From cfd29aa0e81b791985e8428e6507e80e074e6730 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Tue, 10 Sep 2013 12:50:48 +0200 Subject: serial: tegra: fix tty-kref leak Fix potential tty-kref leak in stop_rx path. Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold Tested-by: Stephen Warren Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/serial-tegra.c b/drivers/tty/serial/serial-tegra.c index d0d972f..0489a2b 100644 --- a/drivers/tty/serial/serial-tegra.c +++ b/drivers/tty/serial/serial-tegra.c @@ -732,7 +732,7 @@ static irqreturn_t tegra_uart_isr(int irq, void *data) static void tegra_uart_stop_rx(struct uart_port *u) { struct tegra_uart_port *tup = to_tegra_uport(u); - struct tty_struct *tty = tty_port_tty_get(&tup->uport.state->port); + struct tty_struct *tty; struct tty_port *port = &u->state->port; struct dma_tx_state state; unsigned long ier; @@ -744,6 +744,8 @@ static void tegra_uart_stop_rx(struct uart_port *u) if (!tup->rx_in_progress) return; + tty = tty_port_tty_get(&tup->uport.state->port); + tegra_uart_wait_sym_time(tup, 1); /* wait a character interval */ ier = tup->ier_shadow; -- cgit v0.10.2 From fc0919c68cb2f75bb1af759315f9d7e2a9443c28 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Tue, 10 Sep 2013 12:50:49 +0200 Subject: serial: pch_uart: fix tty-kref leak in rx-error path Fix tty-kref leak introduced by commit 384e301e ("pch_uart: fix a deadlock when pch_uart as console") which never put its tty reference. Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/pch_uart.c b/drivers/tty/serial/pch_uart.c index 52379e5..f0161d6 100644 --- a/drivers/tty/serial/pch_uart.c +++ b/drivers/tty/serial/pch_uart.c @@ -1098,6 +1098,8 @@ static void pch_uart_err_ir(struct eg20t_port *priv, unsigned int lsr) if (tty == NULL) { for (i = 0; error_msg[i] != NULL; i++) dev_err(&priv->pdev->dev, error_msg[i]); + } else { + tty_kref_put(tty); } } -- cgit v0.10.2 From 19b85cfb190eb9980eaf416bff96aef4159a430e Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Tue, 10 Sep 2013 12:50:50 +0200 Subject: serial: pch_uart: fix tty-kref leak in dma-rx path Fix tty_kref leak when tty_buffer_request room fails in dma-rx path. Note that the tty ref isn't really needed anymore, but as the leak has always been there, fixing it before removing should makes it easier to backport the fix. Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/pch_uart.c b/drivers/tty/serial/pch_uart.c index f0161d6..2c5a3e4 100644 --- a/drivers/tty/serial/pch_uart.c +++ b/drivers/tty/serial/pch_uart.c @@ -685,11 +685,12 @@ static int dma_push_rx(struct eg20t_port *priv, int size) dev_warn(port->dev, "Rx overrun: dropping %u bytes\n", size - room); if (!room) - return room; + goto out; tty_insert_flip_string(tport, sg_virt(&priv->sg_rx), size); port->icount.rx += room; +out: tty_kref_put(tty); return room; -- cgit v0.10.2 From 0b53861230729f60a4a02a838fb51a2345b16f71 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Tue, 10 Sep 2013 12:50:51 +0200 Subject: serial: pch_uart: remove unnecessary tty_port_tty_get Remove unused tty-reference from dma-rx path which was left after the recent tty-port conversions. Also remove a redundant port initialisation while at it. Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/serial/pch_uart.c b/drivers/tty/serial/pch_uart.c index 2c5a3e4..44077c0 100644 --- a/drivers/tty/serial/pch_uart.c +++ b/drivers/tty/serial/pch_uart.c @@ -667,31 +667,21 @@ static int pop_tx_x(struct eg20t_port *priv, unsigned char *buf) static int dma_push_rx(struct eg20t_port *priv, int size) { - struct tty_struct *tty; int room; struct uart_port *port = &priv->port; struct tty_port *tport = &port->state->port; - port = &priv->port; - tty = tty_port_tty_get(tport); - if (!tty) { - dev_dbg(priv->port.dev, "%s:tty is busy now", __func__); - return 0; - } - room = tty_buffer_request_room(tport, size); if (room < size) dev_warn(port->dev, "Rx overrun: dropping %u bytes\n", size - room); if (!room) - goto out; + return 0; tty_insert_flip_string(tport, sg_virt(&priv->sg_rx), size); port->icount.rx += room; -out: - tty_kref_put(tty); return room; } -- cgit v0.10.2 From 93a8d4163ecd3fd8aac264707f1d497f193ea88d Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Tue, 17 Sep 2013 12:34:14 -0400 Subject: n_tty: Fix EOF push index when termios changes Commit 40d5e0905a03601d40cd4e46b8690093c2355d03, 'n_tty: Fix EOF push handling' introduced a subtle state change error wrt EOF push handling when the termios is changed from non-canonical to canonical mode. Reset line_start to the current read_tail index, not 0. Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c index c9a9ddd..01bf5eb 100644 --- a/drivers/tty/n_tty.c +++ b/drivers/tty/n_tty.c @@ -1758,8 +1758,7 @@ static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old) canon_change = (old->c_lflag ^ tty->termios.c_lflag) & ICANON; if (canon_change) { bitmap_zero(ldata->read_flags, N_TTY_BUF_SIZE); - ldata->line_start = 0; - ldata->canon_head = ldata->read_tail; + ldata->line_start = ldata->canon_head = ldata->read_tail; ldata->erasing = 0; ldata->lnext = 0; } -- cgit v0.10.2 From c89efa731e70a12908a0c155518ebbfd2d6ca605 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Tue, 17 Sep 2013 19:16:38 +0100 Subject: ARM: 7836/1: add __get_user_unaligned/__put_user_unaligned BTRFS is now relying on those since v3.12-rc1. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h index 7e1f760..72abdc5 100644 --- a/arch/arm/include/asm/uaccess.h +++ b/arch/arm/include/asm/uaccess.h @@ -19,6 +19,13 @@ #include #include +#if __LINUX_ARM_ARCH__ < 6 +#include +#else +#define __get_user_unaligned __get_user +#define __put_user_unaligned __put_user +#endif + #define VERIFY_READ 0 #define VERIFY_WRITE 1 -- cgit v0.10.2 From 700870119f49084da004ab588ea2b799689efaf7 Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Thu, 18 Apr 2013 07:51:34 -0700 Subject: x86, efi: Don't map Boot Services on i386 Add patch to fix 32bit EFI service mapping (rhbz 726701) Multiple people are reporting hitting the following WARNING on i386, WARNING: at arch/x86/mm/ioremap.c:102 __ioremap_caller+0x3d3/0x440() Modules linked in: Pid: 0, comm: swapper Not tainted 3.9.0-rc7+ #95 Call Trace: [] warn_slowpath_common+0x5f/0x80 [] ? __ioremap_caller+0x3d3/0x440 [] ? __ioremap_caller+0x3d3/0x440 [] warn_slowpath_null+0x1d/0x20 [] __ioremap_caller+0x3d3/0x440 [] ? get_usage_chars+0xfb/0x110 [] ? vprintk_emit+0x147/0x480 [] ? efi_enter_virtual_mode+0x1e4/0x3de [] ioremap_cache+0x1a/0x20 [] ? efi_enter_virtual_mode+0x1e4/0x3de [] efi_enter_virtual_mode+0x1e4/0x3de [] start_kernel+0x286/0x2f4 [] ? repair_env_string+0x51/0x51 [] i386_start_kernel+0x12c/0x12f Due to the workaround described in commit 916f676f8 ("x86, efi: Retain boot service code until after switching to virtual mode") EFI Boot Service regions are mapped for a period during boot. Unfortunately, with the limited size of the i386 direct kernel map it's possible that some of the Boot Service regions will not be directly accessible, which causes them to be ioremap()'d, triggering the above warning as the regions are marked as E820_RAM in the e820 memmap. There are currently only two situations where we need to map EFI Boot Service regions, 1. To workaround the firmware bug described in 916f676f8 2. To access the ACPI BGRT image but since we haven't seen an i386 implementation that requires either, this simple fix should suffice for now. [ Added to changelog - Matt ] Reported-by: Bryan O'Donoghue Acked-by: Tom Zanussi Acked-by: Darren Hart Cc: Josh Triplett Cc: Matthew Garrett Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Greg Kroah-Hartman Cc: Signed-off-by: Josh Boyer Signed-off-by: Matt Fleming diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 90f6ed1..c7e22ab 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -912,10 +912,13 @@ void __init efi_enter_virtual_mode(void) for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { md = p; - if (!(md->attribute & EFI_MEMORY_RUNTIME) && - md->type != EFI_BOOT_SERVICES_CODE && - md->type != EFI_BOOT_SERVICES_DATA) - continue; + if (!(md->attribute & EFI_MEMORY_RUNTIME)) { +#ifdef CONFIG_X86_64 + if (md->type != EFI_BOOT_SERVICES_CODE && + md->type != EFI_BOOT_SERVICES_DATA) +#endif + continue; + } size = md->num_pages << EFI_PAGE_SHIFT; end = md->phys_addr + size; -- cgit v0.10.2 From cc9d3c382bc1674884c2e5e468d51230a9503dee Mon Sep 17 00:00:00 2001 From: Jun'ichi Nomura Date: Fri, 13 Sep 2013 14:54:30 +0900 Subject: dm mpath: do not fail path on -ENOSPC Since ENOSPC is a target-side error, dm-mpath should just pass the error information to upper layer instead of retrying itself with path failover. Otherwise it will end up failing all paths down while path checkers find all paths ok. ENOSPC can now be returned from SCSI device after commit a9d6ceb8 ("[SCSI] return ENOSPC on thin provisioning failure"). Signed-off-by: Jun'ichi Nomura Acked-by: Hannes Reinecke Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index b759a12..075747e 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -1268,6 +1268,7 @@ static int noretry_error(int error) case -EREMOTEIO: case -EILSEQ: case -ENODATA: + case -ENOSPC: return 1; } -- cgit v0.10.2 From bbf3f8cbdc139860a14c4fc2bb25432427045aaa Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Fri, 13 Sep 2013 17:42:24 -0400 Subject: dm stats: fix possible counter corruption on 32-bit systems There was a deliberate race condition in dm_stat_for_entry() to avoid the overhead of disabling and enabling interrupts. The race could result in some events not being counted on 64-bit architectures. However, on 32-bit architectures, operations on long long variables are not atomic, so the race condition could cause the counter to jump by 2^32. Such jumps could be disruptive, so we need to do proper locking on 32-bit architectures. Signed-off-by: Mikulas Patocka Cc: Alasdair G. Kergon Signed-off-by: Mike Snitzer diff --git a/drivers/md/dm-stats.c b/drivers/md/dm-stats.c index 8ae31e8..3d404c1 100644 --- a/drivers/md/dm-stats.c +++ b/drivers/md/dm-stats.c @@ -451,19 +451,26 @@ static void dm_stat_for_entry(struct dm_stat *s, size_t entry, struct dm_stat_percpu *p; /* - * For strict correctness we should use local_irq_disable/enable + * For strict correctness we should use local_irq_save/restore * instead of preempt_disable/enable. * - * This is racy if the driver finishes bios from non-interrupt - * context as well as from interrupt context or from more different - * interrupts. + * preempt_disable/enable is racy if the driver finishes bios + * from non-interrupt context as well as from interrupt context + * or from more different interrupts. * - * However, the race only results in not counting some events, - * so it is acceptable. + * On 64-bit architectures the race only results in not counting some + * events, so it is acceptable. On 32-bit architectures the race could + * cause the counter going off by 2^32, so we need to do proper locking + * there. * * part_stat_lock()/part_stat_unlock() have this race too. */ +#if BITS_PER_LONG == 32 + unsigned long flags; + local_irq_save(flags); +#else preempt_disable(); +#endif p = &s->stat_percpu[smp_processor_id()][entry]; if (!end) { @@ -478,7 +485,11 @@ static void dm_stat_for_entry(struct dm_stat *s, size_t entry, p->ticks[idx] += duration; } +#if BITS_PER_LONG == 32 + local_irq_restore(flags); +#else preempt_enable(); +#endif } static void __dm_stat_bio(struct dm_stat *s, unsigned long bi_rw, -- cgit v0.10.2 From 0fd9560257e9491b9f8ff55e882099e171277fb5 Mon Sep 17 00:00:00 2001 From: Sathya Prakash M R Date: Thu, 12 Sep 2013 14:12:55 +0530 Subject: OMAPDSS: Return right error during connector probe While using HDMI connector driver with sil9022 encoder came across issue where connector driver is probed first. This resulted in error. A deffered probe solved this. Most connector drivers need a encoder driver as their video source. This patch ensures we do a probe defferal if video source is not present for connector drivers. Signed-off-by: Sathya Prakash M R Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/omap2/displays-new/connector-analog-tv.c b/drivers/video/omap2/displays-new/connector-analog-tv.c index 1b60698..ccd9073 100644 --- a/drivers/video/omap2/displays-new/connector-analog-tv.c +++ b/drivers/video/omap2/displays-new/connector-analog-tv.c @@ -191,7 +191,7 @@ static int tvc_probe_pdata(struct platform_device *pdev) in = omap_dss_find_output(pdata->source); if (in == NULL) { dev_err(&pdev->dev, "Failed to find video source\n"); - return -ENODEV; + return -EPROBE_DEFER; } ddata->in = in; diff --git a/drivers/video/omap2/displays-new/connector-dvi.c b/drivers/video/omap2/displays-new/connector-dvi.c index bc5f8ce..63d88ee 100644 --- a/drivers/video/omap2/displays-new/connector-dvi.c +++ b/drivers/video/omap2/displays-new/connector-dvi.c @@ -263,7 +263,7 @@ static int dvic_probe_pdata(struct platform_device *pdev) in = omap_dss_find_output(pdata->source); if (in == NULL) { dev_err(&pdev->dev, "Failed to find video source\n"); - return -ENODEV; + return -EPROBE_DEFER; } ddata->in = in; diff --git a/drivers/video/omap2/displays-new/connector-hdmi.c b/drivers/video/omap2/displays-new/connector-hdmi.c index c582671..9abe2c0 100644 --- a/drivers/video/omap2/displays-new/connector-hdmi.c +++ b/drivers/video/omap2/displays-new/connector-hdmi.c @@ -290,7 +290,7 @@ static int hdmic_probe_pdata(struct platform_device *pdev) in = omap_dss_find_output(pdata->source); if (in == NULL) { dev_err(&pdev->dev, "Failed to find video source\n"); - return -ENODEV; + return -EPROBE_DEFER; } ddata->in = in; -- cgit v0.10.2 From 48664b21aeeffb40c5fa06843f18052e2c4ec9ef Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Thu, 19 Sep 2013 12:59:57 +0300 Subject: OMAPDSS: DISPC: set irq_safe for runtime PM We have a bug with omapdrm, where omapdrm calls dispc's pm_runtime function in atomic context, and dispc's pm_runtime is not marked as irq_safe: BUG: sleeping function called from invalid context at drivers/base/power/runtime.c:952 Dispc's runtime PM callbacks are irq safe, so we can just set the irq_safe flag to fix the issue. However, in the long term, I'd rather have omapdrm manage the runtime PM calls in a better way. Calling get/put for every small operation that touches the dispc registers is very inefficient. It'd be better and cleaner to have clear "in-use" and "not-in-use" states for dispc, so that we don't need to do register context restore for small operations, only to turn dispc off right afterwards. Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/omap2/dss/dispc.c b/drivers/video/omap2/dss/dispc.c index 02a7340..4779750 100644 --- a/drivers/video/omap2/dss/dispc.c +++ b/drivers/video/omap2/dss/dispc.c @@ -3691,6 +3691,7 @@ static int __init omap_dispchw_probe(struct platform_device *pdev) } pm_runtime_enable(&pdev->dev); + pm_runtime_irq_safe(&pdev->dev); r = dispc_runtime_get(); if (r) -- cgit v0.10.2 From 33cbbdccf39f6d7e01c282c4320d818a75b67d54 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 12 Sep 2013 15:27:06 +0200 Subject: perf tools: Sharpen the libaudit dependencies test MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are older libaudit versions that don't have an audit_errno_to_name() method, resulting in a builtin-trace.c build error: builtin-trace.c: In function ‘trace__sys_exit’: builtin-trace.c:794: warning: implicit declaration of function ‘audit_errno_to_name’ Expand the libaudit test to detect this. Signed-off-by: Ingo Molnar Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: H. Peter Anvin Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20130912132706.GD23826@gmail.com [ Fix the test by escaping the double quotes ] Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/config/feature-tests.mak b/tools/perf/config/feature-tests.mak index 708fb8e..7f550c1 100644 --- a/tools/perf/config/feature-tests.mak +++ b/tools/perf/config/feature-tests.mak @@ -210,6 +210,7 @@ define SOURCE_LIBAUDIT int main(void) { + printf(\"error message: %s\n\", audit_errno_to_name(0)); return audit_open(); } endef -- cgit v0.10.2 From 456857bded19a4160bbe083bae926e23f483b7f7 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 12 Sep 2013 15:29:00 +0200 Subject: perf tools: Fill in new definitions for madvise()/mmap() flags builtin-trace.c started using various new syscall features not defined in the header files of older distros - resulting in build failures. Fill in the (ABI) constants if they are not defined. (There might be a better place to put this than builtin-trace.c, into a compat header or so.) Signed-off-by: Ingo Molnar Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: H. Peter Anvin Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: linux-tip-commits@vger.kernel.org Link: http://lkml.kernel.org/r/20130912132900.GE23826@gmail.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index f5aa637..fd48534 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -16,6 +16,23 @@ #include #include +/* For older distros: */ +#ifndef MAP_STACK +# define MAP_STACK 0x20000 +#endif + +#ifndef MADV_HWPOISON +# define MADV_HWPOISON 100 +#endif + +#ifndef MADV_MERGEABLE +# define MADV_MERGEABLE 12 +#endif + +#ifndef MADV_UNMERGEABLE +# define MADV_UNMERGEABLE 13 +#endif + static size_t syscall_arg__scnprintf_hex(char *bf, size_t size, unsigned long arg, u8 arg_idx __maybe_unused, -- cgit v0.10.2 From 886b37baa6274350fb271b2044dc50eb93cb1fc2 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 12 Sep 2013 21:15:21 +0300 Subject: perf annotate: Fix objdump line parsing offset validation When parsing lines from objdump a line containing source code starting with a numeric label is mistaken for a line of disassembly starting with a memory address. Current validation fails to recognise that the "memory address" is out of range and calculates an invalid offset which later causes this segfault: Program received signal SIGSEGV, Segmentation fault. 0x0000000000457315 in disasm__calc_percent (notes=0xc98970, evidx=0, offset=143705, end=2127526177, path=0x7fffffffbf50) at util/annotate.c:631 631 hits += h->addr[offset++]; (gdb) bt #0 0x0000000000457315 in disasm__calc_percent (notes=0xc98970, evidx=0, offset=143705, end=2127526177, path=0x7fffffffbf50) at util/annotate.c:631 #1 0x00000000004d65e3 in annotate_browser__calc_percent (browser=0x7fffffffd130, evsel=0xa01da0) at ui/browsers/annotate.c:364 #2 0x00000000004d7433 in annotate_browser__run (browser=0x7fffffffd130, evsel=0xa01da0, hbt=0x0) at ui/browsers/annotate.c:672 #3 0x00000000004d80c9 in symbol__tui_annotate (sym=0xc989a0, map=0xa02660, evsel=0xa01da0, hbt=0x0) at ui/browsers/annotate.c:962 #4 0x00000000004d7aa0 in hist_entry__tui_annotate (he=0xdf73f0, evsel=0xa01da0, hbt=0x0) at ui/browsers/annotate.c:823 #5 0x00000000004dd648 in perf_evsel__hists_browse (evsel=0xa01da0, nr_events=1, helpline= 0x58b768 "For a higher level overview, try: perf report --sort comm,dso", ev_name=0xa02cd0 "cycles", left_exits=false, hbt= 0x0, min_pcnt=0, env=0xa011e0) at ui/browsers/hists.c:1659 #6 0x00000000004de372 in perf_evlist__tui_browse_hists (evlist=0xa01520, help= 0x58b768 "For a higher level overview, try: perf report --sort comm,dso", hbt=0x0, min_pcnt=0, env=0xa011e0) at ui/browsers/hists.c:1950 #7 0x000000000042cf6b in __cmd_report (rep=0x7fffffffd6c0) at builtin-report.c:581 #8 0x000000000042e25d in cmd_report (argc=0, argv=0x7fffffffe4b0, prefix=0x0) at builtin-report.c:965 #9 0x000000000041a0e1 in run_builtin (p=0x801548, argc=1, argv=0x7fffffffe4b0) at perf.c:319 #10 0x000000000041a319 in handle_internal_command (argc=1, argv=0x7fffffffe4b0) at perf.c:376 #11 0x000000000041a465 in run_argv (argcp=0x7fffffffe38c, argv=0x7fffffffe380) at perf.c:420 #12 0x000000000041a707 in main (argc=1, argv=0x7fffffffe4b0) at perf.c:521 After the fix is applied the symbol can be annotated showing the problematic line "1: rep" copy_user_generic_string /usr/lib/debug/lib/modules/3.9.10-100.fc17.x86_64/vmlinux */ ENTRY(copy_user_generic_string) CFI_STARTPROC ASM_STAC andl %edx,%edx and %edx,%edx jz 4f je 37 cmpl $8,%edx cmp $0x8,%edx jb 2f /* less than 8 bytes, go to byte copy loop */ jb 33 ALIGN_DESTINATION mov %edi,%ecx and $0x7,%ecx je 28 sub $0x8,%ecx neg %ecx sub %ecx,%edx 1a: mov (%rsi),%al mov %al,(%rdi) inc %rsi inc %rdi dec %ecx jne 1a movl %edx,%ecx 28: mov %edx,%ecx shrl $3,%ecx shr $0x3,%ecx andl $7,%edx and $0x7,%edx 1: rep 100.00 rep movsq %ds:(%rsi),%es:(%rdi) movsq 2: movl %edx,%ecx 33: mov %edx,%ecx 3: rep rep movsb %ds:(%rsi),%es:(%rdi) movsb 4: xorl %eax,%eax 37: xor %eax,%eax data32 xchg %ax,%ax ASM_CLAC ret retq Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1379009721-27667-1-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c index bfc5a27..7eae548 100644 --- a/tools/perf/util/annotate.c +++ b/tools/perf/util/annotate.c @@ -809,7 +809,7 @@ static int symbol__parse_objdump_line(struct symbol *sym, struct map *map, end = map__rip_2objdump(map, sym->end); offset = line_ip - start; - if (offset < 0 || (u64)line_ip > end) + if ((u64)line_ip < start || (u64)line_ip > end) offset = -1; else parsed_line = tmp2 + 1; -- cgit v0.10.2 From 5b6a42fcb38f3230922239255698dd8c39382978 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 12 Sep 2013 21:19:19 +0300 Subject: perf tools: Fix buildid cache handling of kallsyms with kcore When kallsyms is used with kcore the dso long_name becomes the kcore file name. That prevents the buildid cache from caching kallsyms. (There is no support at present for caching kcore). Fix by changing it so that the kallsyms name is used in that case instead. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1379009959-28046-1-git-send-email-adrian.hunter@intel.com [ Kept 'struct foo' pointer as first parameter of foo__ prefixed functions ] Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index 26441d0..ce69901 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -199,9 +199,11 @@ static int write_buildid(char *name, size_t name_len, u8 *build_id, return write_padded(fd, name, name_len + 1, len); } -static int __dsos__write_buildid_table(struct list_head *head, pid_t pid, - u16 misc, int fd) +static int __dsos__write_buildid_table(struct list_head *head, + struct machine *machine, + pid_t pid, u16 misc, int fd) { + char nm[PATH_MAX]; struct dso *pos; dsos__for_each_with_build_id(pos, head) { @@ -215,6 +217,10 @@ static int __dsos__write_buildid_table(struct list_head *head, pid_t pid, if (is_vdso_map(pos->short_name)) { name = (char *) VDSO__MAP_NAME; name_len = sizeof(VDSO__MAP_NAME) + 1; + } else if (dso__is_kcore(pos)) { + machine__mmap_name(machine, nm, sizeof(nm)); + name = nm; + name_len = strlen(nm) + 1; } else { name = pos->long_name; name_len = pos->long_name_len + 1; @@ -240,10 +246,10 @@ static int machine__write_buildid_table(struct machine *machine, int fd) umisc = PERF_RECORD_MISC_GUEST_USER; } - err = __dsos__write_buildid_table(&machine->kernel_dsos, machine->pid, - kmisc, fd); + err = __dsos__write_buildid_table(&machine->kernel_dsos, machine, + machine->pid, kmisc, fd); if (err == 0) - err = __dsos__write_buildid_table(&machine->user_dsos, + err = __dsos__write_buildid_table(&machine->user_dsos, machine, machine->pid, umisc, fd); return err; } @@ -375,23 +381,31 @@ out_free: return err; } -static int dso__cache_build_id(struct dso *dso, const char *debugdir) +static int dso__cache_build_id(struct dso *dso, struct machine *machine, + const char *debugdir) { bool is_kallsyms = dso->kernel && dso->long_name[0] != '/'; bool is_vdso = is_vdso_map(dso->short_name); + char *name = dso->long_name; + char nm[PATH_MAX]; - return build_id_cache__add_b(dso->build_id, sizeof(dso->build_id), - dso->long_name, debugdir, - is_kallsyms, is_vdso); + if (dso__is_kcore(dso)) { + is_kallsyms = true; + machine__mmap_name(machine, nm, sizeof(nm)); + name = nm; + } + return build_id_cache__add_b(dso->build_id, sizeof(dso->build_id), name, + debugdir, is_kallsyms, is_vdso); } -static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir) +static int __dsos__cache_build_ids(struct list_head *head, + struct machine *machine, const char *debugdir) { struct dso *pos; int err = 0; dsos__for_each_with_build_id(pos, head) - if (dso__cache_build_id(pos, debugdir)) + if (dso__cache_build_id(pos, machine, debugdir)) err = -1; return err; @@ -399,8 +413,9 @@ static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir) static int machine__cache_build_ids(struct machine *machine, const char *debugdir) { - int ret = __dsos__cache_build_ids(&machine->kernel_dsos, debugdir); - ret |= __dsos__cache_build_ids(&machine->user_dsos, debugdir); + int ret = __dsos__cache_build_ids(&machine->kernel_dsos, machine, + debugdir); + ret |= __dsos__cache_build_ids(&machine->user_dsos, machine, debugdir); return ret; } -- cgit v0.10.2 From e955d5c434009b661dbb200c601254b0c2d9940a Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Fri, 13 Sep 2013 16:49:30 +0300 Subject: perf tools: Fix compile with libelf without get_phdrnum Add a feature check for get_phdrnum() and implement a replacement if it is not present. