73a0")) + !of_machine_is_compatible("renesas,sh7372")) return 0; } -- cgit v0.10.2 From 230f259ffe4ad6d334bb100ce1c0c6a73ff18989 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 17 Feb 2015 16:43:03 +0100 Subject: drivers: sh: Disable PM runtime for multi-platform r8a73a4 with genpd If the default PM domain using PM_CLK is used for PM runtime, the real PM domain(s) cannot be registered from DT later. Hence do not enable it when running a multi-platform kernel with genpd support on an r8a73a4. The R-Mobile PM domain driver will take care of PM runtime management of the module clocks. The default PM domain is still needed for: - platforms without genpd support, - the legacy (non-DT) case, where genpd may take over later, except for the C5 "always on" PM domain. Signed-off-by: Geert Uytterhoeven Signed-off-by: Simon Horman diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c index 6b16a67..89b7785 100644 --- a/drivers/sh/pm_runtime.c +++ b/drivers/sh/pm_runtime.c @@ -80,8 +80,8 @@ static int __init sh_pm_runtime_init(void) if (IS_ENABLED(CONFIG_ARCH_SHMOBILE_MULTI)) { if (!of_machine_is_compatible("renesas,emev2") && !of_machine_is_compatible("renesas,r7s72100") && - !of_machine_is_compatible("renesas,r8a73a4") && #ifndef CONFIG_PM_GENERIC_DOMAINS_OF + !of_machine_is_compatible("renesas,r8a73a4") && !of_machine_is_compatible("renesas,r8a7740") && !of_machine_is_compatible("renesas,sh73a0") && #endif -- cgit v0.10.2 From 00170528f0486c08a18475af7fefa0a7679e042d Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 19 Feb 2015 18:46:46 +0100 Subject: drivers: sh: Remove test for now unsupported sh7372 Signed-off-by: Geert Uytterhoeven Signed-off-by: Simon Horman diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c index 89b7785..fe8875f 100644 --- a/drivers/sh/pm_runtime.c +++ b/drivers/sh/pm_runtime.c @@ -91,8 +91,7 @@ static int __init sh_pm_runtime_init(void) !of_machine_is_compatible("renesas,r8a7791") && !of_machine_is_compatible("renesas,r8a7792") && !of_machine_is_compatible("renesas,r8a7793") && - !of_machine_is_compatible("renesas,r8a7794") && - !of_machine_is_compatible("renesas,sh7372")) + !of_machine_is_compatible("renesas,r8a7794")) return 0; } -- cgit v0.10.2 From 53d2669844263fd5fdc70f0eb6a2eb8a21086d8e Mon Sep 17 00:00:00 2001 From: Ulf Hansson Date: Mon, 20 Apr 2015 16:02:31 +0200 Subject: ARM: ux500: Move GPIO regulator for SD-card into board DTSs The GPIO regulator for the SD-card isn't a ux500 SOC configuration, but instead it's specific to the board. Move the definition of it, into the board DTSs. Fixes: c94a4ab7af3f ("ARM: ux500: Disable the MMCI gpio-regulator by default") Signed-off-by: Ulf Hansson Reviewed-by: Bjorn Andersson Signed-off-by: Linus Walleij diff --git a/arch/arm/boot/dts/ste-dbx5x0.dtsi b/arch/arm/boot/dts/ste-dbx5x0.dtsi index bfd3f1c..2201cd5 100644 --- a/arch/arm/boot/dts/ste-dbx5x0.dtsi +++ b/arch/arm/boot/dts/ste-dbx5x0.dtsi @@ -1017,23 +1017,6 @@ status = "disabled"; }; - vmmci: regulator-gpio { - compatible = "regulator-gpio"; - - regulator-min-microvolt = <1800000>; - regulator-max-microvolt = <2900000>; - regulator-name = "mmci-reg"; - regulator-type = "voltage"; - - startup-delay-us = <100>; - enable-active-high; - - states = <1800000 0x1 - 2900000 0x0>; - - status = "disabled"; - }; - mcde@a0350000 { compatible = "stericsson,mcde"; reg = <0xa0350000 0x1000>, /* MCDE */ diff --git a/arch/arm/boot/dts/ste-href.dtsi b/arch/arm/boot/dts/ste-href.dtsi index bf8f0ed..8cf499a 100644 --- a/arch/arm/boot/dts/ste-href.dtsi +++ b/arch/arm/boot/dts/ste-href.dtsi @@ -111,6 +111,23 @@ pinctrl-1 = <&i2c3_sleep_mode>; }; + vmmci: regulator-gpio { + compatible = "regulator-gpio"; + + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <2900000>; + regulator-name = "mmci-reg"; + regulator-type = "voltage"; + + startup-delay-us = <100>; + enable-active-high; + + states = <1800000 0x1 + 2900000 0x0>; + + status = "disabled"; + }; + // External Micro SD slot sdi0_per1@80126000 { arm,primecell-periphid = <0x10480180>; diff --git a/arch/arm/boot/dts/ste-snowball.dts b/arch/arm/boot/dts/ste-snowball.dts index 206826a..65a7f63 100644 --- a/arch/arm/boot/dts/ste-snowball.dts +++ b/arch/arm/boot/dts/ste-snowball.dts @@ -146,8 +146,23 @@ }; vmmci: regulator-gpio { + compatible = "regulator-gpio"; + gpios = <&gpio7 4 0x4>; enable-gpio = <&gpio6 25 0x4>; + + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <2900000>; + regulator-name = "mmci-reg"; + regulator-type = "voltage"; + + startup-delay-us = <100>; + enable-active-high; + + states = <1800000 0x1 + 2900000 0x0>; + + status = "disabled"; }; // External Micro SD slot -- cgit v0.10.2 From f9a8c3914ba85f19c3360b19612d77c47adb8942 Mon Sep 17 00:00:00 2001 From: Ulf Hansson Date: Mon, 20 Apr 2015 16:02:32 +0200 Subject: ARM: ux500: Enable GPIO regulator for SD-card for HREF boards Fixes: c94a4ab7af3f ("ARM: ux500: Disable the MMCI gpio-regulator by default") Signed-off-by: Ulf Hansson Reviewed-by: Bjorn Andersson Signed-off-by: Linus Walleij diff --git a/arch/arm/boot/dts/ste-href.dtsi b/arch/arm/boot/dts/ste-href.dtsi index 8cf499a..744c1e3 100644 --- a/arch/arm/boot/dts/ste-href.dtsi +++ b/arch/arm/boot/dts/ste-href.dtsi @@ -124,8 +124,6 @@ states = <1800000 0x1 2900000 0x0>; - - status = "disabled"; }; // External Micro SD slot -- cgit v0.10.2 From 11133db7a836b0cb411faa048f07a38e994d1382 Mon Sep 17 00:00:00 2001 From: Ulf Hansson Date: Mon, 20 Apr 2015 16:02:33 +0200 Subject: ARM: ux500: Enable GPIO regulator for SD-card for snowball Fixes: c94a4ab7af3f ("ARM: ux500: Disable the MMCI gpio-regulator by default") Signed-off-by: Ulf Hansson Reviewed-by: Bjorn Andersson Signed-off-by: Linus Walleij diff --git a/arch/arm/boot/dts/ste-snowball.dts b/arch/arm/boot/dts/ste-snowball.dts index 65a7f63..1bc84eb 100644 --- a/arch/arm/boot/dts/ste-snowball.dts +++ b/arch/arm/boot/dts/ste-snowball.dts @@ -161,8 +161,6 @@ states = <1800000 0x1 2900000 0x0>; - - status = "disabled"; }; // External Micro SD slot -- cgit v0.10.2 From 750e30d4076ae5e02ad13a376e96c95a2627742c Mon Sep 17 00:00:00 2001 From: Gregory CLEMENT Date: Tue, 14 Apr 2015 11:50:13 +0200 Subject: ARM: mvebu: armada-xp-openblocks-ax3-4: Disable internal RTC There is no crystal connected to the internal RTC on the Open Block AX3. So let's disable it in order to prevent the kernel probing the driver uselessly. Eventually this patches removes the following warning message from the boot log: "rtc-mv d0010300.rtc: internal RTC not ticking" Acked-by: Andrew Lunn Signed-off-by: Gregory CLEMENT Cc: # v3.8 + diff --git a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts index e3b08fb..990e8a2 100644 --- a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts +++ b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts @@ -105,6 +105,10 @@ }; internal-regs { + rtc@10300 { + /* No crystal connected to the internal RTC */ + status = "disabled"; + }; serial@12000 { status = "okay"; }; -- cgit v0.10.2 From 878a84d5a8a18a4ab241d40cebb791d6aedf5605 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Mon, 20 Apr 2015 11:14:19 +0100 Subject: arm64: add missing data types in smp_load_acquire/smp_store_release Commit 8053871d0f7f ("smp: Fix smp_call_function_single_async() locking") introduced a call to smp_load_acquire() with a u16 argument, but we only cared about u32 and u64 types in that function so far. This resulted in a compiler warning fortunately, pointing at an uninitialized use. Due to the implementation structure the compiler misses that bug in the smp_store_release(), though. Add the u16 and u8 variants using ldarh/stlrh and ldarb/stlrb, respectively. Together with the compiletime_assert_atomic_type() check this should cover all cases now. Acked-by: Will Deacon Signed-off-by: Andre Przywara Signed-off-by: Will Deacon diff --git a/arch/arm64/include/asm/barrier.h b/arch/arm64/include/asm/barrier.h index a5abb00..71f19c4 100644 --- a/arch/arm64/include/asm/barrier.h +++ b/arch/arm64/include/asm/barrier.h @@ -65,6 +65,14 @@ do { \ do { \ compiletime_assert_atomic_type(*p); \ switch (sizeof(*p)) { \ + case 1: \ + asm volatile ("stlrb %w1, %0" \ + : "=Q" (*p) : "r" (v) : "memory"); \ + break; \ + case 2: \ + asm volatile ("stlrh %w1, %0" \ + : "=Q" (*p) : "r" (v) : "memory"); \ + break; \ case 4: \ asm volatile ("stlr %w1, %0" \ : "=Q" (*p) : "r" (v) : "memory"); \ @@ -81,6 +89,14 @@ do { \ typeof(*p) ___p1; \ compiletime_assert_atomic_type(*p); \ switch (sizeof(*p)) { \ + case 1: \ + asm volatile ("ldarb %w0, %1" \ + : "=r" (___p1) : "Q" (*p) : "memory"); \ + break; \ + case 2: \ + asm volatile ("ldarh %w0, %1" \ + : "=r" (___p1) : "Q" (*p) : "memory"); \ + break; \ case 4: \ asm volatile ("ldar %w0, %1" \ : "=r" (___p1) : "Q" (*p) : "memory"); \ -- cgit v0.10.2 From 6544e67bfb1bf55bcf3c0f6b37631917e9acfb74 Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Wed, 22 Apr 2015 18:16:33 +0100 Subject: ARM64: Enable CONFIG_GENERIC_IRQ_SHOW_LEVEL Since several interrupt controllers including GIC support both edge and level triggered interrupts, it's useful to provide that information in /proc/interrupts even on ARM64 similar to ARM and PPC. This is based on Geert Uytterhoeven's commit 7c07005eea96 ("ARM: 8339/1: Enable CONFIG_GENERIC_IRQ_SHOW_LEVEL") Signed-off-by: Sudeep Holla Signed-off-by: Will Deacon diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 4269dba..7796af4 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -31,6 +31,7 @@ config ARM64 select GENERIC_EARLY_IOREMAP select GENERIC_IRQ_PROBE select GENERIC_IRQ_SHOW + select GENERIC_IRQ_SHOW_LEVEL select GENERIC_PCI_IOMAP select GENERIC_SCHED_CLOCK select GENERIC_SMP_IDLE_THREAD -- cgit v0.10.2 From 6829e274a623187c24f7cfc0e3d35f25d087fcc5 Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Thu, 23 Apr 2015 12:46:16 +0100 Subject: arm64: dma-mapping: always clear allocated buffers Buffers allocated by dma_alloc_coherent() are always zeroed on Alpha, ARM (32bit), MIPS, PowerPC, x86/x86_64 and probably other architectures. It turned out that some drivers rely on this 'feature'. Allocated buffer might be also exposed to userspace with dma_mmap() call, so clearing it is desired from security point of view to avoid exposing random memory to userspace. This patch unifies dma_alloc_coherent() behavior on ARM64 architecture with other implementations by unconditionally zeroing allocated buffer. Cc: # v3.14+ Signed-off-by: Marek Szyprowski Signed-off-by: Will Deacon diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c index ef7d112..e0f14ee 100644 --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -67,8 +67,7 @@ static void *__alloc_from_pool(size_t size, struct page **ret_page, gfp_t flags) *ret_page = phys_to_page(phys); ptr = (void *)val; - if (flags & __GFP_ZERO) - memset(ptr, 0, size); + memset(ptr, 0, size); } return ptr; @@ -113,8 +112,7 @@ static void *__dma_alloc_coherent(struct device *dev, size_t size, *dma_handle = phys_to_dma(dev, page_to_phys(page)); addr = page_address(page); - if (flags & __GFP_ZERO) - memset(addr, 0, size); + memset(addr, 0, size); return addr; } else { return swiotlb_alloc_coherent(dev, size, dma_handle, flags); -- cgit v0.10.2 From 325d73bf8fea8af2227240b7305253fb052d3a68 Mon Sep 17 00:00:00 2001 From: Bob Liu Date: Fri, 3 Apr 2015 14:42:58 +0800 Subject: xen/blkback: safely unmap purge persistent grants MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit c43cf3ea8385 ("xen-blkback: safely unmap grants in case they are still in use") use gnttab_unmap_refs_async() to wait until the mapped pages are no longer in use before unmapping them, but that commit missed the persistent case. Purge persistent pages can't be unmapped either unless no longer in use. Signed-off-by: Bob Liu Acked-by: Roger Pau Monné Signed-off-by: David Vrabel diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index bd2b3bb..48e98f2 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -329,8 +329,18 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work) struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST]; struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST]; struct persistent_gnt *persistent_gnt; - int ret, segs_to_unmap = 0; + int segs_to_unmap = 0; struct xen_blkif *blkif = container_of(work, typeof(*blkif), persistent_purge_work); + struct gntab_unmap_queue_data unmap_data; + struct completion unmap_completion; + + init_completion(&unmap_completion); + + unmap_data.data = &unmap_completion; + unmap_data.done = &free_persistent_gnts_unmap_callback; + unmap_data.pages = pages; + unmap_data.unmap_ops = unmap; + unmap_data.kunmap_ops = NULL; while(!list_empty(&blkif->persistent_purge_list)) { persistent_gnt = list_first_entry(&blkif->persistent_purge_list, @@ -346,17 +356,19 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work) pages[segs_to_unmap] = persistent_gnt->page; if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST) { - ret = gnttab_unmap_refs(unmap, NULL, pages, - segs_to_unmap); - BUG_ON(ret); + unmap_data.count = segs_to_unmap; + gnttab_unmap_refs_async(&unmap_data); + wait_for_completion(&unmap_completion); + put_free_pages(blkif, pages, segs_to_unmap); segs_to_unmap = 0; } kfree(persistent_gnt); } if (segs_to_unmap > 0) { - ret = gnttab_unmap_refs(unmap, NULL, pages, segs_to_unmap); - BUG_ON(ret); + unmap_data.count = segs_to_unmap; + gnttab_unmap_refs_async(&unmap_data); + wait_for_completion(&unmap_completion); put_free_pages(blkif, pages, segs_to_unmap); } } -- cgit v0.10.2 From b44166cd46e28dd608d5baa5873047a40f32919c Mon Sep 17 00:00:00 2001 From: Bob Liu Date: Fri, 3 Apr 2015 14:42:59 +0800 Subject: xen/grant: introduce func gnttab_unmap_refs_sync() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are several place using gnttab async unmap and wait for completion, so move the common code to a function gnttab_unmap_refs_sync(). Signed-off-by: Bob Liu Acked-by: Roger Pau Monné Acked-by: Konrad Rzeszutek Wilk Signed-off-by: David Vrabel diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 48e98f2..713fc9f 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -265,17 +265,6 @@ static void put_persistent_gnt(struct xen_blkif *blkif, atomic_dec(&blkif->persistent_gnt_in_use); } -static void free_persistent_gnts_unmap_callback(int result, - struct gntab_unmap_queue_data *data) -{ - struct completion *c = data->data; - - /* BUG_ON used to reproduce existing behaviour, - but is this the best way to deal with this? */ - BUG_ON(result); - complete(c); -} - static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root, unsigned int num) { @@ -285,12 +274,7 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root, struct rb_node *n; int segs_to_unmap = 0; struct gntab_unmap_queue_data unmap_data; - struct completion unmap_completion; - - init_completion(&unmap_completion); - unmap_data.data = &unmap_completion; - unmap_data.done = &free_persistent_gnts_unmap_callback; unmap_data.pages = pages; unmap_data.unmap_ops = unmap; unmap_data.kunmap_ops = NULL; @@ -310,8 +294,7 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root, !rb_next(&persistent_gnt->node)) { unmap_data.count = segs_to_unmap; - gnttab_unmap_refs_async(&unmap_data); - wait_for_completion(&unmap_completion); + BUG_ON(gnttab_unmap_refs_sync(&unmap_data)); put_free_pages(blkif, pages, segs_to_unmap); segs_to_unmap = 0; @@ -332,12 +315,7 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work) int segs_to_unmap = 0; struct xen_blkif *blkif = container_of(work, typeof(*blkif), persistent_purge_work); struct gntab_unmap_queue_data unmap_data; - struct completion unmap_completion; - init_completion(&unmap_completion); - - unmap_data.data = &unmap_completion; - unmap_data.done = &free_persistent_gnts_unmap_callback; unmap_data.pages = pages; unmap_data.unmap_ops = unmap; unmap_data.kunmap_ops = NULL; @@ -357,9 +335,7 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work) if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST) { unmap_data.count = segs_to_unmap; - gnttab_unmap_refs_async(&unmap_data); - wait_for_completion(&unmap_completion); - + BUG_ON(gnttab_unmap_refs_sync(&unmap_data)); put_free_pages(blkif, pages, segs_to_unmap); segs_to_unmap = 0; } @@ -367,8 +343,7 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work) } if (segs_to_unmap > 0) { unmap_data.count = segs_to_unmap; - gnttab_unmap_refs_async(&unmap_data); - wait_for_completion(&unmap_completion); + BUG_ON(gnttab_unmap_refs_sync(&unmap_data)); put_free_pages(blkif, pages, segs_to_unmap); } } diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c index d5bb1a3..8927485 100644 --- a/drivers/xen/gntdev.c +++ b/drivers/xen/gntdev.c @@ -327,30 +327,10 @@ static int map_grant_pages(struct grant_map *map) return err; } -struct unmap_grant_pages_callback_data -{ - struct completion completion; - int result; -}; - -static void unmap_grant_callback(int result, - struct gntab_unmap_queue_data *data) -{ - struct unmap_grant_pages_callback_data* d = data->data; - - d->result = result; - complete(&d->completion); -} - static int __unmap_grant_pages(struct grant_map *map, int offset, int pages) { int i, err = 0; struct gntab_unmap_queue_data unmap_data; - struct unmap_grant_pages_callback_data data; - - init_completion(&data.completion); - unmap_data.data = &data; - unmap_data.done= &unmap_grant_callback; if (map->notify.flags & UNMAP_NOTIFY_CLEAR_BYTE) { int pgno = (map->notify.addr >> PAGE_SHIFT); @@ -367,11 +347,9 @@ static int __unmap_grant_pages(struct grant_map *map, int offset, int pages) unmap_data.pages = map->pages + offset; unmap_data.count = pages; - gnttab_unmap_refs_async(&unmap_data); - - wait_for_completion(&data.completion); - if (data.result) - return data.result; + err = gnttab_unmap_refs_sync(&unmap_data); + if (err) + return err; for (i = 0; i < pages; i++) { if (map->unmap_ops[offset+i].status) diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index 17972fb..b1c7170 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -123,6 +123,11 @@ struct gnttab_ops { int (*query_foreign_access)(grant_ref_t ref); }; +struct unmap_refs_callback_data { + struct completion completion; + int result; +}; + static struct gnttab_ops *gnttab_interface; static int grant_table_version; @@ -863,6 +868,29 @@ void gnttab_unmap_refs_async(struct gntab_unmap_queue_data* item) } EXPORT_SYMBOL_GPL(gnttab_unmap_refs_async); +static void unmap_refs_callback(int result, + struct gntab_unmap_queue_data *data) +{ + struct unmap_refs_callback_data *d = data->data; + + d->result = result; + complete(&d->completion); +} + +int gnttab_unmap_refs_sync(struct gntab_unmap_queue_data *item) +{ + struct unmap_refs_callback_data data; + + init_completion(&data.completion); + item->data = &data; + item->done = &unmap_refs_callback; + gnttab_unmap_refs_async(item); + wait_for_completion(&data.completion); + + return data.result; +} +EXPORT_SYMBOL_GPL(gnttab_unmap_refs_sync); + static int gnttab_map_frames_v1(xen_pfn_t *frames, unsigned int nr_gframes) { int rc; diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h index 143ca5f..4478f4b 100644 --- a/include/xen/grant_table.h +++ b/include/xen/grant_table.h @@ -191,6 +191,7 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops, struct gnttab_unmap_grant_ref *kunmap_ops, struct page **pages, unsigned int count); void gnttab_unmap_refs_async(struct gntab_unmap_queue_data* item); +int gnttab_unmap_refs_sync(struct gntab_unmap_queue_data *item); /* Perform a batch of grant map/copy operations. Retry every batch slot -- cgit v0.10.2 From cfe8c59762244251fd9a5e281d48808095ff4090 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Tue, 14 Apr 2015 11:05:04 -0300 Subject: ARM: dts: imx23-olinuxino: Fix polarity of LED GPIO On imx23-olinuxino the LED turns on when level logic high is aplied to GPIO2_1. Fix the gpios property accordingly. Fixes: b34aa1850244 ("ARM: dts: imx23-olinuxino: Remove unneeded "default-on"") Reported-by: Stefan Wahren Signed-off-by: Fabio Estevam Tested-by: Stefan Wahren Cc: stable@vger.kernel.org Signed-off-by: Shawn Guo diff --git a/arch/arm/boot/dts/imx23-olinuxino.dts b/arch/arm/boot/dts/imx23-olinuxino.dts index 7e6eef2..839a3b9 100644 --- a/arch/arm/boot/dts/imx23-olinuxino.dts +++ b/arch/arm/boot/dts/imx23-olinuxino.dts @@ -12,6 +12,7 @@ */ /dts-v1/; +#include #include "imx23.dtsi" / { @@ -122,7 +123,7 @@ user { label = "green"; - gpios = <&gpio2 1 1>; + gpios = <&gpio2 1 GPIO_ACTIVE_HIGH>; }; }; }; -- cgit v0.10.2 From 0fdebe1a2f4d3a8fc03754022fabf8ba95e131a3 Mon Sep 17 00:00:00 2001 From: Stefan Wahren Date: Tue, 14 Apr 2015 20:37:26 +0000 Subject: ARM: dts: imx23-olinuxino: Fix dr_mode of usb0 The dr_mode of usb0 on imx233-olinuxino is left to default "otg". Since the green LED (GPIO2_1) on imx233-olinuxino is connected to the same pin as USB_OTG_ID it's possible to disable USB host by LED toggling: echo 0 > /sys/class/leds/green/brightness [ 1068.890000] ci_hdrc ci_hdrc.0: remove, state 1 [ 1068.890000] usb usb1: USB disconnect, device number 1 [ 1068.920000] usb 1-1: USB disconnect, device number 2 [ 1068.920000] usb 1-1.1: USB disconnect, device number 3 [ 1069.070000] usb 1-1.2: USB disconnect, device number 4 [ 1069.450000] ci_hdrc ci_hdrc.0: USB bus 1 deregistered [ 1074.460000] ci_hdrc ci_hdrc.0: timeout waiting for 00000800 in 11 This patch fixes the issue by setting dr_mode to "host" in the dts file. Reported-by: Harald Geyer Signed-off-by: Stefan Wahren Reviewed-by: Fabio Estevam Reviewed-by: Marek Vasut Acked-by: Peter Chen Fixes: b49312948285 ("ARM: dts: imx23-olinuxino: Add USB host support") Cc: stable@vger.kernel.org Signed-off-by: Shawn Guo diff --git a/arch/arm/boot/dts/imx23-olinuxino.dts b/arch/arm/boot/dts/imx23-olinuxino.dts index 839a3b9..8204539 100644 --- a/arch/arm/boot/dts/imx23-olinuxino.dts +++ b/arch/arm/boot/dts/imx23-olinuxino.dts @@ -94,6 +94,7 @@ ahb@80080000 { usb0: usb@80080000 { + dr_mode = "host"; vbus-supply = <®_usb0_vbus>; status = "okay"; }; -- cgit v0.10.2 From 61b8c7b58450846b63b35e90866d2eb4fde44834 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Mon, 20 Apr 2015 15:51:38 +0200 Subject: ARM: mach-imx: devices: platform-sdhci-esdhc-imx: fix broken email address My Pengutronix address is not valid anymore, redirect people to the Pengutronix kernel team. Reported-by: Harald Geyer Signed-off-by: Wolfram Sang Acked-by: Robert Schwebel Signed-off-by: Shawn Guo diff --git a/arch/arm/mach-imx/devices/platform-sdhci-esdhc-imx.c b/arch/arm/mach-imx/devices/platform-sdhci-esdhc-imx.c index fb8d4a2..a5edd7d 100644 --- a/arch/arm/mach-imx/devices/platform-sdhci-esdhc-imx.c +++ b/arch/arm/mach-imx/devices/platform-sdhci-esdhc-imx.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2010 Pengutronix, Wolfram Sang + * Copyright (C) 2010 Pengutronix, Wolfram Sang * * This program is free software; you can redistribute it and/or modify it under * the terms of the GNU General Public License version 2 as published by the -- cgit v0.10.2 From a4811622fea15fdc78102069f573061fc87f3570 Mon Sep 17 00:00:00 2001 From: Qipeng Zha Date: Fri, 10 Apr 2015 06:25:10 +0800 Subject: gpiolib: change gpio pin from unsigned to signed in acpi callback The signed error will be wrongly used as valid gpio offset Reported-by: David Binderman Signed-off-by: Alan Cox Signed-off-by: Qipeng Zha Acked-by: Mika Westerberg Signed-off-by: Linus Walleij diff --git a/drivers/gpio/gpiolib-acpi.c b/drivers/gpio/gpiolib-acpi.c index d2303d5..725d161 100644 --- a/drivers/gpio/gpiolib-acpi.c +++ b/drivers/gpio/gpiolib-acpi.c @@ -550,7 +550,7 @@ acpi_gpio_adr_space_handler(u32 function, acpi_physical_address address, length = min(agpio->pin_table_length, (u16)(pin_index + bits)); for (i = pin_index; i < length; ++i) { - unsigned pin = agpio->pin_table[i]; + int pin = agpio->pin_table[i]; struct acpi_gpio_connection *conn; struct gpio_desc *desc; bool found; -- cgit v0.10.2 From a526973e0291b245aefe12023a7f775f3c0e59a2 Mon Sep 17 00:00:00 2001 From: Andrew Andrianov Date: Sat, 11 Apr 2015 23:29:19 +0300 Subject: pinctrl: mvebu: Fix mapping of pin 63 (gpo -> gpio) Signed-off-by: Andrew Andrianov Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/mvebu/pinctrl-armada-370.c b/drivers/pinctrl/mvebu/pinctrl-armada-370.c index 42f930f..03aa58c 100644 --- a/drivers/pinctrl/mvebu/pinctrl-armada-370.c +++ b/drivers/pinctrl/mvebu/pinctrl-armada-370.c @@ -364,7 +364,7 @@ static struct mvebu_mpp_mode mv88f6710_mpp_modes[] = { MPP_FUNCTION(0x5, "audio", "mclk"), MPP_FUNCTION(0x6, "uart0", "cts")), MPP_MODE(63, - MPP_FUNCTION(0x0, "gpo", NULL), + MPP_FUNCTION(0x0, "gpio", NULL), MPP_FUNCTION(0x1, "spi0", "sck"), MPP_FUNCTION(0x2, "tclk", NULL)), MPP_MODE(64, -- cgit v0.10.2 From dc391502fdbf97a9cabdc58ba8c915175383f681 Mon Sep 17 00:00:00 2001 From: "Ivan T. Ivanov" Date: Fri, 17 Apr 2015 17:50:49 +0300 Subject: pinctrl: qcom-spmi: Fix pin direction configuration Pin direction configuration was incorrectly overwritten by output and function values in set_mux(). Fix this. Signed-off-by: Ivan T. Ivanov Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c index b2d2221..de684ca 100644 --- a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c +++ b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c @@ -260,6 +260,7 @@ static int pmic_gpio_set_mux(struct pinctrl_dev *pctldev, unsigned function, val = 1; } + val = val << PMIC_GPIO_REG_MODE_DIR_SHIFT; val |= pad->function << PMIC_GPIO_REG_MODE_FUNCTION_SHIFT; val |= pad->out_value & PMIC_GPIO_REG_MODE_VALUE_SHIFT; diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c b/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c index 8f36c5f..890df16 100644 --- a/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c +++ b/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c @@ -370,6 +370,7 @@ static int pmic_mpp_set_mux(struct pinctrl_dev *pctldev, unsigned function, } } + val = val << PMIC_MPP_REG_MODE_DIR_SHIFT; val |= pad->function << PMIC_MPP_REG_MODE_FUNCTION_SHIFT; val |= pad->out_value & PMIC_MPP_REG_MODE_VALUE_MASK; -- cgit v0.10.2 From 5dca0d9147458be9b9363b8a484aa77d710b412a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Radim=20Kr=C4=8Dm=C3=A1=C5=99?= Date: Wed, 25 Mar 2015 12:08:14 +0100 Subject: kvm: x86: fix kvmclock update protocol MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The kvmclock spec says that the host will increment a version field to an odd number, then update stuff, then increment it to an even number. The host is buggy and doesn't do this, and the result is observable when one vcpu reads another vcpu's kvmclock data. There's no good way for a guest kernel to keep its vdso from reading a different vcpu's kvmclock data, but we don't need to care about changing VCPUs as long as we read a consistent data from kvmclock. (VCPU can change outside of this loop too, so it doesn't matter if we return a value not fit for this VCPU.) Based on a patch by Radim Krčmář. Reviewed-by: Radim Krčmář Acked-by: Marcelo Tosatti Signed-off-by: Paolo Bonzini diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ed31c31..c73efcd 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1669,12 +1669,28 @@ static int kvm_guest_time_update(struct kvm_vcpu *v) &guest_hv_clock, sizeof(guest_hv_clock)))) return 0; - /* - * The interface expects us to write an even number signaling that the - * update is finished. Since the guest won't see the intermediate - * state, we just increase by 2 at the end. + /* This VCPU is paused, but it's legal for a guest to read another + * VCPU's kvmclock, so we really have to follow the specification where + * it says that version is odd if data is being modified, and even after + * it is consistent. + * + * Version field updates must be kept separate. This is because + * kvm_write_guest_cached might use a "rep movs" instruction, and + * writes within a string instruction are weakly ordered. So there + * are three writes overall. + * + * As a small optimization, only write the version field in the first + * and third write. The vcpu->pv_time cache is still valid, because the + * version field is the first in the struct. */ - vcpu->hv_clock.version = guest_hv_clock.version + 2; + BUILD_BUG_ON(offsetof(struct pvclock_vcpu_time_info, version) != 0); + + vcpu->hv_clock.version = guest_hv_clock.version + 1; + kvm_write_guest_cached(v->kvm, &vcpu->pv_time, + &vcpu->hv_clock, + sizeof(vcpu->hv_clock.version)); + + smp_wmb(); /* retain PVCLOCK_GUEST_STOPPED if set in guest copy */ pvclock_flags = (guest_hv_clock.flags & PVCLOCK_GUEST_STOPPED); @@ -1695,6 +1711,13 @@ static int kvm_guest_time_update(struct kvm_vcpu *v) kvm_write_guest_cached(v->kvm, &vcpu->pv_time, &vcpu->hv_clock, sizeof(vcpu->hv_clock)); + + smp_wmb(); + + vcpu->hv_clock.version++; + kvm_write_guest_cached(v->kvm, &vcpu->pv_time, + &vcpu->hv_clock, + sizeof(vcpu->hv_clock.version)); return 0; } -- cgit v0.10.2 From 73459e2a1ada09a68c02cc5b73f3116fc8194b3d Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 23 Apr 2015 13:20:18 +0200 Subject: x86: pvclock: Really remove the sched notifier for cross-cpu migrations This reverts commits 0a4e6be9ca17c54817cf814b4b5aa60478c6df27 and 80f7fdb1c7f0f9266421f823964fd1962681f6ce. The task migration notifier was originally introduced in order to support the pvclock vsyscall with non-synchronized TSC, but KVM only supports it with synchronized TSC. Hence, on KVM the race condition is only needed due to a bad implementation on the host side, and even then it's so rare that it's mostly theoretical. As far as KVM is concerned it's possible to fix the host, avoiding the additional complexity in the vDSO and the (re)introduction of the task migration notifier. Xen, on the other hand, hasn't yet implemented vsyscall support at all, so we do not care about its plans for non-synchronized TSC. Reported-by: Peter Zijlstra Suggested-by: Marcelo Tosatti Signed-off-by: Paolo Bonzini diff --git a/arch/x86/include/asm/pvclock.h b/arch/x86/include/asm/pvclock.h index 25b1cc0..d6b078e 100644 --- a/arch/x86/include/asm/pvclock.h +++ b/arch/x86/include/asm/pvclock.h @@ -95,7 +95,6 @@ unsigned __pvclock_read_cycles(const struct pvclock_vcpu_time_info *src, struct pvclock_vsyscall_time_info { struct pvclock_vcpu_time_info pvti; - u32 migrate_count; } __attribute__((__aligned__(SMP_CACHE_BYTES))); #define PVTI_SIZE sizeof(struct pvclock_vsyscall_time_info) diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index e5ecd20..2f355d2 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -141,46 +141,7 @@ void pvclock_read_wallclock(struct pvclock_wall_clock *wall_clock, set_normalized_timespec(ts, now.tv_sec, now.tv_nsec); } -static struct pvclock_vsyscall_time_info *pvclock_vdso_info; - -static struct pvclock_vsyscall_time_info * -pvclock_get_vsyscall_user_time_info(int cpu) -{ - if (!pvclock_vdso_info) { - BUG(); - return NULL; - } - - return &pvclock_vdso_info[cpu]; -} - -struct pvclock_vcpu_time_info *pvclock_get_vsyscall_time_info(int cpu) -{ - return &pvclock_get_vsyscall_user_time_info(cpu)->pvti; -} - #ifdef CONFIG_X86_64 -static int pvclock_task_migrate(struct notifier_block *nb, unsigned long l, - void *v) -{ - struct task_migration_notifier *mn = v; - struct pvclock_vsyscall_time_info *pvti; - - pvti = pvclock_get_vsyscall_user_time_info(mn->from_cpu); - - /* this is NULL when pvclock vsyscall is not initialized */ - if (unlikely(pvti == NULL)) - return NOTIFY_DONE; - - pvti->migrate_count++; - - return NOTIFY_DONE; -} - -static struct notifier_block pvclock_migrate = { - .notifier_call = pvclock_task_migrate, -}; - /* * Initialize the generic pvclock vsyscall state. This will allocate * a/some page(s) for the per-vcpu pvclock information, set up a @@ -194,17 +155,12 @@ int __init pvclock_init_vsyscall(struct pvclock_vsyscall_time_info *i, WARN_ON (size != PVCLOCK_VSYSCALL_NR_PAGES*PAGE_SIZE); - pvclock_vdso_info = i; - for (idx = 0; idx <= (PVCLOCK_FIXMAP_END-PVCLOCK_FIXMAP_BEGIN); idx++) { __set_fixmap(PVCLOCK_FIXMAP_BEGIN + idx, __pa(i) + (idx*PAGE_SIZE), PAGE_KERNEL_VVAR); } - - register_task_migration_notifier(&pvclock_migrate); - return 0; } #endif diff --git a/arch/x86/vdso/vclock_gettime.c b/arch/x86/vdso/vclock_gettime.c index 40d2473..9793322 100644 --- a/arch/x86/vdso/vclock_gettime.c +++ b/arch/x86/vdso/vclock_gettime.c @@ -82,15 +82,18 @@ static notrace cycle_t vread_pvclock(int *mode) cycle_t ret; u64 last; u32 version; - u32 migrate_count; u8 flags; unsigned cpu, cpu1; /* - * When looping to get a consistent (time-info, tsc) pair, we - * also need to deal with the possibility we can switch vcpus, - * so make sure we always re-fetch time-info for the current vcpu. + * Note: hypervisor must guarantee that: + * 1. cpu ID number maps 1:1 to per-CPU pvclock time info. + * 2. that per-CPU pvclock time info is updated if the + * underlying CPU changes. + * 3. that version is increased whenever underlying CPU + * changes. + * */ do { cpu = __getcpu() & VGETCPU_CPU_MASK; @@ -99,27 +102,20 @@ static notrace cycle_t vread_pvclock(int *mode) * __getcpu() calls (Gleb). */ - /* Make sure migrate_count will change if we leave the VCPU. */ - do { - pvti = get_pvti(cpu); - migrate_count = pvti->migrate_count; - - cpu1 = cpu; - cpu = __getcpu() & VGETCPU_CPU_MASK; - } while (unlikely(cpu != cpu1)); + pvti = get_pvti(cpu); version = __pvclock_read_cycles(&pvti->pvti, &ret, &flags); /* * Test we're still on the cpu as well as the version. - * - We must read TSC of pvti's VCPU. - * - KVM doesn't follow the versioning protocol, so data could - * change before version if we left the VCPU. + * We could have been migrated just after the first + * vgetcpu but before fetching the version, so we + * wouldn't notice a version change. */ - smp_rmb(); - } while (unlikely((pvti->pvti.version & 1) || - pvti->pvti.version != version || - pvti->migrate_count != migrate_count)); + cpu1 = __getcpu() & VGETCPU_CPU_MASK; + } while (unlikely(cpu != cpu1 || + (pvti->pvti.version & 1) || + pvti->pvti.version != version)); if (unlikely(!(flags & PVCLOCK_TSC_STABLE_BIT))) *mode = VCLOCK_NONE; diff --git a/include/linux/sched.h b/include/linux/sched.h index 8222ae4..26a2e61 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -175,14 +175,6 @@ extern void get_iowait_load(unsigned long *nr_waiters, unsigned long *load); extern void calc_global_load(unsigned long ticks); extern void update_cpu_load_nohz(void); -/* Notifier for when a task gets migrated to a new CPU */ -struct task_migration_notifier { - struct task_struct *task; - int from_cpu; - int to_cpu; -}; -extern void register_task_migration_notifier(struct notifier_block *n); - extern unsigned long get_parent_ip(unsigned long addr); extern void dump_cpu_task(int cpu); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index f9123a8..fe22f75 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1016,13 +1016,6 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) rq_clock_skip_update(rq, true); } -static ATOMIC_NOTIFIER_HEAD(task_migration_notifier); - -void register_task_migration_notifier(struct notifier_block *n) -{ - atomic_notifier_chain_register(&task_migration_notifier, n); -} - #ifdef CONFIG_SMP void set_task_cpu(struct task_struct *p, unsigned int new_cpu) { @@ -1053,18 +1046,10 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) trace_sched_migrate_task(p, new_cpu); if (task_cpu(p) != new_cpu) { - struct task_migration_notifier tmn; - if (p->sched_class->migrate_task_rq) p->sched_class->migrate_task_rq(p, new_cpu); p->se.nr_migrations++; perf_sw_event_sched(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 0); - - tmn.task = p; - tmn.from_cpu = task_cpu(p); - tmn.to_cpu = new_cpu; - - atomic_notifier_call_chain(&task_migration_notifier, 0, &tmn); } __set_task_cpu(p, new_cpu); -- cgit v0.10.2 From b421ed15d2c3039eb724680e4de1e4b2bd196a9a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Thu, 16 Apr 2015 11:17:27 +0900 Subject: drm/radeon: Use drm_calloc_ab for CS relocs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The number of relocs is passed in by userspace and can be large. It has been observed to cause kcalloc failures in the wild. Cc: stable@vger.kernel.org Reviewed-by: Christian König Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 4d0f96c..ab39b85 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -88,7 +88,7 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) p->dma_reloc_idx = 0; /* FIXME: we assume that each relocs use 4 dwords */ p->nrelocs = chunk->length_dw / 4; - p->relocs = kcalloc(p->nrelocs, sizeof(struct radeon_bo_list), GFP_KERNEL); + p->relocs = drm_calloc_large(p->nrelocs, sizeof(struct radeon_bo_list)); if (p->relocs == NULL) { return -ENOMEM; } @@ -428,7 +428,7 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo } } kfree(parser->track); - kfree(parser->relocs); + drm_free_large(parser->relocs); drm_free_large(parser->vm_bos); for (i = 0; i < parser->nchunks; i++) drm_free_large(parser->chunks[i].kdata); -- cgit v0.10.2 From 304f07e9c8d302cf3183248cbabb40598eb5b732 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 31 Mar 2015 10:33:05 -0400 Subject: drm/radeon: fix ordering of AVI packet setup Set the line first, then enable the stream. May fix pink line problems on some displays. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c index c18d4ec..2379979 100644 --- a/drivers/gpu/drm/radeon/evergreen_hdmi.c +++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c @@ -219,13 +219,13 @@ void evergreen_set_avi_packet(struct radeon_device *rdev, u32 offset, WREG32(AFMT_AVI_INFO3 + offset, frame[0xC] | (frame[0xD] << 8) | (buffer[1] << 24)); - WREG32_OR(HDMI_INFOFRAME_CONTROL0 + offset, - HDMI_AVI_INFO_SEND | /* enable AVI info frames */ - HDMI_AVI_INFO_CONT); /* required for audio info values to be updated */ - WREG32_P(HDMI_INFOFRAME_CONTROL1 + offset, - HDMI_AVI_INFO_LINE(2), /* anything other than 0 */ - ~HDMI_AVI_INFO_LINE_MASK); + HDMI_AVI_INFO_LINE(2), /* anything other than 0 */ + ~HDMI_AVI_INFO_LINE_MASK); + + WREG32_OR(HDMI_INFOFRAME_CONTROL0 + offset, + HDMI_AVI_INFO_SEND | /* enable AVI info frames */ + HDMI_AVI_INFO_CONT); /* required for audio info values to be updated */ } void dce4_hdmi_audio_set_dto(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c index dd6606b..e85894a 100644 --- a/drivers/gpu/drm/radeon/r600_hdmi.c +++ b/drivers/gpu/drm/radeon/r600_hdmi.c @@ -228,12 +228,13 @@ void r600_set_avi_packet(struct radeon_device *rdev, u32 offset, WREG32(HDMI0_AVI_INFO3 + offset, frame[0xC] | (frame[0xD] << 8) | (buffer[1] << 24)); + WREG32_OR(HDMI0_INFOFRAME_CONTROL1 + offset, + HDMI0_AVI_INFO_LINE(2)); /* anything other than 0 */ + WREG32_OR(HDMI0_INFOFRAME_CONTROL0 + offset, - HDMI0_AVI_INFO_SEND | /* enable AVI info frames */ - HDMI0_AVI_INFO_CONT); /* send AVI info frames every frame/field */ + HDMI0_AVI_INFO_SEND | /* enable AVI info frames */ + HDMI0_AVI_INFO_CONT); /* send AVI info frames every frame/field */ - WREG32_OR(HDMI0_INFOFRAME_CONTROL1 + offset, - HDMI0_AVI_INFO_LINE(2)); /* anything other than 0 */ } /* -- cgit v0.10.2 From 12428327bbd1180b5d8ef83fdf9482b878d0502a Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 31 Mar 2015 11:38:48 -0400 Subject: drm/radeon: drop dce6_dp_enable It's mostly duplicated with evergreen_dp_enable. This is a prerequisite for fix implemented in another patch. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c index 3adc2af..68fd9fc 100644 --- a/drivers/gpu/drm/radeon/dce6_afmt.c +++ b/drivers/gpu/drm/radeon/dce6_afmt.c @@ -295,28 +295,3 @@ void dce6_dp_audio_set_dto(struct radeon_device *rdev, WREG32(DCCG_AUDIO_DTO1_MODULE, clock); } } - -void dce6_dp_enable(struct drm_encoder *encoder, bool enable) -{ - struct drm_device *dev = encoder->dev; - struct radeon_device *rdev = dev->dev_private; - struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - - if (!dig || !dig->afmt) - return; - - if (enable) { - WREG32(EVERGREEN_DP_SEC_TIMESTAMP + dig->afmt->offset, - EVERGREEN_DP_SEC_TIMESTAMP_MODE(1)); - WREG32(EVERGREEN_DP_SEC_CNTL + dig->afmt->offset, - EVERGREEN_DP_SEC_ASP_ENABLE | /* Audio packet transmission */ - EVERGREEN_DP_SEC_ATP_ENABLE | /* Audio timestamp packet transmission */ - EVERGREEN_DP_SEC_AIP_ENABLE | /* Audio infoframe packet transmission */ - EVERGREEN_DP_SEC_STREAM_ENABLE); /* Master enable for secondary stream engine */ - } else { - WREG32(EVERGREEN_DP_SEC_CNTL + dig->afmt->offset, 0); - } - - dig->afmt->enabled = enable; -} diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c index 2379979..7264ccd 100644 --- a/drivers/gpu/drm/radeon/evergreen_hdmi.c +++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c @@ -437,7 +437,7 @@ void evergreen_dp_enable(struct drm_encoder *encoder, bool enable) WREG32(EVERGREEN_DP_SEC_TIMESTAMP + dig->afmt->offset, EVERGREEN_DP_SEC_TIMESTAMP_MODE(1)); - if (radeon_connector->con_priv) { + if (!ASIC_IS_DCE6(rdev) && radeon_connector->con_priv) { dig_connector = radeon_connector->con_priv; val = RREG32(EVERGREEN_DP_SEC_AUD_N + dig->afmt->offset); val &= ~EVERGREEN_DP_SEC_N_BASE_MULTIPLE(0xf); diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c index 48d49e6..2c17df9 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.c +++ b/drivers/gpu/drm/radeon/radeon_audio.c @@ -102,7 +102,6 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, void r600_hdmi_enable(struct drm_encoder *encoder, bool enable); void evergreen_hdmi_enable(struct drm_encoder *encoder, bool enable); void evergreen_dp_enable(struct drm_encoder *encoder, bool enable); -void dce6_dp_enable(struct drm_encoder *encoder, bool enable); static const u32 pin_offsets[7] = { @@ -240,7 +239,7 @@ static struct radeon_audio_funcs dce6_dp_funcs = { .set_avi_packet = evergreen_set_avi_packet, .set_audio_packet = dce4_set_audio_packet, .mode_set = radeon_audio_dp_mode_set, - .dpms = dce6_dp_enable, + .dpms = evergreen_dp_enable, }; static void radeon_audio_interface_init(struct radeon_device *rdev) -- cgit v0.10.2 From 362ff251390f3d1f8fe94666f4fc4e5876381114 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 31 Mar 2015 11:43:12 -0400 Subject: drm/radeon/audio: don't enable packets until the end Don't enable the audio and avi infoframes and audio stream until all the state is set up. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c index 7264ccd..9e1cd0c 100644 --- a/drivers/gpu/drm/radeon/evergreen_hdmi.c +++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c @@ -222,10 +222,6 @@ void evergreen_set_avi_packet(struct radeon_device *rdev, u32 offset, WREG32_P(HDMI_INFOFRAME_CONTROL1 + offset, HDMI_AVI_INFO_LINE(2), /* anything other than 0 */ ~HDMI_AVI_INFO_LINE_MASK); - - WREG32_OR(HDMI_INFOFRAME_CONTROL0 + offset, - HDMI_AVI_INFO_SEND | /* enable AVI info frames */ - HDMI_AVI_INFO_CONT); /* required for audio info values to be updated */ } void dce4_hdmi_audio_set_dto(struct radeon_device *rdev, @@ -370,9 +366,13 @@ void dce4_set_audio_packet(struct drm_encoder *encoder, u32 offset) WREG32(AFMT_AUDIO_PACKET_CONTROL2 + offset, AFMT_AUDIO_CHANNEL_ENABLE(0xff)); + WREG32(HDMI_AUDIO_PACKET_CONTROL + offset, + HDMI_AUDIO_DELAY_EN(1) | /* set the default audio delay */ + HDMI_AUDIO_PACKETS_PER_LINE(3)); /* should be suffient for all audio modes and small enough for all hblanks */ + /* allow 60958 channel status and send audio packets fields to be updated */ - WREG32(AFMT_AUDIO_PACKET_CONTROL + offset, - AFMT_AUDIO_SAMPLE_SEND | AFMT_RESET_FIFO_WHEN_AUDIO_DIS | AFMT_60958_CS_UPDATE); + WREG32_OR(AFMT_AUDIO_PACKET_CONTROL + offset, + AFMT_RESET_FIFO_WHEN_AUDIO_DIS | AFMT_60958_CS_UPDATE); } @@ -398,17 +398,16 @@ void evergreen_hdmi_enable(struct drm_encoder *encoder, bool enable) return; if (enable) { - WREG32(HDMI_INFOFRAME_CONTROL1 + dig->afmt->offset, - HDMI_AUDIO_INFO_LINE(2)); /* anything other than 0 */ - - WREG32(HDMI_AUDIO_PACKET_CONTROL + dig->afmt->offset, - HDMI_AUDIO_DELAY_EN(1) | /* set the default audio delay */ - HDMI_AUDIO_PACKETS_PER_LINE(3)); /* should be suffient for all audio modes and small enough for all hblanks */ - WREG32(HDMI_INFOFRAME_CONTROL0 + dig->afmt->offset, + HDMI_AVI_INFO_SEND | /* enable AVI info frames */ + HDMI_AVI_INFO_CONT | /* required for audio info values to be updated */ HDMI_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */ HDMI_AUDIO_INFO_CONT); /* required for audio info values to be updated */ + WREG32_OR(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, + AFMT_AUDIO_SAMPLE_SEND); } else { + WREG32_AND(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, + ~AFMT_AUDIO_SAMPLE_SEND); WREG32(HDMI_INFOFRAME_CONTROL0 + dig->afmt->offset, 0); } @@ -434,6 +433,9 @@ void evergreen_dp_enable(struct drm_encoder *encoder, bool enable) struct radeon_connector_atom_dig *dig_connector; uint32_t val; + WREG32_OR(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, + AFMT_AUDIO_SAMPLE_SEND); + WREG32(EVERGREEN_DP_SEC_TIMESTAMP + dig->afmt->offset, EVERGREEN_DP_SEC_TIMESTAMP_MODE(1)); @@ -457,6 +459,8 @@ void evergreen_dp_enable(struct drm_encoder *encoder, bool enable) EVERGREEN_DP_SEC_STREAM_ENABLE); /* Master enable for secondary stream engine */ } else { WREG32(EVERGREEN_DP_SEC_CNTL + dig->afmt->offset, 0); + WREG32_AND(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, + ~AFMT_AUDIO_SAMPLE_SEND); } dig->afmt->enabled = enable; -- cgit v0.10.2 From 0f55db36d49d45b80eff0c0a2a498766016f458b Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 7 Apr 2015 09:52:42 -0400 Subject: drm/radeon: only mark audio as connected if the monitor supports it (v3) Otherwise the driver may try and send audio which may confuse the monitor. v2: set pin to NULL if no audio v3: avoid crash with analog encoders Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c index 2c17df9..8b82abb 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.c +++ b/drivers/gpu/drm/radeon/radeon_audio.c @@ -460,30 +460,33 @@ void radeon_audio_detect(struct drm_connector *connector, if (!connector || !connector->encoder) return; + if (!radeon_encoder_is_digital(connector->encoder)) + return; + rdev = connector->encoder->dev->dev_private; radeon_encoder = to_radeon_encoder(connector->encoder); dig = radeon_encoder->enc_priv; - if (status == connector_status_connected) { - struct radeon_connector *radeon_connector; - int sink_type; - - if (!drm_detect_monitor_audio(radeon_connector_edid(connector))) { - radeon_encoder->audio = NULL; - return; - } + if (!dig->afmt) + return; - radeon_connector = to_radeon_connector(connector); - sink_type = radeon_dp_getsinktype(radeon_connector); + if (status == connector_status_connected) { + struct radeon_connector *radeon_connector = to_radeon_connector(connector); if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort && - sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT) + radeon_dp_getsinktype(radeon_connector) == + CONNECTOR_OBJECT_ID_DISPLAYPORT) radeon_encoder->audio = rdev->audio.dp_funcs; else radeon_encoder->audio = rdev->audio.hdmi_funcs; dig->afmt->pin = radeon_audio_get_pin(connector->encoder); - radeon_audio_enable(rdev, dig->afmt->pin, 0xf); + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + radeon_audio_enable(rdev, dig->afmt->pin, 0xf); + } else { + radeon_audio_enable(rdev, dig->afmt->pin, 0); + dig->afmt->pin = NULL; + } } else { radeon_audio_enable(rdev, dig->afmt->pin, 0); dig->afmt->pin = NULL; diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index cebb65e..d17d251 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1379,8 +1379,10 @@ out: /* updated in get modes as well since we need to know if it's analog or digital */ radeon_connector_update_scratch_regs(connector, ret); - if (radeon_audio != 0) + if (radeon_audio != 0) { + radeon_connector_get_edid(connector); radeon_audio_detect(connector, ret); + } exit: pm_runtime_mark_last_busy(connector->dev->dev); @@ -1717,8 +1719,10 @@ radeon_dp_detect(struct drm_connector *connector, bool force) radeon_connector_update_scratch_regs(connector, ret); - if (radeon_audio != 0) + if (radeon_audio != 0) { + radeon_connector_get_edid(connector); radeon_audio_detect(connector, ret); + } out: pm_runtime_mark_last_busy(connector->dev->dev); -- cgit v0.10.2 From 38aef1549b18539eaecd804383a6ccb6588a9ce1 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 7 Apr 2015 10:20:49 -0400 Subject: drm/radeon: only enable audio streams if the monitor supports it Selectively enable which packets we send based on monitor caps. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index f57c1ab..dd39f43 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -1761,17 +1761,15 @@ radeon_atom_encoder_dpms(struct drm_encoder *encoder, int mode) struct drm_device *dev = encoder->dev; struct radeon_device *rdev = dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); int encoder_mode = atombios_get_encoder_mode(encoder); DRM_DEBUG_KMS("encoder dpms %d to mode %d, devices %08x, active_devices %08x\n", radeon_encoder->encoder_id, mode, radeon_encoder->devices, radeon_encoder->active_device); - if (connector && (radeon_audio != 0) && + if ((radeon_audio != 0) && ((encoder_mode == ATOM_ENCODER_MODE_HDMI) || - (ENCODER_MODE_IS_DP(encoder_mode) && - drm_detect_monitor_audio(radeon_connector_edid(connector))))) + ENCODER_MODE_IS_DP(encoder_mode))) radeon_audio_dpms(encoder, mode); switch (radeon_encoder->encoder_id) { diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c index 9e1cd0c..0926739 100644 --- a/drivers/gpu/drm/radeon/evergreen_hdmi.c +++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c @@ -398,13 +398,23 @@ void evergreen_hdmi_enable(struct drm_encoder *encoder, bool enable) return; if (enable) { - WREG32(HDMI_INFOFRAME_CONTROL0 + dig->afmt->offset, - HDMI_AVI_INFO_SEND | /* enable AVI info frames */ - HDMI_AVI_INFO_CONT | /* required for audio info values to be updated */ - HDMI_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */ - HDMI_AUDIO_INFO_CONT); /* required for audio info values to be updated */ - WREG32_OR(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, - AFMT_AUDIO_SAMPLE_SEND); + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); + + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + WREG32(HDMI_INFOFRAME_CONTROL0 + dig->afmt->offset, + HDMI_AVI_INFO_SEND | /* enable AVI info frames */ + HDMI_AVI_INFO_CONT | /* required for audio info values to be updated */ + HDMI_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */ + HDMI_AUDIO_INFO_CONT); /* required for audio info values to be updated */ + WREG32_OR(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, + AFMT_AUDIO_SAMPLE_SEND); + } else { + WREG32(HDMI_INFOFRAME_CONTROL0 + dig->afmt->offset, + HDMI_AVI_INFO_SEND | /* enable AVI info frames */ + HDMI_AVI_INFO_CONT); /* required for audio info values to be updated */ + WREG32_AND(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, + ~AFMT_AUDIO_SAMPLE_SEND); + } } else { WREG32_AND(AFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, ~AFMT_AUDIO_SAMPLE_SEND); @@ -423,11 +433,12 @@ void evergreen_dp_enable(struct drm_encoder *encoder, bool enable) struct radeon_device *rdev = dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); if (!dig || !dig->afmt) return; - if (enable) { + if (enable && drm_detect_monitor_audio(radeon_connector_edid(connector))) { struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); struct radeon_connector *radeon_connector = to_radeon_connector(connector); struct radeon_connector_atom_dig *dig_connector; -- cgit v0.10.2 From 7fe04d6fa824ccea704535a597dc417c8687f990 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sun, 19 Apr 2015 12:01:00 -0400 Subject: drm/radeon: adjust pll when audio is not enabled Fixes display problems with some monitors when audio is not enabled. Bugs: https://bugs.freedesktop.org/show_bug.cgi?id=89505 https://bugzilla.kernel.org/show_bug.cgi?id=94171 Plus several reports on IRC. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index dac78ad..42b2ea3 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -580,6 +580,9 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, else radeon_crtc->pll_flags |= RADEON_PLL_PREFER_LOW_REF_DIV; + /* if there is no audio, set MINM_OVER_MAXP */ + if (!drm_detect_monitor_audio(radeon_connector_edid(connector))) + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP; if (rdev->family < CHIP_RV770) radeon_crtc->pll_flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP; /* use frac fb div on APUs */ -- cgit v0.10.2 From cd17e02ff4db58ec32d35cf331c705d295779930 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 27 Apr 2015 09:51:43 -0400 Subject: drm/radeon: add SI DPM quirk for Sapphire R9 270 Dual-X 2G GDDR5 Seems to have problems with high mclks. bug: https://bugs.freedesktop.org/show_bug.cgi?id=76490 Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index b35bccf..ff8b83f 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2924,6 +2924,7 @@ struct si_dpm_quirk { static struct si_dpm_quirk si_dpm_quirk_list[] = { /* PITCAIRN - https://bugs.freedesktop.org/show_bug.cgi?id=76490 */ { PCI_VENDOR_ID_ATI, 0x6810, 0x1462, 0x3036, 0, 120000 }, + { PCI_VENDOR_ID_ATI, 0x6811, 0x174b, 0xe271, 0, 120000 }, { 0, 0, 0, 0 }, }; -- cgit v0.10.2 From 7f8d49dcc66a3dd3a8fc3078330b8fb9e616ad3f Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Tue, 21 Apr 2015 15:59:53 +0200 Subject: ARM: dts: imx6: phyFLEX: USB VBUS control is active-high The fixed-regulator bindings require a separate property enable-active-high, the standard gpio phandle property polarity setting is ignored. Signed-off-by: Philipp Zabel Fixes: 4fe69a934b1f ("ARM: dts: Add Phytec pfla02 with i.MX6 DualLite/Solo") Cc: stable@vger.kernel.org Signed-off-by: Shawn Guo diff --git a/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi b/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi index 19cc269..1ce6133 100644 --- a/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi +++ b/arch/arm/boot/dts/imx6qdl-phytec-pfla02.dtsi @@ -31,6 +31,7 @@ regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>; gpio = <&gpio4 15 0>; + enable-active-high; }; reg_usb_h1_vbus: regulator@1 { @@ -40,6 +41,7 @@ regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>; gpio = <&gpio1 0 0>; + enable-active-high; }; }; -- cgit v0.10.2 From f90d3f0d0a11fa77918fd5497cb616dd2faa8431 Mon Sep 17 00:00:00 2001 From: Markus Pargmann Date: Fri, 24 Apr 2015 09:27:33 +0200 Subject: ARM: dts: imx25: Add #pwm-cells to pwm4 The property '#pwm-cells' is currently missing. It is not possible to use pwm4 without this property. Signed-off-by: Markus Pargmann Fixes: 5658a68fb578 ("ARM i.MX25: Add devicetree") Cc: Reviewed-by: Fabio Estevam Signed-off-by: Shawn Guo diff --git a/arch/arm/boot/dts/imx25.dtsi b/arch/arm/boot/dts/imx25.dtsi index e4d3aec..677f81d 100644 --- a/arch/arm/boot/dts/imx25.dtsi +++ b/arch/arm/boot/dts/imx25.dtsi @@ -428,6 +428,7 @@ pwm4: pwm@53fc8000 { compatible = "fsl,imx25-pwm", "fsl,imx27-pwm"; + #pwm-cells = <2>; reg = <0x53fc8000 0x4000>; clocks = <&clks 108>, <&clks 52>; clock-names = "ipg", "per"; -- cgit v0.10.2 From 4ada77e37a773168fea484899201e272ab44ba8b Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Fri, 24 Apr 2015 13:29:47 +0200 Subject: ARM: dts: imx28: Fix AUART4 TX-DMA interrupt name Fix a typo in the TX DMA interrupt name for AUART4. This patch makes AUART4 operational again. Signed-off-by: Marek Vasut Fixes: f30fb03d4d3a ("ARM: dts: add generic DMA device tree binding for mxs-dma") Cc: stable@vger.kernel.org Acked-by: Stefan Wahren Signed-off-by: Shawn Guo diff --git a/Documentation/devicetree/bindings/dma/fsl-mxs-dma.txt b/Documentation/devicetree/bindings/dma/fsl-mxs-dma.txt index a4873e5..e30e184 100644 --- a/Documentation/devicetree/bindings/dma/fsl-mxs-dma.txt +++ b/Documentation/devicetree/bindings/dma/fsl-mxs-dma.txt @@ -38,7 +38,7 @@ dma_apbx: dma-apbx@80024000 { 80 81 68 69 70 71 72 73 74 75 76 77>; - interrupt-names = "auart4-rx", "aurat4-tx", "spdif-tx", "empty", + interrupt-names = "auart4-rx", "auart4-tx", "spdif-tx", "empty", "saif0", "saif1", "i2c0", "i2c1", "auart0-rx", "auart0-tx", "auart1-rx", "auart1-tx", "auart2-rx", "auart2-tx", "auart3-rx", "auart3-tx"; diff --git a/arch/arm/boot/dts/imx28.dtsi b/arch/arm/boot/dts/imx28.dtsi index 25e25f8..4e073e8 100644 --- a/arch/arm/boot/dts/imx28.dtsi +++ b/arch/arm/boot/dts/imx28.dtsi @@ -913,7 +913,7 @@ 80 81 68 69 70 71 72 73 74 75 76 77>; - interrupt-names = "auart4-rx", "aurat4-tx", "spdif-tx", "empty", + interrupt-names = "auart4-rx", "auart4-tx", "spdif-tx", "empty", "saif0", "saif1", "i2c0", "i2c1", "auart0-rx", "auart0-tx", "auart1-rx", "auart1-tx", "auart2-rx", "auart2-tx", "auart3-rx", "auart3-tx"; -- cgit v0.10.2 From 393a33970540ac6a2c894b0d6ef3f5d485860884 Mon Sep 17 00:00:00 2001 From: Wang YanQing Date: Sun, 26 Apr 2015 16:43:31 +0800 Subject: block:bounce: fix call inc_|dec_zone_page_state on different pages confuse value of NR_BOUNCE Commit d2c5e30c9a1420902262aa923794d2ae4e0bc391 ("[PATCH] zoned vm counters: conversion of nr_bounce to per zone counter") convert statistic of nr_bounce to per zone and one global value in vm_stat, but it call inc_|dec_zone_page_state on different pages, then different zones, and cause us to get unexpected value of NR_BOUNCE. Below is the result on my machine: Mar 2 09:26:08 udknight kernel: [144766.778265] Mem-Info: Mar 2 09:26:08 udknight kernel: [144766.778266] DMA per-cpu: Mar 2 09:26:08 udknight kernel: [144766.778268] CPU 0: hi: 0, btch: 1 usd: 0 Mar 2 09:26:08 udknight kernel: [144766.778269] CPU 1: hi: 0, btch: 1 usd: 0 Mar 2 09:26:08 udknight kernel: [144766.778270] Normal per-cpu: Mar 2 09:26:08 udknight kernel: [144766.778271] CPU 0: hi: 186, btch: 31 usd: 0 Mar 2 09:26:08 udknight kernel: [144766.778273] CPU 1: hi: 186, btch: 31 usd: 0 Mar 2 09:26:08 udknight kernel: [144766.778274] HighMem per-cpu: Mar 2 09:26:08 udknight kernel: [144766.778275] CPU 0: hi: 186, btch: 31 usd: 0 Mar 2 09:26:08 udknight kernel: [144766.778276] CPU 1: hi: 186, btch: 31 usd: 0 Mar 2 09:26:08 udknight kernel: [144766.778279] active_anon:46926 inactive_anon:287406 isolated_anon:0 Mar 2 09:26:08 udknight kernel: [144766.778279] active_file:105085 inactive_file:139432 isolated_file:0 Mar 2 09:26:08 udknight kernel: [144766.778279] unevictable:653 dirty:0 writeback:0 unstable:0 Mar 2 09:26:08 udknight kernel: [144766.778279] free:178957 slab_reclaimable:6419 slab_unreclaimable:9966 Mar 2 09:26:08 udknight kernel: [144766.778279] mapped:4426 shmem:305277 pagetables:784 bounce:0 Mar 2 09:26:08 udknight kernel: [144766.778279] free_cma:0 Mar 2 09:26:08 udknight kernel: [144766.778286] DMA free:3324kB min:68kB low:84kB high:100kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15976kB managed:15900kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? yes Mar 2 09:26:08 udknight kernel: [144766.778287] lowmem_reserve[]: 0 822 3754 3754 Mar 2 09:26:08 udknight kernel: [144766.778293] Normal free:26828kB min:3632kB low:4540kB high:5448kB active_anon:4872kB inactive_anon:68kB active_file:1796kB inactive_file:1796kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:892920kB managed:842560kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:4144kB slab_reclaimable:25676kB slab_unreclaimable:39864kB kernel_stack:1944kB pagetables:3136kB unstable:0kB bounce:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:2412612 all_unreclaimable? yes Mar 2 09:26:08 udknight kernel: [144766.778294] lowmem_reserve[]: 0 0 23451 23451 Mar 2 09:26:08 udknight kernel: [144766.778299] HighMem free:685676kB min:512kB low:3748kB high:6984kB active_anon:182832kB inactive_anon:1149556kB active_file:418544kB inactive_file:555932kB unevictable:2612kB isolated(anon):0kB isolated(file):0kB present:3001732kB managed:3001732kB mlocked:0kB dirty:0kB writeback:0kB mapped:17704kB shmem:1216964kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:75771152kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no Mar 2 09:26:08 udknight kernel: [144766.778300] lowmem_reserve[]: 0 0 0 0 You can see bounce:75771152kB for HighMem, but bounce:0 for lowmem and global. This patch fix it. Signed-off-by: Wang YanQing Signed-off-by: Jens Axboe diff --git a/block/bounce.c b/block/bounce.c index ab21ba2..ed9dd80 100644 --- a/block/bounce.c +++ b/block/bounce.c @@ -221,8 +221,8 @@ bounce: if (page_to_pfn(page) <= queue_bounce_pfn(q) && !force) continue; - inc_zone_page_state(to->bv_page, NR_BOUNCE); to->bv_page = mempool_alloc(pool, q->bounce_gfp); + inc_zone_page_state(to->bv_page, NR_BOUNCE); if (rw == WRITE) { char *vto, *vfrom; -- cgit v0.10.2 From 26d4d129b6042197b4cbc8341c0618f99231af2f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 27 Apr 2015 17:04:34 +0200 Subject: drm/radeon: fix lockup when BOs aren't part of the VM on release MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If we unmap BOs before releasing them them the intervall tree locks up because we try to remove an entry not inside the tree. Based on a patch from Michel Dänzer. Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c index 2a5a4a9..16d8e95 100644 --- a/drivers/gpu/drm/radeon/radeon_vm.c +++ b/drivers/gpu/drm/radeon/radeon_vm.c @@ -1107,7 +1107,8 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev, list_del(&bo_va->bo_list); mutex_lock(&vm->mutex); - interval_tree_remove(&bo_va->it, &vm->va); + if (bo_va->it.start || bo_va->it.last) + interval_tree_remove(&bo_va->it, &vm->va); spin_lock(&vm->status_lock); list_del(&bo_va->vm_status); -- cgit v0.10.2 From 48afbd70ac7b6aa62e8d452091023941d8085f8a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 27 Apr 2015 17:04:35 +0200 Subject: drm/radeon: reset BOs address after clearing it. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Otherwise it is possible that we will have page table corruption if we change a BOs address multiple times. Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c index 16d8e95..cabcb0a 100644 --- a/drivers/gpu/drm/radeon/radeon_vm.c +++ b/drivers/gpu/drm/radeon/radeon_vm.c @@ -490,6 +490,8 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev, spin_lock(&vm->status_lock); list_add(&tmp->vm_status, &vm->freed); spin_unlock(&vm->status_lock); + + bo_va->addr = 0; } interval_tree_remove(&bo_va->it, &vm->va); -- cgit v0.10.2 From c29c0876ec05d51a93508a39b90b92c29ba6423d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 27 Apr 2015 17:04:36 +0200 Subject: drm/radeon: check new address before removing old one MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Otherwise the change isn't atomic. Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c index cabcb0a..de42fc4 100644 --- a/drivers/gpu/drm/radeon/radeon_vm.c +++ b/drivers/gpu/drm/radeon/radeon_vm.c @@ -473,6 +473,23 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev, } mutex_lock(&vm->mutex); + soffset /= RADEON_GPU_PAGE_SIZE; + eoffset /= RADEON_GPU_PAGE_SIZE; + if (soffset || eoffset) { + struct interval_tree_node *it; + it = interval_tree_iter_first(&vm->va, soffset, eoffset - 1); + if (it && it != &bo_va->it) { + struct radeon_bo_va *tmp; + tmp = container_of(it, struct radeon_bo_va, it); + /* bo and tmp overlap, invalid offset */ + dev_err(rdev->dev, "bo %p va 0x%010Lx conflict with " + "(bo %p 0x%010lx 0x%010lx)\n", bo_va->bo, + soffset, tmp->bo, tmp->it.start, tmp->it.last); + mutex_unlock(&vm->mutex); + return -EINVAL; + } + } + if (bo_va->it.start || bo_va->it.last) { if (bo_va->addr) { /* add a clone of the bo_va to clear the old address */ @@ -499,21 +516,7 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev, bo_va->it.last = 0; } - soffset /= RADEON_GPU_PAGE_SIZE; - eoffset /= RADEON_GPU_PAGE_SIZE; if (soffset || eoffset) { - struct interval_tree_node *it; - it = interval_tree_iter_first(&vm->va, soffset, eoffset - 1); - if (it) { - struct radeon_bo_va *tmp; - tmp = container_of(it, struct radeon_bo_va, it); - /* bo and tmp overlap, invalid offset */ - dev_err(rdev->dev, "bo %p va 0x%010Lx conflict with " - "(bo %p 0x%010lx 0x%010lx)\n", bo_va->bo, - soffset, tmp->bo, tmp->it.start, tmp->it.last); - mutex_unlock(&vm->mutex); - return -EINVAL; - } bo_va->it.start = soffset; bo_va->it.last = eoffset - 1; interval_tree_insert(&bo_va->it, &vm->va); -- cgit v0.10.2 From 9fb2bcf928ed5c664affaeb7edf9cc19f590b9f3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 27 Apr 2015 17:04:37 +0200 Subject: drm/radeon: fix userptr return value checking (v2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Otherwise we print false warning from time to time. v2: agd5f: rebase Signed-off-by: Christian König Signed-off-by: Jack Xiao CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_mn.c b/drivers/gpu/drm/radeon/radeon_mn.c index 0170137..535bf40 100644 --- a/drivers/gpu/drm/radeon/radeon_mn.c +++ b/drivers/gpu/drm/radeon/radeon_mn.c @@ -135,7 +135,7 @@ static void radeon_mn_invalidate_range_start(struct mmu_notifier *mn, while (it) { struct radeon_mn_node *node; struct radeon_bo *bo; - int r; + long r; node = container_of(it, struct radeon_mn_node, it); it = interval_tree_iter_next(it, start, end); @@ -144,19 +144,19 @@ static void radeon_mn_invalidate_range_start(struct mmu_notifier *mn, r = radeon_bo_reserve(bo, true); if (r) { - DRM_ERROR("(%d) failed to reserve user bo\n", r); + DRM_ERROR("(%ld) failed to reserve user bo\n", r); continue; } r = reservation_object_wait_timeout_rcu(bo->tbo.resv, true, false, MAX_SCHEDULE_TIMEOUT); - if (r) - DRM_ERROR("(%d) failed to wait for user bo\n", r); + if (r <= 0) + DRM_ERROR("(%ld) failed to wait for user bo\n", r); radeon_ttm_placement_from_domain(bo, RADEON_GEM_DOMAIN_CPU); r = ttm_bo_validate(&bo->tbo, &bo->placement, false, false); if (r) - DRM_ERROR("(%d) failed to validate user bo\n", r); + DRM_ERROR("(%ld) failed to validate user bo\n", r); radeon_bo_unreserve(bo); } -- cgit v0.10.2 From c671835021798c1c40ca0b55b49feff76ed5e0e1 Mon Sep 17 00:00:00 2001 From: Wang Nan Date: Sat, 25 Apr 2015 07:25:03 +0000 Subject: perf top: Fix a segfault when kernel map is restricted. Perf top raise a warning if a kernel sample is collected but kernel map is restricted. The warning message needs to dereference al.map->dso... However, previous perf_event__preprocess_sample() doesn't always guarantee al.map != NULL, for example, when kernel map is restricted. This patch validates al.map before dereferencing, avoid the segfault. Before this patch: $ cat /proc/sys/kernel/kptr_restrict 1 $ perf top -p 120183 perf: Segmentation fault -------- backtrace -------- /path/to/perf[0x509868] /lib64/libc.so.6(+0x3545f)[0x7f9a1540045f] /path/to/perf[0x448820] /path/to/perf(cmd_top+0xe3c)[0x44a5dc] /path/to/perf[0x4766a2] /path/to/perf(main+0x5f5)[0x42e545] /lib64/libc.so.6(__libc_start_main+0xf4)[0x7f9a153ecbd4] /path/to/perf[0x42e674] And gdb call trace: Program received signal SIGSEGV, Segmentation fault. perf_event__process_sample (machine=0xa44030, sample=0x7fffffffa4c0, evsel=0xa43b00, event=0x7ffff41c3000, tool=0x7fffffffa8a0) at builtin-top.c:736 736 !RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION]) ? (gdb) bt #0 perf_event__process_sample (machine=0xa44030, sample=0x7fffffffa4c0, evsel=0xa43b00, event=0x7ffff41c3000, tool=0x7fffffffa8a0) at builtin-top.c:736 #1 perf_top__mmap_read_idx (top=top@entry=0x7fffffffa8a0, idx=idx@entry=0) at builtin-top.c:855 #2 0x000000000044a5dd in perf_top__mmap_read (top=0x7fffffffa8a0) at builtin-top.c:872 #3 __cmd_top (top=0x7fffffffa8a0) at builtin-top.c:997 #4 cmd_top (argc=, argv=, prefix=) at builtin-top.c:1267 #5 0x00000000004766a3 in run_builtin (p=p@entry=0x8a6ce8 , argc=argc@entry=3, argv=argv@entry=0x7fffffffdf70) at perf.c:371 #6 0x000000000042e546 in handle_internal_command (argv=0x7fffffffdf70, argc=3) at perf.c:430 #7 run_argv (argv=0x7fffffffdcf0, argcp=0x7fffffffdcfc) at perf.c:474 #8 main (argc=3, argv=0x7fffffffdf70) at perf.c:589 (gdb) Signed-off-by: Wang Nan Tested-by: Arnaldo Carvalho de Melo Cc: David Ahern Cc: Jiri Olsa Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Zefan Li Link: http://lkml.kernel.org/r/1429946703-80807-1-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 1cb3436..6a4d5d4 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -733,7 +733,7 @@ static void perf_event__process_sample(struct perf_tool *tool, "Kernel address maps (/proc/{kallsyms,modules}) are restricted.\n\n" "Check /proc/sys/kernel/kptr_restrict.\n\n" "Kernel%s samples will not be resolved.\n", - !RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION]) ? + al.map && !RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION]) ? " modules" : ""); if (use_browser <= 0) sleep(5); -- cgit v0.10.2 From 6cd18e711dd8075da9d78cfc1239f912ff28968a Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 27 Apr 2015 14:12:22 +1000 Subject: block: destroy bdi before blockdev is unregistered. Because of the peculiar way that md devices are created (automatically when the device node is opened), a new device can be created and registered immediately after the blk_unregister_region(disk_devt(disk), disk->minors); call in del_gendisk(). Therefore it is important that all visible artifacts of the previous device are removed before this call. In particular, the 'bdi'. Since: commit c4db59d31e39ea067c32163ac961e9c80198fd37 Author: Christoph Hellwig fs: don't reassign dirty inodes to default_backing_dev_info moved the device_unregister(bdi->dev); call from bdi_unregister() to bdi_destroy() it has been quite easy to lose a race and have a new (e.g.) "md127" be created after the blk_unregister_region() call and before bdi_destroy() is ultimately called by the final 'put_disk', which must come after del_gendisk(). The new device finds that the bdi name is already registered in sysfs and complains > [ 9627.630029] WARNING: CPU: 18 PID: 3330 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x5a/0x70() > [ 9627.630032] sysfs: cannot create duplicate filename '/devices/virtual/bdi/9:127' We can fix this by moving the bdi_destroy() call out of blk_release_queue() (which can happen very late when a refcount reaches zero) and into blk_cleanup_queue() - which happens exactly when the md device driver calls it. Then it is only necessary for md to call blk_cleanup_queue() before del_gendisk(). As loop.c devices are also created on demand by opening the device node, we make the same change there. Fixes: c4db59d31e39ea067c32163ac961e9c80198fd37 Reported-by: Azat Khuzhin Cc: Christoph Hellwig Cc: stable@vger.kernel.org (v4.0) Signed-off-by: NeilBrown Reviewed-by: Christoph Hellwig Signed-off-by: Jens Axboe diff --git a/block/blk-core.c b/block/blk-core.c index fd154b9..7871603 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -552,6 +552,8 @@ void blk_cleanup_queue(struct request_queue *q) q->queue_lock = &q->__queue_lock; spin_unlock_irq(lock); + bdi_destroy(&q->backing_dev_info); + /* @q is and will stay empty, shutdown and put */ blk_put_queue(q); } diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index faaf36a..2b8fd30 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -522,8 +522,6 @@ static void blk_release_queue(struct kobject *kobj) blk_trace_shutdown(q); - bdi_destroy(&q->backing_dev_info); - ida_simple_remove(&blk_queue_ida, q->id); call_rcu(&q->rcu_head, blk_free_queue_rcu); } diff --git a/drivers/block/loop.c b/drivers/block/loop.c index ae3fcb4..d7173cb 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1620,8 +1620,8 @@ out: static void loop_remove(struct loop_device *lo) { - del_gendisk(lo->lo_disk); blk_cleanup_queue(lo->lo_queue); + del_gendisk(lo->lo_disk); blk_mq_free_tag_set(&lo->tag_set); put_disk(lo->lo_disk); kfree(lo); diff --git a/drivers/md/md.c b/drivers/md/md.c index e617878..e47d1dd 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -4754,12 +4754,12 @@ static void md_free(struct kobject *ko) if (mddev->sysfs_state) sysfs_put(mddev->sysfs_state); + if (mddev->queue) + blk_cleanup_queue(mddev->queue); if (mddev->gendisk) { del_gendisk(mddev->gendisk); put_disk(mddev->gendisk); } - if (mddev->queue) - blk_cleanup_queue(mddev->queue); kfree(mddev); } -- cgit v0.10.2 From 35e9a9f93994d7f7d12afa41169c7ba05513721b Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 20 Apr 2015 22:42:24 -0500 Subject: SCSI: add 1024 max sectors black list flag This works around a issue with qnap iscsi targets not handling large IOs very well. The target returns: VPD INQUIRY: Block limits page (SBC) Maximum compare and write length: 1 blocks Optimal transfer length granularity: 1 blocks Maximum transfer length: 4294967295 blocks Optimal transfer length: 4294967295 blocks Maximum prefetch, xdread, xdwrite transfer length: 0 blocks Maximum unmap LBA count: 8388607 Maximum unmap block descriptor count: 1 Optimal unmap granularity: 16383 Unmap granularity alignment valid: 0 Unmap granularity alignment: 0 Maximum write same length: 0xffffffff blocks Maximum atomic transfer length: 0 Atomic alignment: 0 Atomic transfer length granularity: 0 and it is *sometimes* able to handle at least one IO of size up to 8 MB. We have seen in traces where it will sometimes work, but other times it looks like it fails and it looks like it returns failures if we send multiple large IOs sometimes. Also it looks like it can return 2 different errors. It will sometimes send iscsi reject errors indicating out of resources or it will send invalid cdb illegal requests check conditions. And then when it sends iscsi rejects it does not seem to handle retries when there are command sequence holes, so I could not just add code to try and gracefully handle that error code. The problem is that we do not have a good contact for the company, so we are not able to determine under what conditions it returns which error and why it sometimes works. So, this patch just adds a new black list flag to set targets like this to the old max safe sectors of 1024. The max_hw_sectors changes added in 3.19 caused this regression, so I also ccing stable. Reported-by: Christian Hesse Signed-off-by: Mike Christie Cc: stable@vger.kernel.org Reviewed-by: Christoph Hellwig Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c index 262ab83..9f77d23 100644 --- a/drivers/scsi/scsi_devinfo.c +++ b/drivers/scsi/scsi_devinfo.c @@ -226,6 +226,7 @@ static struct { {"PIONEER", "CD-ROM DRM-624X", NULL, BLIST_FORCELUN | BLIST_SINGLELUN}, {"Promise", "VTrak E610f", NULL, BLIST_SPARSELUN | BLIST_NO_RSOC}, {"Promise", "", NULL, BLIST_SPARSELUN}, + {"QNAP", "iSCSI Storage", NULL, BLIST_MAX_1024}, {"QUANTUM", "XP34301", "1071", BLIST_NOTQ}, {"REGAL", "CDC-4X", NULL, BLIST_MAX5LUN | BLIST_SINGLELUN}, {"SanDisk", "ImageMate CF-SD1", NULL, BLIST_FORCELUN}, diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 60aae01..6efab1c 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -897,6 +897,12 @@ static int scsi_add_lun(struct scsi_device *sdev, unsigned char *inq_result, */ if (*bflags & BLIST_MAX_512) blk_queue_max_hw_sectors(sdev->request_queue, 512); + /* + * Max 1024 sector transfer length for targets that report incorrect + * max/optimal lengths and relied on the old block layer safe default + */ + else if (*bflags & BLIST_MAX_1024) + blk_queue_max_hw_sectors(sdev->request_queue, 1024); /* * Some devices may not want to have a start command automatically diff --git a/include/scsi/scsi_devinfo.h b/include/scsi/scsi_devinfo.h index 183eaab..96e3f56 100644 --- a/include/scsi/scsi_devinfo.h +++ b/include/scsi/scsi_devinfo.h @@ -36,5 +36,6 @@ for sequential scan */ #define BLIST_TRY_VPD_PAGES 0x10000000 /* Attempt to read VPD pages */ #define BLIST_NO_RSOC 0x20000000 /* don't try to issue RSOC */ +#define BLIST_MAX_1024 0x40000000 /* maximum 1024 sector cdb length */ #endif -- cgit v0.10.2 From d13855ef18e1852b2c4dc86ddf5759c5b34628cb Mon Sep 17 00:00:00 2001 From: He Kuang Date: Sat, 25 Apr 2015 16:08:58 +0800 Subject: perf probe: Fix bug with global variables handling There are missing curly braces which causes find_variable() return wrong value when probing with global variables. This problem can be reproduced as following: $ perf probe -v --add='generic_perform_write global_variable_for_test' ... Try to find probe point from debuginfo. Probe point found: generic_perform_write+0 Searching 'global_variable_for_test' variable in context. An error occurred in debuginfo analysis (-2). Error: Failed to add events. Reason: No such file or directory (Code: -2) After this patch: $ perf probe -v --add='generic_perform_write global_variable_for_test' ... Converting variable global_variable_for_test into trace event. global_variable_for_test type is int. Found 1 probe_trace_events. Opening /sys/kernel/debug/tracing/kprobe_events write=1 Added new event: Writing event: p:probe/generic_perform_write _stext+1237464 global_variable_for_test=@global_variable_for_test+0:s32 probe:generic_perform_write (on generic_perform_write with global_variable_for_test) You can now use it in all perf tools, such as: perf record -e probe:generic_perform_write -aR sleep 1 Signed-off-by: He Kuang Acked-by: Masami Hiramatsu Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Wang Nan Link: http://lkml.kernel.org/r/1429949338-18678-1-git-send-email-hekuang@huawei.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 44554c3..1c3cc07 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -578,10 +578,12 @@ static int find_variable(Dwarf_Die *sc_die, struct probe_finder *pf) /* Search child die for local variables and parameters. */ if (!die_find_variable_at(sc_die, pf->pvar->var, pf->addr, &vr_die)) { /* Search again in global variables */ - if (!die_find_variable_at(&pf->cu_die, pf->pvar->var, 0, &vr_die)) + if (!die_find_variable_at(&pf->cu_die, pf->pvar->var, + 0, &vr_die)) { pr_warning("Failed to find '%s' in this function.