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1379080170-6608-1-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile index 214e17e..346ee92 100644 --- a/tools/perf/config/Makefile +++ b/tools/perf/config/Makefile @@ -180,6 +180,9 @@ FLAGS_LIBELF=$(CFLAGS) $(LDFLAGS) $(EXTLIBS) ifeq ($(call try-cc,$(SOURCE_ELF_MMAP),$(FLAGS_LIBELF),-DLIBELF_MMAP),y) CFLAGS += -DLIBELF_MMAP endif +ifeq ($(call try-cc,$(SOURCE_ELF_GETPHDRNUM),$(FLAGS_LIBELF),-DHAVE_ELF_GETPHDRNUM),y) + CFLAGS += -DHAVE_ELF_GETPHDRNUM +endif # include ARCH specific config -include $(src-perf)/arch/$(ARCH)/Makefile diff --git a/tools/perf/config/feature-tests.mak b/tools/perf/config/feature-tests.mak index 7f550c1..d5a8dd4 100644 --- a/tools/perf/config/feature-tests.mak +++ b/tools/perf/config/feature-tests.mak @@ -61,6 +61,15 @@ int main(void) } endef +define SOURCE_ELF_GETPHDRNUM +#include +int main(void) +{ + size_t dst; + return elf_getphdrnum(0, &dst); +} +endef + ifndef NO_SLANG define SOURCE_SLANG #include diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index a7b9ab5..a9c829b 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -8,6 +8,22 @@ #include "symbol.h" #include "debug.h" +#ifndef HAVE_ELF_GETPHDRNUM +static int elf_getphdrnum(Elf *elf, size_t *dst) +{ + GElf_Ehdr gehdr; + GElf_Ehdr *ehdr; + + ehdr = gelf_getehdr(elf, &gehdr); + if (!ehdr) + return -1; + + *dst = ehdr->e_phnum; + + return 0; +} +#endif + #ifndef NT_GNU_BUILD_ID #define NT_GNU_BUILD_ID 3 #endif -- cgit v0.10.2 From 33e940a25daaea71be054e8a4bdb61730cc9ebbc Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 17 Sep 2013 16:34:28 -0300 Subject: perf session: Check for SIGINT in more loops When processing big files we were not checking if session_done was set by the SIGINT signal handler, for instance in 'perf report'. Fix it. Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-pyad42lgrtq7xhg2dpsoauq7@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c index 423875c..afe377b 100644 --- a/tools/perf/builtin-inject.c +++ b/tools/perf/builtin-inject.c @@ -321,8 +321,6 @@ found: return perf_event__repipe(tool, event_sw, &sample_sw, machine); } -extern volatile int session_done; - static void sig_handler(int sig __maybe_unused) { session_done = 1; diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 8e50d8d..72eae74 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -401,8 +401,6 @@ static int perf_report__setup_sample_type(struct perf_report *rep) return 0; } -extern volatile int session_done; - static void sig_handler(int sig __maybe_unused) { session_done = 1; @@ -568,6 +566,9 @@ static int __cmd_report(struct perf_report *rep) } } + if (session_done()) + return 0; + if (nr_samples == 0) { ui__error("The %s file has no samples!\n", session->filename); return 0; diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index 7f31a3d..9c333ff 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -553,8 +553,6 @@ static struct perf_tool perf_script = { .ordering_requires_timestamps = true, }; -extern volatile int session_done; - static void sig_handler(int sig __maybe_unused) { session_done = 1; diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c index 46a0d35..9ff6cf3 100644 --- a/tools/perf/util/hist.c +++ b/tools/perf/util/hist.c @@ -611,6 +611,8 @@ void hists__collapse_resort(struct hists *hists) next = rb_first(root); while (next) { + if (session_done()) + break; n = rb_entry(next, struct hist_entry, rb_node_in); next = rb_next(&n->rb_node_in); diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 51f5edf..70ffa41 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -531,6 +531,9 @@ static int flush_sample_queue(struct perf_session *s, return 0; list_for_each_entry_safe(iter, tmp, head, list) { + if (session_done()) + return 0; + if (iter->timestamp > limit) break; @@ -1160,7 +1163,6 @@ static void perf_session__warn_about_errors(const struct perf_session *session, } } -#define session_done() (*(volatile int *)(&session_done)) volatile int session_done; static int __perf_session__process_pipe_events(struct perf_session *self, @@ -1372,10 +1374,13 @@ more: "Processing events..."); } + err = 0; + if (session_done()) + goto out_err; + if (file_pos < file_size) goto more; - err = 0; /* do the final flush for ordered samples */ session->ordered_samples.next_flush = ULLONG_MAX; err = flush_sample_queue(session, tool); diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h index 3aa75fb..04bf737 100644 --- a/tools/perf/util/session.h +++ b/tools/perf/util/session.h @@ -124,4 +124,8 @@ int __perf_session__set_tracepoints_handlers(struct perf_session *session, #define perf_session__set_tracepoints_handlers(session, array) \ __perf_session__set_tracepoints_handlers(session, array, ARRAY_SIZE(array)) + +extern volatile int session_done; + +#define session_done() (*(volatile int *)(&session_done)) #endif /* __PERF_SESSION_H */ -- cgit v0.10.2 From 0dbb1cac1dbdddf64cdd0f972c062a26484b7177 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 23 Apr 2012 12:24:36 +0900 Subject: perf probe: Fix finder to find lines of given function The commit ba28c59bc9ed8fb7b9a753cd88ee54a2c4f6265b fixed a declaration entry bug in probe_point_search_cb(). There are same bugs in line finder and call_probe_finder(). This introduces a new dwarf utility function to determine given DIE is a function definition, not declaration. Signed-off-by: Masami Hiramatsu Cc: Ananth N Mavinakayanahalli Cc: Prashanth Nageshappa Cc: Srikar Dronamraju Cc: Steven Rostedt Cc: yrl.pp-manager.tt@hitachi.com Link: http://lkml.kernel.org/r/20120423032435.8737.80064.stgit@localhost.localdomain Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c index 3e5f543..e23bde1 100644 --- a/tools/perf/util/dwarf-aux.c +++ b/tools/perf/util/dwarf-aux.c @@ -263,6 +263,21 @@ bool die_is_signed_type(Dwarf_Die *tp_die) } /** + * die_is_func_def - Ensure that this DIE is a subprogram and definition + * @dw_die: a DIE + * + * Ensure that this DIE is a subprogram and NOT a declaration. This + * returns true if @dw_die is a function definition. + **/ +bool die_is_func_def(Dwarf_Die *dw_die) +{ + Dwarf_Attribute attr; + + return (dwarf_tag(dw_die) == DW_TAG_subprogram && + dwarf_attr(dw_die, DW_AT_declaration, &attr) == NULL); +} + +/** * die_get_data_member_location - Get the data-member offset * @mb_die: a DIE of a member of a data structure * @offs: The offset of the member in the data structure @@ -392,6 +407,10 @@ static int __die_search_func_cb(Dwarf_Die *fn_die, void *data) { struct __addr_die_search_param *ad = data; + /* + * Since a declaration entry doesn't has given pc, this always returns + * function definition entry. + */ if (dwarf_tag(fn_die) == DW_TAG_subprogram && dwarf_haspc(fn_die, ad->addr)) { memcpy(ad->die_mem, fn_die, sizeof(Dwarf_Die)); diff --git a/tools/perf/util/dwarf-aux.h b/tools/perf/util/dwarf-aux.h index 6ce1717..8658d41 100644 --- a/tools/perf/util/dwarf-aux.h +++ b/tools/perf/util/dwarf-aux.h @@ -38,6 +38,9 @@ extern int cu_find_lineinfo(Dwarf_Die *cudie, unsigned long addr, extern int cu_walk_functions_at(Dwarf_Die *cu_die, Dwarf_Addr addr, int (*callback)(Dwarf_Die *, void *), void *data); +/* Ensure that this DIE is a subprogram and definition (not declaration) */ +extern bool die_is_func_def(Dwarf_Die *dw_die); + /* Compare diename and tname */ extern bool die_compare_name(Dwarf_Die *dw_die, const char *tname); diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index be03293..20c7299 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -734,7 +734,7 @@ static int call_probe_finder(Dwarf_Die *sc_die, struct probe_finder *pf) } /* If not a real subprogram, find a real one */ - if (dwarf_tag(sc_die) != DW_TAG_subprogram) { + if (!die_is_func_def(sc_die)) { if (!die_find_realfunc(&pf->cu_die, pf->addr, &pf->sp_die)) { pr_warning("Failed to find probe point in any " "functions.\n"); @@ -980,12 +980,10 @@ static int probe_point_search_cb(Dwarf_Die *sp_die, void *data) struct dwarf_callback_param *param = data; struct probe_finder *pf = param->data; struct perf_probe_point *pp = &pf->pev->point; - Dwarf_Attribute attr; /* Check tag and diename */ - if (dwarf_tag(sp_die) != DW_TAG_subprogram || - !die_compare_name(sp_die, pp->function) || - dwarf_attr(sp_die, DW_AT_declaration, &attr)) + if (!die_is_func_def(sp_die) || + !die_compare_name(sp_die, pp->function)) return DWARF_CB_OK; /* Check declared file */ @@ -1474,7 +1472,7 @@ static int line_range_inline_cb(Dwarf_Die *in_die, void *data) return 0; } -/* Search function from function name */ +/* Search function definition from function name */ static int line_range_search_cb(Dwarf_Die *sp_die, void *data) { struct dwarf_callback_param *param = data; @@ -1485,7 +1483,7 @@ static int line_range_search_cb(Dwarf_Die *sp_die, void *data) if (lr->file && strtailcmp(lr->file, dwarf_decl_file(sp_die))) return DWARF_CB_OK; - if (dwarf_tag(sp_die) == DW_TAG_subprogram && + if (die_is_func_def(sp_die) && die_compare_name(sp_die, lr->function)) { lf->fname = dwarf_decl_file(sp_die); dwarf_decl_line(sp_die, &lr->offset); -- cgit v0.10.2 From 4028b6c4c03f213260e9290ff3a6b5439aad07ce Mon Sep 17 00:00:00 2001 From: Liam Girdwood Date: Fri, 13 Sep 2013 17:43:17 +0100 Subject: ALSA: compress: Fix compress device unregister. snd_unregister_device() should return the device type and not stream direction. Signed-off-by: Liam Girdwood Acked-by: Vinod Koul Tested-by: Vinod Koul Cc: Signed-off-by: Takashi Iwai diff --git a/sound/core/compress_offload.c b/sound/core/compress_offload.c index 9896954..5863ba6 100644 --- a/sound/core/compress_offload.c +++ b/sound/core/compress_offload.c @@ -837,7 +837,8 @@ static int snd_compress_dev_disconnect(struct snd_device *device) struct snd_compr *compr; compr = device->device_data; - snd_unregister_device(compr->direction, compr->card, compr->device); + snd_unregister_device(SNDRV_DEVICE_TYPE_COMPRESS, compr->card, + compr->device); return 0; } -- cgit v0.10.2 From 0f9654296eabf7f238661d20a6af545986cf713f Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 12 Sep 2013 15:16:49 +0200 Subject: perf tools: Fix old GCC build error in trace-event-parse.c:parse_proc_kallsyms() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Old GCC (4.1) does not see through the code flow of parse_proc_kallsyms() and gets confused about the status of 'fmt': util/trace-event-parse.c: In function ‘parse_proc_kallsyms’: util/trace-event-parse.c:189: warning: ‘fmt’ may be used uninitialized in this function make: *** [util/trace-event-parse.o] Error 1 Help out GCC by initializing 'fmt' to NULL. Signed-off-by: Ingo Molnar Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: H. Peter Anvin Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20130912131649.GC23826@gmail.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c index fe7a27d..e9e1c03 100644 --- a/tools/perf/util/trace-event-parse.c +++ b/tools/perf/util/trace-event-parse.c @@ -186,7 +186,7 @@ void parse_proc_kallsyms(struct pevent *pevent, char *next = NULL; char *addr_str; char *mod; - char *fmt; + char *fmt = NULL; line = strtok_r(file, "\n", &next); while (line) { -- cgit v0.10.2 From ce7eebe5c3deef8e19c177c24ee75843256e69ca Mon Sep 17 00:00:00 2001 From: Vinson Lee Date: Wed, 18 Sep 2013 16:16:40 -0700 Subject: tools lib lk: Uninclude linux/magic.h in debugfs.c The compilation only looks for linux/magic.h from the default include paths, which does not include the source tree. This results in a build error if linux/magic.h is not available or not installed. For example, this build error occurs on CentOS 5. $ make -C tools/lib/lk V=1 [...] gcc -o debugfs.o -c -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -D_FORTIFY_SOURCE=2 -Wbad-function-cast -Wdeclaration-after-statement -Wformat-security -Wformat-y2k -Winit-self -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wno-system-headers -Wold-style-definition -Wpacked -Wredundant-decls -Wshadow -Wstrict-aliasing=3 -Wstrict-prototypes -Wswitch-default -Wswitch-enum -Wundef -Wwrite-strings -Wformat -fPIC -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 debugfs.c debugfs.c:8:25: error: linux/magic.h: No such file or directory The only symbol from linux/magic.h needed by debugfs.c is DEBUGFS_MAGIC, and that is already defined in debugfs.h. linux/magic.h isn't providing any extra symbols and can unincluded. This is similar to the approach by perf, which has its own magic.h wrapper at tools/perf/util/include/linux/magic.h Signed-off-by: Vinson Lee Acked-by: Borislav Petkov Cc: Borislav Petkov Cc: Vinson Lee Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1379546200-17028-1-git-send-email-vlee@freedesktop.org Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/lib/lk/debugfs.c b/tools/lib/lk/debugfs.c index 099e7cd..7c43479 100644 --- a/tools/lib/lk/debugfs.c +++ b/tools/lib/lk/debugfs.c @@ -5,7 +5,6 @@ #include #include #include -#include #include #include "debugfs.h" -- cgit v0.10.2 From c5ecceefdb840af45db436adc58219ae97b6ef3c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 13 Sep 2013 23:39:17 +0200 Subject: perf: Update ABI comment For some mysterious reason the sample_id field of PERF_RECORD_MMAP went AWOL. Reported-by: Vince Weaver Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index 40a1fb8..7f6d584 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -528,6 +528,7 @@ enum perf_event_type { * u64 len; * u64 pgoff; * char filename[]; + * struct sample_id sample_id; * }; */ PERF_RECORD_MMAP = 1, -- cgit v0.10.2 From 73c4427c6ca3b32fa0441791e9c6eadceff7242f Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Tue, 17 Sep 2013 14:48:13 +0800 Subject: perf/x86/intel/uncore: Don't use smp_processor_id() in validate_group() uncore_validate_group() can't call smp_processor_id() because it is in preemptible context. Pass NUMA_NO_NODE to the allocator instead. Signed-off-by: Yan, Zheng Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1379400493-11505-1-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/arch/x86/kernel/cpu/perf_event_intel_uncore.c index 8ed4458..4118f9f 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_uncore.c +++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.c @@ -2706,14 +2706,14 @@ static void uncore_pmu_init_hrtimer(struct intel_uncore_box *box) box->hrtimer.function = uncore_pmu_hrtimer; } -struct intel_uncore_box *uncore_alloc_box(struct intel_uncore_type *type, int cpu) +static struct intel_uncore_box *uncore_alloc_box(struct intel_uncore_type *type, int node) { struct intel_uncore_box *box; int i, size; size = sizeof(*box) + type->num_shared_regs * sizeof(struct intel_uncore_extra_reg); - box = kzalloc_node(size, GFP_KERNEL, cpu_to_node(cpu)); + box = kzalloc_node(size, GFP_KERNEL, node); if (!box) return NULL; @@ -3031,7 +3031,7 @@ static int uncore_validate_group(struct intel_uncore_pmu *pmu, struct intel_uncore_box *fake_box; int ret = -EINVAL, n; - fake_box = uncore_alloc_box(pmu->type, smp_processor_id()); + fake_box = uncore_alloc_box(pmu->type, NUMA_NO_NODE); if (!fake_box) return -ENOMEM; @@ -3294,7 +3294,7 @@ static int uncore_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id } type = pci_uncores[UNCORE_PCI_DEV_TYPE(id->driver_data)]; - box = uncore_alloc_box(type, 0); + box = uncore_alloc_box(type, NUMA_NO_NODE); if (!box) return -ENOMEM; @@ -3499,7 +3499,7 @@ static int uncore_cpu_prepare(int cpu, int phys_id) if (pmu->func_id < 0) pmu->func_id = j; - box = uncore_alloc_box(type, cpu); + box = uncore_alloc_box(type, cpu_to_node(cpu)); if (!box) return -ENOMEM; -- cgit v0.10.2 From fa7315871046b9a4c48627905691dbde57e51033 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 19 Sep 2013 10:16:42 +0200 Subject: perf: Fix capabilities bitfield compatibility in 'struct perf_event_mmap_page' Solve the problems around the broken definition of perf_event_mmap_page:: cap_usr_time and cap_usr_rdpmc fields which used to overlap, partially fixed by: 860f085b74e9 ("perf: Fix broken union in 'struct perf_event_mmap_page'") The problem with the fix (merged in v3.12-rc1 and not yet released officially), noticed by Vince Weaver is that the new behavior is not detectable by new user-space, and that due to the reuse of the field names it's easy to mis-compile a binary if old headers are used on a new kernel or new headers are used on an old kernel. To solve all that make this change explicit, detectable and self-contained, by iterating the ABI the following way: - Always clear bit 0, and rename it to usrpage->cap_bit0, to at least not confuse old user-space binaries. RDPMC will be marked as unavailable to old binaries but that's within the ABI, this is a capability bit. - Rename bit 1 to ->cap_bit0_is_deprecated and always set it to 1, so new libraries can reliably detect that bit 0 is deprecated and perma-zero without having to check the kernel version. - Use bits 2, 3, 4 for the newly defined, correct functionality: cap_user_rdpmc : 1, /* The RDPMC instruction can be used to read counts */ cap_user_time : 1, /* The time_* fields are used */ cap_user_time_zero : 1, /* The time_zero field is used */ - Rename all the bitfield names in perf_event.h to be different from the old names, to make sure it's not possible to mis-compile it accidentally with old assumptions. The 'size' field can then be used in the future to add new fields and it will act as a natural ABI version indicator as well. Also adjust tools/perf/ userspace for the new definitions, noticed by Adrian Hunter. Reported-by: Vince Weaver Signed-off-by: Peter Zijlstra Also-Fixed-by: Adrian Hunter Link: http://lkml.kernel.org/n/tip-zr03yxjrpXesOzzupszqglbv@git.kernel.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 8355c84..a9c606b 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1883,9 +1883,9 @@ static struct pmu pmu = { void arch_perf_update_userpage(struct perf_event_mmap_page *userpg, u64 now) { - userpg->cap_usr_time = 0; - userpg->cap_usr_time_zero = 0; - userpg->cap_usr_rdpmc = x86_pmu.attr_rdpmc; + userpg->cap_user_time = 0; + userpg->cap_user_time_zero = 0; + userpg->cap_user_rdpmc = x86_pmu.attr_rdpmc; userpg->pmc_width = x86_pmu.cntval_bits; if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) @@ -1894,13 +1894,13 @@ void arch_perf_update_userpage(struct perf_event_mmap_page *userpg, u64 now) if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC)) return; - userpg->cap_usr_time = 1; + userpg->cap_user_time = 1; userpg->time_mult = this_cpu_read(cyc2ns); userpg->time_shift = CYC2NS_SCALE_FACTOR; userpg->time_offset = this_cpu_read(cyc2ns_offset) - now; if (sched_clock_stable && !check_tsc_disabled()) { - userpg->cap_usr_time_zero = 1; + userpg->cap_user_time_zero = 1; userpg->time_zero = this_cpu_read(cyc2ns_offset); } } diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index 7f6d584..009a655 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -380,10 +380,13 @@ struct perf_event_mmap_page { union { __u64 capabilities; struct { - __u64 cap_usr_time : 1, - cap_usr_rdpmc : 1, - cap_usr_time_zero : 1, - cap_____res : 61; + __u64 cap_bit0 : 1, /* Always 0, deprecated, see commit 860f085b74e9 */ + cap_bit0_is_deprecated : 1, /* Always 1, signals that bit 0 is zero */ + + cap_user_rdpmc : 1, /* The RDPMC instruction can be used to read counts */ + cap_user_time : 1, /* The time_* fields are used */ + cap_user_time_zero : 1, /* The time_zero field is used */ + cap_____res : 59; }; }; @@ -442,12 +445,13 @@ struct perf_event_mmap_page { * ((rem * time_mult) >> time_shift); */ __u64 time_zero; + __u32 size; /* Header size up to __reserved[] fields. */ /* * Hole for extension of the self monitor capabilities */ - __u64 __reserved[119]; /* align to 1k */ + __u8 __reserved[118*8+4]; /* align to 1k. */ /* * Control data for the mmap() data buffer. diff --git a/kernel/events/core.c b/kernel/events/core.c index dd236b6..cb4238e 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -3660,6 +3660,26 @@ static void calc_timer_values(struct perf_event *event, *running = ctx_time - event->tstamp_running; } +static void perf_event_init_userpage(struct perf_event *event) +{ + struct perf_event_mmap_page *userpg; + struct ring_buffer *rb; + + rcu_read_lock(); + rb = rcu_dereference(event->rb); + if (!rb) + goto unlock; + + userpg = rb->user_page; + + /* Allow new userspace to detect that bit 0 is deprecated */ + userpg->cap_bit0_is_deprecated = 1; + userpg->size = offsetof(struct perf_event_mmap_page, __reserved); + +unlock: + rcu_read_unlock(); +} + void __weak arch_perf_update_userpage(struct perf_event_mmap_page *userpg, u64 now) { } @@ -4044,6 +4064,7 @@ again: ring_buffer_attach(event, rb); rcu_assign_pointer(event->rb, rb); + perf_event_init_userpage(event); perf_event_update_userpage(event); unlock: diff --git a/tools/perf/arch/x86/util/tsc.c b/tools/perf/arch/x86/util/tsc.c index 9570c2b..b2519e4 100644 --- a/tools/perf/arch/x86/util/tsc.c +++ b/tools/perf/arch/x86/util/tsc.c @@ -32,7 +32,7 @@ u64 tsc_to_perf_time(u64 cyc, struct perf_tsc_conversion *tc) int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc, struct perf_tsc_conversion *tc) { - bool cap_usr_time_zero; + bool cap_user_time_zero; u32 seq; int i = 0; @@ -42,7 +42,7 @@ int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc, tc->time_mult = pc->time_mult; tc->time_shift = pc->time_shift; tc->time_zero = pc->time_zero; - cap_usr_time_zero = pc->cap_usr_time_zero; + cap_user_time_zero = pc->cap_user_time_zero; rmb(); if (pc->lock == seq && !(seq & 1)) break; @@ -52,7 +52,7 @@ int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc, } } - if (!cap_usr_time_zero) + if (!cap_user_time_zero) return -EOPNOTSUPP; return 0; -- cgit v0.10.2 From b18855500fc40da050512d9df82d2f1471e59642 Mon Sep 17 00:00:00 2001 From: Vladimir Davydov Date: Sun, 15 Sep 2013 17:49:13 +0400 Subject: sched/balancing: Fix 'local->avg_load > sds->avg_load' case in calculate_imbalance() In busiest->group_imb case we can come to calculate_imbalance() with local->avg_load >= busiest->avg_load >= sds->avg_load. This can result in imbalance overflow, because it is calculated as follows env->imbalance = min( max_pull * busiest->group_power, (sds->avg_load - local->avg_load) * local->group_power) / SCHED_POWER_SCALE; As a result we can end up constantly bouncing tasks from one cpu to another if there are pinned tasks. Fix this by skipping the assignment and assuming imbalance=0 in case local->avg_load > sds->avg_load. [ The bug can be caught by running 2*N cpuhogs pinned to two logical cpus belonging to different cores on an HT-enabled machine with N logical cpus: just look at se.nr_migrations growth. ] Signed-off-by: Vladimir Davydov Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/8f596cc6bc0e5e655119dc892c9bfcad26e971f4.1379252740.git.vdavydov@parallels.com Signed-off-by: Ingo Molnar diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 11cd136..0b99aae 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4896,7 +4896,8 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s * max load less than avg load(as we skip the groups at or below * its cpu_power, while calculating max_load..) */ - if (busiest->avg_load < sds->avg_load) { + if (busiest->avg_load <= sds->avg_load || + local->avg_load >= sds->avg_load) { env->imbalance = 0; return fix_small_imbalance(env, sds); } -- cgit v0.10.2 From 3029ede39373c368f402a76896600d85a4f7121b Mon Sep 17 00:00:00 2001 From: Vladimir Davydov Date: Sun, 15 Sep 2013 17:49:14 +0400 Subject: sched/balancing: Fix 'local->avg_load > busiest->avg_load' case in fix_small_imbalance() In busiest->group_imb case we can come to fix_small_imbalance() with local->avg_load > busiest->avg_load. This can result in wrong imbalance fix-up, because there is the following check there where all the members are unsigned: if (busiest->avg_load - local->avg_load + scaled_busy_load_per_task >= (scaled_busy_load_per_task * imbn)) { env->imbalance = busiest->load_per_task; return; } As a result we can end up constantly bouncing tasks from one cpu to another if there are pinned tasks. Fix it by substituting the subtraction with an equivalent addition in the check. [ The bug can be caught by running 2*N cpuhogs pinned to two logical cpus belonging to different cores on an HT-enabled machine with N logical cpus: just look at se.nr_migrations growth. ] Signed-off-by: Vladimir Davydov Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/ef167822e5c5b2d96cf5b0e3e4f4bdff3f0414a2.1379252740.git.vdavydov@parallels.com Signed-off-by: Ingo Molnar diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 0b99aae..2aedacc 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4823,8 +4823,8 @@ void fix_small_imbalance(struct lb_env *env, struct sd_lb_stats *sds) (busiest->load_per_task * SCHED_POWER_SCALE) / busiest->group_power; - if (busiest->avg_load - local->avg_load + scaled_busy_load_per_task >= - (scaled_busy_load_per_task * imbn)) { + if (busiest->avg_load + scaled_busy_load_per_task >= + local->avg_load + (scaled_busy_load_per_task * imbn)) { env->imbalance = busiest->load_per_task; return; } -- cgit v0.10.2 From 7e3115ef5149fc502e3a2e80719dba54a8e7409d Mon Sep 17 00:00:00 2001 From: Vladimir Davydov Date: Sat, 14 Sep 2013 19:39:46 +0400 Subject: sched/balancing: Fix cfs_rq->task_h_load calculation Patch a003a2 (sched: Consider runnable load average in move_tasks()) sets all top-level cfs_rqs' h_load to rq->avg.load_avg_contrib, which is always 0. This mistype leads to all tasks having weight 0 when load balancing in a cpu-cgroup enabled setup. There obviously should be sum of weights of all runnable tasks there instead. Fix it. Signed-off-by: Vladimir Davydov Reviewed-by: Paul Turner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1379173186-11944-1-git-send-email-vdavydov@parallels.com Signed-off-by: Ingo Molnar diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 2aedacc..7c70201 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4242,7 +4242,7 @@ static void update_cfs_rq_h_load(struct cfs_rq *cfs_rq) } if (!se) { - cfs_rq->h_load = rq->avg.load_avg_contrib; + cfs_rq->h_load = cfs_rq->runnable_load_avg; cfs_rq->last_h_load_update = now; } -- cgit v0.10.2 From 5ea330a75bd86b2b2a01d7b85c516983238306fb Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Wed, 18 Sep 2013 19:14:22 -0400 Subject: dm snapshot: workaround for a false positive lockdep warning The kernel reports a lockdep