\n", pf->pvar->var); ret = -ENOENT; + } } if (ret >= 0) ret = convert_variable(&vr_die, pf); -- cgit v0.10.2 From 052b0f6eaf8b1f02669884a177bc3ce463133a42 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Fri, 24 Apr 2015 10:00:48 -0700 Subject: perf bench futex: Fix hung wakeup tasks after requeueing The futex-requeue benchmark can hang because of missing wakeups once the benchmark is done, ie: [Run 1]: Requeued 1024 of 1024 threads in 0.3290 ms perf: couldn't wakeup all tasks (135/1024) This bug, while perhaps suggesting missing wakeups in kernel futex code, is merely a consequence of the crappy FUTEX_CMP_REQUEUE man page, incorrectly mentioning that the number of requeued tasks is in fact returned, not the wakeups. This patch acknowledges this and updates the corresponding futex_wake code around it. Signed-off-by: Davidlohr Bueso Cc: Mel Gorman Link: http://lkml.kernel.org/r/1429894848.10273.44.camel@stgolabs.net Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/bench/futex-requeue.c b/tools/perf/bench/futex-requeue.c index bedff6b..ad0d9b5 100644 --- a/tools/perf/bench/futex-requeue.c +++ b/tools/perf/bench/futex-requeue.c @@ -132,6 +132,9 @@ int bench_futex_requeue(int argc, const char **argv, if (!fshared) futex_flag = FUTEX_PRIVATE_FLAG; + if (nrequeue > nthreads) + nrequeue = nthreads; + printf("Run summary [PID %d]: Requeuing %d threads (from [%s] %p to %p), " "%d at a time.\n\n", getpid(), nthreads, fshared ? "shared":"private", &futex1, &futex2, nrequeue); @@ -161,20 +164,18 @@ int bench_futex_requeue(int argc, const char **argv, /* Ok, all threads are patiently blocked, start requeueing */ gettimeofday(&start, NULL); - for (nrequeued = 0; nrequeued < nthreads; nrequeued += nrequeue) { + while (nrequeued < nthreads) { /* * Do not wakeup any tasks blocked on futex1, allowing * us to really measure futex_wait functionality. */ - futex_cmp_requeue(&futex1, 0, &futex2, 0, - nrequeue, futex_flag); + nrequeued += futex_cmp_requeue(&futex1, 0, &futex2, 0, + nrequeue, futex_flag); } + gettimeofday(&end, NULL); timersub(&end, &start, &runtime); - if (nrequeued > nthreads) - nrequeued = nthreads; - update_stats(&requeued_stats, nrequeued); update_stats(&requeuetime_stats, runtime.tv_usec); @@ -184,7 +185,7 @@ int bench_futex_requeue(int argc, const char **argv, } /* everybody should be blocked on futex2, wake'em up */ - nrequeued = futex_wake(&futex2, nthreads, futex_flag); + nrequeued = futex_wake(&futex2, nrequeued, futex_flag); if (nthreads != nrequeued) warnx("couldn't wakeup all tasks (%d/%d)", nrequeued, nthreads); -- cgit v0.10.2 From 24f1ced167e5e011040b4c3aae75aee45a79eed5 Mon Sep 17 00:00:00 2001 From: Petr Holasek Date: Thu, 16 Apr 2015 17:38:17 +0200 Subject: perf bench numa: Fixes of --quiet argument Corrected description and fixed function of --quiet argument. Signed-off-by: Petr Holasek Reviewed-by: Ingo Molnar Cc: Jiri Olsa Link: http://lkml.kernel.org/r/1429198699-25039-2-git-send-email-pholasek@redhat.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c index ebfa163..cd872e9c 100644 --- a/tools/perf/bench/numa.c +++ b/tools/perf/bench/numa.c @@ -180,7 +180,7 @@ static const struct option options[] = { OPT_INTEGER('H', "thp" , &p0.thp, "MADV_NOHUGEPAGE < 0 < MADV_HUGEPAGE"), OPT_BOOLEAN('c', "show_convergence", &p0.show_convergence, "show convergence details"), OPT_BOOLEAN('m', "measure_convergence", &p0.measure_convergence, "measure convergence latency"), - OPT_BOOLEAN('q', "quiet" , &p0.show_quiet, "bzero the initial allocations"), + OPT_BOOLEAN('q', "quiet" , &p0.show_quiet, "quiet mode"), OPT_BOOLEAN('S', "serialize-startup", &p0.serialize_startup,"serialize thread startup"), /* Special option string parsing callbacks: */ @@ -1395,7 +1395,7 @@ static void print_res(const char *name, double val, if (!name) name = "main,"; - if (g->p.show_quiet) + if (!g->p.show_quiet) printf(" %-30s %15.3f, %-15s %s\n", name, val, txt_unit, txt_short); else printf(" %14.3f %s\n", val, txt_long); -- cgit v0.10.2 From 1d90a685eb75a56648d7dd22c704a1a6da516de9 Mon Sep 17 00:00:00 2001 From: Petr Holasek Date: Thu, 16 Apr 2015 17:38:19 +0200 Subject: perf bench numa: Fix immediate meeting of convergence condition This patch fixes the race in the beginning of benchmark run when some threads hasn't got assigned curr_cpu yet so they don't occur in nodes-of-process stats and benchmark concludes that all remaining threads are converged already. The race can be reproduced with small amount of threads and some bigger amount of shared process memory, e.g. one process, two threads and 5GB of process memory. Signed-off-by: Petr Holasek Reviewed-by: Ingo Molnar Cc: Jiri Olsa Link: http://lkml.kernel.org/r/1429198699-25039-4-git-send-email-pholasek@redhat.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c index cd872e9c..ba5efa4 100644 --- a/tools/perf/bench/numa.c +++ b/tools/perf/bench/numa.c @@ -828,6 +828,9 @@ static int count_process_nodes(int process_nr) td = g->threads + task_nr; node = numa_node_of_cpu(td->curr_cpu); + if (node < 0) /* curr_cpu was likely still -1 */ + return 0; + node_present[node] = 1; } @@ -882,6 +885,11 @@ static void calc_convergence_compression(int *strong) for (p = 0; p < g->p.nr_proc; p++) { unsigned int nodes = count_process_nodes(p); + if (!nodes) { + *strong = 0; + return; + } + nodes_min = min(nodes, nodes_min); nodes_max = max(nodes, nodes_max); } -- cgit v0.10.2 From 8c08a6215a432a8f3629da0e8ccfd534ffa480f5 Mon Sep 17 00:00:00 2001 From: Ondrej Zary Date: Tue, 21 Apr 2015 23:27:50 +0200 Subject: aha1542: Allocate memory before taking a lock The driver currently calls kmalloc with GFP_KERNEL while holding a lock. Fix it by doing the allocation earlier, before taking the lock. Tested on AHA-1542B. Signed-off-by: Ondrej Zary Signed-off-by: James Bottomley diff --git a/drivers/scsi/aha1542.c b/drivers/scsi/aha1542.c index ec43276..b95d277 100644 --- a/drivers/scsi/aha1542.c +++ b/drivers/scsi/aha1542.c @@ -375,9 +375,10 @@ static int aha1542_queuecommand(struct Scsi_Host *sh, struct scsi_cmnd *cmd) u8 lun = cmd->device->lun; unsigned long flags; int bufflen = scsi_bufflen(cmd); - int mbo; + int mbo, sg_count; struct mailbox *mb = aha1542->mb; struct ccb *ccb = aha1542->ccb; + struct chain *cptr; if (*cmd->cmnd == REQUEST_SENSE) { /* Don't do the command - we have the sense data already */ @@ -397,6 +398,13 @@ static int aha1542_queuecommand(struct Scsi_Host *sh, struct scsi_cmnd *cmd) print_hex_dump_bytes("command: ", DUMP_PREFIX_NONE, cmd->cmnd, cmd->cmd_len); } #endif + if (bufflen) { /* allocate memory before taking host_lock */ + sg_count = scsi_sg_count(cmd); + cptr = kmalloc(sizeof(*cptr) * sg_count, GFP_KERNEL | GFP_DMA); + if (!cptr) + return SCSI_MLQUEUE_HOST_BUSY; + } + /* Use the outgoing mailboxes in a round-robin fashion, because this is how the host adapter will scan for them */ @@ -441,19 +449,10 @@ static int aha1542_queuecommand(struct Scsi_Host *sh, struct scsi_cmnd *cmd) if (bufflen) { struct scatterlist *sg; - struct chain *cptr; - int i, sg_count = scsi_sg_count(cmd); + int i; ccb[mbo].op = 2; /* SCSI Initiator Command w/scatter-gather */ - cmd->host_scribble = kmalloc(sizeof(*cptr)*sg_count, - GFP_KERNEL | GFP_DMA); - cptr = (struct chain *) cmd->host_scribble; - if (cptr == NULL) { - /* free the claimed mailbox slot */ - aha1542->int_cmds[mbo] = NULL; - spin_unlock_irqrestore(sh->host_lock, flags); - return SCSI_MLQUEUE_HOST_BUSY; - } + cmd->host_scribble = (void *)cptr; scsi_for_each_sg(cmd, sg, sg_count, i) { any2scsi(cptr[i].dataptr, isa_page_to_bus(sg_page(sg)) + sg->offset); -- cgit v0.10.2 From 579d69bc1fd56d5af5761969aa529d1d1c188300 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Apr 2015 09:48:49 +0200 Subject: 3w-sas: fix command completion race The 3w-sas driver needs to tear down the dma mappings before returning the command to the midlayer, as there is no guarantee the sglist and count are valid after that point. Also remove the dma mapping helpers which have another inherent race due to the request_id index. Signed-off-by: Christoph Hellwig Reported-by: Torsten Luettgert Tested-by: Bernd Kardatzki Cc: stable@vger.kernel.org Acked-by: Adam Radford Signed-off-by: James Bottomley diff --git a/drivers/scsi/3w-sas.c b/drivers/scsi/3w-sas.c index 2361772..f837485 100644 --- a/drivers/scsi/3w-sas.c +++ b/drivers/scsi/3w-sas.c @@ -290,26 +290,6 @@ static int twl_post_command_packet(TW_Device_Extension *tw_dev, int request_id) return 0; } /* End twl_post_command_packet() */ -/* This function will perform a pci-dma mapping for a scatter gather list */ -static int twl_map_scsi_sg_data(TW_Device_Extension *tw_dev, int request_id) -{ - int use_sg; - struct scsi_cmnd *cmd = tw_dev->srb[request_id]; - - use_sg = scsi_dma_map(cmd); - if (!use_sg) - return 0; - else if (use_sg < 0) { - TW_PRINTK(tw_dev->host, TW_DRIVER, 0x1, "Failed to map scatter gather list"); - return 0; - } - - cmd->SCp.phase = TW_PHASE_SGLIST; - cmd->SCp.have_data_in = use_sg; - - return use_sg; -} /* End twl_map_scsi_sg_data() */ - /* This function hands scsi cdb's to the firmware */ static int twl_scsiop_execute_scsi(TW_Device_Extension *tw_dev, int request_id, char *cdb, int use_sg, TW_SG_Entry_ISO *sglistarg) { @@ -357,8 +337,8 @@ static int twl_scsiop_execute_scsi(TW_Device_Extension *tw_dev, int request_id, if (!sglistarg) { /* Map sglist from scsi layer to cmd packet */ if (scsi_sg_count(srb)) { - sg_count = twl_map_scsi_sg_data(tw_dev, request_id); - if (sg_count == 0) + sg_count = scsi_dma_map(srb); + if (sg_count <= 0) goto out; scsi_for_each_sg(srb, sg, sg_count, i) { @@ -1102,15 +1082,6 @@ out: return retval; } /* End twl_initialize_device_extension() */ -/* This function will perform a pci-dma unmap */ -static void twl_unmap_scsi_data(TW_Device_Extension *tw_dev, int request_id) -{ - struct scsi_cmnd *cmd = tw_dev->srb[request_id]; - - if (cmd->SCp.phase == TW_PHASE_SGLIST) - scsi_dma_unmap(cmd); -} /* End twl_unmap_scsi_data() */ - /* This function will handle attention interrupts */ static int twl_handle_attention_interrupt(TW_Device_Extension *tw_dev) { @@ -1251,11 +1222,11 @@ static irqreturn_t twl_interrupt(int irq, void *dev_instance) } /* Now complete the io */ + scsi_dma_unmap(cmd); + cmd->scsi_done(cmd); tw_dev->state[request_id] = TW_S_COMPLETED; twl_free_request_id(tw_dev, request_id); tw_dev->posted_request_count--; - tw_dev->srb[request_id]->scsi_done(tw_dev->srb[request_id]); - twl_unmap_scsi_data(tw_dev, request_id); } /* Check for another response interrupt */ @@ -1400,10 +1371,12 @@ static int twl_reset_device_extension(TW_Device_Extension *tw_dev, int ioctl_res if ((tw_dev->state[i] != TW_S_FINISHED) && (tw_dev->state[i] != TW_S_INITIAL) && (tw_dev->state[i] != TW_S_COMPLETED)) { - if (tw_dev->srb[i]) { - tw_dev->srb[i]->result = (DID_RESET << 16); - tw_dev->srb[i]->scsi_done(tw_dev->srb[i]); - twl_unmap_scsi_data(tw_dev, i); + struct scsi_cmnd *cmd = tw_dev->srb[i]; + + if (cmd) { + cmd->result = (DID_RESET << 16); + scsi_dma_unmap(cmd); + cmd->scsi_done(cmd); } } } @@ -1507,9 +1480,6 @@ static int twl_scsi_queue_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_ /* Save the scsi command for use by the ISR */ tw_dev->srb[request_id] = SCpnt; - /* Initialize phase to zero */ - SCpnt->SCp.phase = TW_PHASE_INITIAL; - retval = twl_scsiop_execute_scsi(tw_dev, request_id, NULL, 0, NULL); if (retval) { tw_dev->state[request_id] = TW_S_COMPLETED; diff --git a/drivers/scsi/3w-sas.h b/drivers/scsi/3w-sas.h index d474892..fec6449 100644 --- a/drivers/scsi/3w-sas.h +++ b/drivers/scsi/3w-sas.h @@ -103,10 +103,6 @@ static char *twl_aen_severity_table[] = #define TW_CURRENT_DRIVER_BUILD 0 #define TW_CURRENT_DRIVER_BRANCH 0 -/* Phase defines */ -#define TW_PHASE_INITIAL 0 -#define TW_PHASE_SGLIST 2 - /* Misc defines */ #define TW_SECTOR_SIZE 512 #define TW_MAX_UNITS 32 -- cgit v0.10.2 From 9cd9554615cba14f0877cc9972a6537ad2bdde61 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Apr 2015 09:48:50 +0200 Subject: 3w-xxxx: fix command completion race The 3w-xxxx driver needs to tear down the dma mappings before returning the command to the midlayer, as there is no guarantee the sglist and count are valid after that point. Also remove the dma mapping helpers which have another inherent race due to the request_id index. Signed-off-by: Christoph Hellwig Cc: stable@vger.kernel.org Acked-by: Adam Radford Signed-off-by: James Bottomley diff --git a/drivers/scsi/3w-xxxx.c b/drivers/scsi/3w-xxxx.c index c75f204..2940bd7 100644 --- a/drivers/scsi/3w-xxxx.c +++ b/drivers/scsi/3w-xxxx.c @@ -1271,32 +1271,6 @@ static int tw_initialize_device_extension(TW_Device_Extension *tw_dev) return 0; } /* End tw_initialize_device_extension() */ -static int tw_map_scsi_sg_data(struct pci_dev *pdev, struct scsi_cmnd *cmd) -{ - int use_sg; - - dprintk(KERN_WARNING "3w-xxxx: tw_map_scsi_sg_data()\n"); - - use_sg = scsi_dma_map(cmd); - if (use_sg < 0) { - printk(KERN_WARNING "3w-xxxx: tw_map_scsi_sg_data(): pci_map_sg() failed.\n"); - return 0; - } - - cmd->SCp.phase = TW_PHASE_SGLIST; - cmd->SCp.have_data_in = use_sg; - - return use_sg; -} /* End tw_map_scsi_sg_data() */ - -static void tw_unmap_scsi_data(struct pci_dev *pdev, struct scsi_cmnd *cmd) -{ - dprintk(KERN_WARNING "3w-xxxx: tw_unmap_scsi_data()\n"); - - if (cmd->SCp.phase == TW_PHASE_SGLIST) - scsi_dma_unmap(cmd); -} /* End tw_unmap_scsi_data() */ - /* This function will reset a device extension */ static int tw_reset_device_extension(TW_Device_Extension *tw_dev) { @@ -1319,8 +1293,8 @@ static int tw_reset_device_extension(TW_Device_Extension *tw_dev) srb = tw_dev->srb[i]; if (srb != NULL) { srb->result = (DID_RESET << 16); - tw_dev->srb[i]->scsi_done(tw_dev->srb[i]); - tw_unmap_scsi_data(tw_dev->tw_pci_dev, tw_dev->srb[i]); + scsi_dma_unmap(srb); + srb->scsi_done(srb); } } } @@ -1767,8 +1741,8 @@ static int tw_scsiop_read_write(TW_Device_Extension *tw_dev, int request_id) command_packet->byte8.io.lba = lba; command_packet->byte6.block_count = num_sectors; - use_sg = tw_map_scsi_sg_data(tw_dev->tw_pci_dev, tw_dev->srb[request_id]); - if (!use_sg) + use_sg = scsi_dma_map(srb); + if (use_sg <= 0) return 1; scsi_for_each_sg(tw_dev->srb[request_id], sg, use_sg, i) { @@ -1955,9 +1929,6 @@ static int tw_scsi_queue_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_c /* Save the scsi command for use by the ISR */ tw_dev->srb[request_id] = SCpnt; - /* Initialize phase to zero */ - SCpnt->SCp.phase = TW_PHASE_INITIAL; - switch (*command) { case READ_10: case READ_6: @@ -2185,12 +2156,11 @@ static irqreturn_t tw_interrupt(int irq, void *dev_instance) /* Now complete the io */ if ((error != TW_ISR_DONT_COMPLETE)) { + scsi_dma_unmap(tw_dev->srb[request_id]); + tw_dev->srb[request_id]->scsi_done(tw_dev->srb[request_id]); tw_dev->state[request_id] = TW_S_COMPLETED; tw_state_request_finish(tw_dev, request_id); tw_dev->posted_request_count--; - tw_dev->srb[request_id]->scsi_done(tw_dev->srb[request_id]); - - tw_unmap_scsi_data(tw_dev->tw_pci_dev, tw_dev->srb[request_id]); } } diff --git a/drivers/scsi/3w-xxxx.h b/drivers/scsi/3w-xxxx.h index 29b0b84e..6f65e66 100644 --- a/drivers/scsi/3w-xxxx.h +++ b/drivers/scsi/3w-xxxx.h @@ -195,11 +195,6 @@ static unsigned char tw_sense_table[][4] = #define TW_AEN_SMART_FAIL 0x000F #define TW_AEN_SBUF_FAIL 0x0024 -/* Phase defines */ -#define TW_PHASE_INITIAL 0 -#define TW_PHASE_SINGLE 1 -#define TW_PHASE_SGLIST 2 - /* Misc defines */ #define TW_ALIGNMENT_6000 64 /* 64 bytes */ #define TW_ALIGNMENT_7000 4 /* 4 bytes */ -- cgit v0.10.2 From 118c855b5623f3e2e6204f02623d88c09e0c34de Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Apr 2015 09:48:51 +0200 Subject: 3w-9xxx: fix command completion race The 3w-9xxx driver needs to tear down the dma mappings before returning the command to the midlayer, as there is no guarantee the sglist and count are valid after that point. Also remove the dma mapping helpers which have another inherent race due to the request_id index. Signed-off-by: Christoph Hellwig Cc: stable@vger.kernel.org Acked-by: Adam Radford Signed-off-by: James Bottomley diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c index 7600639..add419d 100644 --- a/drivers/scsi/3w-9xxx.c +++ b/drivers/scsi/3w-9xxx.c @@ -149,7 +149,6 @@ static int twa_reset_sequence(TW_Device_Extension *tw_dev, int soft_reset); static int twa_scsiop_execute_scsi(TW_Device_Extension *tw_dev, int request_id, char *cdb, int use_sg, TW_SG_Entry *sglistarg); static void twa_scsiop_execute_scsi_complete(TW_Device_Extension *tw_dev, int request_id); static char *twa_string_lookup(twa_message_type *table, unsigned int aen_code); -static void twa_unmap_scsi_data(TW_Device_Extension *tw_dev, int request_id); /* Functions */ @@ -1340,11 +1339,11 @@ static irqreturn_t twa_interrupt(int irq, void *dev_instance) } /* Now complete the io */ + scsi_dma_unmap(cmd); + cmd->scsi_done(cmd); tw_dev->state[request_id] = TW_S_COMPLETED; twa_free_request_id(tw_dev, request_id); tw_dev->posted_request_count--; - tw_dev->srb[request_id]->scsi_done(tw_dev->srb[request_id]); - twa_unmap_scsi_data(tw_dev, request_id); } /* Check for valid status after each drain */ @@ -1402,26 +1401,6 @@ static void twa_load_sgl(TW_Device_Extension *tw_dev, TW_Command_Full *full_comm } } /* End twa_load_sgl() */ -/* This function will perform a pci-dma mapping for a scatter gather list */ -static int twa_map_scsi_sg_data(TW_Device_Extension *tw_dev, int request_id) -{ - int use_sg; - struct scsi_cmnd *cmd = tw_dev->srb[request_id]; - - use_sg = scsi_dma_map(cmd); - if (!use_sg) - return 0; - else if (use_sg < 0) { - TW_PRINTK(tw_dev->host, TW_DRIVER, 0x1c, "Failed to map scatter gather list"); - return 0; - } - - cmd->SCp.phase = TW_PHASE_SGLIST; - cmd->SCp.have_data_in = use_sg; - - return use_sg; -} /* End twa_map_scsi_sg_data() */ - /* This function will poll for a response interrupt of a request */ static int twa_poll_response(TW_Device_Extension *tw_dev, int request_id, int seconds) { @@ -1600,9 +1579,11 @@ static int twa_reset_device_extension(TW_Device_Extension *tw_dev) (tw_dev->state[i] != TW_S_INITIAL) && (tw_dev->state[i] != TW_S_COMPLETED)) { if (tw_dev->srb[i]) { - tw_dev->srb[i]->result = (DID_RESET << 16); - tw_dev->srb[i]->scsi_done(tw_dev->srb[i]); - twa_unmap_scsi_data(tw_dev, i); + struct scsi_cmnd *cmd = tw_dev->srb[i]; + + cmd->result = (DID_RESET << 16); + scsi_dma_unmap(cmd); + cmd->scsi_done(cmd); } } } @@ -1781,21 +1762,18 @@ static int twa_scsi_queue_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_ /* Save the scsi command for use by the ISR */ tw_dev->srb[request_id] = SCpnt; - /* Initialize phase to zero */ - SCpnt->SCp.phase = TW_PHASE_INITIAL; - retval = twa_scsiop_execute_scsi(tw_dev, request_id, NULL, 0, NULL); switch (retval) { case SCSI_MLQUEUE_HOST_BUSY: + scsi_dma_unmap(SCpnt); twa_free_request_id(tw_dev, request_id); - twa_unmap_scsi_data(tw_dev, request_id); break; case 1: - tw_dev->state[request_id] = TW_S_COMPLETED; - twa_free_request_id(tw_dev, request_id); - twa_unmap_scsi_data(tw_dev, request_id); SCpnt->result = (DID_ERROR << 16); + scsi_dma_unmap(SCpnt); done(SCpnt); + tw_dev->state[request_id] = TW_S_COMPLETED; + twa_free_request_id(tw_dev, request_id); retval = 0; } out: @@ -1863,8 +1841,8 @@ static int twa_scsiop_execute_scsi(TW_Device_Extension *tw_dev, int request_id, command_packet->sg_list[0].address = TW_CPU_TO_SGL(tw_dev->generic_buffer_phys[request_id]); command_packet->sg_list[0].length = cpu_to_le32(TW_MIN_SGL_LENGTH); } else { - sg_count = twa_map_scsi_sg_data(tw_dev, request_id); - if (sg_count == 0) + sg_count = scsi_dma_map(srb); + if (sg_count < 0) goto out; scsi_for_each_sg(srb, sg, sg_count, i) { @@ -1979,15 +1957,6 @@ static char *twa_string_lookup(twa_message_type *table, unsigned int code) return(table[index].text); } /* End twa_string_lookup() */ -/* This function will perform a pci-dma unmap */ -static void twa_unmap_scsi_data(TW_Device_Extension *tw_dev, int request_id) -{ - struct scsi_cmnd *cmd = tw_dev->srb[request_id]; - - if (cmd->SCp.phase == TW_PHASE_SGLIST) - scsi_dma_unmap(cmd); -} /* End twa_unmap_scsi_data() */ - /* This function gets called when a disk is coming on-line */ static int twa_slave_configure(struct scsi_device *sdev) { diff --git a/drivers/scsi/3w-9xxx.h b/drivers/scsi/3w-9xxx.h index 040f721..0fdc83c 100644 --- a/drivers/scsi/3w-9xxx.h +++ b/drivers/scsi/3w-9xxx.h @@ -324,11 +324,6 @@ static twa_message_type twa_error_table[] = { #define TW_CURRENT_DRIVER_BUILD 0 #define TW_CURRENT_DRIVER_BRANCH 0 -/* Phase defines */ -#define TW_PHASE_INITIAL 0 -#define TW_PHASE_SINGLE 1 -#define TW_PHASE_SGLIST 2 - /* Misc defines */ #define TW_9550SX_DRAIN_COMPLETED 0xFFFF #define TW_SECTOR_SIZE 512 -- cgit v0.10.2 From 129d23a56623eea0947a05288158d76dc7f2f0ac Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 27 Apr 2015 13:20:34 -0400 Subject: netfilter; Add some missing default cases to switch statements in nft_reject. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This fixes: ==================== net/netfilter/nft_reject.c: In function ‘nft_reject_dump’: net/netfilter/nft_reject.c:61:2: warning: enumeration value ‘NFT_REJECT_TCP_RST’ not handled in switch [-Wswitch] switch (priv->type) { ^ net/netfilter/nft_reject.c:61:2: warning: enumeration value ‘NFT_REJECT_ICMPX_UNREACH’ not handled in switch [-Wswi\ tch] net/netfilter/nft_reject_inet.c: In function ‘nft_reject_inet_dump’: net/netfilter/nft_reject_inet.c:105:2: warning: enumeration value ‘NFT_REJECT_TCP_RST’ not handled in switch [-Wswi\ tch] switch (priv->type) { ^ ==================== Signed-off-by: David S. Miller diff --git a/net/netfilter/nft_reject.c b/net/netfilter/nft_reject.c index 57d3e1a..0522fc9 100644 --- a/net/netfilter/nft_reject.c +++ b/net/netfilter/nft_reject.c @@ -63,6 +63,8 @@ int nft_reject_dump(struct sk_buff *skb, const struct nft_expr *expr) if (nla_put_u8(skb, NFTA_REJECT_ICMP_CODE, priv->icmp_code)) goto nla_put_failure; break; + default: + break; } return 0; diff --git a/net/netfilter/nft_reject_inet.c b/net/netfilter/nft_reject_inet.c index 62cabee..635dbba 100644 --- a/net/netfilter/nft_reject_inet.c +++ b/net/netfilter/nft_reject_inet.c @@ -108,6 +108,8 @@ static int nft_reject_inet_dump(struct sk_buff *skb, if (nla_put_u8(skb, NFTA_REJECT_ICMP_CODE, priv->icmp_code)) goto nla_put_failure; break; + default: + break; } return 0; -- cgit v0.10.2 From 33df10e2ee414183953cc89453bf9c85027a188b Mon Sep 17 00:00:00 2001 From: Rojhalat Ibrahim Date: Mon, 27 Apr 2015 10:37:31 +0200 Subject: mdio-mux-gpio: use new gpiod_get_array and gpiod_put_array functions Use the new gpiod_get_array and gpiod_put_array functions (added to mainline in the v4.1 merge window) for obtaining and disposing of GPIO descriptors. Cc: David Miller Cc: Linus Walleij Signed-off-by: Rojhalat Ibrahim Signed-off-by: David S. Miller diff --git a/drivers/net/phy/mdio-mux-gpio.c b/drivers/net/phy/mdio-mux-gpio.c index 1a87a58..66edd99 100644 --- a/drivers/net/phy/mdio-mux-gpio.c +++ b/drivers/net/phy/mdio-mux-gpio.c @@ -12,33 +12,30 @@ #include #include #include -#include +#include #define DRV_VERSION "1.1" #define DRV_DESCRIPTION "GPIO controlled MDIO bus multiplexer driver" -#define MDIO_MUX_GPIO_MAX_BITS 8 - struct mdio_mux_gpio_state { - struct gpio_desc *gpio[MDIO_MUX_GPIO_MAX_BITS]; - unsigned int num_gpios; + struct gpio_descs *gpios; void *mux_handle; }; static int mdio_mux_gpio_switch_fn(int current_child, int desired_child, void *data) { - int values[MDIO_MUX_GPIO_MAX_BITS]; - unsigned int n; struct mdio_mux_gpio_state *s = data; + int values[s->gpios->ndescs]; + unsigned int n; if (current_child == desired_child) return 0; - for (n = 0; n < s->num_gpios; n++) { + for (n = 0; n < s->gpios->ndescs; n++) values[n] = (desired_child >> n) & 1; - } - gpiod_set_array_cansleep(s->num_gpios, s->gpio, values); + + gpiod_set_array_cansleep(s->gpios->ndescs, s->gpios->desc, values); return 0; } @@ -46,56 +43,33 @@ static int mdio_mux_gpio_switch_fn(int current_child, int desired_child, static int mdio_mux_gpio_probe(struct platform_device *pdev) { struct mdio_mux_gpio_state *s; - int num_gpios; - unsigned int n; int r; - if (!pdev->dev.of_node) - return -ENODEV; - - num_gpios = of_gpio_count(pdev->dev.of_node); - if (num_gpios <= 0 || num_gpios > MDIO_MUX_GPIO_MAX_BITS) - return -ENODEV; - s = devm_kzalloc(&pdev->dev, sizeof(*s), GFP_KERNEL); if (!s) return -ENOMEM; - s->num_gpios = num_gpios; - - for (n = 0; n < num_gpios; ) { - struct gpio_desc *gpio = gpiod_get_index(&pdev->dev, NULL, n, - GPIOD_OUT_LOW); - if (IS_ERR(gpio)) { - r = PTR_ERR(gpio); - goto err; - } - s->gpio[n] = gpio; - n++; - } + s->gpios = gpiod_get_array(&pdev->dev, NULL, GPIOD_OUT_LOW); + if (IS_ERR(s->gpios)) + return PTR_ERR(s->gpios); r = mdio_mux_init(&pdev->dev, mdio_mux_gpio_switch_fn, &s->mux_handle, s); - if (r == 0) { - pdev->dev.platform_data = s; - return 0; - } -err: - while (n) { - n--; - gpiod_put(s->gpio[n]); + if (r != 0) { + gpiod_put_array(s->gpios); + return r; } - return r; + + pdev->dev.platform_data = s; + return 0; } static int mdio_mux_gpio_remove(struct platform_device *pdev) { - unsigned int n; struct mdio_mux_gpio_state *s = dev_get_platdata(&pdev->dev); mdio_mux_uninit(s->mux_handle); - for (n = 0; n < s->num_gpios; n++) - gpiod_put(s->gpio[n]); + gpiod_put_array(s->gpios); return 0; } -- cgit v0.10.2 From b37069090b7c5615610a8aa6b36533d67b364d38 Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Mon, 27 Apr 2015 13:40:56 +0300 Subject: net/mlx4_en: Prevent setting invalid RSS hash function mlx4_en_check_rxfh_func() was checking for hardware support before setting a known RSS hash function, but didn't do any check before setting unknown RSS hash function. Need to make it fail on such values. In this occasion, moved the actual setting of the new value from the check function into mlx4_en_set_rxfh(). Fixes: 947cbb0 ("net/mlx4_en: Support for configurable RSS hash function") Signed-off-by: Amir Vadai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index 3f44e2b..a2ddf3d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1102,20 +1102,21 @@ static int mlx4_en_check_rxfh_func(struct net_device *dev, u8 hfunc) struct mlx4_en_priv *priv = netdev_priv(dev); /* check if requested function is supported by the device */ - if ((hfunc == ETH_RSS_HASH_TOP && - !(priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_RSS_TOP)) || - (hfunc == ETH_RSS_HASH_XOR && - !(priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_RSS_XOR))) - return -EINVAL; + if (hfunc == ETH_RSS_HASH_TOP) { + if (!(priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_RSS_TOP)) + return -EINVAL; + if (!(dev->features & NETIF_F_RXHASH)) + en_warn(priv, "Toeplitz hash function should be used in conjunction with RX hashing for optimal performance\n"); + return 0; + } else if (hfunc == ETH_RSS_HASH_XOR) { + if (!(priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_RSS_XOR)) + return -EINVAL; + if (dev->features & NETIF_F_RXHASH) + en_warn(priv, "Enabling both XOR Hash function and RX Hashing can limit RPS functionality\n"); + return 0; + } - priv->rss_hash_fn = hfunc; - if (hfunc == ETH_RSS_HASH_TOP && !(dev->features & NETIF_F_RXHASH)) - en_warn(priv, - "Toeplitz hash function should be used in conjunction with RX hashing for optimal performance\n"); - if (hfunc == ETH_RSS_HASH_XOR && (dev->features & NETIF_F_RXHASH)) - en_warn(priv, - "Enabling both XOR Hash function and RX Hashing can limit RPS functionality\n"); - return 0; + return -EINVAL; } static int mlx4_en_get_rxfh(struct net_device *dev, u32 *ring_index, u8 *key, @@ -1189,6 +1190,8 @@ static int mlx4_en_set_rxfh(struct net_device *dev, const u32 *ring_index, priv->prof->rss_rings = rss_rings; if (key) memcpy(priv->rss_key, key, MLX4_EN_RSS_KEY_SIZE); + if (hfunc != ETH_RSS_HASH_NO_CHANGE) + priv->rss_hash_fn = hfunc; if (port_up) { err = mlx4_en_start_port(dev); -- cgit v0.10.2 From 94435f764cc5838a7e94008f17628ad63384bf06 Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Mon, 27 Apr 2015 23:14:57 +0900 Subject: net:treewide: Fix typo in drivers/net This patch fix spelling typo in printk. Signed-off-by: Masanari Iida Signed-off-by: David S. Miller diff --git a/drivers/net/can/usb/kvaser_usb.c b/drivers/net/can/usb/kvaser_usb.c index 4643914..8b17a90 100644 --- a/drivers/net/can/usb/kvaser_usb.c +++ b/drivers/net/can/usb/kvaser_usb.c @@ -1102,7 +1102,7 @@ static void kvaser_usb_rx_can_err(const struct kvaser_usb_net_priv *priv, if (msg->u.rx_can_header.flag & (MSG_FLAG_ERROR_FRAME | MSG_FLAG_NERR)) { - netdev_err(priv->netdev, "Unknow error (flags: 0x%02x)\n", + netdev_err(priv->netdev, "Unknown error (flags: 0x%02x)\n", msg->u.rx_can_header.flag); stats->rx_errors++; diff --git a/drivers/net/ethernet/8390/etherh.c b/drivers/net/ethernet/8390/etherh.c index b36ee9e..d686b9c 100644 --- a/drivers/net/ethernet/8390/etherh.c +++ b/drivers/net/ethernet/8390/etherh.c @@ -523,7 +523,7 @@ static int etherh_addr(char *addr, struct expansion_card *ec) char *s; if (!ecard_readchunk(&cd, ec, 0xf5, 0)) { - printk(KERN_ERR "%s: unable to read podule description string\n", + printk(KERN_ERR "%s: unable to read module description string\n", dev_name(&ec->dev)); goto no_addr; } -- cgit v0.10.2 From 22a8f237c0551bae95ffcd2a7ff17d6f5fcce7e7 Mon Sep 17 00:00:00 2001 From: Michal Schmidt Date: Mon, 27 Apr 2015 17:20:38 +0200 Subject: bnx2x: really disable TPA if 'disable_tpa' option is set bnx2x's 'disable_tpa=1' module option is not respected properly and TPA (transparent packet aggregation) remains enabled. Even though the module option causes LRO to be disabled, TPA is enabled in GRO mode. Additionally, disabling GRO via ethtool then has no effect. One can still observe tpa_* statistics increase and large packets being received in tcpdump. The bug was an unintended consequence of commit aebf6244cd39 "bnx2x: Be more forgiving toward SW GRO". Fix it by following the bp->disable_tpa flag when initializing fp's. Signed-off-by: Michal Schmidt Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 6f7dc81..3558a36 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -2485,8 +2485,10 @@ static void bnx2x_bz_fp(struct bnx2x *bp, int index) else if (bp->flags & GRO_ENABLE_FLAG) fp->mode = TPA_MODE_GRO; - /* We don't want TPA on an FCoE L2 ring */ - if (IS_FCOE_FP(fp)) + /* We don't want TPA if it's disabled in bp + * or if this is an FCoE L2 ring. + */ + if (bp->disable_tpa || IS_FCOE_FP(fp)) fp->disable_tpa = 1; } -- cgit v0.10.2 From 2a700d8edffdbfb8200332d96c3147e042b337f1 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Mon, 13 Apr 2015 11:18:51 -0300 Subject: [media] marvell-ccic: fix Y'CbCr ordering Various formats had their byte ordering implemented incorrectly, and the V4L2_PIX_FMT_UYVY is actually impossible to create, instead you get V4L2_PIX_FMT_YVYU. This was working before commit ad6ac452227b7cb93ac79beec092850d178740b1 ("add new formats support for marvell-ccic driver"). That commit broke the original format support and the OLPC XO-1 laptop showed wrong colors ever since (if you are crazy enough to attempt to run the latest kernel on it, like I did). The email addresses of the authors of that patch are no longer valid, so without a way to reach them and ask them about their test setup I am going with what I can test on the OLPC laptop. If this breaks something for someone on their non-OLPC setup, then contact the linux-media mailinglist. My suspicion however is that that commit went in untested. Signed-off-by: Hans Verkuil Acked-by: Jonathan Corbet Cc: # for v3.19 and up Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/platform/marvell-ccic/mcam-core.c b/drivers/media/platform/marvell-ccic/mcam-core.c index 9c64b5d..110fd70 100644 --- a/drivers/media/platform/marvell-ccic/mcam-core.c +++ b/drivers/media/platform/marvell-ccic/mcam-core.c @@ -116,8 +116,8 @@ static struct mcam_format_struct { .planar = false, }, { - .desc = "UYVY 4:2:2", - .pixelformat = V4L2_PIX_FMT_UYVY, + .desc = "YVYU 4:2:2", + .pixelformat = V4L2_PIX_FMT_YVYU, .mbus_code = MEDIA_BUS_FMT_YUYV8_2X8, .bpp = 2, .planar = false, @@ -748,7 +748,7 @@ static void mcam_ctlr_image(struct mcam_camera *cam) switch (fmt->pixelformat) { case V4L2_PIX_FMT_YUYV: - case V4L2_PIX_FMT_UYVY: + case V4L2_PIX_FMT_YVYU: widthy = fmt->width * 2; widthuv = 0; break; @@ -784,15 +784,15 @@ static void mcam_ctlr_image(struct mcam_camera *cam) case V4L2_PIX_FMT_YUV420: case V4L2_PIX_FMT_YVU420: mcam_reg_write_mask(cam, REG_CTRL0, - C0_DF_YUV | C0_YUV_420PL | C0_YUVE_YVYU, C0_DF_MASK); + C0_DF_YUV | C0_YUV_420PL | C0_YUVE_VYUY, C0_DF_MASK); break; case V4L2_PIX_FMT_YUYV: mcam_reg_write_mask(cam, REG_CTRL0, - C0_DF_YUV | C0_YUV_PACKED | C0_YUVE_UYVY, C0_DF_MASK); + C0_DF_YUV | C0_YUV_PACKED | C0_YUVE_NOSWAP, C0_DF_MASK); break; - case V4L2_PIX_FMT_UYVY: + case V4L2_PIX_FMT_YVYU: mcam_reg_write_mask(cam, REG_CTRL0, - C0_DF_YUV | C0_YUV_PACKED | C0_YUVE_YUYV, C0_DF_MASK); + C0_DF_YUV | C0_YUV_PACKED | C0_YUVE_SWAP24, C0_DF_MASK); break; case V4L2_PIX_FMT_JPEG: mcam_reg_write_mask(cam, REG_CTRL0, diff --git a/drivers/media/platform/marvell-ccic/mcam-core.h b/drivers/media/platform/marvell-ccic/mcam-core.h index aa0c6ea..7ffdf4d 100644 --- a/drivers/media/platform/marvell-ccic/mcam-core.h +++ b/drivers/media/platform/marvell-ccic/mcam-core.h @@ -330,10 +330,10 @@ int mccic_resume(struct mcam_camera *cam); #define C0_YUVE_YVYU 0x00010000 /* Y1CrY0Cb */ #define C0_YUVE_VYUY 0x00020000 /* CrY1CbY0 */ #define C0_YUVE_UYVY 0x00030000 /* CbY1CrY0 */ -#define C0_YUVE_XYUV 0x00000000 /* 420: .YUV */ -#define C0_YUVE_XYVU 0x00010000 /* 420: .YVU */ -#define C0_YUVE_XUVY 0x00020000 /* 420: .UVY */ -#define C0_YUVE_XVUY 0x00030000 /* 420: .VUY */ +#define C0_YUVE_NOSWAP 0x00000000 /* no bytes swapping */ +#define C0_YUVE_SWAP13 0x00010000 /* swap byte 1 and 3 */ +#define C0_YUVE_SWAP24 0x00020000 /* swap byte 2 and 4 */ +#define C0_YUVE_SWAP1324 0x00030000 /* swap bytes 1&3 and 2&4 */ /* Bayer bits 18,19 if needed */ #define C0_EOF_VSYNC 0x00400000 /* Generate EOF by VSYNC */ #define C0_VEDGE_CTRL 0x00800000 /* Detect falling edge of VSYNC */ -- cgit v0.10.2 From 5a9b06a27db6b006605421658418fb8943a6e217 Mon Sep 17 00:00:00 2001 From: Koji Matsuoka Date: Sun, 29 Mar 2015 10:04:56 -0300 Subject: [media] media: soc_camera: rcar_vin: Fix wait_for_completion When stopping abnormally, a driver can't return from wait_for_completion. This patch resolved this problem by changing wait_for_completion_timeout from wait_for_completion. Signed-off-by: Koji Matsuoka Signed-off-by: Yoshihiro Kaneko Signed-off-by: Guennadi Liakhovetski Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/platform/soc_camera/rcar_vin.c b/drivers/media/platform/soc_camera/rcar_vin.c index 9351f64..6460f8e 100644 --- a/drivers/media/platform/soc_camera/rcar_vin.c +++ b/drivers/media/platform/soc_camera/rcar_vin.c @@ -135,6 +135,8 @@ #define VIN_MAX_WIDTH 2048 #define VIN_MAX_HEIGHT 2048 +#define TIMEOUT_MS 100 + enum chip_id { RCAR_GEN2, RCAR_H1, @@ -820,7 +822,10 @@ static void rcar_vin_wait_stop_streaming(struct rcar_vin_priv *priv) if (priv->state == STOPPING) { priv->request_to_stop = true; spin_unlock_irq(&priv->lock); - wait_for_completion(&priv->capture_stop); + if (!wait_for_completion_timeout( + &priv->capture_stop, + msecs_to_jiffies(TIMEOUT_MS))) + priv->state = STOPPED; spin_lock_irq(&priv->lock); } } -- cgit v0.10.2 From 53f9b3baa937e0cbdd75ea11b3c824462e4359b2 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 21 Apr 2015 12:19:49 +0800 Subject: ASoC: rt5645: Fix mask for setting RT5645_DMIC_2_DP_GPIO12 bit Current code uses wrong mask when setting RT5645_DMIC_2_DP_GPIO12 bit, fix it. Signed-off-by: Axel Lin Signed-off-by: Mark Brown diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index e16724a..3153aa0 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -2742,7 +2742,7 @@ static int rt5645_i2c_probe(struct i2c_client *i2c, case RT5645_DMIC_DATA_GPIO12: regmap_update_bits(rt5645->regmap, RT5645_DMIC_CTRL1, - RT5645_DMIC_1_DP_MASK, RT5645_DMIC_2_DP_GPIO12); + RT5645_DMIC_2_DP_MASK, RT5645_DMIC_2_DP_GPIO12); regmap_update_bits(rt5645->regmap, RT5645_GPIO_CTRL1, RT5645_GP12_PIN_MASK, RT5645_GP12_PIN_DMIC2_SDA); -- cgit v0.10.2 From 68ce9a1f24b8e8894955032a4af74998315be9fe Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Tue, 28 Apr 2015 09:05:21 +0800 Subject: ARM: dts: imx6qdl-sabreauto: remove pinctrl-assert-gpios The pinctrl-assert-gpios is an invalid pinctrl property. It was probably sneaked from vendor tree. Remove it. Fixes: 4e18a2243a87 ("ARM: imx6qdl-sabreauto.dtsi: add max7310 support") Signed-off-by: Shawn Guo diff --git a/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi b/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi index 46b2fed..3b24b126 100644 --- a/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi +++ b/arch/arm/boot/dts/imx6qdl-sabreauto.dtsi @@ -185,7 +185,6 @@ &i2c3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_i2c3>; - pinctrl-assert-gpios = <&gpio5 4 GPIO_ACTIVE_HIGH>; status = "okay"; max7310_a: gpio@30 { -- cgit v0.10.2 From 8e046d68ba719a5bb95912c154c6314472edccfa Mon Sep 17 00:00:00 2001 From: "Karicheri, Muralidharan" Date: Mon, 27 Apr 2015 14:12:43 -0400 Subject: net: netcp: remove call to netif_carrier_(on/off) for MAC to Phy interface Currently when interface type is MAC to Phy, netif_carrier_(on/off) is called which is not needed as Phy lib already updates the carrier status to net stack. This is needed only for other interface types Signed-off-by: Murali Karicheri Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 2bef655..9b7e0a3 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -1765,7 +1765,9 @@ static void netcp_ethss_link_state_action(struct gbe_priv *gbe_dev, ALE_PORT_STATE, ALE_PORT_STATE_FORWARD); - if (ndev && slave->open) + if (ndev && slave->open && + slave->link_interface != SGMII_LINK_MAC_PHY && + slave->link_interface != XGMII_LINK_MAC_PHY) netif_carrier_on(ndev); } else { writel(mac_control, GBE_REG_ADDR(slave, emac_regs, @@ -1773,7 +1775,9 @@ static void netcp_ethss_link_state_action(struct gbe_priv *gbe_dev, cpsw_ale_control_set(gbe_dev->ale, slave->port_num, ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); - if (ndev) + if (ndev && + slave->link_interface != SGMII_LINK_MAC_PHY && + slave->link_interface != XGMII_LINK_MAC_PHY) netif_carrier_off(ndev); } -- cgit v0.10.2 From 876a7ae65b86d8cec8efe7d15d050ac61116874e Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Mon, 27 Apr 2015 14:40:37 -0700 Subject: bpf: fix 64-bit divide ALU64_DIV instruction should be dividing 64-bit by 64-bit, whereas do_div() does 64-bit by 32-bit divide. x64 and arm64 JITs correctly implement 64 by 64 unsigned divide. llvm BPF backend emits code assuming that ALU64_DIV does 64 by 64. Fixes: 89aa075832b0 ("net: sock: allow eBPF programs to be attached to sockets") Reported-by: Michael Holzheu Acked-by: Daniel Borkmann Signed-off-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 4139a0f..54f0e7f 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -357,8 +357,8 @@ select_insn: ALU64_MOD_X: if (unlikely(SRC == 0)) return 0; - tmp = DST; - DST = do_div(tmp, SRC); + div64_u64_rem(DST, SRC, &tmp); + DST = tmp; CONT; ALU_MOD_X: if (unlikely(SRC == 0)) @@ -367,8 +367,8 @@ select_insn: DST = do_div(tmp, (u32) SRC); CONT; ALU64_MOD_K: - tmp = DST; - DST = do_div(tmp, IMM); + div64_u64_rem(DST, IMM, &tmp); + DST = tmp; CONT; ALU_MOD_K: tmp = (u32) DST; @@ -377,7 +377,7 @@ select_insn: ALU64_DIV_X: if (unlikely(SRC == 0)) return 0; - do_div(DST, SRC); + DST = div64_u64(DST, SRC); CONT; ALU_DIV_X: if (unlikely(SRC == 0)) @@ -387,7 +387,7 @@ select_insn: DST = (u32) tmp; CONT; ALU64_DIV_K: - do_div(DST, IMM); + DST = div64_u64(DST, IMM); CONT; ALU_DIV_K: tmp = (u32) DST; -- cgit v0.10.2 From a5011d44f0e1117a6db14b19b57c51f8be5673a0 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Tue, 21 Apr 2015 11:20:30 +0200 Subject: uas: Allow uas_use_uas_driver to return usb-storage flags uas_use_uas_driver may set some US_FL_foo flags during detection, currently these are stored in a local variable and then throw away, but these may be of interest to the caller, so add an extra parameter to (optionally) return the detected flags, and use this in the uas driver. Cc: stable@vger.kernel.org # 3.16 Signed-off-by: Hans de Goede Acked-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/uas-detect.h b/drivers/usb/storage/uas-detect.h index 9893d69..63ae161 100644 --- a/drivers/usb/storage/uas-detect.h +++ b/drivers/usb/storage/uas-detect.h @@ -51,7 +51,8 @@ static int uas_find_endpoints(struct usb_host_interface *alt, } static int uas_use_uas_driver(struct usb_interface *intf, - const struct usb_device_id *id) + const struct usb_device_id *id, + unsigned long *flags_ret) { struct usb_host_endpoint *eps[4] = { }; struct usb_device *udev = interface_to_usbdev(intf); @@ -132,5 +133,8 @@ static int uas_use_uas_driver(struct usb_interface *intf, return 0; } + if (flags_ret) + *flags_ret = flags; + return 1; } diff --git a/drivers/usb/storage/uas.c b/drivers/usb/storage/uas.c index 6cdabdc..c6109c1 100644 --- a/drivers/usb/storage/uas.c +++ b/drivers/usb/storage/uas.c @@ -887,8 +887,9 @@ static int uas_probe(struct usb_interface *intf, const struct usb_device_id *id) struct Scsi_Host *shost = NULL; struct uas_dev_info *devinfo; struct usb_device *udev = interface_to_usbdev(intf); + unsigned long dev_flags; - if (!uas_use_uas_driver(intf, id)) + if (!uas_use_uas_driver(intf, id, &dev_flags)) return -ENODEV; if (uas_switch_interface(udev, intf)) @@ -910,8 +911,7 @@ static int uas_probe(struct usb_interface *intf, const struct usb_device_id *id) devinfo->udev = udev; devinfo->resetting = 0; devinfo->shutdown = 0; - devinfo->flags = id->driver_info; - usb_stor_adjust_quirks(udev, &devinfo->flags); + devinfo->flags = dev_flags; init_usb_anchor(&devinfo->cmd_urbs); init_usb_anchor(&devinfo->sense_urbs); init_usb_anchor(&devinfo->data_urbs); diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c index 5600c33..db6f6b5 100644 --- a/drivers/usb/storage/usb.c +++ b/drivers/usb/storage/usb.c @@ -1080,7 +1080,7 @@ static int storage_probe(struct usb_interface *intf, /* If uas is enabled and this device can do uas then ignore it. */ #if IS_ENABLED(CONFIG_USB_UAS) - if (uas_use_uas_driver(intf, id)) + if (uas_use_uas_driver(intf, id, NULL)) return -ENXIO; #endif -- cgit v0.10.2 From ee136af4a064c2f61e2025873584d2c7ec93f4ae Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Tue, 21 Apr 2015 11:20:31 +0200 Subject: uas: Add US_FL_MAX_SECTORS_240 flag The usb-storage driver sets max_sectors = 240 in its scsi-host template, for uas we do not want to do that for all devices, but testing has shown that some devices need it. This commit adds a US_FL_MAX_SECTORS_240 flag for such devices, and implements support for it in uas.c, while at it it also adds support for US_FL_MAX_SECTORS_64 to uas.c. Cc: stable@vger.kernel.org # 3.16 Signed-off-by: Hans de Goede Acked-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index f6befa9..61ab162 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -3787,6 +3787,8 @@ bytes respectively. Such letter suffixes can also be entirely omitted. READ_CAPACITY_16 command); f = NO_REPORT_OPCODES (don't use report opcodes command, uas only); + g = MAX_SECTORS_240 (don't transfer more than + 240 sectors at a time, uas only); h = CAPACITY_HEURISTICS (decrease the reported device capacity by one sector if the number is odd); diff --git a/drivers/usb/storage/uas.c b/drivers/usb/storage/uas.c index c6109c1..6d3122a 100644 --- a/drivers/usb/storage/uas.c +++ b/drivers/usb/storage/uas.c @@ -759,7 +759,10 @@ static int uas_eh_bus_reset_handler(struct scsi_cmnd *cmnd) static int uas_slave_alloc(struct scsi_device *sdev) { - sdev->hostdata = (void *)sdev->host->hostdata; + struct uas_dev_info *devinfo = + (struct uas_dev_info *)sdev->host->hostdata; + + sdev->hostdata = devinfo; /* USB has unusual DMA-alignment requirements: Although the * starting address of each scatter-gather element doesn't matter, @@ -778,6 +781,11 @@ static int uas_slave_alloc(struct scsi_device *sdev) */ blk_queue_update_dma_alignment(sdev->request_queue, (512 - 1)); + if (devinfo->flags & US_FL_MAX_SECTORS_64) + blk_queue_max_hw_sectors(sdev->request_queue, 64); + else if (devinfo->flags & US_FL_MAX_SECTORS_240) + blk_queue_max_hw_sectors(sdev->request_queue, 240); + return 0; } diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c index db6f6b5..6c10c88 100644 --- a/drivers/usb/storage/usb.c +++ b/drivers/usb/storage/usb.c @@ -479,7 +479,8 @@ void usb_stor_adjust_quirks(struct usb_device *udev, unsigned long *fflags) US_FL_SINGLE_LUN | US_FL_NO_WP_DETECT | US_FL_NO_READ_DISC_INFO | US_FL_NO_READ_CAPACITY_16 | US_FL_INITIAL_READ10 | US_FL_WRITE_CACHE | - US_FL_NO_ATA_1X | US_FL_NO_REPORT_OPCODES); + US_FL_NO_ATA_1X | US_FL_NO_REPORT_OPCODES | + US_FL_MAX_SECTORS_240); p = quirks; while (*p) { @@ -520,6 +521,9 @@ void usb_stor_adjust_quirks(struct usb_device *udev, unsigned long *fflags) case 'f': f |= US_FL_NO_REPORT_OPCODES; break; + case 'g': + f |= US_FL_MAX_SECTORS_240; + break; case 'h': f |= US_FL_CAPACITY_HEURISTICS; break; diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h index a7f2604..7f5f78b 100644 --- a/include/linux/usb_usual.h +++ b/include/linux/usb_usual.h @@ -77,6 +77,8 @@ /* Cannot handle ATA_12 or ATA_16 CDBs */ \ US_FLAG(NO_REPORT_OPCODES, 0x04000000) \ /* Cannot handle MI_REPORT_SUPPORTED_OPERATION_CODES */ \ + US_FLAG(MAX_SECTORS_240, 0x08000000) \ + /* Sets max_sectors to 240 */ \ #define US_FLAG(name, value) US_FL_##name = value , enum { US_DO_ALL_FLAGS }; -- cgit v0.10.2 From 8e779c6c4a398763c21371fe40f649206041dc1e Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Tue, 21 Apr 2015 11:20:32 +0200 Subject: uas: Set max_sectors_240 quirk for ASM1053 devices Testing has shown that ASM1053 devices do not work properly with transfers larger than 240 sectors, so set max_sectors to 240 on these. Cc: stable@vger.kernel.org # 3.16 Reported-by: Steve Bangert Signed-off-by: Hans de Goede Tested-by: Steve Bangert Acked-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/uas-detect.h b/drivers/usb/storage/uas-detect.h index 63ae161..f58caa9 100644 --- a/drivers/usb/storage/uas-detect.h +++ b/drivers/usb/storage/uas-detect.h @@ -74,7 +74,7 @@ static int uas_use_uas_driver(struct usb_interface *intf, * this writing the following versions exist: * ASM1051 - no uas support version * ASM1051 - with broken (*) uas support - * ASM1053 - with working uas support + * ASM1053 - with working uas support, but problems with large xfers * ASM1153 - with working uas support * * Devices with these chips re-use a number of device-ids over the @@ -104,6 +104,9 @@ static int uas_use_uas_driver(struct usb_interface *intf, } else if (usb_ss_max_streams(&eps[1]->ss_ep_comp) == 32) { /* Possibly an ASM1051, disable uas */ flags |= US_FL_IGNORE_UAS; + } else { + /* ASM1053, these have issues with large transfers */ + flags |= US_FL_MAX_SECTORS_240; } } -- cgit v0.10.2 From a5a356cee89f86ff86cc3ce24136ca1f802c1bf1 Mon Sep 17 00:00:00 2001 From: Li Jun Date: Sun, 12 Apr 2015 17:51:02 +0800 Subject: usb: chipidea: otg: remove mutex unlock and lock while stop and start role Wrongly release mutex lock during otg_statemachine may result in re-enter otg_statemachine, which is not allowed, we should do next state transtition after previous one completed. Fixes: 826cfe751f3e ("usb: chipidea: add OTG fsm operation functions implementation") Cc: # v3.16+ Signed-off-by: Li Jun Signed-off-by: Peter Chen Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/chipidea/otg_fsm.c b/drivers/usb/chipidea/otg_fsm.c index 083acf4..19d655a 100644 --- a/drivers/usb/chipidea/otg_fsm.c +++ b/drivers/usb/chipidea/otg_fsm.c @@ -520,7 +520,6 @@ static int ci_otg_start_host(struct otg_fsm *fsm, int on) { struct ci_hdrc *ci = container_of(fsm, struct ci_hdrc, fsm); - mutex_unlock(&fsm->lock); if (on) { ci_role_stop(ci); ci_role_start(ci, CI_ROLE_HOST); @@ -529,7 +528,6 @@ static int ci_otg_start_host(struct otg_fsm *fsm, int on) hw_device_reset(ci); ci_role_start(ci, CI_ROLE_GADGET); } - mutex_lock(&fsm->lock); return 0; } @@ -537,12 +535,10 @@ static int ci_otg_start_gadget(struct otg_fsm *fsm, int on) { struct ci_hdrc *ci = container_of(fsm, struct ci_hdrc, fsm); - mutex_unlock(&fsm->lock); if (on) usb_gadget_vbus_connect(&ci->gadget); else usb_gadget_vbus_disconnect(&ci->gadget); - mutex_lock(&fsm->lock); return 0; } -- cgit v0.10.2 From bb304b71f8dbcb284f0f876dfb4eecbadf079773 Mon Sep 17 00:00:00 2001 From: "Ivan T. Ivanov" Date: Mon, 27 Apr 2015 15:27:36 +0300 Subject: Revert "usb: host: ehci-msm: Use devm_ioremap_resource instead of devm_ioremap" This reverts commit 70843f623b58 ("usb: host: ehci-msm: Use devm_ioremap_resource instead of devm_ioremap") and commit e507bf577e5a ("host: ehci-msm: remove duplicate check on resource"), because msm_otg and this driver are using same address space to access AHB mode and USB command registers. Signed-off-by: Ivan T. Ivanov Acked-by: Alan Stern Acked-by: Vivek Gautam Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ehci-msm.c b/drivers/usb/host/ehci-msm.c index 9db74ca..275c92e 100644 --- a/drivers/usb/host/ehci-msm.c +++ b/drivers/usb/host/ehci-msm.c @@ -88,13 +88,20 @@ static int ehci_msm_probe(struct platform_device *pdev) } res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - hcd->regs = devm_ioremap_resource(&pdev->dev, res); - if (IS_ERR(hcd->regs)) { - ret = PTR_ERR(hcd->regs); + if (!res) { + dev_err(&pdev->dev, "Unable to get memory resource\n"); + ret = -ENODEV; goto put_hcd; } + hcd->rsrc_start = res->start; hcd->rsrc_len = resource_size(res); + hcd->regs = devm_ioremap(&pdev->dev, hcd->rsrc_start, hcd->rsrc_len); + if (!hcd->regs) { + dev_err(&pdev->dev, "ioremap failed\n"); + ret = -ENOMEM; + goto put_hcd; + } /* * OTG driver takes care of PHY initialization, clock management, -- cgit v0.10.2 From 0d3bba0287d4e284c3ec7d3397e81eec920d5e7e Mon Sep 17 00:00:00 2001 From: Quentin Casasnovas Date: Tue, 14 Apr 2015 11:25:43 +0200 Subject: cdc-acm: prevent infinite loop when parsing CDC headers. Phil and I found out a problem with commit: 7e860a6e7aa6 ("cdc-acm: add sanity checks") It added some sanity checks to ignore potential garbage in CDC headers but also introduced a potential infinite loop. This can happen at the first loop iteration (elength = 0 in that case) if the description isn't a DT_CS_INTERFACE or later if 'buffer[0]' is zero. It should also be noted that the wrong length was being added to 'buffer' in case 'buffer[1]' was not a DT_CS_INTERFACE descriptor, since elength was assigned after that check in the loop. A specially crafted USB device could be used to trigger this infinite loop. Fixes: 7e860a6e7aa6 ("cdc-acm: add sanity checks") Signed-off-by: Phil Turnbull Signed-off-by: Quentin Casasnovas CC: Sergei Shtylyov CC: Oliver Neukum CC: Adam Lee CC: Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 3e15add..5c8f581 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1142,11 +1142,16 @@ static int acm_probe(struct usb_interface *intf, } while (buflen > 0) { + elength = buffer[0]; + if (!elength) { + dev_err(&intf->dev, "skipping garbage byte\n"); + elength = 1; + goto next_desc; + } if (buffer[1] != USB_DT_CS_INTERFACE) { dev_err(&intf->dev, "skipping garbage\n"); goto next_desc; } - elength = buffer[0]; switch (buffer[2]) { case USB_CDC_UNION_TYPE: /* we've found it */ -- cgit v0.10.2 From fefad2d54beb8aad6bf4ac6daeb74f86f52565de Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Sun, 12 Apr 2015 09:09:05 -0300 Subject: [media] v4l: omap4iss: Replace outdated OMAP4 control pad API with syscon The omap4_ctrl_pad_readl and omap4_ctrl_pad_writel functions have been removed by commit efde234674d9 but are still used by the OMAP4 ISS driver, resulting in a compilation breakage: drivers/staging/media/omap4iss/iss_csiphy.c: In function 'omap4iss_csiphy_config': drivers/staging/media/omap4iss/iss_csiphy.c:167:2: error: implicit declaration of function 'omap4_ctrl_pad_writel' [-Werror=implicit-function-declaration] omap4_ctrl_pad_writel(cam_rx_ctrl, Fix the problem by using the syscon API to reaplace the control pad API. Lookup the syscon instance by compatible name for now as the OMAP4 ISS driver doesn't support DT yet. Fixes: efde234674d9 ("ARM: OMAP4+: control: remove support for legacy pad read/write") Signed-off-by: Laurent Pinchart Acked-by: Sakari Alius Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/staging/media/omap4iss/Kconfig b/drivers/staging/media/omap4iss/Kconfig index b78643f..072dac0 100644 --- a/drivers/staging/media/omap4iss/Kconfig +++ b/drivers/staging/media/omap4iss/Kconfig @@ -2,6 +2,7 @@ config VIDEO_OMAP4 bool "OMAP 4 Camera support" depends on VIDEO_V4L2=y && VIDEO_V4L2_SUBDEV_API && I2C=y && ARCH_OMAP4 depends on HAS_DMA + select MFD_SYSCON select VIDEOBUF2_DMA_CONTIG ---help--- Driver for an OMAP 4 ISS controller. diff --git a/drivers/staging/media/omap4iss/iss.c b/drivers/staging/media/omap4iss/iss.c index e0ad5e5..7ced940 100644 --- a/drivers/staging/media/omap4iss/iss.c +++ b/drivers/staging/media/omap4iss/iss.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -1386,6 +1387,16 @@ static int iss_probe(struct platform_device *pdev) platform_set_drvdata(pdev, iss); + /* + * TODO: When implementing DT support switch to syscon regmap lookup by + * phandle. + */ + iss->syscon = syscon_regmap_lookup_by_compatible("syscon"); + if (IS_ERR(iss->syscon)) { + ret = PTR_ERR(iss->syscon); + goto error; + } + /* Clocks */ ret = iss_map_mem_resource(pdev, iss, OMAP4_ISS_MEM_TOP); if (ret < 0) diff --git a/drivers/staging/media/omap4iss/iss.h b/drivers/staging/media/omap4iss/iss.h index 734cfee..35df8b4 100644 --- a/drivers/staging/media/omap4iss/iss.h +++ b/drivers/staging/media/omap4iss/iss.h @@ -29,6 +29,8 @@ #include "iss_ipipe.h" #include "iss_resizer.h" +struct regmap; + #define to_iss_device(ptr_module) \ container_of(ptr_module, struct iss_device, ptr_module) #define to_device(ptr_module) \ @@ -79,6 +81,7 @@ struct iss_reg { /* * struct iss_device - ISS device structure. + * @syscon: Regmap for the syscon register space * @crashed: Bitmask of crashed entities (indexed by entity ID) */ struct iss_device { @@ -93,6 +96,7 @@ struct iss_device { struct resource *res[OMAP4_ISS_MEM_LAST]; void __iomem *regs[OMAP4_ISS_MEM_LAST]; + struct regmap *syscon; u64 raw_dmamask; diff --git a/drivers/staging/media/omap4iss/iss_csiphy.c b/drivers/staging/media/omap4iss/iss_csiphy.c index 7c3d55d..748607f 100644 --- a/drivers/staging/media/omap4iss/iss_csiphy.c +++ b/drivers/staging/media/omap4iss/iss_csiphy.c @@ -13,6 +13,7 @@ #include #include +#include #include "../../../../arch/arm/mach-omap2/control.h" @@ -140,9 +141,11 @@ int omap4iss_csiphy_config(struct iss_device *iss, * - bit [18] : CSIPHY1 CTRLCLK enable * - bit [17:16] : CSIPHY1 config: 00 d-phy, 01/10 ccp2 */ - cam_rx_ctrl = omap4_ctrl_pad_readl( - OMAP4_CTRL_MODULE_PAD_CORE_CONTROL_CAMERA_RX); - + /* + * TODO: When implementing DT support specify the CONTROL_CAMERA_RX + * register offset in the syscon property instead of hardcoding it. + */ + regmap_read(iss->syscon, 0x68, &cam_rx_ctrl); if (subdevs->interface == ISS_INTERFACE_CSI2A_PHY1) { cam_rx_ctrl &= ~(OMAP4_CAMERARX_CSI21_LANEENABLE_MASK | @@ -166,8 +169,7 @@ int omap4iss_csiphy_config(struct iss_device *iss, cam_rx_ctrl |= OMAP4_CAMERARX_CSI22_CTRLCLKEN_MASK; } - omap4_ctrl_pad_writel(cam_rx_ctrl, - OMAP4_CTRL_MODULE_PAD_CORE_CONTROL_CAMERA_RX); + regmap_write(iss->syscon, 0x68, cam_rx_ctrl); /* Reset used lane count */ csi2->phy->used_data_lanes = 0; -- cgit v0.10.2 From b00f5c2dc01450bed9fed1a41a637fa917e03c5c Mon Sep 17 00:00:00 2001 From: Frederic Danis Date: Fri, 10 Apr 2015 15:13:05 +0200 Subject: tty: Re-add external interface for tty_set_termios() This is needed by Bluetooth hci_uart module to be able to change speed of Bluetooth controller and local UART. Signed-off-by: Frederic Danis Reviewed-by: Peter Hurley Cc: Marcel Holtmann Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/tty_ioctl.c b/drivers/tty/tty_ioctl.c index 632fc81..8e53fe4 100644 --- a/drivers/tty/tty_ioctl.c +++ b/drivers/tty/tty_ioctl.c @@ -536,7 +536,7 @@ EXPORT_SYMBOL(tty_termios_hw_change); * Locking: termios_rwsem */ -static int tty_set_termios(struct tty_struct *tty, struct ktermios *new_termios) +int tty_set_termios(struct tty_struct *tty, struct ktermios *new_termios) { struct ktermios old_termios; struct tty_ldisc *ld; @@ -569,6 +569,7 @@ static int tty_set_termios(struct tty_struct *tty, struct ktermios *new_termios) up_write(&tty->termios_rwsem); return 0; } +EXPORT_SYMBOL_GPL(tty_set_termios); /** * set_termios - set termios values for a tty diff --git a/include/linux/tty.h b/include/linux/tty.h index 358a337..fe5623c 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -491,6 +491,7 @@ static inline speed_t tty_get_baud_rate(struct tty_struct *tty) extern void tty_termios_copy_hw(struct ktermios *new, struct ktermios *old); extern int tty_termios_hw_change(struct ktermios *a, struct ktermios *b); +extern int tty_set_termios(struct tty_struct *tty, struct ktermios *kt); extern struct tty_ldisc *tty_ldisc_ref(struct tty_struct *); extern void tty_ldisc_deref(struct tty_ldisc *); -- cgit v0.10.2 From 10afbe346bba125e2eedfd1a89b7bd5c92900859 Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Sat, 11 Apr 2015 10:05:07 -0400 Subject: serial: core: Fix kernel-doc build warnings Fix uart_console_write() kernel-doc build warnings. Warning(drivers/tty/serial/serial_core.c:1778): No description found for parameter 'putchar' Warning(drivers/tty/serial/serial_core.c:1778): Excess function parameter 'write' description in 'uart_console_write' Fixes: 1cfe42b7fd29 ("serial: core: Fix kernel doc for uart_console_write()") Reported-by: Fengguang Wu Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c index eb5b03b..0b7bb12 100644 --- a/drivers/tty/serial/serial_core.c +++ b/drivers/tty/serial/serial_core.c @@ -1770,7 +1770,7 @@ static const struct file_operations uart_proc_fops = { * @port: the port to write the message * @s: array of characters * @count: number of characters in string to write - * @write: function to write character to port + * @putchar: function to write character to port */ void uart_console_write(struct uart_port *port, const char *s, unsigned int count, -- cgit v0.10.2 From 5c90c07b98c02198d9777a7c4f3047b0a94bf7ed Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Mon, 13 Apr 2015 16:34:21 +0200 Subject: serial: xilinx: Use platform_get_irq to get irq description structure For systems with CONFIG_SERIAL_OF_PLATFORM=y and device_type = "serial"; property in DT of_serial.c driver maps and unmaps IRQ (because driver probe fails). Then a driver is called but irq mapping is not created that's why driver is failing again in again on request_irq(). Based on this use platform_get_irq() instead of platform_get_resource() which is doing irq_desc allocation and driver itself can request IRQ. Fix both xilinx serial drivers in the tree. Signed-off-by: Michal Simek CC: Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/uartlite.c b/drivers/tty/serial/uartlite.c index 708eead..b1c6bd3 100644 --- a/drivers/tty/serial/uartlite.c +++ b/drivers/tty/serial/uartlite.c @@ -632,7 +632,8 @@ MODULE_DEVICE_TABLE(of, ulite_of_match); static int ulite_probe(struct platform_device *pdev) { - struct resource *res, *res2; + struct resource *res; + int irq; int id = pdev->id; #ifdef CONFIG_OF const __be32 *prop; @@ -646,11 +647,11 @@ static int ulite_probe(struct platform_device *pdev) if (!res) return -ENODEV; - res2 = platform_get_resource(pdev, IORESOURCE_IRQ, 0); - if (!res2) - return -ENODEV; + irq = platform_get_irq(pdev, 0); + if (irq <= 0) + return -ENXIO; - return ulite_assign(&pdev->dev, id, res->start, res2->start); + return ulite_assign(&pdev->dev, id, res->start, irq); } static int ulite_remove(struct platform_device *pdev) diff --git a/drivers/tty/serial/xilinx_uartps.c b/drivers/tty/serial/xilinx_uartps.c index f218ec6..3ddbac7 100644 --- a/drivers/tty/serial/xilinx_uartps.c +++ b/drivers/tty/serial/xilinx_uartps.c @@ -1331,9 +1331,9 @@ static SIMPLE_DEV_PM_OPS(cdns_uart_dev_pm_ops, cdns_uart_suspend, */ static int cdns_uart_probe(struct platform_device *pdev) { - int rc, id; + int rc, id, irq; struct uart_port *port; - struct resource *res, *res2; + struct resource *res; struct cdns_uart *cdns_uart_data; cdns_uart_data = devm_kzalloc(&pdev->dev, sizeof(*cdns_uart_data), @@ -1380,9 +1380,9 @@ static int cdns_uart_probe(struct platform_device *pdev) goto err_out_clk_disable; } - res2 = platform_get_resource(pdev, IORESOURCE_IRQ, 0); - if (!res2) { - rc = -ENODEV; + irq = platform_get_irq(pdev, 0); + if (irq <= 0) { + rc = -ENXIO; goto err_out_clk_disable; } @@ -1411,7 +1411,7 @@ static int cdns_uart_probe(struct platform_device *pdev) * and triggers invocation of the config_port() entry point. */ port->mapbase = res->start; - port->irq = res2->start; + port->irq = irq; port->dev = &pdev->dev; port->uartclk = clk_get_rate(cdns_uart_data->uartclk); port->private_data = cdns_uart_data; -- cgit v0.10.2 From 6befa9d883385c580369a2cc9e53fbf329771f6d Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Tue, 14 Apr 2015 12:03:09 +0200 Subject: serial: of-serial: Remove device_type = "serial" registration Do not probe all serial drivers by of_serial.c which are using device_type = "serial"; property. Only drivers which have valid compatible strings listed in the driver should be probed. When PORT_UNKNOWN is setup probe will fail anyway. Arnd quotation about driver historical background: "when I wrote that driver initially, the idea was that it would get used as a stub to hook up all other serial drivers but after that, the common code learned to create platform devices from DT" This patch fix the problem with on the system with xilinx_uartps and 16550a where of_serial failed to register for xilinx_uartps and because of irq_dispose_mapping() removed irq_desc. Then when xilinx_uartps was asking for irq with request_irq() EINVAL is returned. Signed-off-by: Michal Simek CC: Acked-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/of_serial.c b/drivers/tty/serial/of_serial.c index 5b73afb..137381e 100644 --- a/drivers/tty/serial/of_serial.c +++ b/drivers/tty/serial/of_serial.c @@ -346,7 +346,6 @@ static const struct of_device_id of_platform_serial_table[] = { { .compatible = "ibm,qpace-nwp-serial", .data = (void *)PORT_NWPSERIAL, }, #endif - { .type = "serial", .data = (void *)PORT_UNKNOWN, }, { /* end of list */ }, }; -- cgit v0.10.2 From a8d4e01637902311c5643b69a5c80e2805f04054 Mon Sep 17 00:00:00 2001 From: Ludovic Desroches Date: Thu, 16 Apr 2015 16:58:12 +0200 Subject: tty/serial: at91: maxburst was missing for dma transfers Maxburst was not set when doing the dma slave configuration. This value is checked by the recently introduced xdmac. It causes an error when doing the slave configuration and so prevents from using dma. Signed-off-by: Ludovic Desroches Cc: # 3.12 and later Acked-by: Nicolas Ferre Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c index d58fe47..27dade2 100644 --- a/drivers/tty/serial/atmel_serial.c +++ b/drivers/tty/serial/atmel_serial.c @@ -880,6 +880,7 @@ static int atmel_prepare_tx_dma(struct uart_port *port) config.direction = DMA_MEM_TO_DEV; config.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE; config.dst_addr = port->mapbase + ATMEL_US_THR; + config.dst_maxburst = 1; ret = dmaengine_slave_config(atmel_port->chan_tx, &config); @@ -1059,6 +1060,7 @@ static int atmel_prepare_rx_dma(struct uart_port *port) config.direction = DMA_DEV_TO_MEM; config.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE; config.src_addr = port->mapbase + ATMEL_US_RHR; + config.src_maxburst = 1; ret = dmaengine_slave_config(atmel_port->chan_rx, &config); -- cgit v0.10.2 From 31c6ba97d1671b0a853197fe54a05e0b07ce2631 Mon Sep 17 00:00:00 2001 From: Robert Baldyga Date: Fri, 17 Apr 2015 08:43:09 +0200 Subject: serial: samsung: fix serial console break This patch fixes problems with serial console break. When function s3c64xx_serial_startup() was started while serial console has been working, it caused lose of characters written to TX FIFO. This effect was particularly observable with systemd, which closes serial port every time when it's not currently needed, hence function s3c64xx_serial_startup() is called quite often there. To fix this problem we avoid resetting TX FIFO if port is used as serial console. Example of broken console log: [ 1086.7 Expecting device dev-ttySAC1.device... [ 1086.[ OK ] Reached target Paths. [ 1086.756416] s[ OK ] Reached target Swap. [ 1086.776413] systemd[1]: Reached target Swap. [ 1086.776642] systemd[1]: Starting Root Slice. [ 5.53403[ OK ] Created slice Root Slice. [ 5.548433] systemd[1]: Create[ OK ] Created slice User and Session Slice. [ 5.568414] sys[ OK ] Listening on /dev/initctl Compatibility Named Pipe. [ 5.588388] s[ OK ] Listening on Delayed Shutdown Socket. [ 5.608376] sy[ OK ] Listening on Journal Socket (/dev/log). [ 5.628361] [ OK ] Listening on udev Kernel Socket. [ 5.648357] s[ OK ] Listening on udev Control Socket. [ 5.668353] s[ OK ] Listening on Journal Socket. [ 5.688366] systemd[1]: Listeni[ OK ] Created slice System Slice. [ 5.708393] Mounting Temporary Directory... [ 7139.067436] Starting prepare device daemon... [ 7139.091726] sy Starting Generate environment from /etc/profile.d... [ 5.792867] system Starting Create Static Device Nodes in /dev... [ 7848.718 Mounting Debug File System... [ 7848.7384 Mounting Configuration File System... [ 5.852 Starting Apply Kernel Variables... [ 5.8720 Starting Setup Virtual Console... [ 7848.798 Starting udev Coldplug all Devices... [ 7848.817 Starting Journal Service... [ OK ] Started Journal Service. [ 7848.854222] s[ OK ] Reached target Slices. Starting Remount Root and Kernel File Systems... [ OK ] Mounted Configuration File System. Reported-by: Chanwoo Choi Signed-off-by: Robert Baldyga Reviewed-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/samsung.c b/drivers/tty/serial/samsung.c index cf08876..a0ae942 100644 --- a/drivers/tty/serial/samsung.c +++ b/drivers/tty/serial/samsung.c @@ -1068,8 +1068,9 @@ static int s3c64xx_serial_startup(struct uart_port *port) spin_lock_irqsave(&port->lock, flags); ufcon = rd_regl(port, S3C2410_UFCON); - ufcon |= S3C2410_UFCON_RESETRX | S3C2410_UFCON_RESETTX | - S5PV210_UFCON_RXTRIG8; + ufcon |= S3C2410_UFCON_RESETRX | S5PV210_UFCON_RXTRIG8; + if (!uart_console(port)) + ufcon |= S3C2410_UFCON_RESETTX; wr_regl(port, S3C2410_UFCON, ufcon); enable_rx_pio(ourport); -- cgit v0.10.2 From 982df6aec06479c844f46f7a5cc960151f8fc005 Mon Sep 17 00:00:00 2001 From: "Ivan T. Ivanov" Date: Thu, 9 Apr 2015 18:18:35 +0300 Subject: pinctrl: qcom-spmi-gpio: Fix output type configuration GPIO output type configuration was incorrectly overwritten by strength value. Fix this. Signed-off-by: Ivan T. Ivanov Acked-by: Bjorn Andersson Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c index b2d2221..4b21aac 100644 --- a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c +++ b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c @@ -417,7 +417,7 @@ static int pmic_gpio_config_set(struct pinctrl_dev *pctldev, unsigned int pin, return ret; val = pad->buffer_type << PMIC_GPIO_REG_OUT_TYPE_SHIFT; - val = pad->strength << PMIC_GPIO_REG_OUT_STRENGTH_SHIFT; + val |= pad->strength << PMIC_GPIO_REG_OUT_STRENGTH_SHIFT; ret = pmic_gpio_write(state, pad, PMIC_GPIO_REG_DIG_OUT_CTL, val); if (ret < 0) -- cgit v0.10.2 From 24a66618d63548e1012bf65ed30a3c031f410ca2 Mon Sep 17 00:00:00 2001 From: "Ivan T. Ivanov" Date: Thu, 9 Apr 2015 18:18:36 +0300 Subject: pinctrl: qcom-spmi-gpio: Fix input value report Read input buffer when input is enabled, not when it is disabled. Also fix interpretation of the pmic_gpio_read() return code, negative value means an error. Signed-off-by: Ivan T. Ivanov Reviewed-by: Bjorn Andersson Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c index 4b21aac..e8b74c6 100644 --- a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c +++ b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c @@ -466,12 +466,13 @@ static void pmic_gpio_config_dbg_show(struct pinctrl_dev *pctldev, seq_puts(s, " ---"); } else { - if (!pad->input_enabled) { + if (pad->input_enabled) { ret = pmic_gpio_read(state, pad, PMIC_MPP_REG_RT_STS); - if (!ret) { - ret &= PMIC_MPP_REG_RT_STS_VAL_MASK; - pad->out_value = ret; - } + if (ret < 0) + return; + + ret &= PMIC_MPP_REG_RT_STS_VAL_MASK; + pad->out_value = ret; } seq_printf(s, " %-4s", pad->output_enabled ? "out" : "in"); -- cgit v0.10.2 From 4e637ac212b63f4b5dd1da626aca34ffcbfd5daa Mon Sep 17 00:00:00 2001 From: "Ivan T. Ivanov" Date: Thu, 9 Apr 2015 18:18:37 +0300 Subject: pinctrl: qcom-spmi-mpp: Fix input value report Fix interpretation of the pmic_mpp_read() return code, negative value means an error. Signed-off-by: Ivan T. Ivanov Reviewed-by: Bjorn Andersson Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c b/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c index 890df16..211b942 100644 --- a/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c +++ b/drivers/pinctrl/qcom/pinctrl-spmi-mpp.c @@ -577,10 +577,11 @@ static void pmic_mpp_config_dbg_show(struct pinctrl_dev *pctldev, if (pad->input_enabled) { ret = pmic_mpp_read(state, pad, PMIC_MPP_REG_RT_STS); - if (!ret) { - ret &= PMIC_MPP_REG_RT_STS_VAL_MASK; - pad->out_value = ret; - } + if (ret < 0) + return; + + ret &= PMIC_MPP_REG_RT_STS_VAL_MASK; + pad->out_value = ret; } seq_printf(s, " %-4s", pad->output_enabled ? "out" : "in"); -- cgit v0.10.2 From a04f90a33fab74789b91fc9739999012f11022d1 Mon Sep 17 00:00:00 2001 From: Deepak S Date: Thu, 16 Apr 2015 08:51:28 +0530 Subject: drm/i915/chv: Implement WaDisableShadowRegForCpd MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This WA is avoid problem between shadow vs wake FIFO unload problem during CPD/RC6 transactions on CHV. v2: Define individual bits GTFIFOCTL (Ville) v3: move WA to uncore_early_sanitize (ville) Signed-off-by: Deepak S Reviewed-by: Ville Syrjälä [Jani: fixed some whitespace issues while applying] Signed-off-by: Jani Nikula diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 3da1af4..773d1d2 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -6074,6 +6074,8 @@ enum skl_disp_power_wells { #define GTFIFOCTL 0x120008 #define GT_FIFO_FREE_ENTRIES_MASK 0x7f #define GT_FIFO_NUM_RESERVED_ENTRIES 20 +#define GT_FIFO_CTL_BLOCK_ALL_POLICY_STALL (1 << 12) +#define GT_FIFO_CTL_RC6_POLICY_STALL (1 << 11) #define HSW_IDICR 0x9008 #define IDIHASHMSK(x) (((x) & 0x3f) << 16) diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index ab5cc94..ff2a746 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -360,6 +360,14 @@ static void __intel_uncore_early_sanitize(struct drm_device *dev, __raw_i915_write32(dev_priv, GTFIFODBG, __raw_i915_read32(dev_priv, GTFIFODBG)); + /* WaDisableShadowRegForCpd:chv */ + if (IS_CHERRYVIEW(dev)) { + __raw_i915_write32(dev_priv, GTFIFOCTL, + __raw_i915_read32(dev_priv, GTFIFOCTL) | + GT_FIFO_CTL_BLOCK_ALL_POLICY_STALL | + GT_FIFO_CTL_RC6_POLICY_STALL); + } + intel_uncore_forcewake_reset(dev, restore_forcewake); } -- cgit v0.10.2 From 9b0f5d63e74a987bf56cc1774baca80a291c9d8d Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Tue, 28 Apr 2015 10:36:45 -0400 Subject: tile: properly use node_isset() on a nodemask_t The code accidentally used cpu_isset() previously in one place (though properly node_isset() elsewhere). Signed-off-by: Chris Metcalf diff --git a/arch/tile/kernel/setup.c b/arch/tile/kernel/setup.c index 6873f00..d366675 100644 --- a/arch/tile/kernel/setup.c +++ b/arch/tile/kernel/setup.c @@ -774,7 +774,7 @@ static void __init zone_sizes_init(void) * though, there'll be no lowmem, so we just alloc_bootmem * the memmap. There will be no percpu memory either. */ - if (i != 0 && cpumask_test_cpu(i, &isolnodes)) { + if (i != 0 && node_isset(i, isolnodes)) { node_memmap_pfn[i] = alloc_bootmem_pfn(0, memmap_size, 0); BUG_ON(node_percpu[i] != 0); -- cgit v0.10.2 From 9d7dd6cd2a1d0d307be6bf783e40e7fda17d6dba Mon Sep 17 00:00:00 2001 From: Rajeev Kumar Date: Tue, 28 Apr 2015 11:30:08 +0530 Subject: ASoC: Update email-id of Rajeev Kumar rajeev-dlh.kumar@st.com email-id doesn't exist anymore as I have left the company. Replace ST's id with Rajeev Kumar Signed-off-by: Rajeev Kumar Signed-off-by: Mark Brown diff --git a/include/sound/designware_i2s.h b/include/sound/designware_i2s.h index 26f406e..3a8fca9 100644 --- a/include/sound/designware_i2s.h +++ b/include/sound/designware_i2s.h @@ -1,5 +1,5 @@ /* - * Copyright (ST) 2012 Rajeev Kumar (rajeev-dlh.kumar@st.com) + * Copyright (ST) 2012 Rajeev Kumar (rajeevkumar.linux@gmail.com) * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by diff --git a/include/sound/spear_dma.h b/include/sound/spear_dma.h index 65aca51..e290de4 100644 --- a/include/sound/spear_dma.h +++ b/include/sound/spear_dma.h @@ -1,7 +1,7 @@ /* * linux/spear_dma.h * -* Copyright (ST) 2012 Rajeev Kumar (rajeev-dlh.kumar@st.com) +* Copyright (ST) 2012 Rajeev Kumar (rajeevkumar.linux@gmail.com) * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by -- cgit v0.10.2 From 96a5d18bc1338786fecac73599f1681f59a59a8e Mon Sep 17 00:00:00 2001 From: Soeren Grunewald Date: Tue, 28 Apr 2015 16:29:49 +0200 Subject: serial: 8250_pci: Add support for 16 port Exar boards The Exar XR17V358 chip usually provides only 8 ports. But two chips can be combined to act as a single 16 port chip. Therefor one chip is configured as master the second as slave by connecting the mode pin to VCC (master) or GND (slave). Then the master chip is reporting a different device-id depending on whether a slave is detected or not. The UARTs 8-15 are addressed from 0x2000-0x3fff. So the offset of 0x400 from UART to UART can be used to address all 16 ports as before. See: https://www.exar.com/common/content/document.ashx?id=1587 page 11 Signed-off-by: Soeren Grunewald Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c index 08da4d3..46bcebb 100644 --- a/drivers/tty/serial/8250/8250_pci.c +++ b/drivers/tty/serial/8250/8250_pci.c @@ -1998,6 +1998,8 @@ pci_wch_ch38x_setup(struct serial_private *priv, #define PCIE_DEVICE_ID_WCH_CH382_2S1P 0x3250 #define PCIE_DEVICE_ID_WCH_CH384_4S 0x3470 +#define PCI_DEVICE_ID_EXAR_XR17V8358 0x8358 + /* Unknown vendors/cards - this should not be in linux/pci_ids.h */ #define PCI_SUBDEVICE_ID_UNKNOWN_0x1584 0x1584 #define PCI_SUBDEVICE_ID_UNKNOWN_0x1588 0x1588 @@ -2520,6 +2522,13 @@ static struct pci_serial_quirk pci_serial_quirks[] __refdata = { .subdevice = PCI_ANY_ID, .setup = pci_xr17v35x_setup, }, + { + .vendor = PCI_VENDOR_ID_EXAR, + .device = PCI_DEVICE_ID_EXAR_XR17V8358, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .setup = pci_xr17v35x_setup, + }, /* * Xircom cards */ @@ -2999,6 +3008,7 @@ enum pci_board_num_t { pbn_exar_XR17V352, pbn_exar_XR17V354, pbn_exar_XR17V358, + pbn_exar_XR17V8358, pbn_exar_ibm_saturn, pbn_pasemi_1682M, pbn_ni8430_2, @@ -3685,6 +3695,14 @@ static struct pciserial_board pci_boards[] = { .reg_shift = 0, .first_offset = 0, }, + [pbn_exar_XR17V8358] = { + .flags = FL_BASE0, + .num_ports = 16, + .base_baud = 7812500, + .uart_offset = 0x400, + .reg_shift = 0, + .first_offset = 0, + }, [pbn_exar_ibm_saturn] = { .flags = FL_BASE0, .num_ports = 1, @@ -5080,7 +5098,7 @@ static struct pci_device_id serial_pci_tbl[] = { 0, 0, pbn_exar_XR17C158 }, /* - * Exar Corp. XR17V35[248] Dual/Quad/Octal PCIe UARTs + * Exar Corp. XR17V[48]35[248] Dual/Quad/Octal/Hexa PCIe UARTs */ { PCI_VENDOR_ID_EXAR, PCI_DEVICE_ID_EXAR_XR17V352, PCI_ANY_ID, PCI_ANY_ID, @@ -5094,7 +5112,10 @@ static struct pci_device_id serial_pci_tbl[] = { PCI_ANY_ID, PCI_ANY_ID, 0, 0, pbn_exar_XR17V358 }, - + { PCI_VENDOR_ID_EXAR, PCI_DEVICE_ID_EXAR_XR17V8358, + PCI_ANY_ID, PCI_ANY_ID, + 0, + 0, pbn_exar_XR17V8358 }, /* * Topic TP560 Data/Fax/Voice 56k modem (reported by Evan Clarke) */ -- cgit v0.10.2 From 9e9d55e69a95f8583283d9f01b04562d1278c95d Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Tue, 28 Apr 2015 16:54:04 +0200 Subject: ACPICA: remove duplicate u8 typedef During commit e252652fb266 ("ACPICA: acpidump: Remove integer types translation protection.") two 'unsigned char' types got converted to 'u8'. The result does not compile with gcc-4.5, it can not cope with duplicate typedefs. Signed-off-by: Olaf Hering Signed-off-by: Rafael J. Wysocki diff --git a/include/acpi/actypes.h b/include/acpi/actypes.h index f5ca0e9..1c3002e 100644 --- a/include/acpi/actypes.h +++ b/include/acpi/actypes.h @@ -124,7 +124,6 @@ #ifndef ACPI_USE_SYSTEM_INTTYPES typedef unsigned char u8; -typedef unsigned char u8; typedef unsigned short u16; typedef short s16; typedef COMPILER_DEPENDENT_UINT64 u64; -- cgit v0.10.2 From 61f8ff693923e4b19748b0e8287b99778f2661c7 Mon Sep 17 00:00:00 2001 From: Chris Bainbridge Date: Wed, 22 Apr 2015 16:40:21 +0100 Subject: ACPI / SBS: Enable battery manager when present Commit 9faf6136ff46 (ACPI / SBS: Disable smart battery manager on Apple) introduced a regression disabling the SBS battery manager. The battery manager should be marked as present when acpi_manager_get_info() returns 0. Fixes: 9faf6136ff46 (ACPI / SBS: Disable smart battery manager on Apple) Signed-off-by: Chris Bainbridge Cc: 3.18+ # 3.18+ Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/sbs.c b/drivers/acpi/sbs.c index cd82762..01504c8 100644 --- a/drivers/acpi/sbs.c +++ b/drivers/acpi/sbs.c @@ -684,7 +684,7 @@ static int acpi_sbs_add(struct acpi_device *device) if (!sbs_manager_broken) { result = acpi_manager_get_info(sbs); if (!result) { - sbs->manager_present = 0; + sbs->manager_present = 1; for (id = 0; id < MAX_SBS_BAT; ++id) if ((sbs->batteries_supported & (1 << id))) acpi_battery_add(sbs, id); -- cgit v0.10.2 From 433c5c20c505fef92be84c6afab70f1c2ab5eda3 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 28 Apr 2015 10:34:35 +1000 Subject: powerpc/kvm: Fix SMP=n build error in book3s_xics.c Commit 34cb7954c0aa "Convert ICS mutex lock to spin lock" added an include of asm/spinlock.h, which does not work in the SMP=n case. It should instead include linux/spinlock.h Fixes: 34cb7954c0aa ("KVM: PPC: Book3S HV: Convert ICS mutex lock to spin lock") Acked-by: Paul Mackerras Reviewed-by: Alexander Graf Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/kvm/book3s_xics.c b/arch/powerpc/kvm/book3s_xics.c index 8f3e6cc..c6ca7db 100644 --- a/arch/powerpc/kvm/book3s_xics.c +++ b/arch/powerpc/kvm/book3s_xics.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -20,7 +21,6 @@ #include #include #include -#include #include #include -- cgit v0.10.2 From d33047fd7e7d93662622888681861ba84d43c506 Mon Sep 17 00:00:00 2001 From: Daniel Axtens Date: Mon, 27 Apr 2015 12:33:51 +1000 Subject: powerpc/powernv: Fix early pci_controller_ops loading. Load the PowerNV platform pci controller ops into pci controllers after all the operations are loaded into the platform ops struct, not before. Otherwise we aren't actually setting the ops properly which can break IO for some devices. Fixes: 65ebf4b63 ("powerpc/powernv: Move controller ops from ppc_md to controller_ops") Reported-by: Gavin Shan Reviewed-by: Gavin Shan Signed-off-by: Daniel Axtens Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 920c252..f8bc950 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -2693,7 +2693,6 @@ static void __init pnv_pci_init_ioda_phb(struct device_node *np, hose->last_busno = 0xff; } hose->private_data = phb; - hose->controller_ops = pnv_pci_controller_ops; phb->hub_id = hub_id; phb->opal_id = phb_id; phb->type = ioda_type; @@ -2812,6 +2811,7 @@ static void __init pnv_pci_init_ioda_phb(struct device_node *np, pnv_pci_controller_ops.enable_device_hook = pnv_pci_enable_device_hook; pnv_pci_controller_ops.window_alignment = pnv_pci_window_alignment; pnv_pci_controller_ops.reset_secondary_bus = pnv_pci_reset_secondary_bus; + hose->controller_ops = pnv_pci_controller_ops; #ifdef CONFIG_PCI_IOV ppc_md.pcibios_fixup_sriov = pnv_pci_ioda_fixup_iov_resources; -- cgit v0.10.2 From 63f89caad0e32dcfa17b2d17919816253de48996 Mon Sep 17 00:00:00 2001 From: Christopher Freeman Date: Wed, 4 Mar 2015 01:16:58 -0800 Subject: dmaengine: increment privatecnt when using dma_get_any_slave_channel Channels allocated via dma_get_any_slave_channel were not increasing the counter tracking private allocations. When these channels were released, privatecnt may erroneously fall to zero. The DMA device would then lose its DMA_PRIVATE cap and fail to allocate future private channels (via private_candidate) as any allocations still outstanding would incorrectly be seen as public allocations. Signed-off-by: Christopher Freeman Signed-off-by: Vinod Koul diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 0e035a8..2890d74 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -571,11 +571,15 @@ struct dma_chan *dma_get_any_slave_channel(struct dma_device *device) chan = private_candidate(&mask, device, NULL, NULL); if (chan) { + dma_cap_set(DMA_PRIVATE, device->cap_mask); + device->privatecnt++; err = dma_chan_get(chan); if (err) { pr_debug("%s: failed to get %s: (%d)\n", __func__, dma_chan_name(chan), err); chan = NULL; + if (--device->privatecnt == 0) + dma_cap_clear(DMA_PRIVATE, device->cap_mask); } } -- cgit v0.10.2 From 60a8d62b8497c23eb3d48149af7e55dac2dd83a2 Mon Sep 17 00:00:00 2001 From: Bard Liao Date: Tue, 28 Apr 2015 11:27:39 +0800 Subject: ASoC: rt5677: fixed wrong DMIC ref clock DMIC clock source is not from codec system clock directly. it is generated from the division of system clock. And it should be 256 * sample rate of AIF1. Signed-off-by: Bard Liao Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/codecs/rt5677.c b/sound/soc/codecs/rt5677.c index c6d4e8f..84d162d 100644 --- a/sound/soc/codecs/rt5677.c +++ b/sound/soc/codecs/rt5677.c @@ -904,7 +904,7 @@ static int set_dmic_clk(struct snd_soc_dapm_widget *w, { struct snd_soc_codec *codec = snd_soc_dapm_to_codec(w->dapm); struct rt5677_priv *rt5677 = snd_soc_codec_get_drvdata(codec); - int idx = rl6231_calc_dmic_clk(rt5677->sysclk); + int idx = rl6231_calc_dmic_clk(rt5677->lrck[RT5677_AIF1] << 8); if (idx < 0) dev_err(codec->dev, "Failed to set DMIC clock\n"); -- cgit v0.10.2 From df8d9eeadd0f7a216f2476351d5aee43c6550bf0 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 29 Apr 2015 15:19:21 +0200 Subject: cpuidle: Run tick_broadcast_exit() with disabled interrupts Commit 335f49196fd6 (sched/idle: Use explicit broadcast oneshot control function) replaced clockevents_notify() invocations in cpuidle_idle_call() with direct calls to tick_broadcast_enter() and tick_broadcast_exit(), but it overlooked the fact that interrupts were already enabled before calling the latter which led to functional breakage on systems using idle states with the CPUIDLE_FLAG_TIMER_STOP flag set. Fix that by moving the invocations of tick_broadcast_enter() and tick_broadcast_exit() down into cpuidle_enter_state() where interrupts are still disabled when tick_broadcast_exit() is called. Also ensure that interrupts will be disabled before running tick_broadcast_exit() even if they have been enabled by the idle state's ->enter callback. Trigger a WARN_ON_ONCE() in that case, as we generally don't want that to happen for states with CPUIDLE_FLAG_TIMER_STOP set. Fixes: 335f49196fd6 (sched/idle: Use explicit broadcast oneshot control function) Reported-and-tested-by: Linus Walleij Acked-by: Peter Zijlstra (Intel) Acked-by: Daniel Lezcano Reported-and-tested-by: Sudeep Holla Signed-off-by: Rafael J. Wysocki diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 7a73a27..61c417b 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -158,9 +158,18 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv, int entered_state; struct cpuidle_state *target_state = &drv->states[index]; + bool broadcast = !!(target_state->flags & CPUIDLE_FLAG_TIMER_STOP); ktime_t time_start, time_end; s64 diff; + /* + * Tell the time framework to switch to a broadcast timer because our + * local timer will be shut down. If a local timer is used from another + * CPU as a broadcast timer, this call may fail if it is not available. + */ + if (broadcast && tick_broadcast_enter()) + return -EBUSY; + trace_cpu_idle_rcuidle(index, dev->cpu); time_start = ktime_get(); @@ -169,6 +178,13 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv, time_end = ktime_get(); trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, dev->cpu); + if (broadcast) { + if (WARN_ON_ONCE(!irqs_disabled())) + local_irq_disable(); + + tick_broadcast_exit(); + } + if (!cpuidle_state_is_coupled(dev, drv, entered_state)) local_irq_enable(); diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index deef1ca..fefcb1f 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -81,7 +81,6 @@ static void cpuidle_idle_call(void) struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices); struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev); int next_state, entered_state; - unsigned int broadcast; bool reflect; /* @@ -150,17 +149,6 @@ static void cpuidle_idle_call(void) goto exit_idle; } - broadcast = drv->states[next_state].flags & CPUIDLE_FLAG_TIMER_STOP; - - /* - * Tell the time framework to switch to a broadcast timer - * because our local timer will be shutdown. If a local timer - * is used from another cpu as a broadcast timer, this call may - * fail if it is not available - */ - if (broadcast && tick_broadcast_enter()) - goto use_default; - /* Take note of the planned idle state. */ idle_set_state(this_rq(), &drv->states[next_state]); @@ -174,8 +162,8 @@ static void cpuidle_idle_call(void) /* The cpu is no longer idle or about to enter idle. */ idle_set_state(this_rq(), NULL); - if (broadcast) - tick_broadcast_exit(); + if (entered_state == -EBUSY) + goto use_default; /* * Give the governor an opportunity to reflect on the outcome -- cgit v0.10.2 From 2b953a5e994ce279904ec70220f7d4f31d380a0a Mon Sep 17 00:00:00 2001 From: Boris Ostrovsky Date: Tue, 28 Apr 2015 18:46:20 -0400 Subject: xen: Suspend ticks on all CPUs during suspend Commit 77e32c89a711 ("clockevents: Manage device's state separately for the core") decouples clockevent device's modes from states. With this change when a Xen guest tries to resume, it won't be calling its set_mode op which needs to be done on each VCPU in order to make the hypervisor aware that we are in oneshot mode. This happens because clockevents_tick_resume() (which is an intermediate step of resuming ticks on a processor) doesn't call clockevents_set_state() anymore and because during suspend clockevent devices on all VCPUs (except for the one doing the suspend) are left in ONESHOT state. As result, during resume the clockevents state machine will assume that device is already where it should be and doesn't need to be updated. To avoid this problem we should suspend ticks on all VCPUs during suspend. Signed-off-by: Boris Ostrovsky Signed-off-by: David Vrabel diff --git a/arch/x86/xen/suspend.c b/arch/x86/xen/suspend.c index d949769..53b4c08 100644 --- a/arch/x86/xen/suspend.c +++ b/arch/x86/xen/suspend.c @@ -88,7 +88,17 @@ static void xen_vcpu_notify_restore(void *data) tick_resume_local(); } +static void xen_vcpu_notify_suspend(void *data) +{ + tick_suspend_local(); +} + void xen_arch_resume(void) { on_each_cpu(xen_vcpu_notify_restore, NULL, 1); } + +void xen_arch_suspend(void) +{ + on_each_cpu(xen_vcpu_notify_suspend, NULL, 1); +} diff --git a/drivers/xen/manage.c b/drivers/xen/manage.c index bf19407..9e6a851 100644 --- a/drivers/xen/manage.c +++ b/drivers/xen/manage.c @@ -131,6 +131,8 @@ static void do_suspend(void) goto out_resume; } + xen_arch_suspend(); + si.cancelled = 1; err = stop_machine(xen_suspend, &si, cpumask_of(0)); @@ -148,11 +150,12 @@ static void do_suspend(void) si.cancelled = 1; } + xen_arch_resume(); + out_resume: - if (!si.cancelled) { - xen_arch_resume(); + if (!si.cancelled) xs_resume(); - } else + else xs_suspend_cancel(); dpm_resume_end(si.cancelled ? PMSG_THAW : PMSG_RESTORE); diff --git a/include/xen/xen-ops.h b/include/xen/xen-ops.h index c643e6a..0ce4f32 100644 --- a/include/xen/xen-ops.h +++ b/include/xen/xen-ops.h @@ -13,6 +13,7 @@ void xen_arch_post_suspend(int suspend_cancelled); void xen_timer_resume(void); void xen_arch_resume(void); +void xen_arch_suspend(void); void xen_resume_notifier_register(struct notifier_block *nb); void xen_resume_notifier_unregister(struct notifier_block *nb); -- cgit v0.10.2 From 8014bcc86ef112eab9ee1db312dba4e6b608cf89 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Mon, 13 Apr 2015 00:26:35 +0100 Subject: xen-pciback: Add name prefix to global 'permissive' variable The variable for the 'permissive' module parameter used to be static but was recently changed to be extern. This puts it in the kernel global namespace if the driver is built-in, so its name should begin with a prefix identifying the driver. Signed-off-by: Ben Hutchings Fixes: af6fc858a35b ("xen-pciback: limit guest control of command register") Signed-off-by: David Vrabel diff --git a/drivers/xen/xen-pciback/conf_space.c b/drivers/xen/xen-pciback/conf_space.c index 75fe3d4..9c23420 100644 --- a/drivers/xen/xen-pciback/conf_space.c +++ b/drivers/xen/xen-pciback/conf_space.c @@ -16,8 +16,8 @@ #include "conf_space.h" #include "conf_space_quirks.h" -bool permissive; -module_param(permissive, bool, 0644); +bool xen_pcibk_permissive; +module_param_named(permissive, xen_pcibk_permissive, bool, 0644); /* This is where xen_pcibk_read_config_byte, xen_pcibk_read_config_word, * xen_pcibk_write_config_word, and xen_pcibk_write_config_byte are created. */ @@ -262,7 +262,7 @@ int xen_pcibk_config_write(struct pci_dev *dev, int offset, int size, u32 value) * This means that some fields may still be read-only because * they have entries in the config_field list that intercept * the write and do nothing. */ - if (dev_data->permissive || permissive) { + if (dev_data->permissive || xen_pcibk_permissive) { switch (size) { case 1: err = pci_write_config_byte(dev, offset, diff --git a/drivers/xen/xen-pciback/conf_space.h b/drivers/xen/xen-pciback/conf_space.h index 2e1d73d..62461a8 100644 --- a/drivers/xen/xen-pciback/conf_space.h +++ b/drivers/xen/xen-pciback/conf_space.h @@ -64,7 +64,7 @@ struct config_field_entry { void *data; }; -extern bool permissive; +extern bool xen_pcibk_permissive; #define OFFSET(cfg_entry) ((cfg_entry)->base_offset+(cfg_entry)->field->offset) diff --git a/drivers/xen/xen-pciback/conf_space_header.c b/drivers/xen/xen-pciback/conf_space_header.c index c2260a0..ad3d17d 100644 --- a/drivers/xen/xen-pciback/conf_space_header.c +++ b/drivers/xen/xen-pciback/conf_space_header.c @@ -118,7 +118,7 @@ static int command_write(struct pci_dev *dev, int offset, u16 value, void *data) cmd->val = value; - if (!permissive && (!dev_data || !dev_data->permissive)) + if (!xen_pcibk_permissive && (!dev_data || !dev_data->permissive)) return 0; /* Only allow the guest to control certain bits. */ -- cgit v0.10.2 From 2cff98b99c469880ce830cbcde015b53b67e0a7b Mon Sep 17 00:00:00 2001 From: Dean Nelson Date: Wed, 29 Apr 2015 16:09:18 +0100 Subject: arm64: add missing PAGE_ALIGN() to __dma_free() __dma_alloc() does a PAGE_ALIGN() on the passed in size argument before doing anything else. __dma_free() does not. And because it doesn't, it is possible to leak memory should size not be an integer multiple of PAGE_SIZE. The solution is to add a PAGE_ALIGN() to __dma_free() like is done in __dma_alloc(). Additionally, this patch removes a redundant PAGE_ALIGN() from __dma_alloc_coherent(), since __dma_alloc_coherent() can only be called from __dma_alloc(), which already does a PAGE_ALIGN() before the call. Cc: stable@vger.kernel.org Acked-by: Catalin Marinas Signed-off-by: Dean Nelson Signed-off-by: Will Deacon diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c index e0f14ee..b0bd4e5 100644 --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -104,7 +104,6 @@ static void *__dma_alloc_coherent(struct device *dev, size_t size, struct page *page; void *addr; - size = PAGE_ALIGN(size); page = dma_alloc_from_contiguous(dev, size >> PAGE_SHIFT, get_order(size)); if (!page) @@ -193,6 +192,8 @@ static void __dma_free(struct device *dev, size_t size, { void *swiotlb_addr = phys_to_virt(dma_to_phys(dev, dma_handle)); + size = PAGE_ALIGN(size); + if (!is_device_dma_coherent(dev)) { if (__free_from_pool(vaddr, size)) return; -- cgit v0.10.2 From cb6ccf09d6b94bec4def1ac5cf4678d12b216474 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 28 Apr 2015 11:43:15 +0800 Subject: route: Use ipv4_mtu instead of raw rt_pmtu The commit 3cdaa5be9e81a914e633a6be7b7d2ef75b528562 ("ipv4: Don't increase PMTU with Datagram Too Big message") broke PMTU in cases where the rt_pmtu value has expired but is smaller than the new PMTU value. This obsolete rt_pmtu then prevents the new PMTU value from being installed. Fixes: 3cdaa5be9e81 ("ipv4: Don't increase PMTU with Datagram Too Big message") Reported-by: Gerd v. Egidy Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/net/ipv4/route.c b/net/ipv4/route.c index a78540f..bff62fc 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -962,10 +962,7 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) if (dst_metric_locked(dst, RTAX_MTU)) return; - if (dst->dev->mtu < mtu) - return; - - if (rt->rt_pmtu && rt->rt_pmtu < mtu) + if (ipv4_mtu(dst) < mtu) return; if (mtu < ip_rt_min_pmtu) -- cgit v0.10.2 From 3f300ff41d89fe9674b8dbab950ba2572639ee8d Mon Sep 17 00:00:00 2001 From: Simon Xiao Date: Tue, 28 Apr 2015 01:05:17 -0700 Subject: hv_netvsc: introduce netif-msg into netvsc module 1. Introduce netif-msg to netvsc to control debug logging output and keep msg_enable in netvsc_device_context so that it is kept persistently. 2. Only call dump_rndis_message() when NETIF_MSG_RX_ERR or above is specified in netvsc module debug param. In non-debug mode, in current code, dump_rndis_message() will not dump anything but it still initialize some local variables and process the switch logic which is unnecessary, especially in high network throughput situation. Signed-off-by: Simon Xiao Reviewed-by: K. Y. Srinivasan Reviewed-by: Haiyang Zhang Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index a10b316..e55c8f4f 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -612,6 +612,15 @@ struct multi_send_data { u32 count; /* counter of batched packets */ }; +/* The context of the netvsc device */ +struct net_device_context { + /* point back to our device context */ + struct hv_device *device_ctx; + struct delayed_work dwork; + struct work_struct work; + u32 msg_enable; /* debug level */ +}; + /* Per netvsc device */ struct netvsc_device { struct hv_device *dev; @@ -667,6 +676,9 @@ struct netvsc_device { struct multi_send_data msd[NR_CPUS]; u32 max_pkt; /* max number of pkt in one send, e.g. 8 */ u32 pkt_align; /* alignment bytes, e.g. 8 */ + + /* The net device context */ + struct net_device_context *nd_ctx; }; /* NdisInitialize message */ diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 2e8ad06..c651d4d 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -1197,6 +1197,9 @@ int netvsc_device_add(struct hv_device *device, void *additional_info) */ ndev = net_device->ndev; + /* Add netvsc_device context to netvsc_device */ + net_device->nd_ctx = netdev_priv(ndev); + /* Initialize the NetVSC channel extension */ init_completion(&net_device->channel_init_wait); diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index a3a9d38..66c4b0c 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -40,18 +40,21 @@ #include "hyperv_net.h" -struct net_device_context { - /* point back to our device context */ - struct hv_device *device_ctx; - struct delayed_work dwork; - struct work_struct work; -}; #define RING_SIZE_MIN 64 static int ring_size = 128; module_param(ring_size, int, S_IRUGO); MODULE_PARM_DESC(ring_size, "Ring buffer size (# of pages)"); +static const u32 default_msg = NETIF_MSG_DRV | NETIF_MSG_PROBE | + NETIF_MSG_LINK | NETIF_MSG_IFUP | + NETIF_MSG_IFDOWN | NETIF_MSG_RX_ERR | + NETIF_MSG_TX_ERR; + +static int debug = -1; +module_param(debug, int, S_IRUGO); +MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)"); + static void do_set_multicast(struct work_struct *w) { struct net_device_context *ndevctx = @@ -888,6 +891,11 @@ static int netvsc_probe(struct hv_device *dev, net_device_ctx = netdev_priv(net); net_device_ctx->device_ctx = dev; + net_device_ctx->msg_enable = netif_msg_init(debug, default_msg); + if (netif_msg_probe(net_device_ctx)) + netdev_dbg(net, "netvsc msg_enable: %d\n", + net_device_ctx->msg_enable); + hv_set_drvdata(dev, net); INIT_DELAYED_WORK(&net_device_ctx->dwork, netvsc_link_change); INIT_WORK(&net_device_ctx->work, do_set_multicast); diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 0d92efe..9118cea 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -429,7 +429,8 @@ int rndis_filter_receive(struct hv_device *dev, rndis_msg = pkt->data; - dump_rndis_message(dev, rndis_msg); + if (netif_msg_rx_err(net_dev->nd_ctx)) + dump_rndis_message(dev, rndis_msg); switch (rndis_msg->ndis_msg_type) { case RNDIS_MSG_PACKET: -- cgit v0.10.2 From d9b9e860cef1d904832bc6e92e444adfeabe49e9 Mon Sep 17 00:00:00 2001 From: Michal Schmidt Date: Tue, 28 Apr 2015 11:34:21 +0200 Subject: bnx2x: mark LRO as a fixed disabled feature if disable_tpa is set If disable_tpa is set, remove NETIF_F_LRO from hw_features, so ethtool sees it as "off [fixed]". Note that setting the NETIF_F_LRO bit in dev->features in the 'else' branch is not needed, because the bit was already set by bnx2x_init_dev(). Then the check for disable_tpa in in bnx2x_fix_features() becomes unnecessary. Signed-off-by: Michal Schmidt Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 3558a36..dcdbd00 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -4834,10 +4834,6 @@ netdev_features_t bnx2x_fix_features(struct net_device *dev, features &= ~NETIF_F_GRO; } - /* Note: do not disable SW GRO in kernel when HW GRO is off */ - if (bp->disable_tpa) - features &= ~NETIF_F_LRO; - return features; } diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index b9f85fcc..2d067c9 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -12108,10 +12108,10 @@ static int bnx2x_init_bp(struct bnx2x *bp) /* Set TPA flags */ if (bp->disable_tpa) { bp->flags &= ~(TPA_ENABLE_FLAG | GRO_ENABLE_FLAG); + bp->dev->hw_features &= ~NETIF_F_LRO; bp->dev->features &= ~NETIF_F_LRO; } else { bp->flags |= (TPA_ENABLE_FLAG | GRO_ENABLE_FLAG); - bp->dev->features |= NETIF_F_LRO; } if (CHIP_IS_E1(bp)) -- cgit v0.10.2 From 7e6b4d440b0ae9062b84dfb417ea6d51a45dab76 Mon Sep 17 00:00:00 2001 From: Michal Schmidt Date: Tue, 28 Apr 2015 11:34:22 +0200 Subject: bnx2x: merge fp->disable_tpa with fp->mode It is simpler to have the TPA mode as one three-state variable. Signed-off-by: Michal Schmidt Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 355d5fe..2a41604 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -521,6 +521,7 @@ struct bnx2x_fp_txdata { }; enum bnx2x_tpa_mode_t { + TPA_MODE_DISABLED, TPA_MODE_LRO, TPA_MODE_GRO }; @@ -589,7 +590,6 @@ struct bnx2x_fastpath { /* TPA related */ struct bnx2x_agg_info *tpa_info; - u8 disable_tpa; #ifdef BNX2X_STOP_ON_ERROR u64 tpa_queue_used; #endif diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index dcdbd00..84612b0 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -947,10 +947,10 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget) u16 frag_size, pages; #ifdef BNX2X_STOP_ON_ERROR /* sanity check */ - if (fp->disable_tpa && + if (fp->mode == TPA_MODE_DISABLED && (CQE_TYPE_START(cqe_fp_type) || CQE_TYPE_STOP(cqe_fp_type))) - BNX2X_ERR("START/STOP packet while disable_tpa type %x\n", + BNX2X_ERR("START/STOP packet while TPA disabled, type %x\n", CQE_TYPE(cqe_fp_type)); #endif @@ -1396,7 +1396,7 @@ void bnx2x_init_rx_rings(struct bnx2x *bp) DP(NETIF_MSG_IFUP, "mtu %d rx_buf_size %d\n", bp->dev->mtu, fp->rx_buf_size); - if (!fp->disable_tpa) { + if (fp->mode != TPA_MODE_DISABLED) { /* Fill the per-aggregation pool */ for (i = 0; i < MAX_AGG_QS(bp); i++) { struct bnx2x_agg_info *tpa_info = @@ -1410,7 +1410,7 @@ void bnx2x_init_rx_rings(struct bnx2x *bp) BNX2X_ERR("Failed to allocate TPA skb pool for queue[%d] - disabling TPA on this queue!\n", j); bnx2x_free_tpa_pool(bp, fp, i); - fp->disable_tpa = 1; + fp->mode = TPA_MODE_DISABLED; break; } dma_unmap_addr_set(first_buf, mapping, 0); @@ -1438,7 +1438,7 @@ void bnx2x_init_rx_rings(struct bnx2x *bp) ring_prod); bnx2x_free_tpa_pool(bp, fp, MAX_AGG_QS(bp)); - fp->disable_tpa = 1; + fp->mode = TPA_MODE_DISABLED; ring_prod = 0; break; } @@ -1560,7 +1560,7 @@ static void bnx2x_free_rx_skbs(struct bnx2x *bp) bnx2x_free_rx_bds(fp); - if (!fp->disable_tpa) + if (fp->mode != TPA_MODE_DISABLED) bnx2x_free_tpa_pool(bp, fp, MAX_AGG_QS(bp)); } } @@ -2477,19 +2477,19 @@ static void bnx2x_bz_fp(struct bnx2x *bp, int index) /* set the tpa flag for each queue. The tpa flag determines the queue * minimal size so it must be set prior to queue memory allocation */ - fp->disable_tpa = !(bp->flags & TPA_ENABLE_FLAG || - (bp->flags & GRO_ENABLE_FLAG && - bnx2x_mtu_allows_gro(bp->dev->mtu))); if (bp->flags & TPA_ENABLE_FLAG) fp->mode = TPA_MODE_LRO; - else if (bp->flags & GRO_ENABLE_FLAG) + else if (bp->flags & GRO_ENABLE_FLAG && + bnx2x_mtu_allows_gro(bp->dev->mtu)) fp->mode = TPA_MODE_GRO; + else + fp->mode = TPA_MODE_DISABLED; /* We don't want TPA if it's disabled in bp * or if this is an FCoE L2 ring. */ if (bp->disable_tpa || IS_FCOE_FP(fp)) - fp->disable_tpa = 1; + fp->mode = TPA_MODE_DISABLED; } int bnx2x_load_cnic(struct bnx2x *bp) @@ -2610,7 +2610,7 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode) /* * Zero fastpath structures preserving invariants like napi, which are * allocated only once, fp index, max_cos, bp pointer. - * Also set fp->disable_tpa and txdata_ptr. + * Also set fp->mode and txdata_ptr. */ DP(NETIF_MSG_IFUP, "num queues: %d", bp->num_queues); for_each_queue(bp, i) @@ -4545,7 +4545,7 @@ alloc_mem_err: * In these cases we disable the queue * Min size is different for OOO, TPA and non-TPA queues */ - if (ring_size < (fp->disable_tpa ? + if (ring_size < (fp->mode == TPA_MODE_DISABLED ? MIN_RX_SIZE_NONTPA : MIN_RX_SIZE_TPA)) { /* release memory allocated for this queue */ bnx2x_free_fp_mem_at(bp, index); diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index adcacda..d7a7175 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -969,7 +969,7 @@ static inline void bnx2x_free_rx_sge_range(struct bnx2x *bp, { int i; - if (fp->disable_tpa) + if (fp->mode == TPA_MODE_DISABLED) return; for (i = 0; i < last; i++) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 2d067c9..64ea5ad 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -3128,7 +3128,7 @@ static unsigned long bnx2x_get_q_flags(struct bnx2x *bp, __set_bit(BNX2X_Q_FLG_FORCE_DEFAULT_PRI, &flags); } - if (!fp->disable_tpa) { + if (fp->mode != TPA_MODE_DISABLED) { __set_bit(BNX2X_Q_FLG_TPA, &flags); __set_bit(BNX2X_Q_FLG_TPA_IPV6, &flags); if (fp->mode == TPA_MODE_GRO) @@ -3176,7 +3176,7 @@ static void bnx2x_pf_rx_q_prep(struct bnx2x *bp, u16 sge_sz = 0; u16 tpa_agg_size = 0; - if (!fp->disable_tpa) { + if (fp->mode != TPA_MODE_DISABLED) { pause->sge_th_lo = SGE_TH_LO(bp); pause->sge_th_hi = SGE_TH_HI(bp); diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c index 15b2d16..06b8c0d 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c @@ -594,7 +594,7 @@ int bnx2x_vfpf_setup_q(struct bnx2x *bp, struct bnx2x_fastpath *fp, bnx2x_vfpf_prep(bp, &req->first_tlv, CHANNEL_TLV_SETUP_Q, sizeof(*req)); /* select tpa mode to request */ - if (!fp->disable_tpa) { + if (fp->mode != TPA_MODE_DISABLED) { flags |= VFPF_QUEUE_FLG_TPA; flags |= VFPF_QUEUE_FLG_TPA_IPV6; if (fp->mode == TPA_MODE_GRO) -- cgit v0.10.2 From f8dcb5e3365a23b620f5744f23f1918b9c38d883 Mon Sep 17 00:00:00 2001 From: Michal Schmidt Date: Tue, 28 Apr 2015 11:34:23 +0200 Subject: bnx2x: remove {TPA,GRO}_ENABLE_FLAG These flags are redundant with dev->features. Remove them. Just make sure to set dev->features ourselves in bnx2x_set_features() before performing the reload of the card. Signed-off-by: Michal Schmidt Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 2a41604..a3b0f7a 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -1545,9 +1545,7 @@ struct bnx2x { #define USING_MSIX_FLAG (1 << 5) #define USING_MSI_FLAG (1 << 6) #define DISABLE_MSI_FLAG (1 << 7) -#define TPA_ENABLE_FLAG (1 << 8) #define NO_MCP_FLAG (1 << 9) -#define GRO_ENABLE_FLAG (1 << 10) #define MF_FUNC_DIS (1 << 11) #define OWN_CNIC_IRQ (1 << 12) #define NO_ISCSI_OOO_FLAG (1 << 13) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 84612b0..a8bb8f6 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -2477,9 +2477,9 @@ static void bnx2x_bz_fp(struct bnx2x *bp, int index) /* set the tpa flag for each queue. The tpa flag determines the queue * minimal size so it must be set prior to queue memory allocation */ - if (bp->flags & TPA_ENABLE_FLAG) + if (bp->dev->features & NETIF_F_LRO) fp->mode = TPA_MODE_LRO; - else if (bp->flags & GRO_ENABLE_FLAG && + else if (bp->dev->features & NETIF_F_GRO && bnx2x_mtu_allows_gro(bp->dev->mtu)) fp->mode = TPA_MODE_GRO; else @@ -3249,7 +3249,7 @@ int bnx2x_low_latency_recv(struct napi_struct *napi) if ((bp->state == BNX2X_STATE_CLOSED) || (bp->state == BNX2X_STATE_ERROR) || - (bp->flags & (TPA_ENABLE_FLAG | GRO_ENABLE_FLAG))) + (bp->dev->features & (NETIF_F_LRO | NETIF_F_GRO))) return LL_FLUSH_FAILED; if (!bnx2x_fp_lock_poll(fp)) @@ -4840,19 +4840,9 @@ netdev_features_t bnx2x_fix_features(struct net_device *dev, int bnx2x_set_features(struct net_device *dev, netdev_features_t features) { struct bnx2x *bp = netdev_priv(dev); - u32 flags = bp->flags; - u32 changes; + netdev_features_t changes = features ^ dev->features; bool bnx2x_reload = false; - - if (features & NETIF_F_LRO) - flags |= TPA_ENABLE_FLAG; - else - flags &= ~TPA_ENABLE_FLAG; - - if (features & NETIF_F_GRO) - flags |= GRO_ENABLE_FLAG; - else - flags &= ~GRO_ENABLE_FLAG; + int rc; /* VFs or non SRIOV PFs should be able to change loopback feature */ if (!pci_num_vf(bp->pdev)) { @@ -4869,24 +4859,23 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features) } } - changes = flags ^ bp->flags; - /* if GRO is changed while LRO is enabled, don't force a reload */ - if ((changes & GRO_ENABLE_FLAG) && (flags & TPA_ENABLE_FLAG)) - changes &= ~GRO_ENABLE_FLAG; + if ((changes & NETIF_F_GRO) && (features & NETIF_F_LRO)) + changes &= ~NETIF_F_GRO; /* if GRO is changed while HW TPA is off, don't force a reload */ - if ((changes & GRO_ENABLE_FLAG) && bp->disable_tpa) - changes &= ~GRO_ENABLE_FLAG; + if ((changes & NETIF_F_GRO) && bp->disable_tpa) + changes &= ~NETIF_F_GRO; if (changes) bnx2x_reload = true; - bp->flags = flags; - if (bnx2x_reload) { - if (bp->recovery_state == BNX2X_RECOVERY_DONE) - return bnx2x_reload_if_running(dev); + if (bp->recovery_state == BNX2X_RECOVERY_DONE) { + dev->features = features; + rc = bnx2x_reload_if_running(dev); + return rc ? rc : 1; + } /* else: bnx2x_nic_load() will be called at end of recovery */ } diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 64ea5ad..60d2b8e 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -3304,7 +3304,7 @@ static void bnx2x_pf_init(struct bnx2x *bp) /* This flag is relevant for E1x only. * E2 doesn't have a TPA configuration in a function level. */ - flags |= (bp->flags & TPA_ENABLE_FLAG) ? FUNC_FLG_TPA : 0; + flags |= (bp->dev->features & NETIF_F_LRO) ? FUNC_FLG_TPA : 0; func_init.func_flgs = flags; func_init.pf_id = BP_FUNC(bp); @@ -12107,11 +12107,8 @@ static int bnx2x_init_bp(struct bnx2x *bp) /* Set TPA flags */ if (bp->disable_tpa) { - bp->flags &= ~(TPA_ENABLE_FLAG | GRO_ENABLE_FLAG); bp->dev->hw_features &= ~NETIF_F_LRO; bp->dev->features &= ~NETIF_F_LRO; - } else { - bp->flags |= (TPA_ENABLE_FLAG | GRO_ENABLE_FLAG); } if (CHIP_IS_E1(bp)) -- cgit v0.10.2 From 18fe369a5a4f175ee652ba8b7d8d7e49f140e281 Mon Sep 17 00:00:00 2001 From: Antonio Ospite Date: Tue, 28 Apr 2015 13:11:27 +0200 Subject: trivial: net: atl1e: atl1e_hw.h: fix 0x0x prefix Fix the 0x0x prefix in an integer constant. Signed-off-by: Antonio Ospite Cc: Jay Cliburn Cc: Chris Snook Cc: netdev@vger.kernel.org Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h b/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h index 74df16a..88a6271 100644 --- a/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h +++ b/drivers/net/ethernet/atheros/atl1e/atl1e_hw.h @@ -129,7 +129,7 @@ s32 atl1e_restart_autoneg(struct atl1e_hw *hw); #define TWSI_CTRL_LD_SLV_ADDR_SHIFT 8 #define TWSI_CTRL_SW_LDSTART 0x800 #define TWSI_CTRL_HW_LDSTART 0x1000 -#define TWSI_CTRL_SMB_SLV_ADDR_MASK 0x0x7F +#define TWSI_CTRL_SMB_SLV_ADDR_MASK 0x7F #define TWSI_CTRL_SMB_SLV_ADDR_SHIFT 15 #define TWSI_CTRL_LD_EXIST 0x400000 #define TWSI_CTRL_READ_FREQ_SEL_MASK 0x3 -- cgit v0.10.2 From 165996bd351f0e711feff92479c0796098afaa38 Mon Sep 17 00:00:00 2001 From: Antonio Ospite Date: Tue, 28 Apr 2015 13:11:29 +0200 Subject: trivial: net: systemport: bcmsysport.h: fix 0x0x prefix Fix the 0x0x prefix in an integer constant. In this case, while at it, also fix a typo (s/unitcast/unicast/). Signed-off-by: Antonio Ospite Cc: Florian Fainelli Cc: netdev@vger.kernel.org Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bcmsysport.h b/drivers/net/ethernet/broadcom/bcmsysport.h index 7e3d87a..e2c043e 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.h +++ b/drivers/net/ethernet/broadcom/bcmsysport.h @@ -543,7 +543,7 @@ struct bcm_sysport_tx_counters { u32 jbr; /* RO # of xmited jabber count*/ u32 bytes; /* RO # of xmited byte count */ u32 pok; /* RO # of xmited good pkt */ - u32 uc; /* RO (0x0x4f0)# of xmited unitcast pkt */ + u32 uc; /* RO (0x4f0) # of xmited unicast pkt */ }; struct bcm_sysport_mib { -- cgit v0.10.2 From 2b70fe5aba0dd00d81173243c7ab04c66aeb67d8 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 28 Apr 2015 13:33:21 +0200 Subject: net: sched: act_connmark: don't zap skb->nfct This action is meant to be passive, i.e. we should not alter skb->nfct: If nfct is present just leave it alone. Compile tested only. Cc: Jamal Hadi Salim Signed-off-by: Florian Westphal Acked-by: Pablo Neira Ayuso Signed-off-by: David S. Miller diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index 8e47251..295d14b 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -63,7 +63,6 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a, skb->mark = c->mark; /* using overlimits stats to count how many packets marked */ ca->tcf_qstats.overlimits++; - nf_ct_put(c); goto out; } @@ -82,7 +81,6 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a, nf_ct_put(c); out: - skb->nfct = NULL; spin_unlock(&ca->tcf_lock); return ca->tcf_action; } -- cgit v0.10.2 From 821996795973fd52703c35811a03db9fec1ac141 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 28 Apr 2015 18:33:48 +0200 Subject: bridge/mdb: remove wrong use of NLM_F_MULTI NLM_F_MULTI must be used only when a NLMSG_DONE message is sent. In fact, it is sent only at the end of a dump. Libraries like libnl will wait forever for NLMSG_DONE. Fixes: 37a393bc4932 ("bridge: notify mdb changes via netlink") CC: Cong Wang CC: Stephen Hemminger CC: bridge@lists.linux-foundation.org Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c index 4096089..e29ad70 100644 --- a/net/bridge/br_mdb.c +++ b/net/bridge/br_mdb.c @@ -170,7 +170,7 @@ static int nlmsg_populate_mdb_fill(struct sk_buff *skb, struct br_port_msg *bpm; struct nlattr *nest, *nest2; - nlh = nlmsg_put(skb, pid, seq, type, sizeof(*bpm), NLM_F_MULTI); + nlh = nlmsg_put(skb, pid, seq, type, sizeof(*bpm), 0); if (!nlh) return -EMSGSIZE; -- cgit v0.10.2 From 46c264daaaa569e24f8aba877d0fd8167c42a9a4 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 28 Apr 2015 18:33:49 +0200 Subject: bridge/nl: remove wrong use of NLM_F_MULTI NLM_F_MULTI must be used only when a NLMSG_DONE message is sent. In fact, it is sent only at the end of a dump. Libraries like libnl will wait forever for NLMSG_DONE. Fixes: e5a55a898720 ("net: create generic bridge ops") Fixes: 815cccbf10b2 ("ixgbe: add setlink, getlink support to ixgbe and ixgbevf") CC: John Fastabend CC: Sathya Perla CC: Subbu Seetharaman CC: Ajit Khaparde CC: Jeff Kirsher CC: intel-wired-lan@lists.osuosl.org CC: Jiri Pirko CC: Scott Feldman CC: Stephen Hemminger CC: bridge@lists.linux-foundation.org Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index fb0bc3c..a6dcbf8 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -4846,7 +4846,8 @@ err: } static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, - struct net_device *dev, u32 filter_mask) + struct net_device *dev, u32 filter_mask, + int nlflags) { struct be_adapter *adapter = netdev_priv(dev); int status = 0; @@ -4868,7 +4869,7 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, return ndo_dflt_bridge_getlink(skb, pid, seq, dev, hsw_mode == PORT_FWD_TYPE_VEPA ? BRIDGE_MODE_VEPA : BRIDGE_MODE_VEB, - 0, 0); + 0, 0, nlflags); } #ifdef CONFIG_BE2NET_VXLAN diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 24481cd..a54c144 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -8053,10 +8053,10 @@ static int i40e_ndo_bridge_setlink(struct net_device *dev, #ifdef HAVE_BRIDGE_FILTER static int i40e_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, - u32 __always_unused filter_mask) + u32 __always_unused filter_mask, int nlflags) #else static int i40e_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, - struct net_device *dev) + struct net_device *dev, int nlflags) #endif /* HAVE_BRIDGE_FILTER */ { struct i40e_netdev_priv *np = netdev_priv(dev); @@ -8078,7 +8078,8 @@ static int i40e_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, if (!veb) return 0; - return ndo_dflt_bridge_getlink(skb, pid, seq, dev, veb->bridge_mode); + return ndo_dflt_bridge_getlink(skb, pid, seq, dev, veb->bridge_mode, + nlflags); } #endif /* HAVE_BRIDGE_ATTRIBS */ diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index d3f4b0c..5be12a0 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -8044,7 +8044,7 @@ static int ixgbe_ndo_bridge_setlink(struct net_device *dev, static int ixgbe_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, - u32 filter_mask) + u32 filter_mask, int nlflags) { struct ixgbe_adapter *adapter = netdev_priv(dev); @@ -8052,7 +8052,7 @@ static int ixgbe_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, return 0; return ndo_dflt_bridge_getlink(skb, pid, seq, dev, - adapter->bridge_mode, 0, 0); + adapter->bridge_mode, 0, 0, nlflags); } static void *ixgbe_fwd_add(struct net_device *pdev, struct net_device *vdev) diff --git a/drivers/net/ethernet/rocker/rocker.c b/drivers/net/ethernet/rocker/rocker.c index a570a60..ec25153 100644 --- a/drivers/net/ethernet/rocker/rocker.c +++ b/drivers/net/ethernet/rocker/rocker.c @@ -4176,14 +4176,15 @@ static int rocker_port_bridge_setlink(struct net_device *dev, static int rocker_port_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, - u32 filter_mask) + u32 filter_mask, int nlflags) { struct rocker_port *rocker_port = netdev_priv(dev); u16 mode = BRIDGE_MODE_UNDEF; u32 mask = BR_LEARNING | BR_LEARNING_SYNC; return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, - rocker_port->brport_flags, mask); + rocker_port->brport_flags, mask, + nlflags); } static int rocker_port_get_phys_port_name(struct net_device *dev, diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index dbad4d7..1899c74 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -977,7 +977,8 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev, * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh, * u16 flags) * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq, - * struct net_device *dev, u32 filter_mask) + * struct net_device *dev, u32 filter_mask, + * int nlflags) * int (*ndo_bridge_dellink)(struct net_device *dev, struct nlmsghdr *nlh, * u16 flags); * @@ -1173,7 +1174,8 @@ struct net_device_ops { int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, - u32 filter_mask); + u32 filter_mask, + int nlflags); int (*ndo_bridge_dellink)(struct net_device *dev, struct nlmsghdr *nlh, u16 flags); diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 2da5d10..7b8e260 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -122,5 +122,5 @@ extern int ndo_dflt_fdb_del(struct ndmsg *ndm, extern int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, u16 mode, - u32 flags, u32 mask); + u32 flags, u32 mask, int nlflags); #endif /* __LINUX_RTNETLINK_H */ diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 0e4ddb8..4b5c236 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -394,7 +394,7 @@ errout: * Dump information about all ports, in response to GETLINK */ int br_getlink(struct sk_buff *skb, u32 pid, u32 seq, - struct net_device *dev, u32 filter_mask) + struct net_device *dev, u32 filter_mask, int nlflags) { struct net_bridge_port *port = br_port_get_rtnl(dev); @@ -402,7 +402,7 @@ int br_getlink(struct sk_buff *skb, u32 pid, u32 seq, !(filter_mask & RTEXT_FILTER_BRVLAN_COMPRESSED)) return 0; - return br_fill_ifinfo(skb, port, pid, seq, RTM_NEWLINK, NLM_F_MULTI, + return br_fill_ifinfo(skb, port, pid, seq, RTM_NEWLINK, nlflags, filter_mask, dev); } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 6ca0251..3362c29 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -828,7 +828,7 @@ void br_ifinfo_notify(int event, struct net_bridge_port *port); int br_setlink(struct net_device *dev, struct nlmsghdr *nlmsg, u16 flags); int br_dellink(struct net_device *dev, struct nlmsghdr *nlmsg, u16 flags); int br_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, - u32 filter_mask); + u32 filter_mask, int nlflags); #ifdef CONFIG_SYSFS /* br_sysfs_if.c */ diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 358d52a..666e092 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -2854,7 +2854,7 @@ static int brport_nla_put_flag(struct sk_buff *skb, u32 flags, u32 mask, int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, u16 mode, - u32 flags, u32 mask) + u32 flags, u32 mask, int nlflags) { struct nlmsghdr *nlh; struct ifinfomsg *ifm; @@ -2863,7 +2863,7 @@ int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, u8 operstate = netif_running(dev) ? dev->operstate : IF_OPER_DOWN; struct net_device *br_dev = netdev_master_upper_dev_get(dev); - nlh = nlmsg_put(skb, pid, seq, RTM_NEWLINK, sizeof(*ifm), NLM_F_MULTI); + nlh = nlmsg_put(skb, pid, seq, RTM_NEWLINK, sizeof(*ifm), nlflags); if (nlh == NULL) return -EMSGSIZE; @@ -2969,7 +2969,8 @@ static int rtnl_bridge_getlink(struct sk_buff *skb, struct netlink_callback *cb) if (br_dev && br_dev->netdev_ops->ndo_bridge_getlink) { if (idx >= cb->args[0] && br_dev->netdev_ops->ndo_bridge_getlink( - skb, portid, seq, dev, filter_mask) < 0) + skb, portid, seq, dev, filter_mask, + NLM_F_MULTI) < 0) break; idx++; } @@ -2977,7 +2978,8 @@ static int rtnl_bridge_getlink(struct sk_buff *skb, struct netlink_callback *cb) if (ops->ndo_bridge_getlink) { if (idx >= cb->args[0] && ops->ndo_bridge_getlink(skb, portid, seq, dev, - filter_mask) < 0) + filter_mask, + NLM_F_MULTI) < 0) break; idx++; } @@ -3018,7 +3020,7 @@ static int rtnl_bridge_notify(struct net_device *dev) goto errout; } - err = dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev, 0); + err = dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev, 0, 0); if (err < 0) goto errout; -- cgit v0.10.2 From f2f67390a4b961dae83733732e96e1a394a53c4e Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 28 Apr 2015 18:33:50 +0200 Subject: tipc: remove wrong use of NLM_F_MULTI NLM_F_MULTI must be used only when a NLMSG_DONE message is sent. In fact, it is sent only at the end of a dump. Libraries like libnl will wait forever for NLMSG_DONE. Fixes: 35b9dd7607f0 ("tipc: add bearer get/dump to new netlink api") Fixes: 7be57fc69184 ("tipc: add link get/dump to new netlink api") Fixes: 46f15c6794fb ("tipc: add media get/dump to new netlink api") CC: Richard Alpe CC: Jon Maloy CC: Ying Xue CC: tipc-discussion@lists.sourceforge.net Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 3613e72..70e3dac 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -591,14 +591,14 @@ void tipc_bearer_stop(struct net *net) /* Caller should hold rtnl_lock to protect the bearer */ static int __tipc_nl_add_bearer(struct tipc_nl_msg *msg, - struct tipc_bearer *bearer) + struct tipc_bearer *bearer, int nlflags) { void *hdr; struct nlattr *attrs; struct nlattr *prop; hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, - NLM_F_MULTI, TIPC_NL_BEARER_GET); + nlflags, TIPC_NL_BEARER_GET); if (!hdr) return -EMSGSIZE; @@ -657,7 +657,7 @@ int tipc_nl_bearer_dump(struct sk_buff *skb, struct netlink_callback *cb) if (!bearer) continue; - err = __tipc_nl_add_bearer(&msg, bearer); + err = __tipc_nl_add_bearer(&msg, bearer, NLM_F_MULTI); if (err) break; } @@ -705,7 +705,7 @@ int tipc_nl_bearer_get(struct sk_buff *skb, struct genl_info *info) goto err_out; } - err = __tipc_nl_add_bearer(&msg, bearer); + err = __tipc_nl_add_bearer(&msg, bearer, 0); if (err) goto err_out; rtnl_unlock(); @@ -857,14 +857,14 @@ int tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info) } static int __tipc_nl_add_media(struct tipc_nl_msg *msg, - struct tipc_media *media) + struct tipc_media *media, int nlflags) { void *hdr; struct nlattr *attrs; struct nlattr *prop; hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, - NLM_F_MULTI, TIPC_NL_MEDIA_GET); + nlflags, TIPC_NL_MEDIA_GET); if (!hdr) return -EMSGSIZE; @@ -916,7 +916,8 @@ int tipc_nl_media_dump(struct sk_buff *skb, struct netlink_callback *cb) rtnl_lock(); for (; media_info_array[i] != NULL; i++) { - err = __tipc_nl_add_media(&msg, media_info_array[i]); + err = __tipc_nl_add_media(&msg, media_info_array[i], + NLM_F_MULTI); if (err) break; } @@ -963,7 +964,7 @@ int tipc_nl_media_get(struct sk_buff *skb, struct genl_info *info) goto err_out; } - err = __tipc_nl_add_media(&msg, media); + err = __tipc_nl_add_media(&msg, media, 0); if (err) goto err_out; rtnl_unlock(); diff --git a/net/tipc/link.c b/net/tipc/link.c index 57be6e6..a11cc5e 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -2013,7 +2013,7 @@ msg_full: /* Caller should hold appropriate locks to protect the link */ static int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg, - struct tipc_link *link) + struct tipc_link *link, int nlflags) { int err; void *hdr; @@ -2022,7 +2022,7 @@ static int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg, struct tipc_net *tn = net_generic(net, tipc_net_id); hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, - NLM_F_MULTI, TIPC_NL_LINK_GET); + nlflags, TIPC_NL_LINK_GET); if (!hdr) return -EMSGSIZE; @@ -2095,7 +2095,7 @@ static int __tipc_nl_add_node_links(struct net *net, struct tipc_nl_msg *msg, if (!node->links[i]) continue; - err = __tipc_nl_add_link(net, msg, node->links[i]); + err = __tipc_nl_add_link(net, msg, node->links[i], NLM_F_MULTI); if (err) return err; } @@ -2209,7 +2209,7 @@ int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info) goto err_out; } - err = __tipc_nl_add_link(net, &msg, link); + err = __tipc_nl_add_link(net, &msg, link, 0); if (err) goto err_out; -- cgit v0.10.2 From 0d699f28ee5d0641470a603ab5904e463cb1532a Mon Sep 17 00:00:00 2001 From: Jon Paul Maloy Date: Tue, 28 Apr 2015 16:59:04 -0400 Subject: tipc: fix problem with parallel link synchronization mechanism Currently, we try to accumulate arrived packets in the links's 'deferred' queue during the parallel link syncronization phase. This entails two problems: - With an unlucky combination of arriving packets the algorithm may go into a lockstep with the out-of-sequence handling function, where the synch mechanism is adding a packet to the deferred queue, while the out-of-sequence handling is retrieving it again, thus ending up in a loop inside the node_lock scope. - Even if this is avoided, the link will very often send out unnecessary protocol messages, in the worst case leading to redundant retransmissions. We fix this by just dropping arriving packets on the upcoming link during the synchronization phase, thus relying on the retransmission protocol to resolve the situation once the two links have arrived to a synchronized state. Reviewed-by: Erik Hugne Reviewed-by: Ying Xue Signed-off-by: Jon Maloy Signed-off-by: David S. Miller diff --git a/net/tipc/link.c b/net/tipc/link.c index a11cc5e..43a515d 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -1145,11 +1145,8 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr) } /* Synchronize with parallel link if applicable */ if (unlikely((l_ptr->flags & LINK_SYNCHING) && !msg_dup(msg))) { - link_handle_out_of_seq_msg(l_ptr, skb); - if (link_synch(l_ptr)) - link_retrieve_defq(l_ptr, &head); - skb = NULL; - goto unlock; + if (!link_synch(l_ptr)) + goto unlock; } l_ptr->next_in_no++; if (unlikely(!skb_queue_empty(&l_ptr->deferdq))) -- cgit v0.10.2 From 42eab005a5dd5d7ea2b0328aecc4d6cc0c23c9c2 Mon Sep 17 00:00:00 2001 From: Benjamin Poirier Date: Tue, 28 Apr 2015 14:49:29 -0700 Subject: mlx4: Fix tx ring affinity_mask creation By default, the number of tx queues is limited by the number of online cpus in mlx4_en_get_profile(). However, this limit no longer holds after the ethtool .set_channels method has been called. In that situation, the driver may access invalid bits of certain cpumask variables when queue_index >= nr_cpu_ids. Signed-off-by: Benjamin Poirier Acked-by: Ido Shamay Fixes: d03a68f ("net/mlx4_en: Configure the XPS queue mapping on driver load") Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 1783705..f7bf312 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -143,8 +143,10 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, ring->hwtstamp_tx_type = priv->hwtstamp_config.tx_type; ring->queue_index = queue_index; - if (queue_index < priv->num_tx_rings_p_up && cpu_online(queue_index)) - cpumask_set_cpu(queue_index, &ring->affinity_mask); + if (queue_index < priv->num_tx_rings_p_up) + cpumask_set_cpu_local_first(queue_index, + priv->mdev->dev->numa_node, + &ring->affinity_mask); *pring = ring; return 0; @@ -213,7 +215,7 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv, err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context, &ring->qp, &ring->qp_state); - if (!user_prio && cpu_online(ring->queue_index)) + if (!cpumask_empty(&ring->affinity_mask)) netif_set_xps_queue(priv->dev, &ring->affinity_mask, ring->queue_index); -- cgit v0.10.2 From 48734994ac268eb391a66dca4cde7d08a08aba08 Mon Sep 17 00:00:00 2001 From: Vlastimil Setka Date: Wed, 29 Apr 2015 00:17:11 +0200 Subject: altera_tse: Correct rx packet length Altera TSE MAC rx DMA transfer starts with the 2 additional bytes for IP payload alignment. This patch fixes tse_rx() function loop which reads DMA rx status and extracts packet length from it. Status signalises a whole DMA transfer length, which is 2 bytes longer than the packet itself. Signed-off-by: Vlastimil Setka Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 0533c05..da48e66 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -391,6 +391,12 @@ static int tse_rx(struct altera_tse_private *priv, int limit) "RCV pktstatus %08X pktlength %08X\n", pktstatus, pktlength); + /* DMA trasfer from TSE starts with 2 aditional bytes for + * IP payload alignment. Status returned by get_rx_status() + * contains DMA transfer length. Packet is 2 bytes shorter. + */ + pktlength -= 2; + count++; next_entry = (++priv->rx_cons) % priv->rx_ring_size; -- cgit v0.10.2 From b56fc3c536541c8081cd5f1f1d101a16c002a365 Mon Sep 17 00:00:00 2001 From: KY Srinivasan Date: Tue, 28 Apr 2015 17:59:48 -0700 Subject: hv_netvsc: Fix a bug in netvsc_start_xmit() Commit b08cc79155fc26d0d112b1470d1ece5034651a4b eliminated memory allocation in the packet send path: "hv_netvsc: Eliminate memory allocation in the packet send path The network protocol used to communicate with the host is the remote ndis (rndis) protocol. We need to decorate each outgoing packet with a rndis header and additional rndis state (rndis per-packet state). To manage this state, we currently allocate memory in the transmit path. Eliminate this allocation by requesting additional head room in the skb." This commit introduced a bug since it did not account for the case if the skb was cloned. Fix this bug. Signed-off-by: K. Y. Srinivasan Tested-by: Dexuan Cui Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index e55c8f4f..41071d3 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -128,7 +128,6 @@ struct ndis_tcp_ip_checksum_info; struct hv_netvsc_packet { /* Bookkeeping stuff */ u32 status; - bool part_of_skb; bool is_data_pkt; bool xmit_more; /* from skb */ diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index c651d4d..2d9ef53 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -889,11 +889,6 @@ int netvsc_send(struct hv_device *device, } else { packet->page_buf_cnt = 0; packet->total_data_buflen += msd_len; - if (!packet->part_of_skb) { - skb = (struct sk_buff *)(unsigned long)packet-> - send_completion_tid; - packet->send_completion_tid = 0; - } } if (msdp->pkt) diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 66c4b0c..5993c7e 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -238,9 +238,6 @@ void netvsc_xmit_completion(void *context) struct sk_buff *skb = (struct sk_buff *) (unsigned long)packet->send_completion_tid; - if (!packet->part_of_skb) - kfree(packet); - if (skb) dev_kfree_skb_any(skb); } @@ -392,7 +389,6 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net) u32 net_trans_info; u32 hash; u32 skb_length; - u32 head_room; u32 pkt_sz; struct hv_page_buffer page_buf[MAX_PAGE_BUFFER_COUNT]; @@ -405,7 +401,6 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net) check_size: skb_length = skb->len; - head_room = skb_headroom(skb); num_data_pgs = netvsc_get_slots(skb) + 2; if (num_data_pgs > MAX_PAGE_BUFFER_COUNT && linear) { net_alert_ratelimited("packet too big: %u pages (%u bytes)\n", @@ -424,20 +419,14 @@ check_size: pkt_sz = sizeof(struct hv_netvsc_packet) + RNDIS_AND_PPI_SIZE; - if (head_room < pkt_sz) { - packet = kmalloc(pkt_sz, GFP_ATOMIC); - if (!packet) { - /* out of memory, drop packet */ - netdev_err(net, "unable to alloc hv_netvsc_packet\n"); - ret = -ENOMEM; - goto drop; - } - packet->part_of_skb = false; - } else { - /* Use the headroom for building up the packet */ - packet = (struct hv_netvsc_packet *)skb->head; - packet->part_of_skb = true; + ret = skb_cow_head(skb, pkt_sz); + if (ret) { + netdev_err(net, "unable to alloc hv_netvsc_packet\n"); + ret = -ENOMEM; + goto drop; } + /* Use the headroom for building up the packet */ + packet = (struct hv_netvsc_packet *)skb->head; packet->status = 0; packet->xmit_more = skb->xmit_more; @@ -594,8 +583,6 @@ drop: net->stats.tx_bytes += skb_length; net->stats.tx_packets++; } else { - if (packet && !packet->part_of_skb) - kfree(packet); if (ret != -EAGAIN) { dev_kfree_skb_any(skb); net->stats.tx_dropped++; -- cgit v0.10.2 From d4c216c54197d741ed8b7ca54f13645dfb3eacde Mon Sep 17 00:00:00 2001 From: Punnaiah Choudary Kalluri Date: Wed, 29 Apr 2015 08:34:46 +0530 Subject: net: macb: Fix race condition in driver when Rx frame is dropped Under heavy Rx load, observed that the Hw is updating the USED bit and it is not updating the received frame status to the BD control field. This could be lack of resources for processing the BDs at high data rates. Driver drops the frame associated with this BD but not clearing the USED bit. So, this is causing hang condition as Hw expects USED bit to be cleared for this BD. Signed-off-by: Punnaiah Choudary Kalluri Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 665c290..4104d49 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -707,6 +707,9 @@ static void gem_rx_refill(struct macb *bp) /* properly align Ethernet header */ skb_reserve(skb, NET_IP_ALIGN); + } else { + bp->rx_ring[entry].addr &= ~MACB_BIT(RX_USED); + bp->rx_ring[entry].ctrl = 0; } } -- cgit v0.10.2 From 50d4964f1d0411d82cca593f2664bfab7f82dbbf Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Wed, 29 Apr 2015 10:56:15 -0700 Subject: net: dsa: Fix scope of eeprom-length property eeprom-length is a switch property, not a dsa property, and thus needs to be attached to the switch node, not to the dsa node. Reported-by: Andrew Lunn Fixes: 6793abb4e849 ("net: dsa: Add support for switch EEPROM access") Signed-off-by: Guenter Roeck Acked-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 079a224..e6f6cc3 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -633,7 +633,7 @@ static int dsa_of_probe(struct device *dev) if (cd->sw_addr > PHY_MAX_ADDR) continue; - if (!of_property_read_u32(np, "eeprom-length", &eeprom_len)) + if (!of_property_read_u32(child, "eeprom-length", &eeprom_len)) cd->eeprom_len = eeprom_len; for_each_available_child_of_node(child, port) { -- cgit v0.10.2 From e913fb279c564f2af05658b3f01041757d2e9303 Mon Sep 17 00:00:00 2001 From: Pai Date: Wed, 29 Apr 2015 14:24:23 -0400 Subject: net: Fix Kernel Panic in bonding driver debugfs file: rlb_hash_table This patch fixes a Kernel Panic in bonding driver debugfs file: rlb_hash_table. $> modprobe bonding mode=6 $> cat /sys/kernel/debug/bonding/bond0/rlb_hash_table This will crash the kernel. The struct alb_bond_info is initialized only when the bonding interface is initialized (ip link set bond0 up) and not at the time it is allocated. If we try to read the table before that, it'll result in a kernel panic. The patch applies against both net and net-next Signed-off-by: Vishwanath Pai Signed-off-by: Andy Gospodarek Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 3a10551..d5fe5d5 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4544,6 +4544,8 @@ unsigned int bond_get_num_tx_queues(void) int bond_create(struct net *net, const char *name) { struct net_device *bond_dev; + struct bonding *bond; + struct alb_bond_info *bond_info; int res; rtnl_lock(); @@ -4557,6 +4559,14 @@ int bond_create(struct net *net, const char *name) return -ENOMEM; } + /* + * Initialize rx_hashtbl_used_head to RLB_NULL_INDEX. + * It is set to 0 by default which is wrong. + */ + bond = netdev_priv(bond_dev); + bond_info = &(BOND_ALB_INFO(bond)); + bond_info->rx_hashtbl_used_head = RLB_NULL_INDEX; + dev_net_set(bond_dev, net); bond_dev->rtnl_link_ops = &bond_link_ops; -- cgit v0.10.2 From 12a8541d5c82d17d42424d46ec36929cfef06a10 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Wed, 29 Apr 2015 08:09:49 +0300 Subject: bnx2x: Delay during kdump load In a kdump environment interfaces might be re-loaded without a proper unload sequence in the previous running kernel. bnx2x management FW and driver maintains a `pulse' that notifies the FW that the driver is still up and running. Driver load on the kdump kernel should be performed only after the pulse has been out-of-sync long enough for the management FW to identify that the driver has crashed, on which point it will perform some necessary cleanup of the HW. In today's distros kdump loading is quite fast, sometimes too fast for our FW to get out-of-sync. This patch delays the bnx2x's probe during kdump to allow a proper re-load on the kdump kernel. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 60d2b8e..556dcc1 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -13368,6 +13368,12 @@ static int bnx2x_init_one(struct pci_dev *pdev, bool is_vf; int cnic_cnt; + /* Management FW 'remembers' living interfaces. Allow it some time + * to forget previously living interfaces, allowing a proper re-load. + */ + if (is_kdump_kernel()) + msleep(5000); + /* An estimated maximum supported CoS number according to the chip * version. * We will try to roughly estimate the maximum number of CoSes this chip -- cgit v0.10.2 From 7f0b8a56c978b0a3315ac84c6cbb065413afb8e9 Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Wed, 29 Apr 2015 17:19:05 +0530 Subject: cxgb4: Fix MC1 memory offset calculation Commit 6559a7e8296002b4 ("cxgb4: Cleanup macros so they follow the same style and look consistent") introduced a regression where reading MC1 memory in adapters where MC0 isn't present or MC0 size is not equal to MC1 size caused the adapter to crash due to incorrect computation of memoffset. Fix is to read the size of MC0 instead of MC1 for offset calculation Signed-off-by: Steve Wise Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index 5959e3a..e8578a7 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -492,7 +492,7 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr, memoffset = (mtype * (edc_size * 1024 * 1024)); else { mc_size = EXT_MEM0_SIZE_G(t4_read_reg(adap, - MA_EXT_MEMORY1_BAR_A)); + MA_EXT_MEMORY0_BAR_A)); memoffset = (MEM_MC0 * edc_size + mc_size) * 1024 * 1024; } -- cgit v0.10.2 From 5d2361db48899789fb466ff62db5d5fc7b070e86 Mon Sep 17 00:00:00 2001 From: Forrest Liu Date: Mon, 9 Feb 2015 17:31:45 +0800 Subject: Btrfs: btrfs_release_extent_buffer_page didn't free pages of dummy extent btrfs_release_extent_buffer_page() can't handle dummy extent that allocated by btrfs_clone_extent_buffer() properly. That is because reference count of pages that allocated by btrfs_clone_extent_buffer() was 2, 1 by alloc_page(), and another by attach_extent_buffer_page(). Running following command repeatly can check this memory leak problem btrfs inspect-internal inode-resolve 256 /mnt/btrfs Signed-off-by: Chien-Kuan Yeh Signed-off-by: Forrest Liu Reviewed-by: Filipe Manana Tested-by: Filipe Manana Signed-off-by: Chris Mason diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index ea100eb..43af5a6 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4560,36 +4560,37 @@ static void btrfs_release_extent_buffer_page(struct extent_buffer *eb) do { index--; page = eb->pages[index]; - if (page && mapped) { + if (!page) + continue; + if (mapped) spin_lock(&page->mapping->private_lock); + /* + * We do this since we'll remove the pages after we've + * removed the eb from the radix tree, so we could race + * and have this page now attached to the new eb. So + * only clear page_private if it's still connected to + * this eb. + */ + if (PagePrivate(page) && + page->private == (unsigned long)eb) { + BUG_ON(test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)); + BUG_ON(PageDirty(page)); + BUG_ON(PageWriteback(page)); /* - * We do this since we'll remove the pages after we've - * removed the eb from the radix tree, so we could race - * and have this page now attached to the new eb. So - * only clear page_private if it's still connected to - * this eb. + * We need to make sure we haven't be attached + * to a new eb. */ - if (PagePrivate(page) && - page->private == (unsigned long)eb) { - BUG_ON(test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)); - BUG_ON(PageDirty(page)); - BUG_ON(PageWriteback(page)); - /* - * We need to make sure we haven't be attached - * to a new eb. - */ - ClearPagePrivate(page); - set_page_private(page, 0); - /* One for the page private */ - page_cache_release(page); - } - spin_unlock(&page->mapping->private_lock); - - } - if (page) { - /* One for when we alloced the page */ + ClearPagePrivate(page); + set_page_private(page, 0); + /* One for the page private */ page_cache_release(page); } + + if (mapped) + spin_unlock(&page->mapping->private_lock); + + /* One for when we alloced the page */ + page_cache_release(page); } while (index != 0); } -- cgit v0.10.2 From 483d821108791092798f5d230686868112927044 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Tue, 21 Apr 2015 17:42:09 +0200 Subject: gpio: sysfs: fix memory leaks and device hotplug Unregister GPIOs requested through sysfs at chip remove to avoid leaking the associated memory and sysfs entries. The stale sysfs entries prevented the gpio numbers from being exported when the gpio range was later reused (e.g. at device reconnect). This also fixes the related module-reference leak. Note that kernfs makes sure that any on-going sysfs operations finish before the class devices are unregistered and that further accesses fail. The chip exported flag is used to prevent gpiod exports during removal. This also makes it harder to trigger, but does not fix, the related race between gpiochip_remove and export_store, which is really a race with gpiod_request that needs to be addressed separately. Also note that this would prevent the crashes (e.g. NULL-dereferences) at reconnect that affects pre-3.18 kernels, as well as use-after-free on operations on open attribute files on pre-3.14 kernels (prior to kernfs). Fixes: d8f388d8dc8d ("gpio: sysfs interface") Cc: stable # v2.6.27: 01cca93a9491 Signed-off-by: Johan Hovold Signed-off-by: Linus Walleij diff --git a/drivers/gpio/gpiolib-sysfs.c b/drivers/gpio/gpiolib-sysfs.c index 7722ed5..af3bc7a 100644 --- a/drivers/gpio/gpiolib-sysfs.c +++ b/drivers/gpio/gpiolib-sysfs.c @@ -551,6 +551,7 @@ static struct class gpio_class = { */ int gpiod_export(struct gpio_desc *desc, bool direction_may_change) { + struct gpio_chip *chip; unsigned long flags; int status; const char *ioname = NULL; @@ -568,8 +569,16 @@ int gpiod_export(struct gpio_desc *desc, bool direction_may_change) return -EINVAL; } + chip = desc->chip; + mutex_lock(&sysfs_lock); + /* check if chip is being removed */ + if (!chip || !chip->exported) { + status = -ENODEV; + goto fail_unlock; + } + spin_lock_irqsave(&gpio_lock, flags); if (!test_bit(FLAG_REQUESTED, &desc->flags) || test_bit(FLAG_EXPORT, &desc->flags)) { @@ -783,12 +792,15 @@ void gpiochip_unexport(struct gpio_chip *chip) { int status; struct device *dev; + struct gpio_desc *desc; + unsigned int i; mutex_lock(&sysfs_lock); dev = class_find_device(&gpio_class, NULL, chip, match_export); if (dev) { put_device(dev); device_unregister(dev); + /* prevent further gpiod exports */ chip->exported = false; status = 0; } else @@ -797,6 +809,13 @@ void gpiochip_unexport(struct gpio_chip *chip) if (status) chip_dbg(chip, "%s: status %d\n", __func__, status); + + /* unregister gpiod class devices owned by sysfs */ + for (i = 0; i < chip->ngpio; i++) { + desc = &chip->desc[i]; + if (test_and_clear_bit(FLAG_SYSFS, &desc->flags)) + gpiod_free(desc); + } } static int __init gpiolib_sysfs_init(void) -- cgit v0.10.2 From 68fc378ce332cc4efd7f314d3e6e15e83f53ebf2 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Thu, 30 Apr 2015 15:13:14 +1000 Subject: Revert "powerpc/tm: Abort syscalls in active transactions" This reverts commit feba40362b11341bee6d8ed58d54b896abbd9f84. Although the principle of this change is good, the implementation has a few issues. Firstly we can sometimes fail to abort a syscall because r12 may have been clobbered by C code if we went down the virtual CPU accounting path, or if syscall tracing was enabled. Secondly we have decided that it is safer to abort the syscall even earlier in the syscall entry path, so that we avoid the syscall tracing path when we are transactional. So that we have time to thoroughly test those changes we have decided to revert this for this merge window and will merge the fixed version in the next window. NB. Rather than reverting the selftest we just drop tm-syscall from TEST_PROGS so that it's not run by default. Fixes: feba40362b11 ("powerpc/tm: Abort syscalls in active transactions") Signed-off-by: Michael Ellerman diff --git a/Documentation/powerpc/transactional_memory.txt b/Documentation/powerpc/transactional_memory.txt index ba0a2a4..ded6979 100644 --- a/Documentation/powerpc/transactional_memory.txt +++ b/Documentation/powerpc/transactional_memory.txt @@ -74,23 +74,22 @@ Causes of transaction aborts Syscalls ======== -Syscalls made from within an active transaction will not be performed and the -transaction will be doomed by the kernel with the failure code TM_CAUSE_SYSCALL -| TM_CAUSE_PERSISTENT. +Performing syscalls from within transaction is not recommended, and can lead +to unpredictable results. -Syscalls made from within a suspended transaction are performed as normal and -the transaction is not explicitly doomed by the kernel. However, what the -kernel does to perform the syscall may result in the transaction being doomed -by the hardware. The syscall is performed in suspended mode so any side -effects will be persistent, independent of transaction success or failure. No -guarantees are provided by the kernel about which syscalls will affect -transaction success. +Syscalls do not by design abort transactions, but beware: The kernel code will +not be running in transactional state. The effect of syscalls will always +remain visible, but depending on the call they may abort your transaction as a +side-effect, read soon-to-be-aborted transactional data that should not remain +invisible, etc. If you constantly retry a transaction that constantly aborts +itself by calling a syscall, you'll have a livelock & make no progress. -Care must be taken when relying on syscalls to abort during active transactions -if the calls are made via a library. Libraries may cache values (which may -give the appearance of success) or perform operations that cause transaction -failure before entering the kernel (which may produce different failure codes). -Examples are glibc's getpid() and lazy symbol resolution. +Simple syscalls (e.g. sigprocmask()) "could" be OK. Even things like write() +from, say, printf() should be OK as long as the kernel does not access any +memory that was accessed transactionally. + +Consider any syscalls that happen to work as debug-only -- not recommended for +production use. Best to queue them up till after the transaction is over. Signals @@ -177,7 +176,8 @@ kernel aborted a transaction: TM_CAUSE_RESCHED Thread was rescheduled. TM_CAUSE_TLBI Software TLB invalid. TM_CAUSE_FAC_UNAV FP/VEC/VSX unavailable trap. - TM_CAUSE_SYSCALL Syscall from active transaction. + TM_CAUSE_SYSCALL Currently unused; future syscalls that must abort + transactions for consistency will use this. TM_CAUSE_SIGNAL Signal delivered. TM_CAUSE_MISC Currently unused. TM_CAUSE_ALIGNMENT Alignment fault. diff --git a/arch/powerpc/include/uapi/asm/tm.h b/arch/powerpc/include/uapi/asm/tm.h index 5047659..5d836b7 100644 --- a/arch/powerpc/include/uapi/asm/tm.h +++ b/arch/powerpc/include/uapi/asm/tm.h @@ -11,7 +11,7 @@ #define TM_CAUSE_RESCHED 0xde #define TM_CAUSE_TLBI 0xdc #define TM_CAUSE_FAC_UNAV 0xda -#define TM_CAUSE_SYSCALL 0xd8 +#define TM_CAUSE_SYSCALL 0xd8 /* future use */ #define TM_CAUSE_MISC 0xd6 /* future use */ #define TM_CAUSE_SIGNAL 0xd4 #define TM_CAUSE_ALIGNMENT 0xd2 diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index 8ca9434..afbc200 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -34,7 +34,6 @@ #include #include #include -#include /* * System calls. @@ -146,24 +145,6 @@ END_FW_FTR_SECTION_IFSET(FW_FEATURE_SPLPAR) andi. r11,r10,_TIF_SYSCALL_DOTRACE bne syscall_dotrace .Lsyscall_dotrace_cont: -#ifdef CONFIG_PPC_TRANSACTIONAL_MEM -BEGIN_FTR_SECTION - b 1f -END_FTR_SECTION_IFCLR(CPU_FTR_TM) - extrdi. r11, r12, 1, (63-MSR_TS_T_LG) /* transaction active? */ - beq+ 1f - - /* Doom the transaction and don't perform the syscall: */ - mfmsr r11 - li r12, 1 - rldimi r11, r12, MSR_TM_LG, 63-MSR_TM_LG - mtmsrd r11, 0 - li r11, (TM_CAUSE_SYSCALL|TM_CAUSE_PERSISTENT) - TABORT(R11) - - b .Lsyscall_exit -1: -#endif cmpldi 0,r0,NR_syscalls bge- syscall_enosys diff --git a/tools/testing/selftests/powerpc/tm/Makefile b/tools/testing/selftests/powerpc/tm/Makefile index 1b616fa..6bff955 100644 --- a/tools/testing/selftests/powerpc/tm/Makefile +++ b/tools/testing/selftests/powerpc/tm/Makefile @@ -1,4 +1,4 @@ -TEST_PROGS := tm-resched-dscr tm-syscall +TEST_PROGS := tm-resched-dscr all: $(TEST_PROGS) -- cgit v0.10.2 From d795ef9aa8311ca3c5158bda1edbcd14479c101c Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 17 Apr 2015 14:41:29 +0100 Subject: arm64: perf: don't warn about missing interrupt-affinity property for PPIs PPIs are affine by nature, so the interrupt-affinity property is not used and therefore we shouldn't print a warning in its absence. Reported-by: Maxime Ripard Reviewed-by: Maxime Ripard Signed-off-by: Will Deacon diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c index 195991d..2a9cbcb 100644 --- a/arch/arm64/kernel/perf_event.c +++ b/arch/arm64/kernel/perf_event.c @@ -1310,7 +1310,7 @@ static const struct of_device_id armpmu_of_device_ids[] = { static int armpmu_device_probe(struct platform_device *pdev) { - int i, *irqs; + int i, irq, *irqs; if (!cpu_pmu) return -ENODEV; @@ -1319,6 +1319,11 @@ static int armpmu_device_probe(struct platform_device *pdev) if (!irqs) return -ENOMEM; + /* Don't bother with PPIs; they're already affine */ + irq = platform_get_irq(pdev, 0); + if (irq >= 0 && irq_is_percpu(irq)) + return 0; + for (i = 0; i < pdev->num_resources; ++i) { struct device_node *dn; int cpu; -- cgit v0.10.2 From 8291fd04d86b97869bd34e796bcac3141b9d5432 Mon Sep 17 00:00:00 2001 From: "Suzuki K. Poulose" Date: Mon, 13 Apr 2015 10:17:55 +0100 Subject: arm64: perf: Fix the pmu node name in warning message With commit d5efd9cc9cf2 ("arm64: pmu: add support for interrupt-affinity property"), we print a warning when we find a PMU SPI with a missing missing interrupt-affinity property in a pmu node. Unfortunately, we pass the wrong (NULL) device node to of_node_full_name, resulting in unhelpful messages such as: hw perfevents: Failed to parse /interrupt-affinity[0] This patch fixes the name to that of the pmu node. Fixes: d5efd9cc9cf2 (arm64: pmu: add support for interrupt-affinity property) Acked-by: Mark Rutland Signed-off-by: Suzuki K. Poulose Signed-off-by: Will Deacon diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c index 2a9cbcb..23f25ac 100644 --- a/arch/arm64/kernel/perf_event.c +++ b/arch/arm64/kernel/perf_event.c @@ -1332,7 +1332,7 @@ static int armpmu_device_probe(struct platform_device *pdev) i); if (!dn) { pr_warn("Failed to parse %s/interrupt-affinity[%d]\n", - of_node_full_name(dn), i); + of_node_full_name(pdev->dev.of_node), i); break; } -- cgit v0.10.2 From 3e6180f0c82b3790a9ec6d13d67aae359bf1ce84 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 30 Apr 2015 10:10:36 -0400 Subject: dm: only initialize the request_queue once Commit bfebd1cdb4 ("dm: add full blk-mq support to request-based DM") didn't properly account for the need to short-circuit re-initializing DM's blk-mq request_queue if it was already initialized. Otherwise, reloading a blk-mq request-based DM table (either manually or via multipathd) resulted in errors, see: https://www.redhat.com/archives/dm-devel/2015-April/msg00132.html Fix is to only initialize the request_queue on the initial table load (when the mapped_device type is assigned). This is better than having dm_init_request_based_blk_mq_queue() return early if the queue was already initialized because it elevates the constraint to a more meaningful location in DM core. As such the pre-existing early return in dm_init_request_based_queue() can now be removed. Fixes: bfebd1cdb4 ("dm: add full blk-mq support to request-based DM") Signed-off-by: Christoph Hellwig Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index c8a18e4..720ceeb 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1298,21 +1298,22 @@ static int table_load(struct dm_ioctl *param, size_t param_size) goto err_unlock_md_type; } - if (dm_get_md_type(md) == DM_TYPE_NONE) + if (dm_get_md_type(md) == DM_TYPE_NONE) { /* Initial table load: acquire type of table. */ dm_set_md_type(md, dm_table_get_type(t)); - else if (dm_get_md_type(md) != dm_table_get_type(t)) { + + /* setup md->queue to reflect md's type (may block) */ + r = dm_setup_md_queue(md); + if (r) { + DMWARN("unable to set up device queue for new table."); + goto err_unlock_md_type; + } + } else if (dm_get_md_type(md) != dm_table_get_type(t)) { DMWARN("can't change device type after initial table load."); r = -EINVAL; goto err_unlock_md_type; } - /* setup md->queue to reflect md's type (may block) */ - r = dm_setup_md_queue(md); - if (r) { - DMWARN("unable to set up device queue for new table."); - goto err_unlock_md_type; - } dm_unlock_md_type(md); /* stage inactive table */ diff --git a/drivers/md/dm.c b/drivers/md/dm.c index f8c7ca3..923496b 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -2662,9 +2662,6 @@ static int dm_init_request_based_queue(struct mapped_device *md) { struct request_queue *q = NULL; - if (md->queue->elevator) - return 0; - /* Fully initialize the queue */ q = blk_init_allocated_queue(md->queue, dm_request_fn, NULL); if (!q) -- cgit v0.10.2 From aa6df8dd28c01d9a3d2cfcfe9dd0a4a334d1cd81 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Wed, 29 Apr 2015 10:48:09 -0400 Subject: dm: fix free_rq_clone() NULL pointer when requeueing unmapped request Commit 022333427a ("dm: optimize dm_mq_queue_rq to _not_ use kthread if using pure blk-mq") mistakenly removed free_rq_clone()'s clone->q check before testing clone->q->mq_ops. It was an oversight to discontinue that check for 1 of the 2 use-cases for free_rq_clone(): 1) free_rq_clone() called when an unmapped original request is requeued 2) free_rq_clone() called in the request-based IO completion path The clone->q check made sense for case #1 but not for #2. However, we cannot just reinstate the check as it'd mask a serious bug in the IO completion case #2 -- no in-flight request should have an uninitialized request_queue (basic block layer refcounting _should_ ensure this). The NULL pointer seen for case #1 is detailed here: https://www.redhat.com/archives/dm-devel/2015-April/msg00160.html Fix this free_rq_clone() NULL pointer by simply checking if the mapped_device's type is DM_TYPE_MQ_REQUEST_BASED (clone's queue is blk-mq) rather than checking clone->q->mq_ops. This avoids the need to dereference clone->q, but a WARN_ON_ONCE is added to let us know if an uninitialized clone request is being completed. Reported-by: Bart Van Assche Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 923496b..a930b72 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1082,18 +1082,26 @@ static void rq_completed(struct mapped_device *md, int rw, bool run_queue) dm_put(md); } -static void free_rq_clone(struct request *clone) +static void free_rq_clone(struct request *clone, bool must_be_mapped) { struct dm_rq_target_io *tio = clone->end_io_data; struct mapped_device *md = tio->md; + WARN_ON_ONCE(must_be_mapped && !clone->q); + blk_rq_unprep_clone(clone); - if (clone->q->mq_ops) + if (md->type == DM_TYPE_MQ_REQUEST_BASED) + /* stacked on blk-mq queue(s) */ tio->ti->type->release_clone_rq(clone); else if (!md->queue->mq_ops) /* request_fn queue stacked on request_fn queue(s) */ free_clone_request(md, clone); + /* + * NOTE: for the blk-mq queue stacked on request_fn queue(s) case: + * no need to call free_clone_request() because we leverage blk-mq by + * allocating the clone at the end of the blk-mq pdu (see: clone_rq) + */ if (!md->queue->mq_ops) free_rq_tio(tio); @@ -1124,7 +1132,7 @@ static void dm_end_request(struct request *clone, int error) rq->sense_len = clone->sense_len; } - free_rq_clone(clone); + free_rq_clone(clone, true); if (!rq->q->mq_ops) blk_end_request_all(rq, error); else @@ -1143,7 +1151,7 @@ static void dm_unprep_request(struct request *rq) } if (clone) - free_rq_clone(clone); + free_rq_clone(clone, false); } /* -- cgit v0.10.2 From 9c4249c8e0221e5cfae758d35b768aee84abf6c0 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 30 Apr 2015 14:58:43 +0100 Subject: modsign: change default key details Change default key details to be more obviously unspecified. Reported-by: Linus Torvalds Signed-off-by: David Howells Acked-by: James Morris Signed-off-by: Linus Torvalds diff --git a/Documentation/module-signing.txt b/Documentation/module-signing.txt index 09c2382..c72702e 100644 --- a/Documentation/module-signing.txt +++ b/Documentation/module-signing.txt @@ -119,9 +119,9 @@ Most notably, in the x509.genkey file, the req_distinguished_name section should