warning if a snapshot is invalidated because it runs out of space. The lockdep warning was triggered by commit 0976dfc1d0cd80a4e9dfaf87bd87 ("workqueue: Catch more locking problems with flush_work()") in v3.5. The warning is false positive. The real cause for the warning is that the lockdep engine treats different instances of md->lock as a single lock. This patch is a workaround - we use flush_workqueue instead of flush_work. This code path is not performance sensitive (it is called only on initialization or invalidation), thus it doesn't matter that we flush the whole workqueue. The real fix for the problem would be to teach the lockdep engine to treat different instances of md->lock as separate locks. Signed-off-by: Mikulas Patocka Acked-by: Alasdair G Kergon Signed-off-by: Mike Snitzer Cc: stable@vger.kernel.org # 3.5+ diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index 3ac4156..4caa8e6 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -256,7 +256,7 @@ static int chunk_io(struct pstore *ps, void *area, chunk_t chunk, int rw, */ INIT_WORK_ONSTACK(&req.work, do_metadata); queue_work(ps->metadata_wq, &req.work); - flush_work(&req.work); + flush_workqueue(ps->metadata_wq); return req.result; } -- cgit v0.10.2 From 60e356f381954d79088d0455e357db48cfdd6857 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Wed, 18 Sep 2013 19:40:42 -0400 Subject: dm-snapshot: fix performance degradation due to small hash size LVM2, since version 2.02.96, creates origin with zero size, then loads the snapshot driver and then loads the origin. Consequently, the snapshot driver sees the origin size zero and sets the hash size to the lower bound 64. Such small hash table causes performance degradation. This patch changes it so that the hash size is determined by the size of snapshot volume, not minimum of origin and snapshot size. It doesn't make sense to set the snapshot size significantly larger than the origin size, so we do not need to take origin size into account when calculating the hash size. Signed-off-by: Mikulas Patocka Signed-off-by: Mike Snitzer Cc: stable@vger.kernel.org diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index c434e5a..aec57d7 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -725,17 +725,16 @@ static int calc_max_buckets(void) */ static int init_hash_tables(struct dm_snapshot *s) { - sector_t hash_size, cow_dev_size, origin_dev_size, max_buckets; + sector_t hash_size, cow_dev_size, max_buckets; /* * Calculate based on the size of the original volume or * the COW volume... */ cow_dev_size = get_dev_size(s->cow->bdev); - origin_dev_size = get_dev_size(s->origin->bdev); max_buckets = calc_max_buckets(); - hash_size = min(origin_dev_size, cow_dev_size) >> s->store->chunk_shift; + hash_size = cow_dev_size >> s->store->chunk_shift; hash_size = min(hash_size, max_buckets); if (hash_size < 64) -- cgit v0.10.2 From f84cb8a46a771f36a04a02c61ea635c968ed5f6a Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Thu, 19 Sep 2013 12:13:58 -0400 Subject: dm mpath: disable WRITE SAME if it fails Workaround the SCSI layer's problematic WRITE SAME heuristics by disabling WRITE SAME in the DM multipath device's queue_limits if an underlying device disabled it. The WRITE SAME heuristics, with both the original commit 5db44863b6eb ("[SCSI] sd: Implement support for WRITE SAME") and the updated commit 66c28f971 ("[SCSI] sd: Update WRITE SAME heuristics"), default to enabling WRITE SAME(10) even without successfully determining it is supported. After the first failed WRITE SAME the SCSI layer will disable WRITE SAME for the device (by setting sdkp->device->no_write_same which results in 'max_write_same_sectors' in device's queue_limits to be set to 0). When a device is stacked ontop of such a SCSI device any changes to that SCSI device's queue_limits do not automatically propagate up the stack. As such, a DM multipath device will not have its WRITE SAME support disabled. This causes the block layer to continue to issue WRITE SAME requests to the mpath device which causes paths to fail and (if mpath IO isn't configured to queue when no paths are available) it will result in actual IO errors to the upper layers. This fix doesn't help configurations that have additional devices stacked ontop of the mpath device (e.g. LVM created linear DM devices ontop). A proper fix that restacks all the queue_limits from the bottom of the device stack up will need to be explored if SCSI will continue to use this model of optimistically allowing op codes and then disabling them after they fail for the first time. Before this patch: EXT4-fs (dm-6): mounted filesystem with ordered data mode. Opts: (null) device-mapper: multipath: XXX snitm debugging: got -EREMOTEIO (-121) device-mapper: multipath: XXX snitm debugging: failing WRITE SAME IO with error=-121 end_request: critical target error, dev dm-6, sector 528 dm-6: WRITE SAME failed. Manually zeroing. device-mapper: multipath: Failing path 8:112. end_request: I/O error, dev dm-6, sector 4616 dm-6: WRITE SAME failed. Manually zeroing. end_request: I/O error, dev dm-6, sector 4616 end_request: I/O error, dev dm-6, sector 5640 end_request: I/O error, dev dm-6, sector 6664 end_request: I/O error, dev dm-6, sector 7688 end_request: I/O error, dev dm-6, sector 524288 Buffer I/O error on device dm-6, logical block 65536 lost page write due to I/O error on dm-6 JBD2: Error -5 detected when updating journal superblock for dm-6-8. end_request: I/O error, dev dm-6, sector 524296 Aborting journal on device dm-6-8. end_request: I/O error, dev dm-6, sector 524288 Buffer I/O error on device dm-6, logical block 65536 lost page write due to I/O error on dm-6 JBD2: Error -5 detected when updating journal superblock for dm-6-8. # cat /sys/block/sdh/queue/write_same_max_bytes 0 # cat /sys/block/dm-6/queue/write_same_max_bytes 33553920 After this patch: EXT4-fs (dm-6): mounted filesystem with ordered data mode. Opts: (null) device-mapper: multipath: XXX snitm debugging: got -EREMOTEIO (-121) device-mapper: multipath: XXX snitm debugging: WRITE SAME I/O failed with error=-121 end_request: critical target error, dev dm-6, sector 528 dm-6: WRITE SAME failed. Manually zeroing. # cat /sys/block/sdh/queue/write_same_max_bytes 0 # cat /sys/block/dm-6/queue/write_same_max_bytes 0 It should be noted that WRITE SAME support wasn't enabled in DM multipath until v3.10. Signed-off-by: Mike Snitzer Cc: Martin K. Petersen Cc: Hannes Reinecke Cc: stable@vger.kernel.org # 3.10+ diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 075747e..e08be94 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -1299,8 +1299,17 @@ static int do_end_io(struct multipath *m, struct request *clone, if (!error && !clone->errors) return 0; /* I/O complete */ - if (noretry_error(error)) + if (noretry_error(error)) { + if ((clone->cmd_flags & REQ_WRITE_SAME) && + !clone->q->limits.max_write_same_sectors) { + struct queue_limits *limits; + + /* device doesn't really support WRITE SAME, disable it */ + limits = dm_get_queue_limits(dm_table_get_md(m->ti->table)); + limits->max_write_same_sectors = 0; + } return error; + } if (mpio->pgpath) fail_path(mpio->pgpath); diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 6a5e9ed..7b0ccdc 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -2278,6 +2278,17 @@ struct target_type *dm_get_immutable_target_type(struct mapped_device *md) } /* + * The queue_limits are only valid as long as you have a reference + * count on 'md'. + */ +struct queue_limits *dm_get_queue_limits(struct mapped_device *md) +{ + BUG_ON(!atomic_read(&md->holders)); + return &md->queue->limits; +} +EXPORT_SYMBOL_GPL(dm_get_queue_limits); + +/* * Fully initialize a request-based queue (->elevator, ->request_fn, etc). */ static int dm_init_request_based_queue(struct mapped_device *md) diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 653073d..ed419c6 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -406,13 +406,14 @@ int dm_noflush_suspending(struct dm_target *ti); union map_info *dm_get_mapinfo(struct bio *bio); union map_info *dm_get_rq_mapinfo(struct request *rq); +struct queue_limits *dm_get_queue_limits(struct mapped_device *md); + /* * Geometry functions. */ int dm_get_geometry(struct mapped_device *md, struct hd_geometry *geo); int dm_set_geometry(struct mapped_device *md, struct hd_geometry *geo); - /*----------------------------------------------------------------- * Functions for manipulating device-mapper tables. *---------------------------------------------------------------*/ -- cgit v0.10.2 From 4ca5a6cba53e13b8fd153b0762b4128fab6a3cfb Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sun, 15 Sep 2013 23:23:07 -0400 Subject: drm/radeon: avoid UVD corruption on AGP cards using GPU gart MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If the user has forced the driver to use the internal GPU gart rather than AGP on an AGP card, force the buffers to vram as well. Signed-off-by: Alex Deucher Reviewed-by: Christian König Tested-by: Dieter Nützel Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index ac6ece6..80285e3 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -85,7 +85,7 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) VRAM, also but everything into VRAM on AGP cards to avoid image corruptions */ if (p->ring == R600_RING_TYPE_UVD_INDEX && - (i == 0 || p->rdev->flags & RADEON_IS_AGP)) { + (i == 0 || drm_pci_device_is_agp(p->rdev->ddev))) { /* TODO: is this still needed for NI+ ? */ p->relocs[i].lobj.domain = RADEON_GEM_DOMAIN_VRAM; -- cgit v0.10.2 From a7ee824a6255e347ea76e2f00827e81bbe01004e Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 16 Sep 2013 17:46:00 -0400 Subject: drm/radeon: additional gcc fixes for radeon_atombios.c Newer versions of gcc seem to wander off into the weeds when dealing with variable sizes arrays in structs. Rather than indexing the arrays, use pointer arithmetic. Fix up spread spectrum tables. See bugs: https://bugs.freedesktop.org/show_bug.cgi?id=66932 https://bugs.freedesktop.org/show_bug.cgi?id=66972 https://bugs.freedesktop.org/show_bug.cgi?id=66945 Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index 404e25d..f79ee18 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -1367,6 +1367,7 @@ bool radeon_atombios_get_ppll_ss_info(struct radeon_device *rdev, int index = GetIndexIntoMasterTable(DATA, PPLL_SS_Info); uint16_t data_offset, size; struct _ATOM_SPREAD_SPECTRUM_INFO *ss_info; + struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT *ss_assign; uint8_t frev, crev; int i, num_indices; @@ -1378,18 +1379,21 @@ bool radeon_atombios_get_ppll_ss_info(struct radeon_device *rdev, num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) / sizeof(ATOM_SPREAD_SPECTRUM_ASSIGNMENT); - + ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT*) + ((u8 *)&ss_info->asSS_Info[0]); for (i = 0; i < num_indices; i++) { - if (ss_info->asSS_Info[i].ucSS_Id == id) { + if (ss_assign->ucSS_Id == id) { ss->percentage = - le16_to_cpu(ss_info->asSS_Info[i].usSpreadSpectrumPercentage); - ss->type = ss_info->asSS_Info[i].ucSpreadSpectrumType; - ss->step = ss_info->asSS_Info[i].ucSS_Step; - ss->delay = ss_info->asSS_Info[i].ucSS_Delay; - ss->range = ss_info->asSS_Info[i].ucSS_Range; - ss->refdiv = ss_info->asSS_Info[i].ucRecommendedRef_Div; + le16_to_cpu(ss_assign->usSpreadSpectrumPercentage); + ss->type = ss_assign->ucSpreadSpectrumType; + ss->step = ss_assign->ucSS_Step; + ss->delay = ss_assign->ucSS_Delay; + ss->range = ss_assign->ucSS_Range; + ss->refdiv = ss_assign->ucRecommendedRef_Div; return true; } + ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT*) + ((u8 *)ss_assign + sizeof(struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT)); } } return false; @@ -1477,6 +1481,12 @@ union asic_ss_info { struct _ATOM_ASIC_INTERNAL_SS_INFO_V3 info_3; }; +union asic_ss_assignment { + struct _ATOM_ASIC_SS_ASSIGNMENT v1; + struct _ATOM_ASIC_SS_ASSIGNMENT_V2 v2; + struct _ATOM_ASIC_SS_ASSIGNMENT_V3 v3; +}; + bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev, struct radeon_atom_ss *ss, int id, u32 clock) @@ -1485,6 +1495,7 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev, int index = GetIndexIntoMasterTable(DATA, ASIC_InternalSS_Info); uint16_t data_offset, size; union asic_ss_info *ss_info; + union asic_ss_assignment *ss_assign; uint8_t frev, crev; int i, num_indices; @@ -1509,45 +1520,52 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev, num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) / sizeof(ATOM_ASIC_SS_ASSIGNMENT); + ss_assign = (union asic_ss_assignment *)((u8 *)&ss_info->info.asSpreadSpectrum[0]); for (i = 0; i < num_indices; i++) { - if ((ss_info->info.asSpreadSpectrum[i].ucClockIndication == id) && - (clock <= le32_to_cpu(ss_info->info.asSpreadSpectrum[i].ulTargetClockRange))) { + if ((ss_assign->v1.ucClockIndication == id) && + (clock <= le32_to_cpu(ss_assign->v1.ulTargetClockRange))) { ss->percentage = - le16_to_cpu(ss_info->info.asSpreadSpectrum[i].usSpreadSpectrumPercentage); - ss->type = ss_info->info.asSpreadSpectrum[i].ucSpreadSpectrumMode; - ss->rate = le16_to_cpu(ss_info->info.asSpreadSpectrum[i].usSpreadRateInKhz); + le16_to_cpu(ss_assign->v1.usSpreadSpectrumPercentage); + ss->type = ss_assign->v1.ucSpreadSpectrumMode; + ss->rate = le16_to_cpu(ss_assign->v1.usSpreadRateInKhz); return true; } + ss_assign = (union asic_ss_assignment *) + ((u8 *)ss_assign + sizeof(ATOM_ASIC_SS_ASSIGNMENT)); } break; case 2: num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) / sizeof(ATOM_ASIC_SS_ASSIGNMENT_V2); + ss_assign = (union asic_ss_assignment *)((u8 *)&ss_info->info_2.asSpreadSpectrum[0]); for (i = 0; i < num_indices; i++) { - if ((ss_info->info_2.asSpreadSpectrum[i].ucClockIndication == id) && - (clock <= le32_to_cpu(ss_info->info_2.asSpreadSpectrum[i].ulTargetClockRange))) { + if ((ss_assign->v2.ucClockIndication == id) && + (clock <= le32_to_cpu(ss_assign->v2.ulTargetClockRange))) { ss->percentage = - le16_to_cpu(ss_info->info_2.asSpreadSpectrum[i].usSpreadSpectrumPercentage); - ss->type = ss_info->info_2.asSpreadSpectrum[i].ucSpreadSpectrumMode; - ss->rate = le16_to_cpu(ss_info->info_2.asSpreadSpectrum[i].usSpreadRateIn10Hz); + le16_to_cpu(ss_assign->v2.usSpreadSpectrumPercentage); + ss->type = ss_assign->v2.ucSpreadSpectrumMode; + ss->rate = le16_to_cpu(ss_assign->v2.usSpreadRateIn10Hz); if ((crev == 2) && ((id == ASIC_INTERNAL_ENGINE_SS) || (id == ASIC_INTERNAL_MEMORY_SS))) ss->rate /= 100; return true; } + ss_assign = (union asic_ss_assignment *) + ((u8 *)ss_assign + sizeof(ATOM_ASIC_SS_ASSIGNMENT_V2)); } break; case 3: num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) / sizeof(ATOM_ASIC_SS_ASSIGNMENT_V3); + ss_assign = (union asic_ss_assignment *)((u8 *)&ss_info->info_3.asSpreadSpectrum[0]); for (i = 0; i < num_indices; i++) { - if ((ss_info->info_3.asSpreadSpectrum[i].ucClockIndication == id) && - (clock <= le32_to_cpu(ss_info->info_3.asSpreadSpectrum[i].ulTargetClockRange))) { + if ((ss_assign->v3.ucClockIndication == id) && + (clock <= le32_to_cpu(ss_assign->v3.ulTargetClockRange))) { ss->percentage = - le16_to_cpu(ss_info->info_3.asSpreadSpectrum[i].usSpreadSpectrumPercentage); - ss->type = ss_info->info_3.asSpreadSpectrum[i].ucSpreadSpectrumMode; - ss->rate = le16_to_cpu(ss_info->info_3.asSpreadSpectrum[i].usSpreadRateIn10Hz); + le16_to_cpu(ss_assign->v3.usSpreadSpectrumPercentage); + ss->type = ss_assign->v3.ucSpreadSpectrumMode; + ss->rate = le16_to_cpu(ss_assign->v3.usSpreadRateIn10Hz); if ((id == ASIC_INTERNAL_ENGINE_SS) || (id == ASIC_INTERNAL_MEMORY_SS)) ss->rate /= 100; @@ -1555,6 +1573,8 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev, radeon_atombios_get_igp_ss_overrides(rdev, ss, id); return true; } + ss_assign = (union asic_ss_assignment *) + ((u8 *)ss_assign + sizeof(ATOM_ASIC_SS_ASSIGNMENT_V3)); } break; default: -- cgit v0.10.2 From 328a50c7b09d313ab9278f972950da414d348eb1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Wed, 18 Sep 2013 15:39:40 +0200 Subject: drm/radeon/cik: Fix printing of client name on VM protection fault MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The string is encoded from the MSB to the LSB of the register. Cc: stable@vger.kernel.org Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index adbdb65..c0714d7 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -4735,12 +4735,13 @@ static void cik_vm_decode_fault(struct radeon_device *rdev, u32 mc_id = (status & MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT; u32 vmid = (status & FAULT_VMID_MASK) >> FAULT_VMID_SHIFT; u32 protections = (status & PROTECTIONS_MASK) >> PROTECTIONS_SHIFT; - char *block = (char *)&mc_client; + char block[5] = { mc_client >> 24, (mc_client >> 16) & 0xff, + (mc_client >> 8) & 0xff, mc_client & 0xff, 0 }; - printk("VM fault (0x%02x, vmid %d) at page %u, %s from %s (%d)\n", + printk("VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, addr, (status & MEMORY_CLIENT_RW_MASK) ? "write" : "read", - block, mc_id); + block, mc_client, mc_id); } /** -- cgit v0.10.2 From a537314e0b539e22934d3cffeb0b1f476e56491c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Wed, 18 Sep 2013 15:39:41 +0200 Subject: drm/radeon/cik: Fix encoding of number of banks in tiling configuration info MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are multiple valid values, not just 0 or 1. Required to properly support 2D tiling in the userspace drivers. Cc: stable@vger.kernel.org Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index c0714d7..5e6802d 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -2845,10 +2845,8 @@ static void cik_gpu_init(struct radeon_device *rdev) rdev->config.cik.tile_config |= (3 << 0); break; } - if ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) - rdev->config.cik.tile_config |= 1 << 4; - else - rdev->config.cik.tile_config |= 0 << 4; + rdev->config.cik.tile_config |= + ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) << 4; rdev->config.cik.tile_config |= ((gb_addr_config & PIPE_INTERLEAVE_SIZE_MASK) >> PIPE_INTERLEAVE_SIZE_SHIFT) << 8; rdev->config.cik.tile_config |= -- cgit v0.10.2 From 42baf21d91d4f52f5b9a4d11dc8e7b1e3b93de7c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Wed, 18 Sep 2013 18:23:51 +0200 Subject: drm/radeon/cik: Add tiling mode index for 1D tiled depth/stencil surfaces MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CIK uses a different index for 1D DST surfaces compared to SI. Expose the new index so libdrm_radeon can use it properly for userspace drivers. Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher diff --git a/include/uapi/drm/radeon_drm.h b/include/uapi/drm/radeon_drm.h index fa8b3ad..46d41e8 100644 --- a/include/uapi/drm/radeon_drm.h +++ b/include/uapi/drm/radeon_drm.h @@ -1007,4 +1007,6 @@ struct drm_radeon_info { #define SI_TILE_MODE_DEPTH_STENCIL_2D_4AA 3 #define SI_TILE_MODE_DEPTH_STENCIL_2D_8AA 2 +#define CIK_TILE_MODE_DEPTH_STENCIL_1D 5 + #endif -- cgit v0.10.2 From 0eb3448aa6b31fbf24c31756aba7940cac5ad6b8 Mon Sep 17 00:00:00 2001 From: Alex Ivanov Date: Fri, 20 Sep 2013 17:36:06 +0400 Subject: drm/radeon: Make r100_cp_ring_info() and radeon_ring_gfx() safe (v2) Prevent NULL pointer dereference in case when radeon_ring_fini() did it's job. Reading of r100_cp_ring_info and radeon_ring_gfx debugfs entries will lead to a KP if ring buffer was deallocated, e.g. on failed ring test. Seen on PA-RISC machine having "radeon: ring test failed (scratch(0x8504)=0xCAFEDEAD)" issue. v2: agd5f: add some parens around ring->ready check Signed-off-by: Alex Ivanov Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 2417571..d713330 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -2933,9 +2933,11 @@ static int r100_debugfs_cp_ring_info(struct seq_file *m, void *data) seq_printf(m, "CP_RB_RPTR 0x%08x\n", rdp); seq_printf(m, "%u free dwords in ring\n", ring->ring_free_dw); seq_printf(m, "%u dwords in ring\n", count); - for (j = 0; j <= count; j++) { - i = (rdp + j) & ring->ptr_mask; - seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]); + if (ring->ready) { + for (j = 0; j <= count; j++) { + i = (rdp + j) & ring->ptr_mask; + seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]); + } } return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c index 46a25f0..18254e1 100644 --- a/drivers/gpu/drm/radeon/radeon_ring.c +++ b/drivers/gpu/drm/radeon/radeon_ring.c @@ -839,9 +839,11 @@ static int radeon_debugfs_ring_info(struct seq_file *m, void *data) * packet that is the root issue */ i = (ring->rptr + ring->ptr_mask + 1 - 32) & ring->ptr_mask; - for (j = 0; j <= (count + 32); j++) { - seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]); - i = (i + 1) & ring->ptr_mask; + if (ring->ready) { + for (j = 0; j <= (count + 32); j++) { + seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]); + i = (i + 1) & ring->ptr_mask; + } } return 0; } -- cgit v0.10.2 From d95bc2501da97e2884b957c48df37c258a34db8d Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 19 Sep 2013 10:32:20 +0100 Subject: ARM: 7839/1: entry: fix tracing of ARM-private syscalls Commit 377747c40657 ("ARM: entry: allow ARM-private syscalls to be restarted") reworked the low-level syscall dispatcher to allow restarting of ARM-private syscalls. Unfortunately, this relocated the label used to dispatch a private syscall from the trace path, so that the invocation would be bypassed altogether! This causes applications to fail under strace as soon as they rely on a private syscall (e.g. set_tls): set_tls(0xb6fad4c0, 0xb6fadb98, 0xb6fb1050, 0xb6fad4c0, 0xb6fb1050) = -1 ENOSYS (Function not implemented) This patch fixes the label so that we correctly dispatch private syscalls from the trace path. Reported-by: Jason Gunthorpe Tested-by: Jason Gunthorpe Signed-off-by: Will Deacon Signed-off-by: Russell King diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index 74ad15d1..bc6bd96 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -442,10 +442,10 @@ local_restart: ldrcc pc, [tbl, scno, lsl #2] @ call sys_* routine add r1, sp, #S_OFF - cmp scno, #(__ARM_NR_BASE - __NR_SYSCALL_BASE) +2: cmp scno, #(__ARM_NR_BASE - __NR_SYSCALL_BASE) eor r0, scno, #__NR_SYSCALL_BASE @ put OS number back bcs arm_syscall -2: mov why, #0 @ no longer a real syscall + mov why, #0 @ no longer a real syscall b sys_ni_syscall @ not private func #if defined(CONFIG_OABI_COMPAT) || !defined(CONFIG_AEABI) -- cgit v0.10.2 From c4a30c3b2997bbc7d81cd0f5fde43599700834f5 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 22 Sep 2013 11:08:50 +0100 Subject: ARM: only allow kernel mode neon with AEABI This prevents the linker erroring with: arm-linux-ld: error: arch/arm/lib/xor-neon.o uses VFP instructions, whereas arch/arm/lib/built-in.o does not arm-linux-ld: failed to merge target specific data of file arch/arm/lib/xor-neon.o This is due to the non-neon files being marked as containing FPA data/ instructions (even though they do not) being mixed with files which contain VFP, which is an incompatible floating point format. Signed-off-by: Russell King diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 3f7714d..1ad6fb6 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -2217,8 +2217,7 @@ config NEON config KERNEL_MODE_NEON bool "Support for NEON in kernel mode" - default n - depends on NEON + depends on NEON && AEABI help Say Y to include support for NEON in kernel mode. -- cgit v0.10.2 From 40190c85f427dcfdbab5dbef4ffd2510d649da1f Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Sat, 21 Sep 2013 11:23:50 +0100 Subject: ARM: 7837/3: fix Thumb-2 bug in AES assembler code Patch 638591c enabled building the AES assembler code in Thumb2 mode. However, this code used arithmetic involving PC rather than adr{l} instructions to generate PC-relative references to the lookup tables, and this needs to take into account the different PC offset when running in Thumb mode. Signed-off-by: Ard Biesheuvel Acked-by: Nicolas Pitre Cc: stable@vger.kernel.org Signed-off-by: Russell King diff --git a/arch/arm/crypto/aes-armv4.S b/arch/arm/crypto/aes-armv4.S index 19d6cd6..3a14ea8 100644 --- a/arch/arm/crypto/aes-armv4.S +++ b/arch/arm/crypto/aes-armv4.S @@ -148,7 +148,7 @@ AES_Te: @ const AES_KEY *key) { .align 5 ENTRY(AES_encrypt) - sub r3,pc,#8 @ AES_encrypt + adr r3,AES_encrypt stmdb sp!,{r1,r4-r12,lr} mov r12,r0 @ inp mov r11,r2 @@ -381,7 +381,7 @@ _armv4_AES_encrypt: .align 5 ENTRY(private_AES_set_encrypt_key) _armv4_AES_set_encrypt_key: - sub r3,pc,#8 @ AES_set_encrypt_key + adr r3,_armv4_AES_set_encrypt_key teq r0,#0 moveq r0,#-1 beq .Labrt @@ -843,7 +843,7 @@ AES_Td: @ const AES_KEY *key) { .align 5 ENTRY(AES_decrypt) - sub r3,pc,#8 @ AES_decrypt + adr r3,AES_decrypt stmdb sp!,{r1,r4-r12,lr} mov r12,r0 @ inp mov r11,r2 -- cgit v0.10.2 From cf3b425dd8d99e01214515a6754f9e69ecc6dce8 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Sun, 22 Sep 2013 16:19:13 +0800 Subject: perf/x86/intel: Add model number for Avoton Silvermont Signed-off-by: Yan, Zheng Cc: a.p.zijlstra@chello.nl Cc: eranian@google.com Cc: ak@linux.intel.com Link: http://lkml.kernel.org/r/1379837953-17755-1-git-send-email-zheng.z.yan@intel.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 9db76c3..f31a165 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -2325,6 +2325,7 @@ __init int intel_pmu_init(void) break; case 55: /* Atom 22nm "Silvermont" */ + case 77: /* Avoton "Silvermont" */ memcpy(hw_cache_event_ids, slm_hw_cache_event_ids, sizeof(hw_cache_event_ids)); memcpy(hw_cache_extra_regs, slm_hw_cache_extra_regs, -- cgit v0.10.2 From 4f0acd31c31f03ba42494c8baf6c0465150e2621 Mon Sep 17 00:00:00 2001 From: Masoud Sharbiani Date: Fri, 20 Sep 2013 15:59:07 -0700 Subject: x86/reboot: Add quirk to make Dell C6100 use reboot=pci automatically Dell PowerEdge C6100 machines fail to completely reboot about 20% of the time. Signed-off-by: Masoud Sharbiani Signed-off-by: Vinson Lee Cc: Robin Holt Cc: Russell King Cc: Guan Xuetao Cc: Link: http://lkml.kernel.org/r/1379717947-18042-1-git-send-email-vlee@freedesktop.