be altered from the default: [ req_distinguished_name ] - O = Magrathea - CN = Glacier signing key - emailAddress = slartibartfast@magrathea.h2g2 + #O = Unspecified company + CN = Build time autogenerated kernel key + #emailAddress = unspecified.user@unspecified.company The generated RSA key size can also be set with: diff --git a/kernel/Makefile b/kernel/Makefile index 0f8f8b0..60c302c 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -197,9 +197,9 @@ x509.genkey: @echo >>x509.genkey "x509_extensions = myexts" @echo >>x509.genkey @echo >>x509.genkey "[ req_distinguished_name ]" - @echo >>x509.genkey "O = Magrathea" - @echo >>x509.genkey "CN = Glacier signing key" - @echo >>x509.genkey "emailAddress = slartibartfast@magrathea.h2g2" + @echo >>x509.genkey "#O = Unspecified company" + @echo >>x509.genkey "CN = Build time autogenerated kernel key" + @echo >>x509.genkey "#emailAddress = unspecified.user@unspecified.company" @echo >>x509.genkey @echo >>x509.genkey "[ myexts ]" @echo >>x509.genkey "basicConstraints=critical,CA:FALSE" -- cgit v0.10.2 From 27796e755a2fc6707cdf9b484a56395f703e4f3d Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Thu, 30 Apr 2015 11:57:41 +0300 Subject: ASoC: davinci-mcasp: Correct pm status check in suspend callback pm_runtime_enabled() will only tell if the pm runtime has been enabled for the device, which is done at probe time but will not tell the actual power state of the device. pm_runtime_active() provides this information. This patch fixes a kernel crash when doing suspend when McASP is not active. Signed-off-by: Peter Ujfalusi Signed-off-by: Mark Brown diff --git a/sound/soc/davinci/davinci-mcasp.c b/sound/soc/davinci/davinci-mcasp.c index bb4b78e..23c91fa 100644 --- a/sound/soc/davinci/davinci-mcasp.c +++ b/sound/soc/davinci/davinci-mcasp.c @@ -1247,7 +1247,7 @@ static int davinci_mcasp_suspend(struct snd_soc_dai *dai) u32 reg; int i; - context->pm_state = pm_runtime_enabled(mcasp->dev); + context->pm_state = pm_runtime_active(mcasp->dev); if (!context->pm_state) pm_runtime_get_sync(mcasp->dev); -- cgit v0.10.2 From 2c62e8492ed7358bbe7da51666c7e0f6da9474ee Mon Sep 17 00:00:00 2001 From: Jiang Liu Date: Thu, 30 Apr 2015 12:41:28 +0800 Subject: x86/PCI/ACPI: Make all resources except [io 0xcf8-0xcff] available on PCI bus An IO port or MMIO resource assigned to a PCI host bridge may be consumed by the host bridge itself or available to its child bus/devices. The ACPI specification defines a bit (Producer/Consumer) to tell whether the resource is consumed by the host bridge itself, but firmware hasn't used that bit consistently, so we can't rely on it. Before commit 593669c2ac0f ("x86/PCI/ACPI: Use common ACPI resource interfaces to simplify implementation"), arch/x86/pci/acpi.c ignored all IO port resources defined by acpi_resource_io and acpi_resource_fixed_io to filter out IO ports consumed by the host bridge itself. Commit 593669c2ac0f ("x86/PCI/ACPI: Use common ACPI resource interfaces to simplify implementation") started accepting all IO port and MMIO resources, which caused a regression that IO port resources consumed by the host bridge itself became available to its child devices. Then commit 63f1789ec716 ("x86/PCI/ACPI: Ignore resources consumed by host bridge itself") ignored resources consumed by the host bridge itself by checking the IORESOURCE_WINDOW flag, which accidently removed MMIO resources defined by acpi_resource_memory24, acpi_resource_memory32 and acpi_resource_fixed_memory32. On x86 and IA64 platforms, all IO port and MMIO resources are assumed to be available to child bus/devices except one special case: IO port [0xCF8-0xCFF] is consumed by the host bridge itself to access PCI configuration space. So explicitly filter out PCI CFG IO ports[0xCF8-0xCFF]. This solution will also ease the way to consolidate ACPI PCI host bridge common code from x86, ia64 and ARM64. Related ACPI table are archived at: https://bugzilla.kernel.org/show_bug.cgi?id=94221 Related discussions at: http://patchwork.ozlabs.org/patch/461633/ https://lkml.org/lkml/2015/3/29/304 Fixes: 63f1789ec716 (Ignore resources consumed by host bridge itself) Reported-by: Bernhard Thaler Signed-off-by: Jiang Liu Cc: 4.0+ # 4.0+ Reviewed-by: Bjorn Helgaas Signed-off-by: Rafael J. Wysocki diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index e469598..d939633 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -325,6 +325,26 @@ static void release_pci_root_info(struct pci_host_bridge *bridge) kfree(info); } +/* + * An IO port or MMIO resource assigned to a PCI host bridge may be + * consumed by the host bridge itself or available to its child + * bus/devices. The ACPI specification defines a bit (Producer/Consumer) + * to tell whether the resource is consumed by the host bridge itself, + * but firmware hasn't used that bit consistently, so we can't rely on it. + * + * On x86 and IA64 platforms, all IO port and MMIO resources are assumed + * to be available to child bus/devices except one special case: + * IO port [0xCF8-0xCFF] is consumed by the host bridge itself + * to access PCI configuration space. + * + * So explicitly filter out PCI CFG IO ports[0xCF8-0xCFF]. + */ +static bool resource_is_pcicfg_ioport(struct resource *res) +{ + return (res->flags & IORESOURCE_IO) && + res->start == 0xCF8 && res->end == 0xCFF; +} + static void probe_pci_root_info(struct pci_root_info *info, struct acpi_device *device, int busnum, int domain, @@ -346,8 +366,8 @@ static void probe_pci_root_info(struct pci_root_info *info, "no IO and memory resources present in _CRS\n"); else resource_list_for_each_entry_safe(entry, tmp, list) { - if ((entry->res->flags & IORESOURCE_WINDOW) == 0 || - (entry->res->flags & IORESOURCE_DISABLED)) + if ((entry->res->flags & IORESOURCE_DISABLED) || + resource_is_pcicfg_ioport(entry->res)) resource_list_destroy_entry(entry); else entry->res->name = info->name; diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index 5589a6e..8244f01 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -573,7 +573,7 @@ EXPORT_SYMBOL_GPL(acpi_dev_get_resources); * @ares: Input ACPI resource object. * @types: Valid resource types of IORESOURCE_XXX * - * This is a hepler function to support acpi_dev_get_resources(), which filters + * This is a helper function to support acpi_dev_get_resources(), which filters * ACPI resource objects according to resource types. */ int acpi_dev_filter_resource_type(struct acpi_resource *ares, -- cgit v0.10.2 From f94813f3c1d02090cc02dcfcbed339897830acb8 Mon Sep 17 00:00:00 2001 From: Benjamin Poirier Date: Wed, 29 Apr 2015 15:59:35 -0700 Subject: mlx4_en: Use correct loop cursor in error path. Signed-off-by: Benjamin Poirier Fixes: 9e311e7 ("net/mlx4_en: Use affinity hint") Acked-by: Amir Vadai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 0f1afc0..bf173d7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -1721,7 +1721,7 @@ mac_err: cq_err: while (rx_index--) { mlx4_en_deactivate_cq(priv, priv->rx_cq[rx_index]); - mlx4_en_free_affinity_hint(priv, i); + mlx4_en_free_affinity_hint(priv, rx_index); } for (i = 0; i < priv->rx_ring_num; i++) mlx4_en_deactivate_rx_ring(priv, priv->rx_ring[i]); -- cgit v0.10.2 From 17d5ceb6e43ea545d6d92db2f3ddb035233ba335 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 29 Apr 2015 16:52:51 -0400 Subject: net/mlx4_core: Fix unaligned accesses Addresses the following kernel logs seen during boot: Kernel unaligned access at TPC[100ee150] mlx4_QUERY_HCA+0x80/0x248 [mlx4_core] Kernel unaligned access at TPC[100f071c] mlx4_QUERY_ADAPTER+0x100/0x12c [mlx4_core] Kernel unaligned access at TPC[100f071c] mlx4_QUERY_ADAPTER+0x100/0x12c [mlx4_core] Kernel unaligned access at TPC[100f071c] mlx4_QUERY_ADAPTER+0x100/0x12c [mlx4_core] Kernel unaligned access at TPC[100f071c] mlx4_QUERY_ADAPTER+0x100/0x12c [mlx4_core] Signed-off-by: David Ahern Acked-by: Or Gerlitz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index a407981..e30bf57 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -56,11 +56,13 @@ MODULE_PARM_DESC(enable_qos, "Enable Enhanced QoS support (default: on)"); #define MLX4_GET(dest, source, offset) \ do { \ void *__p = (char *) (source) + (offset); \ + u64 val; \ switch (sizeof (dest)) { \ case 1: (dest) = *(u8 *) __p; break; \ case 2: (dest) = be16_to_cpup(__p); break; \ case 4: (dest) = be32_to_cpup(__p); break; \ - case 8: (dest) = be64_to_cpup(__p); break; \ + case 8: val = get_unaligned((u64 *)__p); \ + (dest) = be64_to_cpu(val); break; \ default: __buggy_use_of_MLX4_GET(); \ } \ } while (0) @@ -1605,9 +1607,17 @@ static void get_board_id(void *vsd, char *board_id) * swaps each 4-byte word before passing it back to * us. Therefore we need to swab it before printing. */ - for (i = 0; i < 4; ++i) - ((u32 *) board_id)[i] = - swab32(*(u32 *) (vsd + VSD_OFFSET_MLX_BOARD_ID + i * 4)); + u32 *bid_u32 = (u32 *)board_id; + + for (i = 0; i < 4; ++i) { + u32 *addr; + u32 val; + + addr = (u32 *) (vsd + VSD_OFFSET_MLX_BOARD_ID + i * 4); + val = get_unaligned(addr); + val = swab32(val); + put_unaligned(val, &bid_u32[i]); + } } } -- cgit v0.10.2 From c232d8a8bb1416f7ec21bb1aabc7c4ec8a5a899e Mon Sep 17 00:00:00 2001 From: Tony Camuso Date: Thu, 30 Apr 2015 07:51:27 -0400 Subject: netxen_nic: use spin_[un]lock_bh around tx_clean_lock While testing this driver with DEBUG_LOCKDEP and DEBUG_SPINLOCK enabled did not produce any traces, it would be more prudent in the case of tx_clean_lock to use spin_[un]lock_bh, since this lock is manipulated in both the process and softirq contexts. This patch was tested for functionality and regressions with netperf and DEBUG_LOCKDEP and DEBUG_SPINLOCK enabled. Signed-off-by: Tony Camuso Acked-by: Neil Horman Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c index 5c40683..8da7c3f 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_init.c @@ -135,7 +135,7 @@ void netxen_release_tx_buffers(struct netxen_adapter *adapter) int i, j; struct nx_host_tx_ring *tx_ring = adapter->tx_ring; - spin_lock(&adapter->tx_clean_lock); + spin_lock_bh(&adapter->tx_clean_lock); cmd_buf = tx_ring->cmd_buf_arr; for (i = 0; i < tx_ring->num_desc; i++) { buffrag = cmd_buf->frag_array; @@ -159,7 +159,7 @@ void netxen_release_tx_buffers(struct netxen_adapter *adapter) } cmd_buf++; } - spin_unlock(&adapter->tx_clean_lock); + spin_unlock_bh(&adapter->tx_clean_lock); } void netxen_free_sw_resources(struct netxen_adapter *adapter) -- cgit v0.10.2 From 07841f9d94c11afe00c0498cf242edf4075729f4 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 30 Apr 2015 17:32:46 +0300 Subject: net/mlx4_en: Schedule napi when RX buffers allocation fails When system is out of memory, refilling of RX buffers fails while the driver continue to pass the received packets to the kernel stack. At some point, when all RX buffers deplete, driver may fall into a sleep, and not recover when memory for new RX buffers is once again availible. This is because hardware does not have valid descriptors, so no interrupt will be generated for the driver to return to work in napi context. Fix it by schedule the napi poll function from stats_task delayed workqueue, as long as the allocations fail. Signed-off-by: Ido Shamay Signed-off-by: Amir Vadai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index bf173d7..32f5ec7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -1467,6 +1467,7 @@ static void mlx4_en_service_task(struct work_struct *work) if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS) mlx4_en_ptp_overflow_check(mdev); + mlx4_en_recover_from_oom(priv); queue_delayed_work(mdev->workqueue, &priv->service_task, SERVICE_TASK_DELAY); } diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 4fdd3c3..2a77a6b 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -244,6 +244,12 @@ static int mlx4_en_prepare_rx_desc(struct mlx4_en_priv *priv, return mlx4_en_alloc_frags(priv, rx_desc, frags, ring->page_alloc, gfp); } +static inline bool mlx4_en_is_ring_empty(struct mlx4_en_rx_ring *ring) +{ + BUG_ON((u32)(ring->prod - ring->cons) > ring->actual_size); + return ring->prod == ring->cons; +} + static inline void mlx4_en_update_rx_prod_db(struct mlx4_en_rx_ring *ring) { *ring->wqres.db.db = cpu_to_be32(ring->prod & 0xffff); @@ -315,8 +321,7 @@ static void mlx4_en_free_rx_buf(struct mlx4_en_priv *priv, ring->cons, ring->prod); /* Unmap and free Rx buffers */ - BUG_ON((u32) (ring->prod - ring->cons) > ring->actual_size); - while (ring->cons != ring->prod) { + while (!mlx4_en_is_ring_empty(ring)) { index = ring->cons & ring->size_mask; en_dbg(DRV, priv, "Processing descriptor:%d\n", index); mlx4_en_free_rx_desc(priv, ring, index); @@ -491,6 +496,23 @@ err_allocator: return err; } +/* We recover from out of memory by scheduling our napi poll + * function (mlx4_en_process_cq), which tries to allocate + * all missing RX buffers (call to mlx4_en_refill_rx_buffers). + */ +void mlx4_en_recover_from_oom(struct mlx4_en_priv *priv) +{ + int ring; + + if (!priv->port_up) + return; + + for (ring = 0; ring < priv->rx_ring_num; ring++) { + if (mlx4_en_is_ring_empty(priv->rx_ring[ring])) + napi_reschedule(&priv->rx_cq[ring]->napi); + } +} + void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv, struct mlx4_en_rx_ring **pring, u32 size, u16 stride) diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 9de3021..d021f07 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -774,6 +774,7 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv, void mlx4_en_deactivate_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring); void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev); +void mlx4_en_recover_from_oom(struct mlx4_en_priv *priv); int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv, struct mlx4_en_rx_ring **pring, u32 size, u16 stride, int node); -- cgit v0.10.2 From e813bb2b955d9f72c94be7d592746b49929a499b Mon Sep 17 00:00:00 2001 From: Markus Pargmann Date: Thu, 30 Apr 2015 17:07:50 +0200 Subject: net: fec: Fix RGMII-ID mode RGMII-ID uses an internal delay within the transmitter or receiver. This feature is phy specific. The rest of the communication is normal RGMII. So the fec driver has to check for all RGMII modes, not only 'PHY_INTERFACE_MODE_RGMII'. Signed-off-by: Markus Pargmann Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index f6a3a7a..66d47e4 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -988,7 +988,10 @@ fec_restart(struct net_device *ndev) rcntl |= 0x40000000 | 0x00000020; /* RGMII, RMII or MII */ - if (fep->phy_interface == PHY_INTERFACE_MODE_RGMII) + if (fep->phy_interface == PHY_INTERFACE_MODE_RGMII || + fep->phy_interface == PHY_INTERFACE_MODE_RGMII_ID || + fep->phy_interface == PHY_INTERFACE_MODE_RGMII_RXID || + fep->phy_interface == PHY_INTERFACE_MODE_RGMII_TXID) rcntl |= (1 << 6); else if (fep->phy_interface == PHY_INTERFACE_MODE_RMII) rcntl |= (1 << 8); -- cgit v0.10.2 From aa8d6b73ea33c2167c543663ab66039ec94d58f1 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Fri, 24 Apr 2015 22:22:19 +0200 Subject: ARM: pxa: pxa_cplds: add lubbock and mainstone IO Historically, this support was in arch/arm/mach-pxa/lubbock.c and arch/arm/mach-pxa/mainstone.c. When gpio-pxa was moved to drivers/pxa, it became a driver, and its initialization and probing happened at postcore initcall. The lubbock code used to install the chained lubbock interrupt handler at init_irq() time. The consequence of the gpio-pxa change is that the installed chained irq handler lubbock_irq_handler() was overwritten in pxa_gpio_probe(_dt)(), removing : - the handler - the falling edge detection setting of GPIO0, which revealed the interrupt request from the lubbock IO board. As a fix, move the gpio0 chained handler setup to a place where we have the guarantee that pxa_gpio_probe() was called before, so that lubbock handler becomes the true IRQ chained handler of GPIO0, demuxing the lubbock IO board interrupts. This patch moves all that handling to a mfd driver. It's only purpose for the time being is the interrupt handling, but in the future it should encompass all the motherboard CPLDs handling : - leds - switches - hexleds The same logic applies to mainstone board. Fixes: 157d2644cb0c ("ARM: pxa: change gpio to platform device") Signed-off-by: Robert Jarzmik Acked-by: Arnd Bergmann diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig index 8896e71..f096836 100644 --- a/arch/arm/mach-pxa/Kconfig +++ b/arch/arm/mach-pxa/Kconfig @@ -691,4 +691,13 @@ config SHARPSL_PM_MAX1111 config PXA310_ULPI bool +config PXA_SYSTEMS_CPLDS + tristate "Motherboard cplds" + default ARCH_LUBBOCK || MACH_MAINSTONE + help + This driver supports the Lubbock and Mainstone multifunction chip + found on the pxa25x development platform system (Lubbock) and pxa27x + development platform system (Mainstone). This IO board supports the + interrupts handling, ethernet controller, flash chips, etc ... + endif diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile index eb0bf76..4087d33 100644 --- a/arch/arm/mach-pxa/Makefile +++ b/arch/arm/mach-pxa/Makefile @@ -90,4 +90,5 @@ obj-$(CONFIG_MACH_RAUMFELD_CONNECTOR) += raumfeld.o obj-$(CONFIG_MACH_RAUMFELD_SPEAKER) += raumfeld.o obj-$(CONFIG_MACH_ZIPIT2) += z2.o +obj-$(CONFIG_PXA_SYSTEMS_CPLDS) += pxa_cplds_irqs.o obj-$(CONFIG_TOSA_BT) += tosa-bt.o diff --git a/arch/arm/mach-pxa/pxa_cplds_irqs.c b/arch/arm/mach-pxa/pxa_cplds_irqs.c new file mode 100644 index 0000000..f1aeb54 --- /dev/null +++ b/arch/arm/mach-pxa/pxa_cplds_irqs.c @@ -0,0 +1,200 @@ +/* + * Intel Reference Systems cplds + * + * Copyright (C) 2014 Robert Jarzmik + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * Cplds motherboard driver, supporting lubbock and mainstone SoC board. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define FPGA_IRQ_MASK_EN 0x0 +#define FPGA_IRQ_SET_CLR 0x10 + +#define CPLDS_NB_IRQ 32 + +struct cplds { + void __iomem *base; + int irq; + unsigned int irq_mask; + struct gpio_desc *gpio0; + struct irq_domain *irqdomain; +}; + +static irqreturn_t cplds_irq_handler(int in_irq, void *d) +{ + struct cplds *fpga = d; + unsigned long pending; + unsigned int bit; + + pending = readl(fpga->base + FPGA_IRQ_SET_CLR) & fpga->irq_mask; + for_each_set_bit(bit, &pending, CPLDS_NB_IRQ) + generic_handle_irq(irq_find_mapping(fpga->irqdomain, bit)); + + return IRQ_HANDLED; +} + +static void cplds_irq_mask_ack(struct irq_data *d) +{ + struct cplds *fpga = irq_data_get_irq_chip_data(d); + unsigned int cplds_irq = irqd_to_hwirq(d); + unsigned int set, bit = BIT(cplds_irq); + + fpga->irq_mask &= ~bit; + writel(fpga->irq_mask, fpga->base + FPGA_IRQ_MASK_EN); + set = readl(fpga->base + FPGA_IRQ_SET_CLR); + writel(set & ~bit, fpga->base + FPGA_IRQ_SET_CLR); +} + +static void cplds_irq_unmask(struct irq_data *d) +{ + struct cplds *fpga = irq_data_get_irq_chip_data(d); + unsigned int cplds_irq = irqd_to_hwirq(d); + unsigned int bit = BIT(cplds_irq); + + fpga->irq_mask |= bit; + writel(fpga->irq_mask, fpga->base + FPGA_IRQ_MASK_EN); +} + +static struct irq_chip cplds_irq_chip = { + .name = "pxa_cplds", + .irq_mask_ack = cplds_irq_mask_ack, + .irq_unmask = cplds_irq_unmask, + .flags = IRQCHIP_MASK_ON_SUSPEND | IRQCHIP_SKIP_SET_WAKE, +}; + +static int cplds_irq_domain_map(struct irq_domain *d, unsigned int irq, + irq_hw_number_t hwirq) +{ + struct cplds *fpga = d->host_data; + + irq_set_chip_and_handler(irq, &cplds_irq_chip, handle_level_irq); + irq_set_chip_data(irq, fpga); + + return 0; +} + +static const struct irq_domain_ops cplds_irq_domain_ops = { + .xlate = irq_domain_xlate_twocell, + .map = cplds_irq_domain_map, +}; + +static int cplds_resume(struct platform_device *pdev) +{ + struct cplds *fpga = platform_get_drvdata(pdev); + + writel(fpga->irq_mask, fpga->base + FPGA_IRQ_MASK_EN); + + return 0; +} + +static int cplds_probe(struct platform_device *pdev) +{ + struct resource *res; + struct cplds *fpga; + int ret; + unsigned int base_irq = 0; + unsigned long irqflags = 0; + + fpga = devm_kzalloc(&pdev->dev, sizeof(*fpga), GFP_KERNEL); + if (!fpga) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_IRQ, 0); + if (res) { + fpga->irq = (unsigned int)res->start; + irqflags = res->flags; + } + if (!fpga->irq) + return -ENODEV; + + base_irq = platform_get_irq(pdev, 1); + if (base_irq < 0) + base_irq = 0; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + fpga->base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(fpga->base)) + return PTR_ERR(fpga->base); + + platform_set_drvdata(pdev, fpga); + + writel(fpga->irq_mask, fpga->base + FPGA_IRQ_MASK_EN); + writel(0, fpga->base + FPGA_IRQ_SET_CLR); + + ret = devm_request_irq(&pdev->dev, fpga->irq, cplds_irq_handler, + irqflags, dev_name(&pdev->dev), fpga); + if (ret == -ENOSYS) + return -EPROBE_DEFER; + + if (ret) { + dev_err(&pdev->dev, "couldn't request main irq%d: %d\n", + fpga->irq, ret); + return ret; + } + + irq_set_irq_wake(fpga->irq, 1); + fpga->irqdomain = irq_domain_add_linear(pdev->dev.of_node, + CPLDS_NB_IRQ, + &cplds_irq_domain_ops, fpga); + if (!fpga->irqdomain) + return -ENODEV; + + if (base_irq) { + ret = irq_create_strict_mappings(fpga->irqdomain, base_irq, 0, + CPLDS_NB_IRQ); + if (ret) { + dev_err(&pdev->dev, "couldn't create the irq mapping %d..%d\n", + base_irq, base_irq + CPLDS_NB_IRQ); + return ret; + } + } + + return 0; +} + +static int cplds_remove(struct platform_device *pdev) +{ + struct cplds *fpga = platform_get_drvdata(pdev); + + irq_set_chip_and_handler(fpga->irq, NULL, NULL); + + return 0; +} + +static const struct of_device_id cplds_id_table[] = { + { .compatible = "intel,lubbock-cplds-irqs", }, + { .compatible = "intel,mainstone-cplds-irqs", }, + { } +}; +MODULE_DEVICE_TABLE(of, cplds_id_table); + +static struct platform_driver cplds_driver = { + .driver = { + .name = "pxa_cplds_irqs", + .of_match_table = of_match_ptr(cplds_id_table), + }, + .probe = cplds_probe, + .remove = cplds_remove, + .resume = cplds_resume, +}; + +module_platform_driver(cplds_driver); + +MODULE_DESCRIPTION("PXA Cplds interrupts driver"); +MODULE_AUTHOR("Robert Jarzmik "); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 277688639f98a9e34a6f109f9cd6129f92e718c1 Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Fri, 24 Apr 2015 23:22:35 +0200 Subject: ARM: pxa: mainstone: use new pxa_cplds driver As the interrupt handling was transferred to the pxa_cplds driver, make the switch in mainstone platform code. Fixes: 157d2644cb0c ("ARM: pxa: change gpio to platform device") Signed-off-by: Robert Jarzmik Acked-by: Arnd Bergmann diff --git a/arch/arm/mach-pxa/include/mach/mainstone.h b/arch/arm/mach-pxa/include/mach/mainstone.h index 1bfc4e8..e82a7d3 100644 --- a/arch/arm/mach-pxa/include/mach/mainstone.h +++ b/arch/arm/mach-pxa/include/mach/mainstone.h @@ -120,7 +120,9 @@ #define MST_PCMCIA_PWR_VCC_50 0x4 /* voltage VCC = 5.0V */ /* board specific IRQs */ -#define MAINSTONE_IRQ(x) (IRQ_BOARD_START + (x)) +#define MAINSTONE_NR_IRQS IRQ_BOARD_START + +#define MAINSTONE_IRQ(x) (MAINSTONE_NR_IRQS + (x)) #define MAINSTONE_MMC_IRQ MAINSTONE_IRQ(0) #define MAINSTONE_USIM_IRQ MAINSTONE_IRQ(1) #define MAINSTONE_USBC_IRQ MAINSTONE_IRQ(2) @@ -136,6 +138,4 @@ #define MAINSTONE_S1_STSCHG_IRQ MAINSTONE_IRQ(14) #define MAINSTONE_S1_IRQ MAINSTONE_IRQ(15) -#define MAINSTONE_NR_IRQS (IRQ_BOARD_START + 16) - #endif diff --git a/arch/arm/mach-pxa/mainstone.c b/arch/arm/mach-pxa/mainstone.c index 78b84c0..2c0658c 100644 --- a/arch/arm/mach-pxa/mainstone.c +++ b/arch/arm/mach-pxa/mainstone.c @@ -13,6 +13,7 @@ * published by the Free Software Foundation. */ #include +#include #include #include #include @@ -122,92 +123,6 @@ static unsigned long mainstone_pin_config[] = { GPIO1_GPIO | WAKEUP_ON_EDGE_BOTH, }; -static unsigned long mainstone_irq_enabled; - -static void mainstone_mask_irq(struct irq_data *d) -{ - int mainstone_irq = (d->irq - MAINSTONE_IRQ(0)); - MST_INTMSKENA = (mainstone_irq_enabled &= ~(1 << mainstone_irq)); -} - -static void mainstone_unmask_irq(struct irq_data *d) -{ - int mainstone_irq = (d->irq - MAINSTONE_IRQ(0)); - /* the irq can be acknowledged only if deasserted, so it's done here */ - MST_INTSETCLR &= ~(1 << mainstone_irq); - MST_INTMSKENA = (mainstone_irq_enabled |= (1 << mainstone_irq)); -} - -static struct irq_chip mainstone_irq_chip = { - .name = "FPGA", - .irq_ack = mainstone_mask_irq, - .irq_mask = mainstone_mask_irq, - .irq_unmask = mainstone_unmask_irq, -}; - -static void mainstone_irq_handler(unsigned int irq, struct irq_desc *desc) -{ - unsigned long pending = MST_INTSETCLR & mainstone_irq_enabled; - do { - /* clear useless edge notification */ - desc->irq_data.chip->irq_ack(&desc->irq_data); - if (likely(pending)) { - irq = MAINSTONE_IRQ(0) + __ffs(pending); - generic_handle_irq(irq); - } - pending = MST_INTSETCLR & mainstone_irq_enabled; - } while (pending); -} - -static void __init mainstone_init_irq(void) -{ - int irq; - - pxa27x_init_irq(); - - /* setup extra Mainstone irqs */ - for(irq = MAINSTONE_IRQ(0); irq <= MAINSTONE_IRQ(15); irq++) { - irq_set_chip_and_handler(irq, &mainstone_irq_chip, - handle_level_irq); - if (irq == MAINSTONE_IRQ(10) || irq == MAINSTONE_IRQ(14)) - set_irq_flags(irq, IRQF_VALID | IRQF_PROBE | IRQF_NOAUTOEN); - else - set_irq_flags(irq, IRQF_VALID | IRQF_PROBE); - } - set_irq_flags(MAINSTONE_IRQ(8), 0); - set_irq_flags(MAINSTONE_IRQ(12), 0); - - MST_INTMSKENA = 0; - MST_INTSETCLR = 0; - - irq_set_chained_handler(PXA_GPIO_TO_IRQ(0), mainstone_irq_handler); - irq_set_irq_type(PXA_GPIO_TO_IRQ(0), IRQ_TYPE_EDGE_FALLING); -} - -#ifdef CONFIG_PM - -static void mainstone_irq_resume(void) -{ - MST_INTMSKENA = mainstone_irq_enabled; -} - -static struct syscore_ops mainstone_irq_syscore_ops = { - .resume = mainstone_irq_resume, -}; - -static int __init mainstone_irq_device_init(void) -{ - if (machine_is_mainstone()) - register_syscore_ops(&mainstone_irq_syscore_ops); - - return 0; -} - -device_initcall(mainstone_irq_device_init); - -#endif - - static struct resource smc91x_resources[] = { [0] = { .start = (MST_ETH_PHYS + 0x300), @@ -487,11 +402,37 @@ static struct platform_device mst_gpio_keys_device = { }, }; +static struct resource mst_cplds_resources[] = { + [0] = { + .start = MST_FPGA_PHYS + 0xc0, + .end = MST_FPGA_PHYS + 0xe0 - 1, + .flags = IORESOURCE_MEM, + }, + [1] = { + .start = PXA_GPIO_TO_IRQ(0), + .end = PXA_GPIO_TO_IRQ(0), + .flags = IORESOURCE_IRQ | IORESOURCE_IRQ_LOWEDGE, + }, + [2] = { + .start = MAINSTONE_IRQ(0), + .end = MAINSTONE_IRQ(15), + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device mst_cplds_device = { + .name = "pxa_cplds_irqs", + .id = -1, + .resource = &mst_cplds_resources[0], + .num_resources = 3, +}; + static struct platform_device *platform_devices[] __initdata = { &smc91x_device, &mst_flash_device[0], &mst_flash_device[1], &mst_gpio_keys_device, + &mst_cplds_device, }; static struct pxaohci_platform_data mainstone_ohci_platform_data = { @@ -718,7 +659,7 @@ MACHINE_START(MAINSTONE, "Intel HCDDBBVA0 Development Platform (aka Mainstone)") .atag_offset = 0x100, /* BLOB boot parameter setting */ .map_io = mainstone_map_io, .nr_irqs = MAINSTONE_NR_IRQS, - .init_irq = mainstone_init_irq, + .init_irq = pxa27x_init_irq, .handle_irq = pxa27x_handle_irq, .init_time = pxa_timer_init, .init_machine = mainstone_init, -- cgit v0.10.2 From fc9e38c0f4d38bfc68b405cf48365d65f7b6319e Mon Sep 17 00:00:00 2001 From: Robert Jarzmik Date: Sun, 14 Dec 2014 23:04:14 +0100 Subject: ARM: pxa: lubbock: use new pxa_cplds driver As the interrupt handling was transferred to the pxa_cplds driver, make the switch in lubbock platform code. Fixes: 157d2644cb0c ("ARM: pxa: change gpio to platform device") Signed-off-by: Robert Jarzmik Acked-by: Arnd Bergmann diff --git a/arch/arm/mach-pxa/include/mach/lubbock.h b/arch/arm/mach-pxa/include/mach/lubbock.h index 958cd6af..1eecf79 100644 --- a/arch/arm/mach-pxa/include/mach/lubbock.h +++ b/arch/arm/mach-pxa/include/mach/lubbock.h @@ -37,7 +37,9 @@ #define LUB_GP __LUB_REG(LUBBOCK_FPGA_PHYS + 0x100) /* Board specific IRQs */ -#define LUBBOCK_IRQ(x) (IRQ_BOARD_START + (x)) +#define LUBBOCK_NR_IRQS IRQ_BOARD_START + +#define LUBBOCK_IRQ(x) (LUBBOCK_NR_IRQS + (x)) #define LUBBOCK_SD_IRQ LUBBOCK_IRQ(0) #define LUBBOCK_SA1111_IRQ LUBBOCK_IRQ(1) #define LUBBOCK_USB_IRQ LUBBOCK_IRQ(2) /* usb connect */ @@ -47,8 +49,7 @@ #define LUBBOCK_USB_DISC_IRQ LUBBOCK_IRQ(6) /* usb disconnect */ #define LUBBOCK_LAST_IRQ LUBBOCK_IRQ(6) -#define LUBBOCK_SA1111_IRQ_BASE (IRQ_BOARD_START + 16) -#define LUBBOCK_NR_IRQS (IRQ_BOARD_START + 16 + 55) +#define LUBBOCK_SA1111_IRQ_BASE (LUBBOCK_NR_IRQS + 32) #ifndef __ASSEMBLY__ extern void lubbock_set_misc_wr(unsigned int mask, unsigned int set); diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c index d8a1be6..4ac9ab8 100644 --- a/arch/arm/mach-pxa/lubbock.c +++ b/arch/arm/mach-pxa/lubbock.c @@ -12,6 +12,7 @@ * published by the Free Software Foundation. */ #include +#include #include #include #include @@ -123,84 +124,6 @@ void lubbock_set_misc_wr(unsigned int mask, unsigned int set) } EXPORT_SYMBOL(lubbock_set_misc_wr); -static unsigned long lubbock_irq_enabled; - -static void lubbock_mask_irq(struct irq_data *d) -{ - int lubbock_irq = (d->irq - LUBBOCK_IRQ(0)); - LUB_IRQ_MASK_EN = (lubbock_irq_enabled &= ~(1 << lubbock_irq)); -} - -static void lubbock_unmask_irq(struct irq_data *d) -{ - int lubbock_irq = (d->irq - LUBBOCK_IRQ(0)); - /* the irq can be acknowledged only if deasserted, so it's done here */ - LUB_IRQ_SET_CLR &= ~(1 << lubbock_irq); - LUB_IRQ_MASK_EN = (lubbock_irq_enabled |= (1 << lubbock_irq)); -} - -static struct irq_chip lubbock_irq_chip = { - .name = "FPGA", - .irq_ack = lubbock_mask_irq, - .irq_mask = lubbock_mask_irq, - .irq_unmask = lubbock_unmask_irq, -}; - -static void lubbock_irq_handler(unsigned int irq, struct irq_desc *desc) -{ - unsigned long pending = LUB_IRQ_SET_CLR & lubbock_irq_enabled; - do { - /* clear our parent irq */ - desc->irq_data.chip->irq_ack(&desc->irq_data); - if (likely(pending)) { - irq = LUBBOCK_IRQ(0) + __ffs(pending); - generic_handle_irq(irq); - } - pending = LUB_IRQ_SET_CLR & lubbock_irq_enabled; - } while (pending); -} - -static void __init lubbock_init_irq(void) -{ - int irq; - - pxa25x_init_irq(); - - /* setup extra lubbock irqs */ - for (irq = LUBBOCK_IRQ(0); irq <= LUBBOCK_LAST_IRQ; irq++) { - irq_set_chip_and_handler(irq, &lubbock_irq_chip, - handle_level_irq); - set_irq_flags(irq, IRQF_VALID | IRQF_PROBE); - } - - irq_set_chained_handler(PXA_GPIO_TO_IRQ(0), lubbock_irq_handler); - irq_set_irq_type(PXA_GPIO_TO_IRQ(0), IRQ_TYPE_EDGE_FALLING); -} - -#ifdef CONFIG_PM - -static void lubbock_irq_resume(void) -{ - LUB_IRQ_MASK_EN = lubbock_irq_enabled; -} - -static struct syscore_ops lubbock_irq_syscore_ops = { - .resume = lubbock_irq_resume, -}; - -static int __init lubbock_irq_device_init(void) -{ - if (machine_is_lubbock()) { - register_syscore_ops(&lubbock_irq_syscore_ops); - return 0; - } - return -ENODEV; -} - -device_initcall(lubbock_irq_device_init); - -#endif - static int lubbock_udc_is_connected(void) { return (LUB_MISC_RD & (1 << 9)) == 0; @@ -383,11 +306,38 @@ static struct platform_device lubbock_flash_device[2] = { }, }; +static struct resource lubbock_cplds_resources[] = { + [0] = { + .start = LUBBOCK_FPGA_PHYS + 0xc0, + .end = LUBBOCK_FPGA_PHYS + 0xe0 - 1, + .flags = IORESOURCE_MEM, + }, + [1] = { + .start = PXA_GPIO_TO_IRQ(0), + .end = PXA_GPIO_TO_IRQ(0), + .flags = IORESOURCE_IRQ | IORESOURCE_IRQ_LOWEDGE, + }, + [2] = { + .start = LUBBOCK_IRQ(0), + .end = LUBBOCK_IRQ(6), + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device lubbock_cplds_device = { + .name = "pxa_cplds_irqs", + .id = -1, + .resource = &lubbock_cplds_resources[0], + .num_resources = 3, +}; + + static struct platform_device *devices[] __initdata = { &sa1111_device, &smc91x_device, &lubbock_flash_device[0], &lubbock_flash_device[1], + &lubbock_cplds_device, }; static struct pxafb_mode_info sharp_lm8v31_mode = { @@ -648,7 +598,7 @@ MACHINE_START(LUBBOCK, "Intel DBPXA250 Development Platform (aka Lubbock)") /* Maintainer: MontaVista Software Inc. */ .map_io = lubbock_map_io, .nr_irqs = LUBBOCK_NR_IRQS, - .init_irq = lubbock_init_irq, + .init_irq = pxa25x_init_irq, .handle_irq = pxa25x_handle_irq, .init_time = pxa_timer_init, .init_machine = lubbock_init, -- cgit v0.10.2 From 3349fb64b2927407017d970dd5c4daf3c5ad69f8 Mon Sep 17 00:00:00 2001 From: Chris Bainbridge Date: Wed, 29 Apr 2015 21:21:40 +0100 Subject: ACPI / SBS: Add 5 us delay to fix SBS hangs on MacBook Commit 7bc5a2bad0b8 'ACPI: Support _OSI("Darwin") correctly' caused the MacBook firmware to expose the SBS, resulting in intermittent hangs of several minutes on boot, and failure to detect or report the battery. Fix this by adding a 5 us delay to the start of each SMBUS transaction. This timing is the result of experimentation - hangs were observed with 3 us but never with 5 us. Fixes: 7bc5a2bad0b8 'ACPI: Support _OSI("Darwin") correctly' Link: https://bugzilla.kernel.org/show_bug.cgi?id=94651 Signed-off-by: Chris Bainbridge Cc: 3.18+ # 3.18+ [ rjw: Subject and changelog ] Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/sbshc.c b/drivers/acpi/sbshc.c index 26e5b50..bf034f8 100644 --- a/drivers/acpi/sbshc.c +++ b/drivers/acpi/sbshc.c @@ -14,6 +14,7 @@ #include #include #include +#include #include "sbshc.h" #define PREFIX "ACPI: " @@ -87,6 +88,8 @@ enum acpi_smb_offset { ACPI_SMB_ALARM_DATA = 0x26, /* 2 bytes alarm data */ }; +static bool macbook; + static inline int smb_hc_read(struct acpi_smb_hc *hc, u8 address, u8 *data) { return ec_read(hc->offset + address, data); @@ -132,6 +135,8 @@ static int acpi_smbus_transaction(struct acpi_smb_hc *hc, u8 protocol, } mutex_lock(&hc->lock); + if (macbook) + udelay(5); if (smb_hc_read(hc, ACPI_SMB_PROTOCOL, &temp)) goto end; if (temp) { @@ -257,12 +262,29 @@ extern int acpi_ec_add_query_handler(struct acpi_ec *ec, u8 query_bit, acpi_handle handle, acpi_ec_query_func func, void *data); +static int macbook_dmi_match(const struct dmi_system_id *d) +{ + pr_debug("Detected MacBook, enabling workaround\n"); + macbook = true; + return 0; +} + +static struct dmi_system_id acpi_smbus_dmi_table[] = { + { macbook_dmi_match, "Apple MacBook", { + DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBook") }, + }, + { }, +}; + static int acpi_smbus_hc_add(struct acpi_device *device) { int status; unsigned long long val; struct acpi_smb_hc *hc; + dmi_check_system(acpi_smbus_dmi_table); + if (!device) return -EINVAL; -- cgit v0.10.2 From 2375a212ca06d35f90841bc511b3e9ae8a95a82e Mon Sep 17 00:00:00 2001 From: Antonio Ospite Date: Wed, 29 Apr 2015 10:37:24 +0200 Subject: ACPI / documentation: fix a sentence about GPIO resources The sentence "These resources are used be used to pass ..." contains a suspicious repetition, likely the author meant "These resources can be used to pass ...". Simplify the wording. Signed-off-by: Antonio Ospite Signed-off-by: Rafael J. Wysocki diff --git a/Documentation/acpi/enumeration.txt b/Documentation/acpi/enumeration.txt index 750401f..15dfce7 100644 --- a/Documentation/acpi/enumeration.txt +++ b/Documentation/acpi/enumeration.txt @@ -253,7 +253,7 @@ input driver: GPIO support ~~~~~~~~~~~~ ACPI 5 introduced two new resources to describe GPIO connections: GpioIo -and GpioInt. These resources are used be used to pass GPIO numbers used by +and GpioInt. These resources can be used to pass GPIO numbers used by the device to the driver. ACPI 5.1 extended this with _DSD (Device Specific Data) which made it possible to name the GPIOs among other things. -- cgit v0.10.2 From 12392f5896c9affc03c1ddd95a83c32238862a58 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Wed, 29 Apr 2015 16:55:43 +0300 Subject: ARM: shmobile: koelsch: Fix adv7511 IRQ sensing The adv7511 IRQ is low level triggered, not falling edge triggered. The wrong sense configuration results in no interrupt being triggered at all, breaking hotplug detection. Fix it. Signed-off-by: Laurent Pinchart Fixes: 83a0731b39f3 ("ARM: shmobile: koelsch: Add DU HDMI output support") Signed-off-by: Simon Horman diff --git a/arch/arm/boot/dts/r8a7791-koelsch.dts b/arch/arm/boot/dts/r8a7791-koelsch.dts index 74c3212..824ddab 100644 --- a/arch/arm/boot/dts/r8a7791-koelsch.dts +++ b/arch/arm/boot/dts/r8a7791-koelsch.dts @@ -545,7 +545,7 @@ compatible = "adi,adv7511w"; reg = <0x39>; interrupt-parent = <&gpio3>; - interrupts = <29 IRQ_TYPE_EDGE_FALLING>; + interrupts = <29 IRQ_TYPE_LEVEL_LOW>; adi,input-depth = <8>; adi,input-colorspace = "rgb"; -- cgit v0.10.2 From 2fa30fe957f3b3599a2f75688405495771877c14 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 1 May 2015 11:10:09 +1000 Subject: selftests/powerpc: Fix the pmu install rule My patch to add install support for the powerpc selftests had a typo, leading to the three tests in the pmu directory itself not being installed. Fixes: 6faeeea44b84 ("selftests: Add install support for the powerpc tests") Signed-off-by: Michael Ellerman diff --git a/tools/testing/selftests/powerpc/pmu/Makefile b/tools/testing/selftests/powerpc/pmu/Makefile index 5a16117..a9099d9 100644 --- a/tools/testing/selftests/powerpc/pmu/Makefile +++ b/tools/testing/selftests/powerpc/pmu/Makefile @@ -26,7 +26,7 @@ override define EMIT_TESTS $(MAKE) -s -C ebb emit_tests endef -DEFAULT_INSTALL := $(INSTALL_RULE) +DEFAULT_INSTALL_RULE := $(INSTALL_RULE) override define INSTALL_RULE $(DEFAULT_INSTALL_RULE) $(MAKE) -C ebb install -- cgit v0.10.2 From f32393c943e297b8ae180c8f83d81a156c7d0412 Mon Sep 17 00:00:00 2001 From: Nathan Fontenot Date: Wed, 29 Apr 2015 20:42:06 -0500 Subject: powerpc/pseries: Correct cpu affinity for dlpar added cpus The incorrect ordering of operations during cpu dlpar add results in invalid affinity for the cpu being added. The ibm,associativity property in the device tree is populated with all zeroes for the added cpu which results in invalid affinity mappings and all cpus appear to belong to node 0. This occurs because rtas configure-connector is called prior to making the rtas set-indicator calls. Phyp does not assign affinity information for a cpu until the rtas set-indicator calls are made to set the isolation and allocation state. Correct the order of operations to make the rtas set-indicator calls (done in dlpar_acquire_drc) before calling rtas configure-connector. Fixes: 1a8061c46c46 ("powerpc/pseries: Add kernel based CPU DLPAR handling") Signed-off-by: Nathan Fontenot Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/platforms/pseries/dlpar.c b/arch/powerpc/platforms/pseries/dlpar.c index b4b1109..019d34a 100644 --- a/arch/powerpc/platforms/pseries/dlpar.c +++ b/arch/powerpc/platforms/pseries/dlpar.c @@ -412,6 +412,10 @@ static ssize_t dlpar_cpu_probe(const char *buf, size_t count) if (rc) return -EINVAL; + rc = dlpar_acquire_drc(drc_index); + if (rc) + return -EINVAL; + parent = of_find_node_by_path("/cpus"); if (!parent) return -ENODEV; @@ -422,12 +426,6 @@ static ssize_t dlpar_cpu_probe(const char *buf, size_t count) of_node_put(parent); - rc = dlpar_acquire_drc(drc_index); - if (rc) { - dlpar_free_cc_nodes(dn); - return -EINVAL; - } - rc = dlpar_attach_node(dn); if (rc) { dlpar_release_drc(drc_index); -- cgit v0.10.2 From 1ae79b78bc52b910a224f3795122538516e07b5f Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Fri, 1 May 2015 09:14:11 +1000 Subject: powerpc/eeh: Fix race condition in pcibios_set_pcie_reset_state() When asserting reset in pcibios_set_pcie_reset_state(), the PE is enforced to (hardware) frozen state in order to drop unexpected PCI transactions (except PCI config read/write) automatically by hardware during reset, which would cause recursive EEH error. However, the (software) frozen state EEH_PE_ISOLATED is missed. When users get 0xFF from PCI config or MMIO read, EEH_PE_ISOLATED is set in PE state retrival backend. Unfortunately, nobody (the reset handler or the EEH recovery functinality in host) will clear EEH_PE_ISOLATED when the PE has been passed through to guest. The patch sets and clears EEH_PE_ISOLATED properly during reset in function pcibios_set_pcie_reset_state() to fix the issue. Fixes: 28158cd ("Enhance pcibios_set_pcie_reset_state()") Reported-by: Carol L. Soto Signed-off-by: Gavin Shan Tested-by: Carol L. Soto Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/kernel/eeh.c b/arch/powerpc/kernel/eeh.c index 44b480e..fa046ca 100644 --- a/arch/powerpc/kernel/eeh.c +++ b/arch/powerpc/kernel/eeh.c @@ -749,21 +749,24 @@ int pcibios_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state stat eeh_unfreeze_pe(pe, false); eeh_pe_state_clear(pe, EEH_PE_CFG_BLOCKED); eeh_pe_dev_traverse(pe, eeh_restore_dev_state, dev); + eeh_pe_state_clear(pe, EEH_PE_ISOLATED); break; case pcie_hot_reset: + eeh_pe_state_mark(pe, EEH_PE_ISOLATED); eeh_ops->set_option(pe, EEH_OPT_FREEZE_PE); eeh_pe_dev_traverse(pe, eeh_disable_and_save_dev_state, dev); eeh_pe_state_mark(pe, EEH_PE_CFG_BLOCKED); eeh_ops->reset(pe, EEH_RESET_HOT); break; case pcie_warm_reset: + eeh_pe_state_mark(pe, EEH_PE_ISOLATED); eeh_ops->set_option(pe, EEH_OPT_FREEZE_PE); eeh_pe_dev_traverse(pe, eeh_disable_and_save_dev_state, dev); eeh_pe_state_mark(pe, EEH_PE_CFG_BLOCKED); eeh_ops->reset(pe, EEH_RESET_FUNDAMENTAL); break; default: - eeh_pe_state_clear(pe, EEH_PE_CFG_BLOCKED); + eeh_pe_state_clear(pe, EEH_PE_ISOLATED | EEH_PE_CFG_BLOCKED); return -EINVAL; }; -- cgit v0.10.2 From d91dafc02f42e23c1a906202ebde5d7c49ef058d Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Fri, 1 May 2015 09:22:15 +1000 Subject: powerpc/eeh: Delay probing EEH device during hotplug Commit 1c509148b ("powerpc/eeh: Do probe on pci_dn") probes EEH devices in early stage, which is reasonable to pSeries platform. However, it's wrong for PowerNV platform because the PE# isn't determined until the resources (IO and MMIO) are assigned to PE in hotplug case. So we have to delay probing EEH devices for PowerNV platform until the PE# is assigned. Fixes: ff57b454ddb9 ("powerpc/eeh: Do probe on pci_dn") Signed-off-by: Gavin Shan Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/kernel/eeh.c b/arch/powerpc/kernel/eeh.c index fa046ca..9ee61d1 100644 --- a/arch/powerpc/kernel/eeh.c +++ b/arch/powerpc/kernel/eeh.c @@ -1061,6 +1061,9 @@ void eeh_add_device_early(struct pci_dn *pdn) if (!edev || !eeh_enabled()) return; + if (!eeh_has_flag(EEH_PROBE_MODE_DEVTREE)) + return; + /* USB Bus children of PCI devices will not have BUID's */ phb = edev->phb; if (NULL == phb || @@ -1115,6 +1118,9 @@ void eeh_add_device_late(struct pci_dev *dev) return; } + if (eeh_has_flag(EEH_PROBE_MODE_DEV)) + eeh_ops->probe(pdn, NULL); + /* * The EEH cache might not be removed correctly because of * unbalanced kref to the device during unplug time, which -- cgit v0.10.2 From 0aab3747091db309b8a484cfd382a41644552aa3 Mon Sep 17 00:00:00 2001 From: Sam Bobroff Date: Fri, 1 May 2015 16:50:34 +1000 Subject: powerpc/powernv: Restore non-volatile CRs after nap Patches 7cba160ad "powernv/cpuidle: Redesign idle states management" and 77b54e9f2 "powernv/powerpc: Add winkle support for offline cpus" use non-volatile condition registers (cr2, cr3 and cr4) early in the system reset interrupt handler (system_reset_pSeries()) before it has been determined if state loss has occurred. If state loss has not occurred, control returns via the power7_wakeup_noloss() path which does not restore those condition registers, leaving them corrupted. Fix this by restoring the condition registers in the power7_wakeup_noloss() case. This is apparent when running a KVM guest on hardware that does not support winkle or sleep and the guest makes use of secondary threads. In practice this means Power7 machines, though some early unreleased Power8 machines may also be susceptible. The secondary CPUs are taken off line before the guest is started and they call pnv_smp_cpu_kill_self(). This checks support for sleep states (in this case there is no support) and power7_nap() is called. When the CPU is woken, power7_nap() returns and because the CPU is still off line, the main while loop executes again. The sleep states support test is executed again, but because the tested values cannot have changed, the compiler has optimized the test away and instead we rely on the result of the first test, which has been left in cr3 and/or cr4. With the result overwritten, the wrong branch is taken and power7_winkle() is called on a CPU that does not support it, leading to it stalling. Fixes: 7cba160ad789 ("powernv/cpuidle: Redesign idle states management") Fixes: 77b54e9f213f ("powernv/powerpc: Add winkle support for offline cpus") [mpe: Massage change log a bit more] Signed-off-by: Sam Bobroff Signed-off-by: Michael Ellerman diff --git a/arch/powerpc/kernel/idle_power7.S b/arch/powerpc/kernel/idle_power7.S index eeaa0d5..ccde8f0 100644 --- a/arch/powerpc/kernel/idle_power7.S +++ b/arch/powerpc/kernel/idle_power7.S @@ -501,9 +501,11 @@ BEGIN_FTR_SECTION CHECK_HMI_INTERRUPT END_FTR_SECTION_IFSET(CPU_FTR_HVMODE) ld r1,PACAR1(r13) + ld r6,_CCR(r1) ld r4,_MSR(r1) ld r5,_NIP(r1) addi r1,r1,INT_FRAME_SIZE + mtcr r6 mtspr SPRN_SRR1,r4 mtspr SPRN_SRR0,r5 rfid -- cgit v0.10.2 From 88776f366ede7d9cdce60bd2c9753dd6d6fa8b77 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 1 May 2015 09:20:34 +0200 Subject: ALSA: hda - Add headphone quirk for Lifebook E752 Fujitsu Lifebook E752 laptop needs a similar quirk done for Lifebook T731. Otherwise the headphone is always muted. Reported-and-tested-by: Christian Weber Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index e2afd53..eb2fc79 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5119,6 +5119,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x104d, 0x9099, "Sony VAIO S13", ALC275_FIXUP_SONY_DISABLE_AAMIX), SND_PCI_QUIRK(0x10cf, 0x1475, "Lifebook", ALC269_FIXUP_LIFEBOOK), SND_PCI_QUIRK(0x10cf, 0x15dc, "Lifebook T731", ALC269_FIXUP_LIFEBOOK_HP_PIN), + SND_PCI_QUIRK(0x10cf, 0x1757, "Lifebook E752", ALC269_FIXUP_LIFEBOOK_HP_PIN), SND_PCI_QUIRK(0x10cf, 0x1845, "Lifebook U904", ALC269_FIXUP_LIFEBOOK_EXTMIC), SND_PCI_QUIRK(0x144d, 0xc109, "Samsung Ativ book 9 (NP900X3G)", ALC269_FIXUP_INV_DMIC), SND_PCI_QUIRK(0x1458, 0xfa53, "Gigabyte BXBT-2807", ALC283_FIXUP_BXBT2807_MIC), -- cgit v0.10.2 From e944ec2ca00fb0170ba9d7f2aeec32c22dc0d4ec Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Wed, 29 Apr 2015 21:08:48 +0900 Subject: perf report: Fix -T/--threads option to work again The commit 512ae1bd6acb ("perf tools: Consolidate management of default sort orders") changed default value of the 'sort_order' variable to NULL indicating that users don't set any sort keys on the command line. However it missed to update a check in perf_evlist__tty_browse_hists() so that 'perf report -T' cannot show the per-thread values after the normal output. This patch fixes it to work again. Note that the -T option only works on --stdio and neither --sort nor --parent option was given. Signed-off-by: Namhyung Kim Cc: Andi Kleen Cc: David Ahern Cc: Jiri Olsa Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1430309328-28317-1-git-send-email-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 476cdf7..b63aeda 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -329,7 +329,7 @@ static int perf_evlist__tty_browse_hists(struct perf_evlist *evlist, fprintf(stdout, "\n\n"); } - if (sort_order == default_sort_order && + if (sort_order == NULL && parent_pattern == default_parent_pattern) { fprintf(stdout, "#\n# (%s)\n#\n", help); -- cgit v0.10.2 From 545774bd6e1427d98dde77244329d2311c5eca6f Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Mon, 27 Apr 2015 14:51:35 +0800 Subject: ASoC: mc13783: Fix wrong mask value used in mc13xxx_reg_rmw() calls mc13xxx_reg_rmw() won't change any bit if passing 0 to the mask field. Pass AUDIO_SSI_SEL instead of 0 for the mask field to set AUDIO_SSI_SEL bit. Signed-off-by: Axel Lin Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/codecs/mc13783.c b/sound/soc/codecs/mc13783.c index 2ffb9a0..3d44fc5 100644 --- a/sound/soc/codecs/mc13783.c +++ b/sound/soc/codecs/mc13783.c @@ -623,14 +623,14 @@ static int mc13783_probe(struct snd_soc_codec *codec) AUDIO_SSI_SEL, 0); else mc13xxx_reg_rmw(priv->mc13xxx, MC13783_AUDIO_CODEC, - 0, AUDIO_SSI_SEL); + AUDIO_SSI_SEL, AUDIO_SSI_SEL); if (priv->dac_ssi_port == MC13783_SSI1_PORT) mc13xxx_reg_rmw(priv->mc13xxx, MC13783_AUDIO_DAC, AUDIO_SSI_SEL, 0); else mc13xxx_reg_rmw(priv->mc13xxx, MC13783_AUDIO_DAC, - 0, AUDIO_SSI_SEL); + AUDIO_SSI_SEL, AUDIO_SSI_SEL); return 0; } -- cgit v0.10.2 From d3cc2e865055f58378de86c6b51a5636328ca827 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 10 Apr 2015 11:49:44 -0700 Subject: MAINTAINERS: Update mach-bcm maintainers list Add Ray Jui and Scott Branden for the mach-bcm/ entries. Signed-off-by: Florian Fainelli diff --git a/MAINTAINERS b/MAINTAINERS index 2e5bbc0..ba352d9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2118,6 +2118,8 @@ F: drivers/net/ethernet/broadcom/bnx2x/ BROADCOM BCM281XX/BCM11XXX/BCM216XX ARM ARCHITECTURE M: Christian Daudt M: Florian Fainelli +M: Ray Jui +M: Scott Branden L: bcm-kernel-feedback-list@broadcom.com T: git git://github.com/broadcom/mach-bcm S: Maintained -- cgit v0.10.2 From 049a4c506586c293b6fe54b5c61762230417dbf8 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 10 Apr 2015 11:50:40 -0700 Subject: MAINTAINERS: Remove Christian Daudt for mach-bcm Christian has not been responding to mach-bcm related emails since around July 2013, and with Scott and Ray maintaining the platform, we should have enough coverage to get that going now. Signed-off-by: Florian Fainelli diff --git a/MAINTAINERS b/MAINTAINERS index ba352d9..7222089 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2116,7 +2116,6 @@ S: Supported F: drivers/net/ethernet/broadcom/bnx2x/ BROADCOM BCM281XX/BCM11XXX/BCM216XX ARM ARCHITECTURE -M: Christian Daudt M: Florian Fainelli M: Ray Jui M: Scott Branden -- cgit v0.10.2 From 1de2a662d0f2d3b211b7d10ea6578abe5c39eda3 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 10 Apr 2015 11:52:40 -0700 Subject: MAINTAINERS: Update brcmstb entry We have not heard back from Marc in a while, since we last asked him to ack/nack this change here: https://lkml.org/lkml/2015/1/14/1004 This was back in January, we are in the 3.21^W4.1 merge window now, so, as outlined in this email thread, remove him from the MAINTAINERS list now. Signed-off-by: Florian Fainelli diff --git a/MAINTAINERS b/MAINTAINERS index 7222089..0aa1a42 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2169,7 +2169,6 @@ S: Maintained F: drivers/usb/gadget/udc/bcm63xx_udc.* BROADCOM BCM7XXX ARM ARCHITECTURE -M: Marc Carino M: Brian Norris M: Gregory Fong M: Florian Fainelli -- cgit v0.10.2 From 5f55d2ae699d1756ad6132786c7f9c27dc456b66 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Tue, 28 Apr 2015 10:23:30 -0600 Subject: vfio-pci: Log device requests more verbosely Log some clues indicating whether the user is receiving device request interfaces or not listening. This can help indicate why a driver unbind is blocked or explain why QEMU automatically unplugged a device from the VM. Signed-off-by: Alex Williamson diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/vfio/pci/vfio_pci.c index 69fab0f..e9851ad 100644 --- a/drivers/vfio/pci/vfio_pci.c +++ b/drivers/vfio/pci/vfio_pci.c @@ -907,8 +907,14 @@ static void vfio_pci_request(void *device_data, unsigned int count) mutex_lock(&vdev->igate); if (vdev->req_trigger) { - dev_dbg(&vdev->pdev->dev, "Requesting device from user\n"); + if (!(count % 10)) + dev_notice_ratelimited(&vdev->pdev->dev, + "Relaying device request to user (#%u)\n", + count); eventfd_signal(vdev->req_trigger, 1); + } else if (count == 0) { + dev_warn(&vdev->pdev->dev, + "No device request channel registered, blocked until released by user\n"); } mutex_unlock(&vdev->igate); -- cgit v0.10.2 From 5de0b4d0cd153c471640b13aae6ae6d18d0a4603 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 1 May 2015 16:56:45 -0400 Subject: ext4 crypto: simplify and speed up filename encryption Avoid using SHA-1 when calculating the user-visible filename when the encryption key is available, and avoid decrypting lots of filenames when searching for a directory entry in a directory block. Change-Id: If4655f144784978ba0305b597bfa1c8d7bb69e63 Signed-off-by: Theodore Ts'o diff --git a/fs/ext4/crypto_fname.c b/fs/ext4/crypto_fname.c index ca2f594..7a877e6 100644 --- a/fs/ext4/crypto_fname.c +++ b/fs/ext4/crypto_fname.c @@ -198,106 +198,57 @@ static int ext4_fname_decrypt(struct ext4_fname_crypto_ctx *ctx, return oname->len; } +static const char *lookup_table = + "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+,"; + /** * ext4_fname_encode_digest() - * * Encodes the input digest using characters from the set [a-zA-Z0-9_+]. * The encoded string is roughly 4/3 times the size of the input string. */ -int ext4_fname_encode_digest(char *dst, char *src, u32 len) +static int digest_encode(const char *src, int len, char *dst) { - static const char *lookup_table = - "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_+"; - u32 current_chunk, num_chunks, i; - char tmp_buf[3]; - u32 c0, c1, c2, c3; - - current_chunk = 0; - num_chunks = len/3; - for (i = 0; i < num_chunks; i++) { - c0 = src[3*i] & 0x3f; - c1 = (((src[3*i]>>6)&0x3) | ((src[3*i+1] & 0xf)<<2)) & 0x3f; - c2 = (((src[3*i+1]>>4)&0xf) | ((src[3*i+2] & 0x3)<<4)) & 0x3f; - c3 = (src[3*i+2]>>2) & 0x3f; - dst[4*i] = lookup_table[c0]; - dst[4*i+1] = lookup_table[c1]; - dst[4*i+2] = lookup_table[c2]; - dst[4*i+3] = lookup_table[c3]; - } - if (i*3 < len) { - memset(tmp_buf, 0, 3); - memcpy(tmp_buf, &src[3*i], len-3*i); - c0 = tmp_buf[0] & 0x3f; - c1 = (((tmp_buf[0]>>6)&0x3) | ((tmp_buf[1] & 0xf)<<2)) & 0x3f; - c2 = (((tmp_buf[1]>>4)&0xf) | ((tmp_buf[2] & 0x3)<<4)) & 0x3f; - c3 = (tmp_buf[2]>>2) & 0x3f; - dst[4*i] = lookup_table[c0]; - dst[4*i+1] = lookup_table[c1]; - dst[4*i+2] = lookup_table[c2]; - dst[4*i+3] = lookup_table[c3]; + int i = 0, bits = 0, ac = 0; + char *cp = dst; + + while (i < len) { + ac += (((unsigned char) src[i]) << bits); + bits += 8; + do { + *cp++ = lookup_table[ac & 0x3f]; + ac >>= 6; + bits -= 6; + } while (bits >= 6); i++; } - return (i * 4); + if (bits) + *cp++ = lookup_table[ac & 0x3f]; + return cp - dst; } -/** - * ext4_fname_hash() - - * - * This function computes the hash of the input filename, and sets the output - * buffer to the *encoded* digest. It returns the length of the digest as its - * return value. Errors are returned as negative numbers. We trust the caller - * to allocate sufficient memory to oname string. - */ -static int ext4_fname_hash(struct ext4_fname_crypto_ctx *ctx, - const struct ext4_str *iname, - struct ext4_str *oname) +static int digest_decode(const char *src, int len, char *dst) { - struct scatterlist sg; - struct hash_desc desc = { - .tfm = (struct crypto_hash *)ctx->htfm, - .flags = CRYPTO_TFM_REQ_MAY_SLEEP - }; - int res = 0; - - if (iname->len <= EXT4_FNAME_CRYPTO_DIGEST_SIZE) { - res = ext4_fname_encode_digest(oname->name, iname->name, - iname->len); - oname->len = res; - return res; - } - - sg_init_one(&sg, iname->name, iname->len); - res = crypto_hash_init(&desc); - if (res) { - printk(KERN_ERR - "%s: Error initializing crypto hash; res = [%d]\n", - __func__, res); - goto out; - } - res = crypto_hash_update(&desc, &sg, iname->len); - if (res) { - printk(KERN_ERR - "%s: Error updating crypto hash; res = [%d]\n", - __func__, res); - goto out; - } - res = crypto_hash_final(&desc, - &oname->name[EXT4_FNAME_CRYPTO_DIGEST_SIZE]); - if (res) { - printk(KERN_ERR - "%s: Error finalizing crypto hash; res = [%d]\n", - __func__, res); - goto out; + int i = 0, bits = 0, ac = 0; + const char *p; + char *cp = dst; + + while (i < len) { + p = strchr(lookup_table, src[i]); + if (p == NULL || src[i] == 0) + return -2; + ac += (p - lookup_table) << bits; + bits += 6; + if (bits >= 8) { + *cp++ = ac & 0xff; + ac >>= 8; + bits -= 8; + } + i++; } - /* Encode the digest as a printable string--this will increase the - * size of the digest */ - oname->name[0] = 'I'; - res = ext4_fname_encode_digest(oname->name+1, - &oname->name[EXT4_FNAME_CRYPTO_DIGEST_SIZE], - EXT4_FNAME_CRYPTO_DIGEST_SIZE) + 1; - oname->len = res; -out: - return res; + if (ac) + return -1; + return cp - dst; } /** @@ -571,9 +522,13 @@ void ext4_fname_crypto_free_buffer(struct ext4_str *crypto_str) * ext4_fname_disk_to_usr() - converts a filename from disk space to user space */ int _ext4_fname_disk_to_usr(struct ext4_fname_crypto_ctx *ctx, - const struct ext4_str *iname, - struct ext4_str *oname) + struct dx_hash_info *hinfo, + const struct ext4_str *iname, + struct ext4_str *oname) { + char buf[24]; + int ret; + if (ctx == NULL) return -EIO; if (iname->len < 3) { @@ -587,18 +542,33 @@ int _ext4_fname_disk_to_usr(struct ext4_fname_crypto_ctx *ctx, } if (ctx->has_valid_key) return ext4_fname_decrypt(ctx, iname, oname); - else - return ext4_fname_hash(ctx, iname, oname); + + if (iname->len <= EXT4_FNAME_CRYPTO_DIGEST_SIZE) { + ret = digest_encode(iname->name, iname->len, oname->name); + oname->len = ret; + return ret; + } + if (hinfo) { + memcpy(buf, &hinfo->hash, 4); + memcpy(buf+4, &hinfo->minor_hash, 4); + } else + memset(buf, 0, 8); + memcpy(buf + 8, iname->name + iname->len - 16, 16); + oname->name[0] = '_'; + ret = digest_encode(buf, 24, oname->name+1); + oname->len = ret + 1; + return ret + 1; } int ext4_fname_disk_to_usr(struct ext4_fname_crypto_ctx *ctx, + struct dx_hash_info *hinfo, const struct ext4_dir_entry_2 *de, struct ext4_str *oname) { struct ext4_str iname = {.name = (unsigned char *) de->name, .len = de->name_len }; - return _ext4_fname_disk_to_usr(ctx, &iname, oname); + return _ext4_fname_disk_to_usr(ctx, hinfo, &iname, oname); } @@ -640,10 +610,11 @@ int ext4_fname_usr_to_hash(struct ext4_fname_crypto_ctx *ctx, const struct qstr *iname, struct dx_hash_info *hinfo) { - struct ext4_str tmp, tmp2; + struct ext4_str tmp; int ret = 0; + char buf[EXT4_FNAME_CRYPTO_DIGEST_SIZE+1]; - if (!ctx || !ctx->has_valid_key || + if (!ctx || ((iname->name[0] == '.') && ((iname->len == 1) || ((iname->name[1] == '.') && (iname->len == 2))))) { @@ -651,59 +622,90 @@ int ext4_fname_usr_to_hash(struct ext4_fname_crypto_ctx *ctx, return 0; } + if (!ctx->has_valid_key && iname->name[0] == '_') { + if (iname->len != 33) + return -ENOENT; + ret = digest_decode(iname->name+1, iname->len, buf); + if (ret != 24) + return -ENOENT; + memcpy(&hinfo->hash, buf, 4); + memcpy(&hinfo->minor_hash, buf + 4, 4); + return 0; + } + + if (!ctx->has_valid_key && iname->name[0] != '_') { + if (iname->len > 43) + return -ENOENT; + ret = digest_decode(iname->name, iname->len, buf); + ext4fs_dirhash(buf, ret, hinfo); + return 0; + } + /* First encrypt the plaintext name */ ret = ext4_fname_crypto_alloc_buffer(ctx, iname->len, &tmp); if (ret < 0) return ret; ret = ext4_fname_encrypt(ctx, iname, &tmp); - if (ret < 0) - goto out; - - tmp2.len = (4 * ((EXT4_FNAME_CRYPTO_DIGEST_SIZE + 2) / 3)) + 1; - tmp2.name = kmalloc(tmp2.len + 1, GFP_KERNEL); - if (tmp2.name == NULL) { - ret = -ENOMEM; - goto out; + if (ret >= 0) { + ext4fs_dirhash(tmp.name, tmp.len, hinfo); + ret = 0; } - ret = ext4_fname_hash(ctx, &tmp, &tmp2); - if (ret > 0) - ext4fs_dirhash(tmp2.name, tmp2.len, hinfo); - ext4_fname_crypto_free_buffer(&tmp2); -out: ext4_fname_crypto_free_buffer(&tmp); return ret; } -/** - * ext4_fname_disk_to_htree() - converts a filename from disk space to htree-access string - */ -int ext4_fname_disk_to_hash(struct ext4_fname_crypto_ctx *ctx, - const struct ext4_dir_entry_2 *de, - struct dx_hash_info *hinfo) +int ext4_fname_match(struct ext4_fname_crypto_ctx *ctx, struct ext4_str *cstr, + int len, const char * const name, + struct ext4_dir_entry_2 *de) { - struct ext4_str iname = {.name = (unsigned char *) de->name, - .len = de->name_len}; - struct ext4_str tmp; - int ret; + int ret = -ENOENT; + int bigname = (*name == '_'); - if (!ctx || - ((iname.name[0] == '.') && - ((iname.len == 1) || - ((iname.name[1] == '.') && (iname.len == 2))))) { - ext4fs_dirhash(iname.name, iname.len, hinfo); - return 0; + if (ctx->has_valid_key) { + if (cstr->name == NULL) { + struct qstr istr; + + ret = ext4_fname_crypto_alloc_buffer(ctx, len, cstr); + if (ret < 0) + goto errout; + istr.name = name; + istr.len = len; + ret = ext4_fname_encrypt(ctx, &istr, cstr); + if (ret < 0) + goto errout; + } + } else { + if (cstr->name == NULL) { + cstr->name = kmalloc(32, GFP_KERNEL); + if (cstr->name == NULL) + return -ENOMEM; + if ((bigname && (len != 33)) || + (!bigname && (len > 43))) + goto errout; + ret = digest_decode(name+bigname, len-bigname, + cstr->name); + if (ret < 0) { + ret = -ENOENT; + goto errout; + } + cstr->len = ret; + } + if (bigname) { + if (de->name_len < 16) + return 0; + ret = memcmp(de->name + de->name_len - 16, + cstr->name + 8, 16); + return (ret == 0) ? 1 : 0; + } } - - tmp.len = (4 * ((EXT4_FNAME_CRYPTO_DIGEST_SIZE + 2) / 3)) + 1; - tmp.name = kmalloc(tmp.len + 1, GFP_KERNEL); - if (tmp.name == NULL) - return -ENOMEM; - - ret = ext4_fname_hash(ctx, &iname, &tmp); - if (ret > 0) - ext4fs_dirhash(tmp.name, tmp.len, hinfo); - ext4_fname_crypto_free_buffer(&tmp); + if (de->name_len != cstr->len) + return 0; + ret = memcmp(de->name, cstr->name, cstr->len); + return (ret == 0) ? 1 : 0; +errout: + kfree(cstr->name); + cstr->name = NULL; return ret; } diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 61db51a5..5665d82 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -249,7 +249,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx) } else { /* Directory is encrypted */ err = ext4_fname_disk_to_usr(enc_ctx, - de, &fname_crypto_str); + NULL, de, &fname_crypto_str); if (err < 0) goto errout; if (!dir_emit(ctx, diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 0179654..dfb1138 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -2093,9 +2093,11 @@ u32 ext4_fname_crypto_round_up(u32 size, u32 blksize); int ext4_fname_crypto_alloc_buffer(struct ext4_fname_crypto_ctx *ctx, u32 ilen, struct ext4_str *crypto_str); int _ext4_fname_disk_to_usr(struct ext4_fname_crypto_ctx *ctx, + struct dx_hash_info *hinfo, const struct ext4_str *iname, struct ext4_str *oname); int ext4_fname_disk_to_usr(struct ext4_fname_crypto_ctx *ctx, + struct dx_hash_info *hinfo, const struct ext4_dir_entry_2 *de, struct ext4_str *oname); int ext4_fname_usr_to_disk(struct ext4_fname_crypto_ctx *ctx, @@ -2104,11 +2106,12 @@ int ext4_fname_usr_to_disk(struct ext4_fname_crypto_ctx *ctx, int ext4_fname_usr_to_hash(struct ext4_fname_crypto_ctx *ctx, const struct qstr *iname, struct dx_hash_info *hinfo); -int ext4_fname_disk_to_hash(struct ext4_fname_crypto_ctx *ctx, - const struct ext4_dir_entry_2 *de, - struct dx_hash_info *hinfo); int ext4_fname_crypto_namelen_on_disk(struct ext4_fname_crypto_ctx *ctx, u32 namelen); +int ext4_fname_match(struct ext4_fname_crypto_ctx *ctx, struct ext4_str *cstr, + int len, const char * const name, + struct ext4_dir_entry_2 *de); + #ifdef CONFIG_EXT4_FS_ENCRYPTION void ext4_put_fname_crypto_ctx(struct ext4_fname_crypto_ctx **ctx); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 4f87127..5ea7371 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -640,7 +640,7 @@ static struct stats dx_show_leaf(struct inode *dir, ext4_put_fname_crypto_ctx(&ctx); ctx = NULL; } - res = ext4_fname_disk_to_usr(ctx, de, + res = ext4_fname_disk_to_usr(ctx, NULL, de, &fname_crypto_str); if (res < 0) { printk(KERN_WARNING "Error " @@ -653,15 +653,8 @@ static struct stats dx_show_leaf(struct inode *dir, name = fname_crypto_str.name; len = fname_crypto_str.len; } - res = ext4_fname_disk_to_hash(ctx, de, - &h); - if (res < 0) { - printk(KERN_WARNING "Error " - "converting filename " - "from disk to htree" - "\n"); - h.hash = 0xDEADBEEF; - } + ext4fs_dirhash(de->name, de->name_len, + &h); printk("%*.s:(E)%x.%u ", len, name, h.hash, (unsigned) ((char *) de - base)); @@ -1008,15 +1001,7 @@ static int htree_dirblock_to_tree(struct file *dir_file, /* silently ignore the rest of the block */ break; } -#ifdef CONFIG_EXT4_FS_ENCRYPTION - err = ext4_fname_disk_to_hash(ctx, de, hinfo); - if (err < 0) { - count = err; - goto errout; - } -#else ext4fs_dirhash(de->name, de->name_len, hinfo); -#endif if ((hinfo->hash < start_hash) || ((hinfo->hash == start_hash) && (hinfo->minor_hash < start_minor_hash))) @@ -1032,7 +1017,7 @@ static int htree_dirblock_to_tree(struct file *dir_file, &tmp_str); } else { /* Directory is encrypted */ - err = ext4_fname_disk_to_usr(ctx, de, + err = ext4_fname_disk_to_usr(ctx, hinfo, de, &fname_crypto_str); if (err < 0) { count = err; @@ -1193,26 +1178,10 @@ static int dx_make_map(struct inode *dir, struct ext4_dir_entry_2 *de, int count = 0; char *base = (char *) de; struct dx_hash_info h = *hinfo; -#ifdef CONFIG_EXT4_FS_ENCRYPTION - struct ext4_fname_crypto_ctx *ctx = NULL; - int err; - - ctx = ext4_get_fname_crypto_ctx(dir, EXT4_NAME_LEN); - if (IS_ERR(ctx)) - return PTR_ERR(ctx); -#endif while ((char *) de < base + blocksize) { if (de->name_len && de->inode) { -#ifdef CONFIG_EXT4_FS_ENCRYPTION - err = ext4_fname_disk_to_hash(ctx, de, &h); - if (err < 0) { - ext4_put_fname_crypto_ctx(&ctx); - return err; - } -#else ext4fs_dirhash(de->name, de->name_len, &h); -#endif map_tail--; map_tail->hash = h.hash; map_tail->offs = ((char *) de - base)>>2; @@ -1223,9 +1192,6 @@ static int dx_make_map(struct inode *dir, struct ext4_dir_entry_2 *de, /* XXX: do we need to check rec_len == 0 case? -Chris */ de = ext4_next_entry(de, blocksize); } -#ifdef CONFIG_EXT4_FS_ENCRYPTION - ext4_put_fname_crypto_ctx(&ctx); -#endif return count; } @@ -1287,16 +1253,8 @@ static inline int ext4_match(struct ext4_fname_crypto_ctx *ctx, return 0; #ifdef CONFIG_EXT4_FS_ENCRYPTION - if (ctx) { - /* Directory is encrypted */ - res = ext4_fname_disk_to_usr(ctx, de, fname_crypto_str); - if (res < 0) - return res; - if (len != res) - return 0; - res = memcmp(name, fname_crypto_str->name, len); - return (res == 0) ? 1 : 0; - } + if (ctx) + return ext4_fname_match(ctx, fname_crypto_str, len, name, de); #endif if (len != de->name_len) return 0; @@ -1324,16 +1282,6 @@ int search_dir(struct buffer_head *bh, char *search_buf, int buf_size, if (IS_ERR(ctx)) return -1; - if (ctx != NULL) { - /* Allocate buffer to hold maximum name length */ - res = ext4_fname_crypto_alloc_buffer(ctx, EXT4_NAME_LEN, - &fname_crypto_str); - if (res < 0) { - ext4_put_fname_crypto_ctx(&ctx); - return -1; - } - } - de = (struct ext4_dir_entry_2 *)search_buf; dlimit = search_buf + buf_size; while ((char *) de < dlimit) { @@ -1872,14 +1820,6 @@ int ext4_find_dest_de(struct inode *dir, struct inode *inode, return res; } reclen = EXT4_DIR_REC_LEN(res); - - /* Allocate buffer to hold maximum name length */ - res = ext4_fname_crypto_alloc_buffer(ctx, EXT4_NAME_LEN, - &fname_crypto_str); - if (res < 0) { - ext4_put_fname_crypto_ctx(&ctx); - return -1; - } } de = (struct ext4_dir_entry_2 *)buf; diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c index 136ca0e..ce2ed28 100644 --- a/fs/ext4/symlink.c +++ b/fs/ext4/symlink.c @@ -74,7 +74,7 @@ static void *ext4_follow_link(struct dentry *dentry, struct nameidata *nd) goto errout; } pstr.name = paddr; - res = _ext4_fname_disk_to_usr(ctx, &cstr, &pstr); + res = _ext4_fname_disk_to_usr(ctx, NULL, &cstr, &pstr); if (res < 0) goto errout; /* Null-terminate the name */ -- cgit v0.10.2 From a44cd7a05496d60fd2ba8cca080e3db8f481549b Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 1 May 2015 16:56:50 -0400 Subject: ext4 crypto: add padding to filenames before encrypting This obscures the length of the filenames, to decrease the amount of information leakage. By default, we pad the filenames to the next 4 byte boundaries. This costs nothing, since the directory entries are aligned to 4 byte boundaries anyway. Filenames can also be padded to 8, 16, or 32 bytes, which will consume more directory space. Change-Id: Ibb7a0fb76d2c48e2061240a709358ff40b14f322 Signed-off-by: Theodore Ts'o diff --git a/fs/ext4/crypto_fname.c b/fs/ext4/crypto_fname.c index 7a877e6..fded02f 100644 --- a/fs/ext4/crypto_fname.c +++ b/fs/ext4/crypto_fname.c @@ -66,6 +66,7 @@ static int ext4_fname_encrypt(struct ext4_fname_crypto_ctx *ctx, int res = 0; char iv[EXT4_CRYPTO_BLOCK_SIZE]; struct scatterlist sg[1]; + int padding = 4 << (ctx->flags & EXT4_POLICY_FLAGS_PAD_MASK); char *workbuf; if (iname->len <= 0 || iname->len > ctx->lim) @@ -73,6 +74,7 @@ static int ext4_fname_encrypt(struct ext4_fname_crypto_ctx *ctx, ciphertext_len = (iname->len < EXT4_CRYPTO_BLOCK_SIZE) ? EXT4_CRYPTO_BLOCK_SIZE : iname->len; + ciphertext_len = ext4_fname_crypto_round_up(ciphertext_len, padding); ciphertext_len = (ciphertext_len > ctx->lim) ? ctx->lim : ciphertext_len; @@ -101,7 +103,7 @@ static int ext4_fname_encrypt(struct ext4_fname_crypto_ctx *ctx, /* Create encryption request */ sg_init_table(sg, 1); sg_set_page(sg, ctx->workpage, PAGE_SIZE, 0); - ablkcipher_request_set_crypt(req, sg, sg, iname->len, iv); + ablkcipher_request_set_crypt(req, sg, sg, ciphertext_len, iv); res = crypto_ablkcipher_encrypt(req); if (res == -EINPROGRESS || res == -EBUSY) { BUG_ON(req->base.data != &ecr); @@ -356,6 +358,7 @@ struct ext4_fname_crypto_ctx *ext4_get_fname_crypto_ctx( if (IS_ERR(ctx)) return ctx; + ctx->flags = ei->i_crypt_policy_flags; if (ctx->has_valid_key) { if (ctx->key.mode != EXT4_ENCRYPTION_MODE_AES_256_CTS) { printk_once(KERN_WARNING @@ -468,6 +471,7 @@ int ext4_fname_crypto_namelen_on_disk(struct ext4_fname_crypto_ctx *ctx, u32 namelen) { u32 ciphertext_len; + int padding = 4 << (ctx->flags & EXT4_POLICY_FLAGS_PAD_MASK); if (ctx == NULL) return -EIO; @@ -475,6 +479,7 @@ int ext4_fname_crypto_namelen_on_disk(struct ext4_fname_crypto_ctx *ctx, return -EACCES; ciphertext_len = (namelen < EXT4_CRYPTO_BLOCK_SIZE) ? EXT4_CRYPTO_BLOCK_SIZE : namelen; + ciphertext_len = ext4_fname_crypto_round_up(ciphertext_len, padding); ciphertext_len = (ciphertext_len > ctx->lim) ? ctx->lim : ciphertext_len; return (int) ciphertext_len; @@ -490,10 +495,13 @@ int ext4_fname_crypto_alloc_buffer(struct ext4_fname_crypto_ctx *ctx, u32 ilen, struct ext4_str *crypto_str) { unsigned int olen; + int padding = 4 << (ctx->flags & EXT4_POLICY_FLAGS_PAD_MASK); if (!ctx) return -EIO; - olen = ext4_fname_crypto_round_up(ilen, EXT4_CRYPTO_BLOCK_SIZE); + if (padding < EXT4_CRYPTO_BLOCK_SIZE) + padding = EXT4_CRYPTO_BLOCK_SIZE; + olen = ext4_fname_crypto_round_up(ilen, padding); crypto_str->len = olen; if (olen < EXT4_FNAME_CRYPTO_DIGEST_SIZE*2) olen = EXT4_FNAME_CRYPTO_DIGEST_SIZE*2; diff --git a/fs/ext4/crypto_key.c b/fs/ext4/crypto_key.c index c8392af..52170d0 100644 --- a/fs/ext4/crypto_key.c +++ b/fs/ext4/crypto_key.c @@ -110,6 +110,7 @@ int ext4_generate_encryption_key(struct inode *inode) } res = 0; + ei->i_crypt_policy_flags = ctx.flags; if (S_ISREG(inode->i_mode)) crypt_key->mode = ctx.contents_encryption_mode; else if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) diff --git a/fs/ext4/crypto_policy.c b/fs/ext4/crypto_policy.c index 30eaf9e..a6d6291 100644 --- a/fs/ext4/crypto_policy.c +++ b/fs/ext4/crypto_policy.c @@ -37,6 +37,8 @@ static int ext4_is_encryption_context_consistent_with_policy( return 0; return (memcmp(ctx.master_key_descriptor, policy->master_key_descriptor, EXT4_KEY_DESCRIPTOR_SIZE) == 0 && + (ctx.flags == + policy->flags) && (ctx.contents_encryption_mode == policy->contents_encryption_mode) && (ctx.filenames_encryption_mode == @@ -56,25 +58,25 @@ static int ext4_create_encryption_context_from_policy( printk(KERN_WARNING "%s: Invalid contents encryption mode %d\n", __func__, policy->contents_encryption_mode); - res = -EINVAL; - goto out; + return -EINVAL; } if (!ext4_valid_filenames_enc_mode(policy->filenames_encryption_mode)) { printk(KERN_WARNING "%s: Invalid filenames encryption mode %d\n", __func__, policy->filenames_encryption_mode); - res = -EINVAL; - goto out; + return -EINVAL; } + if (policy->flags & ~EXT4_POLICY_FLAGS_VALID) + return -EINVAL; ctx.contents_encryption_mode = policy->contents_encryption_mode; ctx.filenames_encryption_mode = policy->filenames_encryption_mode; + ctx.flags = policy->flags; BUILD_BUG_ON(sizeof(ctx.nonce) != EXT4_KEY_DERIVATION_NONCE_SIZE); get_random_bytes(ctx.nonce, EXT4_KEY_DERIVATION_NONCE_SIZE); res = ext4_xattr_set(inode, EXT4_XATTR_INDEX_ENCRYPTION, EXT4_XATTR_NAME_ENCRYPTION_CONTEXT, &ctx, sizeof(ctx), 0); -out: if (!res) ext4_set_inode_flag(inode, EXT4_INODE_ENCRYPT); return res; @@ -115,6 +117,7 @@ int ext4_get_policy(struct inode *inode, struct ext4_encryption_policy *policy) policy->version = 0; policy->contents_encryption_mode = ctx.contents_encryption_mode; policy->filenames_encryption_mode = ctx.filenames_encryption_mode; + policy->flags = ctx.flags; memcpy(&policy->master_key_descriptor, ctx.master_key_descriptor, EXT4_KEY_DESCRIPTOR_SIZE); return 0; @@ -176,6 +179,7 @@ int ext4_inherit_context(struct inode *parent, struct inode *child) EXT4_ENCRYPTION_MODE_AES_256_XTS; ctx.filenames_encryption_mode = EXT4_ENCRYPTION_MODE_AES_256_CTS; + ctx.flags = 0; memset(ctx.master_key_descriptor, 0x42, EXT4_KEY_DESCRIPTOR_SIZE); res = 0; diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index dfb1138..bca1bdc 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -911,6 +911,7 @@ struct ext4_inode_info { /* on-disk additional length */ __u16 i_extra_isize; + char i_crypt_policy_flags; /* Indicate the inline data space. */ u16 i_inline_off; diff --git a/fs/ext4/ext4_crypto.h b/fs/ext4/ext4_crypto.h index c2ba35a..d75159c 100644 --- a/fs/ext4/ext4_crypto.h +++ b/fs/ext4/ext4_crypto.h @@ -20,12 +20,20 @@ struct ext4_encryption_policy { char version; char contents_encryption_mode; char filenames_encryption_mode; + char flags; char master_key_descriptor[EXT4_KEY_DESCRIPTOR_SIZE]; } __attribute__((__packed__)); #define EXT4_ENCRYPTION_CONTEXT_FORMAT_V1 1 #define EXT4_KEY_DERIVATION_NONCE_SIZE 16 +#define EXT4_POLICY_FLAGS_PAD_4 0x00 +#define EXT4_POLICY_FLAGS_PAD_8 0x01 +#define EXT4_POLICY_FLAGS_PAD_16 0x02 +#define EXT4_POLICY_FLAGS_PAD_32 0x03 +#define EXT4_POLICY_FLAGS_PAD_MASK 0x03 +#define EXT4_POLICY_FLAGS_VALID 0x03 + /** * Encryption context for inode * @@ -41,7 +49,7 @@ struct ext4_encryption_context { char format; char contents_encryption_mode; char filenames_encryption_mode; - char reserved; + char flags; char master_key_descriptor[EXT4_KEY_DESCRIPTOR_SIZE]; char nonce[EXT4_KEY_DERIVATION_NONCE_SIZE]; } __attribute__((__packed__)); @@ -120,6 +128,7 @@ struct ext4_fname_crypto_ctx { struct crypto_hash *htfm; struct page *workpage; struct ext4_encryption_key key; + unsigned flags : 8; unsigned has_valid_key : 1; unsigned ctfm_key_is_ready : 1; }; -- cgit v0.10.2 From db7d4d7f40215843000cb9d441c9149fd42ea36b Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Fri, 1 May 2015 16:31:41 -0600 Subject: vfio: Fix runaway interruptible timeout Commit 13060b64b819 ("vfio: Add and use device request op for vfio bus drivers") incorrectly makes use of an interruptible timeout. When interrupted, the signal remains pending resulting in subsequent timeouts occurring instantly. This makes the loop spin at a much higher rate than intended. Instead of making this completely non-interruptible, we can change this into a sort of interruptible-once behavior and use the "once" to log debug information. The driver API doesn't allow us to abort and return an error code. Signed-off-by: Alex Williamson Fixes: 13060b64b819 Cc: stable@vger.kernel.org # v4.0 diff --git a/drivers/vfio/vfio.c b/drivers/vfio/vfio.c index 0d33662..e1278fe 100644 --- a/drivers/vfio/vfio.c +++ b/drivers/vfio/vfio.c @@ -710,6 +710,8 @@ void *vfio_del_group_dev(struct device *dev) void *device_data = device->device_data; struct vfio_unbound_dev *unbound; unsigned int i = 0; + long ret; + bool interrupted = false; /* * The group exists so long as we have a device reference. Get @@ -755,9 +757,22 @@ void *vfio_del_group_dev(struct device *dev) vfio_device_put(device); - } while (wait_event_interruptible_timeout(vfio.release_q, - !vfio_dev_present(group, dev), - HZ * 10) <= 0); + if (interrupted) { + ret = wait_event_timeout(vfio.release_q, + !vfio_dev_present(group, dev), HZ * 10); + } else { + ret = wait_event_interruptible_timeout(vfio.release_q, + !vfio_dev_present(group, dev), HZ * 10); + if (ret == -ERESTARTSYS) { + interrupted = true; + dev_warn(dev, + "Device is currently in use, task" + " \"%s\" (%d) " + "blocked until device is released", + current->comm, task_pid_nr(current)); + } + } + } while (ret <= 0); vfio_group_put(group); -- cgit v0.10.2 From 082a75dad84d79d1c15ea9e50f31cb4bb4fa7fd6 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Sat, 25 Apr 2015 15:56:15 +0300 Subject: rbd: end I/O the entire obj_request on error When we end I/O struct request with error, we need to pass obj_request->length as @nr_bytes so that the entire obj_request worth of bytes is completed. Otherwise block layer ends up confused and we trip on rbd_assert(more ^ (which == img_request->obj_request_count)); in rbd_img_obj_callback() due to more being true no matter what. We already do it in most cases but we are missing some, in particular those where we don't even get a chance to submit any obj_requests, due to an early -ENOMEM for example. A number of obj_request->xferred assignments seem to be redundant but I haven't touched any of obj_request->xferred stuff to keep this small and isolated. Cc: Alex Elder Cc: stable@vger.kernel.org # 3.10+ Reported-by: Shawn Edwards Reviewed-by: Sage Weil Signed-off-by: Ilya Dryomov diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 8125233..ec6c5c6 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -2264,6 +2264,11 @@ static bool rbd_img_obj_end_request(struct rbd_obj_request *obj_request) result, xferred); if (!img_request->result) img_request->result = result; + /* + * Need to end I/O on the entire obj_request worth of + * bytes in case of error. + */ + xferred = obj_request->length; } /* Image object requests don't own their page array */ -- cgit v0.10.2 From a134f083e79fb4c3d0a925691e732c56911b4326 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 1 May 2015 22:02:47 -0400 Subject: ipv4: Missing sk_nulls_node_init() in ping_unhash(). If we don't do that, then the poison value is left in the ->pprev backlink. This can cause crashes if we do a disconnect, followed by a connect(). Tested-by: Linus Torvalds Reported-by: Wen Xu Signed-off-by: David S. Miller diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index a93f260..05ff44b 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -158,6 +158,7 @@ void ping_unhash(struct sock *sk) if (sk_hashed(sk)) { write_lock_bh(&ping_table.lock); hlist_nulls_del(&sk->sk_nulls_node); + sk_nulls_node_init(&sk->sk_nulls_node); sock_put(sk); isk->inet_num = 0; isk->inet_sport = 0; -- cgit v0.10.2 From feda5f939eafa4af94dfb547847806e0f2df73b8 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 2 May 2015 08:42:38 +0930 Subject: virtio: pass baton to Michael Tsirkin With my job change kernel work will be "own time"; I'm keeping lguest and modules (and the virtio standards work), but virtio kernel has to go. This makes it clear that Michael is in charge. He's good, but having me watch over his shoulder won't help. Good luck Michael! Signed-off-by: Rusty Russell Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 2e5bbc0..1622775 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10523,7 +10523,6 @@ F: include/linux/virtio_console.h F: include/uapi/linux/virtio_console.h VIRTIO CORE, NET AND BLOCK DRIVERS -M: Rusty Russell M: "Michael S. Tsirkin" L: virtualization@lists.linux-foundation.org S: Maintained -- cgit v0.10.2 From e412d3a32badcf17541d7443b033769fdf39b545 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Sat, 2 May 2015 08:42:29 +0930 Subject: virtio: fix typo in vring_need_event() doc comment Here the "other side" refers to the guest or host. Signed-off-by: Stefan Hajnoczi Signed-off-by: Rusty Russell Signed-off-by: Linus Torvalds diff --git a/include/uapi/linux/virtio_ring.h b/include/uapi/linux/virtio_ring.h index a3318f3..915980a 100644 --- a/include/uapi/linux/virtio_ring.h +++ b/include/uapi/linux/virtio_ring.h @@ -155,7 +155,7 @@ static inline unsigned vring_size(unsigned int num, unsigned long align) } /* The following is used with USED_EVENT_IDX and AVAIL_EVENT_IDX */ -/* Assuming a given event_idx value from the other size, if +/* Assuming a given event_idx value from the other side, if * we have just incremented index from old to new_idx, * should we trigger an event? */ static inline int vring_need_event(__u16 event_idx, __u16 new_idx, __u16 old) -- cgit v0.10.2 From fb63e5489f7ef5bb4d1a655984ca7ef98ffc5849 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 2 May 2015 10:29:19 -0400 Subject: ext4 crypto: do not select from EXT4_FS_ENCRYPTION This patch adds a tristate EXT4_ENCRYPTION to do the selections for EXT4_FS_ENCRYPTION because selecting from a bool causes all the selected options to be built-in, even if EXT4 itself is a module. Signed-off-by: Herbert Xu Signed-off-by: Theodore Ts'o diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig index 18228c2..024f228 100644 --- a/fs/ext4/Kconfig +++ b/fs/ext4/Kconfig @@ -64,8 +64,8 @@ config EXT4_FS_SECURITY If you are not using a security module that requires using extended attributes for file security labels, say N. -config EXT4_FS_ENCRYPTION - bool "Ext4 Encryption" +config EXT4_ENCRYPTION + tristate "Ext4 Encryption" depends on EXT4_FS select CRYPTO_AES select CRYPTO_CBC @@ -81,6 +81,11 @@ config EXT4_FS_ENCRYPTION efficient since it avoids caching the encrypted and decrypted pages in the page cache. +config EXT4_FS_ENCRYPTION + bool + default y + depends on EXT4_ENCRYPTION + config EXT4_DEBUG bool "EXT4 debugging support" depends on EXT4_FS -- cgit v0.10.2 From 9402bdcacdfedf7219a17e4d93300058a8e2aa4c Mon Sep 17 00:00:00 2001 From: Chanho Park Date: Sat, 2 May 2015 10:29:22 -0400 Subject: ext4 crypto: remove duplicated encryption mode definitions This patch removes duplicated encryption modes which were already in ext4.h. They were duplicated from commit 3edc18d and commit f542fb. Cc: Theodore Ts'o Cc: Michael Halcrow Cc: Andreas Dilger Signed-off-by: Chanho Park Signed-off-by: Theodore Ts'o diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index bca1bdc..1de8c7f 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1067,12 +1067,6 @@ extern void ext4_set_bits(void *bm, int cur, int len); /* Metadata checksum algorithm codes */ #define EXT4_CRC32C_CHKSUM 1 -/* Encryption algorithms */ -#define EXT4_ENCRYPTION_MODE_INVALID 0 -#define EXT4_ENCRYPTION_MODE_AES_256_XTS 1 -#define EXT4_ENCRYPTION_MODE_AES_256_GCM 2 -#define EXT4_ENCRYPTION_MODE_AES_256_CBC 3 - /* * Structure of the super block */ -- cgit v0.10.2 From d2dc317d564a46dfc683978a2e5a4f91434e9711 Mon Sep 17 00:00:00 2001 From: Lukas Czerner Date: Sat, 2 May 2015 21:36:55 -0400 Subject: ext4: fix data corruption caused by unwritten and delayed extents Currently it is possible to lose whole file system block worth of data when we hit the specific interaction with unwritten and delayed extents in status extent tree. The problem is that when we insert delayed extent into extent status tree the only way to get rid of it is when we write out delayed buffer. However there is a limitation in the extent status tree implementation so that when inserting unwritten extent should there be even a single delayed block the whole unwritten extent would be marked as delayed. At this point, there is no way to get rid of the delayed extents, because there are no delayed buffers to write out. So when a we write into said unwritten extent we will convert it to written, but it still remains delayed. When we try to write into that block later ext4_da_map_blocks() will set the buffer new and delayed and map it to invalid block which causes the rest of the block to be zeroed loosing already written data. For now we can fix this by simply not allowing to set delayed status on written extent in the extent status tree. Also add WARN_ON() to make sure that we notice if this happens in the future. This problem can be easily reproduced by running the following xfs_io. xfs_io -f -c "pwrite -S 0xaa 4096 2048" \ -c "falloc 0 131072" \ -c "pwrite -S 0xbb 65536 2048" \ -c "fsync" /mnt/test/fff echo 3 > /proc/sys/vm/drop_caches xfs_io -c "pwrite -S 0xdd 67584 2048" /mnt/test/fff This can be theoretically also reproduced by at random by running fsx, but it's not very reliable, though on machines with bigger page size (like ppc) this can be seen more often (especially xfstest generic/127) Signed-off-by: Lukas Czerner Signed-off-by: Theodore Ts'o Cc: stable@vger.kernel.org diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c index d33d5a6..26724ae 100644 --- a/fs/ext4/extents_status.c +++ b/fs/ext4/extents_status.c @@ -703,6 +703,14 @@ int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk, BUG_ON(end < lblk); + if ((status & EXTENT_STATUS_DELAYED) && + (status & EXTENT_STATUS_WRITTEN)) { + ext4_warning(inode->i_sb, "Inserting extent [%u/%u] as " + " delayed and written which can potentially " + " cause data loss.\n", lblk, len); + WARN_ON(1); + } + newes.es_lblk = lblk; newes.es_len = len; ext4_es_store_pblock_status(&newes, pblk, status); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index f6b35d8..4415cea 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -532,6 +532,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode, status = map->m_flags & EXT4_MAP_UNWRITTEN ? EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN; if (!(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) && + !(status & EXTENT_STATUS_WRITTEN) && ext4_find_delalloc_range(inode, map->m_lblk, map->m_lblk + map->m_len - 1)) status |= EXTENT_STATUS_DELAYED; @@ -636,6 +637,7 @@ found: status = map->m_flags & EXT4_MAP_UNWRITTEN ? EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN; if (!(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) && + !(status & EXTENT_STATUS_WRITTEN) && ext4_find_delalloc_range(inode, map->m_lblk, map->m_lblk + map->m_len - 1)) status |= EXTENT_STATUS_DELAYED; -- cgit v0.10.2 From 280227a75b56ab5d35854f3a77ef74a7ad56a203 Mon Sep 17 00:00:00 2001 From: Davide Italiano Date: Sat, 2 May 2015 23:21:15 -0400 Subject: ext4: move check under lock scope to close a race. fallocate() checks that the file is extent-based and returns EOPNOTSUPP in case is not. Other tasks can convert from and to indirect and extent so it's safe to check only after grabbing the inode mutex. Signed-off-by: Davide Italiano Signed-off-by: Theodore Ts'o Cc: stable@vger.kernel.org diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 973816b..d74e0802 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -4927,13 +4927,6 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) if (ret) return ret; - /* - * currently supporting (pre)allocate mode for extent-based - * files _only_ - */ - if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) - return -EOPNOTSUPP; - if (mode & FALLOC_FL_COLLAPSE_RANGE) return ext4_collapse_range(inode, offset, len); @@ -4955,6 +4948,14 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) mutex_lock(&inode->i_mutex); + /* + * We only support preallocation for extent-based files only + */ + if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { + ret = -EOPNOTSUPP; + goto out; + } + if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > i_size_read(inode)) { new_size = offset + len; -- cgit v0.10.2 From 2c869b262a10ca99cb866d04087d75311587a30c Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Sat, 2 May 2015 23:58:32 -0400 Subject: ext4: fix growing of tiny filesystems The estimate of necessary transaction credits in ext4_flex_group_add() is too pessimistic. It reserves credit for sb, resize inode, and resize inode dindirect block for each group added in a flex group although they are always the same block and thus it is enough to account them only once. Also the number of modified GDT block is overestimated since we fit EXT4_DESC_PER_BLOCK(sb) descriptors in one block. Make the estimation more precise. That reduces number of requested credits enough that we can grow 20 MB filesystem (which has 1 MB journal, 79 reserved GDT blocks, and flex group size 16 by default). Signed-off-by: Jan Kara Signed-off-by: Theodore Ts'o Reviewed-by: Eric Sandeen diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 8a8ec62..cf0c472 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -1432,12 +1432,15 @@ static int ext4_flex_group_add(struct super_block *sb, goto exit; /* * We will always be modifying at least the superblock and GDT - * block. If we are adding a group past the last current GDT block, + * blocks. If we are adding a group past the last current GDT block, * we will also modify the inode and the dindirect block. If we * are adding a group with superblock/GDT backups we will also * modify each of the reserved GDT dindirect blocks. */ - credit = flex_gd->count * 4 + reserved_gdb; + credit = 3; /* sb, resize inode, resize inode dindirect */ + /* GDT blocks */ + credit += 1 + DIV_ROUND_UP(flex_gd->count, EXT4_DESC_PER_BLOCK(sb)); + credit += reserved_gdb; /* Reserved GDT dindirect blocks */ handle = ext4_journal_start_sb(sb, EXT4_HT_RESIZE, credit); if (IS_ERR(handle)) { err = PTR_ERR(handle); -- cgit v0.10.2 From 9b071a43553d6b2df4364951639f61076a8dd676 Mon Sep 17 00:00:00 2001 From: Philippe Coval Date: Sat, 2 May 2015 15:14:08 +0200 Subject: ideapad_laptop: Add Lenovo G40-30 to devices without radio switch Lenovo G40-30 does not provide any physical radio switch to user. Therefore disable the rfkill switch identically to the Yoga 2 approach. (Note for later, models ids are sorted alphabetically). Benefit is to make wireless available again without unloading module. It was tested successfully on 4.1.0-rc1 base with this model: (LENOVO_MT_80FY_BU_idea_FM_Lenovo G40-30). BugLink: https://bugs.launchpad.net/ideapad-laptop/+bug/1450946 Cc: platform-driver-x86@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Philippe Coval Signed-off-by: Darren Hart diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index b3d419a..b496db8 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -830,6 +830,13 @@ static void ideapad_acpi_notify(acpi_handle handle, u32 event, void *data) */ static const struct dmi_system_id no_hw_rfkill_list[] = { { + .ident = "Lenovo G40-30", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo G40-30"), + }, + }, + { .ident = "Lenovo Yoga 2 11 / 13 / Pro", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), -- cgit v0.10.2 From 1424532b2163bf1580f4b1091a5801e12310fac5 Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Wed, 29 Apr 2015 11:29:19 +0100 Subject: ARM: 8347/1: dma-mapping: fix off-by-one check in arm_setup_iommu_dma_ops Patch 22b3c181c6c324a46f71aae806d8ddbe61d25761 ("arm: dma-mapping: limit IOMMU mapping size") added a check for IO address space size. However this patch broke IOMMU initialization for typical platforms initialized from device tree, which get the default IO address space size of 4GiB. This value doesn't fit into size_t and fails a check introduced by that commit resulting in failed dma-mapping/iommu initialization. This patch fixes this issue by adding proper support for full 4GiB address space size. Signed-off-by: Marek Szyprowski Acked-by: Will Deacon Signed-off-by: Russell King diff --git a/arch/arm/include/asm/dma-iommu.h b/arch/arm/include/asm/dma-iommu.h index 8e3fcb9..2ef282f 100644 --- a/arch/arm/include/asm/dma-iommu.h +++ b/arch/arm/include/asm/dma-iommu.h @@ -25,7 +25,7 @@ struct dma_iommu_mapping { }; struct dma_iommu_mapping * -arm_iommu_create_mapping(struct bus_type *bus, dma_addr_t base, size_t size); +arm_iommu_create_mapping(struct bus_type *bus, dma_addr_t base, u64 size); void arm_iommu_release_mapping(struct dma_iommu_mapping *mapping); diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c index 09c5fe3..7e7583d 100644 --- a/arch/arm/mm/dma-mapping.c +++ b/arch/arm/mm/dma-mapping.c @@ -1878,7 +1878,7 @@ struct dma_map_ops iommu_coherent_ops = { * arm_iommu_attach_device function. */ struct dma_iommu_mapping * -arm_iommu_create_mapping(struct bus_type *bus, dma_addr_t base, size_t size) +arm_iommu_create_mapping(struct bus_type *bus, dma_addr_t base, u64 size) { unsigned int bits = size >> PAGE_SHIFT; unsigned int bitmap_size = BITS_TO_LONGS(bits) * sizeof(long); @@ -1886,6 +1886,10 @@ arm_iommu_create_mapping(struct bus_type *bus, dma_addr_t base, size_t size) int extensions = 1; int err = -ENOMEM; + /* currently only 32-bit DMA address space is supported */ + if (size > DMA_BIT_MASK(32) + 1) + return ERR_PTR(-ERANGE); + if (!bitmap_size) return ERR_PTR(-EINVAL); @@ -2057,13 +2061,6 @@ static bool arm_setup_iommu_dma_ops(struct device *dev, u64 dma_base, u64 size, if (!iommu) return false; - /* - * currently arm_iommu_create_mapping() takes a max of size_t - * for size param. So check this limit for now. - */ - if (size > SIZE_MAX) - return false; - mapping = arm_iommu_create_mapping(dev->bus, dma_base, size); if (IS_ERR(mapping)) { pr_warn("Failed to create %llu-byte IOMMU mapping for device %s\n", -- cgit v0.10.2 From 024587dc2137a22096372fc8df4b283e84e057b5 Mon Sep 17 00:00:00 2001 From: Valentin Rothberg Date: Thu, 30 Apr 2015 14:34:35 +0100 Subject: ARM: 8348/1: remove comments on CPU_ARM1020_CPU_IDLE CPU_ARM1020_CPU_IDLE is not defined in Kconfig. The last reference on LKML dates back to 2001, so we can safely remove the comments to make static analysis tools happy. Signed-off-by: Valentin Rothberg Signed-off-by: Russell King diff --git a/arch/arm/mm/proc-arm1020.S b/arch/arm/mm/proc-arm1020.S index aa0519e..774ef13 100644 --- a/arch/arm/mm/proc-arm1020.S +++ b/arch/arm/mm/proc-arm1020.S @@ -22,8 +22,6 @@ * * These are the low level assembler for performing cache and TLB * functions on the arm1020. - * - * CONFIG_CPU_ARM1020_CPU_IDLE -> nohlt */ #include #include diff --git a/arch/arm/mm/proc-arm1020e.S b/arch/arm/mm/proc-arm1020e.S index bff4c7f..ae3c27b 100644 --- a/arch/arm/mm/proc-arm1020e.S +++ b/arch/arm/mm/proc-arm1020e.S @@ -22,8 +22,6 @@ * * These are the low level assembler for performing cache and TLB * functions on the arm1020e. - * - * CONFIG_CPU_ARM1020_CPU_IDLE -> nohlt */ #include #include -- cgit v0.10.2 From 13f2fa7cb33d82b95e1e1dab8c95fa619b49f418 Mon Sep 17 00:00:00 2001 From: Valentin Rothberg Date: Thu, 30 Apr 2015 14:36:03 +0100 Subject: ARM: 8349/1: arch/arm/mm/proc-arm925.S: remove dead #ifdef block The block could never be compiled; CPU_ICACHE_STREAMING_DISABLE has not been defined in Kconfig since the very first Git commit. Hence, we can safely remove the entire block. Signed-off-by: Valentin Rothberg Signed-off-by: Russell King diff --git a/arch/arm/mm/proc-arm925.S b/arch/arm/mm/proc-arm925.S index ede8c54..32a47cc 100644 --- a/arch/arm/mm/proc-arm925.S +++ b/arch/arm/mm/proc-arm925.S @@ -441,9 +441,6 @@ ENTRY(cpu_arm925_set_pte_ext) .type __arm925_setup, #function __arm925_setup: mov r0, #0 -#if defined(CONFIG_CPU_ICACHE_STREAMING_DISABLE) - orr r0,r0,#1 << 7 -#endif /* Transparent on, D-cache clean & flush mode. See NOTE2 above */ orr r0,r0,#1 << 1 @ transparent mode on -- cgit v0.10.2 From 59c3191628fd1ced948d099f3138b92b06281520 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Thu, 30 Apr 2015 19:51:43 +0100 Subject: ARM: 8350/1: proc-feroceon: Fix feroceon_proc_info macro bf35706f3d09 ("ARM: 8314/1: replace PROCINFO embedded branch with relative offset") broke booting for Kirkwood. The kernel would say: Starting kernel ... Uncompressing Linux... done, booting the kernel. Error: unrecognized/unsupported processor variant (0x56251311). Fix it by removing the extraneous .long __feroceon_setup from the feroceon_proc_info macro. Fixes: bf35706f3d09 ("ARM: 8314/1: replace PROCINFO embedded branch with relative offset") Reported-by: Florian Fainelli Suggested-by: Arnd Bergmann Signed-off-by: Andrew Lunn Tested-by: Florian Fainelli Acked-by: Ard Biesheuvel Tested-by: Aaro Koskinen Signed-off-by: Russell King diff --git a/arch/arm/mm/proc-feroceon.S b/arch/arm/mm/proc-feroceon.S index e494d6d..92e08bf 100644 --- a/arch/arm/mm/proc-feroceon.S +++ b/arch/arm/mm/proc-feroceon.S @@ -602,7 +602,6 @@ __\name\()_proc_info: PMD_SECT_AP_WRITE | \ PMD_SECT_AP_READ initfn __feroceon_setup, __\name\()_proc_info - .long __feroceon_setup .long cpu_arch_name .long cpu_elf_name .long HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP -- cgit v0.10.2 From 338d9dd3e2aee00a9198e8bf6e7d535d3feeaf32 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 1 May 2015 17:15:23 +0100 Subject: ARM: 8351/1: perf: don't warn about missing interrupt-affinity property for PPIs PPIs are affine by nature, so the interrupt-affinity property is not used and therefore we shouldn't print a warning in its absence. Reported-by: Maxime Ripard Reviewed-by: Maxime Ripard Signed-off-by: Will Deacon Signed-off-by: Russell King diff --git a/arch/arm/kernel/perf_event_cpu.c b/arch/arm/kernel/perf_event_cpu.c index 91c7ba1..becf7ad 100644 --- a/arch/arm/kernel/perf_event_cpu.c +++ b/arch/arm/kernel/perf_event_cpu.c @@ -303,12 +303,17 @@ static int probe_current_pmu(struct arm_pmu *pmu) static int of_pmu_irq_cfg(struct platform_device *pdev) { - int i; + int i, irq; int *irqs = kcalloc(pdev->num_resources, sizeof(*irqs), GFP_KERNEL); if (!irqs) return -ENOMEM; + /* Don't bother with PPIs; they're already affine */ + irq = platform_get_irq(pdev, 0); + if (irq >= 0 && irq_is_percpu(irq)) + return 0; + for (i = 0; i < pdev->num_resources; ++i) { struct device_node *dn; int cpu; -- cgit v0.10.2 From 3b8786ff7a1b31645ae2c26a2ec32dbd42ac1094 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 1 May 2015 17:16:01 +0100 Subject: ARM: 8352/1: perf: Fix the pmu node name in warning message With commit 9fd85eb502a7 ("ARM: pmu: add support for interrupt-affinity property"), we print a warning when we find a PMU SPI with a missing missing interrupt-affinity property in a pmu node. Unfortunately, we pass the wrong (NULL) device node to of_node_full_name, resulting in unhelpful messages such as: hw perfevents: Failed to parse /interrupt-affinity[0] This patch fixes the name to that of the pmu node. Fixes: 9fd85eb502a7 (ARM: pmu: add support for interrupt-affinity property) Acked-by: Mark Rutland Signed-off-by: Will Deacon Signed-off-by: Russell King diff --git a/arch/arm/kernel/perf_event_cpu.c b/arch/arm/kernel/perf_event_cpu.c index becf7ad..213919b 100644 --- a/arch/arm/kernel/perf_event_cpu.c +++ b/arch/arm/kernel/perf_event_cpu.c @@ -322,7 +322,7 @@ static int of_pmu_irq_cfg(struct platform_device *pdev) i); if (!dn) { pr_warn("Failed to parse %s/interrupt-affinity[%d]\n", - of_node_full_name(dn), i); + of_node_full_name(pdev->dev.of_node), i); break; } -- cgit v0.10.2 From 5ebe6afaf0057ac3eaeb98defd5456894b446d22 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 3 May 2015 19:22:23 -0700 Subject: Linux 4.1-rc2 diff --git a/Makefile b/Makefile index 7ff1239..2da553f 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 1 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc2 NAME = Hurr durr I'ma sheep # *DOCUMENTATION* -- cgit v0.10.2 From 184af16b09360d6273fd6160e6ff7f8e2482ef23 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Thu, 23 Apr 2015 13:43:43 +0300 Subject: mmc: core: add missing pm event in mmc_pm_notify to fix hib restore The PM_RESTORE_PREPARE is not handled now in mmc_pm_notify(), as result mmc_rescan() could be scheduled and executed at late hibernation restore stages when MMC device is suspended already - which, in turn, will lead to system crash on TI dra7-evm board: WARNING: CPU: 0 PID: 3188 at drivers/bus/omap_l3_noc.c:148 l3_interrupt_handler+0x258/0x374() 44000000.ocp:L3 Custom Error: MASTER MPU TARGET L4_PER1_P3 (Idle): Data Access in User mode during Functional access Hence, add missed PM_RESTORE_PREPARE PM event in mmc_pm_notify(). Fixes: 4c2ef25fe0b8 (mmc: fix all hangs related to mmc/sd card...) Signed-off-by: Grygorii Strashko Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index c296bc0..92e7671 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -2651,6 +2651,7 @@ int mmc_pm_notify(struct notifier_block *notify_block, switch (mode) { case PM_HIBERNATION_PREPARE: case PM_SUSPEND_PREPARE: + case PM_RESTORE_PREPARE: spin_lock_irqsave(&host->lock, flags); host->rescan_disable = 1; spin_unlock_irqrestore(&host->lock, flags); -- cgit v0.10.2 From 7829fb09a2b4268b30dd9bc782fa5ebee278b137 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 30 Apr 2015 04:13:52 +0200 Subject: lib: make memzero_explicit more robust against dead store elimination In commit 0b053c951829 ("lib: memzero_explicit: use barrier instead of OPTIMIZER_HIDE_VAR"), we made memzero_explicit() more robust in case LTO would decide to inline memzero_explicit() and eventually find out it could be elimiated as dead store. While using barrier() works well for the case of gcc, recent efforts from LLVMLinux people suggest to use llvm as an alternative to gcc, and there, Stephan found in a simple stand-alone user space example that llvm could nevertheless optimize and thus elimitate the memset(). A similar issue has been observed in the referenced llvm bug report, which is regarded as not-a-bug. Based on some experiments, icc is a bit special on its own, while it doesn't seem to eliminate the memset(), it could do so with an own implementation, and then result in similar findings as with llvm. The fix in this patch now works for all three compilers (also tested with more aggressive optimization levels). Arguably, in the current kernel tree it's more of a theoretical issue, but imho, it's better to be pedantic about it. It's clearly visible with gcc/llvm though, with the below code: if we would have used barrier() only here, llvm would have omitted clearing, not so with barrier_data() variant: static inline void memzero_explicit(void *s, size_t count) { memset(s, 0, count); barrier_data(s); } int main(void) { char buff[20]; memzero_explicit(buff, sizeof(buff)); return 0; } $ gcc -O2 test.c $ gdb a.out (gdb) disassemble main Dump of assembler code for function main: 0x0000000000400400 <+0>: lea -0x28(%rsp),%rax 0x0000000000400405 <+5>: movq $0x0,-0x28(%rsp) 0x000000000040040e <+14>: movq $0x0,-0x20(%rsp) 0x0000000000400417 <+23>: movl $0x0,-0x18(%rsp) 0x000000000040041f <+31>: xor %eax,%eax 0x0000000000400421 <+33>: retq End of assembler dump. $ clang -O2 test.c $ gdb a.out (gdb) disassemble main Dump of assembler code for function main: 0x00000000004004f0 <+0>: xorps %xmm0,%xmm0 0x00000000004004f3 <+3>: movaps %xmm0,-0x18(%rsp) 0x00000000004004f8 <+8>: movl $0x0,-0x8(%rsp) 0x0000000000400500 <+16>: lea -0x18(%rsp),%rax 0x0000000000400505 <+21>: xor %eax,%eax 0x0000000000400507 <+23>: retq End of assembler dump. As gcc, clang, but also icc defines __GNUC__, it's sufficient to define this in compiler-gcc.h only to be picked up. For a fallback or otherwise unsupported compiler, we define it as a barrier. Similarly, for ecc which does not support gcc inline asm. Reference: https://llvm.org/bugs/show_bug.cgi?id=15495 Reported-by: Stephan Mueller Tested-by: Stephan Mueller Signed-off-by: Daniel Borkmann Cc: Theodore Ts'o Cc: Stephan Mueller Cc: Hannes Frederic Sowa Cc: mancha security Cc: Mark Charlebois Cc: Behan Webster Signed-off-by: Herbert Xu diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index cdf13ca..371e560 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -9,10 +9,24 @@ + __GNUC_MINOR__ * 100 \ + __GNUC_PATCHLEVEL__) - /* Optimization barrier */ + /* The "volatile" is due to gcc bugs */ #define barrier() __asm__ __volatile__("": : :"memory") +/* + * This version is i.e. to prevent dead stores elimination on @ptr + * where gcc and llvm may behave differently when otherwise using + * normal barrier(): while gcc behavior gets along with a normal + * barrier(), llvm needs an explicit input variable to be assumed + * clobbered. The issue is as follows: while the inline asm might + * access any memory it wants, the compiler could have fit all of + * @ptr into memory registers instead, and since @ptr never escaped + * from that, it proofed that the inline asm wasn't touching any of + * it. This version works well with both compilers, i.e. we're telling + * the compiler that the inline asm absolutely may see the contents + * of @ptr. See also: https://llvm.org/bugs/show_bug.cgi?id=15495 + */ +#define barrier_data(ptr) __asm__ __volatile__("": :"r"(ptr) :"memory") /* * This macro obfuscates arithmetic on a variable address so that gcc diff --git a/include/linux/compiler-intel.h b/include/linux/compiler-intel.h index ba147a1..0c9a2f2 100644 --- a/include/linux/compiler-intel.h +++ b/include/linux/compiler-intel.h @@ -13,9 +13,12 @@ /* Intel ECC compiler doesn't support gcc specific asm stmts. * It uses intrinsics to do the equivalent things. */ +#undef barrier_data #undef RELOC_HIDE #undef OPTIMIZER_HIDE_VAR +#define barrier_data(ptr) barrier() + #define RELOC_HIDE(ptr, off) \ ({ unsigned long __ptr; \ __ptr = (unsigned long) (ptr); \ diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 0e41ca0..8677225 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -169,6 +169,10 @@ void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect); # define barrier() __memory_barrier() #endif +#ifndef barrier_data +# define barrier_data(ptr) barrier() +#endif + /* Unreachable code */ #ifndef unreachable # define unreachable() do { } while (1) diff --git a/lib/string.c b/lib/string.c index a579201..bb3d4b6 100644 --- a/lib/string.c +++ b/lib/string.c @@ -607,7 +607,7 @@ EXPORT_SYMBOL(memset); void memzero_explicit(void *s, size_t count) { memset(s, 0, count); - barrier(); + barrier_data(s); } EXPORT_SYMBOL(memzero_explicit); -- cgit v0.10.2 From f440c4ee3e53f767974fe60bcbc0b6687a5fb53f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=C3=81lvaro=20Fern=C3=A1ndez=20Rojas?= Date: Sat, 2 May 2015 12:08:42 +0200 Subject: hwrng: bcm63xx - Fix driver compilation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - s/clk_didsable_unprepare/clk_disable_unprepare - s/prov/priv - s/error/ret (bcm63xx_rng_probe) Fixes: 6229c16060fe ("hwrng: bcm63xx - make use of devm_hwrng_register") Signed-off-by: Álvaro Fernández Rojas Acked-by: Florian Fainelli Signed-off-by: Herbert Xu diff --git a/drivers/char/hw_random/bcm63xx-rng.c b/drivers/char/hw_random/bcm63xx-rng.c index d1494ec..4b31f13 100644 --- a/drivers/char/hw_random/bcm63xx-rng.c +++ b/drivers/char/hw_random/bcm63xx-rng.c @@ -57,7 +57,7 @@ static void bcm63xx_rng_cleanup(struct hwrng *rng) val &= ~RNG_EN; __raw_writel(val, priv->regs + RNG_CTRL); - clk_didsable_unprepare(prov->clk); + clk_disable_unprepare(priv->clk); } static int bcm63xx_rng_data_present(struct hwrng *rng, int wait) @@ -97,14 +97,14 @@ static int bcm63xx_rng_probe(struct platform_device *pdev) priv->rng.name = pdev->name; priv->rng.init = bcm63xx_rng_init; priv->rng.cleanup = bcm63xx_rng_cleanup; - prov->rng.data_present = bcm63xx_rng_data_present; + priv->rng.data_present = bcm63xx_rng_data_present; priv->rng.data_read = bcm63xx_rng_data_read; priv->clk = devm_clk_get(&pdev->dev, "ipsec"); if (IS_ERR(priv->clk)) { - error = PTR_ERR(priv->clk); - dev_err(&pdev->dev, "no clock for device: %d\n", error); - return error; + ret = PTR_ERR(priv->clk); + dev_err(&pdev->dev, "no clock for device: %d\n", ret); + return ret; } if (!devm_request_mem_region(&pdev->dev, r->start, @@ -120,11 +120,11 @@ static int bcm63xx_rng_probe(struct platform_device *pdev) return -ENOMEM; } - error = devm_hwrng_register(&pdev->dev, &priv->rng); - if (error) { + ret = devm_hwrng_register(&pdev->dev, &priv->rng); + if (ret) { dev_err(&pdev->dev, "failed to register rng device: %d\n", - error); - return error; + ret); + return ret; } dev_info(&pdev->dev, "registered RNG driver\n"); -- cgit v0.10.2 From ec6bfca835b61df360881c1bf89268f69fed2a61 Mon Sep 17 00:00:00 2001 From: Kailang Yang Date: Mon, 4 May 2015 15:50:47 +0800 Subject: ALSA: hda/realtek - Fix typo for ALC286/ALC288 Delete more one break for ALC286/ALC288. Signed-off-by: Kailang Yang Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index eb2fc79..516a0da 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -3738,7 +3738,6 @@ static void alc_headset_mode_default(struct hda_codec *codec) case 0x10ec0288: alc_process_coef_fw(codec, coef0288); break; - break; case 0x10ec0292: alc_process_coef_fw(codec, coef0292); break; -- cgit v0.10.2 From a00212e21928640486d3cc939cf4d908e8522016 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 4 May 2015 01:58:27 +0200 Subject: ACPI / documentation: Fix ambiguity in the GPIO properties document The first paragraph in Documentation/acpi/gpio-properties.txt is ambiguous, so make it more clear. Reported-by: Antonio Ospite Acked-by: Antonio Ospite Acked-by: Mika Westerberg Signed-off-by: Rafael J. Wysocki diff --git a/Documentation/acpi/gpio-properties.txt b/Documentation/acpi/gpio-properties.txt index ae36fcf..f35dad1 100644 --- a/Documentation/acpi/gpio-properties.txt +++ b/Documentation/acpi/gpio-properties.txt @@ -1,9 +1,9 @@ _DSD Device Properties Related to GPIO -------------------------------------- -With the release of ACPI 5.1 and the _DSD configuration objecte names -can finally be given to GPIOs (and other things as well) returned by -_CRS. Previously, we were only able to use an integer index to find +With the release of ACPI 5.1, the _DSD configuration object finally +allows names to be given to GPIOs (and other things as well) returned +by _CRS. Previously, we were only able to use an integer index to find the corresponding GPIO, which is pretty error prone (it depends on the _CRS output ordering, for example). -- cgit v0.10.2 From 74d77e50f23123938fbb7987eba71310864e6a7c Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Mon, 20 Apr 2015 10:59:17 -0500 Subject: pinctrl: mediatek: mtk-common: initialize unmask cppcheck detected an uninitialized variable: [drivers/pinctrl/mediatek/pinctrl-mtk-common.c:897]: (error) Uninitialized variable: unmask unmask should be initialized to zero to ensure unmasking only occurs if a previous mask occurred. The current situation is that the unmask variable could contain any random garbage causing random unexpected unmasking. Signed-off-by: Colin Ian King Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c index 493294c..474812e 100644 --- a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c +++ b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c @@ -881,6 +881,8 @@ static int mtk_gpio_set_debounce(struct gpio_chip *chip, unsigned offset, if (!mtk_eint_get_mask(pctl, eint_num)) { mtk_eint_mask(d); unmask = 1; + } else { + unmask = 0; } clr_bit = 0xff << eint_offset; -- cgit v0.10.2 From 622532bb2fad8fe342fb685727ae0be566f6be5d Mon Sep 17 00:00:00 2001 From: Witold Szczeponik Date: Fri, 1 May 2015 19:05:20 +0200 Subject: ACPI / PNP: add two IDs to list for PNPACPI device enumeration Commit eec15edbb0e1 (ACPI / PNP: use device ID list for PNPACPI device enumeration) changed the way how ACPI devices are enumerated and when they are added to the PNP bus. However, it broke the sound card support on (at least) a vintage IBM ThinkPad 600E: with said commit applied, two of the necessary "CSC01xx" devices are not added to the PNP bus and hence can not be found during the initialization of the "snd-cs4236" module. As a consequence, loading "snd-cs4236" causes null pointer exceptions. The attached patch fixes the problem end re-enables sound on the IBM ThinkPad 600E. Fixes: eec15edbb0e1 (ACPI / PNP: use device ID list for PNPACPI device enumeration) Signed-off-by: Witold Szczeponik Cc: 3.16+ # 3.16+ Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/acpi_pnp.c b/drivers/acpi/acpi_pnp.c index b193f84..ff6d8ad 100644 --- a/drivers/acpi/acpi_pnp.c +++ b/drivers/acpi/acpi_pnp.c @@ -304,6 +304,8 @@ static const struct acpi_device_id acpi_pnp_device_ids[] = { {"PNPb006"}, /* cs423x-pnpbios */ {"CSC0100"}, + {"CSC0103"}, + {"CSC0110"}, {"CSC0000"}, {"GIM0100"}, /* Guillemot Turtlebeach something appears to be cs4232 compatible */ /* es18xx-pnpbios */ -- cgit v0.10.2 From 810e4425c224af6be67dff68c8832af1b5a11f89 Mon Sep 17 00:00:00 2001 From: Vasily Khoruzhick Date: Sun, 3 May 2015 21:11:47 +0300 Subject: ASoC: uda1380: Avoid accessing i2c bus when codec is disabled set_dai_fmt_both() callback is called from snd_soc_runtime_set_dai_fmt() which is called from snd_soc_register_card(), but at this time codec is not powered on yet. Replace direct i2c write with regcache write. Fixes: 5f0acedddf533c (ASoC: rx1950_uda1380: Use static DAI format setup) Fixes: 5cc10b9b77c234 (ASoC: h1940_uda1380: Use static DAI format setup) Signed-off-by: Vasily Khoruzhick Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/codecs/uda1380.c b/sound/soc/codecs/uda1380.c index dc7778b..c3c33bd 100644 --- a/sound/soc/codecs/uda1380.c +++ b/sound/soc/codecs/uda1380.c @@ -437,7 +437,7 @@ static int uda1380_set_dai_fmt_both(struct snd_soc_dai *codec_dai, if ((fmt & SND_SOC_DAIFMT_MASTER_MASK) != SND_SOC_DAIFMT_CBS_CFS) return -EINVAL; - uda1380_write(codec, UDA1380_IFACE, iface); + uda1380_write_reg_cache(codec, UDA1380_IFACE, iface); return 0; } -- cgit v0.10.2 From 102bcb6ed2d1c3ffcc7269afc957c2df11942085 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Mon, 4 May 2015 08:54:41 -0700 Subject: ARM: OMAP2+: Fix omap off idle power consumption creeping up If we use a combination of VMODE and I2C4 for retention modes, eventually the off idle power consumption will creep up by about 23mW, even during off mode with I2C4 always staying enabled. Turns out this is because of erratum i531 "Extra Power Consumed When Repeated Start Operation Mode Is Enabled on I2C Interface Dedicated for Smart Reflex (I2C4)" as pointed out by Nishanth Menon . Let's fix the issue by adding i2c_cfg_clear_mask for the bits to clear when initializing the I2C4 adapter so we can clear SREN bit that drives the I2C4 lines low otherwise when there is no traffic. Fixes: 3b8c4ebb7630 ("ARM: OMAP3: Fix idle mode signaling for Cc: stable@vger.kernel.org # v3.16+ sys_clkreq and sys_off_mode") Cc: Kevin Hilman Cc: Tero Kristo Reviewed-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/arch/arm/mach-omap2/prm-regbits-34xx.h b/arch/arm/mach-omap2/prm-regbits-34xx.h index cbefbd7..661d753 100644 --- a/arch/arm/mach-omap2/prm-regbits-34xx.h +++ b/arch/arm/mach-omap2/prm-regbits-34xx.h @@ -112,6 +112,7 @@ #define OMAP3430_VC_CMD_ONLP_SHIFT 16 #define OMAP3430_VC_CMD_RET_SHIFT 8 #define OMAP3430_VC_CMD_OFF_SHIFT 0 +#define OMAP3430_SREN_MASK (1 << 4) #define OMAP3430_HSEN_MASK (1 << 3) #define OMAP3430_MCODE_MASK (0x7 << 0) #define OMAP3430_VALID_MASK (1 << 24) diff --git a/arch/arm/mach-omap2/prm-regbits-44xx.h b/arch/arm/mach-omap2/prm-regbits-44xx.h index b1c7a33..e794828 100644 --- a/arch/arm/mach-omap2/prm-regbits-44xx.h +++ b/arch/arm/mach-omap2/prm-regbits-44xx.h @@ -35,6 +35,7 @@ #define OMAP4430_GLOBAL_WARM_SW_RST_SHIFT 1 #define OMAP4430_GLOBAL_WUEN_MASK (1 << 16) #define OMAP4430_HSMCODE_MASK (0x7 << 0) +#define OMAP4430_SRMODEEN_MASK (1 << 4) #define OMAP4430_HSMODEEN_MASK (1 << 3) #define OMAP4430_HSSCLL_SHIFT 24 #define OMAP4430_ICEPICK_RST_SHIFT 9 diff --git a/arch/arm/mach-omap2/vc.c b/arch/arm/mach-omap2/vc.c index be9ef83..076fd20 100644 --- a/arch/arm/mach-omap2/vc.c +++ b/arch/arm/mach-omap2/vc.c @@ -316,7 +316,8 @@ static void __init omap3_vc_init_pmic_signaling(struct voltagedomain *voltdm) * idle. And we can also scale voltages to zero for off-idle. * Note that no actual voltage scaling during off-idle will * happen unless the board specific twl4030 PMIC scripts are - * loaded. + * loaded. See also omap_vc_i2c_init for comments regarding + * erratum i531. */ val = voltdm->read(OMAP3_PRM_VOLTCTRL_OFFSET); if (!(val & OMAP3430_PRM_VOLTCTRL_SEL_OFF)) { @@ -704,9 +705,16 @@ static void __init omap_vc_i2c_init(struct voltagedomain *voltdm) return; } + /* + * Note that for omap3 OMAP3430_SREN_MASK clears SREN to work around + * erratum i531 "Extra Power Consumed When Repeated Start Operation + * Mode Is Enabled on I2C Interface Dedicated for Smart Reflex (I2C4)". + * Otherwise I2C4 eventually leads into about 23mW extra power being + * consumed even during off idle using VMODE. + */ i2c_high_speed = voltdm->pmic->i2c_high_speed; if (i2c_high_speed) - voltdm->rmw(vc->common->i2c_cfg_hsen_mask, + voltdm->rmw(vc->common->i2c_cfg_clear_mask, vc->common->i2c_cfg_hsen_mask, vc->common->i2c_cfg_reg); diff --git a/arch/arm/mach-omap2/vc.h b/arch/arm/mach-omap2/vc.h index cdbdd78..89b83b7 100644 --- a/arch/arm/mach-omap2/vc.h +++ b/arch/arm/mach-omap2/vc.h @@ -34,6 +34,7 @@ struct voltagedomain; * @cmd_ret_shift: RET field shift in PRM_VC_CMD_VAL_* register * @cmd_off_shift: OFF field shift in PRM_VC_CMD_VAL_* register * @i2c_cfg_reg: I2C configuration register offset + * @i2c_cfg_clear_mask: high-speed mode bit clear mask in I2C config register * @i2c_cfg_hsen_mask: high-speed mode bit field mask in I2C config register * @i2c_mcode_mask: MCODE field mask for I2C config register * @@ -52,6 +53,7 @@ struct omap_vc_common { u8 cmd_ret_shift; u8 cmd_off_shift; u8 i2c_cfg_reg; + u8 i2c_cfg_clear_mask; u8 i2c_cfg_hsen_mask; u8 i2c_mcode_mask; }; diff --git a/arch/arm/mach-omap2/vc3xxx_data.c b/arch/arm/mach-omap2/vc3xxx_data.c index 75bc4aa..71d74c9 100644 --- a/arch/arm/mach-omap2/vc3xxx_data.c +++ b/arch/arm/mach-omap2/vc3xxx_data.c @@ -40,6 +40,7 @@ static struct omap_vc_common omap3_vc_common = { .cmd_onlp_shift = OMAP3430_VC_CMD_ONLP_SHIFT, .cmd_ret_shift = OMAP3430_VC_CMD_RET_SHIFT, .cmd_off_shift = OMAP3430_VC_CMD_OFF_SHIFT, + .i2c_cfg_clear_mask = OMAP3430_SREN_MASK | OMAP3430_HSEN_MASK, .i2c_cfg_hsen_mask = OMAP3430_HSEN_MASK, .i2c_cfg_reg = OMAP3_PRM_VC_I2C_CFG_OFFSET, .i2c_mcode_mask = OMAP3430_MCODE_MASK, diff --git a/arch/arm/mach-omap2/vc44xx_data.c b/arch/arm/mach-omap2/vc44xx_data.c index 085e5d6..2abd5fa 100644 --- a/arch/arm/mach-omap2/vc44xx_data.c +++ b/arch/arm/mach-omap2/vc44xx_data.c @@ -42,6 +42,7 @@ static const struct omap_vc_common omap4_vc_common = { .cmd_ret_shift = OMAP4430_RET_SHIFT, .cmd_off_shift = OMAP4430_OFF_SHIFT, .i2c_cfg_reg = OMAP4_PRM_VC_CFG_I2C_MODE_OFFSET, + .i2c_cfg_clear_mask = OMAP4430_SRMODEEN_MASK | OMAP4430_HSMODEEN_MASK, .i2c_cfg_hsen_mask = OMAP4430_HSMODEEN_MASK, .i2c_mcode_mask = OMAP4430_HSMCODE_MASK, }; -- cgit v0.10.2 From 1819e3034ee26ffadc71880064ed8b8e7d74f52c Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Sun, 1 Mar 2015 21:07:08 +0200 Subject: ARM: dts: OMAP3-N900: Add microphone bias voltages N900 audio recording needs that codec provides bias voltage for integrated digital microphone and headset microphone depending which one is used. Digital microphone uses 2 V bias and it comes from the codec A part. Codec B part drives the headset microphone bias and that is set to 2.5 V. Cc: stable@vger.kernel.org # v3.16+ Signed-off-by: Pavel Machek [Jarkko: Headset mic bias changed to 2 (2.5 V) as it was before commit e2e8bfdf6157 ("ASoC: tlv320aic3x: Convert mic bias to a supply widget")] Signed-off-by: Jarkko Nikula Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/omap3-n900.dts b/arch/arm/boot/dts/omap3-n900.dts index a293158..5c16145 100644 --- a/arch/arm/boot/dts/omap3-n900.dts +++ b/arch/arm/boot/dts/omap3-n900.dts @@ -498,6 +498,8 @@ DRVDD-supply = <&vmmc2>; IOVDD-supply = <&vio>; DVDD-supply = <&vio>; + + ai3x-micbias-vg = <1>; }; tlv320aic3x_aux: tlv320aic3x@19 { @@ -509,6 +511,8 @@ DRVDD-supply = <&vmmc2>; IOVDD-supply = <&vio>; DVDD-supply = <&vio>; + + ai3x-micbias-vg = <2>; }; tsl2563: tsl2563@29 { -- cgit v0.10.2 From 67defd5c0ee5270948104784c8fc8f6d4a99682b Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Tue, 24 Mar 2015 12:19:19 +0200 Subject: ARM: omap2plus_defconfig: Enable EXTCON_USB_GPIO We need to enable EXTCON_USB_GPIO_USB and not EXTCON_GPIO_USB. Fixes: c08a54c0ebeb ("ARM: omap2plus_defconfig: Enable EXTCON_GPIO_USB") Reported-by: Nishant Menon Signed-off-by: Roger Quadros Acked-by-by: Nishanth Menon Acked-by: Felipe Balbi Signed-off-by: Tony Lindgren diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig index 9ff7b54..3743ca2 100644 --- a/arch/arm/configs/omap2plus_defconfig +++ b/arch/arm/configs/omap2plus_defconfig @@ -393,7 +393,7 @@ CONFIG_TI_EDMA=y CONFIG_DMA_OMAP=y # CONFIG_IOMMU_SUPPORT is not set CONFIG_EXTCON=m -CONFIG_EXTCON_GPIO=m +CONFIG_EXTCON_USB_GPIO=m CONFIG_EXTCON_PALMAS=m CONFIG_TI_EMIF=m CONFIG_PWM=y -- cgit v0.10.2 From 2055088b5e17b450fcfb536568aee00eccd0ced5 Mon Sep 17 00:00:00 2001 From: Sebastian Reichel Date: Tue, 31 Mar 2015 03:28:10 +0200 Subject: ARM: dts: omap3: Add #iommu-cells to isp and iva iommu Add missing #iommu-cells property to the isp and iva iommu nodes. This fixes the binding (property is required according to the generic iommu binding) and removes the following kernel warning triggered once the iommu nodes are referenced: [ 0.647521] /ocp/isp@480bc000: could not get #iommu-cells for /ocp/mmu@480bd400 Signed-off-by: Sebastian Reichel Cc: Laurent Pinchart Acked-by: Sakari Ailus Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/omap3.dtsi b/arch/arm/boot/dts/omap3.dtsi index d18a90f..69a40cf 100644 --- a/arch/arm/boot/dts/omap3.dtsi +++ b/arch/arm/boot/dts/omap3.dtsi @@ -456,6 +456,7 @@ }; mmu_isp: mmu@480bd400 { + #iommu-cells = <0>; compatible = "ti,omap2-iommu"; reg = <0x480bd400 0x80>; interrupts = <24>; @@ -464,6 +465,7 @@ }; mmu_iva: mmu@5d000000 { + #iommu-cells = <0>; compatible = "ti,omap2-iommu"; reg = <0x5d000000 0x80>; interrupts = <28>; -- cgit v0.10.2 From e7a7357341664148cabf98a3ab50b5d449afca0b Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Wed, 8 Apr 2015 18:56:26 -0500 Subject: ARM: dts: am57xx-beagle-x15: Fix IRQ type for mcp7941x The interrupt polarity provided in devicetree is used to configure the interrupt controller(ARM GIC), however, it seems that we have an inverter at the GIC boundary inside AM57xx which inverts the signal input from sys_irq external interrupt source. Further, as per GIC distributor TRM, http://infocenter.arm.com/help/topic/com.arm.doc.ddi0438d/BGBHIACJ.html#BABJFCFB ARM GIC distributor does not support IRQ trigger type IRQ_TYPE_LEVEL_LOW, and only rising or level high signals. However, for some reason, the current configuration(which gets ignored by GIC driver) functions on some platforms, however, on few platforms results in infinite interrupts hogging the system down. Switch over to rising edge for GIC configuration which is also aligned with trigger point from the RTC chip and the internal inversion. Fixes: 5a0f93c6576a ("ARM: dts: Add am57xx-beagle-x15") Signed-off-by: Grygorii Strashko Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index 15f198e..7ac6ee2 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -455,7 +455,7 @@ mcp_rtc: rtc@6f { compatible = "microchip,mcp7941x"; reg = <0x6f>; - interrupts = ; /* IRQ_SYS_1N */ + interrupts = ; /* IRQ_SYS_1N */ pinctrl-names = "default"; pinctrl-0 = <&mcp79410_pins_default>; -- cgit v0.10.2 From 00edd3170c757772f74c533e61298be249447a11 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Wed, 8 Apr 2015 18:56:27 -0500 Subject: ARM: dts: am57xx-beagle-x15: Fix RTC aliases With commit bc078316d86c ("ARM: dts: DRA7: Add node for RTC"), we now have AM57xx RTC register itself as alias 0 even before DS1307 or TPS rtc drivers are loaded up. However, since neither TPS, nor AM57xx RTC are capable of being backedup by battery, we would like to maintain the "primary" rtc as mcp79410 rtc device. This also generates the following warnings in the bootlog highlighting the issue: [ 5.895445] rtc-ds1307 2-006f: /aliases ID 0 not available ... [ 6.476285] palmas-rtc 48070000.i2c:tps659038@58:tps659038_rtc: /aliases ID 1 not available So, add proper aliases to ensure that RTC order is always consistent to userspace immaterial of probe order. Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index 7ac6ee2..be9b6a5 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -18,6 +18,7 @@ aliases { rtc0 = &mcp_rtc; rtc1 = &tps659038_rtc; + rtc2 = &rtc; }; memory { diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi index 5332b57..0f73fea 100644 --- a/arch/arm/boot/dts/dra7.dtsi +++ b/arch/arm/boot/dts/dra7.dtsi @@ -1203,7 +1203,7 @@ status = "disabled"; }; - rtc@48838000 { + rtc: rtc@48838000 { compatible = "ti,am3352-rtc"; reg = <0x48838000 0x100>; interrupts = , -- cgit v0.10.2 From faa4ec1eedbe6d352dc22c945540a64b5c896a95 Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Thu, 9 Apr 2015 10:59:26 -0500 Subject: ARM: dts: am437x-sk: fix for new newhaven display module revision AM437x Starter Kit uses a NewHaven Display module with a 4.3" display and EDT FT5306 touchscreen On that module's new revision, NewHave decided to change the pinout on the 6 pin flat-pcb touchscreen connector so that instead of having WAKE pin, we now have RESETn. The new display module is available on AM437x SK Beta and all new revisions while the older revision is only available on AM437x SK Alpha which, unfortunately, can't be supported anymore in mainline without a revert of this patch. Signed-off-by: Felipe Balbi Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/am437x-sk-evm.dts b/arch/arm/boot/dts/am437x-sk-evm.dts index 8ae29c9..fbc50f9 100644 --- a/arch/arm/boot/dts/am437x-sk-evm.dts +++ b/arch/arm/boot/dts/am437x-sk-evm.dts @@ -473,7 +473,7 @@ interrupt-parent = <&gpio0>; interrupts = <31 0>; - wake-gpios = <&gpio1 28 GPIO_ACTIVE_HIGH>; + reset-gpios = <&gpio1 28 GPIO_ACTIVE_LOW>; touchscreen-size-x = <480>; touchscreen-size-y = <272>; -- cgit v0.10.2 From f6b957fdce50ddbd1899948ab109cca64e571eba Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Thu, 9 Apr 2015 10:59:27 -0500 Subject: ARM: dts: am437x-sk: reduce col-scan-delay-us The new AM437x SK Beta boards have removed the large capacitors on the gpio-matrix column lines which means we can reduce col-scan-delay-us to 5us without loosing functionality. Signed-off-by: Felipe Balbi Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/am437x-sk-evm.dts b/arch/arm/boot/dts/am437x-sk-evm.dts index fbc50f9..c17097d 100644 --- a/arch/arm/boot/dts/am437x-sk-evm.dts +++ b/arch/arm/boot/dts/am437x-sk-evm.dts @@ -49,7 +49,7 @@ pinctrl-0 = <&matrix_keypad_pins>; debounce-delay-ms = <5>; - col-scan-delay-us = <1500>; + col-scan-delay-us = <5>; row-gpios = <&gpio5 5 GPIO_ACTIVE_HIGH /* Bank5, pin5 */ &gpio5 6 GPIO_ACTIVE_HIGH>; /* Bank5, pin6 */ -- cgit v0.10.2 From 5eb6719816faad8e915d4f46ad61a440d641d183 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Wed, 8 Apr 2015 17:40:59 -0500 Subject: ARM: dts: am57xx-beagle-x15: Switch UART mux pins BeagleBoard-X15 pre-production change includes switching over to UART pins that now allow for UART download capability. All original boards should either have been returned for modifications or already modified for the required change and maintaining compatibility for older boards are no longer needed. Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index be9b6a5..dd53ea5 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -131,8 +131,8 @@ uart3_pins_default: uart3_pins_default { pinctrl-single,pins = < - 0x248 (PIN_INPUT_SLEW | MUX_MODE0) /* uart3_rxd.rxd */ - 0x24c (PIN_INPUT_SLEW | MUX_MODE0) /* uart3_txd.txd */ + 0x3f8 (PIN_INPUT_SLEW | MUX_MODE2) /* uart2_ctsn.uart3_rxd */ + 0x3fc (PIN_INPUT_SLEW | MUX_MODE1) /* uart2_rtsn.uart3_txd */ >; }; @@ -479,7 +479,7 @@ &uart3 { status = "okay"; interrupts-extended = <&crossbar_mpu GIC_SPI 69 IRQ_TYPE_LEVEL_HIGH>, - <&dra7_pmx_core 0x248>; + <&dra7_pmx_core 0x3f8>; pinctrl-names = "default"; pinctrl-0 = <&uart3_pins_default>; -- cgit v0.10.2 From ed12f102ba9f55d2617fc6b5e4fbecbf9a132ba7 Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Thu, 9 Apr 2015 17:33:17 -0500 Subject: ARM: dts: am57xx-beagle-x15: Switch GPIO fan number BeagleBoard-X15 pre-production change includes switching the GPIO fan gpio over from 1 to 2 to allow for a potential fix at a later point in time for USB client VBUS detection using PMIC VBUS detect capability. Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts index dd53ea5..7128fad 100644 --- a/arch/arm/boot/dts/am57xx-beagle-x15.dts +++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts @@ -84,7 +84,7 @@ gpio_fan: gpio_fan { /* Based on 5v 500mA AFB02505HHB */ compatible = "gpio-fan"; - gpios = <&tps659038_gpio 1 GPIO_ACTIVE_HIGH>; + gpios = <&tps659038_gpio 2 GPIO_ACTIVE_HIGH>; gpio-fan,speed-map = <0 0>, <13000 1>; #cooling-cells = <2>; -- cgit v0.10.2 From 1822734677d95888d18e3c0e0743727effa62f8e Mon Sep 17 00:00:00 2001 From: Nishanth Menon Date: Thu, 16 Apr 2015 16:56:33 -0500 Subject: ARM: dts: dra7: Fix efuse register size for ABB Fix a typo in DRA7 dtsi where 12 bytes are needed for register description of ABB efuse registers, however only 8 bytes are provided to map. For some weird reason, this does not generate abort at offset 0x8, probably due to default maps already provided in io.c for the bus register ranges. Reported-by: Matt Gessner Reported-by: Suman Anna Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi index 0f73fea..f03a091 100644 --- a/arch/arm/boot/dts/dra7.dtsi +++ b/arch/arm/boot/dts/dra7.dtsi @@ -911,7 +911,7 @@ ti,clock-cycles = <16>; reg = <0x4ae07ddc 0x4>, <0x4ae07de0 0x4>, - <0x4ae06014 0x4>, <0x4a003b20 0x8>, + <0x4ae06014 0x4>, <0x4a003b20 0xc>, <0x4ae0c158 0x4>; reg-names = "setup-address", "control-address", "int-address", "efuse-address", @@ -944,7 +944,7 @@ ti,clock-cycles = <16>; reg = <0x4ae07e34 0x4>, <0x4ae07e24 0x4>, - <0x4ae06010 0x4>, <0x4a0025cc 0x8>, + <0x4ae06010 0x4>, <0x4a0025cc 0xc>, <0x4a002470 0x4>; reg-names = "setup-address", "control-address", "int-address", "efuse-address", @@ -977,7 +977,7 @@ ti,clock-cycles = <16>; reg = <0x4ae07e30 0x4>, <0x4ae07e20 0x4>, - <0x4ae06010 0x4>, <0x4a0025e0 0x8>, + <0x4ae06010 0x4>, <0x4a0025e0 0xc>, <0x4a00246c 0x4>; reg-names = "setup-address", "control-address", "int-address", "efuse-address", @@ -1010,7 +1010,7 @@ ti,clock-cycles = <16>; reg = <0x4ae07de4 0x4>, <0x4ae07de8 0x4>, - <0x4ae06010 0x4>, <0x4a003b08 0x8>, + <0x4ae06010 0x4>, <0x4a003b08 0xc>, <0x4ae0c154 0x4>; reg-names = "setup-address", "control-address", "int-address", "efuse-address", -- cgit v0.10.2 From 4adf82c35572c69e96997641612fc88463b08f6f Mon Sep 17 00:00:00 2001 From: Illia Smyrnov Date: Thu, 16 Apr 2015 17:42:30 -0500 Subject: bus: omap_l3_noc: Fix offset for DRA7 CLK1_HOST_CLK1_2 instance The base address for DRA7 CLK1_HOST_CLK1_2 host instance is 0x44800000, so correct offset is 0x800000. DRA7 TRM rev X(fewb 2015) has updates for this information. With wrong offset these errors are not correctly cleared by the L3 IRQ handler and cause an continuous interrupt scenario and system lockup. Signed-off-by: Illia Smyrnov Signed-off-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/drivers/bus/omap_l3_noc.h b/drivers/bus/omap_l3_noc.h index 9525458..a314d80 100644 --- a/drivers/bus/omap_l3_noc.h +++ b/drivers/bus/omap_l3_noc.h @@ -274,7 +274,7 @@ static struct l3_flagmux_data dra_l3_flagmux_clk1 = { static struct l3_target_data dra_l3_target_data_clk2[] = { {0x0, "HOST CLK1",}, - {0x0, "HOST CLK2",}, + {0x800000, "HOST CLK2",}, {0xdead, L3_TARGET_NOT_SUPPORTED,}, {0x3400, "SHA2_2",}, {0x0900, "BB2D",}, -- cgit v0.10.2 From e7309c2673a389a495fcfad70376d3bae8b9bc89 Mon Sep 17 00:00:00 2001 From: Suman Anna Date: Fri, 24 Apr 2015 12:54:20 -0500 Subject: bus: omap_l3_noc: Fix master id address decoding for OMAP5 The L3 Error handling on OMAP5 for the most part is very similar to that of OMAP4, and had leveraged common data structures and register layout definitions so far. Upon closer inspection, there are a few minor differences causing an incorrect decoding and reporting of the master NIU upon an error: 1. The L3_TARG_STDERRLOG_MSTADDR.STDERRLOG_MSTADDR occupies 11 bits on OMAP5 as against 8 bits on OMAP4, with the master NIU connID encoded in the 6 MSBs of the STDERRLOG_MSTADDR field. 2. The CLK3 FlagMux component has 1 input source on OMAP4 and 3 input sources on OMAP5. The common DEBUGSS source is at a different input on each SoC. Fix the above issues by using a OMAP5-specific compatible property and using SoC-specific data where there are differences. Signed-off-by: Suman Anna Acked-by: Nishanth Menon Signed-off-by: Tony Lindgren diff --git a/Documentation/devicetree/bindings/arm/omap/l3-noc.txt b/Documentation/devicetree/bindings/arm/omap/l3-noc.txt index 974624e..161448d 100644 --- a/Documentation/devicetree/bindings/arm/omap/l3-noc.txt +++ b/Documentation/devicetree/bindings/arm/omap/l3-noc.txt @@ -6,6 +6,7 @@ provided by Arteris. Required properties: - compatible : Should be "ti,omap3-l3-smx" for OMAP3 family Should be "ti,omap4-l3-noc" for OMAP4 family + Should be "ti,omap5-l3-noc" for OMAP5 family Should be "ti,dra7-l3-noc" for DRA7 family Should be "ti,am4372-l3-noc" for AM43 family - reg: Contains L3 register address range for each noc domain. diff --git a/arch/arm/boot/dts/omap5.dtsi b/arch/arm/boot/dts/omap5.dtsi index efe5f73..7d24ae0 100644 --- a/arch/arm/boot/dts/omap5.dtsi +++ b/arch/arm/boot/dts/omap5.dtsi @@ -128,7 +128,7 @@ * hierarchy. */ ocp { - compatible = "ti,omap4-l3-noc", "simple-bus"; + compatible = "ti,omap5-l3-noc", "simple-bus"; #address-cells = <1>; #size-cells = <1>; ranges; diff --git a/drivers/bus/omap_l3_noc.c b/drivers/bus/omap_l3_noc.c index 11f7982..ebee57d 100644 --- a/drivers/bus/omap_l3_noc.c +++ b/drivers/bus/omap_l3_noc.c @@ -1,7 +1,7 @@ /* * OMAP L3 Interconnect error handling driver * - * Copyright (C) 2011-2014 Texas Instruments Incorporated - http://www.ti.com/ + * Copyright (C) 2011-2015 Texas Instruments Incorporated - http://www.ti.com/ * Santosh Shilimkar * Sricharan * @@ -233,7 +233,8 @@ static irqreturn_t l3_interrupt_handler(int irq, void *_l3) } static const struct of_device_id l3_noc_match[] = { - {.compatible = "ti,omap4-l3-noc", .data = &omap_l3_data}, + {.compatible = "ti,omap4-l3-noc", .data = &omap4_l3_data}, + {.compatible = "ti,omap5-l3-noc", .data = &omap5_l3_data}, {.compatible = "ti,dra7-l3-noc", .data = &dra_l3_data}, {.compatible = "ti,am4372-l3-noc", .data = &am4372_l3_data}, {}, diff --git a/drivers/bus/omap_l3_noc.h b/drivers/bus/omap_l3_noc.h index a314d80..73431f8 100644 --- a/drivers/bus/omap_l3_noc.h +++ b/drivers/bus/omap_l3_noc.h @@ -1,7 +1,7 @@ /* * OMAP L3 Interconnect error handling driver header * - * Copyright (C) 2011-2014 Texas Instruments Incorporated - http://www.ti.com/ + * Copyright (C) 2011-2015 Texas Instruments Incorporated - http://www.ti.com/ * Santosh Shilimkar * sricharan * @@ -175,16 +175,14 @@ static struct l3_flagmux_data omap_l3_flagmux_clk2 = { }; -static struct l3_target_data omap_l3_target_data_clk3[] = { - {0x0100, "EMUSS",}, - {0x0300, "DEBUG SOURCE",}, - {0x0, "HOST CLK3",}, +static struct l3_target_data omap4_l3_target_data_clk3[] = { + {0x0100, "DEBUGSS",}, }; -static struct l3_flagmux_data omap_l3_flagmux_clk3 = { +static struct l3_flagmux_data omap4_l3_flagmux_clk3 = { .offset = 0x0200, - .l3_targ = omap_l3_target_data_clk3, - .num_targ_data = ARRAY_SIZE(omap_l3_target_data_clk3), + .l3_targ = omap4_l3_target_data_clk3, + .num_targ_data = ARRAY_SIZE(omap4_l3_target_data_clk3), }; static struct l3_masters_data omap_l3_masters[] = { @@ -215,21 +213,49 @@ static struct l3_masters_data omap_l3_masters[] = { { 0x32, "USBHOSTFS"} }; -static struct l3_flagmux_data *omap_l3_flagmux[] = { +static struct l3_flagmux_data *omap4_l3_flagmux[] = { &omap_l3_flagmux_clk1, &omap_l3_flagmux_clk2, - &omap_l3_flagmux_clk3, + &omap4_l3_flagmux_clk3, }; -static const struct omap_l3 omap_l3_data = { - .l3_flagmux = omap_l3_flagmux, - .num_modules = ARRAY_SIZE(omap_l3_flagmux), +static const struct omap_l3 omap4_l3_data = { + .l3_flagmux = omap4_l3_flagmux, + .num_modules = ARRAY_SIZE(omap4_l3_flagmux), .l3_masters = omap_l3_masters, .num_masters = ARRAY_SIZE(omap_l3_masters), /* The 6 MSBs of register field used to distinguish initiator */ .mst_addr_mask = 0xFC, }; +/* OMAP5 data */ +static struct l3_target_data omap5_l3_target_data_clk3[] = { + {0x0100, "L3INSTR",}, + {0x0300, "DEBUGSS",}, + {0x0, "HOSTCLK3",}, +}; + +static struct l3_flagmux_data omap5_l3_flagmux_clk3 = { + .offset = 0x0200, + .l3_targ = omap5_l3_target_data_clk3, + .num_targ_data = ARRAY_SIZE(omap5_l3_target_data_clk3), +}; + +static struct l3_flagmux_data *omap5_l3_flagmux[] = { + &omap_l3_flagmux_clk1, + &omap_l3_flagmux_clk2, + &omap5_l3_flagmux_clk3, +}; + +static const struct omap_l3 omap5_l3_data = { + .l3_flagmux = omap5_l3_flagmux, + .num_modules = ARRAY_SIZE(omap5_l3_flagmux), + .l3_masters = omap_l3_masters, + .num_masters = ARRAY_SIZE(omap_l3_masters), + /* The 6 MSBs of register field used to distinguish initiator */ + .mst_addr_mask = 0x7E0, +}; + /* DRA7 data */ static struct l3_target_data dra_l3_target_data_clk1[] = { {0x2a00, "AES1",}, -- cgit v0.10.2 From 013ead48a843442e63b9426e3bd5df18ca5d054a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Fri, 1 May 2015 12:34:12 +0200 Subject: drm/radeon: disable semaphores for UVD V1 (v2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Hardware doesn't seem to work correctly, just block userspace in this case. v2: add missing defines Bugs: https://bugs.freedesktop.org/show_bug.cgi?id=85320 Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index fafd8ce..8dbf508 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -1202,7 +1202,7 @@ static struct radeon_asic rs780_asic = { static struct radeon_asic_ring rv770_uvd_ring = { .ib_execute = &uvd_v1_0_ib_execute, .emit_fence = &uvd_v2_2_fence_emit, - .emit_semaphore = &uvd_v1_0_semaphore_emit, + .emit_semaphore = &uvd_v2_2_semaphore_emit, .cs_parse = &radeon_uvd_cs_parse, .ring_test = &uvd_v1_0_ring_test, .ib_test = &uvd_v1_0_ib_test, diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index cf0a90b..a3ca8cd 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -949,6 +949,10 @@ void uvd_v1_0_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib); int uvd_v2_2_resume(struct radeon_device *rdev); void uvd_v2_2_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence); +bool uvd_v2_2_semaphore_emit(struct radeon_device *rdev, + struct radeon_ring *ring, + struct radeon_semaphore *semaphore, + bool emit_wait); /* uvd v3.1 */ bool uvd_v3_1_semaphore_emit(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/rv770d.h b/drivers/gpu/drm/radeon/rv770d.h index 3cf1e29..9ef2064 100644 --- a/drivers/gpu/drm/radeon/rv770d.h +++ b/drivers/gpu/drm/radeon/rv770d.h @@ -989,6 +989,9 @@ ((n) & 0x3FFF) << 16) /* UVD */ +#define UVD_SEMA_ADDR_LOW 0xef00 +#define UVD_SEMA_ADDR_HIGH 0xef04 +#define UVD_SEMA_CMD 0xef08 #define UVD_GPCOM_VCPU_CMD 0xef0c #define UVD_GPCOM_VCPU_DATA0 0xef10 #define UVD_GPCOM_VCPU_DATA1 0xef14 diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c index e72b3cb..c6b1cbc 100644 --- a/drivers/gpu/drm/radeon/uvd_v1_0.c +++ b/drivers/gpu/drm/radeon/uvd_v1_0.c @@ -466,18 +466,8 @@ bool uvd_v1_0_semaphore_emit(struct radeon_device *rdev, struct radeon_semaphore *semaphore, bool emit_wait) { - uint64_t addr = semaphore->gpu_addr; - - radeon_ring_write(ring, PACKET0(UVD_SEMA_ADDR_LOW, 0)); - radeon_ring_write(ring, (addr >> 3) & 0x000FFFFF); - - radeon_ring_write(ring, PACKET0(UVD_SEMA_ADDR_HIGH, 0)); - radeon_ring_write(ring, (addr >> 23) & 0x000FFFFF); - - radeon_ring_write(ring, PACKET0(UVD_SEMA_CMD, 0)); - radeon_ring_write(ring, emit_wait ? 1 : 0); - - return true; + /* disable semaphores for UVD V1 hardware */ + return false; } /** diff --git a/drivers/gpu/drm/radeon/uvd_v2_2.c b/drivers/gpu/drm/radeon/uvd_v2_2.c index 8919351..7ed778c 100644 --- a/drivers/gpu/drm/radeon/uvd_v2_2.c +++ b/drivers/gpu/drm/radeon/uvd_v2_2.c @@ -60,6 +60,35 @@ void uvd_v2_2_fence_emit(struct radeon_device *rdev, } /** + * uvd_v2_2_semaphore_emit - emit semaphore command + * + * @rdev: radeon_device pointer + * @ring: radeon_ring pointer + * @semaphore: semaphore to emit commands for + * @emit_wait: true if we should emit a wait command + * + * Emit a semaphore command (either wait or signal) to the UVD ring. + */ +bool uvd_v2_2_semaphore_emit(struct radeon_device *rdev, + struct radeon_ring *ring, + struct radeon_semaphore *semaphore, + bool emit_wait) +{ + uint64_t addr = semaphore->gpu_addr; + + radeon_ring_write(ring, PACKET0(UVD_SEMA_ADDR_LOW, 0)); + radeon_ring_write(ring, (addr >> 3) & 0x000FFFFF); + + radeon_ring_write(ring, PACKET0(UVD_SEMA_ADDR_HIGH, 0)); + radeon_ring_write(ring, (addr >> 23) & 0x000FFFFF); + + radeon_ring_write(ring, PACKET0(UVD_SEMA_CMD, 0)); + radeon_ring_write(ring, emit_wait ? 1 : 0); + + return true; +} + +/** * uvd_v2_2_resume - memory controller programming * * @rdev: radeon_device pointer -- cgit v0.10.2 From b2387ddcced8de3e6471a2fb16a409577063016f Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Fri, 1 May 2015 09:59:44 -0700 Subject: blk-mq: fix FUA request hang When a FUA request enters its DATA stage of flush pipeline, the request is added to mq requeue list, the request will then be added to ctx->rq_list. blk_mq_attempt_merge() might merge the request with a bio. Later when the request is finished the flush pipeline, the request->__data_len is 0. Then I only saw the bio gets endio called, the original request never finish. Adding REQ_FLUSH_SEQ into REQ_NOMERGE_FLAGS looks an easy fix. stable: 3.15+ Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index a1b25e3..b7299fe 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -220,7 +220,7 @@ enum rq_flag_bits { /* This mask is used for both bio and request merge checking */ #define REQ_NOMERGE_FLAGS \ - (REQ_NOMERGE | REQ_STARTED | REQ_SOFTBARRIER | REQ_FLUSH | REQ_FUA) + (REQ_NOMERGE | REQ_STARTED | REQ_SOFTBARRIER | REQ_FLUSH | REQ_FUA | REQ_FLUSH_SEQ) #define REQ_RAHEAD (1ULL << __REQ_RAHEAD) #define REQ_THROTTLED (1ULL << __REQ_THROTTLED) -- cgit v0.10.2 From 9ba52e5812e53f20f23600d79449a3ec05a0254f Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Mon, 4 May 2015 14:32:48 -0600 Subject: blk-mq: don't lose requests if a stopped queue restarts Normally if driver is busy to dispatch a request the logic is like below: block layer: driver: __blk_mq_run_hw_queue a. blk_mq_stop_hw_queue b. rq add to ctx->dispatch later: 1. blk_mq_start_hw_queue 2. __blk_mq_run_hw_queue But it's possible step 1-2 runs between a and b. And since rq isn't in ctx->dispatch yet, step 2 will not run rq. The rq might get lost if there are no subsequent requests kick in. Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe diff --git a/block/blk-mq.c b/block/blk-mq.c index 76f460e..e68b71b 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -858,6 +858,16 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) spin_lock(&hctx->lock); list_splice(&rq_list, &hctx->dispatch); spin_unlock(&hctx->lock); + /* + * the queue is expected stopped with BLK_MQ_RQ_QUEUE_BUSY, but + * it's possible the queue is stopped and restarted again + * before this. Queue restart will dispatch requests. And since + * requests in rq_list aren't added into hctx->dispatch yet, + * the requests in rq_list might get lost. + * + * blk_mq_run_hw_queue() already checks the STOPPED bit + **/ + blk_mq_run_hw_queue(hctx, true); } } -- cgit v0.10.2 From 5463e7c18e51152104aba9614e6abfc039a8b710 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Tue, 21 Apr 2015 10:40:54 -0700 Subject: Revert "f2fs: enhance multi-threads performance" This reports performance regression by Yuanhan Liu. The basic idea was to reduce one-point mutex, but it turns out this causes another contention like context swithes. https://lkml.org/lkml/2015/4/21/11 Until finishing the analysis on this issue, I'd like to revert this for a while. This reverts commit 78373b7319abdf15050af5b1632c4c8b8b398f33. diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index b91b0e1..1e1aae6 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1513,6 +1513,7 @@ static int f2fs_write_data_pages(struct address_space *mapping, { struct inode *inode = mapping->host; struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + bool locked = false; int ret; long diff; @@ -1533,7 +1534,13 @@ static int f2fs_write_data_pages(struct address_space *mapping, diff = nr_pages_to_write(sbi, DATA, wbc); + if (!S_ISDIR(inode->i_mode)) { + mutex_lock(&sbi->writepages); + locked = true; + } ret = write_cache_pages(mapping, wbc, __f2fs_writepage, mapping); + if (locked) + mutex_unlock(&sbi->writepages); f2fs_submit_merged_bio(sbi, DATA, WRITE); diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index d8921cf..8de34ab 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -625,6 +625,7 @@ struct f2fs_sb_info { struct mutex cp_mutex; /* checkpoint procedure lock */ struct rw_semaphore cp_rwsem; /* blocking FS operations */ struct rw_semaphore node_write; /* locking node writes */ + struct mutex writepages; /* mutex for writepages() */ wait_queue_head_t cp_wait; struct inode_management im[MAX_INO_ENTRY]; /* manage inode cache */ diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 160b883..b2dd1b0 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1035,6 +1035,7 @@ try_onemore: sbi->raw_super = raw_super; sbi->raw_super_buf = raw_super_buf; mutex_init(&sbi->gc_mutex); + mutex_init(&sbi->writepages); mutex_init(&sbi->cp_mutex); init_rwsem(&sbi->node_write); clear_sbi_flag(sbi, SBI_POR_DOING); -- cgit v0.10.2 From 7263b1bd0490fca68ee7eedb0b6973cb86d4701c Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Wed, 22 Apr 2015 11:03:48 -0700 Subject: f2fs: fix wrong error hanlder in f2fs_follow_link The page_follow_link_light returns NULL and its error pointer was remained in nd->path. Reported-by: Dan Carpenter Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 7e3794e..658e807 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -298,16 +298,14 @@ fail: static void *f2fs_follow_link(struct dentry *dentry, struct nameidata *nd) { - struct page *page; + struct page *page = page_follow_link_light(dentry, nd); - page = page_follow_link_light(dentry, nd); - if (IS_ERR(page)) + if (IS_ERR_OR_NULL(page)) return page; /* this is broken symlink case */ if (*nd_get_link(nd) == 0) { - kunmap(page); - page_cache_release(page); + page_put_link(dentry, nd, page); return ERR_PTR(-ENOENT); } return page; -- cgit v0.10.2 From e8a4a2696fecb398b0288c43c0e0dbb91e265bb2 Mon Sep 17 00:00:00 2001 From: Tahsin Erdogan Date: Mon, 4 May 2015 21:15:31 -0700 Subject: x86/spinlocks: Fix regression in spinlock contention detection A spinlock is regarded as contended when there is at least one waiter. Currently, the code that checks whether there are any waiters rely on tail value being greater than head. However, this is not true if tail reaches the max value and wraps back to zero, so arch_spin_is_contended() incorrectly returns 0 (not contended) when tail is smaller than head. The original code (before regression) handled this case by casting the (tail - head) to an unsigned value. This change simply restores that behavior. Fixes: d6abfdb20223 ("x86/spinlocks/paravirt: Fix memory corruption on unlock") Signed-off-by: Tahsin Erdogan Cc: peterz@infradead.org Cc: Waiman.Long@hp.com Cc: borntraeger@de.ibm.com Cc: oleg@redhat.com Cc: raghavendra.kt@linux.vnet.ibm.com Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1430799331-20445-1-git-send-email-tahsin@google.com Signed-off-by: Thomas Gleixner diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index cf87de3..64b6117 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -169,7 +169,7 @@ static inline int arch_spin_is_contended(arch_spinlock_t *lock) struct __raw_tickets tmp = READ_ONCE(lock->tickets); tmp.head &= ~TICKET_SLOWPATH_FLAG; - return (tmp.tail - tmp.head) > TICKET_LOCK_INC; + return (__ticket_t)(tmp.tail - tmp.head) > TICKET_LOCK_INC; } #define arch_spin_is_contended arch_spin_is_contended -- cgit v0.10.2 From 2c674fac5b1603c6a2aacc88116e3fbc75ebd799 Mon Sep 17 00:00:00 2001 From: Kailang Yang Date: Tue, 5 May 2015 15:02:42 +0800 Subject: ALSA: hda/realtek - Add ALC298 alias name for Dell Add ALC3266 for Dell platform. Signed-off-by: Kailang Yang Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 516a0da..4a0a238 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -883,6 +883,7 @@ static struct alc_codec_rename_pci_table rename_pci_tbl[] = { { 0x10ec0668, 0x1028, 0, "ALC3661" }, { 0x10ec0275, 0x1028, 0, "ALC3260" }, { 0x10ec0899, 0x1028, 0, "ALC3861" }, + { 0x10ec0298, 0x1028, 0, "ALC3266" }, { 0x10ec0670, 0x1025, 0, "ALC669X" }, { 0x10ec0676, 0x1025, 0, "ALC679X" }, { 0x10ec0282, 0x1043, 0, "ALC3229" }, -- cgit v0.10.2 From 285214409a9e5fceba2215461b4682b6069d8e77 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Mon, 20 Apr 2015 14:01:11 -0600 Subject: RDMA/CMA: Canonize IPv4 on IPV6 sockets properly When accepting a new IPv4 connect to an IPv6 socket, the CMA tries to canonize the address family to IPv4, but does not properly process the listening sockaddr to get the listening port, and does not properly set the address family of the canonized sockaddr. Fixes: e51060f08a61 ("IB: IP address based RDMA connection manager") Cc: Reported-By: Yotam Kenneth Signed-off-by: Jason Gunthorpe Tested-by: Haggai Eran Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index d570030..06441a4 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -859,19 +859,27 @@ static void cma_save_ib_info(struct rdma_cm_id *id, struct rdma_cm_id *listen_id memcpy(&ib->sib_addr, &path->dgid, 16); } +static __be16 ss_get_port(const struct sockaddr_storage *ss) +{ + if (ss->ss_family == AF_INET) + return ((struct sockaddr_in *)ss)->sin_port; + else if (ss->ss_family == AF_INET6) + return ((struct sockaddr_in6 *)ss)->sin6_port; + BUG(); +} + static void cma_save_ip4_info(struct rdma_cm_id *id, struct rdma_cm_id *listen_id, struct cma_hdr *hdr) { - struct sockaddr_in *listen4, *ip4; + struct sockaddr_in *ip4; - listen4 = (struct sockaddr_in *) &listen_id->route.addr.src_addr; ip4 = (struct sockaddr_in *) &id->route.addr.src_addr; - ip4->sin_family = listen4->sin_family; + ip4->sin_family = AF_INET; ip4->sin_addr.s_addr = hdr->dst_addr.ip4.addr; - ip4->sin_port = listen4->sin_port; + ip4->sin_port = ss_get_port(&listen_id->route.addr.src_addr); ip4 = (struct sockaddr_in *) &id->route.addr.dst_addr; - ip4->sin_family = listen4->sin_family; + ip4->sin_family = AF_INET; ip4->sin_addr.s_addr = hdr->src_addr.ip4.addr; ip4->sin_port = hdr->port; } @@ -879,16 +887,15 @@ static void cma_save_ip4_info(struct rdma_cm_id *id, struct rdma_cm_id *listen_i static void cma_save_ip6_info(struct rdma_cm_id *id, struct rdma_cm_id *listen_id, struct cma_hdr *hdr) { - struct sockaddr_in6 *listen6, *ip6; + struct sockaddr_in6 *ip6; - listen6 = (struct sockaddr_in6 *) &listen_id->route.addr.src_addr; ip6 = (struct sockaddr_in6 *) &id->route.addr.src_addr; - ip6->sin6_family = listen6->sin6_family; + ip6->sin6_family = AF_INET6; ip6->sin6_addr = hdr->dst_addr.ip6; - ip6->sin6_port = listen6->sin6_port; + ip6->sin6_port = ss_get_port(&listen_id->route.addr.src_addr); ip6 = (struct sockaddr_in6 *) &id->route.addr.dst_addr; - ip6->sin6_family = listen6->sin6_family; + ip6->sin6_family = AF_INET6; ip6->sin6_addr = hdr->src_addr.ip6; ip6->sin6_port = hdr->port; } -- cgit v0.10.2 From 0b7410471d59ce2ea30453e68c03bdb941d5951e Mon Sep 17 00:00:00 2001 From: Hariprasad S Date: Wed, 22 Apr 2015 01:44:58 +0530 Subject: iw_cxgb4: Cleanup register defines/MACROS Cleanup macros and register defines for consistency Signed-off-by: Hariprasad Shenai Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 57176dd..0493cca 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -675,7 +675,7 @@ static int send_connect(struct c4iw_ep *ep) if (is_t5(ep->com.dev->rdev.lldi.adapter_type)) { opt2 |= T5_OPT_2_VALID_F; opt2 |= CONG_CNTRL_V(CONG_ALG_TAHOE); - opt2 |= CONG_CNTRL_VALID; /* OPT_2_ISS for T5 */ + opt2 |= T5_ISS_F; } t4_set_arp_err_handler(skb, ep, act_open_req_arp_failure); @@ -2214,7 +2214,7 @@ static void accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, u32 isn = (prandom_u32() & ~7UL) - 1; opt2 |= T5_OPT_2_VALID_F; opt2 |= CONG_CNTRL_V(CONG_ALG_TAHOE); - opt2 |= CONG_CNTRL_VALID; /* OPT_2_ISS for T5 */ + opt2 |= T5_ISS_F; rpl5 = (void *)rpl; memset(&rpl5->iss, 0, roundup(sizeof(*rpl5)-sizeof(*rpl), 16)); if (peer2peer) diff --git a/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h b/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h index 5e53327..343e8daf 100644 --- a/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h +++ b/drivers/infiniband/hw/cxgb4/t4fw_ri_api.h @@ -848,6 +848,8 @@ enum { /* TCP congestion control algorithms */ #define CONG_CNTRL_V(x) ((x) << CONG_CNTRL_S) #define CONG_CNTRL_G(x) (((x) >> CONG_CNTRL_S) & CONG_CNTRL_M) -#define CONG_CNTRL_VALID (1 << 18) +#define T5_ISS_S 18 +#define T5_ISS_V(x) ((x) << T5_ISS_S) +#define T5_ISS_F T5_ISS_V(1U) #endif /* _T4FW_RI_API_H_ */ -- cgit v0.10.2 From 6198dd8d7a6a7f40dc4599cb0676101d9cb82776 Mon Sep 17 00:00:00 2001 From: Hariprasad S Date: Wed, 22 Apr 2015 01:44:59 +0530 Subject: iw_cxgb4: 32b platform fixes - get_dma_mr() was using ~0UL which is should be ~0ULL. This causes the DMA MR to get setup incorrectly in hardware. - wr_log_show() needed a 64b divide function div64_u64() instead of doing division directly. - fixed warnings about recasting a pointer to a u64 Signed-off-by: Steve Wise Signed-off-by: Hariprasad Shenai Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 0493cca..6fb31ba 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -3571,7 +3571,7 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb, * TP will ignore any value > 0 for MSS index. */ req->tcb.opt0 = cpu_to_be64(MSS_IDX_V(0xF)); - req->cookie = (unsigned long)skb; + req->cookie = (uintptr_t)skb; set_wr_txq(req_skb, CPL_PRIORITY_CONTROL, port_id); ret = cxgb4_ofld_send(dev->rdev.lldi.ports[0], req_skb); diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index ab7692a..25dbd69 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -55,7 +55,7 @@ static int destroy_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, FW_RI_RES_WR_NRES_V(1) | FW_WR_COMPL_F); res_wr->len16_pkd = cpu_to_be32(DIV_ROUND_UP(wr_len, 16)); - res_wr->cookie = (unsigned long) &wr_wait; + res_wr->cookie = (uintptr_t)&wr_wait; res = res_wr->res; res->u.cq.restype = FW_RI_RES_TYPE_CQ; res->u.cq.op = FW_RI_RES_OP_RESET; @@ -125,7 +125,7 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, FW_RI_RES_WR_NRES_V(1) | FW_WR_COMPL_F); res_wr->len16_pkd = cpu_to_be32(DIV_ROUND_UP(wr_len, 16)); - res_wr->cookie = (unsigned long) &wr_wait; + res_wr->cookie = (uintptr_t)&wr_wait; res = res_wr->res; res->u.cq.restype = FW_RI_RES_TYPE_CQ; res->u.cq.op = FW_RI_RES_OP_WRITE; @@ -970,8 +970,7 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, int entries, } PDBG("%s cqid 0x%0x chp %p size %u memsize %zu, dma_addr 0x%0llx\n", __func__, chp->cq.cqid, chp, chp->cq.size, - chp->cq.memsize, - (unsigned long long) chp->cq.dma_addr); + chp->cq.memsize, (unsigned long long) chp->cq.dma_addr); return &chp->ibcq; err5: kfree(mm2); diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 8fb295e..7ed3253 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -151,7 +151,7 @@ static int wr_log_show(struct seq_file *seq, void *v) int prev_ts_set = 0; int idx, end; -#define ts2ns(ts) div64_ul((ts) * dev->rdev.lldi.cclk_ps, 1000) +#define ts2ns(ts) div64_u64((ts) * dev->rdev.lldi.cclk_ps, 1000) idx = atomic_read(&dev->rdev.wr_log_idx) & (dev->rdev.wr_log_size - 1); @@ -784,10 +784,10 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) rdev->lldi.vr->qp.size, rdev->lldi.vr->cq.start, rdev->lldi.vr->cq.size); - PDBG("udb len 0x%x udb base %llx db_reg %p gts_reg %p qpshift %lu " + PDBG("udb len 0x%x udb base %p db_reg %p gts_reg %p qpshift %lu " "qpmask 0x%x cqshift %lu cqmask 0x%x\n", (unsigned)pci_resource_len(rdev->lldi.pdev, 2), - (u64)pci_resource_start(rdev->lldi.pdev, 2), + (void *)pci_resource_start(rdev->lldi.pdev, 2), rdev->lldi.db_reg, rdev->lldi.gts_reg, rdev->qpshift, rdev->qpmask, diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c index 3ef0cf9..cff815b 100644 --- a/drivers/infiniband/hw/cxgb4/mem.c +++ b/drivers/infiniband/hw/cxgb4/mem.c @@ -144,7 +144,7 @@ static int _c4iw_write_mem_inline(struct c4iw_rdev *rdev, u32 addr, u32 len, if (i == (num_wqe-1)) { req->wr.wr_hi = cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR) | FW_WR_COMPL_F); - req->wr.wr_lo = (__force __be64)(unsigned long) &wr_wait; + req->wr.wr_lo = (__force __be64)&wr_wait; } else req->wr.wr_hi = cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR)); req->wr.wr_mid = cpu_to_be32( @@ -676,12 +676,12 @@ struct ib_mr *c4iw_get_dma_mr(struct ib_pd *pd, int acc) mhp->attr.zbva = 0; mhp->attr.va_fbo = 0; mhp->attr.page_size = 0; - mhp->attr.len = ~0UL; + mhp->attr.len = ~0ULL; mhp->attr.pbl_size = 0; ret = write_tpt_entry(&rhp->rdev, 0, &stag, 1, php->pdid, FW_RI_STAG_NSMR, mhp->attr.perms, - mhp->attr.mw_bind_enable, 0, 0, ~0UL, 0, 0, 0); + mhp->attr.mw_bind_enable, 0, 0, ~0ULL, 0, 0, 0); if (ret) goto err1; diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index 15cae5a..389ced3 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -275,7 +275,7 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, FW_RI_RES_WR_NRES_V(2) | FW_WR_COMPL_F); res_wr->len16_pkd = cpu_to_be32(DIV_ROUND_UP(wr_len, 16)); - res_wr->cookie = (unsigned long) &wr_wait; + res_wr->cookie = (uintptr_t)&wr_wait; res = res_wr->res; res->u.sqrq.restype = FW_RI_RES_TYPE_SQ; res->u.sqrq.op = FW_RI_RES_OP_WRITE; @@ -1209,7 +1209,7 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp, wqe->flowid_len16 = cpu_to_be32( FW_WR_FLOWID_V(ep->hwtid) | FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*wqe), 16))); - wqe->cookie = (unsigned long) &ep->com.wr_wait; + wqe->cookie = (uintptr_t)&ep->com.wr_wait; wqe->u.fini.type = FW_RI_TYPE_FINI; ret = c4iw_ofld_send(&rhp->rdev, skb); @@ -1279,7 +1279,7 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp) FW_WR_FLOWID_V(qhp->ep->hwtid) | FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*wqe), 16))); - wqe->cookie = (unsigned long) &qhp->ep->com.wr_wait; + wqe->cookie = (uintptr_t)&qhp->ep->com.wr_wait; wqe->u.init.type = FW_RI_TYPE_INIT; wqe->u.init.mpareqbit_p2ptype = @@ -1766,11 +1766,11 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, mm2->len = PAGE_ALIGN(qhp->wq.rq.memsize); insert_mmap(ucontext, mm2); mm3->key = uresp.sq_db_gts_key; - mm3->addr = (__force unsigned long) qhp->wq.sq.udb; + mm3->addr = (__force unsigned long)qhp->wq.sq.udb; mm3->len = PAGE_SIZE; insert_mmap(ucontext, mm3); mm4->key = uresp.rq_db_gts_key; - mm4->addr = (__force unsigned long) qhp->wq.rq.udb; + mm4->addr = (__force unsigned long)qhp->wq.rq.udb; mm4->len = PAGE_SIZE; insert_mmap(ucontext, mm4); if (mm5) { -- cgit v0.10.2 From 09ece8b9e983fe858de6eab7a386d58d194227b6 Mon Sep 17 00:00:00 2001 From: Hariprasad S Date: Wed, 22 Apr 2015 01:45:00 +0530 Subject: iw_cxgb4: use BAR2 GTS register for T5 kernel mode CQs For T5, we must not use the kdb/kgts registers, in order avoid db drops under extreme loads. Signed-off-by: Steve Wise Signed-off-by: Hariprasad Shenai Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 25dbd69..68ddb37 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -156,12 +156,19 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, goto err4; cq->gen = 1; - cq->gts = rdev->lldi.gts_reg; cq->rdev = rdev; if (user) { - cq->ugts = (u64)pci_resource_start(rdev->lldi.pdev, 2) + - (cq->cqid << rdev->cqshift); - cq->ugts &= PAGE_MASK; + u32 off = (cq->cqid << rdev->cqshift) & PAGE_MASK; + + cq->ugts = (u64)rdev->bar2_pa + off; + } else if (is_t4(rdev->lldi.adapter_type)) { + cq->gts = rdev->lldi.gts_reg; + cq->qid_mask = -1U; + } else { + u32 off = ((cq->cqid << rdev->cqshift) & PAGE_MASK) + 12; + + cq->gts = rdev->bar2_kva + off; + cq->qid_mask = rdev->qpmask; } return 0; err4: diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h index 871cdca..7f2a6c2 100644 --- a/drivers/infiniband/hw/cxgb4/t4.h +++ b/drivers/infiniband/hw/cxgb4/t4.h @@ -539,6 +539,7 @@ struct t4_cq { size_t memsize; __be64 bits_type_ts; u32 cqid; + u32 qid_mask; int vector; u16 size; /* including status page */ u16 cidx; @@ -563,12 +564,12 @@ static inline int t4_arm_cq(struct t4_cq *cq, int se) set_bit(CQ_ARMED, &cq->flags); while (cq->cidx_inc > CIDXINC_M) { val = SEINTARM_V(0) | CIDXINC_V(CIDXINC_M) | TIMERREG_V(7) | - INGRESSQID_V(cq->cqid); + INGRESSQID_V(cq->cqid & cq->qid_mask); writel(val, cq->gts); cq->cidx_inc -= CIDXINC_M; } val = SEINTARM_V(se) | CIDXINC_V(cq->cidx_inc) | TIMERREG_V(6) | - INGRESSQID_V(cq->cqid); + INGRESSQID_V(cq->cqid & cq->qid_mask); writel(val, cq->gts); cq->cidx_inc = 0; return 0; @@ -601,7 +602,7 @@ static inline void t4_hwcq_consume(struct t4_cq *cq) u32 val; val = SEINTARM_V(0) | CIDXINC_V(cq->cidx_inc) | TIMERREG_V(7) | - INGRESSQID_V(cq->cqid); + INGRESSQID_V(cq->cqid & cq->qid_mask); writel(val, cq->gts); cq->cidx_inc = 0; } -- cgit v0.10.2 From 4a75a86c8d04390f268d7237cc49fe9a8e36efe7 Mon Sep 17 00:00:00 2001 From: Hariprasad S Date: Wed, 22 Apr 2015 01:45:01 +0530 Subject: iw_cxgb4: enforce qp/cq id requirements Currently the iw_cxgb4 implementation requires the qp and cq qid densities to match as well as the qp and cq id ranges. So fail a device open if the device configuration doesn't meet the requirements. The reason for these restictions has to do with the fact that IQ qid X has a UGTS register in the same bar2 page as EQ qid X. Thus both qids need to be allocated to the same user process for security reasons. The logic that does this (the qpid allocator in iw_cxgb4/resource.c) handles this but requires the above restrictions. Signed-off-by: Steve Wise Signed-off-by: Hariprasad Shenai Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 7ed3253..83209bb 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -765,6 +765,29 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) c4iw_init_dev_ucontext(rdev, &rdev->uctx); /* + * This implementation assumes udb_density == ucq_density! Eventually + * we might need to support this but for now fail the open. Also the + * cqid and qpid range must match for now. + */ + if (rdev->lldi.udb_density != rdev->lldi.ucq_density) { + pr_err(MOD "%s: unsupported udb/ucq densities %u/%u\n", + pci_name(rdev->lldi.pdev), rdev->lldi.udb_density, + rdev->lldi.ucq_density); + err = -EINVAL; + goto err1; + } + if (rdev->lldi.vr->qp.start != rdev->lldi.vr->cq.start || + rdev->lldi.vr->qp.size != rdev->lldi.vr->cq.size) { + pr_err(MOD "%s: unsupported qp and cq id ranges " + "qp start %u size %u cq start %u size %u\n", + pci_name(rdev->lldi.pdev), rdev->lldi.vr->qp.start, + rdev->lldi.vr->qp.size, rdev->lldi.vr->cq.size, + rdev->lldi.vr->cq.size); + err = -EINVAL; + goto err1; + } + + /* * qpshift is the number of bits to shift the qpid left in order * to get the correct address of the doorbell for that qp. */ -- cgit v0.10.2 From 6eec177461751f0fe191cf9977cde692b9481d0a Mon Sep 17 00:00:00 2001 From: Tatyana Nikolova Date: Tue, 21 Apr 2015 16:28:10 -0400 Subject: RDMA/core: Enable the iWarp Port Mapper to provide the actual address of the connecting peer to its clients Add functionality to enable the port mapper on the passive side to provide to its clients the actual (non-mapped) ip/tcp address information of the connecting peer 1) Adding remote_info_cb() to process the address info of the connecting peer The address info is provided by the user space port mapper service when the connection is initiated by the peer 2) Adding a hash list to store the remote address info 3) Adding functionality to add/remove the remote address info After the info has been provided to the port mapper client, it is removed from the hash list Signed-off-by: Tatyana Nikolova Reviewed-by: Steve Wise Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/core/iwpm_msg.c b/drivers/infiniband/core/iwpm_msg.c index b85ddbc..ab08170 100644 --- a/drivers/infiniband/core/iwpm_msg.c +++ b/drivers/infiniband/core/iwpm_msg.c @@ -468,7 +468,8 @@ add_mapping_response_exit: } EXPORT_SYMBOL(iwpm_add_mapping_cb); -/* netlink attribute policy for the response to add and query mapping request */ +/* netlink attribute policy for the response to add and query mapping request + * and response with remote address info */ static const struct nla_policy resp_query_policy[IWPM_NLA_RQUERY_MAPPING_MAX] = { [IWPM_NLA_QUERY_MAPPING_SEQ] = { .type = NLA_U32 }, [IWPM_NLA_QUERY_LOCAL_ADDR] = { .len = sizeof(struct sockaddr_storage) }, @@ -559,6 +560,76 @@ query_mapping_response_exit: } EXPORT_SYMBOL(iwpm_add_and_query_mapping_cb); +/* + * iwpm_remote_info_cb - Process a port mapper message, containing + * the remote connecting peer address info + */ +int iwpm_remote_info_cb(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct nlattr *nltb[IWPM_NLA_RQUERY_MAPPING_MAX]; + struct sockaddr_storage *local_sockaddr, *remote_sockaddr; + struct sockaddr_storage *mapped_loc_sockaddr, *mapped_rem_sockaddr; + struct iwpm_remote_info *rem_info; + const char *msg_type; + u8 nl_client; + int ret = -EINVAL; + + msg_type = "Remote Mapping info"; + if (iwpm_parse_nlmsg(cb, IWPM_NLA_RQUERY_MAPPING_MAX, + resp_query_policy, nltb, msg_type)) + return ret; + + nl_client = RDMA_NL_GET_CLIENT(cb->nlh->nlmsg_type); + if (!iwpm_valid_client(nl_client)) { + pr_info("%s: Invalid port mapper client = %d\n", + __func__, nl_client); + return ret; + } + atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq); + + local_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_QUERY_LOCAL_ADDR]); + remote_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_QUERY_REMOTE_ADDR]); + mapped_loc_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR]); + mapped_rem_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_REM_ADDR]); + + if (mapped_loc_sockaddr->ss_family != local_sockaddr->ss_family || + mapped_rem_sockaddr->ss_family != remote_sockaddr->ss_family) { + pr_info("%s: Sockaddr family doesn't match the requested one\n", + __func__); + return ret; + } + rem_info = kzalloc(sizeof(struct iwpm_remote_info), GFP_ATOMIC); + if (!rem_info) { + pr_err("%s: Unable to allocate a remote info\n", __func__); + ret = -ENOMEM; + return ret; + } + memcpy(&rem_info->mapped_loc_sockaddr, mapped_loc_sockaddr, + sizeof(struct sockaddr_storage)); + memcpy(&rem_info->remote_sockaddr, remote_sockaddr, + sizeof(struct sockaddr_storage)); + memcpy(&rem_info->mapped_rem_sockaddr, mapped_rem_sockaddr, + sizeof(struct sockaddr_storage)); + rem_info->nl_client = nl_client; + + iwpm_add_remote_info(rem_info); + + iwpm_print_sockaddr(local_sockaddr, + "remote_info: Local sockaddr:"); + iwpm_print_sockaddr(mapped_loc_sockaddr, + "remote_info: Mapped local sockaddr:"); + iwpm_print_sockaddr(remote_sockaddr, + "remote_info: Remote sockaddr:"); + iwpm_print_sockaddr(mapped_rem_sockaddr, + "remote_info: Mapped remote sockaddr:"); + return ret; +} +EXPORT_SYMBOL(iwpm_remote_info_cb); + /* netlink attribute policy for the received request for mapping info */ static const struct nla_policy resp_mapinfo_policy[IWPM_NLA_MAPINFO_REQ_MAX] = { [IWPM_NLA_MAPINFO_ULIB_NAME] = { .type = NLA_STRING, diff --git a/drivers/infiniband/core/iwpm_util.c b/drivers/infiniband/core/iwpm_util.c index 69e9f84..a626795 100644 --- a/drivers/infiniband/core/iwpm_util.c +++ b/drivers/infiniband/core/iwpm_util.c @@ -33,8 +33,10 @@ #include "iwpm_util.h" -#define IWPM_HASH_BUCKET_SIZE 512 -#define IWPM_HASH_BUCKET_MASK (IWPM_HASH_BUCKET_SIZE - 1) +#define IWPM_MAPINFO_HASH_SIZE 512 +#define IWPM_MAPINFO_HASH_MASK (IWPM_MAPINFO_HASH_SIZE - 1) +#define IWPM_REMINFO_HASH_SIZE 64 +#define IWPM_REMINFO_HASH_MASK (IWPM_REMINFO_HASH_SIZE - 1) static LIST_HEAD(iwpm_nlmsg_req_list); static DEFINE_SPINLOCK(iwpm_nlmsg_req_lock); @@ -42,31 +44,49 @@ static DEFINE_SPINLOCK(iwpm_nlmsg_req_lock); static struct hlist_head *iwpm_hash_bucket; static DEFINE_SPINLOCK(iwpm_mapinfo_lock); +static struct hlist_head *iwpm_reminfo_bucket; +static DEFINE_SPINLOCK(iwpm_reminfo_lock); + static DEFINE_MUTEX(iwpm_admin_lock); static struct iwpm_admin_data iwpm_admin; int iwpm_init(u8 nl_client) { + int ret = 0; if (iwpm_valid_client(nl_client)) return -EINVAL; mutex_lock(&iwpm_admin_lock); if (atomic_read(&iwpm_admin.refcount) == 0) { - iwpm_hash_bucket = kzalloc(IWPM_HASH_BUCKET_SIZE * + iwpm_hash_bucket = kzalloc(IWPM_MAPINFO_HASH_SIZE * sizeof(struct hlist_head), GFP_KERNEL); if (!iwpm_hash_bucket) { - mutex_unlock(&iwpm_admin_lock); + ret = -ENOMEM; pr_err("%s Unable to create mapinfo hash table\n", __func__); - return -ENOMEM; + goto init_exit; + } + iwpm_reminfo_bucket = kzalloc(IWPM_REMINFO_HASH_SIZE * + sizeof(struct hlist_head), GFP_KERNEL); + if (!iwpm_reminfo_bucket) { + kfree(iwpm_hash_bucket); + ret = -ENOMEM; + pr_err("%s Unable to create reminfo hash table\n", __func__); + goto init_exit; } } atomic_inc(&iwpm_admin.refcount); +init_exit: mutex_unlock(&iwpm_admin_lock); - iwpm_set_valid(nl_client, 1); - return 0; + if (!ret) { + iwpm_set_valid(nl_client, 1); + pr_debug("%s: Mapinfo and reminfo tables are created\n", + __func__); + } + return ret; } EXPORT_SYMBOL(iwpm_init); static void free_hash_bucket(void); +static void free_reminfo_bucket(void); int iwpm_exit(u8 nl_client) { @@ -81,7 +101,8 @@ int iwpm_exit(u8 nl_client) } if (atomic_dec_and_test(&iwpm_admin.refcount)) { free_hash_bucket(); - pr_debug("%s: Mapinfo hash table is destroyed\n", __func__); + free_reminfo_bucket(); + pr_debug("%s: Resources are destroyed\n", __func__); } mutex_unlock(&iwpm_admin_lock); iwpm_set_valid(nl_client, 0); @@ -89,7 +110,7 @@ int iwpm_exit(u8 nl_client) } EXPORT_SYMBOL(iwpm_exit); -static struct hlist_head *get_hash_bucket_head(struct sockaddr_storage *, +static struct hlist_head *get_mapinfo_hash_bucket(struct sockaddr_storage *, struct sockaddr_storage *); int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, @@ -99,9 +120,10 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, struct hlist_head *hash_bucket_head; struct iwpm_mapping_info *map_info; unsigned long flags; + int ret = -EINVAL; if (!iwpm_valid_client(nl_client)) - return -EINVAL; + return ret; map_info = kzalloc(sizeof(struct iwpm_mapping_info), GFP_KERNEL); if (!map_info) { pr_err("%s: Unable to allocate a mapping info\n", __func__); @@ -115,13 +137,16 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, spin_lock_irqsave(&iwpm_mapinfo_lock, flags); if (iwpm_hash_bucket) { - hash_bucket_head = get_hash_bucket_head( + hash_bucket_head = get_mapinfo_hash_bucket( &map_info->local_sockaddr, &map_info->mapped_sockaddr); - hlist_add_head(&map_info->hlist_node, hash_bucket_head); + if (hash_bucket_head) { + hlist_add_head(&map_info->hlist_node, hash_bucket_head); + ret = 0; + } } spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); - return 0; + return ret; } EXPORT_SYMBOL(iwpm_create_mapinfo); @@ -136,9 +161,12 @@ int iwpm_remove_mapinfo(struct sockaddr_storage *local_sockaddr, spin_lock_irqsave(&iwpm_mapinfo_lock, flags); if (iwpm_hash_bucket) { - hash_bucket_head = get_hash_bucket_head( + hash_bucket_head = get_mapinfo_hash_bucket( local_sockaddr, mapped_local_addr); + if (!hash_bucket_head) + goto remove_mapinfo_exit; + hlist_for_each_entry_safe(map_info, tmp_hlist_node, hash_bucket_head, hlist_node) { @@ -152,6 +180,7 @@ int iwpm_remove_mapinfo(struct sockaddr_storage *local_sockaddr, } } } +remove_mapinfo_exit: spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); return ret; } @@ -166,7 +195,7 @@ static void free_hash_bucket(void) /* remove all the mapinfo data from the list */ spin_lock_irqsave(&iwpm_mapinfo_lock, flags); - for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) { + for (i = 0; i < IWPM_MAPINFO_HASH_SIZE; i++) { hlist_for_each_entry_safe(map_info, tmp_hlist_node, &iwpm_hash_bucket[i], hlist_node) { @@ -180,6 +209,96 @@ static void free_hash_bucket(void) spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); } +static void free_reminfo_bucket(void) +{ + struct hlist_node *tmp_hlist_node; + struct iwpm_remote_info *rem_info; + unsigned long flags; + int i; + + /* remove all the remote info from the list */ + spin_lock_irqsave(&iwpm_reminfo_lock, flags); + for (i = 0; i < IWPM_REMINFO_HASH_SIZE; i++) { + hlist_for_each_entry_safe(rem_info, tmp_hlist_node, + &iwpm_reminfo_bucket[i], hlist_node) { + + hlist_del_init(&rem_info->hlist_node); + kfree(rem_info); + } + } + /* free the hash list */ + kfree(iwpm_reminfo_bucket); + iwpm_reminfo_bucket = NULL; + spin_unlock_irqrestore(&iwpm_reminfo_lock, flags); +} + +static struct hlist_head *get_reminfo_hash_bucket(struct sockaddr_storage *, + struct sockaddr_storage *); + +void iwpm_add_remote_info(struct iwpm_remote_info *rem_info) +{ + struct hlist_head *hash_bucket_head; + unsigned long flags; + + spin_lock_irqsave(&iwpm_reminfo_lock, flags); + if (iwpm_reminfo_bucket) { + hash_bucket_head = get_reminfo_hash_bucket( + &rem_info->mapped_loc_sockaddr, + &rem_info->mapped_rem_sockaddr); + if (hash_bucket_head) + hlist_add_head(&rem_info->hlist_node, hash_bucket_head); + } + spin_unlock_irqrestore(&iwpm_reminfo_lock, flags); +} + +int iwpm_get_remote_info(struct sockaddr_storage *mapped_loc_addr, + struct sockaddr_storage *mapped_rem_addr, + struct sockaddr_storage *remote_addr, + u8 nl_client) +{ + struct hlist_node *tmp_hlist_node; + struct hlist_head *hash_bucket_head; + struct iwpm_remote_info *rem_info = NULL; + unsigned long flags; + int ret = -EINVAL; + + if (!iwpm_valid_client(nl_client)) { + pr_info("%s: Invalid client = %d\n", __func__, nl_client); + return ret; + } + spin_lock_irqsave(&iwpm_reminfo_lock, flags); + if (iwpm_reminfo_bucket) { + hash_bucket_head = get_reminfo_hash_bucket( + mapped_loc_addr, + mapped_rem_addr); + if (!hash_bucket_head) + goto get_remote_info_exit; + hlist_for_each_entry_safe(rem_info, tmp_hlist_node, + hash_bucket_head, hlist_node) { + + if (!iwpm_compare_sockaddr(&rem_info->mapped_loc_sockaddr, + mapped_loc_addr) && + !iwpm_compare_sockaddr(&rem_info->mapped_rem_sockaddr, + mapped_rem_addr)) { + + memcpy(remote_addr, &rem_info->remote_sockaddr, + sizeof(struct sockaddr_storage)); + iwpm_print_sockaddr(remote_addr, + "get_remote_info: Remote sockaddr:"); + + hlist_del_init(&rem_info->hlist_node); + kfree(rem_info); + ret = 0; + break; + } + } + } +get_remote_info_exit: + spin_unlock_irqrestore(&iwpm_reminfo_lock, flags); + return ret; +} +EXPORT_SYMBOL(iwpm_get_remote_info); + struct iwpm_nlmsg_request *iwpm_get_nlmsg_request(__u32 nlmsg_seq, u8 nl_client, gfp_t gfp) { @@ -409,31 +528,54 @@ static u32 iwpm_ipv4_jhash(struct sockaddr_in *ipv4_sockaddr) return hash; } -static struct hlist_head *get_hash_bucket_head(struct sockaddr_storage - *local_sockaddr, - struct sockaddr_storage - *mapped_sockaddr) +static int get_hash_bucket(struct sockaddr_storage *a_sockaddr, + struct sockaddr_storage *b_sockaddr, u32 *hash) { - u32 local_hash, mapped_hash, hash; + u32 a_hash, b_hash; - if (local_sockaddr->ss_family == AF_INET) { - local_hash = iwpm_ipv4_jhash((struct sockaddr_in *) local_sockaddr); - mapped_hash = iwpm_ipv4_jhash((struct sockaddr_in *) mapped_sockaddr); + if (a_sockaddr->ss_family == AF_INET) { + a_hash = iwpm_ipv4_jhash((struct sockaddr_in *) a_sockaddr); + b_hash = iwpm_ipv4_jhash((struct sockaddr_in *) b_sockaddr); - } else if (local_sockaddr->ss_family == AF_INET6) { - local_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) local_sockaddr); - mapped_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) mapped_sockaddr); + } else if (a_sockaddr->ss_family == AF_INET6) { + a_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) a_sockaddr); + b_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) b_sockaddr); } else { pr_err("%s: Invalid sockaddr family\n", __func__); - return NULL; + return -EINVAL; } - if (local_hash == mapped_hash) /* if port mapper isn't available */ - hash = local_hash; + if (a_hash == b_hash) /* if port mapper isn't available */ + *hash = a_hash; else - hash = jhash_2words(local_hash, mapped_hash, 0); + *hash = jhash_2words(a_hash, b_hash, 0); + return 0; +} + +static struct hlist_head *get_mapinfo_hash_bucket(struct sockaddr_storage + *local_sockaddr, struct sockaddr_storage + *mapped_sockaddr) +{ + u32 hash; + int ret; - return &iwpm_hash_bucket[hash & IWPM_HASH_BUCKET_MASK]; + ret = get_hash_bucket(local_sockaddr, mapped_sockaddr, &hash); + if (ret) + return NULL; + return &iwpm_hash_bucket[hash & IWPM_MAPINFO_HASH_MASK]; +} + +static struct hlist_head *get_reminfo_hash_bucket(struct sockaddr_storage + *mapped_loc_sockaddr, struct sockaddr_storage + *mapped_rem_sockaddr) +{ + u32 hash; + int ret; + + ret = get_hash_bucket(mapped_loc_sockaddr, mapped_rem_sockaddr, &hash); + if (ret) + return NULL; + return &iwpm_reminfo_bucket[hash & IWPM_REMINFO_HASH_MASK]; } static int send_mapinfo_num(u32 mapping_num, u8 nl_client, int iwpm_pid) @@ -512,7 +654,7 @@ int iwpm_send_mapinfo(u8 nl_client, int iwpm_pid) } skb_num++; spin_lock_irqsave(&iwpm_mapinfo_lock, flags); - for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) { + for (i = 0; i < IWPM_MAPINFO_HASH_SIZE; i++) { hlist_for_each_entry(map_info, &iwpm_hash_bucket[i], hlist_node) { if (map_info->nl_client != nl_client) @@ -595,7 +737,7 @@ int iwpm_mapinfo_available(void) spin_lock_irqsave(&iwpm_mapinfo_lock, flags); if (iwpm_hash_bucket) { - for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) { + for (i = 0; i < IWPM_MAPINFO_HASH_SIZE; i++) { if (!hlist_empty(&iwpm_hash_bucket[i])) { full_bucket = 1; break; diff --git a/drivers/infiniband/core/iwpm_util.h b/drivers/infiniband/core/iwpm_util.h index 9777c86..ee2d9ff 100644 --- a/drivers/infiniband/core/iwpm_util.h +++ b/drivers/infiniband/core/iwpm_util.h @@ -76,6 +76,14 @@ struct iwpm_mapping_info { u8 nl_client; }; +struct iwpm_remote_info { + struct hlist_node hlist_node; + struct sockaddr_storage remote_sockaddr; + struct sockaddr_storage mapped_loc_sockaddr; + struct sockaddr_storage mapped_rem_sockaddr; + u8 nl_client; +}; + struct iwpm_admin_data { atomic_t refcount; atomic_t nlmsg_seq; @@ -128,6 +136,13 @@ int iwpm_wait_complete_req(struct iwpm_nlmsg_request *nlmsg_request); int iwpm_get_nlmsg_seq(void); /** + * iwpm_add_reminfo - Add remote address info of the connecting peer + * to the remote info hash table + * @reminfo: The remote info to be added + */ +void iwpm_add_remote_info(struct iwpm_remote_info *reminfo); + +/** * iwpm_valid_client - Check if the port mapper client is valid * @nl_client: The index of the netlink client * diff --git a/include/rdma/iw_portmap.h b/include/rdma/iw_portmap.h index 928b277..fda3167 100644 --- a/include/rdma/iw_portmap.h +++ b/include/rdma/iw_portmap.h @@ -148,6 +148,16 @@ int iwpm_add_mapping_cb(struct sk_buff *, struct netlink_callback *); int iwpm_add_and_query_mapping_cb(struct sk_buff *, struct netlink_callback *); /** + * iwpm_remote_info_cb - Process remote connecting peer address info, which + * the port mapper has received from the connecting peer + * + * @cb: Contains the received message (payload and netlink header) + * + * Stores the IPv4/IPv6 address info in a hash table + */ +int iwpm_remote_info_cb(struct sk_buff *, struct netlink_callback *); + +/** * iwpm_mapping_error_cb - Process port mapper notification for error * * @skb: @@ -175,6 +185,21 @@ int iwpm_mapping_info_cb(struct sk_buff *, struct netlink_callback *); int iwpm_ack_mapping_info_cb(struct sk_buff *, struct netlink_callback *); /** + * iwpm_get_remote_info - Get the remote connecting peer address info + * + * @mapped_loc_addr: Mapped local address of the listening peer + * @mapped_rem_addr: Mapped remote address of the connecting peer + * @remote_addr: To store the remote address of the connecting peer + * @nl_client: The index of the netlink client + * + * The remote address info is retrieved and provided to the client in + * the remote_addr. After that it is removed from the hash table + */ +int iwpm_get_remote_info(struct sockaddr_storage *mapped_loc_addr, + struct sockaddr_storage *mapped_rem_addr, + struct sockaddr_storage *remote_addr, u8 nl_client); + +/** * iwpm_create_mapinfo - Store local and mapped IPv4/IPv6 address * info in a hash table * @local_addr: Local ip/tcp address diff --git a/include/uapi/rdma/rdma_netlink.h b/include/uapi/rdma/rdma_netlink.h index de69170..6e4bb42 100644 --- a/include/uapi/rdma/rdma_netlink.h +++ b/include/uapi/rdma/rdma_netlink.h @@ -37,6 +37,7 @@ enum { RDMA_NL_IWPM_ADD_MAPPING, RDMA_NL_IWPM_QUERY_MAPPING, RDMA_NL_IWPM_REMOVE_MAPPING, + RDMA_NL_IWPM_REMOTE_INFO, RDMA_NL_IWPM_HANDLE_ERR, RDMA_NL_IWPM_MAPINFO, RDMA_NL_IWPM_MAPINFO_NUM, -- cgit v0.10.2 From 230da36ae919e690dbcc44d1be8a2154214c6e36 Mon Sep 17 00:00:00 2001 From: Tatyana Nikolova Date: Tue, 21 Apr 2015 16:28:25 -0400 Subject: RDMA/nes: Report the actual address of the remote connecting peer Get the actual (non-mapped) ip/tcp address of the connecting peer from the port mapper and report the address info to the user space application at the time of connection establishment Signed-off-by: Tatyana Nikolova Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c index 3b2a6dc..9f9d5c5 100644 --- a/drivers/infiniband/hw/nes/nes.c +++ b/drivers/infiniband/hw/nes/nes.c @@ -116,6 +116,7 @@ static struct ibnl_client_cbs nes_nl_cb_table[] = { [RDMA_NL_IWPM_REG_PID] = {.dump = iwpm_register_pid_cb}, [RDMA_NL_IWPM_ADD_MAPPING] = {.dump = iwpm_add_mapping_cb}, [RDMA_NL_IWPM_QUERY_MAPPING] = {.dump = iwpm_add_and_query_mapping_cb}, + [RDMA_NL_IWPM_REMOTE_INFO] = {.dump = iwpm_remote_info_cb}, [RDMA_NL_IWPM_HANDLE_ERR] = {.dump = iwpm_mapping_error_cb}, [RDMA_NL_IWPM_MAPINFO] = {.dump = iwpm_mapping_info_cb}, [RDMA_NL_IWPM_MAPINFO_NUM] = {.dump = iwpm_ack_mapping_info_cb} diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index 6f09a72..72b4341 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -596,27 +596,52 @@ static void nes_form_reg_msg(struct nes_vnic *nesvnic, memcpy(pm_msg->if_name, nesvnic->netdev->name, IWPM_IFNAME_SIZE); } +static void record_sockaddr_info(struct sockaddr_storage *addr_info, + nes_addr_t *ip_addr, u16 *port_num) +{ + struct sockaddr_in *in_addr = (struct sockaddr_in *)addr_info; + + if (in_addr->sin_family == AF_INET) { + *ip_addr = ntohl(in_addr->sin_addr.s_addr); + *port_num = ntohs(in_addr->sin_port); + } +} + /* * nes_record_pm_msg - Save the received mapping info */ static void nes_record_pm_msg(struct nes_cm_info *cm_info, struct iwpm_sa_data *pm_msg) { - struct sockaddr_in *mapped_loc_addr = - (struct sockaddr_in *)&pm_msg->mapped_loc_addr; - struct sockaddr_in *mapped_rem_addr = - (struct sockaddr_in *)&pm_msg->mapped_rem_addr; - - if (mapped_loc_addr->sin_family == AF_INET) { - cm_info->mapped_loc_addr = - ntohl(mapped_loc_addr->sin_addr.s_addr); - cm_info->mapped_loc_port = ntohs(mapped_loc_addr->sin_port); - } - if (mapped_rem_addr->sin_family == AF_INET) { - cm_info->mapped_rem_addr = - ntohl(mapped_rem_addr->sin_addr.s_addr); - cm_info->mapped_rem_port = ntohs(mapped_rem_addr->sin_port); - } + record_sockaddr_info(&pm_msg->mapped_loc_addr, + &cm_info->mapped_loc_addr, &cm_info->mapped_loc_port); + + record_sockaddr_info(&pm_msg->mapped_rem_addr, + &cm_info->mapped_rem_addr, &cm_info->mapped_rem_port); +} + +/* + * nes_get_reminfo - Get the address info of the remote connecting peer + */ +static int nes_get_remote_addr(struct nes_cm_node *cm_node) +{ + struct sockaddr_storage mapped_loc_addr, mapped_rem_addr; + struct sockaddr_storage remote_addr; + int ret; + + nes_create_sockaddr(htonl(cm_node->mapped_loc_addr), + htons(cm_node->mapped_loc_port), &mapped_loc_addr); + nes_create_sockaddr(htonl(cm_node->mapped_rem_addr), + htons(cm_node->mapped_rem_port), &mapped_rem_addr); + + ret = iwpm_get_remote_info(&mapped_loc_addr, &mapped_rem_addr, + &remote_addr, RDMA_NL_NES); + if (ret) + nes_debug(NES_DBG_CM, "Unable to find remote peer address info\n"); + else + record_sockaddr_info(&remote_addr, &cm_node->rem_addr, + &cm_node->rem_port); + return ret; } /** @@ -1566,9 +1591,14 @@ static struct nes_cm_node *make_cm_node(struct nes_cm_core *cm_core, return NULL; /* set our node specific transport info */ - cm_node->loc_addr = cm_info->loc_addr; + if (listener) { + cm_node->loc_addr = listener->loc_addr; + cm_node->loc_port = listener->loc_port; + } else { + cm_node->loc_addr = cm_info->loc_addr; + cm_node->loc_port = cm_info->loc_port; + } cm_node->rem_addr = cm_info->rem_addr; - cm_node->loc_port = cm_info->loc_port; cm_node->rem_port = cm_info->rem_port; cm_node->mapped_loc_addr = cm_info->mapped_loc_addr; @@ -2151,6 +2181,7 @@ static int handle_ack_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb, cm_node->state = NES_CM_STATE_ESTABLISHED; if (datasize) { cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize; + nes_get_remote_addr(cm_node); handle_rcv_mpa(cm_node, skb); } else { /* rcvd ACK only */ dev_kfree_skb_any(skb); -- cgit v0.10.2 From 5b6b8fe64053b2649660ded2f3c5be25ebddbfdb Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Tue, 21 Apr 2015 16:28:41 -0400 Subject: RDMA/cxgb4: Report the actual address of the remote connecting peer Get the actual (non-mapped) ip/tcp address of the connecting peer from the port mapper Also setup the passive side endpoint to correctly display the actual and mapped addresses for the new connection. Signed-off-by: Steve Wise Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 6fb31ba..3c3b00e 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -583,6 +583,22 @@ static void c4iw_record_pm_msg(struct c4iw_ep *ep, sizeof(ep->com.mapped_remote_addr)); } +static int get_remote_addr(struct c4iw_ep *ep) +{ + int ret; + + print_addr(&ep->com, __func__, "get_remote_addr"); + + ret = iwpm_get_remote_info(&ep->com.mapped_local_addr, + &ep->com.mapped_remote_addr, + &ep->com.remote_addr, RDMA_NL_C4IW); + if (ret) + pr_info(MOD "Unable to find remote peer addr info - err %d\n", + ret); + + return ret; +} + static void best_mtu(const unsigned short *mtus, unsigned short mtu, unsigned int *idx, int use_ts, int ipv6) { @@ -2352,27 +2368,57 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) state_set(&child_ep->com, CONNECTING); child_ep->com.dev = dev; child_ep->com.cm_id = NULL; + + /* + * The mapped_local and mapped_remote addresses get setup with + * the actual 4-tuple. The local address will be based on the + * actual local address of the connection, but on the port number + * of the parent listening endpoint. The remote address is + * setup based on a query to the IWPM since we don't know what it + * originally was before mapping. If no mapping was done, then + * mapped_remote == remote, and mapped_local == local. + */ if (iptype == 4) { struct sockaddr_in *sin = (struct sockaddr_in *) - &child_ep->com.local_addr; + &child_ep->com.mapped_local_addr; + sin->sin_family = PF_INET; sin->sin_port = local_port; sin->sin_addr.s_addr = *(__be32 *)local_ip; - sin = (struct sockaddr_in *)&child_ep->com.remote_addr; + + sin = (struct sockaddr_in *)&child_ep->com.local_addr; + sin->sin_family = PF_INET; + sin->sin_port = ((struct sockaddr_in *) + &parent_ep->com.local_addr)->sin_port; + sin->sin_addr.s_addr = *(__be32 *)local_ip; + + sin = (struct sockaddr_in *)&child_ep->com.mapped_remote_addr; sin->sin_family = PF_INET; sin->sin_port = peer_port; sin->sin_addr.s_addr = *(__be32 *)peer_ip; } else { struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) - &child_ep->com.local_addr; + &child_ep->com.mapped_local_addr; + sin6->sin6_family = PF_INET6; sin6->sin6_port = local_port; memcpy(sin6->sin6_addr.s6_addr, local_ip, 16); - sin6 = (struct sockaddr_in6 *)&child_ep->com.remote_addr; + + sin6 = (struct sockaddr_in6 *)&child_ep->com.local_addr; + sin6->sin6_family = PF_INET6; + sin6->sin6_port = ((struct sockaddr_in6 *) + &parent_ep->com.local_addr)->sin6_port; + memcpy(sin6->sin6_addr.s6_addr, local_ip, 16); + + sin6 = (struct sockaddr_in6 *)&child_ep->com.mapped_remote_addr; sin6->sin6_family = PF_INET6; sin6->sin6_port = peer_port; memcpy(sin6->sin6_addr.s6_addr, peer_ip, 16); } + memcpy(&child_ep->com.remote_addr, &child_ep->com.mapped_remote_addr, + sizeof(child_ep->com.remote_addr)); + get_remote_addr(child_ep); + c4iw_get_ep(&parent_ep->com); child_ep->parent_ep = parent_ep; child_ep->tos = PASS_OPEN_TOS_G(ntohl(req->tos_stid)); diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 83209bb..1ffbd03 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -93,6 +93,7 @@ static struct ibnl_client_cbs c4iw_nl_cb_table[] = { [RDMA_NL_IWPM_ADD_MAPPING] = {.dump = iwpm_add_mapping_cb}, [RDMA_NL_IWPM_QUERY_MAPPING] = {.dump = iwpm_add_and_query_mapping_cb}, [RDMA_NL_IWPM_HANDLE_ERR] = {.dump = iwpm_mapping_error_cb}, + [RDMA_NL_IWPM_REMOTE_INFO] = {.dump = iwpm_remote_info_cb}, [RDMA_NL_IWPM_MAPINFO] = {.dump = iwpm_mapping_info_cb}, [RDMA_NL_IWPM_MAPINFO_NUM] = {.dump = iwpm_ack_mapping_info_cb} }; -- cgit v0.10.2 From c1d383b5785b1e0fb5fb862864712a7208219e6a Mon Sep 17 00:00:00 2001 From: Guy Shapiro Date: Wed, 15 Apr 2015 18:17:56 +0300 Subject: IB/core: dma map/unmap locking optimizations Currently, while mapping or unmapping pages for ODP, the umem mutex is locked and unlocked once for each page. Such lock/unlock operation take few tens to hundreds of nsecs. This makes a significant impact when mapping or unmapping few MBs of memory. To avoid this, the mutex should be locked only once per operation, and not per page. Signed-off-by: Guy Shapiro Acked-by: Shachar Raindel Reviewed-by: Sagi Grimberg Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index 8b8cc6f..aba4739 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -446,7 +446,6 @@ static int ib_umem_odp_map_dma_single_page( int remove_existing_mapping = 0; int ret = 0; - mutex_lock(&umem->odp_data->umem_mutex); /* * Note: we avoid writing if seq is different from the initial seq, to * handle case of a racing notifier. This check also allows us to bail @@ -479,8 +478,6 @@ static int ib_umem_odp_map_dma_single_page( } out: - mutex_unlock(&umem->odp_data->umem_mutex); - /* On Demand Paging - avoid pinning the page */ if (umem->context->invalidate_range || !stored_page) put_page(page); @@ -586,6 +583,7 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, bcnt -= min_t(size_t, npages << PAGE_SHIFT, bcnt); user_virt += npages << PAGE_SHIFT; + mutex_lock(&umem->odp_data->umem_mutex); for (j = 0; j < npages; ++j) { ret = ib_umem_odp_map_dma_single_page( umem, k, base_virt_addr, local_page_list[j], @@ -594,6 +592,7 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, break; k++; } + mutex_unlock(&umem->odp_data->umem_mutex); if (ret < 0) { /* Release left over pages when handling errors. */ @@ -633,9 +632,9 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt, * faults from completion. We might be racing with other * invalidations, so we must make sure we free each page only * once. */ + mutex_lock(&umem->odp_data->umem_mutex); for (addr = virt; addr < bound; addr += (u64)umem->page_size) { idx = (addr - ib_umem_start(umem)) / PAGE_SIZE; - mutex_lock(&umem->odp_data->umem_mutex); if (umem->odp_data->page_list[idx]) { struct page *page = umem->odp_data->page_list[idx]; struct page *head_page = compound_head(page); @@ -663,7 +662,7 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt, umem->odp_data->page_list[idx] = NULL; umem->odp_data->dma_list[idx] = 0; } - mutex_unlock(&umem->odp_data->umem_mutex); } + mutex_unlock(&umem->odp_data->umem_mutex); } EXPORT_SYMBOL(ib_umem_odp_unmap_dma_pages); -- cgit v0.10.2 From 325ad0617adaf163e32dd2d857b90baf65a25b5b Mon Sep 17 00:00:00 2001 From: Guy Shapiro Date: Wed, 15 Apr 2015 18:17:57 +0300 Subject: IB/core: dma unmap optimizations While unmapping an ODP writable page, the dirty bit of the page is set. In order to do so, the head of the compound page is found. Currently, the compound head is found even on non-writable pages, where it is never used, leading to unnecessary cpu barrier that impacts performance. This patch moves the search for the compound head to be done only when needed. Signed-off-by: Guy Shapiro Acked-by: Shachar Raindel Reviewed-by: Sagi Grimberg Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index aba4739..40becdb 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -637,7 +637,6 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt, idx = (addr - ib_umem_start(umem)) / PAGE_SIZE; if (umem->odp_data->page_list[idx]) { struct page *page = umem->odp_data->page_list[idx]; - struct page *head_page = compound_head(page); dma_addr_t dma = umem->odp_data->dma_list[idx]; dma_addr_t dma_addr = dma & ODP_DMA_ADDR_MASK; @@ -645,7 +644,8 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt, ib_dma_unmap_page(dev, dma_addr, PAGE_SIZE, DMA_BIDIRECTIONAL); - if (dma & ODP_WRITE_ALLOWED_BIT) + if (dma & ODP_WRITE_ALLOWED_BIT) { + struct page *head_page = compound_head(page); /* * set_page_dirty prefers being called with * the page lock. However, MMU notifiers are @@ -656,6 +656,7 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt, * be removed. */ set_page_dirty(head_page); + } /* on demand pinning support */ if (!umem->context->invalidate_range) put_page(page); -- cgit v0.10.2 From 87a26e976cb93e26742224bdd39f51f7861aa9b7 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Tue, 21 Apr 2015 14:50:34 -0700 Subject: IB/qib: add acounting for MTRR There is no good reason not to, we eventually delete it as well. Cc: Toshi Kani Cc: Suresh Siddha Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Juergen Gross Cc: Daniel Vetter Cc: Andy Lutomirski Cc: Dave Airlie Cc: Antonino Daplas Cc: Jean-Christophe Plagniol-Villard Cc: Tomi Valkeinen Cc: Mike Marciniszyn Cc: Roland Dreier Cc: Sean Hefty Cc: Hal Rosenstock Cc: linux-rdma@vger.kernel.org Cc: linux-fbdev@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Luis R. Rodriguez Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/qib/qib_wc_x86_64.c b/drivers/infiniband/hw/qib/qib_wc_x86_64.c index 81b225f..fe0850a 100644 --- a/drivers/infiniband/hw/qib/qib_wc_x86_64.c +++ b/drivers/infiniband/hw/qib/qib_wc_x86_64.c @@ -118,7 +118,7 @@ int qib_enable_wc(struct qib_devdata *dd) if (!ret) { int cookie; - cookie = mtrr_add(pioaddr, piolen, MTRR_TYPE_WRCOMB, 0); + cookie = mtrr_add(pioaddr, piolen, MTRR_TYPE_WRCOMB, 1); if (cookie < 0) { { qib_devinfo(dd->pcidev, -- cgit v0.10.2 From d4988623cc605131bed8c77f007082c3555c39ee Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Wed, 22 Apr 2015 11:38:24 -0700 Subject: IB/qib: use arch_phys_wc_add() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This driver already makes use of ioremap_wc() on PIO buffers, so convert it to use arch_phys_wc_add(). The qib driver uses a mmap() special case for when PAT is not used, this behaviour used to be determined with a module parameter but since we have been asked to just remove that module parameter this checks for the WC cookie, if not set we can assume PAT was used. If its set we do what we used to do for the mmap for when MTRR was enabled. The removal of the module parameter is OK given that Andy notes that even if users of module parameter are still around it will not prevent loading of the module on recent kernels. Cc: Doug Ledford Cc: Toshi Kani Cc: Rickard Strandqvist Cc: Mike Marciniszyn Cc: Roland Dreier Cc: Sean Hefty Cc: Hal Rosenstock Cc: Dennis Dalessandro Cc: Andy Lutomirski Cc: Suresh Siddha Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Juergen Gross Cc: Daniel Vetter Cc: Dave Airlie Cc: Bjorn Helgaas Cc: Antonino Daplas Cc: Jean-Christophe Plagniol-Villard Cc: Tomi Valkeinen Cc: Dave Hansen Cc: Arnd Bergmann Cc: Michael S. Tsirkin Cc: Stefan Bader Cc: konrad.wilk@oracle.com Cc: ville.syrjala@linux.intel.com Cc: david.vrabel@citrix.com Cc: jbeulich@suse.com Cc: Roger Pau Monné Cc: infinipath@intel.com Cc: linux-rdma@vger.kernel.org Cc: linux-fbdev@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: xen-devel@lists.xensource.com Signed-off-by: Luis R. Rodriguez Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/qib/qib.h b/drivers/infiniband/hw/qib/qib.h index ffd48bf..ba5173e2 100644 --- a/drivers/infiniband/hw/qib/qib.h +++ b/drivers/infiniband/hw/qib/qib.h @@ -1136,7 +1136,6 @@ extern struct qib_devdata *qib_lookup(int unit); extern u32 qib_cpulist_count; extern unsigned long *qib_cpulist; -extern unsigned qib_wc_pat; extern unsigned qib_cc_table_size; int qib_init(struct qib_devdata *, int); int init_chip_wc_pat(struct qib_devdata *dd, u32); diff --git a/drivers/infiniband/hw/qib/qib_file_ops.c b/drivers/infiniband/hw/qib/qib_file_ops.c index 9ea6c44..7258818 100644 --- a/drivers/infiniband/hw/qib/qib_file_ops.c +++ b/drivers/infiniband/hw/qib/qib_file_ops.c @@ -835,7 +835,8 @@ static int mmap_piobufs(struct vm_area_struct *vma, vma->vm_flags &= ~VM_MAYREAD; vma->vm_flags |= VM_DONTCOPY | VM_DONTEXPAND; - if (qib_wc_pat) + /* We used PAT if wc_cookie == 0 */ + if (!dd->wc_cookie) vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); ret = io_remap_pfn_range(vma, vma->vm_start, phys >> PAGE_SHIFT, diff --git a/drivers/infiniband/hw/qib/qib_iba6120.c b/drivers/infiniband/hw/qib/qib_iba6120.c index 0d2ba59..4b927809 100644 --- a/drivers/infiniband/hw/qib/qib_iba6120.c +++ b/drivers/infiniband/hw/qib/qib_iba6120.c @@ -3315,11 +3315,9 @@ static int init_6120_variables(struct qib_devdata *dd) qib_6120_config_ctxts(dd); qib_set_ctxtcnt(dd); - if (qib_wc_pat) { - ret = init_chip_wc_pat(dd, 0); - if (ret) - goto bail; - } + ret = init_chip_wc_pat(dd, 0); + if (ret) + goto bail; set_6120_baseaddrs(dd); /* set chip access pointers now */ ret = 0; diff --git a/drivers/infiniband/hw/qib/qib_iba7220.c b/drivers/infiniband/hw/qib/qib_iba7220.c index 22affda..00b2af2 100644 --- a/drivers/infiniband/hw/qib/qib_iba7220.c +++ b/drivers/infiniband/hw/qib/qib_iba7220.c @@ -4126,11 +4126,9 @@ static int qib_init_7220_variables(struct qib_devdata *dd) qib_7220_config_ctxts(dd); qib_set_ctxtcnt(dd); /* needed for PAT setup */ - if (qib_wc_pat) { - ret = init_chip_wc_pat(dd, 0); - if (ret) - goto bail; - } + ret = init_chip_wc_pat(dd, 0); + if (ret) + goto bail; set_7220_baseaddrs(dd); /* set chip access pointers now */ ret = 0; diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index ef97b71..f32b462 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -6429,6 +6429,7 @@ static int qib_init_7322_variables(struct qib_devdata *dd) unsigned features, pidx, sbufcnt; int ret, mtu; u32 sbufs, updthresh; + resource_size_t vl15off; /* pport structs are contiguous, allocated after devdata */ ppd = (struct qib_pportdata *)(dd + 1); @@ -6677,29 +6678,27 @@ static int qib_init_7322_variables(struct qib_devdata *dd) qib_7322_config_ctxts(dd); qib_set_ctxtcnt(dd); - if (qib_wc_pat) { - resource_size_t vl15off; - /* - * We do not set WC on the VL15 buffers to avoid - * a rare problem with unaligned writes from - * interrupt-flushed store buffers, so we need - * to map those separately here. We can't solve - * this for the rarely used mtrr case. - */ - ret = init_chip_wc_pat(dd, 0); - if (ret) - goto bail; + /* + * We do not set WC on the VL15 buffers to avoid + * a rare problem with unaligned writes from + * interrupt-flushed store buffers, so we need + * to map those separately here. We can't solve + * this for the rarely used mtrr case. + */ + ret = init_chip_wc_pat(dd, 0); + if (ret) + goto bail; - /* vl15 buffers start just after the 4k buffers */ - vl15off = dd->physaddr + (dd->piobufbase >> 32) + - dd->piobcnt4k * dd->align4k; - dd->piovl15base = ioremap_nocache(vl15off, - NUM_VL15_BUFS * dd->align4k); - if (!dd->piovl15base) { - ret = -ENOMEM; - goto bail; - } + /* vl15 buffers start just after the 4k buffers */ + vl15off = dd->physaddr + (dd->piobufbase >> 32) + + dd->piobcnt4k * dd->align4k; + dd->piovl15base = ioremap_nocache(vl15off, + NUM_VL15_BUFS * dd->align4k); + if (!dd->piovl15base) { + ret = -ENOMEM; + goto bail; } + qib_7322_set_baseaddrs(dd); /* set chip access pointers now */ ret = 0; diff --git a/drivers/infiniband/hw/qib/qib_init.c b/drivers/infiniband/hw/qib/qib_init.c index 2ee3695..7e00470 100644 --- a/drivers/infiniband/hw/qib/qib_init.c +++ b/drivers/infiniband/hw/qib/qib_init.c @@ -91,15 +91,6 @@ MODULE_PARM_DESC(krcvqs, "number of kernel receive queues per IB port"); unsigned qib_cc_table_size; module_param_named(cc_table_size, qib_cc_table_size, uint, S_IRUGO); MODULE_PARM_DESC(cc_table_size, "Congestion control table entries 0 (CCA disabled - default), min = 128, max = 1984"); -/* - * qib_wc_pat parameter: - * 0 is WC via MTRR - * 1 is WC via PAT - * If PAT initialization fails, code reverts back to MTRR - */ -unsigned qib_wc_pat = 1; /* default (1) is to use PAT, not MTRR */ -module_param_named(wc_pat, qib_wc_pat, uint, S_IRUGO); -MODULE_PARM_DESC(wc_pat, "enable write-combining via PAT mechanism"); static void verify_interrupt(unsigned long); @@ -1377,8 +1368,7 @@ static void cleanup_device_data(struct qib_devdata *dd) spin_unlock(&dd->pport[pidx].cc_shadow_lock); } - if (!qib_wc_pat) - qib_disable_wc(dd); + qib_disable_wc(dd); if (dd->pioavailregs_dma) { dma_free_coherent(&dd->pcidev->dev, PAGE_SIZE, @@ -1547,14 +1537,12 @@ static int qib_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto bail; } - if (!qib_wc_pat) { - ret = qib_enable_wc(dd); - if (ret) { - qib_dev_err(dd, - "Write combining not enabled (err %d): performance may be poor\n", - -ret); - ret = 0; - } + ret = qib_enable_wc(dd); + if (ret) { + qib_dev_err(dd, + "Write combining not enabled (err %d): performance may be poor\n", + -ret); + ret = 0; } qib_verify_pioperf(dd); diff --git a/drivers/infiniband/hw/qib/qib_wc_x86_64.c b/drivers/infiniband/hw/qib/qib_wc_x86_64.c index fe0850a..6d61ef9 100644 --- a/drivers/infiniband/hw/qib/qib_wc_x86_64.c +++ b/drivers/infiniband/hw/qib/qib_wc_x86_64.c @@ -116,21 +116,9 @@ int qib_enable_wc(struct qib_devdata *dd) } if (!ret) { - int cookie; - - cookie = mtrr_add(pioaddr, piolen, MTRR_TYPE_WRCOMB, 1); - if (cookie < 0) { - { - qib_devinfo(dd->pcidev, - "mtrr_add() WC for PIO bufs failed (%d)\n", - cookie); - ret = -EINVAL; - } - } else { - dd->wc_cookie = cookie; - dd->wc_base = (unsigned long) pioaddr; - dd->wc_len = (unsigned long) piolen; - } + dd->wc_cookie = arch_phys_wc_add(pioaddr, piolen); + if (dd->wc_cookie < 0) + ret = -EINVAL; } return ret; @@ -142,18 +130,7 @@ int qib_enable_wc(struct qib_devdata *dd) */ void qib_disable_wc(struct qib_devdata *dd) { - if (dd->wc_cookie) { - int r; - - r = mtrr_del(dd->wc_cookie, dd->wc_base, - dd->wc_len); - if (r < 0) - qib_devinfo(dd->pcidev, - "mtrr_del(%lx, %lx, %lx) failed: %d\n", - dd->wc_cookie, dd->wc_base, - dd->wc_len, r); - dd->wc_cookie = 0; /* even on failure */ - } + arch_phys_wc_del(dd->wc_cookie); } /** -- cgit v0.10.2 From d73a824acc705571c0f47596326d7967fba9a1d9 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 4 May 2015 14:35:01 -0400 Subject: drm/radeon: don't setup audio on asics that don't support it bug: https://bugzilla.kernel.org/show_bug.cgi?id=97701 Reported-by: Mikael Pettersson Tested-by: Mikael Pettersson Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c index 8b82abb..dcb7796 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.c +++ b/drivers/gpu/drm/radeon/radeon_audio.c @@ -464,6 +464,10 @@ void radeon_audio_detect(struct drm_connector *connector, return; rdev = connector->encoder->dev->dev_private; + + if (!radeon_audio_chipset_supported(rdev)) + return; + radeon_encoder = to_radeon_encoder(connector->encoder); dig = radeon_encoder->enc_priv; -- cgit v0.10.2 From 6f317cfe42c9d8a7c9c1a327d2f1bcc517a3cd91 Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Sun, 12 Apr 2015 21:10:35 +0200 Subject: drm/i915: Assume dual channel LVDS if pixel clock necessitates it MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Single channel LVDS maxes out at 112 MHz, anything above must be dual channel. This avoids the need to specify i915.lvds_channel_mode=2 on all 17" MacBook Pro models with i915 graphics since they had 1920x1200 (193 MHz), plus those 15" pre-retina models which had a resolution of 1680x1050 (119 MHz) as a BTO option. Source for 112 MHz limit of single channel LVDS is section 2.3 of: https://01.org/linuxgraphics/sites/default/files/documentation/ivb_ihd_os_vol3_part4.pdf v2: Avoid hardcoding 17" models by assuming dual channel LVDS if the resolution necessitates it, suggested by Jani Nikula. v3: Fix typo, thanks Joonas Lahtinen. v4: Split commit in two, suggested by Ville Syrjälä. Signed-off-by: Lukas Wunner Tested-by: Lukas Wunner Reviewed-by: Ville Syrjälä Cc: stable@vger.kernel.org [Jani: included spec reference into the commit message] Signed-off-by: Jani Nikula diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index 5abda1d..ee621e3 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -848,6 +848,11 @@ static bool compute_is_dual_link_lvds(struct intel_lvds_encoder *lvds_encoder) if (i915.lvds_channel_mode > 0) return i915.lvds_channel_mode == 2; + /* single channel LVDS is limited to 112 MHz */ + if (lvds_encoder->attached_connector->base.panel.fixed_mode->clock + > 112999) + return true; + if (dmi_check_system(intel_dual_link_lvds)) return true; @@ -1111,6 +1116,8 @@ void intel_lvds_init(struct drm_device *dev) out: mutex_unlock(&dev->mode_config.mutex); + intel_panel_init(&intel_connector->panel, fixed_mode, downclock_mode); + lvds_encoder->is_dual_link = compute_is_dual_link_lvds(lvds_encoder); DRM_DEBUG_KMS("detected %s-link lvds configuration\n", lvds_encoder->is_dual_link ? "dual" : "single"); @@ -1125,7 +1132,6 @@ out: } drm_connector_register(connector); - intel_panel_init(&intel_connector->panel, fixed_mode, downclock_mode); intel_panel_setup_backlight(connector, INVALID_PIPE); return; -- cgit v0.10.2 From 3916e3fd81021fb795bfbdb17f375b6b3685bced Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Mon, 4 May 2015 15:06:49 +0200 Subject: drm/i915: Add missing MacBook Pro models with dual channel LVDS Single channel LVDS maxes out at 112 MHz. The 15" pre-retina models shipped with 1440x900 (106 MHz) by default or 1680x1050 (119 MHz) as a BTO option, both versions used dual channel LVDS even though the smaller one would have fit into a single channel. Notes: Bug report showing that the MacBookPro8,2 with 1440x900 uses dual channel LVDS (this lead to it being hardcoded in intel_lvds.c by Daniel Vetter with commit 618563e3945b9d0864154bab3c607865b557cecc): https://bugzilla.kernel.org/show_bug.cgi?id=42842 If i915.lvds_channel_mode=2 is missing even though the machine needs it, every other vertical line is white and consequently, only the left half of the screen is visible (verified by myself on a MacBookPro9,1). Forum posting concerning a MacBookPro6,2 with 1440x900, author is using i915.lvds_channel_mode=2 on the kernel command line, proving that the machine uses dual channels: https://bbs.archlinux.org/viewtopic.php?id=185770 Chi Mei N154C6-L04 with 1440x900 is a replacement panel for all MacBook Pro "A1286" models, and that model number encompasses the MacBookPro6,2 / 8,2 / 9,1. Page 17 of the panel's datasheet shows it's driven with dual channel LVDS: http://www.ebay.com/itm/-/400690878560 http://www.everymac.com/ultimate-mac-lookup/?search_keywords=A1286 http://www.taopanel.com/chimei/datasheet/N154C6-L04.pdf Those three 15" models, MacBookPro6,2 / 8,2 / 9,1, are the only ones with i915 graphics and dual channel LVDS, so that list should be complete. And the 8,2 is already in intel_lvds.c. Possible motivation to use dual channel LVDS even on the 1440x900 models: Reduce the number of different parts, i.e. use identical logic boards and display cabling on both versions and the only differing component is the panel. Signed-off-by: Lukas Wunner Acked-by: Jani Nikula Cc: stable@vger.kernel.org [Jani: included notes in the commit message for posterity] Signed-off-by: Jani Nikula diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index ee621e3..fbcc7df 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -813,12 +813,28 @@ static int intel_dual_link_lvds_callback(const struct dmi_system_id *id) static const struct dmi_system_id intel_dual_link_lvds[] = { { .callback = intel_dual_link_lvds_callback, - .ident = "Apple MacBook Pro (Core i5/i7 Series)", + .ident = "Apple MacBook Pro 15\" (2010)", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro6,2"), + }, + }, + { + .callback = intel_dual_link_lvds_callback, + .ident = "Apple MacBook Pro 15\" (2011)", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro8,2"), }, }, + { + .callback = intel_dual_link_lvds_callback, + .ident = "Apple MacBook Pro 15\" (2012)", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro9,1"), + }, + }, { } /* terminating entry */ }; -- cgit v0.10.2 From d67e199611b986b345ea3087ee2e4a15da1c98b3 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 21 Apr 2015 16:46:28 +0300 Subject: efi: Fix error handling in add_sysfs_runtime_map_entry() I spotted two (difficult to hit) bugs while reviewing this. 1) There is a double free bug because we unregister "map_kset" in add_sysfs_runtime_map_entry() and also efi_runtime_map_init(). 2) If we fail to allocate "entry" then we should return ERR_PTR(-ENOMEM) instead of NULL. Signed-off-by: Dan Carpenter Cc: Dave Young Cc: Guangyu Sun Cc: Signed-off-by: Matt Fleming diff --git a/drivers/firmware/efi/runtime-map.c b/drivers/firmware/efi/runtime-map.c index 87b8e3b..5c55227 100644 --- a/drivers/firmware/efi/runtime-map.c +++ b/drivers/firmware/efi/runtime-map.c @@ -120,7 +120,8 @@ add_sysfs_runtime_map_entry(struct kobject *kobj, int nr) entry = kzalloc(sizeof(*entry), GFP_KERNEL); if (!entry) { kset_unregister(map_kset); - return entry; + map_kset = NULL; + return ERR_PTR(-ENOMEM); } memcpy(&entry->md, efi_runtime_map + nr * efi_memdesc_size, @@ -132,6 +133,7 @@ add_sysfs_runtime_map_entry(struct kobject *kobj, int nr) if (ret) { kobject_put(&entry->kobj); kset_unregister(map_kset); + map_kset = NULL; return ERR_PTR(ret); } @@ -195,8 +197,6 @@ out_add_entry: entry = *(map_entries + j); kobject_put(&entry->kobj); } - if (map_kset) - kset_unregister(map_kset); out: return ret; } -- cgit v0.10.2 From e59d29e88f7b7e3d1231202b0203d0af6f15a440 Mon Sep 17 00:00:00 2001 From: Wang Nan Date: Tue, 28 Apr 2015 08:46:09 +0000 Subject: perf probe: Fix segfault if passed with ''. Since parse_perf_probe_point() deals with a user passed argument, we should not assume it to be a valid string. Without this patch, if pass '' to perf probe, a segfault raises: $ perf probe -a '' Segmentation fault This patch checks argument of parse_perf_probe_point() before string processing. After this patch: $ perf probe -a '' usage: perf probe [] 'PROBEDEF' ['PROBEDEF' ...] or: perf probe [] --add 'PROBEDEF' [--add 'PROBEDEF' ...] ... Signed-off-by: Wang Nan Acked-by: Masami Hiramatsu Tested-by: Arnaldo Carvalho de Melo Cc: David Ahern Cc: Jiri Olsa Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Zefan Li Link: http://lkml.kernel.org/r/1430210769-94177-1-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index d8bb616..d05b77c 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -1084,6 +1084,8 @@ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) * * TODO:Group name support */ + if (!arg) + return -EINVAL; ptr = strpbrk(arg, ";=@+%"); if (ptr && *ptr == '=') { /* Event name */ -- cgit v0.10.2 From c0403ec0bb5a8c5b267fb7e16021bec0b17e4964 Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Tue, 5 May 2015 15:15:56 +0200 Subject: Revert "dm crypt: fix deadlock when async crypto algorithm returns -EBUSY" This reverts Linux 4.1-rc1 commit 0618764cb25f6fa9fb31152995de42a8a0496475. The problem which that commit attempts to fix actually lies in the Freescale CAAM crypto driver not dm-crypt. dm-crypt uses CRYPTO_TFM_REQ_MAY_BACKLOG. This means the the crypto driver should internally backlog requests which arrive when the queue is full and process them later. Until the crypto hw's queue becomes full, the driver returns -EINPROGRESS. When the crypto hw's queue if full, the driver returns -EBUSY, and if CRYPTO_TFM_REQ_MAY_BACKLOG is set, is expected to backlog the request and process it when the hardware has queue space. At the point when the driver takes the request from the backlog and starts processing it, it calls the completion function with a status of -EINPROGRESS. The completion function is called (for a second time, in the case of backlogged requests) with a status/err of 0 when a request is done. Crypto drivers for hardware without hardware queueing use the helpers, crypto_init_queue(), crypto_enqueue_request(), crypto_dequeue_request() and crypto_get_backlog() helpers to implement this behaviour correctly, while others implement this behaviour without these helpers (ccp, for example). dm-crypt (before the patch that needs reverting) uses this API correctly. It queues up as many requests as the hw queues will allow (i.e. as long as it gets back -EINPROGRESS from the request function). Then, when it sees at least one backlogged request (gets -EBUSY), it waits till that backlogged request is handled (completion gets called with -EINPROGRESS), and then continues. The references to af_alg_wait_for_completion() and af_alg_complete() in that commit's commit message are irrelevant because those functions only handle one request at a time, unlink dm-crypt. The problem is that the Freescale CAAM driver, which that commit describes as having being tested with, fails to implement the backlogging behaviour correctly. In cam_jr_enqueue(), if the hardware queue is full, it simply returns -EBUSY without backlogging the request. What the observed deadlock was is not described in the commit message but it is obviously the wait_for_completion() in crypto_convert() where dm-crypto would wait for the completion being called with -EINPROGRESS in the case of backlogged requests. This completion will never be completed due to the bug in the CAAM driver. Commit 0618764cb25 incorrectly made dm-crypt wait for every request, even when the driver/hardware queues are not full, which means that dm-crypt will never see -EBUSY. This means that that commit will cause a performance regression on all crypto drivers which implement the API correctly. Revert it. Correct backlog handling should be implemented in the CAAM driver instead. Cc'ing stable purely because commit 0618764cb25 did. If for some reason a stable@ kernel did pick up commit 0618764cb25 it should get reverted. Signed-off-by: Rabin Vincent Reviewed-by: Horia Geanta Cc: stable@vger.kernel.org Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 9eeea19..5503e43 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -925,10 +925,11 @@ static int crypt_convert(struct crypt_config *cc, switch (r) { /* async */ - case -EINPROGRESS: case -EBUSY: wait_for_completion(&ctx->restart); reinit_completion(&ctx->restart); + /* fall through*/ + case -EINPROGRESS: ctx->req = NULL; ctx->cc_sector++; continue; @@ -1345,8 +1346,10 @@ static void kcryptd_async_done(struct crypto_async_request *async_req, struct dm_crypt_io *io = container_of(ctx, struct dm_crypt_io, ctx); struct crypt_config *cc = io->cc; - if (error == -EINPROGRESS) + if (error == -EINPROGRESS) { + complete(&ctx->restart); return; + } if (!error && cc->iv_gen_ops && cc->iv_gen_ops->post) error = cc->iv_gen_ops->post(cc, iv_of_dmreq(cc, dmreq), dmreq); @@ -1357,15 +1360,12 @@ static void kcryptd_async_done(struct crypto_async_request *async_req, crypt_free_req(cc, req_of_dmreq(cc, dmreq), io->base_bio); if (!atomic_dec_and_test(&ctx->cc_pending)) - goto done; + return; if (bio_data_dir(io->base_bio) == READ) kcryptd_crypt_read_done(io); else kcryptd_crypt_write_io_submit(io, 1); -done: - if (!completion_done(&ctx->restart)) - complete(&ctx->restart); } static void kcryptd_crypt(struct work_struct *work) -- cgit v0.10.2 From cb31ef485dd4c6a205d1064b42027f82076d00c8 Mon Sep 17 00:00:00 2001 From: Chen Yu Date: Sun, 3 May 2015 22:35:05 +0800 Subject: init: fix regression by supporting devices with major:minor:offset format Commit 283e7ad02 ("init: stricter checking of major:minor root= values") was so strict that it exposed the fact that a previously unknown device format was being used. Distributions like Ubuntu uses klibc (rather than uswsusp) to resume system from hibernation. klibc expressed the swap partition/file in the form of major:minor:offset. For example, 8:3:0 represents a swap partition in klibc, and klibc's resume process in initrd will finally echo 8:3:0 to /sys/power/resume for manually resuming. However, due to commit 283e7ad02's stricter checking, 8:3:0 will be treated as an invalid device format, and manual resuming from hibernation will fail. Fix this by adding support for devices with major:minor:offset format when resuming from hibernation. Reported-by: Prigent, Christophe Signed-off-by: Chen Yu Acked-by: Rafael J. Wysocki Signed-off-by: Mike Snitzer diff --git a/init/do_mounts.c b/init/do_mounts.c index 8369ffa..a95bbdb 100644 --- a/init/do_mounts.c +++ b/init/do_mounts.c @@ -225,10 +225,11 @@ dev_t name_to_dev_t(const char *name) #endif if (strncmp(name, "/dev/", 5) != 0) { - unsigned maj, min; + unsigned maj, min, offset; char dummy; - if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2) { + if ((sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2) || + (sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3)) { res = MKDEV(maj, min); if (maj != MAJOR(res) || min != MINOR(res)) goto fail; -- cgit v0.10.2 From 471e70583217728955436a3fa6e5201e5c8c296a Mon Sep 17 00:00:00 2001 From: Honggang LI Date: Wed, 29 Apr 2015 17:40:44 +0800 Subject: IB/core: change rdma_gid2ip into void function as it always return zero Signed-off-by: Honggang Li Acked-by: Sean Hefty Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index f80da50..38339d2 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -472,13 +472,8 @@ int rdma_addr_find_dmac_by_grh(union ib_gid *sgid, union ib_gid *dgid, u8 *dmac, } sgid_addr, dgid_addr; - ret = rdma_gid2ip(&sgid_addr._sockaddr, sgid); - if (ret) - return ret; - - ret = rdma_gid2ip(&dgid_addr._sockaddr, dgid); - if (ret) - return ret; + rdma_gid2ip(&sgid_addr._sockaddr, sgid); + rdma_gid2ip(&dgid_addr._sockaddr, dgid); memset(&dev_addr, 0, sizeof(dev_addr)); @@ -512,10 +507,8 @@ int rdma_addr_find_smac_by_sgid(union ib_gid *sgid, u8 *smac, u16 *vlan_id) struct sockaddr_in6 _sockaddr_in6; } gid_addr; - ret = rdma_gid2ip(&gid_addr._sockaddr, sgid); + rdma_gid2ip(&gid_addr._sockaddr, sgid); - if (ret) - return ret; memset(&dev_addr, 0, sizeof(dev_addr)); ret = rdma_translate_ip(&gid_addr._sockaddr, &dev_addr, vlan_id); if (ret) diff --git a/include/rdma/ib_addr.h b/include/rdma/ib_addr.h index ce55906..ac54c27 100644 --- a/include/rdma/ib_addr.h +++ b/include/rdma/ib_addr.h @@ -160,7 +160,7 @@ static inline int rdma_ip2gid(struct sockaddr *addr, union ib_gid *gid) } /* Important - sockaddr should be a union of sockaddr_in and sockaddr_in6 */ -static inline int rdma_gid2ip(struct sockaddr *out, union ib_gid *gid) +static inline void rdma_gid2ip(struct sockaddr *out, union ib_gid *gid) { if (ipv6_addr_v4mapped((struct in6_addr *)gid)) { struct sockaddr_in *out_in = (struct sockaddr_in *)out; @@ -173,7 +173,6 @@ static inline int rdma_gid2ip(struct sockaddr *out, union ib_gid *gid) out_in->sin6_family = AF_INET6; memcpy(&out_in->sin6_addr.s6_addr, gid->raw, 16); } - return 0; } static inline void iboe_addr_get_sgid(struct rdma_dev_addr *dev_addr, -- cgit v0.10.2 From 0d0f738f6a11856a704dcd8fd3a008b200f17625 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Sun, 3 May 2015 09:48:26 -0400 Subject: IB/core: Fix unaligned accesses Addresses the following kernel logs seen during boot of sparc systems: Kernel unaligned access at TPC[103bce50] cm_find_listen+0x34/0xf8 [ib_cm] Kernel unaligned access at TPC[103bce50] cm_find_listen+0x34/0xf8 [ib_cm] Kernel unaligned access at TPC[103bce50] cm_find_listen+0x34/0xf8 [ib_cm] Kernel unaligned access at TPC[103bce50] cm_find_listen+0x34/0xf8 [ib_cm] Kernel unaligned access at TPC[103bce50] cm_find_listen+0x34/0xf8 [ib_cm] Signed-off-by: David Ahern Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index e28a494..0c14191 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -437,39 +437,38 @@ static struct cm_id_private * cm_acquire_id(__be32 local_id, __be32 remote_id) return cm_id_priv; } -static void cm_mask_copy(u8 *dst, u8 *src, u8 *mask) +static void cm_mask_copy(u32 *dst, const u32 *src, const u32 *mask) { int i; - for (i = 0; i < IB_CM_COMPARE_SIZE / sizeof(unsigned long); i++) - ((unsigned long *) dst)[i] = ((unsigned long *) src)[i] & - ((unsigned long *) mask)[i]; + for (i = 0; i < IB_CM_COMPARE_SIZE; i++) + dst[i] = src[i] & mask[i]; } static int cm_compare_data(struct ib_cm_compare_data *src_data, struct ib_cm_compare_data *dst_data) { - u8 src[IB_CM_COMPARE_SIZE]; - u8 dst[IB_CM_COMPARE_SIZE]; + u32 src[IB_CM_COMPARE_SIZE]; + u32 dst[IB_CM_COMPARE_SIZE]; if (!src_data || !dst_data) return 0; cm_mask_copy(src, src_data->data, dst_data->mask); cm_mask_copy(dst, dst_data->data, src_data->mask); - return memcmp(src, dst, IB_CM_COMPARE_SIZE); + return memcmp(src, dst, sizeof(src)); } -static int cm_compare_private_data(u8 *private_data, +static int cm_compare_private_data(u32 *private_data, struct ib_cm_compare_data *dst_data) { - u8 src[IB_CM_COMPARE_SIZE]; + u32 src[IB_CM_COMPARE_SIZE]; if (!dst_data) return 0; cm_mask_copy(src, private_data, dst_data->mask); - return memcmp(src, dst_data->data, IB_CM_COMPARE_SIZE); + return memcmp(src, dst_data->data, sizeof(src)); } /* @@ -538,7 +537,7 @@ static struct cm_id_private * cm_insert_listen(struct cm_id_private *cm_id_priv) static struct cm_id_private * cm_find_listen(struct ib_device *device, __be64 service_id, - u8 *private_data) + u32 *private_data) { struct rb_node *node = cm.listen_service_table.rb_node; struct cm_id_private *cm_id_priv; @@ -953,7 +952,7 @@ int ib_cm_listen(struct ib_cm_id *cm_id, __be64 service_id, __be64 service_mask, cm_mask_copy(cm_id_priv->compare_data->data, compare_data->data, compare_data->mask); memcpy(cm_id_priv->compare_data->mask, compare_data->mask, - IB_CM_COMPARE_SIZE); + sizeof(compare_data->mask)); } cm_id->state = IB_CM_LISTEN; diff --git a/drivers/infiniband/core/cm_msgs.h b/drivers/infiniband/core/cm_msgs.h index be068f4..8b76f0e 100644 --- a/drivers/infiniband/core/cm_msgs.h +++ b/drivers/infiniband/core/cm_msgs.h @@ -103,7 +103,7 @@ struct cm_req_msg { /* local ACK timeout:5, rsvd:3 */ u8 alt_offset139; - u8 private_data[IB_CM_REQ_PRIVATE_DATA_SIZE]; + u32 private_data[IB_CM_REQ_PRIVATE_DATA_SIZE / sizeof(u32)]; } __attribute__ ((packed)); @@ -801,7 +801,7 @@ struct cm_sidr_req_msg { __be16 rsvd; __be64 service_id; - u8 private_data[IB_CM_SIDR_REQ_PRIVATE_DATA_SIZE]; + u32 private_data[IB_CM_SIDR_REQ_PRIVATE_DATA_SIZE / sizeof(u32)]; } __attribute__ ((packed)); struct cm_sidr_rep_msg { diff --git a/include/rdma/ib_cm.h b/include/rdma/ib_cm.h index 0e3ff30..39ed2d2 100644 --- a/include/rdma/ib_cm.h +++ b/include/rdma/ib_cm.h @@ -105,7 +105,8 @@ enum ib_cm_data_size { IB_CM_SIDR_REQ_PRIVATE_DATA_SIZE = 216, IB_CM_SIDR_REP_PRIVATE_DATA_SIZE = 136, IB_CM_SIDR_REP_INFO_LENGTH = 72, - IB_CM_COMPARE_SIZE = 64 + /* compare done u32 at a time */ + IB_CM_COMPARE_SIZE = (64 / sizeof(u32)) }; struct ib_cm_id; @@ -337,8 +338,8 @@ void ib_destroy_cm_id(struct ib_cm_id *cm_id); #define IB_SDP_SERVICE_ID_MASK cpu_to_be64(0xFFFFFFFFFFFF0000ULL) struct ib_cm_compare_data { - u8 data[IB_CM_COMPARE_SIZE]; - u8 mask[IB_CM_COMPARE_SIZE]; + u32 data[IB_CM_COMPARE_SIZE]; + u32 mask[IB_CM_COMPARE_SIZE]; }; /** -- cgit v0.10.2 From 179d03bbfd2ebc63934753a696467d28bf9f5b64 Mon Sep 17 00:00:00 2001 From: Hariprasad S Date: Tue, 5 May 2015 03:55:24 +0530 Subject: iw_cxgb4: Remove negative advice dmesg warnings Remove these log messages in favor of per-endpoint counters as well as device-global counters that can be inspected via debugfs. Signed-off-by: Steve Wise Signed-off-by: Hariprasad Shenai Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 3c3b00e..bb95a6c 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -2058,9 +2058,12 @@ static int act_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb) status, status2errno(status)); if (is_neg_adv(status)) { - dev_warn(&dev->rdev.lldi.pdev->dev, - "Connection problems for atid %u status %u (%s)\n", - atid, status, neg_adv_str(status)); + PDBG("%s Connection problems for atid %u status %u (%s)\n", + __func__, atid, status, neg_adv_str(status)); + ep->stats.connect_neg_adv++; + mutex_lock(&dev->rdev.stats.lock); + dev->rdev.stats.neg_adv++; + mutex_unlock(&dev->rdev.stats.lock); return 0; } @@ -2566,9 +2569,13 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb) ep = lookup_tid(t, tid); if (is_neg_adv(req->status)) { - dev_warn(&dev->rdev.lldi.pdev->dev, - "Negative advice on abort - tid %u status %d (%s)\n", - ep->hwtid, req->status, neg_adv_str(req->status)); + PDBG("%s Negative advice on abort- tid %u status %d (%s)\n", + __func__, ep->hwtid, req->status, + neg_adv_str(req->status)); + ep->stats.abort_neg_adv++; + mutex_lock(&dev->rdev.stats.lock); + dev->rdev.stats.neg_adv++; + mutex_unlock(&dev->rdev.stats.lock); return 0; } PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, @@ -3977,9 +3984,11 @@ static int peer_abort_intr(struct c4iw_dev *dev, struct sk_buff *skb) return 0; } if (is_neg_adv(req->status)) { - dev_warn(&dev->rdev.lldi.pdev->dev, - "Negative advice on abort - tid %u status %d (%s)\n", - ep->hwtid, req->status, neg_adv_str(req->status)); + PDBG("%s Negative advice on abort- tid %u status %d (%s)\n", + __func__, ep->hwtid, req->status, + neg_adv_str(req->status)); + ep->stats.abort_neg_adv++; + dev->rdev.stats.neg_adv++; kfree_skb(skb); return 0; } diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 1ffbd03..cf54d69 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -490,6 +490,7 @@ static int stats_show(struct seq_file *seq, void *v) dev->rdev.stats.act_ofld_conn_fails); seq_printf(seq, "PAS_OFLD_CONN_FAILS: %10llu\n", dev->rdev.stats.pas_ofld_conn_fails); + seq_printf(seq, "NEG_ADV_RCVD: %10llu\n", dev->rdev.stats.neg_adv); seq_printf(seq, "AVAILABLE IRD: %10u\n", dev->avail_ird); return 0; } @@ -561,10 +562,13 @@ static int dump_ep(int id, void *p, void *data) cc = snprintf(epd->buf + epd->pos, space, "ep %p cm_id %p qp %p state %d flags 0x%lx " "history 0x%lx hwtid %d atid %d " + "conn_na %u abort_na %u " "%pI4:%d/%d <-> %pI4:%d/%d\n", ep, ep->com.cm_id, ep->com.qp, (int)ep->com.state, ep->com.flags, ep->com.history, ep->hwtid, ep->atid, + ep->stats.connect_neg_adv, + ep->stats.abort_neg_adv, &lsin->sin_addr, ntohs(lsin->sin_port), ntohs(mapped_lsin->sin_port), &rsin->sin_addr, ntohs(rsin->sin_port), @@ -582,10 +586,13 @@ static int dump_ep(int id, void *p, void *data) cc = snprintf(epd->buf + epd->pos, space, "ep %p cm_id %p qp %p state %d flags 0x%lx " "history 0x%lx hwtid %d atid %d " + "conn_na %u abort_na %u " "%pI6:%d/%d <-> %pI6:%d/%d\n", ep, ep->com.cm_id, ep->com.qp, (int)ep->com.state, ep->com.flags, ep->com.history, ep->hwtid, ep->atid, + ep->stats.connect_neg_adv, + ep->stats.abort_neg_adv, &lsin6->sin6_addr, ntohs(lsin6->sin6_port), ntohs(mapped_lsin6->sin6_port), &rsin6->sin6_addr, ntohs(rsin6->sin6_port), diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index d87e165..97bb555 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -137,6 +137,7 @@ struct c4iw_stats { u64 tcam_full; u64 act_ofld_conn_fails; u64 pas_ofld_conn_fails; + u64 neg_adv; }; struct c4iw_hw_queue { @@ -814,6 +815,11 @@ struct c4iw_listen_ep { int backlog; }; +struct c4iw_ep_stats { + unsigned connect_neg_adv; + unsigned abort_neg_adv; +}; + struct c4iw_ep { struct c4iw_ep_common com; struct c4iw_ep *parent_ep; @@ -846,6 +852,7 @@ struct c4iw_ep { unsigned int retry_count; int snd_win; int rcv_win; + struct c4iw_ep_stats stats; }; static inline void print_addr(struct c4iw_ep_common *epc, const char *func, -- cgit v0.10.2 From 8f71c1a27b84948720be17fffba71a67a1f0942d Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Tue, 5 May 2015 13:01:39 +0200 Subject: IPoIB/CM: Fix indentation level See also patch "IPoIB/cm: Add connected mode support for devices without SRQs" (commit ID 68e995a29572). Detected by smatch. Signed-off-by: Bart Van Assche Cc: Pradeep Satyanarayana Signed-off-by: Doug Ledford diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 56959ad..cf32a778 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -386,8 +386,8 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i rx->rx_ring[i].mapping, GFP_KERNEL)) { ipoib_warn(priv, "failed to allocate receive buffer %d\n", i); - ret = -ENOMEM; - goto err_count; + ret = -ENOMEM; + goto err_count; } ret = ipoib_cm_post_receive_nonsrq(dev, rx, &t->wr, t->sge, i); if (ret) { -- cgit v0.10.2 From 954138dc25dca0263f315c5a3450059df05a4ea1 Mon Sep 17 00:00:00 2001 From: Yann Droneaud Date: Mon, 4 May 2015 14:31:03 +0200 Subject: MAINTAINERS: add include/rdma/ to InfiniBand subsystem Most headers for InfiniBand/RDMA are located under include/rdma/ and include/uapi/rdma. Signed-off-by: Yann Droneaud Reviewed-by: Ira Weiny Signed-off-by: Doug Ledford diff --git a/MAINTAINERS b/MAINTAINERS index 781e099..c80714a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5053,6 +5053,8 @@ S: Supported F: Documentation/infiniband/ F: drivers/infiniband/ F: include/uapi/linux/if_infiniband.h +F: include/uapi/rdma/ +F: include/rdma/ INOTIFY M: John McCutchan -- cgit v0.10.2 From b6b2bbe65b2117afd9766faa7cffea4d8f681455 Mon Sep 17 00:00:00 2001 From: Doug Ledford Date: Tue, 5 May 2015 12:57:09 -0400 Subject: MAINTAINERS: Update InfiniBand subsystem maintainer Since Roland stepped down, the community asked me to take his place, and the nomination was followed by sufficient votes and no dissensions that we can move forward with the change. Signed-off-by: Doug Ledford diff --git a/MAINTAINERS b/MAINTAINERS index c80714a..40c14f1 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5042,13 +5042,13 @@ S: Orphan F: drivers/video/fbdev/imsttfb.c INFINIBAND SUBSYSTEM -M: Roland Dreier +M: Doug Ledford M: Sean Hefty M: Hal Rosenstock L: linux-rdma@vger.kernel.org W: http://www.openfabrics.org/ Q: http://patchwork.kernel.org/project/linux-rdma/list/ -T: git git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband.git +T: git git://github.com/dledford/linux.git S: Supported F: Documentation/infiniband/ F: drivers/infiniband/ -- cgit v0.10.2 From 5cec98834989a014a9560b1841649eaca95cf00e Mon Sep 17 00:00:00 2001 From: Boris Ostrovsky Date: Wed, 29 Apr 2015 17:10:12 -0400 Subject: xen/events: Clear cpu_evtchn_mask before resuming When a guest is resumed, the hypervisor may change event channel assignments. If this happens and the guest uses 2-level events it is possible for the interrupt to be claimed by wrong VCPU since cpu_evtchn_mask bits may be stale. This can happen even though evtchn_2l_bind_to_cpu() attempts to clear old bits: irq_info that is passed in is not necessarily the original one (from pre-migration times) but instead is freshly allocated during resume and so any information about which CPU the channel was bound to is lost. Thus we should clear the mask during resume. We also need to make sure that bits for xenstore and console channels are set when these two subsystems are resumed. While rebind_evtchn_irq() (which is invoked for both of them on a resume) calls irq_set_affinity(), the latter will in fact postpone setting affinity until handling the interrupt. But because cpu_evtchn_mask will have bits for these two cleared we won't be able to take the interrupt. With that in mind, we need to bind those two channels explicitly in rebind_evtchn_irq(). We will keep irq_set_affinity() so that we have a pass through generic irq affinity code later, in case something needs to be updated there as well. (Also replace cpumask_of(0) with cpumask_of(info->cpu) in rebind_evtchn_irq(): it should be set to zero in preceding xen_irq_info_evtchn_setup().) Signed-off-by: Boris Ostrovsky Reported-by: Annie Li Cc: # 3.14+ Signed-off-by: David Vrabel diff --git a/drivers/xen/events/events_2l.c b/drivers/xen/events/events_2l.c index 5db43fc..7dd4631 100644 --- a/drivers/xen/events/events_2l.c +++ b/drivers/xen/events/events_2l.c @@ -345,6 +345,15 @@ irqreturn_t xen_debug_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } +static void evtchn_2l_resume(void) +{ + int i; + + for_each_online_cpu(i) + memset(per_cpu(cpu_evtchn_mask, i), 0, sizeof(xen_ulong_t) * + EVTCHN_2L_NR_CHANNELS/BITS_PER_EVTCHN_WORD); +} + static const struct evtchn_ops evtchn_ops_2l = { .max_channels = evtchn_2l_max_channels, .nr_channels = evtchn_2l_max_channels, @@ -356,6 +365,7 @@ static const struct evtchn_ops evtchn_ops_2l = { .mask = evtchn_2l_mask, .unmask = evtchn_2l_unmask, .handle_events = evtchn_2l_handle_events, + .resume = evtchn_2l_resume, }; void __init xen_evtchn_2l_init(void) diff --git a/drivers/xen/events/events_base.c b/drivers/xen/events/events_base.c index 70fba97..a1ec564 100644 --- a/drivers/xen/events/events_base.c +++ b/drivers/xen/events/events_base.c @@ -1279,8 +1279,9 @@ void rebind_evtchn_irq(int evtchn, int irq) mutex_unlock(&irq_mapping_update_lock); - /* new event channels are always bound to cpu 0 */ - irq_set_affinity(irq, cpumask_of(0)); + bind_evtchn_to_cpu(evtchn, info->cpu); + /* This will be deferred until interrupt is processed */ + irq_set_affinity(irq, cpumask_of(info->cpu)); /* Unmask the event channel. */ enable_irq(irq); -- cgit v0.10.2 From 16f1cf3ba7303228372d3756677bf7d10e79cf9f Mon Sep 17 00:00:00 2001 From: Boris Ostrovsky Date: Wed, 29 Apr 2015 17:10:13 -0400 Subject: xen/xenbus: Update xenbus event channel on resume After a resume the hypervisor/tools may change xenbus event channel number. We should re-query it. Signed-off-by: Boris Ostrovsky Cc: Signed-off-by: David Vrabel diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 564b315..5390a67 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -57,6 +57,7 @@ #include #include #include +#include #include #include @@ -735,6 +736,30 @@ static int __init xenstored_local_init(void) return err; } +static int xenbus_resume_cb(struct notifier_block *nb, + unsigned long action, void *data) +{ + int err = 0; + + if (xen_hvm_domain()) { + uint64_t v; + + err = hvm_get_parameter(HVM_PARAM_STORE_EVTCHN, &v); + if (!err && v) + xen_store_evtchn = v; + else + pr_warn("Cannot update xenstore event channel: %d\n", + err); + } else + xen_store_evtchn = xen_start_info->store_evtchn; + + return err; +} + +static struct notifier_block xenbus_resume_nb = { + .notifier_call = xenbus_resume_cb, +}; + static int __init xenbus_init(void) { int err = 0; @@ -793,6 +818,10 @@ static int __init xenbus_init(void) goto out_error; } + if ((xen_store_domain_type != XS_LOCAL) && + (xen_store_domain_type != XS_UNKNOWN)) + xen_resume_notifier_register(&xenbus_resume_nb); + #ifdef CONFIG_XEN_COMPAT_XENFS /* * Create xenfs mountpoint in /proc for compatibility with -- cgit v0.10.2 From b9d934f27c91b878c4b2e64299d6e419a4022f8d Mon Sep 17 00:00:00 2001 From: Boris Ostrovsky Date: Wed, 29 Apr 2015 17:10:14 -0400 Subject: xen/console: Update console event channel on resume After a resume the hypervisor/tools may change console event channel number. We should re-query it. Signed-off-by: Boris Ostrovsky Cc: Signed-off-by: David Vrabel diff --git a/drivers/tty/hvc/hvc_xen.c b/drivers/tty/hvc/hvc_xen.c index f1e5742..5bab1c6 100644 --- a/drivers/tty/hvc/hvc_xen.c +++ b/drivers/tty/hvc/hvc_xen.c @@ -299,11 +299,27 @@ static int xen_initial_domain_console_init(void) return 0; } +static void xen_console_update_evtchn(struct xencons_info *info) +{ + if (xen_hvm_domain()) { + uint64_t v; + int err; + + err = hvm_get_parameter(HVM_PARAM_CONSOLE_EVTCHN, &v); + if (!err && v) + info->evtchn = v; + } else + info->evtchn = xen_start_info->console.domU.evtchn; +} + void xen_console_resume(void) { struct xencons_info *info = vtermno_to_xencons(HVC_COOKIE); - if (info != NULL && info->irq) + if (info != NULL && info->irq) { + if (!xen_initial_domain()) + xen_console_update_evtchn(info); rebind_evtchn_irq(info->evtchn, info->irq); + } } static void xencons_disconnect_backend(struct xencons_info *info) -- cgit v0.10.2 From 16e6bd5970c88a2ac018b84a5f1dd5c2ff1fdf2c Mon Sep 17 00:00:00 2001 From: Boris Ostrovsky Date: Wed, 29 Apr 2015 17:10:15 -0400 Subject: xen/events: Set irq_info->evtchn before binding the channel to CPU in __startup_pirq() .. because bind_evtchn_to_cpu(evtchn, cpu) will map evtchn to 'info' and pass 'info' down to xen_evtchn_port_bind_to_cpu(). Signed-off-by: Boris Ostrovsky Tested-by: Annie Li Cc: Signed-off-by: David Vrabel diff --git a/drivers/xen/events/events_base.c b/drivers/xen/events/events_base.c index a1ec564..2b8553b 100644 --- a/drivers/xen/events/events_base.c +++ b/drivers/xen/events/events_base.c @@ -529,8 +529,8 @@ static unsigned int __startup_pirq(unsigned int irq) if (rc) goto err; - bind_evtchn_to_cpu(evtchn, 0); info->evtchn = evtchn; + bind_evtchn_to_cpu(evtchn, 0); rc = xen_evtchn_port_setup(info); if (rc) -- cgit v0.10.2 From a71dbdaa8ca2933391b08e0ae5567083e3af0892 Mon Sep 17 00:00:00 2001 From: Boris Ostrovsky Date: Mon, 4 May 2015 11:02:15 -0400 Subject: hypervisor/x86/xen: Unset X86_BUG_SYSRET_SS_ATTRS on Xen PV guests Commit 61f01dd941ba ("x86_64, asm: Work around AMD SYSRET SS descriptor attribute issue") makes AMD processors set SS to __KERNEL_DS in __switch_to() to deal with cases when SS is NULL. This breaks Xen PV guests who do not want to load SS with__KERNEL_DS. Since the problem that the commit is trying to address would have to be fixed in the hypervisor (if it in fact exists under Xen) there is no reason to set X86_BUG_SYSRET_SS_ATTRS flag for PV VPCUs here. This can be easily achieved by adding x86_hyper_xen_hvm.set_cpu_features op which will clear this flag. (And since this structure is no longer HVM-specific we should do some renaming). Signed-off-by: Boris Ostrovsky Reported-by: Sander Eikelenboom Signed-off-by: David Vrabel diff --git a/arch/x86/include/asm/hypervisor.h b/arch/x86/include/asm/hypervisor.h index e42f758..055ea99 100644 --- a/arch/x86/include/asm/hypervisor.h +++ b/arch/x86/include/asm/hypervisor.h @@ -50,7 +50,7 @@ extern const struct hypervisor_x86 *x86_hyper; /* Recognized hypervisors */ extern const struct hypervisor_x86 x86_hyper_vmware; extern const struct hypervisor_x86 x86_hyper_ms_hyperv; -extern const struct hypervisor_x86 x86_hyper_xen_hvm; +extern const struct hypervisor_x86 x86_hyper_xen; extern const struct hypervisor_x86 x86_hyper_kvm; extern void init_hypervisor(struct cpuinfo_x86 *c); diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c index 36ce402..d820d8e 100644 --- a/arch/x86/kernel/cpu/hypervisor.c +++ b/arch/x86/kernel/cpu/hypervisor.c @@ -27,8 +27,8 @@ static const __initconst struct hypervisor_x86 * const hypervisors[] = { -#ifdef CONFIG_XEN_PVHVM - &x86_hyper_xen_hvm, +#ifdef CONFIG_XEN + &x86_hyper_xen, #endif &x86_hyper_vmware, &x86_hyper_ms_hyperv, diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 94578ef..46957ea 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1760,6 +1760,9 @@ static struct notifier_block xen_hvm_cpu_notifier = { static void __init xen_hvm_guest_init(void) { + if (xen_pv_domain()) + return; + init_hvm_pv_info(); xen_hvm_init_shared_info(); @@ -1775,6 +1778,7 @@ static void __init xen_hvm_guest_init(void) xen_hvm_init_time_ops(); xen_hvm_init_mmu_ops(); } +#endif static bool xen_nopv = false; static __init int xen_parse_nopv(char *arg) @@ -1784,14 +1788,11 @@ static __init int xen_parse_nopv(char *arg) } early_param("xen_nopv", xen_parse_nopv); -static uint32_t __init xen_hvm_platform(void) +static uint32_t __init xen_platform(void) { if (xen_nopv) return 0; - if (xen_pv_domain()) - return 0; - return xen_cpuid_base(); } @@ -1809,11 +1810,19 @@ bool xen_hvm_need_lapic(void) } EXPORT_SYMBOL_GPL(xen_hvm_need_lapic); -const struct hypervisor_x86 x86_hyper_xen_hvm __refconst = { - .name = "Xen HVM", - .detect = xen_hvm_platform, +static void xen_set_cpu_features(struct cpuinfo_x86 *c) +{ + if (xen_pv_domain()) + clear_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS); +} + +const struct hypervisor_x86 x86_hyper_xen = { + .name = "Xen", + .detect = xen_platform, +#ifdef CONFIG_XEN_PVHVM .init_platform = xen_hvm_guest_init, +#endif .x2apic_available = xen_x2apic_para_available, + .set_cpu_features = xen_set_cpu_features, }; -EXPORT_SYMBOL(x86_hyper_xen_hvm); -#endif +EXPORT_SYMBOL(x86_hyper_xen); -- cgit v0.10.2 From de71ad2c97862eae1516aa36528cc3b317c17b2f Mon Sep 17 00:00:00 2001 From: Marc Dionne Date: Mon, 4 May 2015 15:16:44 -0300 Subject: x86: Make cpu_tss available to external modules Commit 75182b1632 ("x86/asm/entry: Switch all C consumers of kernel_stack to this_cpu_sp0()") changed current_thread_info to use this_cpu_sp0, and indirectly made it rely on init_tss which was exported with EXPORT_PER_CPU_SYMBOL_GPL. As a result some macros and inline functions such as set/get_fs, test_thread_flag and variants have been made unusable for external modules. Make cpu_tss exported with EXPORT_PER_CPU_SYMBOL so that these functions are accessible again, as they were previously. Signed-off-by: Marc Dionne Acked-by: Andy Lutomirski Link: http://lkml.kernel.org/r/1430763404-21221-1-git-send-email-marc.dionne@your-file-system.com Signed-off-by: Thomas Gleixner diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 8213da6..bfc99b3 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -57,7 +57,7 @@ __visible DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, cpu_tss) = { .io_bitmap = { [0 ... IO_BITMAP_LONGS] = ~0 }, #endif }; -EXPORT_PER_CPU_SYMBOL_GPL(cpu_tss); +EXPORT_PER_CPU_SYMBOL(cpu_tss); #ifdef CONFIG_X86_64 static DEFINE_PER_CPU(unsigned char, is_idle); -- cgit v0.10.2 From d467f7a405cf0e7f06ed8d3175607ebb4ed06671 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Thu, 26 Mar 2015 13:35:18 -0500 Subject: ipmi_ssif: Fix the logic on user-supplied addresses Returning zero is success. Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_ssif.c b/drivers/char/ipmi/ipmi_ssif.c index f40e3bd..1de1914 100644 --- a/drivers/char/ipmi/ipmi_ssif.c +++ b/drivers/char/ipmi/ipmi_ssif.c @@ -1832,7 +1832,7 @@ static int init_ipmi_ssif(void) rv = new_ssif_client(addr[i], adapter_name[i], dbg[i], slave_addrs[i], SI_HARDCODED); - if (!rv) + if (rv) pr_err(PFX "Couldn't add hardcoded device at addr 0x%x\n", addr[i]); -- cgit v0.10.2 From b0e9aaa99dfb3036829e91d4f0aae449639e221a Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Tue, 31 Mar 2015 12:48:53 -0500 Subject: ipmi:ssif: Ignore spaces when comparing I2C adapter names Some of the adapters have spaces in their names, but that's really hard to pass in as a module or kernel parameters. So ignore the spaces. Signed-off-by: Corey Minyard diff --git a/Documentation/IPMI.txt b/Documentation/IPMI.txt index 653d5d7..31d1d65 100644 --- a/Documentation/IPMI.txt +++ b/Documentation/IPMI.txt @@ -505,7 +505,10 @@ at module load time (for a module) with: The addresses are normal I2C addresses. The adapter is the string name of the adapter, as shown in /sys/class/i2c-adapter/i2c-/name. -It is *NOT* i2c- itself. +It is *NOT* i2c- itself. Also, the comparison is done ignoring +spaces, so if the name is "This is an I2C chip" you can say +adapter_name=ThisisanI2cchip. This is because it's hard to pass in +spaces in kernel parameters. The debug flags are bit flags for each BMC found, they are: IPMI messages: 1, driver state: 2, timing: 4, I2C probe: 8 diff --git a/drivers/char/ipmi/ipmi_ssif.c b/drivers/char/ipmi/ipmi_ssif.c index 1de1914..3c3b725 100644 --- a/drivers/char/ipmi/ipmi_ssif.c +++ b/drivers/char/ipmi/ipmi_ssif.c @@ -1258,6 +1258,23 @@ static const struct file_operations smi_stats_proc_ops = { .release = single_release, }; +static int strcmp_nospace(char *s1, char *s2) +{ + while (*s1 && *s2) { + while (isspace(*s1)) + s1++; + while (isspace(*s2)) + s2++; + if (*s1 > *s2) + return 1; + if (*s1 < *s2) + return -1; + s1++; + s2++; + } + return 0; +} + static struct ssif_addr_info *ssif_info_find(unsigned short addr, char *adapter_name, bool match_null_name) @@ -1272,8 +1289,10 @@ restart: /* One is NULL and one is not */ continue; } - if (strcmp(info->adapter_name, adapter_name)) - /* Names to not match */ + if (adapter_name && + strcmp_nospace(info->adapter_name, + adapter_name)) + /* Names do not match */ continue; } found = info; @@ -1407,7 +1426,7 @@ static int ssif_probe(struct i2c_client *client, const struct i2c_device_id *id) } else { no_support: /* Assume no multi-part or PEC support */ - pr_info(PFX "Error fetching SSIF: %d %d %2.2x, your system probably doesn't support this command so using defaults\n", + pr_info(PFX "Error fetching SSIF: %d %d %2.2x, your system probably doesn't support this command so using defaults\n", rv, len, resp[2]); ssif_info->max_xmit_msg_size = 32; -- cgit v0.10.2 From 5e33cd0c5a299772b5ec1a493f0a77548664ae06 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 22 Feb 2015 10:21:07 -0800 Subject: ipmi: Remove incorrect use of seq_has_overflowed commit d6c5dc18d863 ("ipmi: Remove uses of return value of seq_printf") incorrectly changed the return value of various proc_show functions to use seq_has_overflowed(). These functions should return 0 on completion rather than 1/true on overflow. 1 is the same as #define SEQ_SKIP which would cause the output to not be emitted (skipped) instead. This is a logical defect only as the length of these outputs are all smaller than the initial allocation done by the seq filesystem. Signed-off-by: Joe Perches Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 9bb5928..bf75f63 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -2000,7 +2000,7 @@ static int smi_ipmb_proc_show(struct seq_file *m, void *v) seq_printf(m, " %x", intf->channels[i].address); seq_putc(m, '\n'); - return seq_has_overflowed(m); + return 0; } static int smi_ipmb_proc_open(struct inode *inode, struct file *file) @@ -2023,7 +2023,7 @@ static int smi_version_proc_show(struct seq_file *m, void *v) ipmi_version_major(&intf->bmc->id), ipmi_version_minor(&intf->bmc->id)); - return seq_has_overflowed(m); + return 0; } static int smi_version_proc_open(struct inode *inode, struct file *file) diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 5e90a18..468c75e 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -3080,7 +3080,7 @@ static int smi_type_proc_show(struct seq_file *m, void *v) seq_printf(m, "%s\n", si_to_str[smi->si_type]); - return seq_has_overflowed(m); + return 0; } static int smi_type_proc_open(struct inode *inode, struct file *file) @@ -3153,7 +3153,7 @@ static int smi_params_proc_show(struct seq_file *m, void *v) smi->irq, smi->slave_addr); - return seq_has_overflowed(m); + return 0; } static int smi_params_proc_open(struct inode *inode, struct file *file) diff --git a/drivers/char/ipmi/ipmi_ssif.c b/drivers/char/ipmi/ipmi_ssif.c index 3c3b725..ee3b8c5 100644 --- a/drivers/char/ipmi/ipmi_ssif.c +++ b/drivers/char/ipmi/ipmi_ssif.c @@ -1200,7 +1200,7 @@ static int smi_type_proc_show(struct seq_file *m, void *v) { seq_puts(m, "ssif\n"); - return seq_has_overflowed(m); + return 0; } static int smi_type_proc_open(struct inode *inode, struct file *file) -- cgit v0.10.2 From b1e65e71535aa128089d4cb1b6d90db7551fcb05 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Fri, 10 Apr 2015 20:19:18 -0500 Subject: ipmi: Don't report err in the SI driver for SSIF devices Really ignore them by returning -ENODEV from the probe, but not doing anything. Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 468c75e..4612741 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2244,7 +2244,7 @@ static int ipmi_pnp_probe(struct pnp_dev *dev, acpi_handle handle; acpi_status status; unsigned long long tmp; - int rv; + int rv = -EINVAL; acpi_dev = pnp_acpi_device(dev); if (!acpi_dev) @@ -2276,6 +2276,7 @@ static int ipmi_pnp_probe(struct pnp_dev *dev, info->si_type = SI_BT; break; case 4: /* SSIF, just ignore */ + rv = -ENODEV; goto err_free; default: dev_info(&dev->dev, "unknown IPMI type %lld\n", tmp); @@ -2336,7 +2337,7 @@ static int ipmi_pnp_probe(struct pnp_dev *dev, err_free: kfree(info); - return -EINVAL; + return rv; } static void ipmi_pnp_remove(struct pnp_dev *dev) -- cgit v0.10.2 From df6dd1b35b0ec0ac6a5298378ceaf487091f448c Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Mon, 27 Apr 2015 09:45:06 +0200 Subject: thinkpad_acpi: Fix warning for static not at beginning Fix the following warning: warning: "static" is not at beginning of declaration void static hotkey_mask_warn_incomplete_mask(void) ^ Signed-off-by: Jean Delvare Cc: Henrique de Moraes Holschuh Cc: Darren Hart Signed-off-by: Darren Hart diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index 7769575..9bb9ad6 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -2115,7 +2115,7 @@ static int hotkey_mask_get(void) return 0; } -void static hotkey_mask_warn_incomplete_mask(void) +static void hotkey_mask_warn_incomplete_mask(void) { /* log only what the user can fix... */ const u32 wantedmask = hotkey_driver_mask & -- cgit v0.10.2 From 99ebbd30e3640f6addb37f222b4d6ad4b807d9ea Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 5 May 2015 16:23:25 -0700 Subject: revert "zram: move compact_store() to sysfs functions area" Revert commit c72c6160d967ed26a0b136dbab337f821d233509 It was intended to be a cosmetic change that w/o any functional change and was part of a bigger change: http://lkml.iu.edu/hypermail/linux/kernel/1503.1/01818.html Sergey Senozhatsky Cc: Linus Torvalds Cc: Minchan Kim Cc: Nitin Gupta Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index c94386a..8dcbced 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -74,6 +74,27 @@ static inline struct zram *dev_to_zram(struct device *dev) return (struct zram *)dev_to_disk(dev)->private_data; } +static ssize_t compact_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) +{ + unsigned long nr_migrated; + struct zram *zram = dev_to_zram(dev); + struct zram_meta *meta; + + down_read(&zram->init_lock); + if (!init_done(zram)) { + up_read(&zram->init_lock); + return -EINVAL; + } + + meta = zram->meta; + nr_migrated = zs_compact(meta->mem_pool); + atomic64_add(nr_migrated, &zram->stats.num_migrated); + up_read(&zram->init_lock); + + return len; +} + static ssize_t disksize_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -1038,6 +1059,7 @@ static const struct block_device_operations zram_devops = { .owner = THIS_MODULE }; +static DEVICE_ATTR_WO(compact); static DEVICE_ATTR_RW(disksize); static DEVICE_ATTR_RO(initstate); static DEVICE_ATTR_WO(reset); @@ -1114,6 +1136,7 @@ static struct attribute *zram_disk_attrs[] = { &dev_attr_num_writes.attr, &dev_attr_failed_reads.attr, &dev_attr_failed_writes.attr, + &dev_attr_compact.attr, &dev_attr_invalid_io.attr, &dev_attr_notify_free.attr, &dev_attr_zero_pages.attr, -- cgit v0.10.2 From 74f3037c4015f3a440dc4cb4e31477875fa9791c Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Tue, 5 May 2015 16:23:28 -0700 Subject: zram: add Designated Reviewer for zram in MAINTAINERS Sergey Senozhatsky has contributed/reviewed to zram for a long time. He is really helpful for maintaining zram so I want for him to continue helping me as Designated Reviewer unless he hates it. Signed-off-by: Minchan Kim Cc: Sergey Senozhatsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 781e099..1c7fd3c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -11037,6 +11037,7 @@ F: drivers/media/pci/zoran/ ZRAM COMPRESSED RAM BLOCK DEVICE DRVIER M: Minchan Kim M: Nitin Gupta +R: Sergey Senozhatsky L: linux-kernel@vger.kernel.org S: Maintained F: drivers/block/zram/ -- cgit v0.10.2 From 48b945a19cf6e7e548b2ce545ec88f93284ab276 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 5 May 2015 16:23:30 -0700 Subject: MAINTAINERS: add co-maintainer for LED subsystem Add myself (Jacek Anaszewski) as a co-maintainer for the LED subsystem. Signed-off-by: Jacek Anaszewski Acked-by: Bryan Wu Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 1c7fd3c..b399b34 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5805,6 +5805,7 @@ F: drivers/scsi/53c700* LED SUBSYSTEM M: Bryan Wu M: Richard Purdie +M: Jacek Anaszewski L: linux-leds@vger.kernel.org T: git git://git.kernel.org/pub/scm/linux/kernel/git/cooloney/linux-leds.git S: Maintained -- cgit v0.10.2 From 7d616e4ddb9c0754ed6245a43332d5b867e4db11 Mon Sep 17 00:00:00 2001 From: Yury Norov Date: Tue, 5 May 2015 16:23:33 -0700 Subject: lib: delete lib/find_last_bit.c The file lib/find_last_bit.c was no longer used and supposed to be deleted by commit 8f6f19dd51 ("lib: move find_last_bit to lib/find_next_bit.c") but that delete didn't happen. This gets rid of it. Signed-off-by: Yury Norov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/find_last_bit.c b/lib/find_last_bit.c deleted file mode 100644 index 3e3be40..0000000 --- a/lib/find_last_bit.c +++ /dev/null @@ -1,41 +0,0 @@ -/* find_last_bit.c: fallback find next bit implementation - * - * Copyright (C) 2008 IBM Corporation - * Written by Rusty Russell - * (Inspired by David Howell's find_next_bit implementation) - * - * Rewritten by Yury Norov to decrease - * size and improve performance, 2015. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include - -#ifndef find_last_bit - -unsigned long find_last_bit(const unsigned long *addr, unsigned long size) -{ - if (size) { - unsigned long val = BITMAP_LAST_WORD_MASK(size); - unsigned long idx = (size-1) / BITS_PER_LONG; - - do { - val &= addr[idx]; - if (val) - return idx * BITS_PER_LONG + __fls(val); - - val = ~0ul; - } while (idx--); - } - return size; -} -EXPORT_SYMBOL(find_last_bit); - -#endif -- cgit v0.10.2 From 09789e5de18e4e442870b2d700831f5cb802eb05 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Tue, 5 May 2015 16:23:35 -0700 Subject: mm/memory-failure: call shake_page() when error hits thp tail page Currently memory_failure() calls shake_page() to sweep pages out from pcplists only when the victim page is 4kB LRU page or thp head page. But we should do this for a thp tail page too. Consider that a memory error hits a thp tail page whose head page is on a pcplist when memory_failure() runs. Then, the current kernel skips shake_pages() part, so hwpoison_user_mappings() returns without calling split_huge_page() nor try_to_unmap() because PageLRU of the thp head is still cleared due to the skip of shake_page(). As a result, me_huge_page() runs for the thp, which is broken behavior. One effect is a leak of the thp. And another is to fail to isolate the memory error, so later access to the error address causes another MCE, which kills the processes which used the thp. This patch fixes this problem by calling shake_page() for thp tail case. Fixes: 385de35722c9 ("thp: allow a hwpoisoned head page to be put back to LRU") Signed-off-by: Naoya Horiguchi Reviewed-by: Andi Kleen Acked-by: Dean Nelson Cc: Andrea Arcangeli Cc: Hidetoshi Seto Cc: Jin Dongming Cc: [3.4+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memory-failure.c b/mm/memory-failure.c index d9359b7..22e0f27 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1187,10 +1187,10 @@ int memory_failure(unsigned long pfn, int trapno, int flags) * The check (unnecessarily) ignores LRU pages being isolated and * walked by the page reclaim code, however that's not a big loss. */ - if (!PageHuge(p) && !PageTransTail(p)) { - if (!PageLRU(p)) - shake_page(p, 0); - if (!PageLRU(p)) { + if (!PageHuge(p)) { + if (!PageLRU(hpage)) + shake_page(hpage, 0); + if (!PageLRU(hpage)) { /* * shake_page could have turned it free. */ -- cgit v0.10.2 From 01e76903f655a4d88c2e09d3182436c65f6e1213 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 5 May 2015 16:23:38 -0700 Subject: kasan: show gcc version requirements in Kconfig and Documentation The documentation shows a need for gcc > 4.9.2, but it's really >=. The Kconfig entries don't show require versions so add them. Correct a latter/later typo too. Also mention that gcc 5 required to catch out of bounds accesses to global and stack variables. Signed-off-by: Joe Perches Signed-off-by: Andrey Ryabinin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/kasan.txt b/Documentation/kasan.txt index 092fc10..4692241 100644 --- a/Documentation/kasan.txt +++ b/Documentation/kasan.txt @@ -9,7 +9,9 @@ a fast and comprehensive solution for finding use-after-free and out-of-bounds bugs. KASan uses compile-time instrumentation for checking every memory access, -therefore you will need a certain version of GCC > 4.9.2 +therefore you will need a gcc version of 4.9.2 or later. KASan could detect out +of bounds accesses to stack or global variables, but only if gcc 5.0 or later was +used to built the kernel. Currently KASan is supported only for x86_64 architecture and requires that the kernel be built with the SLUB allocator. @@ -23,8 +25,8 @@ To enable KASAN configure kernel with: and choose between CONFIG_KASAN_OUTLINE and CONFIG_KASAN_INLINE. Outline/inline is compiler instrumentation types. The former produces smaller binary the -latter is 1.1 - 2 times faster. Inline instrumentation requires GCC 5.0 or -latter. +latter is 1.1 - 2 times faster. Inline instrumentation requires a gcc version +of 5.0 or later. Currently KASAN works only with the SLUB memory allocator. For better bug detection and nicer report, enable CONFIG_STACKTRACE and put diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan index 4fecaedc..777eda7 100644 --- a/lib/Kconfig.kasan +++ b/lib/Kconfig.kasan @@ -10,8 +10,11 @@ config KASAN help Enables kernel address sanitizer - runtime memory debugger, designed to find out-of-bounds accesses and use-after-free bugs. - This is strictly debugging feature. It consumes about 1/8 - of available memory and brings about ~x3 performance slowdown. + This is strictly a debugging feature and it requires a gcc version + of 4.9.2 or later. Detection of out of bounds accesses to stack or + global variables requires gcc 5.0 or later. + This feature consumes about 1/8 of available memory and brings about + ~x3 performance slowdown. For better error detection enable CONFIG_STACKTRACE, and add slub_debug=U to boot cmdline. @@ -40,6 +43,7 @@ config KASAN_INLINE memory accesses. This is faster than outline (in some workloads it gives about x2 boost over outline instrumentation), but make kernel's .text size much bigger. + This requires a gcc version of 5.0 or later. endchoice -- cgit v0.10.2 From c71f1e05e62f3eb843b6458eeb7298f269b34b1e Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Tue, 5 May 2015 16:23:41 -0700 Subject: Documentation: bindings: add abracon,abx80x Document the bindings for abracon,abx80x and related compatibles. Signed-off-by: Alexandre Belloni Cc: Philippe De Muyter Cc: Alessandro Zummo Cc: Arnd Bergmann Cc: Paul Bolle Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/devicetree/bindings/rtc/abracon,abx80x.txt b/Documentation/devicetree/bindings/rtc/abracon,abx80x.txt new file mode 100644 index 0000000..be78968 --- /dev/null +++ b/Documentation/devicetree/bindings/rtc/abracon,abx80x.txt @@ -0,0 +1,30 @@ +Abracon ABX80X I2C ultra low power RTC/Alarm chip + +The Abracon ABX80X family consist of the ab0801, ab0803, ab0804, ab0805, ab1801, +ab1803, ab1804 and ab1805. The ab0805 is the superset of ab080x and the ab1805 +is the superset of ab180x. + +Required properties: + + - "compatible": should one of: + "abracon,abx80x" + "abracon,ab0801" + "abracon,ab0803" + "abracon,ab0804" + "abracon,ab0805" + "abracon,ab1801" + "abracon,ab1803" + "abracon,ab1804" + "abracon,ab1805" + Using "abracon,abx80x" will enable chip autodetection. + - "reg": I2C bus address of the device + +Optional properties: + +The abx804 and abx805 have a trickle charger that is able to charge the +connected battery or supercap. Both the following properties have to be defined +and valid to enable charging: + + - "abracon,tc-diode": should be "standard" (0.6V) or "schottky" (0.3V) + - "abracon,tc-resistor": should be <0>, <3>, <6> or <11>. 0 disables the output + resistor, the other values are in ohm. -- cgit v0.10.2 From 4d61ff6b9960cb00cf2c12abd5769aa2dd475415 Mon Sep 17 00:00:00 2001 From: Philippe De Muyter Date: Tue, 5 May 2015 16:23:44 -0700 Subject: rtc: add rtc-abx80x, a driver for the Abracon AB x80x i2c rtc This is a basic driver for the ultra-low-power Abracon AB x80x series of RTC chips. It supports in particular, the supersets AB0805 and AB1805. It allows reading and writing the time, and enables the supercapacitor/ battery charger. [arnd@arndb.de: abx805 depends on i2c] [alexandre.belloni@free-electrons.com: renam buffer from date to buf in abx80x_rtc_read_time()] Signed-off-by: Philippe De Muyter Cc: Alessandro Zummo Signed-off-by: Alexandre Belloni Signed-off-by: Arnd Bergmann Cc: Paul Bolle Cc: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index 6149ae0..0fe4ad8 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -164,6 +164,16 @@ config RTC_DRV_ABB5ZES3 This driver can also be built as a module. If so, the module will be called rtc-ab-b5ze-s3. +config RTC_DRV_ABX80X + tristate "Abracon ABx80x" + help + If you say yes here you get support for Abracon AB080X and AB180X + families of ultra-low-power battery- and capacitor-backed real-time + clock chips. + + This driver can also be built as a module. If so, the module + will be called rtc-abx80x. + config RTC_DRV_AS3722 tristate "ams AS3722 RTC driver" depends on MFD_AS3722 diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile index c31731c..2b82e2b 100644 --- a/drivers/rtc/Makefile +++ b/drivers/rtc/Makefile @@ -25,6 +25,7 @@ obj-$(CONFIG_RTC_DRV_88PM80X) += rtc-88pm80x.o obj-$(CONFIG_RTC_DRV_AB3100) += rtc-ab3100.o obj-$(CONFIG_RTC_DRV_AB8500) += rtc-ab8500.o obj-$(CONFIG_RTC_DRV_ABB5ZES3) += rtc-ab-b5ze-s3.o +obj-$(CONFIG_RTC_DRV_ABX80X) += rtc-abx80x.o obj-$(CONFIG_RTC_DRV_ARMADA38X) += rtc-armada38x.o obj-$(CONFIG_RTC_DRV_AS3722) += rtc-as3722.o obj-$(CONFIG_RTC_DRV_AT32AP700X)+= rtc-at32ap700x.o diff --git a/drivers/rtc/rtc-abx80x.c b/drivers/rtc/rtc-abx80x.c new file mode 100644 index 0000000..4337c3b --- /dev/null +++ b/drivers/rtc/rtc-abx80x.c @@ -0,0 +1,307 @@ +/* + * A driver for the I2C members of the Abracon AB x8xx RTC family, + * and compatible: AB 1805 and AB 0805 + * + * Copyright 2014-2015 Macq S.A. + * + * Author: Philippe De Muyter + * Author: Alexandre Belloni + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include +#include +#include +#include + +#define ABX8XX_REG_HTH 0x00 +#define ABX8XX_REG_SC 0x01 +#define ABX8XX_REG_MN 0x02 +#define ABX8XX_REG_HR 0x03 +#define ABX8XX_REG_DA 0x04 +#define ABX8XX_REG_MO 0x05 +#define ABX8XX_REG_YR 0x06 +#define ABX8XX_REG_WD 0x07 + +#define ABX8XX_REG_CTRL1 0x10 +#define ABX8XX_CTRL_WRITE BIT(1) +#define ABX8XX_CTRL_12_24 BIT(6) + +#define ABX8XX_REG_CFG_KEY 0x1f +#define ABX8XX_CFG_KEY_MISC 0x9d + +#define ABX8XX_REG_ID0 0x28 + +#define ABX8XX_REG_TRICKLE 0x20 +#define ABX8XX_TRICKLE_CHARGE_ENABLE 0xa0 +#define ABX8XX_TRICKLE_STANDARD_DIODE 0x8 +#define ABX8XX_TRICKLE_SCHOTTKY_DIODE 0x4 + +static u8 trickle_resistors[] = {0, 3, 6, 11}; + +enum abx80x_chip {AB0801, AB0803, AB0804, AB0805, + AB1801, AB1803, AB1804, AB1805, ABX80X}; + +struct abx80x_cap { + u16 pn; + bool has_tc; +}; + +static struct abx80x_cap abx80x_caps[] = { + [AB0801] = {.pn = 0x0801}, + [AB0803] = {.pn = 0x0803}, + [AB0804] = {.pn = 0x0804, .has_tc = true}, + [AB0805] = {.pn = 0x0805, .has_tc = true}, + [AB1801] = {.pn = 0x1801}, + [AB1803] = {.pn = 0x1803}, + [AB1804] = {.pn = 0x1804, .has_tc = true}, + [AB1805] = {.pn = 0x1805, .has_tc = true}, + [ABX80X] = {.pn = 0} +}; + +static struct i2c_driver abx80x_driver; + +static int abx80x_enable_trickle_charger(struct i2c_client *client, + u8 trickle_cfg) +{ + int err; + + /* + * Write the configuration key register to enable access to the Trickle + * register + */ + err = i2c_smbus_write_byte_data(client, ABX8XX_REG_CFG_KEY, + ABX8XX_CFG_KEY_MISC); + if (err < 0) { + dev_err(&client->dev, "Unable to write configuration key\n"); + return -EIO; + } + + err = i2c_smbus_write_byte_data(client, ABX8XX_REG_TRICKLE, + ABX8XX_TRICKLE_CHARGE_ENABLE | + trickle_cfg); + if (err < 0) { + dev_err(&client->dev, "Unable to write trickle register\n"); + return -EIO; + } + + return 0; +} + +static int abx80x_rtc_read_time(struct device *dev, struct rtc_time *tm) +{ + struct i2c_client *client = to_i2c_client(dev); + unsigned char buf[8]; + int err; + + err = i2c_smbus_read_i2c_block_data(client, ABX8XX_REG_HTH, + sizeof(buf), buf); + if (err < 0) { + dev_err(&client->dev, "Unable to read date\n"); + return -EIO; + } + + tm->tm_sec = bcd2bin(buf[ABX8XX_REG_SC] & 0x7F); + tm->tm_min = bcd2bin(buf[ABX8XX_REG_MN] & 0x7F); + tm->tm_hour = bcd2bin(buf[ABX8XX_REG_HR] & 0x3F); + tm->tm_wday = buf[ABX8XX_REG_WD] & 0x7; + tm->tm_mday = bcd2bin(buf[ABX8XX_REG_DA] & 0x3F); + tm->tm_mon = bcd2bin(buf[ABX8XX_REG_MO] & 0x1F) - 1; + tm->tm_year = bcd2bin(buf[ABX8XX_REG_YR]) + 100; + + err = rtc_valid_tm(tm); + if (err < 0) + dev_err(&client->dev, "retrieved date/time is not valid.\n"); + + return err; +} + +static int abx80x_rtc_set_time(struct device *dev, struct rtc_time *tm) +{ + struct i2c_client *client = to_i2c_client(dev); + unsigned char buf[8]; + int err; + + if (tm->tm_year < 100) + return -EINVAL; + + buf[ABX8XX_REG_HTH] = 0; + buf[ABX8XX_REG_SC] = bin2bcd(tm->tm_sec); + buf[ABX8XX_REG_MN] = bin2bcd(tm->tm_min); + buf[ABX8XX_REG_HR] = bin2bcd(tm->tm_hour); + buf[ABX8XX_REG_DA] = bin2bcd(tm->tm_mday); + buf[ABX8XX_REG_MO] = bin2bcd(tm->tm_mon + 1); + buf[ABX8XX_REG_YR] = bin2bcd(tm->tm_year - 100); + buf[ABX8XX_REG_WD] = tm->tm_wday; + + err = i2c_smbus_write_i2c_block_data(client, ABX8XX_REG_HTH, + sizeof(buf), buf); + if (err < 0) { + dev_err(&client->dev, "Unable to write to date registers\n"); + return -EIO; + } + + return 0; +} + +static const struct rtc_class_ops abx80x_rtc_ops = { + .read_time = abx80x_rtc_read_time, + .set_time = abx80x_rtc_set_time, +}; + +static int abx80x_dt_trickle_cfg(struct device_node *np) +{ + const char *diode; + int trickle_cfg = 0; + int i, ret; + u32 tmp; + + ret = of_property_read_string(np, "abracon,tc-diode", &diode); + if (ret) + return ret; + + if (!strcmp(diode, "standard")) + trickle_cfg |= ABX8XX_TRICKLE_STANDARD_DIODE; + else if (!strcmp(diode, "schottky")) + trickle_cfg |= ABX8XX_TRICKLE_SCHOTTKY_DIODE; + else + return -EINVAL; + + ret = of_property_read_u32(np, "abracon,tc-resistor", &tmp); + if (ret) + return ret; + + for (i = 0; i < sizeof(trickle_resistors); i++) + if (trickle_resistors[i] == tmp) + break; + + if (i == sizeof(trickle_resistors)) + return -EINVAL; + + return (trickle_cfg | i); +} + +static int abx80x_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct device_node *np = client->dev.of_node; + struct rtc_device *rtc; + int i, data, err, trickle_cfg = -EINVAL; + char buf[7]; + unsigned int part = id->driver_data; + unsigned int partnumber; + unsigned int majrev, minrev; + unsigned int lot; + unsigned int wafer; + unsigned int uid; + + if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) + return -ENODEV; + + err = i2c_smbus_read_i2c_block_data(client, ABX8XX_REG_ID0, + sizeof(buf), buf); + if (err < 0) { + dev_err(&client->dev, "Unable to read partnumber\n"); + return -EIO; + } + + partnumber = (buf[0] << 8) | buf[1]; + majrev = buf[2] >> 3; + minrev = buf[2] & 0x7; + lot = ((buf[4] & 0x80) << 2) | ((buf[6] & 0x80) << 1) | buf[3]; + uid = ((buf[4] & 0x7f) << 8) | buf[5]; + wafer = (buf[6] & 0x7c) >> 2; + dev_info(&client->dev, "model %04x, revision %u.%u, lot %x, wafer %x, uid %x\n", + partnumber, majrev, minrev, lot, wafer, uid); + + data = i2c_smbus_read_byte_data(client, ABX8XX_REG_CTRL1); + if (data < 0) { + dev_err(&client->dev, "Unable to read control register\n"); + return -EIO; + } + + err = i2c_smbus_write_byte_data(client, ABX8XX_REG_CTRL1, + ((data & ~ABX8XX_CTRL_12_24) | + ABX8XX_CTRL_WRITE)); + if (err < 0) { + dev_err(&client->dev, "Unable to write control register\n"); + return -EIO; + } + + /* part autodetection */ + if (part == ABX80X) { + for (i = 0; abx80x_caps[i].pn; i++) + if (partnumber == abx80x_caps[i].pn) + break; + if (abx80x_caps[i].pn == 0) { + dev_err(&client->dev, "Unknown part: %04x\n", + partnumber); + return -EINVAL; + } + part = i; + } + + if (partnumber != abx80x_caps[part].pn) { + dev_err(&client->dev, "partnumber mismatch %04x != %04x\n", + partnumber, abx80x_caps[part].pn); + return -EINVAL; + } + + if (np && abx80x_caps[part].has_tc) + trickle_cfg = abx80x_dt_trickle_cfg(np); + + if (trickle_cfg > 0) { + dev_info(&client->dev, "Enabling trickle charger: %02x\n", + trickle_cfg); + abx80x_enable_trickle_charger(client, trickle_cfg); + } + + rtc = devm_rtc_device_register(&client->dev, abx80x_driver.driver.name, + &abx80x_rtc_ops, THIS_MODULE); + + if (IS_ERR(rtc)) + return PTR_ERR(rtc); + + i2c_set_clientdata(client, rtc); + + return 0; +} + +static int abx80x_remove(struct i2c_client *client) +{ + return 0; +} + +static const struct i2c_device_id abx80x_id[] = { + { "abx80x", ABX80X }, + { "ab0801", AB0801 }, + { "ab0803", AB0803 }, + { "ab0804", AB0804 }, + { "ab0805", AB0805 }, + { "ab1801", AB1801 }, + { "ab1803", AB1803 }, + { "ab1804", AB1804 }, + { "ab1805", AB1805 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, abx80x_id); + +static struct i2c_driver abx80x_driver = { + .driver = { + .name = "rtc-abx80x", + }, + .probe = abx80x_probe, + .remove = abx80x_remove, + .id_table = abx80x_id, +}; + +module_i2c_driver(abx80x_driver); + +MODULE_AUTHOR("Philippe De Muyter "); +MODULE_AUTHOR("Alexandre Belloni "); +MODULE_DESCRIPTION("Abracon ABX80X RTC driver"); +MODULE_LICENSE("GPL v2"); -- cgit v0.10.2 From 602498f9aa43d4951eece3fd6ad95a6d0a78d537 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Tue, 5 May 2015 16:23:46 -0700 Subject: mm: soft-offline: fix num_poisoned_pages counting on concurrent events If multiple soft offline events hit one free page/hugepage concurrently, soft_offline_page() can handle the free page/hugepage multiple times, which makes num_poisoned_pages counter increased more than once. This patch fixes this wrong counting by checking TestSetPageHWPoison for normal papes and by checking the return value of dequeue_hwpoisoned_huge_page() for hugepages. Signed-off-by: Naoya Horiguchi Acked-by: Dean Nelson Cc: Andi Kleen Cc: [3.14+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 22e0f27..501820c 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1777,12 +1777,12 @@ int soft_offline_page(struct page *page, int flags) } else if (ret == 0) { /* for free pages */ if (PageHuge(page)) { set_page_hwpoison_huge_page(hpage); - dequeue_hwpoisoned_huge_page(hpage); - atomic_long_add(1 << compound_order(hpage), + if (!dequeue_hwpoisoned_huge_page(hpage)) + atomic_long_add(1 << compound_order(hpage), &num_poisoned_pages); } else { - SetPageHWPoison(page); - atomic_long_inc(&num_poisoned_pages); + if (!TestSetPageHWPoison(page)) + atomic_long_inc(&num_poisoned_pages); } } unset_migratetype_isolate(page, MIGRATE_MOVABLE); -- cgit v0.10.2 From 7ea434a4eb49db83d17cc076f2267704c52938ae Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Tue, 5 May 2015 16:23:49 -0700 Subject: mm/hwpoison-inject: fix refcounting in no-injection case Hwpoison injection via debugfs:hwpoison/corrupt-pfn takes a refcount of the target page. But current code doesn't release it if the target page is not supposed to be injected, which results in memory leak. This patch simply adds the refcount releasing code. Signed-off-by: Naoya Horiguchi Acked-by: Dean Nelson Cc: Andi Kleen Cc: Andrea Arcangeli Cc: Hidetoshi Seto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/hwpoison-inject.c b/mm/hwpoison-inject.c index 329caf5..2b3f933 100644 --- a/mm/hwpoison-inject.c +++ b/mm/hwpoison-inject.c @@ -40,7 +40,7 @@ static int hwpoison_inject(void *data, u64 val) * This implies unable to support non-LRU pages. */ if (!PageLRU(p) && !PageHuge(p)) - return 0; + goto put_out; /* * do a racy check with elevated page count, to make sure PG_hwpoison @@ -52,11 +52,14 @@ static int hwpoison_inject(void *data, u64 val) err = hwpoison_filter(hpage); unlock_page(hpage); if (err) - return 0; + goto put_out; inject: pr_info("Injecting memory failure at pfn %#lx\n", pfn); return memory_failure(pfn, 18, MF_COUNT_INCREASED); +put_out: + put_page(hpage); + return 0; } static int hwpoison_unpoison(void *data, u64 val) -- cgit v0.10.2 From e386eed89c764f102fcc3c0d4c78c65a357f7399 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Tue, 5 May 2015 16:23:52 -0700 Subject: mm/hwpoison-inject: check PageLRU of hpage Hwpoison injector checks PageLRU of the raw target page to find out whether the page is an appropriate target, but current code now filters out thp tail pages, which prevents us from testing for such cases via this interface. So let's check hpage instead of p. Signed-off-by: Naoya Horiguchi Acked-by: Dean Nelson Cc: Andi Kleen Cc: Andrea Arcangeli Cc: Hidetoshi Seto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/hwpoison-inject.c b/mm/hwpoison-inject.c index 2b3f933..4ca5fe0 100644 --- a/mm/hwpoison-inject.c +++ b/mm/hwpoison-inject.c @@ -34,12 +34,12 @@ static int hwpoison_inject(void *data, u64 val) if (!hwpoison_filter_enable) goto inject; - if (!PageLRU(p) && !PageHuge(p)) - shake_page(p, 0); + if (!PageLRU(hpage) && !PageHuge(p)) + shake_page(hpage, 0); /* * This implies unable to support non-LRU pages. */ - if (!PageLRU(p) && !PageHuge(p)) + if (!PageLRU(hpage) && !PageHuge(p)) goto put_out; /* -- cgit v0.10.2 From f5b697700c86d7d01489202bfd37d86665754afd Mon Sep 17 00:00:00 2001 From: Daniel Baluta Date: Tue, 5 May 2015 16:23:54 -0700 Subject: configfs: init configfs module earlier at boot time We need this earlier in the boot process to allow various subsystems to use configfs (e.g Industrial IIO). Also, debugfs is at core_initcall level and configfs should be on the same level from infrastructure point of view. Signed-off-by: Daniel Baluta Suggested-by: Lars-Peter Clausen Reviewed-by: Christoph Hellwig Cc: Al Viro Cc: Joel Becker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index da94e41..5373567 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -173,5 +173,5 @@ MODULE_LICENSE("GPL"); MODULE_VERSION("0.0.2"); MODULE_DESCRIPTION("Simple RAM filesystem for user driven kernel subsystem configuration."); -module_init(configfs_init); +core_initcall(configfs_init); module_exit(configfs_exit); -- cgit v0.10.2 From 05836c378c7af9527b98a83746f32c7289a5f3c8 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Tue, 5 May 2015 16:23:57 -0700 Subject: util_macros.h: have array pointer point to array of constants Using the new find_closest() macro can result in the following sparse warnings. drivers/hwmon/lm85.c:194:16: warning: incorrect type in initializer (different modifiers) drivers/hwmon/lm85.c:194:16: expected int *__fc_a drivers/hwmon/lm85.c:194:16: got int static const [toplevel] * drivers/hwmon/lm85.c:210:16: warning: incorrect type in initializer (different modifiers) drivers/hwmon/lm85.c:210:16: expected int *__fc_a drivers/hwmon/lm85.c:210:16: got int const *map This is because the array passed to find_closest() will typically be declared as array of constants, but the macro declares a non-constant pointer to it. Signed-off-by: Guenter Roeck Cc: Bartosz Golaszewski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/util_macros.h b/include/linux/util_macros.h index d5f4fb6..f9b2ce5 100644 --- a/include/linux/util_macros.h +++ b/include/linux/util_macros.h @@ -5,7 +5,7 @@ ({ \ typeof(as) __fc_i, __fc_as = (as) - 1; \ typeof(x) __fc_x = (x); \ - typeof(*a) *__fc_a = (a); \ + typeof(*a) const *__fc_a = (a); \ for (__fc_i = 0; __fc_i < __fc_as; __fc_i++) { \ if (__fc_x op DIV_ROUND_CLOSEST(__fc_a[__fc_i] + \ __fc_a[__fc_i + 1], 2)) \ -- cgit v0.10.2 From d8fd150fe3935e1692bf57c66691e17409ebb9c1 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Tue, 5 May 2015 16:24:00 -0700 Subject: nilfs2: fix sanity check of btree level in nilfs_btree_root_broken() The range check for b-tree level parameter in nilfs_btree_root_broken() is wrong; it accepts the case of "level == NILFS_BTREE_LEVEL_MAX" even though the level is limited to values in the range of 0 to (NILFS_BTREE_LEVEL_MAX - 1). Since the level parameter is read from storage device and used to index nilfs_btree_path array whose element count is NILFS_BTREE_LEVEL_MAX, it can cause memory overrun during btree operations if the boundary value is set to the level parameter on device. This fixes the broken sanity check and adds a comment to clarify that the upper bound NILFS_BTREE_LEVEL_MAX is exclusive. Signed-off-by: Ryusuke Konishi Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nilfs2/btree.c b/fs/nilfs2/btree.c index 059f371..919fd5b 100644 --- a/fs/nilfs2/btree.c +++ b/fs/nilfs2/btree.c @@ -388,7 +388,7 @@ static int nilfs_btree_root_broken(const struct nilfs_btree_node *node, nchildren = nilfs_btree_node_get_nchildren(node); if (unlikely(level < NILFS_BTREE_LEVEL_NODE_MIN || - level > NILFS_BTREE_LEVEL_MAX || + level >= NILFS_BTREE_LEVEL_MAX || nchildren < 0 || nchildren > NILFS_BTREE_ROOT_NCHILDREN_MAX)) { pr_crit("NILFS: bad btree root (inode number=%lu): level = %d, flags = 0x%x, nchildren = %d\n", diff --git a/include/linux/nilfs2_fs.h b/include/linux/nilfs2_fs.h index ff3fea3..9abb763 100644 --- a/include/linux/nilfs2_fs.h +++ b/include/linux/nilfs2_fs.h @@ -460,7 +460,7 @@ struct nilfs_btree_node { /* level */ #define NILFS_BTREE_LEVEL_DATA 0 #define NILFS_BTREE_LEVEL_NODE_MIN (NILFS_BTREE_LEVEL_DATA + 1) -#define NILFS_BTREE_LEVEL_MAX 14 +#define NILFS_BTREE_LEVEL_MAX 14 /* Max level (exclusive) */ /** * struct nilfs_palloc_group_desc - block group descriptor -- cgit v0.10.2 From b1432a2a35565f538586774a03bf277c27fc267d Mon Sep 17 00:00:00 2001 From: Junxiao Bi Date: Tue, 5 May 2015 16:24:02 -0700 Subject: ocfs2: dlm: fix race between purge and get lock resource There is a race window in dlm_get_lock_resource(), which may return a lock resource which has been purged. This will cause the process to hang forever in dlmlock() as the ast msg can't be handled due to its lock resource not existing. dlm_get_lock_resource { ... spin_lock(&dlm->spinlock); tmpres = __dlm_lookup_lockres_full(dlm, lockid, namelen, hash); if (tmpres) { spin_unlock(&dlm->spinlock); >>>>>>>> race window, dlm_run_purge_list() may run and purge the lock resource spin_lock(&tmpres->spinlock); ... spin_unlock(&tmpres->spinlock); } } Signed-off-by: Junxiao Bi Cc: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index a6944b2..fdf4b41 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -757,6 +757,19 @@ lookup: if (tmpres) { spin_unlock(&dlm->spinlock); spin_lock(&tmpres->spinlock); + + /* + * Right after dlm spinlock was released, dlm_thread could have + * purged the lockres. Check if lockres got unhashed. If so + * start over. + */ + if (hlist_unhashed(&tmpres->hash_node)) { + spin_unlock(&tmpres->spinlock); + dlm_lockres_put(tmpres); + tmpres = NULL; + goto lookup; + } + /* Wait on the thread that is mastering the resource */ if (tmpres->owner == DLM_LOCK_RES_OWNER_UNKNOWN) { __dlm_wait_on_lockres(tmpres); -- cgit v0.10.2 From 489405fe5ed38e65f6f82f131a39c67f3bae6045 Mon Sep 17 00:00:00 2001 From: Gregory CLEMENT Date: Tue, 5 May 2015 16:24:05 -0700 Subject: rtc: armada38x: fix concurrency access in armada38x_rtc_set_time While setting the time, the RTC TIME register should not be accessed. However due to hardware constraints, setting the RTC time involves sleeping during 100ms. This sleep was done outside the critical section protected by the spinlock, so it was possible to read the RTC TIME register and get an incorrect value. This patch introduces a mutex for protecting the RTC TIME access, unlike the spinlock it is allowed to sleep in a critical section protected by a mutex. The RTC STATUS register can still be used from the interrupt handler but it has no effect on setting the time. Signed-off-by: Gregory CLEMENT Acked-by: Alexandre Belloni Acked-by: Andrew Lunn Cc: Alessandro Zummo Cc: [4.0] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-armada38x.c b/drivers/rtc/rtc-armada38x.c index 43e04af..cb70ced 100644 --- a/drivers/rtc/rtc-armada38x.c +++ b/drivers/rtc/rtc-armada38x.c @@ -40,6 +40,13 @@ struct armada38x_rtc { void __iomem *regs; void __iomem *regs_soc; spinlock_t lock; + /* + * While setting the time, the RTC TIME register should not be + * accessed. Setting the RTC time involves sleeping during + * 100ms, so a mutex instead of a spinlock is used to protect + * it + */ + struct mutex mutex_time; int irq; }; @@ -59,8 +66,7 @@ static int armada38x_rtc_read_time(struct device *dev, struct rtc_time *tm) struct armada38x_rtc *rtc = dev_get_drvdata(dev); unsigned long time, time_check, flags; - spin_lock_irqsave(&rtc->lock, flags); - + mutex_lock(&rtc->mutex_time); time = readl(rtc->regs + RTC_TIME); /* * WA for failing time set attempts. As stated in HW ERRATA if @@ -71,7 +77,7 @@ static int armada38x_rtc_read_time(struct device *dev, struct rtc_time *tm) if ((time_check - time) > 1) time_check = readl(rtc->regs + RTC_TIME); - spin_unlock_irqrestore(&rtc->lock, flags); + mutex_unlock(&rtc->mutex_time); rtc_time_to_tm(time_check, tm); @@ -94,19 +100,12 @@ static int armada38x_rtc_set_time(struct device *dev, struct rtc_time *tm) * then wait for 100ms before writing to the time register to be * sure that the data will be taken into account. */ - spin_lock_irqsave(&rtc->lock, flags); - + mutex_lock(&rtc->mutex_time); rtc_delayed_write(0, rtc, RTC_STATUS); - - spin_unlock_irqrestore(&rtc->lock, flags); - msleep(100); - - spin_lock_irqsave(&rtc->lock, flags); - rtc_delayed_write(time, rtc, RTC_TIME); + mutex_unlock(&rtc->mutex_time); - spin_unlock_irqrestore(&rtc->lock, flags); out: return ret; } @@ -230,6 +229,7 @@ static __init int armada38x_rtc_probe(struct platform_device *pdev) return -ENOMEM; spin_lock_init(&rtc->lock); + mutex_init(&rtc->mutex_time); res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "rtc"); rtc->regs = devm_ioremap_resource(&pdev->dev, res); -- cgit v0.10.2 From 15c5725e6b86cb8dfc4ca655a22005cc678a6f6f Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Thu, 16 Apr 2015 21:09:53 +0800 Subject: ipmi: Remove unused including Remove including that don't need it. Signed-off-by: Wei Yongjun Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_ssif.c b/drivers/char/ipmi/ipmi_ssif.c index ee3b8c5..f6ea4fa 100644 --- a/drivers/char/ipmi/ipmi_ssif.c +++ b/drivers/char/ipmi/ipmi_ssif.c @@ -31,7 +31,6 @@ * interface into the I2C driver, I believe. */ -#include #if defined(MODVERSIONS) #include #endif -- cgit v0.10.2 From a182a4b2b3e85a559ea2cd3545f4311db41325f2 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 22 Apr 2015 13:25:40 -0500 Subject: ipmi: Report an error if ACPI _IFT doesn't exist When probing an ACPI table, report a specific error, instead of just returning an error, if _IFT doesn't exist. Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 4612741..b5a1b45 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2262,8 +2262,10 @@ static int ipmi_pnp_probe(struct pnp_dev *dev, /* _IFT tells us the interface type: KCS, BT, etc */ status = acpi_evaluate_integer(handle, "_IFT", NULL, &tmp); - if (ACPI_FAILURE(status)) + if (ACPI_FAILURE(status)) { + dev_err(&dev->dev, "Could not find ACPI IPMI interface type\n"); goto err_free; + } switch (tmp) { case 1: -- cgit v0.10.2 From 9f8127048ab8b47b43f8aeaaec9fec2da44be9a1 Mon Sep 17 00:00:00 2001 From: Hidehiro Kawai Date: Thu, 23 Apr 2015 11:16:44 +0900 Subject: ipmi: Fix a problem that messages are not issued in run_to_completion mode start_next_msg() issues a message placed in smi_info->waiting_msg if it is non-NULL. However, sender() sets a message to smi_info->curr_msg and NULL to smi_info->waiting_msg in the context of run_to_completion mode. As the result, it leads an infinite loop by waiting the completion of unissued message when leaving dying message after kernel panic. sender() should set the message to smi_info->waiting_msg not curr_msg. Signed-off-by: Hidehiro Kawai Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index b5a1b45..8a45e92 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -942,8 +942,7 @@ static void sender(void *send_info, * If we are running to completion, start it and run * transactions until everything is clear. */ - smi_info->curr_msg = msg; - smi_info->waiting_msg = NULL; + smi_info->waiting_msg = msg; /* * Run to completion means we are single-threaded, no -- cgit v0.10.2 From 9162052173d2381e2bbabc224c3c1457acb4c54c Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Fri, 24 Apr 2015 07:46:06 -0500 Subject: ipmi: Add alert handling to SSIF The SSIF interface can optionally have an SMBus alert come in when data is ready. Unfortunately, the IPMI spec gives wiggle room to the implementer to allow them to always have the alert enabled, even if the driver doesn't enable it. So implement alerts. If you don't in this situation, the SMBus alert handling will constantly complain. Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_ssif.c b/drivers/char/ipmi/ipmi_ssif.c index f6ea4fa..5b82d99 100644 --- a/drivers/char/ipmi/ipmi_ssif.c +++ b/drivers/char/ipmi/ipmi_ssif.c @@ -165,6 +165,9 @@ enum ssif_stat_indexes { /* Number of watchdog pretimeouts. */ SSIF_STAT_watchdog_pretimeouts, + /* Number of alers received. */ + SSIF_STAT_alerts, + /* Always add statistics before this value, it must be last. */ SSIF_NUM_STATS }; @@ -213,7 +216,16 @@ struct ssif_info { #define WDT_PRE_TIMEOUT_INT 0x08 unsigned char msg_flags; + u8 global_enables; bool has_event_buffer; + bool supports_alert; + + /* + * Used to tell what we should do with alerts. If we are + * waiting on a response, read the data immediately. + */ + bool got_alert; + bool waiting_alert; /* * If set to true, this will request events the next time the @@ -517,14 +529,10 @@ static int ssif_i2c_send(struct ssif_info *ssif_info, static void msg_done_handler(struct ssif_info *ssif_info, int result, unsigned char *data, unsigned int len); -static void retry_timeout(unsigned long data) +static void start_get(struct ssif_info *ssif_info) { - struct ssif_info *ssif_info = (void *) data; int rv; - if (ssif_info->stopping) - return; - ssif_info->rtc_us_timer = 0; rv = ssif_i2c_send(ssif_info, msg_done_handler, I2C_SMBUS_READ, @@ -539,6 +547,46 @@ static void retry_timeout(unsigned long data) } } +static void retry_timeout(unsigned long data) +{ + struct ssif_info *ssif_info = (void *) data; + unsigned long oflags, *flags; + bool waiting; + + if (ssif_info->stopping) + return; + + flags = ipmi_ssif_lock_cond(ssif_info, &oflags); + waiting = ssif_info->waiting_alert; + ssif_info->waiting_alert = false; + ipmi_ssif_unlock_cond(ssif_info, flags); + + if (waiting) + start_get(ssif_info); +} + + +static void ssif_alert(struct i2c_client *client, unsigned int data) +{ + struct ssif_info *ssif_info = i2c_get_clientdata(client); + unsigned long oflags, *flags; + bool do_get = false; + + ssif_inc_stat(ssif_info, alerts); + + flags = ipmi_ssif_lock_cond(ssif_info, &oflags); + if (ssif_info->waiting_alert) { + ssif_info->waiting_alert = false; + del_timer(&ssif_info->retry_timer); + do_get = true; + } else if (ssif_info->curr_msg) { + ssif_info->got_alert = true; + } + ipmi_ssif_unlock_cond(ssif_info, flags); + if (do_get) + start_get(ssif_info); +} + static int start_resend(struct ssif_info *ssif_info); static void msg_done_handler(struct ssif_info *ssif_info, int result, @@ -558,9 +606,12 @@ static void msg_done_handler(struct ssif_info *ssif_info, int result, if (ssif_info->retries_left > 0) { ssif_inc_stat(ssif_info, receive_retries); + flags = ipmi_ssif_lock_cond(ssif_info, &oflags); + ssif_info->waiting_alert = true; + ssif_info->rtc_us_timer = SSIF_MSG_USEC; mod_timer(&ssif_info->retry_timer, jiffies + SSIF_MSG_JIFFIES); - ssif_info->rtc_us_timer = SSIF_MSG_USEC; + ipmi_ssif_unlock_cond(ssif_info, flags); return; } @@ -649,7 +700,7 @@ static void msg_done_handler(struct ssif_info *ssif_info, int result, if (rv < 0) { if (ssif_info->ssif_debug & SSIF_DEBUG_MSG) pr_info(PFX - "Error from i2c_non_blocking_op(2)\n"); + "Error from ssif_i2c_send\n"); result = -EIO; } else @@ -863,15 +914,32 @@ static void msg_written_handler(struct ssif_info *ssif_info, int result, msg_done_handler(ssif_info, -EIO, NULL, 0); } } else { + unsigned long oflags, *flags; + bool got_alert; + ssif_inc_stat(ssif_info, sent_messages); ssif_inc_stat(ssif_info, sent_messages_parts); - /* Wait a jiffie then request the next message */ - ssif_info->retries_left = SSIF_RECV_RETRIES; - ssif_info->rtc_us_timer = SSIF_MSG_PART_USEC; - mod_timer(&ssif_info->retry_timer, - jiffies + SSIF_MSG_PART_JIFFIES); - return; + flags = ipmi_ssif_lock_cond(ssif_info, &oflags); + got_alert = ssif_info->got_alert; + if (got_alert) { + ssif_info->got_alert = false; + ssif_info->waiting_alert = false; + } + + if (got_alert) { + ipmi_ssif_unlock_cond(ssif_info, flags); + /* The alert already happened, try now. */ + retry_timeout((unsigned long) ssif_info); + } else { + /* Wait a jiffie then request the next message */ + ssif_info->waiting_alert = true; + ssif_info->retries_left = SSIF_RECV_RETRIES; + ssif_info->rtc_us_timer = SSIF_MSG_PART_USEC; + mod_timer(&ssif_info->retry_timer, + jiffies + SSIF_MSG_PART_JIFFIES); + ipmi_ssif_unlock_cond(ssif_info, flags); + } } } @@ -880,6 +948,8 @@ static int start_resend(struct ssif_info *ssif_info) int rv; int command; + ssif_info->got_alert = false; + if (ssif_info->data_len > 32) { command = SSIF_IPMI_MULTI_PART_REQUEST_START; ssif_info->multi_data = ssif_info->data; @@ -1242,6 +1312,8 @@ static int smi_stats_proc_show(struct seq_file *m, void *v) ssif_get_stat(ssif_info, events)); seq_printf(m, "watchdog_pretimeouts: %u\n", ssif_get_stat(ssif_info, watchdog_pretimeouts)); + seq_printf(m, "alerts: %u\n", + ssif_get_stat(ssif_info, alerts)); return 0; } @@ -1324,6 +1396,12 @@ static bool check_acpi(struct ssif_info *ssif_info, struct device *dev) return false; } +/* + * Global enables we care about. + */ +#define GLOBAL_ENABLES_MASK (IPMI_BMC_EVT_MSG_BUFF | IPMI_BMC_RCV_MSG_INTR | \ + IPMI_BMC_EVT_MSG_INTR) + static int ssif_probe(struct i2c_client *client, const struct i2c_device_id *id) { unsigned char msg[3]; @@ -1454,6 +1532,8 @@ static int ssif_probe(struct i2c_client *client, const struct i2c_device_id *id) goto found; } + ssif_info->global_enables = resp[3]; + if (resp[3] & IPMI_BMC_EVT_MSG_BUFF) { ssif_info->has_event_buffer = true; /* buffer is already enabled, nothing to do. */ @@ -1462,18 +1542,37 @@ static int ssif_probe(struct i2c_client *client, const struct i2c_device_id *id) msg[0] = IPMI_NETFN_APP_REQUEST << 2; msg[1] = IPMI_SET_BMC_GLOBAL_ENABLES_CMD; - msg[2] = resp[3] | IPMI_BMC_EVT_MSG_BUFF; + msg[2] = ssif_info->global_enables | IPMI_BMC_EVT_MSG_BUFF; rv = do_cmd(client, 3, msg, &len, resp); if (rv || (len < 2)) { - pr_warn(PFX "Error getting global enables: %d %d %2.2x\n", + pr_warn(PFX "Error setting global enables: %d %d %2.2x\n", rv, len, resp[2]); rv = 0; /* Not fatal */ goto found; } - if (resp[2] == 0) + if (resp[2] == 0) { /* A successful return means the event buffer is supported. */ ssif_info->has_event_buffer = true; + ssif_info->global_enables |= IPMI_BMC_EVT_MSG_BUFF; + } + + msg[0] = IPMI_NETFN_APP_REQUEST << 2; + msg[1] = IPMI_SET_BMC_GLOBAL_ENABLES_CMD; + msg[2] = ssif_info->global_enables | IPMI_BMC_RCV_MSG_INTR; + rv = do_cmd(client, 3, msg, &len, resp); + if (rv || (len < 2)) { + pr_warn(PFX "Error setting global enables: %d %d %2.2x\n", + rv, len, resp[2]); + rv = 0; /* Not fatal */ + goto found; + } + + if (resp[2] == 0) { + /* A successful return means the alert is supported. */ + ssif_info->supports_alert = true; + ssif_info->global_enables |= IPMI_BMC_RCV_MSG_INTR; + } found: ssif_info->intf_num = atomic_inc_return(&next_intf); @@ -1831,6 +1930,7 @@ static struct i2c_driver ssif_i2c_driver = { }, .probe = ssif_probe, .remove = ssif_remove, + .alert = ssif_alert, .id_table = ssif_id, .detect = ssif_detect }; -- cgit v0.10.2 From 3d69d43baa2749c3d187ce70940d7aebe609e149 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 29 Apr 2015 17:59:21 -0500 Subject: ipmi: Fix multi-part message handling Lots of little fixes for multi-part messages: The values was not being re-initialized, if something went wrong handling a multi-part message and it got left in a bad state, it might be an issue. The commands were not correct when issuing multi-part reads, the code was not passing in the proper value for commands. Also clean up some minor formatting issues. Get the block number from the right location, limit the maximum send message size to 63 bytes and explain why, and fix some minor sylistic issues. Signed-off-by: Corey Minyard diff --git a/drivers/char/ipmi/ipmi_ssif.c b/drivers/char/ipmi/ipmi_ssif.c index 5b82d99..207689c 100644 --- a/drivers/char/ipmi/ipmi_ssif.c +++ b/drivers/char/ipmi/ipmi_ssif.c @@ -489,13 +489,13 @@ static int ipmi_ssif_thread(void *data) if (ssif_info->i2c_read_write == I2C_SMBUS_WRITE) { result = i2c_smbus_write_block_data( - ssif_info->client, SSIF_IPMI_REQUEST, + ssif_info->client, ssif_info->i2c_command, ssif_info->i2c_data[0], ssif_info->i2c_data + 1); ssif_info->done_handler(ssif_info, result, NULL, 0); } else { result = i2c_smbus_read_block_data( - ssif_info->client, SSIF_IPMI_RESPONSE, + ssif_info->client, ssif_info->i2c_command, ssif_info->i2c_data); if (result < 0) ssif_info->done_handler(ssif_info, result, @@ -534,6 +534,7 @@ static void start_get(struct ssif_info *ssif_info) int rv; ssif_info->rtc_us_timer = 0; + ssif_info->multi_pos = 0; rv = ssif_i2c_send(ssif_info, msg_done_handler, I2C_SMBUS_READ, SSIF_IPMI_RESPONSE, @@ -631,9 +632,9 @@ static void msg_done_handler(struct ssif_info *ssif_info, int result, ssif_inc_stat(ssif_info, received_message_parts); /* Remove the multi-part read marker. */ - for (i = 0; i < (len-2); i++) - ssif_info->data[i] = data[i+2]; len -= 2; + for (i = 0; i < len; i++) + ssif_info->data[i] = data[i+2]; ssif_info->multi_len = len; ssif_info->multi_pos = 1; @@ -660,9 +661,9 @@ static void msg_done_handler(struct ssif_info *ssif_info, int result, goto continue_op; } - blocknum = data[ssif_info->multi_len]; + blocknum = data[0]; - if (ssif_info->multi_len+len-1 > IPMI_MAX_MSG_LENGTH) { + if (ssif_info->multi_len + len - 1 > IPMI_MAX_MSG_LENGTH) { /* Received message too big, abort the operation. */ result = -E2BIG; if (ssif_info->ssif_debug & SSIF_DEBUG_MSG) @@ -672,15 +673,15 @@ static void msg_done_handler(struct ssif_info *ssif_info, int result, } /* Remove the blocknum from the data. */ - for (i = 0; i < (len-1); i++) - ssif_info->data[i+ssif_info->multi_len] = data[i+1]; len--; + for (i = 0; i < len; i++) + ssif_info->data[i + ssif_info->multi_len] = data[i + 1]; ssif_info->multi_len += len; if (blocknum == 0xff) { /* End of read */ len = ssif_info->multi_len; data = ssif_info->data; - } else if ((blocknum+1) != ssif_info->multi_pos) { + } else if (blocknum + 1 != ssif_info->multi_pos) { /* * Out of sequence block, just abort. Block * numbers start at zero for the second block, @@ -880,7 +881,11 @@ static void msg_written_handler(struct ssif_info *ssif_info, int result, } if (ssif_info->multi_data) { - /* In the middle of a multi-data write. */ + /* + * In the middle of a multi-data write. See the comment + * in the SSIF_MULTI_n_PART case in the probe function + * for details on the intricacies of this. + */ int left; ssif_inc_stat(ssif_info, sent_messages_parts); @@ -984,7 +989,7 @@ static int start_send(struct ssif_info *ssif_info, return -E2BIG; ssif_info->retries_left = SSIF_SEND_RETRIES; - memcpy(ssif_info->data+1, data, len); + memcpy(ssif_info->data + 1, data, len); ssif_info->data_len = len; return start_resend(ssif_info); } @@ -1487,13 +1492,33 @@ static int ssif_probe(struct i2c_client *client, const struct i2c_device_id *id) break; case SSIF_MULTI_2_PART: - if (ssif_info->max_xmit_msg_size > 64) - ssif_info->max_xmit_msg_size = 64; + if (ssif_info->max_xmit_msg_size > 63) + ssif_info->max_xmit_msg_size = 63; if (ssif_info->max_recv_msg_size > 62) ssif_info->max_recv_msg_size = 62; break; case SSIF_MULTI_n_PART: + /* + * The specification is rather confusing at + * this point, but I think I understand what + * is meant. At least I have a workable + * solution. With multi-part messages, you + * cannot send a message that is a multiple of + * 32-bytes in length, because the start and + * middle messages are 32-bytes and the end + * message must be at least one byte. You + * can't fudge on an extra byte, that would + * screw up things like fru data writes. So + * we limit the length to 63 bytes. That way + * a 32-byte message gets sent as a single + * part. A larger message will be a 32-byte + * start and the next message is always going + * to be 1-31 bytes in length. Not ideal, but + * it should work. + */ + if (ssif_info->max_xmit_msg_size > 63) + ssif_info->max_xmit_msg_size = 63; break; default: -- cgit v0.10.2 From 9fcb1704d1d51b12e2f03c78bca013d0cbbb7c98 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Tue, 5 May 2015 16:32:12 +0300 Subject: drm/i915/dp: there is no audio on port A The eDP port A register on PCH split platforms has a slightly different register layout from the other ports, with bit 6 being either alternate scrambler reset or reserved, depending on the generation. Our misinterpretation of the bit as audio has lead to warning. Fix this by not enabling audio on port A, since none of our platforms support audio on port A anyway. v2: DDI doesn't have audio on port A either (Sivakumar Thulasimani) Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89958 Reported-and-tested-by: Chris Bainbridge Cc: stable@vger.kernel.org Reviewed-by: Sivakumar Thulasimani Signed-off-by: Jani Nikula diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index d023710..8d534f4 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1348,7 +1348,7 @@ intel_dp_compute_config(struct intel_encoder *encoder, pipe_config->has_dp_encoder = true; pipe_config->has_drrs = false; - pipe_config->has_audio = intel_dp->has_audio; + pipe_config->has_audio = intel_dp->has_audio && port != PORT_A; if (is_edp(intel_dp) && intel_connector->panel.fixed_mode) { intel_fixed_panel_mode(intel_connector->panel.fixed_mode, @@ -2211,8 +2211,8 @@ static void intel_dp_get_config(struct intel_encoder *encoder, int dotclock; tmp = I915_READ(intel_dp->output_reg); - if (tmp & DP_AUDIO_OUTPUT_ENABLE) - pipe_config->has_audio = true; + + pipe_config->has_audio = tmp & DP_AUDIO_OUTPUT_ENABLE && port != PORT_A; if ((port == PORT_A) || !HAS_PCH_CPT(dev)) { if (tmp & DP_SYNC_HS_HIGH) -- cgit v0.10.2 From c88d47480d300eaad80c213d50c9bf6077fc49bc Mon Sep 17 00:00:00 2001 From: Bobby Powers Date: Mon, 27 Apr 2015 08:10:41 -0700 Subject: x86/fpu: Always restore_xinit_state() when use_eager_cpu() The following commit: f893959b0898 ("x86/fpu: Don't abuse drop_init_fpu() in flush_thread()") removed drop_init_fpu() usage from flush_thread(). This seems to break things for me - the Go 1.4 test suite fails all over the place with floating point comparision errors (offending commit found through bisection). The functional change was that flush_thread() after this commit only calls restore_init_xstate() when both use_eager_fpu() and !used_math() are true. drop_init_fpu() (now fpu_reset_state()) calls restore_init_xstate() regardless of whether current used_math() - apply the same logic here. Switch used_math() -> tsk_used_math(tsk) to consistently use the grabbed tsk instead of current, like in the rest of flush_thread(). Tested-by: Dave Hansen Signed-off-by: Bobby Powers Signed-off-by: Borislav Petkov Acked-by: Oleg Nesterov Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Fenghua Yu Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Pekka Riikonen Cc: Quentin Casasnovas Cc: Rik van Riel Cc: Suresh Siddha Cc: Thomas Gleixner Fixes: f893959b ("x86/fpu: Don't abuse drop_init_fpu() in flush_thread()") Link: http://lkml.kernel.org/r/1430147441-9820-1-git-send-email-bobbypowers@gmail.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index bfc99b3..6e338e3 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -156,11 +156,13 @@ void flush_thread(void) /* FPU state will be reallocated lazily at the first use. */ drop_fpu(tsk); free_thread_xstate(tsk); - } else if (!used_math()) { - /* kthread execs. TODO: cleanup this horror. */ - if (WARN_ON(init_fpu(tsk))) - force_sig(SIGKILL, tsk); - user_fpu_begin(); + } else { + if (!tsk_used_math(tsk)) { + /* kthread execs. TODO: cleanup this horror. */ + if (WARN_ON(init_fpu(tsk))) + force_sig(SIGKILL, tsk); + user_fpu_begin(); + } restore_init_xstate(); } } -- cgit v0.10.2 From bad4371d87d1d1ed1aecd9c9cc21c41ac3f289c8 Mon Sep 17 00:00:00 2001 From: Takeshi Kihara Date: Thu, 30 Apr 2015 02:03:51 +0900 Subject: mmc: sh_mmcif: Fix timeout value for command request f9fd54f22e ("mmc: sh_mmcif: Use msecs_to_jiffies() for host->timeout") changed the timeout value from 1000 jiffies to 1s. In the case where HZ is 1000 the values are the same. However, for smaller HZ values the timeout is now smaller, 1s instead of 10s in the case of HZ=100. Since the timeout occurs in spite of a normal data transfer a timeout of 10s seems more appropriate. This restores the previous timeout in the case where HZ=100 and results in an increase over the previous timeout for larger values of HZ. Fixes: f9fd54f22e ("mmc: sh_mmcif: Use msecs_to_jiffies() for host->timeout") Signed-off-by: Takeshi Kihara [horms: rewrote changelog to refer to HZ] Signed-off-by: Simon Horman Signed-off-by: Yoshihiro Kaneko Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c index 2b6ef6b..7eff087 100644 --- a/drivers/mmc/host/sh_mmcif.c +++ b/drivers/mmc/host/sh_mmcif.c @@ -1408,7 +1408,7 @@ static int sh_mmcif_probe(struct platform_device *pdev) host = mmc_priv(mmc); host->mmc = mmc; host->addr = reg; - host->timeout = msecs_to_jiffies(1000); + host->timeout = msecs_to_jiffies(10000); host->ccs_enable = !pd || !pd->ccs_unsupported; host->clk_ctrl2_enable = pd && pd->clk_ctrl2_present; -- cgit v0.10.2 From 4e93b9a6abc0d028daf3c8a00cb77b679d8a4df4 Mon Sep 17 00:00:00 2001 From: Chuanxiao Dong Date: Tue, 12 Aug 2014 12:01:30 +0800 Subject: mmc: card: Don't access RPMB partitions for normal read/write During kernel boot, it will try to read some logical sectors of each block device node for the possible partition table. But since RPMB partition is special and can not be accessed by normal eMMC read / write CMDs, it will cause below error messages during kernel boot: ... mmc0: Got data interrupt 0x00000002 even though no data operation was in progress. mmcblk0rpmb: error -110 transferring data, sector 0, nr 32, cmd response 0x900, card status 0xb00 mmcblk0rpmb: retrying using single block read mmcblk0rpmb: timed out sending r/w cmd command, card status 0x400900 mmcblk0rpmb: timed out sending r/w cmd command, card status 0x400900 mmcblk0rpmb: timed out sending r/w cmd command, card status 0x400900 mmcblk0rpmb: timed out sending r/w cmd command, card status 0x400900 mmcblk0rpmb: timed out sending r/w cmd command, card status 0x400900 mmcblk0rpmb: timed out sending r/w cmd command, card status 0x400900 end_request: I/O error, dev mmcblk0rpmb, sector 0 Buffer I/O error on device mmcblk0rpmb, logical block 0 end_request: I/O error, dev mmcblk0rpmb, sector 8 Buffer I/O error on device mmcblk0rpmb, logical block 1 end_request: I/O error, dev mmcblk0rpmb, sector 16 Buffer I/O error on device mmcblk0rpmb, logical block 2 end_request: I/O error, dev mmcblk0rpmb, sector 24 Buffer I/O error on device mmcblk0rpmb, logical block 3 ... This patch will discard the access request in eMMC queue if it is RPMB partition access request. By this way, it avoids trigger above error messages. Fixes: 090d25fe224c ("mmc: core: Expose access to RPMB partition") Signed-off-by: Yunpeng Gao Signed-off-by: Chuanxiao Dong Tested-by: Michael Shigorin Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c index 2c25271..60f7141 100644 --- a/drivers/mmc/card/block.c +++ b/drivers/mmc/card/block.c @@ -1029,6 +1029,18 @@ static inline void mmc_blk_reset_success(struct mmc_blk_data *md, int type) md->reset_done &= ~type; } +int mmc_access_rpmb(struct mmc_queue *mq) +{ + struct mmc_blk_data *md = mq->data; + /* + * If this is a RPMB partition access, return ture + */ + if (md && md->part_type == EXT_CSD_PART_CONFIG_ACC_RPMB) + return true; + + return false; +} + static int mmc_blk_issue_discard_rq(struct mmc_queue *mq, struct request *req) { struct mmc_blk_data *md = mq->data; diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c index 236d194..8efa368 100644 --- a/drivers/mmc/card/queue.c +++ b/drivers/mmc/card/queue.c @@ -38,7 +38,7 @@ static int mmc_prep_request(struct request_queue *q, struct request *req) return BLKPREP_KILL; } - if (mq && mmc_card_removed(mq->card)) + if (mq && (mmc_card_removed(mq->card) || mmc_access_rpmb(mq))) return BLKPREP_KILL; req->cmd_flags |= REQ_DONTPREP; diff --git a/drivers/mmc/card/queue.h b/drivers/mmc/card/queue.h index 5752d50..99e6521 100644 --- a/drivers/mmc/card/queue.h +++ b/drivers/mmc/card/queue.h @@ -73,4 +73,6 @@ extern void mmc_queue_bounce_post(struct mmc_queue_req *); extern int mmc_packed_init(struct mmc_queue *, struct mmc_card *); extern void mmc_packed_clean(struct mmc_queue *); +extern int mmc_access_rpmb(struct mmc_queue *); + #endif -- cgit v0.10.2 From d2d05c65c40e067ca5898399069053f095c67d6f Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Thu, 23 Apr 2015 16:54:17 -0700 Subject: gpio: omap: Fix regression for MPUIO interrupts At some point with all the GPIO clean-up we've broken the MPUIO interrupts. Those are just a little bit different from the GPIO interrupts, so we can fix it up just by setting different irqchip functions for it. And then we can just remove all old code trying to do the same. Cc: Aaro Koskinen Cc: Javier Martinez Canillas Cc: Kevin Hilman Cc: Nishanth Menon Signed-off-by: Tony Lindgren Reviewed-by: Grygorii Strashko Acked-by: Santosh Shilimkar Reviewed-by: Felipe Balbi Signed-off-by: Linus Walleij diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c index cd1d5bf..b232397 100644 --- a/drivers/gpio/gpio-omap.c +++ b/drivers/gpio/gpio-omap.c @@ -1054,38 +1054,8 @@ static void omap_gpio_mod_init(struct gpio_bank *bank) dev_err(bank->dev, "Could not get gpio dbck\n"); } -static void -omap_mpuio_alloc_gc(struct gpio_bank *bank, unsigned int irq_start, - unsigned int num) -{ - struct irq_chip_generic *gc; - struct irq_chip_type *ct; - - gc = irq_alloc_generic_chip("MPUIO", 1, irq_start, bank->base, - handle_simple_irq); - if (!gc) { - dev_err(bank->dev, "Memory alloc failed for gc\n"); - return; - } - - ct = gc->chip_types; - - /* NOTE: No ack required, reading IRQ status clears it. */ - ct->chip.irq_mask = irq_gc_mask_set_bit; - ct->chip.irq_unmask = irq_gc_mask_clr_bit; - ct->chip.irq_set_type = omap_gpio_irq_type; - - if (bank->regs->wkup_en) - ct->chip.irq_set_wake = omap_gpio_wake_enable; - - ct->regs.mask = OMAP_MPUIO_GPIO_INT / bank->stride; - irq_setup_generic_chip(gc, IRQ_MSK(num), IRQ_GC_INIT_MASK_CACHE, - IRQ_NOREQUEST | IRQ_NOPROBE, 0); -} - static int omap_gpio_chip_init(struct gpio_bank *bank, struct irq_chip *irqc) { - int j; static int gpio; int irq_base = 0; int ret; @@ -1132,6 +1102,15 @@ static int omap_gpio_chip_init(struct gpio_bank *bank, struct irq_chip *irqc) } #endif + /* MPUIO is a bit different, reading IRQ status clears it */ + if (bank->is_mpuio) { + irqc->irq_ack = dummy_irq_chip.irq_ack; + irqc->irq_mask = irq_gc_mask_set_bit; + irqc->irq_unmask = irq_gc_mask_clr_bit; + if (!bank->regs->wkup_en) + irqc->irq_set_wake = NULL; + } + ret = gpiochip_irqchip_add(&bank->chip, irqc, irq_base, omap_gpio_irq_handler, IRQ_TYPE_NONE); @@ -1145,15 +1124,6 @@ static int omap_gpio_chip_init(struct gpio_bank *bank, struct irq_chip *irqc) gpiochip_set_chained_irqchip(&bank->chip, irqc, bank->irq, omap_gpio_irq_handler); - for (j = 0; j < bank->width; j++) { - int irq = irq_find_mapping(bank->chip.irqdomain, j); - if (bank->is_mpuio) { - omap_mpuio_alloc_gc(bank, irq, bank->width); - irq_set_chip_and_handler(irq, NULL, NULL); - set_irq_flags(irq, 0); - } - } - return 0; } -- cgit v0.10.2 From 8746515d7f04c9ea94cf43e2db1fd2cfca93276d Mon Sep 17 00:00:00 2001 From: Stefano Stabellini Date: Fri, 24 Apr 2015 10:16:40 +0100 Subject: xen: Add __GFP_DMA flag when xen_swiotlb_init gets free pages on ARM Make sure that xen_swiotlb_init allocates buffers that are DMA capable when at least one memblock is available below 4G. Otherwise we assume that all devices on the SoC can cope with >4G addresses. We do this on ARM and ARM64, where dom0 is mapped 1:1, so pfn == mfn in this case. No functional changes on x86. From: Chen Baozi Signed-off-by: Chen Baozi Signed-off-by: Stefano Stabellini Tested-by: Chen Baozi Acked-by: Konrad Rzeszutek Wilk Signed-off-by: David Vrabel diff --git a/arch/arm/include/asm/xen/page.h b/arch/arm/include/asm/xen/page.h index 2f7e6ff..0b579b2 100644 --- a/arch/arm/include/asm/xen/page.h +++ b/arch/arm/include/asm/xen/page.h @@ -110,5 +110,6 @@ static inline bool set_phys_to_machine(unsigned long pfn, unsigned long mfn) bool xen_arch_need_swiotlb(struct device *dev, unsigned long pfn, unsigned long mfn); +unsigned long xen_get_swiotlb_free_pages(unsigned int order); #endif /* _ASM_ARM_XEN_PAGE_H */ diff --git a/arch/arm/xen/mm.c b/arch/arm/xen/mm.c index 793551d..4983250 100644 --- a/arch/arm/xen/mm.c +++ b/arch/arm/xen/mm.c @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include @@ -21,6 +22,20 @@ #include #include +unsigned long xen_get_swiotlb_free_pages(unsigned int order) +{ + struct memblock_region *reg; + gfp_t flags = __GFP_NOWARN; + + for_each_memblock(memory, reg) { + if (reg->base < (phys_addr_t)0xffffffff) { + flags |= __GFP_DMA; + break; + } + } + return __get_free_pages(flags, order); +} + enum dma_cache_op { DMA_UNMAP, DMA_MAP, diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index 358dcd3..c44a5d5 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -269,4 +269,9 @@ static inline bool xen_arch_need_swiotlb(struct device *dev, return false; } +static inline unsigned long xen_get_swiotlb_free_pages(unsigned int order) +{ + return __get_free_pages(__GFP_NOWARN, order); +} + #endif /* _ASM_X86_XEN_PAGE_H */ diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c index 810ad41..4c54932 100644 --- a/drivers/xen/swiotlb-xen.c +++ b/drivers/xen/swiotlb-xen.c @@ -235,7 +235,7 @@ retry: #define SLABS_PER_PAGE (1 << (PAGE_SHIFT - IO_TLB_SHIFT)) #define IO_TLB_MIN_SLABS ((1<<20) >> IO_TLB_SHIFT) while ((SLABS_PER_PAGE << order) > IO_TLB_MIN_SLABS) { - xen_io_tlb_start = (void *)__get_free_pages(__GFP_NOWARN, order); + xen_io_tlb_start = (void *)xen_get_swiotlb_free_pages(order); if (xen_io_tlb_start) break; order--; -- cgit v0.10.2 From c5272a28566b00cce79127ad382406e0a8650690 Mon Sep 17 00:00:00 2001 From: Doug Anderson Date: Fri, 1 May 2015 09:01:27 -0700 Subject: pinctrl: Don't just pretend to protect pinctrl_maps, do it for real Way back, when the world was a simpler place and there was no war, no evil, and no kernel bugs, there was just a single pinctrl lock. That was how the world was when (57291ce pinctrl: core device tree mapping table parsing support) was written. In that case, there were instances where the pinctrl mutex was already held when pinctrl_register_map() was called, hence a "locked" parameter was passed to the function to indicate that the mutex was already locked (so we shouldn't lock it again). A few years ago in (42fed7b pinctrl: move subsystem mutex to pinctrl_dev struct), we switched to a separate pinctrl_maps_mutex. ...but (oops) we forgot to re-think about the whole "locked" parameter for pinctrl_register_map(). Basically the "locked" parameter appears to still refer to whether the bigger pinctrl_dev mutex is locked, but we're using it to skip locks of our (now separate) pinctrl_maps_mutex. That's kind of a bad thing(TM). Probably nobody noticed because most of the calls to pinctrl_register_map happen at boot time and we've got synchronous device probing. ...and even cases where we're asynchronous don't end up actually hitting the race too often. ...but after banging my head against the wall for a bug that reproduced 1 out of 1000 reboots and lots of looking through kgdb, I finally noticed this. Anyway, we can now safely remove the "locked" parameter and go back to a war-free, evil-free, and kernel-bug-free world. Fixes: 42fed7ba44e4 ("pinctrl: move subsystem mutex to pinctrl_dev struct") Signed-off-by: Doug Anderson Signed-off-by: Linus Walleij diff --git a/drivers/pinctrl/core.c b/drivers/pinctrl/core.c index 89dca77..18ee208 100644 --- a/drivers/pinctrl/core.c +++ b/drivers/pinctrl/core.c @@ -1110,7 +1110,7 @@ void devm_pinctrl_put(struct pinctrl *p) EXPORT_SYMBOL_GPL(devm_pinctrl_put); int pinctrl_register_map(struct pinctrl_map const *maps, unsigned num_maps, - bool dup, bool locked) + bool dup) { int i, ret; struct pinctrl_maps *maps_node; @@ -1178,11 +1178,9 @@ int pinctrl_register_map(struct pinctrl_map const *maps, unsigned num_maps, maps_node->maps = maps; } - if (!locked) - mutex_lock(&pinctrl_maps_mutex); + mutex_lock(&pinctrl_maps_mutex); list_add_tail(&maps_node->node, &pinctrl_maps); - if (!locked) - mutex_unlock(&pinctrl_maps_mutex); + mutex_unlock(&pinctrl_maps_mutex); return 0; } @@ -1197,7 +1195,7 @@ int pinctrl_register_map(struct pinctrl_map const *maps, unsigned num_maps, int pinctrl_register_mappings(struct pinctrl_map const *maps, unsigned num_maps) { - return pinctrl_register_map(maps, num_maps, true, false); + return pinctrl_register_map(maps, num_maps, true); } void pinctrl_unregister_map(struct pinctrl_map const *map) diff --git a/drivers/pinctrl/core.h b/drivers/pinctrl/core.h index 75476b3..b24ea84 100644 --- a/drivers/pinctrl/core.h +++ b/drivers/pinctrl/core.h @@ -183,7 +183,7 @@ static inline struct pin_desc *pin_desc_get(struct pinctrl_dev *pctldev, } int pinctrl_register_map(struct pinctrl_map const *maps, unsigned num_maps, - bool dup, bool locked); + bool dup); void pinctrl_unregister_map(struct pinctrl_map const *map); extern int pinctrl_force_sleep(struct pinctrl_dev *pctldev); diff --git a/drivers/pinctrl/devicetree.c b/drivers/pinctrl/devicetree.c index eda13de..0bbf7d7 100644 --- a/drivers/pinctrl/devicetree.c +++ b/drivers/pinctrl/devicetree.c @@ -92,7 +92,7 @@ static int dt_remember_or_free_map(struct pinctrl *p, const char *statename, dt_map->num_maps = num_maps; list_add_tail(&dt_map->node, &p->dt_maps); - return pinctrl_register_map(map, num_maps, false, true); + return pinctrl_register_map(map, num_maps, false); } struct pinctrl_dev *of_pinctrl_get(struct device_node *np) -- cgit v0.10.2 From 0ff28d9f4674d781e492bcff6f32f0fe48cf0fed Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Wed, 6 May 2015 17:26:47 +0200 Subject: splice: sendfile() at once fails for big files Using sendfile with below small program to get MD5 sums of some files, it appear that big files (over 64kbytes with 4k pages system) get a wrong MD5 sum while small files get the correct sum. This program uses sendfile() to send a file to an AF_ALG socket for hashing. /* md5sum2.c */ #include #include #include #include #include #include #include #include #include int main(int argc, char **argv) { int sk = socket(AF_ALG, SOCK_SEQPACKET, 0); struct stat st; struct sockaddr_alg sa = { .salg_family = AF_ALG, .salg_type = "hash", .salg_name = "md5", }; int n; bind(sk, (struct sockaddr*)&sa, sizeof(sa)); for (n = 1; n < argc; n++) { int size; int offset = 0; char buf[4096]; int fd; int sko; int i; fd = open(argv[n], O_RDONLY); sko = accept(sk, NULL, 0); fstat(fd, &st); size = st.st_size; sendfile(sko, fd, &offset, size); size = read(sko, buf, sizeof(buf)); for (i = 0; i < size; i++) printf("%2.2x", buf[i]); printf(" %s\n", argv[n]); close(fd); close(sko); } exit(0); } Test below is done using official linux patch files. First result is with a software based md5sum. Second result is with the program above. root@vgoip:~# ls -l patch-3.6.* -rw-r--r-- 1 root root 64011 Aug 24 12:01 patch-3.6.2.gz -rw-r--r-- 1 root root 94131 Aug 24 12:01 patch-3.6.3.gz root@vgoip:~# md5sum patch-3.6.* b3ffb9848196846f31b2ff133d2d6443 patch-3.6.2.gz c5e8f687878457db77cb7158c38a7e43 patch-3.6.3.gz root@vgoip:~# ./md5sum2 patch-3.6.* b3ffb9848196846f31b2ff133d2d6443 patch-3.6.2.gz 5fd77b24e68bb24dcc72d6e57c64790e patch-3.6.3.gz After investivation, it appears that sendfile() sends the files by blocks of 64kbytes (16 times PAGE_SIZE). The problem is that at the end of each block, the SPLICE_F_MORE flag is missing, therefore the hashing operation is reset as if it was the end of the file. This patch adds SPLICE_F_MORE to the flags when more data is pending. With the patch applied, we get the correct sums: root@vgoip:~# md5sum patch-3.6.* b3ffb9848196846f31b2ff133d2d6443 patch-3.6.2.gz c5e8f687878457db77cb7158c38a7e43 patch-3.6.3.gz root@vgoip:~# ./md5sum2 patch-3.6.* b3ffb9848196846f31b2ff133d2d6443 patch-3.6.2.gz c5e8f687878457db77cb7158c38a7e43 patch-3.6.3.gz Signed-off-by: Christophe Leroy Signed-off-by: Jens Axboe diff --git a/fs/splice.c b/fs/splice.c index 476024b..bfe62ae 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1161,7 +1161,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, long ret, bytes; umode_t i_mode; size_t len; - int i, flags; + int i, flags, more; /* * We require the input being a regular file, as we don't want to @@ -1204,6 +1204,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, * Don't block on output, we have to drain the direct pipe. */ sd->flags &= ~SPLICE_F_NONBLOCK; + more = sd->flags & SPLICE_F_MORE; while (len) { size_t read_len; @@ -1217,6 +1218,15 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, sd->total_len = read_len; /* + * If more data is pending, set SPLICE_F_MORE + * If this is the last data and SPLICE_F_MORE was not set + * initially, clears it. + */ + if (read_len < len) + sd->flags |= SPLICE_F_MORE; + else if (!more) + sd->flags &= ~SPLICE_F_MORE; + /* * NOTE: nonblocking mode only applies to the input. We * must not do the output in nonblocking mode as then we * could get stuck data in the internal pipe: -- cgit v0.10.2 From 1d3c61c2eb3fe4f96d3192212f1bdcee49ea55aa Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 6 May 2015 11:17:01 +0100 Subject: Btrfs: fix wrong mapping flags for free space inode We were passing a flags value that differed from the intention in commit 2b108268006e ("Btrfs: don't use highmem for free space cache pages"). This caused problems in a ARM machine, leaving btrfs unusable there. Reported-by: Merlijn Wajer Tested-by: Merlijn Wajer Signed-off-by: Filipe Manana Signed-off-by: Chris Mason diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 41c510b..5e020d7 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -86,7 +86,7 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root, mapping_set_gfp_mask(inode->i_mapping, mapping_gfp_mask(inode->i_mapping) & - ~(GFP_NOFS & ~__GFP_HIGHMEM)); + ~(__GFP_FS | __GFP_HIGHMEM)); return inode; } -- cgit v0.10.2 From ac01ce1410fc2c7b5f3af5e9c972e6a412eee54f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Alex=20Benn=C3=A9e?= Date: Wed, 29 Apr 2015 16:18:46 +0100 Subject: tracing: Make ftrace_print_array_seq compute buf_len MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The only caller to this function (__print_array) was getting it wrong by passing the array length instead of buffer length. As the element size was already being passed for other reasons it seems reasonable to push the calculation of buffer length into the function. Link: http://lkml.kernel.org/r/1430320727-14582-1-git-send-email-alex.bennee@linaro.org Signed-off-by: Alex Bennée Signed-off-by: Steven Rostedt diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 46e83c2..f9ecf63 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -46,7 +46,7 @@ const char *ftrace_print_hex_seq(struct trace_seq *p, const unsigned char *buf, int len); const char *ftrace_print_array_seq(struct trace_seq *p, - const void *buf, int buf_len, + const void *buf, int count, size_t el_size); struct trace_iterator; diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index 692bf71..25a086b 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -178,12 +178,13 @@ ftrace_print_hex_seq(struct trace_seq *p, const unsigned char *buf, int buf_len) EXPORT_SYMBOL(ftrace_print_hex_seq); const char * -ftrace_print_array_seq(struct trace_seq *p, const void *buf, int buf_len, +ftrace_print_array_seq(struct trace_seq *p, const void *buf, int count, size_t el_size) { const char *ret = trace_seq_buffer_ptr(p); const char *prefix = ""; void *ptr = (void *)buf; + size_t buf_len = count * el_size; trace_seq_putc(p, '{'); -- cgit v0.10.2 From af77b9741300616e7d04264d0ee7ac9f8abb448f Mon Sep 17 00:00:00 2001 From: Sonika Jindal Date: Thu, 7 May 2015 13:59:28 +0530 Subject: drm/i915: Sink rate read should be saved in deca-kHz MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The sink rate read from supported link rate table is in KHz as per spec while in drm, the saved clock is in deca-KHz. So divide the link rate by 10 before storing. Reading of rates was added by: commit fc0f8e25318f ("drm/i915/skl: Read sink supported rates from edp panel") Cc: Ville Syrjälä Signed-off-by: Sonika Jindal Reviewed-by: Ville Syrjälä Signed-off-by: Jani Nikula diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 8d534f4..f27346e 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -3812,7 +3812,8 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp) if (val == 0) break; - intel_dp->sink_rates[i] = val * 200; + /* Value read is in kHz while drm clock is saved in deca-kHz */ + intel_dp->sink_rates[i] = (val * 200) / 10; } intel_dp->num_sink_rates = i; } -- cgit v0.10.2 From a3bd92519879ddc3052140e6776173f266c934eb Mon Sep 17 00:00:00 2001 From: "Fang, Yang A" Date: Thu, 23 Apr 2015 16:35:17 -0700 Subject: ASoC: rt5645: Add ACPI match ID This patch adds the ACPI match ID for rt5645/5650 codec Signed-off-by: Fang, Yang A Signed-off-by: Mark Brown diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 69528ae..61ca49f 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -2656,6 +2657,15 @@ static const struct i2c_device_id rt5645_i2c_id[] = { }; MODULE_DEVICE_TABLE(i2c, rt5645_i2c_id); +#ifdef CONFIG_ACPI +static struct acpi_device_id rt5645_acpi_match[] = { + { "10EC5645", 0 }, + { "10EC5650", 0 }, + {}, +}; +MODULE_DEVICE_TABLE(acpi, rt5645_acpi_match); +#endif + static int rt5645_i2c_probe(struct i2c_client *i2c, const struct i2c_device_id *id) { @@ -2872,6 +2882,7 @@ static struct i2c_driver rt5645_i2c_driver = { .driver = { .name = "rt5645", .owner = THIS_MODULE, + .acpi_match_table = ACPI_PTR(rt5645_acpi_match), }, .probe = rt5645_i2c_probe, .remove = rt5645_i2c_remove, -- cgit v0.10.2 From 5023281b1870fdc99bddf95a7bc00b801801d13e Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 21 Apr 2015 12:19:49 +0800 Subject: ASoC: rt5645: Fix mask for setting RT5645_DMIC_2_DP_GPIO12 bit Current code uses wrong mask when setting RT5645_DMIC_2_DP_GPIO12 bit, fix it. Signed-off-by: Axel Lin Signed-off-by: Mark Brown diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 61ca49f..be4d741 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -2780,7 +2780,7 @@ static int rt5645_i2c_probe(struct i2c_client *i2c, case RT5645_DMIC_DATA_GPIO12: regmap_update_bits(rt5645->regmap, RT5645_DMIC_CTRL1, - RT5645_DMIC_1_DP_MASK, RT5645_DMIC_2_DP_GPIO12); + RT5645_DMIC_2_DP_MASK, RT5645_DMIC_2_DP_GPIO12); regmap_update_bits(rt5645->regmap, RT5645_GPIO_CTRL1, RT5645_GP12_PIN_MASK, RT5645_GP12_PIN_DMIC2_SDA); -- cgit v0.10.2 From 736a69ca8c99a595a523d2fece66491b89168da6 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 7 May 2015 11:16:26 +0100 Subject: drm/i915: Drop PIPE-A quirk for 945GSE HP Mini Since the introduction of BIOS fb preservation, circa 3.17, we began encountering a failure during boot when trying to use force-detect before GEM was initialised. That bug is from commit 7fad798e16fecddd41c6a91728a09f0b9507e40c Author: Daniel Vetter Date: Wed Jul 4 17:51:47 2012 +0200 drm/i915: ensure the force pipe A quirk is actually followed but investigation of the affected machine revealed that it was using a PIPE-A quirk even though it was a 945GSE and the quirk is only supposed to be used to workaround a hardware issue on 830/845. That quirk was added for this HP Mini in commit 6b93afc564a5e74b0eaaa46c95f557449951b3b9 Author: Bryce Harrington Date: Wed May 27 03:40:52 2009 -0700 add pipe a force quirk for Dell mini in order to workaround an issue with the BIOS behaving strangely during lid-close. Since then we have a much larger hammer to thwart the BIOS after opening the lid and the PIPE-A quirk is no longer required. Reported-and-tested-by: Apostolos B. References: https://bugs.freedesktop.org/show_bug.cgi?id=21960 Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=87521 Signed-off-by: Chris Wilson Signed-off-by: Jani Nikula diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index d547d9c8..d0f3cbc 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -13635,9 +13635,6 @@ static const struct intel_dmi_quirk intel_dmi_quirks[] = { }; static struct intel_quirk intel_quirks[] = { - /* HP Mini needs pipe A force quirk (LP: #322104) */ - { 0x27ae, 0x103c, 0x361a, quirk_pipea_force }, - /* Toshiba Protege R-205, S-209 needs pipe A force quirk */ { 0x2592, 0x1179, 0x0001, quirk_pipea_force }, -- cgit v0.10.2 From 1e5ec956a057585adaa1365615c82810b2f5356f Mon Sep 17 00:00:00 2001 From: Oded Gabbay Date: Tue, 14 Apr 2015 14:13:18 +0300 Subject: drm/amdkfd: allow unregister process with queues Sometimes we might unregister process that have queues, because we couldn't preempt the queues. Until now we blocked it with BUG_ON but instead just print it as debug. Reviewed-by: Ben Goz Signed-off-by: Oded Gabbay Cc: stable@vger.kernel.org Reviewed-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 69af73f..7b1d510 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -430,9 +430,10 @@ static int unregister_process_nocpsch(struct device_queue_manager *dqm, BUG_ON(!dqm || !qpd); - BUG_ON(!list_empty(&qpd->queues_list)); + pr_debug("In func %s\n", __func__); - pr_debug("kfd: In func %s\n", __func__); + pr_debug("qpd->queues_list is %s\n", + list_empty(&qpd->queues_list) ? "empty" : "not empty"); retval = 0; mutex_lock(&dqm->lock); -- cgit v0.10.2 From 42e08c78360e58516b6ac8af18a75a494f2967a2 Mon Sep 17 00:00:00 2001 From: Oded Gabbay Date: Tue, 5 May 2015 11:15:07 +0300 Subject: drm/amdkfd: Don't report local memory size This patch sets the local memory size that is reported to userspace to 0. This is done to make sure that userspace won't try to allocate local memory for HSA. As long as amdkfd doesn't support allocating local memory for HSA, we need this patch. Signed-off-by: Oded Gabbay Cc: stable@vger.kernel.org Reviewed-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 661c660..e469c4b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -728,9 +728,9 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, sysfs_show_32bit_prop(buffer, "max_engine_clk_fcompute", dev->gpu->kfd2kgd->get_max_engine_clock_in_mhz( dev->gpu->kgd)); + sysfs_show_64bit_prop(buffer, "local_mem_size", - dev->gpu->kfd2kgd->get_vmem_size( - dev->gpu->kgd)); + (unsigned long long int) 0); sysfs_show_32bit_prop(buffer, "fw_version", dev->gpu->kfd2kgd->get_fw_version( -- cgit v0.10.2 From 79b066bd76d501cfe8328142153da301f5ca11d1 Mon Sep 17 00:00:00 2001 From: Xihan Zhang Date: Tue, 28 Apr 2015 23:48:40 +0800 Subject: drm/amdkfd: Initialize sdma vm when creating sdma queue This patch fixes a bug where sdma vm wasn't initialized when an sdma queue was created in HWS mode. This caused GPUVM faults to appear on dmesg and it is one of the causes that SDMA queues are not working. Signed-off-by: Xihan Zhang Reviewed-by: Ben Goz Signed-off-by: Oded Gabbay Cc: stable@vger.kernel.org Reviewed-by: Alex Deucher diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 7b1d510..596ee5c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -883,6 +883,8 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, return -ENOMEM; } + init_sdma_vm(dqm, q, qpd); + retval = mqd->init_mqd(mqd, &q->mqd, &q->mqd_mem_obj, &q->gart_mqd_addr, &q->properties); if (retval != 0) -- cgit v0.10.2 From db12973cd581d4e79f4aadd0960948f268d15af7 Mon Sep 17 00:00:00 2001 From: "monk.liu" Date: Tue, 5 May 2015 09:24:17 +0200 Subject: drm/radeon: fix userptr BO unpin bug v3 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixing a memory leak with userptrs. v2: clean up the loop, use an iterator instead v3: remove unused variable Signed-off-by: monk.liu Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index b292aca..edafd3c 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -591,8 +591,7 @@ static void radeon_ttm_tt_unpin_userptr(struct ttm_tt *ttm) { struct radeon_device *rdev = radeon_get_rdev(ttm->bdev); struct radeon_ttm_tt *gtt = (void *)ttm; - struct scatterlist *sg; - int i; + struct sg_page_iter sg_iter; int write = !(gtt->userflags & RADEON_GEM_USERPTR_READONLY); enum dma_data_direction direction = write ? @@ -605,9 +604,8 @@ static void radeon_ttm_tt_unpin_userptr(struct ttm_tt *ttm) /* free the sg table and pages again */ dma_unmap_sg(rdev->dev, ttm->sg->sgl, ttm->sg->nents, direction); - for_each_sg(ttm->sg->sgl, sg, ttm->sg->nents, i) { - struct page *page = sg_page(sg); - + for_each_sg_page(ttm->sg->sgl, &sg_iter, ttm->sg->nents, 0) { + struct page *page = sg_page_iter_page(&sg_iter); if (!(gtt->userflags & RADEON_GEM_USERPTR_READONLY)) set_page_dirty(page); -- cgit v0.10.2 From 247c405098ab731ad9b58971e2cfbab116b54b45 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Tue, 5 May 2015 09:52:12 +0200 Subject: drm/radeon: fix userptr lockup MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We shouldn't try to reserve and wait for a BO that isn't bound. Otherwise we can run into a deadlock if we have a fault during binding the BO. Signed-off-by: Christian König Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_mn.c b/drivers/gpu/drm/radeon/radeon_mn.c index 535bf40..eef006c 100644 --- a/drivers/gpu/drm/radeon/radeon_mn.c +++ b/drivers/gpu/drm/radeon/radeon_mn.c @@ -142,6 +142,9 @@ static void radeon_mn_invalidate_range_start(struct mmu_notifier *mn, list_for_each_entry(bo, &node->bos, mn_list) { + if (!bo->tbo.ttm || bo->tbo.ttm->state != tt_bound) + continue; + r = radeon_bo_reserve(bo, true); if (r) { DRM_ERROR("(%ld) failed to reserve user bo\n", r); -- cgit v0.10.2 From 29c63fe22a17c64e54016040cd882481bd45ee5a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 7 May 2015 15:19:22 +0200 Subject: drm/radeon: make VCE handle check more strict MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Invalid handles can crash the hw. Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 24f849f..0de5711 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -493,18 +493,27 @@ int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi, * * @p: parser context * @handle: handle to validate + * @allocated: allocated a new handle? * * Validates the handle and return the found session index or -EINVAL * we we don't have another free session index. */ -int radeon_vce_validate_handle(struct radeon_cs_parser *p, uint32_t handle) +static int radeon_vce_validate_handle(struct radeon_cs_parser *p, + uint32_t handle, bool *allocated) { unsigned i; + *allocated = false; + /* validate the handle */ for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i) { - if (atomic_read(&p->rdev->vce.handles[i]) == handle) + if (atomic_read(&p->rdev->vce.handles[i]) == handle) { + if (p->rdev->vce.filp[i] != p->filp) { + DRM_ERROR("VCE handle collision detected!\n"); + return -EINVAL; + } return i; + } } /* handle not found try to alloc a new one */ @@ -512,6 +521,7 @@ int radeon_vce_validate_handle(struct radeon_cs_parser *p, uint32_t handle) if (!atomic_cmpxchg(&p->rdev->vce.handles[i], 0, handle)) { p->rdev->vce.filp[i] = p->filp; p->rdev->vce.img_size[i] = 0; + *allocated = true; return i; } } @@ -529,10 +539,10 @@ int radeon_vce_validate_handle(struct radeon_cs_parser *p, uint32_t handle) int radeon_vce_cs_parse(struct radeon_cs_parser *p) { int session_idx = -1; - bool destroyed = false; + bool destroyed = false, created = false, allocated = false; uint32_t tmp, handle = 0; uint32_t *size = &tmp; - int i, r; + int i, r = 0; while (p->idx < p->chunk_ib->length_dw) { uint32_t len = radeon_get_ib_value(p, p->idx); @@ -540,18 +550,21 @@ int radeon_vce_cs_parse(struct radeon_cs_parser *p) if ((len < 8) || (len & 3)) { DRM_ERROR("invalid VCE command length (%d)!\n", len); - return -EINVAL; + r = -EINVAL; + goto out; } if (destroyed) { DRM_ERROR("No other command allowed after destroy!\n"); - return -EINVAL; + r = -EINVAL; + goto out; } switch (cmd) { case 0x00000001: // session handle = radeon_get_ib_value(p, p->idx + 2); - session_idx = radeon_vce_validate_handle(p, handle); + session_idx = radeon_vce_validate_handle(p, handle, + &allocated); if (session_idx < 0) return session_idx; size = &p->rdev->vce.img_size[session_idx]; @@ -561,6 +574,13 @@ int radeon_vce_cs_parse(struct radeon_cs_parser *p) break; case 0x01000001: // create + created = true; + if (!allocated) { + DRM_ERROR("Handle already in use!\n"); + r = -EINVAL; + goto out; + } + *size = radeon_get_ib_value(p, p->idx + 8) * radeon_get_ib_value(p, p->idx + 10) * 8 * 3 / 2; @@ -578,12 +598,12 @@ int radeon_vce_cs_parse(struct radeon_cs_parser *p) r = radeon_vce_cs_reloc(p, p->idx + 10, p->idx + 9, *size); if (r) - return r; + goto out; r = radeon_vce_cs_reloc(p, p->idx + 12, p->idx + 11, *size / 3); if (r) - return r; + goto out; break; case 0x02000001: // destroy @@ -594,7 +614,7 @@ int radeon_vce_cs_parse(struct radeon_cs_parser *p) r = radeon_vce_cs_reloc(p, p->idx + 3, p->idx + 2, *size * 2); if (r) - return r; + goto out; break; case 0x05000004: // video bitstream buffer @@ -602,36 +622,47 @@ int radeon_vce_cs_parse(struct radeon_cs_parser *p) r = radeon_vce_cs_reloc(p, p->idx + 3, p->idx + 2, tmp); if (r) - return r; + goto out; break; case 0x05000005: // feedback buffer r = radeon_vce_cs_reloc(p, p->idx + 3, p->idx + 2, 4096); if (r) - return r; + goto out; break; default: DRM_ERROR("invalid VCE command (0x%x)!\n", cmd); - return -EINVAL; + r = -EINVAL; + goto out; } if (session_idx == -1) { DRM_ERROR("no session command at start of IB\n"); - return -EINVAL; + r = -EINVAL; + goto out; } p->idx += len / 4; } - if (destroyed) { - /* IB contains a destroy msg, free the handle */ + if (allocated && !created) { + DRM_ERROR("New session without create command!\n"); + r = -ENOENT; + } + +out: + if ((!r && destroyed) || (r && allocated)) { + /* + * IB contains a destroy msg or we have allocated an + * handle and got an error, anyway free the handle + */ for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i) atomic_cmpxchg(&p->rdev->vce.handles[i], handle, 0); } - return 0; + return r; } /** -- cgit v0.10.2 From a1b403da70e038ca6c6c6fe434d1d873546873a3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 7 May 2015 15:19:23 +0200 Subject: drm/radeon: make UVD handle checking more strict MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Invalid messages can crash the hw otherwise. Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index c10b2ae..f67a6aa 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -436,50 +436,64 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo, return -EINVAL; } - if (msg_type == 1) { + switch (msg_type) { + case 0: + /* it's a create msg, calc image size (width * height) */ + img_size = msg[7] * msg[8]; + radeon_bo_kunmap(bo); + + /* try to alloc a new handle */ + for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) { + if (atomic_read(&p->rdev->uvd.handles[i]) == handle) { + DRM_ERROR("Handle 0x%x already in use!\n", handle); + return -EINVAL; + } + + if (!atomic_cmpxchg(&p->rdev->uvd.handles[i], 0, handle)) { + p->rdev->uvd.filp[i] = p->filp; + p->rdev->uvd.img_size[i] = img_size; + return 0; + } + } + + DRM_ERROR("No more free UVD handles!\n"); + return -EINVAL; + + case 1: /* it's a decode msg, calc buffer sizes */ r = radeon_uvd_cs_msg_decode(msg, buf_sizes); - /* calc image size (width * height) */ - img_size = msg[6] * msg[7]; radeon_bo_kunmap(bo); if (r) return r; - } else if (msg_type == 2) { + /* validate the handle */ + for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) { + if (atomic_read(&p->rdev->uvd.handles[i]) == handle) { + if (p->rdev->uvd.filp[i] != p->filp) { + DRM_ERROR("UVD handle collision detected!\n"); + return -EINVAL; + } + return 0; + } + } + + DRM_ERROR("Invalid UVD handle 0x%x!\n", handle); + return -ENOENT; + + case 2: /* it's a destroy msg, free the handle */ for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) atomic_cmpxchg(&p->rdev->uvd.handles[i], handle, 0); radeon_bo_kunmap(bo); return 0; - } else { - /* it's a create msg, calc image size (width * height) */ - img_size = msg[7] * msg[8]; - radeon_bo_kunmap(bo); - if (msg_type != 0) { - DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); - return -EINVAL; - } - - /* it's a create msg, no special handling needed */ - } - - /* create or decode, validate the handle */ - for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) { - if (atomic_read(&p->rdev->uvd.handles[i]) == handle) - return 0; - } + default: - /* handle not found try to alloc a new one */ - for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) { - if (!atomic_cmpxchg(&p->rdev->uvd.handles[i], 0, handle)) { - p->rdev->uvd.filp[i] = p->filp; - p->rdev->uvd.img_size[i] = img_size; - return 0; - } + DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); + return -EINVAL; } - DRM_ERROR("No more free UVD handles!\n"); + BUG(); return -EINVAL; } -- cgit v0.10.2 From d52cdfa4a0c6406bbfb33206341eaf1fb1555994 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 7 May 2015 15:19:24 +0200 Subject: drm/radeon: more strictly validate the UVD codec MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit MPEG 2/4 are only supported since UVD3. Signed-off-by: Christian König CC: stable@vger.kernel.org Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index f67a6aa..cd63028 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -396,6 +396,29 @@ static int radeon_uvd_cs_msg_decode(uint32_t *msg, unsigned buf_sizes[]) return 0; } +static int radeon_uvd_validate_codec(struct radeon_cs_parser *p, + unsigned stream_type) +{ + switch (stream_type) { + case 0: /* H264 */ + case 1: /* VC1 */ + /* always supported */ + return 0; + + case 3: /* MPEG2 */ + case 4: /* MPEG4 */ + /* only since UVD 3 */ + if (p->rdev->family >= CHIP_PALM) + return 0; + + /* fall through */ + default: + DRM_ERROR("UVD codec not supported by hardware %d!\n", + stream_type); + return -EINVAL; + } +} + static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo, unsigned offset, unsigned buf_sizes[]) { @@ -440,7 +463,11 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo, case 0: /* it's a create msg, calc image size (width * height) */ img_size = msg[7] * msg[8]; + + r = radeon_uvd_validate_codec(p, msg[4]); radeon_bo_kunmap(bo); + if (r) + return r; /* try to alloc a new handle */ for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) { @@ -460,8 +487,10 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo, return -EINVAL; case 1: - /* it's a decode msg, calc buffer sizes */ - r = radeon_uvd_cs_msg_decode(msg, buf_sizes); + /* it's a decode msg, validate codec and calc buffer sizes */ + r = radeon_uvd_validate_codec(p, msg[4]); + if (!r) + r = radeon_uvd_cs_msg_decode(msg, buf_sizes); radeon_bo_kunmap(bo); if (r) return r; -- cgit v0.10.2 From 12e49feadff6d7b7ebbe852b36943a71524d8d34 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Thu, 7 May 2015 15:19:25 +0200 Subject: drm/radeon: stop trying to suspend UVD sessions MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Saving the current UVD state on suspend and restoring it on resume just doesn't work reliable. Just close cleanup all sessions on suspend. Signed-off-by: Christian König Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index d2abe48..46eb0fa 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -1673,7 +1673,6 @@ struct radeon_uvd { struct radeon_bo *vcpu_bo; void *cpu_addr; uint64_t gpu_addr; - void *saved_bo; atomic_t handles[RADEON_MAX_UVD_HANDLES]; struct drm_file *filp[RADEON_MAX_UVD_HANDLES]; unsigned img_size[RADEON_MAX_UVD_HANDLES]; diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index cd63028..6edcb54 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -204,28 +204,32 @@ void radeon_uvd_fini(struct radeon_device *rdev) int radeon_uvd_suspend(struct radeon_device *rdev) { - unsigned size; - void *ptr; - int i; + int i, r; if (rdev->uvd.vcpu_bo == NULL) return 0; - for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) - if (atomic_read(&rdev->uvd.handles[i])) - break; + for (i = 0; i < RADEON_MAX_UVD_HANDLES; ++i) { + uint32_t handle = atomic_read(&rdev->uvd.handles[i]); + if (handle != 0) { + struct radeon_fence *fence; - if (i == RADEON_MAX_UVD_HANDLES) - return 0; + radeon_uvd_note_usage(rdev); - size = radeon_bo_size(rdev->uvd.vcpu_bo); - size -= rdev->uvd_fw->size; + r = radeon_uvd_get_destroy_msg(rdev, + R600_RING_TYPE_UVD_INDEX, handle, &fence); + if (r) { + DRM_ERROR("Error destroying UVD (%d)!\n", r); + continue; + } - ptr = rdev->uvd.cpu_addr; - ptr += rdev->uvd_fw->size; + radeon_fence_wait(fence, false); + radeon_fence_unref(&fence); - rdev->uvd.saved_bo = kmalloc(size, GFP_KERNEL); - memcpy(rdev->uvd.saved_bo, ptr, size); + rdev->uvd.filp[i] = NULL; + atomic_set(&rdev->uvd.handles[i], 0); + } + } return 0; } @@ -246,12 +250,7 @@ int radeon_uvd_resume(struct radeon_device *rdev) ptr = rdev->uvd.cpu_addr; ptr += rdev->uvd_fw->size; - if (rdev->uvd.saved_bo != NULL) { - memcpy(ptr, rdev->uvd.saved_bo, size); - kfree(rdev->uvd.saved_bo); - rdev->uvd.saved_bo = NULL; - } else - memset(ptr, 0, size); + memset(ptr, 0, size); return 0; } -- cgit v0.10.2 From 454be2af5b49612e7f20ceb6683d5809ce848bee Mon Sep 17 00:00:00 2001 From: Mark Salter Date: Tue, 28 Apr 2015 13:09:32 -0400 Subject: drivers: CCI: fix used_mask init in validate_group() Currently in validate_group(), there is a static initializer for fake_pmu.used_mask which is based on CPU_BITS_NONE but the used_mask array size is based on CCI_PMU_MAX_HW_EVENTS. CCI_PMU_MAX_HW_EVENTS is not based on NR_CPUS, so CPU_BITS_NONE is not correct and will cause a build failure if NR_CPUS is set high enough to make CPU_BITS_NONE larger than used_mask. Reviewed-by: Mark Rutland Signed-off-by: Mark Salter Signed-off-by: Arnd Bergmann diff --git a/drivers/bus/arm-cci.c b/drivers/bus/arm-cci.c index b854125..5340604 100644 --- a/drivers/bus/arm-cci.c +++ b/drivers/bus/arm-cci.c @@ -660,7 +660,7 @@ validate_group(struct perf_event *event) * Initialise the fake PMU. We only need to populate the * used_mask for the purposes of validation. */ - .used_mask = CPU_BITS_NONE, + .used_mask = { 0 }, }; if (!validate_event(event->pmu, &fake_pmu, leader)) -- cgit v0.10.2 From d68b35f845cc56472fa0d782449ff1549f00ec47 Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Wed, 29 Apr 2015 11:57:18 +0200 Subject: MAINTAINERS: replace an AT91 maintainer As some help is needed from an active maintainer, replace Andrew Victor by Alexandre Belloni in the ARM/Atmel MAINTAINERS' entry (aka AT91). Add an entry to the CREDITS file. Thanks Andrew for the great role you played during the early days of this product family. Signed-off-by: Nicolas Ferre Acked-by: Andrew Victor Acked-by: Alexandre Belloni Signed-off-by: Arnd Bergmann diff --git a/CREDITS b/CREDITS index 40cc4bf..ec7e6c7 100644 --- a/CREDITS +++ b/CREDITS @@ -3709,6 +3709,13 @@ N: Dirk Verworner D: Co-author of German book ``Linux-Kernel-Programmierung'' D: Co-founder of Berlin Linux User Group +N: Andrew Victor +E: linux@maxim.org.za +W: http://maxim.org.za/at91_26.html +D: First maintainer of Atmel ARM-based SoC, aka AT91 +D: Introduced support for at91rm9200, the first chip of AT91 family +S: South Africa + N: Riku Voipio E: riku.voipio@iki.fi D: Author of PCA9532 LED and Fintek f75375s hwmon driver diff --git a/MAINTAINERS b/MAINTAINERS index f639c04..3ab4597 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -892,11 +892,10 @@ S: Maintained F: arch/arm/mach-alpine/ ARM/ATMEL AT91RM9200 AND AT91SAM ARM ARCHITECTURES -M: Andrew Victor M: Nicolas Ferre +M: Alexandre Belloni M: Jean-Christophe Plagniol-Villard L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) -W: http://maxim.org.za/at91_26.html W: http://www.linux4sam.org S: Supported F: arch/arm/mach-at91/ -- cgit v0.10.2 From dc0e3db4ec7c1e73e3283ad39b68eff32b7dc8a1 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Tue, 5 May 2015 14:35:49 -0300 Subject: ARM: multi_v7_defconfig: Select more FSL SoCs Select IMX50, IMX6SX and LS1021A SoC support. Signed-off-by: Fabio Estevam Signed-off-by: Arnd Bergmann diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig index ab86655..0ca4a3e 100644 --- a/arch/arm/configs/multi_v7_defconfig +++ b/arch/arm/configs/multi_v7_defconfig @@ -39,11 +39,14 @@ CONFIG_ARCH_HIP04=y CONFIG_ARCH_KEYSTONE=y CONFIG_ARCH_MESON=y CONFIG_ARCH_MXC=y +CONFIG_SOC_IMX50=y CONFIG_SOC_IMX51=y CONFIG_SOC_IMX53=y CONFIG_SOC_IMX6Q=y CONFIG_SOC_IMX6SL=y +CONFIG_SOC_IMX6SX=y CONFIG_SOC_VF610=y +CONFIG_SOC_LS1021A=y CONFIG_ARCH_OMAP3=y CONFIG_ARCH_OMAP4=y CONFIG_SOC_OMAP5=y -- cgit v0.10.2 From 4b2447248325ac7b1654d362c0e9f050f79e0ffe Mon Sep 17 00:00:00 2001 From: Zhangfei Gao Date: Thu, 30 Apr 2015 22:16:28 +0800 Subject: mmc: dw_mmc: init desc in dw_mci_idmac_init Set 0 to des1 in 32bit case. Otherwise the random value of des1 will be used in dw_mci_translate_sglist: IDMAC_SET_BUFFER1_SIZE(desc, length) Signed-off-by: Fei Wang Signed-off-by: Zhangfei Gao Signed-off-by: Jaehoon Chung Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c index 38b2926..69952b2 100644 --- a/drivers/mmc/host/dw_mmc.c +++ b/drivers/mmc/host/dw_mmc.c @@ -589,9 +589,11 @@ static int dw_mci_idmac_init(struct dw_mci *host) host->ring_size = PAGE_SIZE / sizeof(struct idmac_desc); /* Forward link the descriptor list */ - for (i = 0, p = host->sg_cpu; i < host->ring_size - 1; i++, p++) + for (i = 0, p = host->sg_cpu; i < host->ring_size - 1; i++, p++) { p->des3 = cpu_to_le32(host->sg_dma + (sizeof(struct idmac_desc) * (i + 1))); + p->des1 = 0; + } /* Set the last descriptor as the end-of-ring descriptor */ p->des3 = cpu_to_le32(host->sg_dma); -- cgit v0.10.2 From 4de3bf66c61ef708d8f22d7f990339668a858e7d Mon Sep 17 00:00:00 2001 From: Zhangfei Gao Date: Tue, 5 May 2015 16:54:49 +0800 Subject: mmc: dw_mmc: dw_mci_get_cd check MMC_CAP_NONREMOVABLE When non-removable is used for emmc, MMC_CAP_NONREMOVABLE should also be checked, otherwise detection fail since present=0 Signed-off-by: Zhangfei Gao Signed-off-by: Jaehoon Chung Signed-off-by: Ulf Hansson diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c index 69952b2..5f5adaf 100644 --- a/drivers/mmc/host/dw_mmc.c +++ b/drivers/mmc/host/dw_mmc.c @@ -1302,7 +1302,8 @@ static int dw_mci_get_cd(struct mmc_host *mmc) int gpio_cd = mmc_gpio_get_cd(mmc); /* Use platform get_cd function, else try onboard card detect */ - if (brd->quirks & DW_MCI_QUIRK_BROKEN_CARD_DETECTION) + if ((brd->quirks & DW_MCI_QUIRK_BROKEN_CARD_DETECTION) || + (mmc->caps & MMC_CAP_NONREMOVABLE)) present = 1; else if (!IS_ERR_VALUE(gpio_cd)) present = gpio_cd; -- cgit v0.10.2 From 3790e395b8f4b66fe4e53629f304505c110a2be7 Mon Sep 17 00:00:00 2001 From: Mario Kleiner Date: Mon, 4 May 2015 06:29:44 +0200 Subject: drm/tegra: Don't use vblank_disable_immediate on incapable driver. Tegra would not only need a hardware vblank counter that increments at leading edge of vblank, but also support for instantaneous high precision vblank timestamp queries, ie. a proper implementation of dev->driver->get_vblank_timestamp(). Without these, there can be off-by-one errors during vblank disable/enable if the scanout is inside vblank at en/disable time, and additionally clients will never see any useable vblank timestamps when querying via drmWaitVblank ioctl. This would negatively affect swap scheduling under X11 and Wayland. Signed-off-by: Mario Kleiner Acked-by: Thierry Reding Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index 1833abd..bfad15a 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -173,7 +173,6 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags) drm->irq_enabled = true; /* syncpoints are used for full 32-bit hardware VBLANK counters */ - drm->vblank_disable_immediate = true; drm->max_vblank_count = 0xffffffff; err = drm_vblank_init(drm, drm->mode_config.num_crtc); -- cgit v0.10.2 From efadb751692f3c10e72cf96c8409e40e60852ff2 Mon Sep 17 00:00:00 2001 From: Dinh Nguyen Date: Mon, 20 Apr 2015 14:13:12 -0500 Subject: MAINTAINERS: socfpga: update the git repo for SoCFPGA The git tree at rocketboards.org is going away. Update the entry to reflect the address of the new location. Also add an entry for all the socfpga_* dts files. Signed-off-by: Dinh Nguyen Signed-off-by: Arnd Bergmann diff --git a/MAINTAINERS b/MAINTAINERS index 3ab4597..948b493 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1438,9 +1438,10 @@ ARM/SOCFPGA ARCHITECTURE M: Dinh Nguyen S: Maintained F: arch/arm/mach-socfpga/ +F: arch/arm/boot/dts/socfpga* +F: arch/arm/configs/socfpga_defconfig W: http://www.rocketboards.org -T: git://git.rocketboards.org/linux-socfpga.git -T: git://git.rocketboards.org/linux-socfpga-next.git +T: git git://git.kernel.org/pub/scm/linux/kernel/git/dinguyen/linux.git ARM/SOCFPGA CLOCK FRAMEWORK SUPPORT M: Dinh Nguyen -- cgit v0.10.2 From c9d862c48c48883a4327ae82f4a6de1eef928d60 Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Tue, 28 Apr 2015 13:59:43 +0300 Subject: MAINTAINERS: add Conexant Digicolor machines entry This adds Baruch as the maintainer for the Digicolor platform. Signed-off-by: Baruch Siach Signed-off-by: Arnd Bergmann diff --git a/MAINTAINERS b/MAINTAINERS index 948b493..9d39c38 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -989,6 +989,12 @@ F: drivers/clocksource/timer-prima2.c F: drivers/clocksource/timer-atlas7.c N: [^a-z]sirf +ARM/CONEXANT DIGICOLOR MACHINE SUPPORT +M: Baruch Siach +L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) +S: Maintained +N: digicolor + ARM/EBSA110 MACHINE SUPPORT M: Russell King L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) -- cgit v0.10.2 From 0d747762966e47e3660cbd1b7806791c83e04f91 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Sat, 9 May 2015 02:33:42 +0900 Subject: ARM: dts: Make DP a consumer of DISP1 power domain on Exynos5420 Commit ea08de16eb1b ("ARM: dts: Add DISP1 power domain for exynos5420") added a device node for the Exynos5420 DISP1 power domain but dit not make the DP controller a consumer of that power domain. This causes an "Unhandled fault: imprecise external abort" error if the exynos-dp driver tries to access the DP controller registers and the PD was turned off. This lead to a kernel panic and a complete system hang. Make the DP controller device node a consumer of the DISP1 power domain to ensure that the PD is turned on when the exynos-dp driver is probed. Fixes: ea08de16eb1b ("ARM: dts: Add DISP1 power domain for exynos5420") Signed-off-by: Javier Martinez Canillas Signed-off-by: Kukjin Kim diff --git a/arch/arm/boot/dts/exynos5420.dtsi b/arch/arm/boot/dts/exynos5420.dtsi index f67b23f..4531753 100644 --- a/arch/arm/boot/dts/exynos5420.dtsi +++ b/arch/arm/boot/dts/exynos5420.dtsi @@ -536,6 +536,7 @@ clock-names = "dp"; phys = <&dp_phy>; phy-names = "dp"; + power-domains = <&disp_pd>; }; mipi_phy: video-phy@10040714 { -- cgit v0.10.2 From ee2020a4ca847e2cde4d5ee73572388061730809 Mon Sep 17 00:00:00 2001 From: Markus Reichl Date: Sat, 9 May 2015 03:05:51 +0900 Subject: ARM: dts: add 'rtc_src' clock to rtc node for exynos4412-odroid boards The Exynos4412 SoC has a s3c6410 RTC where the source clock is now a mandatory property. This patch fixes probe failure of s3c-rtc on Odroid-X2/U2/U3 boards. Signed-off-by: Markus Reichl Tested-by: Tobias Jakobi Reviewed-by: Chanwoo Choi Reviewed-by: Javier Martinez Canillas Tested-by: Krzysztof Kozlowski Signed-off-by: Kukjin Kim diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi index 8de12af7..d6b49e5 100644 --- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi +++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi @@ -9,6 +9,7 @@ #include #include +#include #include "exynos4412.dtsi" / { @@ -105,6 +106,8 @@ rtc@10070000 { status = "okay"; + clocks = <&clock CLK_RTC>, <&max77686 MAX77686_CLK_AP>; + clock-names = "rtc", "rtc_src"; }; g2d@10800000 { -- cgit v0.10.2 From cb0f7c8b40c52124fdb73b5e5a1b28202efbb43b Mon Sep 17 00:00:00 2001 From: Abhilash Kesavan Date: Sat, 9 May 2015 03:11:21 +0900 Subject: ARM: dts: Fix typo in trip point temperature for exynos5420/5440 Remove the extra zero in the "cpu-crit-0" trip point for exynos5420 and exynos5440. Signed-off-by: Abhilash Kesavan Signed-off-by: Kukjin Kim diff --git a/arch/arm/boot/dts/exynos5420-trip-points.dtsi b/arch/arm/boot/dts/exynos5420-trip-points.dtsi index 5d31fc1..2180a01 100644 --- a/arch/arm/boot/dts/exynos5420-trip-points.dtsi +++ b/arch/arm/boot/dts/exynos5420-trip-points.dtsi @@ -28,7 +28,7 @@ trips { type = "active"; }; cpu-crit-0 { - temperature = <1200000>; /* millicelsius */ + temperature = <120000>; /* millicelsius */ hysteresis = <0>; /* millicelsius */ type = "critical"; }; diff --git a/arch/arm/boot/dts/exynos5440-trip-points.dtsi b/arch/arm/boot/dts/exynos5440-trip-points.dtsi index 48adfa8..356e963 100644 --- a/arch/arm/boot/dts/exynos5440-trip-points.dtsi +++ b/arch/arm/boot/dts/exynos5440-trip-points.dtsi @@ -18,7 +18,7 @@ trips { type = "active"; }; cpu-crit-0 { - temperature = <1050000>; /* millicelsius */ + temperature = <105000>; /* millicelsius */ hysteresis = <0>; /* millicelsius */ type = "critical"; }; -- cgit v0.10.2 From 8cf5e6dc8dd55d0f1ad46ab4046c3a8a51d2136d Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Sat, 9 May 2015 03:15:16 +0900 Subject: ARM: dts: Add keep-power-in-suspend to WiFi SDIO node for exynos5250-snow The Marvell mwifiex driver prevents the system to enter into a suspend state if the card power is not preserved during a suspend/resume cycle. So Suspend-to-RAM and Suspend-to-idle are failing on Exynos5250 Snow. Add the keep-power-in-suspend Power Management property to the SDIO/MMC node so the mwifiex suspend handler doesn't fail and the system is able to enter into a suspend state. Signed-off-by: Javier Martinez Canillas Reviewed-by: Doug Anderson Signed-off-by: Kukjin Kim diff --git a/arch/arm/boot/dts/exynos5250-snow.dts b/arch/arm/boot/dts/exynos5250-snow.dts index 2657e84..1eca97e 100644 --- a/arch/arm/boot/dts/exynos5250-snow.dts +++ b/arch/arm/boot/dts/exynos5250-snow.dts @@ -567,6 +567,7 @@ num-slots = <1>; broken-cd; cap-sdio-irq; + keep-power-in-suspend; card-detect-delay = <200>; samsung,dw-mshc-ciu-div = <3>; samsung,dw-mshc-sdr-timing = <2 3>; -- cgit v0.10.2 From 766c4cbfacd8634d7580bac6a1b8456e63de3e84 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 7 May 2015 19:24:57 -0400 Subject: namei: d_is_negative() should be checked before ->d_seq validation Fetching ->d_inode, verifying ->d_seq and finding d_is_negative() to be true does *not* mean that inode we'd fetched had been NULL - that holds only while ->d_seq is still unchanged. Shift d_is_negative() checks into lookup_fast() prior to ->d_seq verification. Reported-by: Steven Rostedt Tested-by: Steven Rostedt Signed-off-by: Al Viro diff --git a/fs/namei.c b/fs/namei.c index 4a8d998b..f67cf6c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1415,6 +1415,7 @@ static int lookup_fast(struct nameidata *nd, */ if (nd->flags & LOOKUP_RCU) { unsigned seq; + bool negative; dentry = __d_lookup_rcu(parent, &nd->last, &seq); if (!dentry) goto unlazy; @@ -1424,8 +1425,11 @@ static int lookup_fast(struct nameidata *nd, * the dentry name information from lookup. */ *inode = dentry->d_inode; + negative = d_is_negative(dentry); if (read_seqcount_retry(&dentry->d_seq, seq)) return -ECHILD; + if (negative) + return -ENOENT; /* * This sequence count validates that the parent had no @@ -1472,6 +1476,10 @@ unlazy: goto need_lookup; } + if (unlikely(d_is_negative(dentry))) { + dput(dentry); + return -ENOENT; + } path->mnt = mnt; path->dentry = dentry; err = follow_managed(path, nd->flags); @@ -1583,10 +1591,10 @@ static inline int walk_component(struct nameidata *nd, struct path *path, goto out_err; inode = path->dentry->d_inode; + err = -ENOENT; + if (d_is_negative(path->dentry)) + goto out_path_put; } - err = -ENOENT; - if (d_is_negative(path->dentry)) - goto out_path_put; if (should_follow_link(path->dentry, follow)) { if (nd->flags & LOOKUP_RCU) { @@ -3036,14 +3044,13 @@ retry_lookup: BUG_ON(nd->flags & LOOKUP_RCU); inode = path->dentry->d_inode; -finish_lookup: - /* we _can_ be in RCU mode here */ error = -ENOENT; if (d_is_negative(path->dentry)) { path_to_nameidata(path, nd); goto out; } - +finish_lookup: + /* we _can_ be in RCU mode here */ if (should_follow_link(path->dentry, !symlink_ok)) { if (nd->flags & LOOKUP_RCU) { if (unlikely(nd->path.mnt != path->mnt || -- cgit v0.10.2 From f15133df088ecadd141ea1907f2c96df67c729f0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 8 May 2015 22:53:15 -0400 Subject: path_openat(): fix double fput() path_openat() jumps to the wrong place after do_tmpfile() - it has already done path_cleanup() (as part of path_lookupat() called by do_tmpfile()), so doing that again can lead to double fput(). Cc: stable@vger.kernel.org # v3.11+ Signed-off-by: Al Viro diff --git a/fs/namei.c b/fs/namei.c index f67cf6c..fe30d3b 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -3233,7 +3233,7 @@ static struct file *path_openat(int dfd, struct filename *pathname, if (unlikely(file->f_flags & __O_TMPFILE)) { error = do_tmpfile(dfd, pathname, nd, flags, op, file, &opened); - goto out; + goto out2; } error = path_init(dfd, pathname, flags, nd); @@ -3263,6 +3263,7 @@ static struct file *path_openat(int dfd, struct filename *pathname, } out: path_cleanup(nd); +out2: if (!(opened & FILE_OPENED)) { BUG_ON(!error); put_filp(file); -- cgit v0.10.2 From 7e96c1b0e0f495c5a7450dc4aa7c9a24ba4305bd Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Fri, 8 May 2015 16:36:50 -0500 Subject: mnt: Fix fs_fully_visible to verify the root directory is visible This fixes a dumb bug in fs_fully_visible that allows proc or sys to be mounted if there is a bind mount of part of /proc/ or /sys/ visible. Cc: stable@vger.kernel.org Reported-by: Eric Windisch Signed-off-by: "Eric W. Biederman" diff --git a/fs/namespace.c b/fs/namespace.c index 1f4f9da..1b9e111 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -3179,6 +3179,12 @@ bool fs_fully_visible(struct file_system_type *type) if (mnt->mnt.mnt_sb->s_type != type) continue; + /* This mount is not fully visible if it's root directory + * is not the root directory of the filesystem. + */ + if (mnt->mnt.mnt_root != mnt->mnt.mnt_sb->s_root) + continue; + /* This mount is not fully visible if there are any child mounts * that cover anything except for empty directories. */ -- cgit v0.10.2 From 1a9f064f516c36d018754407b41921c996439ecf Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 9 May 2015 03:39:52 +0930 Subject: m32r: make flush_cpumask non-volatile. We cast away the volatile, but really, why make it volatile at all? We already do a mb() inside the cpumask_empty() loop. Signed-off-by: Rusty Russell Signed-off-by: Linus Torvalds diff --git a/arch/m32r/kernel/smp.c b/arch/m32r/kernel/smp.c index ce7aea3..c18ddc7 100644 --- a/arch/m32r/kernel/smp.c +++ b/arch/m32r/kernel/smp.c @@ -45,7 +45,7 @@ static volatile unsigned long flushcache_cpumask = 0; /* * For flush_tlb_others() */ -static volatile cpumask_t flush_cpumask; +static cpumask_t flush_cpumask; static struct mm_struct *flush_mm; static struct vm_area_struct *flush_vma; static volatile unsigned long flush_va; @@ -415,7 +415,7 @@ static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm, */ send_IPI_mask(&cpumask, INVALIDATE_TLB_IPI, 0); - while (!cpumask_empty((cpumask_t*)&flush_cpumask)) { + while (!cpumask_empty(&flush_cpumask)) { /* nothing. lockup detection does not belong here */ mb(); } @@ -468,7 +468,7 @@ void smp_invalidate_interrupt(void) __flush_tlb_page(va); } } - cpumask_clear_cpu(cpu_id, (cpumask_t*)&flush_cpumask); + cpumask_clear_cpu(cpu_id, &flush_cpumask); } /*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*/ -- cgit v0.10.2 From fdb68e09bbb1c981f24608d7022c7d93cc47b326 Mon Sep 17 00:00:00 2001 From: Mario Kleiner Date: Tue, 7 Apr 2015 06:31:09 +0200 Subject: drm: Zero out invalid vblank timestamp in drm_update_vblank_count. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit 844b03f27739135fe1fed2fef06da0ffc4c7a081 we make sure that after vblank irq off, we return the last valid (vblank count, vblank timestamp) pair to clients, e.g., during modesets, which is good. An overlooked side effect of that commit for kms drivers without support for precise vblank timestamping is that at vblank irq enable, when we update the vblank counter from the hw counter, we can't update the corresponding vblank timestamp, so now we have a totally mismatched timestamp for the new count to confuse clients. Restore old client visible behaviour from before Linux 3.17, but zero out the timestamp at vblank counter update (instead of disable as in original implementation) if we can't generate a meaningful timestamp immediately for the new vblank counter. This will fix this regression, so callers know they need to retry again later if they need a valid timestamp, but at the same time preserves the improvements made in the commit mentioned above. Signed-off-by: Mario Kleiner Cc: #v3.17+ Cc: Ville Syrjälä Cc: Daniel Vetter Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index c8a3447..af9662e 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -131,12 +131,11 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc) /* Reinitialize corresponding vblank timestamp if high-precision query * available. Skip this step if query unsupported or failed. Will - * reinitialize delayed at next vblank interrupt in that case. + * reinitialize delayed at next vblank interrupt in that case and + * assign 0 for now, to mark the vblanktimestamp as invalid. */ - if (rc) { - tslot = atomic_read(&vblank->count) + diff; - vblanktimestamp(dev, crtc, tslot) = t_vblank; - } + tslot = atomic_read(&vblank->count) + diff; + vblanktimestamp(dev, crtc, tslot) = rc ? t_vblank : (struct timeval) {0, 0}; smp_mb__before_atomic(); atomic_add(diff, &vblank->count); -- cgit v0.10.2 From 030bbdbf4c833bc69f502eae58498bc5572db736 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 10 May 2015 15:12:29 -0700 Subject: Linux 4.1-rc3 diff --git a/Makefile b/Makefile index 2da553f..eae539d 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 1 SUBLEVEL = 0 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc3 NAME = Hurr durr I'ma sheep # *DOCUMENTATION* -- cgit v0.10.2 From 9b5a4e395c2f39fae89f75e4a749be5dba342d22 Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Mon, 11 May 2015 14:04:14 +0200 Subject: ALSA: hda - Add headset mic quirk for Dell Inspiron 5548 This enables the headset microphone on Dell Inspiron 5548, or at least some variants of it. Cc: stable@vger.kernel.org BugLink: https://bugs.launchpad.net/bugs/1452175 Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 4a0a238..37c2c9f 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5346,6 +5346,13 @@ static const struct snd_hda_pin_quirk alc269_pin_fixup_tbl[] = { {0x17, 0x40000000}, {0x1d, 0x40700001}, {0x21, 0x02211050}), + SND_HDA_PIN_QUIRK(0x10ec0255, 0x1028, "Dell Inspiron 5548", ALC255_FIXUP_DELL1_MIC_NO_PRESENCE, + ALC255_STANDARD_PINS, + {0x12, 0x90a60180}, + {0x14, 0x90170130}, + {0x17, 0x40000000}, + {0x1d, 0x40700001}, + {0x21, 0x02211040}), SND_HDA_PIN_QUIRK(0x10ec0256, 0x1028, "Dell", ALC255_FIXUP_DELL1_MIC_NO_PRESENCE, ALC256_STANDARD_PINS, {0x13, 0x40000000}), -- cgit v0.10.2 From fdb6eb0a12871d5bfaf266c5a0d5259a5437a72f Mon Sep 17 00:00:00 2001 From: Koro Chen Date: Mon, 11 May 2015 10:36:53 +0800 Subject: ASoC: dapm: Modify widget stream name according to prefix When there is prefix specified, currently we will add this prefix in widget->name, but not in widget->sname. it causes failure at snd_soc_dapm_link_dai_widgets: if (!w->sname || !strstr(w->sname, dai_w->name)) because dai_w->name has prefix added, but w->sname does not. We should also add prefix for stream name Signed-off-by: Koro Chen Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index defe0f0..158204d 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -3100,11 +3100,16 @@ snd_soc_dapm_new_control(struct snd_soc_dapm_context *dapm, } prefix = soc_dapm_prefix(dapm); - if (prefix) + if (prefix) { w->name = kasprintf(GFP_KERNEL, "%s %s", prefix, widget->name); - else + if (widget->sname) + w->sname = kasprintf(GFP_KERNEL, "%s %s", prefix, + widget->sname); + } else { w->name = kasprintf(GFP_KERNEL, "%s", widget->name); - + if (widget->sname) + w->sname = kasprintf(GFP_KERNEL, "%s", widget->sname); + } if (w->name == NULL) { kfree(w); return NULL; -- cgit v0.10.2 From 85e36a1f4a735d991ba5106781ea48e89a0b8901 Mon Sep 17 00:00:00 2001 From: Zidan Wang Date: Tue, 12 May 2015 14:58:36 +0800 Subject: ASoC: wm8960: fix "RINPUT3" audio route error It should be "RINPUT3" instead of "LINPUT3" route to "Right Input Mixer". Signed-off-by: Zidan Wang Acked-by: Charles Keepax Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/codecs/wm8960.c b/sound/soc/codecs/wm8960.c index 3035d98..e97a761 100644 --- a/sound/soc/codecs/wm8960.c +++ b/sound/soc/codecs/wm8960.c @@ -395,7 +395,7 @@ static const struct snd_soc_dapm_route audio_paths[] = { { "Right Input Mixer", "Boost Switch", "Right Boost Mixer", }, { "Right Input Mixer", NULL, "RINPUT1", }, /* Really Boost Switch */ { "Right Input Mixer", NULL, "RINPUT2" }, - { "Right Input Mixer", NULL, "LINPUT3" }, + { "Right Input Mixer", NULL, "RINPUT3" }, { "Left ADC", NULL, "Left Input Mixer" }, { "Right ADC", NULL, "Right Input Mixer" }, -- cgit v0.10.2 From 17fc2e0a3db11889e942c5ab15a1fcb876638f25 Mon Sep 17 00:00:00 2001 From: Zidan Wang Date: Tue, 12 May 2015 14:58:50 +0800 Subject: ASoC: wm8994: correct BCLK DIV 348 to 384 According to the RM of wm8958, BCLK DIV 348 doesn't exist, correct it to 384. Signed-off-by: Zidan Wang Acked-by: Charles Keepax Signed-off-by: Mark Brown Cc: stable@vger.kernel.org diff --git a/sound/soc/codecs/wm8994.c b/sound/soc/codecs/wm8994.c index 4fbc768..a1c04da 100644 --- a/sound/soc/codecs/wm8994.c +++ b/sound/soc/codecs/wm8994.c @@ -2754,7 +2754,7 @@ static struct { }; static int fs_ratios[] = { - 64, 128, 192, 256, 348, 512, 768, 1024, 1408, 1536 + 64, 128, 192, 256, 384, 512, 768, 1024, 1408, 1536 }; static int bclk_divs[] = { -- cgit v0.10.2 From 1f8b46cdc806dfd76386f8442d9a6d0ae69abb25 Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Tue, 12 May 2015 14:38:15 +0200 Subject: ALSA: hda - Fix headset mic and mic-in for a Dell desktop ALC662 does not need any special verbs to change the jack functionality, and enables mic in through the headphone jack mode by changing the direction of the headphone pin node. BugLink: https://bugs.launchpad.net/bugs/1454235 Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 37c2c9f..4202a80 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -3674,6 +3674,10 @@ static void alc_headset_mode_mic_in(struct hda_codec *codec, hda_nid_t hp_pin, alc_process_coef_fw(codec, coef0293); snd_hda_set_pin_ctl_cache(codec, mic_pin, PIN_VREF50); break; + case 0x10ec0662: + snd_hda_set_pin_ctl_cache(codec, hp_pin, 0); + snd_hda_set_pin_ctl_cache(codec, mic_pin, PIN_VREF50); + break; case 0x10ec0668: alc_write_coef_idx(codec, 0x11, 0x0001); snd_hda_set_pin_ctl_cache(codec, hp_pin, 0); @@ -4012,7 +4016,7 @@ static void alc_update_headset_mode(struct hda_codec *codec) if (new_headset_mode != ALC_HEADSET_MODE_MIC) { snd_hda_set_pin_ctl_cache(codec, hp_pin, AC_PINCTL_OUT_EN | AC_PINCTL_HP_EN); - if (spec->headphone_mic_pin) + if (spec->headphone_mic_pin && spec->headphone_mic_pin != hp_pin) snd_hda_set_pin_ctl_cache(codec, spec->headphone_mic_pin, PIN_VREFHIZ); } @@ -4215,6 +4219,18 @@ static void alc_fixup_dell_xps13(struct hda_codec *codec, } } +static void alc_fixup_headset_mode_alc662(struct hda_codec *codec, + const struct hda_fixup *fix, int action) +{ + struct alc_spec *spec = codec->spec; + + if (action == HDA_FIXUP_ACT_PRE_PROBE) { + spec->parse_flags |= HDA_PINCFG_HEADSET_MIC; + spec->gen.hp_mic = 1; /* Mic-in is same pin as headphone */ + } else + alc_fixup_headset_mode(codec, fix, action); +} + static void alc_fixup_headset_mode_alc668(struct hda_codec *codec, const struct hda_fixup *fix, int action) { @@ -6087,7 +6103,9 @@ enum { ALC662_FIXUP_NO_JACK_DETECT, ALC662_FIXUP_ZOTAC_Z68, ALC662_FIXUP_INV_DMIC, + ALC662_FIXUP_DELL_MIC_NO_PRESENCE, ALC668_FIXUP_DELL_MIC_NO_PRESENCE, + ALC662_FIXUP_HEADSET_MODE, ALC668_FIXUP_HEADSET_MODE, ALC662_FIXUP_BASS_MODE4_CHMAP, ALC662_FIXUP_BASS_16, @@ -6280,6 +6298,20 @@ static const struct hda_fixup alc662_fixups[] = { .chained = true, .chain_id = ALC668_FIXUP_DELL_MIC_NO_PRESENCE }, + [ALC662_FIXUP_DELL_MIC_NO_PRESENCE] = { + .type = HDA_FIXUP_PINS, + .v.pins = (const struct hda_pintbl[]) { + { 0x19, 0x03a1113c }, /* use as headset mic, without its own jack detect */ + /* headphone mic by setting pin control of 0x1b (headphone out) to in + vref_50 */ + { } + }, + .chained = true, + .chain_id = ALC662_FIXUP_HEADSET_MODE + }, + [ALC662_FIXUP_HEADSET_MODE] = { + .type = HDA_FIXUP_FUNC, + .v.func = alc_fixup_headset_mode_alc662, + }, [ALC668_FIXUP_DELL_MIC_NO_PRESENCE] = { .type = HDA_FIXUP_PINS, .v.pins = (const struct hda_pintbl[]) { @@ -6431,6 +6463,18 @@ static const struct hda_model_fixup alc662_fixup_models[] = { }; static const struct snd_hda_pin_quirk alc662_pin_fixup_tbl[] = { + SND_HDA_PIN_QUIRK(0x10ec0662, 0x1028, "Dell", ALC662_FIXUP_DELL_MIC_NO_PRESENCE, + {0x12, 0x4004c000}, + {0x14, 0x01014010}, + {0x15, 0x411111f0}, + {0x16, 0x411111f0}, + {0x18, 0x01a19020}, + {0x19, 0x411111f0}, + {0x1a, 0x0181302f}, + {0x1b, 0x0221401f}, + {0x1c, 0x411111f0}, + {0x1d, 0x4054c601}, + {0x1e, 0x411111f0}), SND_HDA_PIN_QUIRK(0x10ec0668, 0x1028, "Dell", ALC668_FIXUP_AUTO_MUTE, {0x12, 0x99a30130}, {0x14, 0x90170110}, -- cgit v0.10.2 From 6ffc0898b29a2811a6c0569c5dd9b581980110df Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Wed, 13 May 2015 13:28:54 +0200 Subject: ALSA: hda - Add Conexant codecs CX20721, CX20722, CX20723 and CX20724 This patch adds support for Conexant HD Audio codecs CX20721, CX20722, CX20723 and CX20724. Cc: stable@vger.kernel.org BugLink: https://bugs.launchpad.net/bugs/1454656 Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index f8f0dfb..78b719b 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -968,6 +968,14 @@ static const struct hda_codec_preset snd_hda_preset_conexant[] = { .patch = patch_conexant_auto }, { .id = 0x14f150b9, .name = "CX20665", .patch = patch_conexant_auto }, + { .id = 0x14f150f1, .name = "CX20721", + .patch = patch_conexant_auto }, + { .id = 0x14f150f2, .name = "CX20722", + .patch = patch_conexant_auto }, + { .id = 0x14f150f3, .name = "CX20723", + .patch = patch_conexant_auto }, + { .id = 0x14f150f4, .name = "CX20724", + .patch = patch_conexant_auto }, { .id = 0x14f1510f, .name = "CX20751/2", .patch = patch_conexant_auto }, { .id = 0x14f15110, .name = "CX20751/2", @@ -1002,6 +1010,10 @@ MODULE_ALIAS("snd-hda-codec-id:14f150ab"); MODULE_ALIAS("snd-hda-codec-id:14f150ac"); MODULE_ALIAS("snd-hda-codec-id:14f150b8"); MODULE_ALIAS("snd-hda-codec-id:14f150b9"); +MODULE_ALIAS("snd-hda-codec-id:14f150f1"); +MODULE_ALIAS("snd-hda-codec-id:14f150f2"); +MODULE_ALIAS("snd-hda-codec-id:14f150f3"); +MODULE_ALIAS("snd-hda-codec-id:14f150f4"); MODULE_ALIAS("snd-hda-codec-id:14f1510f"); MODULE_ALIAS("snd-hda-codec-id:14f15110"); MODULE_ALIAS("snd-hda-codec-id:14f15111"); -- cgit v0.10.2 From 09ea997677cd44ebe7f42573119aaf46b775c683 Mon Sep 17 00:00:00 2001 From: Ansgar Hegerfeld Date: Thu, 14 May 2015 12:31:32 +0200 Subject: ALSA: hda/realtek - ALC292 dock fix for Thinkpad L450 The Lenovo ThinkPad L450 requires the ALC292_FIXUP_TPT440_DOCK fix in order to get sound output on the docking stations audio port. This patch was tested using a ThinkPad L450 (20DSS00B00) using kernel 4.0.3 and a ThinkPad Pro Dock. Signed-off-by: Ansgar Hegerfeld Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 4202a80..2e246fe 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5165,6 +5165,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x5026, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x17aa, 0x5034, "Thinkpad T450", ALC292_FIXUP_TPT440_DOCK), SND_PCI_QUIRK(0x17aa, 0x5036, "Thinkpad T450s", ALC292_FIXUP_TPT440_DOCK), + SND_PCI_QUIRK(0x17aa, 0x503c, "Thinkpad L450", ALC292_FIXUP_TPT440_DOCK), SND_PCI_QUIRK(0x17aa, 0x5109, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x17aa, 0x3bf8, "Quanta FL1", ALC269_FIXUP_PCM_44K), SND_PCI_QUIRK(0x17aa, 0x9e54, "LENOVO NB", ALC269_FIXUP_LENOVO_EAPD), -- cgit v0.10.2 From 13a988396cab68e6c6afd33e461a9e52ce23ff63 Mon Sep 17 00:00:00 2001 From: Koro Chen Date: Wed, 13 May 2015 22:39:03 +0800 Subject: ALSA: pcm: Modify double acknowledged interrupts check condition Currently in snd_pcm_update_hw_ptr0 during interrupt, we consider there were double acknowledged interrupts when: 1. HW reported pointer is smaller than expected, and 2. Time from last update time (hdelta) is over half a buffer time. However, when HW reported pointer is only a few bytes smaller than expected, and when hdelta is just a little larger than half a buffer time (e.g. ping-pong buffer), it wrongly treats this IRQ as double acknowledged. The condition #2 uses jiffies, but jiffies is not high resolution since it is integer. We should consider jiffies inaccuracy. Signed-off-by: Koro Chen Signed-off-by: Takashi Iwai diff --git a/sound/core/pcm_lib.c b/sound/core/pcm_lib.c index ac6b33f..7d45645 100644 --- a/sound/core/pcm_lib.c +++ b/sound/core/pcm_lib.c @@ -339,7 +339,7 @@ static int snd_pcm_update_hw_ptr0(struct snd_pcm_substream *substream, if (delta > new_hw_ptr) { /* check for double acknowledged interrupts */ hdelta = curr_jiffies - runtime->hw_ptr_jiffies; - if (hdelta > runtime->hw_ptr_buffer_jiffies/2) { + if (hdelta > runtime->hw_ptr_buffer_jiffies/2 + 1) { hw_base += runtime->buffer_size; if (hw_base >= runtime->boundary) { hw_base = 0; -- cgit v0.10.2 From fa94b0d72511add5822dc4124f2a3eae66b8863f Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 19 May 2015 10:46:49 +0200 Subject: ALSA: usb-audio: Add quirk for MS LifeCam Studio Microsoft LifeCam Studio (045e:0772) needs a similar quirk for suppressing the wrong sample rate inquiry. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=98481 Cc: Signed-off-by: Takashi Iwai diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 7c5a701..46facfc 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -1117,6 +1117,7 @@ bool snd_usb_get_sample_rate_quirk(struct snd_usb_audio *chip) switch (chip->usb_id) { case USB_ID(0x045E, 0x075D): /* MS Lifecam Cinema */ case USB_ID(0x045E, 0x076D): /* MS Lifecam HD-5000 */ + case USB_ID(0x045E, 0x0772): /* MS Lifecam Studio */ case USB_ID(0x04D8, 0xFEEA): /* Benchmark DAC1 Pre */ return true; } -- cgit v0.10.2 From 976fa9a3ac9223d134cb4e1e3acf0e57160ee924 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 19 May 2015 17:20:36 +0200 Subject: ALSA: sound/atmel/ac97c.c: remove unused variable The recently added DT support for the ac97 driver is causing a gcc warning: sound/atmel/ac97c.c: In function 'atmel_ac97c_probe_dt': sound/atmel/ac97c.c:919:29: warning: unused variable 'match' [-Wunused-variable] const struct of_device_id *match; The variable is clearly unused, so we can remove it. Signed-off-by: Arnd Bergmann Acked-by: Alexander Stein Signed-off-by: Takashi Iwai diff --git a/sound/atmel/ac97c.c b/sound/atmel/ac97c.c index cf4cedf..6dad042 100644 --- a/sound/atmel/ac97c.c +++ b/sound/atmel/ac97c.c @@ -916,7 +916,6 @@ static struct ac97c_platform_data *atmel_ac97c_probe_dt(struct device *dev) { struct ac97c_platform_data *pdata; struct device_node *node = dev->of_node; - const struct of_device_id *match; if (!node) { dev_err(dev, "Device does not have associated DT data\n"); -- cgit v0.10.2 From 48f4b3a2ece560e37232fe9d681cd2ff0f3ff10d Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 20 May 2015 06:49:37 +0200 Subject: ALSA: hda - Reduce verbs by node power-saves The widget (node) power-saves restore the widget states at each transition from D3 to D0 on each node. This was added in the commit [d545a57c5f84:ALSA: hda - Sync node attributes at resume from widget power saving]. However, the test was rater false-positive; this wasn't needed for any codecs. Since the resync may take significant number of additional verbs to be executed, it's better to reduce it. Let's disable it for now again. Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c index 788f969..1c86787 100644 --- a/sound/pci/hda/hda_generic.c +++ b/sound/pci/hda/hda_generic.c @@ -844,8 +844,16 @@ static hda_nid_t path_power_update(struct hda_codec *codec, snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_POWER_STATE, state); changed = nid; + /* all known codecs seem to be capable to handl + * widgets state even in D3, so far. + * if any new codecs need to restore the widget + * states after D0 transition, call the function + * below. + */ +#if 0 /* disabled */ if (state == AC_PWRST_D0) snd_hdac_regmap_sync_node(&codec->core, nid); +#endif } } return changed; -- cgit v0.10.2 From 219f47e4f96442a7fe8fb646c59c6730690a9e66 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 20 May 2015 06:56:23 +0200 Subject: ALSA: hda - Disable widget power-saving for ALC292 & co We've got reports that ALC3226 (a Dell variant of ALC292) gives click noises at transition from D3 to D0 when the widget power-saving is enabled. Further debugging session showed that avoiding it isn't trivial, unfortunately, since paths are basically activated dynamically while the pins have been already enabled. This patch disables the widget power-saving for such codecs. Reported-by: Jonathan McDowell Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 2e246fe..31f8f13 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5623,7 +5623,8 @@ static int patch_alc269(struct hda_codec *codec) spec = codec->spec; spec->gen.shared_mic_vref_pin = 0x18; - codec->power_save_node = 1; + if (codec->core.vendor_id != 0x10ec0292) + codec->power_save_node = 1; snd_hda_pick_fixup(codec, alc269_fixup_models, alc269_fixup_tbl, alc269_fixups); -- cgit v0.10.2 From ae425bb2a05bebe786a25cc8ae64e9d16c4d9b83 Mon Sep 17 00:00:00 2001 From: "Vittorio G (VittGam)" Date: Fri, 22 May 2015 21:15:19 +0200 Subject: ALSA: usb-audio: Add quirk for MS LifeCam HD-3000 Microsoft LifeCam HD-3000 (045e:0779) needs a similar quirk for suppressing the unsupported sample rate inquiry. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=98481 Cc: Signed-off-by: Vittorio Gambaletta Signed-off-by: Takashi Iwai diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 46facfc..2917534 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -1118,6 +1118,7 @@ bool snd_usb_get_sample_rate_quirk(struct snd_usb_audio *chip) case USB_ID(0x045E, 0x075D): /* MS Lifecam Cinema */ case USB_ID(0x045E, 0x076D): /* MS Lifecam HD-5000 */ case USB_ID(0x045E, 0x0772): /* MS Lifecam Studio */ + case USB_ID(0x045E, 0x0779): /* MS Lifecam HD-3000 */ case USB_ID(0x04D8, 0xFEEA): /* Benchmark DAC1 Pre */ return true; } -- cgit v0.10.2 From 3530febb5c7636f6b26d15637f68296804d26491 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Sun, 24 May 2015 08:27:52 +0200 Subject: Revert "ALSA: hda - Add mute-LED mode control to Thinkpad" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit 7290006d8c0900c56d8c58428134f02c35109d17. Through the regression report, it was revealed that the tpacpi_led_set() call to thinkpad_acpi helper doesn't only toggle the mute LED but actually mutes the sound. This is contradiction to the expectation, and rather confuses user. According to Henrique, it's not trivial to judge which TP model behaves "LED-only" and which model does whatever more intrusive, as Lenovo's implementations vary model by model. So, from the safety reason, we should revert the patch for now. Reported-by: Martin Steigerwald Cc: Pali Rohár Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/thinkpad_helper.c b/sound/pci/hda/thinkpad_helper.c index d51703e..0a4ad5f 100644 --- a/sound/pci/hda/thinkpad_helper.c +++ b/sound/pci/hda/thinkpad_helper.c @@ -72,7 +72,6 @@ static void hda_fixup_thinkpad_acpi(struct hda_codec *codec, if (led_set_func(TPACPI_LED_MUTE, false) >= 0) { old_vmaster_hook = spec->vmaster_mute.hook; spec->vmaster_mute.hook = update_tpacpi_mute_led; - spec->vmaster_mute_enum = 1; removefunc = false; } if (led_set_func(TPACPI_LED_MICMUTE, false) >= 0) { -- cgit v0.10.2 From 823245026ead28a244cb9df5ae79b511da128606 Mon Sep 17 00:00:00 2001 From: Kailang Yang Date: Mon, 25 May 2015 17:16:49 +0800 Subject: ALSA: hda/realtek - Add ALC256 alias name for Dell Add ALC3246 for Dell platform. Signed-off-by: Kailang Yang Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 31f8f13..fc0ccc7 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -884,6 +884,7 @@ static struct alc_codec_rename_pci_table rename_pci_tbl[] = { { 0x10ec0275, 0x1028, 0, "ALC3260" }, { 0x10ec0899, 0x1028, 0, "ALC3861" }, { 0x10ec0298, 0x1028, 0, "ALC3266" }, + { 0x10ec0256, 0x1028, 0, "ALC3246" }, { 0x10ec0670, 0x1025, 0, "ALC669X" }, { 0x10ec0676, 0x1025, 0, "ALC679X" }, { 0x10ec0282, 0x1043, 0, "ALC3229" }, -- cgit v0.10.2 From 49fb18972581a781658a4637de76e6069ed5964e Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 27 May 2015 08:37:19 +0200 Subject: ALSA: hda - Set stream_pm ops automatically by generic parser This allows user to test power_save_node feature via sysfs or patch firmware even on the codecs that don't specify it. It'll also save a few lines. Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c index 1c86787..ac0db16 100644 --- a/sound/pci/hda/hda_generic.c +++ b/sound/pci/hda/hda_generic.c @@ -4926,9 +4926,12 @@ int snd_hda_gen_parse_auto_config(struct hda_codec *codec, dig_only: parse_digital(codec); - if (spec->power_down_unused || codec->power_save_node) + if (spec->power_down_unused || codec->power_save_node) { if (!codec->power_filter) codec->power_filter = snd_hda_gen_path_power_filter; + if (!codec->patch_ops.stream_pm) + codec->patch_ops.stream_pm = snd_hda_gen_stream_pm; + } if (!spec->no_analog && spec->beep_nid) { err = snd_hda_attach_beep_device(codec, spec->beep_nid); diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index fc0ccc7..e3bf72b 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5724,7 +5724,6 @@ static int patch_alc269(struct hda_codec *codec) set_beep_amp(spec, spec->gen.mixer_nid, 0x04, HDA_INPUT); codec->patch_ops = alc_patch_ops; - codec->patch_ops.stream_pm = snd_hda_gen_stream_pm; #ifdef CONFIG_PM codec->patch_ops.suspend = alc269_suspend; codec->patch_ops.resume = alc269_resume; diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 43c99ce..0de0fd9 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4403,7 +4403,6 @@ static const struct hda_codec_ops stac_patch_ops = { #ifdef CONFIG_PM .suspend = stac_suspend, #endif - .stream_pm = snd_hda_gen_stream_pm, .reboot_notify = stac_shutup, }; diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c index 31a95cc..742087e 100644 --- a/sound/pci/hda/patch_via.c +++ b/sound/pci/hda/patch_via.c @@ -472,7 +472,6 @@ static const struct hda_codec_ops via_patch_ops = { .init = via_init, .free = via_free, .unsol_event = snd_hda_jack_unsol_event, - .stream_pm = snd_hda_gen_stream_pm, #ifdef CONFIG_PM .suspend = via_suspend, .check_power_status = via_check_power_status, -- cgit v0.10.2 From 0fa372b6c95013af1334b3d5c9b5f03a70ecedab Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 27 May 2015 16:17:19 +0200 Subject: ALSA: hda - Fix noise on AMD radeon 290x controller A new AMD controller [1002:aac8] seems to need the quirk for other AMD NS HDMI stuff, otherwise it gives noisy sounds. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=99021 Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 34040d2..fea198c 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2089,6 +2089,8 @@ static const struct pci_device_id azx_ids[] = { .driver_data = AZX_DRIVER_ATIHDMI_NS | AZX_DCAPS_PRESET_ATI_HDMI_NS }, { PCI_DEVICE(0x1002, 0xaab0), .driver_data = AZX_DRIVER_ATIHDMI_NS | AZX_DCAPS_PRESET_ATI_HDMI_NS }, + { PCI_DEVICE(0x1002, 0xaac8), + .driver_data = AZX_DRIVER_ATIHDMI_NS | AZX_DCAPS_PRESET_ATI_HDMI_NS }, /* VIA VT8251/VT8237A */ { PCI_DEVICE(0x1106, 0x3288), .driver_data = AZX_DRIVER_VIA | AZX_DCAPS_POSFIX_VIA }, -- cgit v0.10.2 From c545f799c7470682756ca64c63f6bfcdaf28442b Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 27 May 2015 21:20:12 +0200 Subject: ALSA: hda - Disable power_save_node for IDT92HD71bxx We've got a regression report that 4.1-rc causes noises on a Dell laptop. Similar like Realtek codec, this seems also triggered by the recent power_save_node feature. As this kind of issue is quite hard to debug without actual hardware, disable the power_save_node flag for this codec as a workaround. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=98971 Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 0de0fd9..6833c74 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4696,7 +4696,8 @@ static int patch_stac92hd71bxx(struct hda_codec *codec) return err; spec = codec->spec; - codec->power_save_node = 1; + /* disabled power_save_node since it causes noises on a Dell machine */ + /* codec->power_save_node = 1; */ spec->linear_tone_beep = 0; spec->gen.own_eapd_ctl = 1; spec->gen.power_down_unused = 1; -- cgit v0.10.2 From b40eda6408e94ee286cb5720cd3f409f70e01778 Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Thu, 28 May 2015 09:40:23 +0200 Subject: ALSA: hda - Disable Headphone Mic boost for ALC662 When headphone mic boost is above zero, some 10 - 20 second delay might occur before the headphone mic is operational. Therefore disable the headphone mic boost control (recording gain is sufficient even without it). (Note: this patch is not about the headset mic, it's about the less common mic-in only mode.) BugLink: https://bugs.launchpad.net/bugs/1454235 Suggested-by: Kailang Yang Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index e3bf72b..ea3af38 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -4228,6 +4228,11 @@ static void alc_fixup_headset_mode_alc662(struct hda_codec *codec, if (action == HDA_FIXUP_ACT_PRE_PROBE) { spec->parse_flags |= HDA_PINCFG_HEADSET_MIC; spec->gen.hp_mic = 1; /* Mic-in is same pin as headphone */ + + /* Disable boost for mic-in permanently. (This code is only called + from quirks that guarantee that the headphone is at NID 0x1b.) */ + snd_hda_codec_write(codec, 0x1b, 0, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000); + snd_hda_override_wcaps(codec, 0x1b, get_wcaps(codec, 0x1b) & ~AC_WCAP_IN_AMP); } else alc_fixup_headset_mode(codec, fix, action); } -- cgit v0.10.2 From b47eee2e0a7de623c24dbcaf303c3bf2b5155455 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 29 May 2015 09:43:29 +0200 Subject: ALSA: hda - Fix lost sound due to stream_pm ops cleanup The commit [49fb18972581: ALSA: hda - Set stream_pm ops automatically by generic parser] resulted in regressions on some Realtek and VIA codecs because these drivers set patch_ops after calling the generic parser, thus stream_pm got cleared to NULL again. I haven't noticed since I tested with IDT codec. Restore (partial revert) the stream_pm ops for them to fix the regression. Fixes: 49fb18972581 ('ALSA: hda - Set stream_pm ops automatically by generic parser') Reported-by: Jeremiah Mahler Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index ea3af38..4641684 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5729,6 +5729,7 @@ static int patch_alc269(struct hda_codec *codec) set_beep_amp(spec, spec->gen.mixer_nid, 0x04, HDA_INPUT); codec->patch_ops = alc_patch_ops; + codec->patch_ops.stream_pm = snd_hda_gen_stream_pm; #ifdef CONFIG_PM codec->patch_ops.suspend = alc269_suspend; codec->patch_ops.resume = alc269_resume; diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c index 742087e..31a95cc 100644 --- a/sound/pci/hda/patch_via.c +++ b/sound/pci/hda/patch_via.c @@ -472,6 +472,7 @@ static const struct hda_codec_ops via_patch_ops = { .init = via_init, .free = via_free, .unsol_event = snd_hda_jack_unsol_event, + .stream_pm = snd_hda_gen_stream_pm, #ifdef CONFIG_PM .suspend = via_suspend, .check_power_status = via_check_power_status, -- cgit v0.10.2