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 563ed91..5f4ad27 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -358,6 +358,22 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Precision M6600"), }, }, + { /* Handle problems with rebooting on the Dell PowerEdge C6100. */ + .callback = set_pci_reboot, + .ident = "Dell PowerEdge C6100", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "C6100"), + }, + }, + { /* Some C6100 machines were shipped with vendor being 'Dell'. */ + .callback = set_pci_reboot, + .ident = "Dell PowerEdge C6100", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "C6100"), + }, + }, { } }; -- cgit v0.10.2 From 142a4e9079e4c81a003bd966a6bf5ca7f466057a Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 21 Sep 2013 07:55:36 -0400 Subject: drm/radeon: fix missed variable sized access MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit I missed this when I fixed up this file. Noticed-by: Mathias Fröhlich Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c index e65f211..5513d8f 100644 --- a/drivers/gpu/drm/radeon/r600_dpm.c +++ b/drivers/gpu/drm/radeon/r600_dpm.c @@ -1084,7 +1084,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev) rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].dclk = le16_to_cpu(uvd_clk->usDClkLow) | (uvd_clk->ucDClkHigh << 16); rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].v = - le16_to_cpu(limits->entries[i].usVoltage); + le16_to_cpu(entry->usVoltage); entry = (ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record *) ((u8 *)entry + sizeof(ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record)); } -- cgit v0.10.2 From 7102e23288f9fb6f1c53a7ed27707cf1b4f22227 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 21 Sep 2013 13:47:57 -0400 Subject: drm/radeon/dpm: fetch the max clk from voltage dep tables helper This patch adds a helper function to fetch the max clock from the voltage clock dependecy tables. Clocks above that level tend to be unstable and will require additional driver tweaks in order to work properly. This patch implemented the helper function to fetch the max clocks from the dependency tables. The following patches implement the per-asic clock filtering. See bug: https://bugs.freedesktop.org/show_bug.cgi?id=68235 Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c index 05ff315..0d0f065 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.c +++ b/drivers/gpu/drm/radeon/btc_dpm.c @@ -1168,6 +1168,23 @@ static const struct radeon_blacklist_clocks btc_blacklist_clocks[] = { 25000, 30000, RADEON_SCLK_UP } }; +void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table, + u32 *max_clock) +{ + u32 i, clock = 0; + + if ((table == NULL) || (table->count == 0)) { + *max_clock = clock; + return; + } + + for (i = 0; i < table->count; i++) { + if (clock < table->entries[i].clk) + clock = table->entries[i].clk; + } + *max_clock = clock; +} + void btc_apply_voltage_dependency_rules(struct radeon_clock_voltage_dependency_table *table, u32 clock, u16 max_voltage, u16 *voltage) { diff --git a/drivers/gpu/drm/radeon/btc_dpm.h b/drivers/gpu/drm/radeon/btc_dpm.h index 1a15e0e..3b6f12b 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.h +++ b/drivers/gpu/drm/radeon/btc_dpm.h @@ -46,6 +46,8 @@ void btc_adjust_clock_combinations(struct radeon_device *rdev, struct rv7xx_pl *pl); void btc_apply_voltage_dependency_rules(struct radeon_clock_voltage_dependency_table *table, u32 clock, u16 max_voltage, u16 *voltage); +void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table, + u32 *max_clock); void btc_apply_voltage_delta_rules(struct radeon_device *rdev, u16 max_vddc, u16 max_vddci, u16 *vddc, u16 *vddci); -- cgit v0.10.2 From 1f28fb9271b5bd8a232d11e8e3b9a25e47aefac4 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 21 Sep 2013 13:59:15 -0400 Subject: drm/radeon/dpm/btc: filter clocks based on voltage/clk dep tables Filter out mclk and sclk levels higher than listed in the clk voltage dependency tables. Supporting these clocks will require additional driver tweaking that isn't supported yet. See bug: https://bugs.freedesktop.org/show_bug.cgi?id=68235 Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c index 0d0f065..b162e98 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.c +++ b/drivers/gpu/drm/radeon/btc_dpm.c @@ -2097,6 +2097,7 @@ static void btc_apply_state_adjust_rules(struct radeon_device *rdev, bool disable_mclk_switching; u32 mclk, sclk; u16 vddc, vddci; + u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc; if ((rdev->pm.dpm.new_active_crtc_count > 1) || btc_dpm_vblank_too_short(rdev)) @@ -2138,6 +2139,39 @@ static void btc_apply_state_adjust_rules(struct radeon_device *rdev, ps->low.vddci = max_limits->vddci; } + /* limit clocks to max supported clocks based on voltage dependency tables */ + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk, + &max_sclk_vddc); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk, + &max_mclk_vddci); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk, + &max_mclk_vddc); + + if (max_sclk_vddc) { + if (ps->low.sclk > max_sclk_vddc) + ps->low.sclk = max_sclk_vddc; + if (ps->medium.sclk > max_sclk_vddc) + ps->medium.sclk = max_sclk_vddc; + if (ps->high.sclk > max_sclk_vddc) + ps->high.sclk = max_sclk_vddc; + } + if (max_mclk_vddci) { + if (ps->low.mclk > max_mclk_vddci) + ps->low.mclk = max_mclk_vddci; + if (ps->medium.mclk > max_mclk_vddci) + ps->medium.mclk = max_mclk_vddci; + if (ps->high.mclk > max_mclk_vddci) + ps->high.mclk = max_mclk_vddci; + } + if (max_mclk_vddc) { + if (ps->low.mclk > max_mclk_vddc) + ps->low.mclk = max_mclk_vddc; + if (ps->medium.mclk > max_mclk_vddc) + ps->medium.mclk = max_mclk_vddc; + if (ps->high.mclk > max_mclk_vddc) + ps->high.mclk = max_mclk_vddc; + } + /* XXX validate the min clocks required for display */ if (disable_mclk_switching) { -- cgit v0.10.2 From 225b769d3ee96fd55ec19ea5ed3cb72bfff32ce5 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 21 Sep 2013 14:05:48 -0400 Subject: drm/radeon/dpm/ni: filter clocks based on voltage/clk dep tables filter out mclk and sclk levels higher than listed in the clk voltage dependency tables. Supporting these clocks will require additional driver tweaking that isn't supported yet. See bug: https://bugs.freedesktop.org/show_bug.cgi?id=68235 Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c index 6c398a4..f263390 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.c +++ b/drivers/gpu/drm/radeon/ni_dpm.c @@ -787,6 +787,7 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev, bool disable_mclk_switching; u32 mclk, sclk; u16 vddc, vddci; + u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc; int i; if ((rdev->pm.dpm.new_active_crtc_count > 1) || @@ -813,6 +814,29 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev, } } + /* limit clocks to max supported clocks based on voltage dependency tables */ + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk, + &max_sclk_vddc); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk, + &max_mclk_vddci); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk, + &max_mclk_vddc); + + for (i = 0; i < ps->performance_level_count; i++) { + if (max_sclk_vddc) { + if (ps->performance_levels[i].sclk > max_sclk_vddc) + ps->performance_levels[i].sclk = max_sclk_vddc; + } + if (max_mclk_vddci) { + if (ps->performance_levels[i].mclk > max_mclk_vddci) + ps->performance_levels[i].mclk = max_mclk_vddci; + } + if (max_mclk_vddc) { + if (ps->performance_levels[i].mclk > max_mclk_vddc) + ps->performance_levels[i].mclk = max_mclk_vddc; + } + } + /* XXX validate the min clocks required for display */ if (disable_mclk_switching) { -- cgit v0.10.2 From 78fbdf0e9dcd84d54044fa1470880b689e34f9c6 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 21 Sep 2013 14:08:28 -0400 Subject: drm/radeon/dpm/si: filter clocks based on voltage/clk dep tables Filter out mclk and sclk levels higher than listed in the clk voltage dependency tables. Supporting these clocks will require additional driver tweaking that isn't supported yet. See bug: https://bugs.freedesktop.org/show_bug.cgi?id=68235 Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index cfe5d4d..9ace287 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2910,6 +2910,7 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev, bool disable_sclk_switching = false; u32 mclk, sclk; u16 vddc, vddci; + u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc; int i; if ((rdev->pm.dpm.new_active_crtc_count > 1) || @@ -2943,6 +2944,29 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev, } } + /* limit clocks to max supported clocks based on voltage dependency tables */ + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk, + &max_sclk_vddc); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk, + &max_mclk_vddci); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk, + &max_mclk_vddc); + + for (i = 0; i < ps->performance_level_count; i++) { + if (max_sclk_vddc) { + if (ps->performance_levels[i].sclk > max_sclk_vddc) + ps->performance_levels[i].sclk = max_sclk_vddc; + } + if (max_mclk_vddci) { + if (ps->performance_levels[i].mclk > max_mclk_vddci) + ps->performance_levels[i].mclk = max_mclk_vddci; + } + if (max_mclk_vddc) { + if (ps->performance_levels[i].mclk > max_mclk_vddc) + ps->performance_levels[i].mclk = max_mclk_vddc; + } + } + /* XXX validate the min clocks required for display */ if (disable_mclk_switching) { -- cgit v0.10.2 From a52b5eb647c728252b75365a0c44d78836661b27 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 21 Sep 2013 14:16:01 -0400 Subject: drm/radeon/dpm/ci: filter clocks based on voltage/clk dep tables Filter out mclk and sclk levels higher than listed in the clk voltage dependency tables. Supporting these clocks will require additional driver tweaking that isn't supported yet. See bug: https://bugs.freedesktop.org/show_bug.cgi?id=68235 Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index 8996274..51e947a 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -146,6 +146,8 @@ static const struct ci_pt_config_reg didt_config_ci[] = }; extern u8 rv770_get_memory_module_index(struct radeon_device *rdev); +extern void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table, + u32 *max_clock); extern int ni_copy_and_switch_arb_sets(struct radeon_device *rdev, u32 arb_freq_src, u32 arb_freq_dest); extern u8 si_get_ddr3_mclk_frequency_ratio(u32 memory_clock); @@ -712,6 +714,7 @@ static void ci_apply_state_adjust_rules(struct radeon_device *rdev, struct radeon_clock_and_voltage_limits *max_limits; bool disable_mclk_switching; u32 sclk, mclk; + u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc; int i; if ((rdev->pm.dpm.new_active_crtc_count > 1) || @@ -739,6 +742,29 @@ static void ci_apply_state_adjust_rules(struct radeon_device *rdev, } } + /* limit clocks to max supported clocks based on voltage dependency tables */ + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk, + &max_sclk_vddc); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk, + &max_mclk_vddci); + btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk, + &max_mclk_vddc); + + for (i = 0; i < ps->performance_level_count; i++) { + if (max_sclk_vddc) { + if (ps->performance_levels[i].sclk > max_sclk_vddc) + ps->performance_levels[i].sclk = max_sclk_vddc; + } + if (max_mclk_vddci) { + if (ps->performance_levels[i].mclk > max_mclk_vddci) + ps->performance_levels[i].mclk = max_mclk_vddci; + } + if (max_mclk_vddc) { + if (ps->performance_levels[i].mclk > max_mclk_vddc) + ps->performance_levels[i].mclk = max_mclk_vddc; + } + } + /* XXX validate the min clocks required for display */ if (disable_mclk_switching) { -- cgit v0.10.2 From 360991867d83e10827d907ef67206986a98953b3 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 21 Sep 2013 14:37:49 -0400 Subject: drm/radeon: don't set default clocks for SI when DPM is disabled This is a partial revert of c6cf7777a32da874fabec4fd1c2a579f0ba4e4dd. We need to take into account the clk voltage dependencies of the board. Not doing so can lead to stability issues on certain boards if the clks exceed the levels in the dep tables. DPM already takes that into account, so for optimal performance, use DPM. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index 87e1d69..ac07ad1 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -1002,7 +1002,7 @@ static void radeon_pm_resume_old(struct radeon_device *rdev) { /* set up the default clocks if the MC ucode is loaded */ if ((rdev->family >= CHIP_BARTS) && - (rdev->family <= CHIP_HAINAN) && + (rdev->family <= CHIP_CAYMAN) && rdev->mc_fw) { if (rdev->pm.default_vddc) radeon_atom_set_voltage(rdev, rdev->pm.default_vddc, @@ -1046,7 +1046,7 @@ static void radeon_pm_resume_dpm(struct radeon_device *rdev) if (ret) { DRM_ERROR("radeon: dpm resume failed\n"); if ((rdev->family >= CHIP_BARTS) && - (rdev->family <= CHIP_HAINAN) && + (rdev->family <= CHIP_CAYMAN) && rdev->mc_fw) { if (rdev->pm.default_vddc) radeon_atom_set_voltage(rdev, rdev->pm.default_vddc, @@ -1097,7 +1097,7 @@ static int radeon_pm_init_old(struct radeon_device *rdev) radeon_pm_init_profile(rdev); /* set up the default clocks if the MC ucode is loaded */ if ((rdev->family >= CHIP_BARTS) && - (rdev->family <= CHIP_HAINAN) && + (rdev->family <= CHIP_CAYMAN) && rdev->mc_fw) { if (rdev->pm.default_vddc) radeon_atom_set_voltage(rdev, rdev->pm.default_vddc, @@ -1183,7 +1183,7 @@ static int radeon_pm_init_dpm(struct radeon_device *rdev) if (ret) { rdev->pm.dpm_enabled = false; if ((rdev->family >= CHIP_BARTS) && - (rdev->family <= CHIP_HAINAN) && + (rdev->family <= CHIP_CAYMAN) && rdev->mc_fw) { if (rdev->pm.default_vddc) radeon_atom_set_voltage(rdev, rdev->pm.default_vddc, -- cgit v0.10.2 From b60ab990ccdf34b0159bf5ff52f4acee7c940d78 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Thu, 19 Sep 2013 18:49:11 -0400 Subject: dm thin: do not expose non-zero discard limits if discards disabled Fix issue where the block layer would stack the discard limits of the pool's data device even if the "ignore_discard" pool feature was specified. The pool and thin device(s) still had discards disabled because the QUEUE_FLAG_DISCARD request_queue flag wasn't set. But to avoid user confusion when "ignore_discard" is used: both the pool device and the thin device(s) have zeroes for all discard limits. Also, always set discard_zeroes_data_unsupported in targets because they should never advertise the 'discard_zeroes_data' capability (even if the pool's data device supports it). Signed-off-by: Mike Snitzer Acked-by: Joe Thornber diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index ed06342..2c0cf51 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -2095,6 +2095,7 @@ static int pool_ctr(struct dm_target *ti, unsigned argc, char **argv) * them down to the data device. The thin device's discard * processing will cause mappings to be removed from the btree. */ + ti->discard_zeroes_data_unsupported = true; if (pf.discard_enabled && pf.discard_passdown) { ti->num_discard_bios = 1; @@ -2104,7 +2105,6 @@ static int pool_ctr(struct dm_target *ti, unsigned argc, char **argv) * thin devices' discard limits consistent). */ ti->discards_supported = true; - ti->discard_zeroes_data_unsupported = true; } ti->private = pt; @@ -2689,8 +2689,16 @@ static void pool_io_hints(struct dm_target *ti, struct queue_limits *limits) * They get transferred to the live pool in bind_control_target() * called from pool_preresume(). */ - if (!pt->adjusted_pf.discard_enabled) + if (!pt->adjusted_pf.discard_enabled) { + /* + * Must explicitly disallow stacking discard limits otherwise the + * block layer will stack them if pool's data device has support. + * QUEUE_FLAG_DISCARD wouldn't be set but there is no way for the + * user to see that, so make sure to set all discard limits to 0. + */ + limits->discard_granularity = 0; return; + } disable_passdown_if_not_supported(pt); @@ -2826,10 +2834,10 @@ static int thin_ctr(struct dm_target *ti, unsigned argc, char **argv) ti->per_bio_data_size = sizeof(struct dm_thin_endio_hook); /* In case the pool supports discards, pass them on. */ + ti->discard_zeroes_data_unsupported = true; if (tc->pool->pf.discard_enabled) { ti->discards_supported = true; ti->num_discard_bios = 1; - ti->discard_zeroes_data_unsupported = true; /* Discard bios must be split on a block boundary */ ti->split_discard_bios = true; } -- cgit v0.10.2 From 6cfa58573f9b4a543005baa002e137820504c7af Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Thu, 12 Sep 2013 18:06:11 -0400 Subject: dm: lower bio-based mempool reservation Bio-based device mapper processing doesn't need larger mempools (like request-based DM does), so lower the number of reserved entries for bio-based operation. 16 was already used for bio-based DM's bioset but mistakenly wasn't used for it's _io_cache. Formalize difference between bio-based and request-based defaults by introducing RESERVED_BIO_BASED_IOS and RESERVED_REQUEST_BASED_IOS. (based on older code from Mikulas Patocka) Signed-off-by: Mike Snitzer Signed-off-by: Frank Mayhar Acked-by: Mikulas Patocka diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 7b0ccdc..f2e5a50 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -211,7 +211,8 @@ struct dm_md_mempools { struct bio_set *bs; }; -#define MIN_IOS 256 +#define RESERVED_BIO_BASED_IOS 16 +#define RESERVED_REQUEST_BASED_IOS 256 static struct kmem_cache *_io_cache; static struct kmem_cache *_rq_tio_cache; @@ -2873,18 +2874,18 @@ struct dm_md_mempools *dm_alloc_md_mempools(unsigned type, unsigned integrity, u if (type == DM_TYPE_BIO_BASED) { cachep = _io_cache; - pool_size = 16; + pool_size = RESERVED_BIO_BASED_IOS; front_pad = roundup(per_bio_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone); } else if (type == DM_TYPE_REQUEST_BASED) { cachep = _rq_tio_cache; - pool_size = MIN_IOS; + pool_size = RESERVED_REQUEST_BASED_IOS; front_pad = offsetof(struct dm_rq_clone_bio_info, clone); /* per_bio_data_size is not used. See __bind_mempools(). */ WARN_ON(per_bio_data_size != 0); } else goto out; - pools->io_pool = mempool_create_slab_pool(MIN_IOS, cachep); + pools->io_pool = mempool_create_slab_pool(pool_size, cachep); if (!pools->io_pool) goto out; -- cgit v0.10.2 From f47908269fb53d522a956b78612a0037f5faf8e7 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Thu, 12 Sep 2013 18:06:12 -0400 Subject: dm: add reserved_rq_based_ios module parameter Allow user to change the number of IOs that are reserved by request-based DM's mempools by writing to this file: /sys/module/dm_mod/parameters/reserved_rq_based_ios The default value is RESERVED_REQUEST_BASED_IOS (256). The maximum allowed value is RESERVED_MAX_IOS (1024). Export dm_get_reserved_rq_based_ios() for use by DM targets and core code. Switch to sizing dm-mpath's mempool using DM core's configurable 'reserved_rq_based_ios'. Signed-off-by: Mike Snitzer Signed-off-by: Frank Mayhar Acked-by: Mikulas Patocka diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index e08be94..de570a5 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -7,6 +7,7 @@ #include +#include "dm.h" #include "dm-path-selector.h" #include "dm-uevent.h" @@ -116,8 +117,6 @@ struct dm_mpath_io { typedef int (*action_fn) (struct pgpath *pgpath); -#define MIN_IOS 256 /* Mempool size */ - static struct kmem_cache *_mpio_cache; static struct workqueue_struct *kmultipathd, *kmpath_handlerd; @@ -190,6 +189,7 @@ static void free_priority_group(struct priority_group *pg, static struct multipath *alloc_multipath(struct dm_target *ti) { struct multipath *m; + unsigned min_ios = dm_get_reserved_rq_based_ios(); m = kzalloc(sizeof(*m), GFP_KERNEL); if (m) { @@ -202,7 +202,7 @@ static struct multipath *alloc_multipath(struct dm_target *ti) INIT_WORK(&m->trigger_event, trigger_event); init_waitqueue_head(&m->pg_init_wait); mutex_init(&m->work_mutex); - m->mpio_pool = mempool_create_slab_pool(MIN_IOS, _mpio_cache); + m->mpio_pool = mempool_create_slab_pool(min_ios, _mpio_cache); if (!m->mpio_pool) { kfree(m); return NULL; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index f2e5a50..1e85f1d 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -213,9 +213,41 @@ struct dm_md_mempools { #define RESERVED_BIO_BASED_IOS 16 #define RESERVED_REQUEST_BASED_IOS 256 +#define RESERVED_MAX_IOS 1024 static struct kmem_cache *_io_cache; static struct kmem_cache *_rq_tio_cache; +/* + * Request-based DM's mempools' reserved IOs set by the user. + */ +static unsigned reserved_rq_based_ios = RESERVED_REQUEST_BASED_IOS; + +static unsigned __dm_get_reserved_ios(unsigned *reserved_ios, + unsigned def, unsigned max) +{ + unsigned ios = ACCESS_ONCE(*reserved_ios); + unsigned modified_ios = 0; + + if (!ios) + modified_ios = def; + else if (ios > max) + modified_ios = max; + + if (modified_ios) { + (void)cmpxchg(reserved_ios, ios, modified_ios); + ios = modified_ios; + } + + return ios; +} + +unsigned dm_get_reserved_rq_based_ios(void) +{ + return __dm_get_reserved_ios(&reserved_rq_based_ios, + RESERVED_REQUEST_BASED_IOS, RESERVED_MAX_IOS); +} +EXPORT_SYMBOL_GPL(dm_get_reserved_rq_based_ios); + static int __init local_init(void) { int r = -ENOMEM; @@ -2878,7 +2910,7 @@ struct dm_md_mempools *dm_alloc_md_mempools(unsigned type, unsigned integrity, u front_pad = roundup(per_bio_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone); } else if (type == DM_TYPE_REQUEST_BASED) { cachep = _rq_tio_cache; - pool_size = RESERVED_REQUEST_BASED_IOS; + pool_size = dm_get_reserved_rq_based_ios(); front_pad = offsetof(struct dm_rq_clone_bio_info, clone); /* per_bio_data_size is not used. See __bind_mempools(). */ WARN_ON(per_bio_data_size != 0); @@ -2936,6 +2968,10 @@ module_exit(dm_exit); module_param(major, uint, 0); MODULE_PARM_DESC(major, "The major number of the device mapper"); + +module_param(reserved_rq_based_ios, uint, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(reserved_rq_based_ios, "Reserved IOs in request-based mempools"); + MODULE_DESCRIPTION(DM_NAME " driver"); MODULE_AUTHOR("Joe Thornber "); MODULE_LICENSE("GPL"); diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 5e604cc..1539650 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -184,6 +184,8 @@ void dm_free_md_mempools(struct dm_md_mempools *pools); /* * Helpers that are used by DM core */ +unsigned dm_get_reserved_rq_based_ios(void); + static inline bool dm_message_test_buffer_overflow(char *result, unsigned maxlen) { return !maxlen || strlen(result) + 1 >= maxlen; -- cgit v0.10.2 From e8603136cb04ec2d0c9b4b5be7a071fc003cb399 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Thu, 12 Sep 2013 18:06:12 -0400 Subject: dm: add reserved_bio_based_ios module parameter Allow user to change the number of IOs that are reserved by bio-based DM's mempools by writing to this file: /sys/module/dm_mod/parameters/reserved_bio_based_ios The default value is RESERVED_BIO_BASED_IOS (16). The maximum allowed value is RESERVED_MAX_IOS (1024). Export dm_get_reserved_bio_based_ios() for use by DM targets and core code. Switch to sizing dm-io's mempool and bioset using DM core's configurable 'reserved_bio_based_ios'. Signed-off-by: Mike Snitzer Signed-off-by: Frank Mayhar diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c index ea49834..2a20986 100644 --- a/drivers/md/dm-io.c +++ b/drivers/md/dm-io.c @@ -19,8 +19,6 @@ #define DM_MSG_PREFIX "io" #define DM_IO_MAX_REGIONS BITS_PER_LONG -#define MIN_IOS 16 -#define MIN_BIOS 16 struct dm_io_client { mempool_t *pool; @@ -50,16 +48,17 @@ static struct kmem_cache *_dm_io_cache; struct dm_io_client *dm_io_client_create(void) { struct dm_io_client *client; + unsigned min_ios = dm_get_reserved_bio_based_ios(); client = kmalloc(sizeof(*client), GFP_KERNEL); if (!client) return ERR_PTR(-ENOMEM); - client->pool = mempool_create_slab_pool(MIN_IOS, _dm_io_cache); + client->pool = mempool_create_slab_pool(min_ios, _dm_io_cache); if (!client->pool) goto bad; - client->bios = bioset_create(MIN_BIOS, 0); + client->bios = bioset_create(min_ios, 0); if (!client->bios) goto bad; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 1e85f1d..b3e26c7 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -218,6 +218,11 @@ static struct kmem_cache *_io_cache; static struct kmem_cache *_rq_tio_cache; /* + * Bio-based DM's mempools' reserved IOs set by the user. + */ +static unsigned reserved_bio_based_ios = RESERVED_BIO_BASED_IOS; + +/* * Request-based DM's mempools' reserved IOs set by the user. */ static unsigned reserved_rq_based_ios = RESERVED_REQUEST_BASED_IOS; @@ -241,6 +246,13 @@ static unsigned __dm_get_reserved_ios(unsigned *reserved_ios, return ios; } +unsigned dm_get_reserved_bio_based_ios(void) +{ + return __dm_get_reserved_ios(&reserved_bio_based_ios, + RESERVED_BIO_BASED_IOS, RESERVED_MAX_IOS); +} +EXPORT_SYMBOL_GPL(dm_get_reserved_bio_based_ios); + unsigned dm_get_reserved_rq_based_ios(void) { return __dm_get_reserved_ios(&reserved_rq_based_ios, @@ -2906,7 +2918,7 @@ struct dm_md_mempools *dm_alloc_md_mempools(unsigned type, unsigned integrity, u if (type == DM_TYPE_BIO_BASED) { cachep = _io_cache; - pool_size = RESERVED_BIO_BASED_IOS; + pool_size = dm_get_reserved_bio_based_ios(); front_pad = roundup(per_bio_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone); } else if (type == DM_TYPE_REQUEST_BASED) { cachep = _rq_tio_cache; @@ -2969,6 +2981,9 @@ module_exit(dm_exit); module_param(major, uint, 0); MODULE_PARM_DESC(major, "The major number of the device mapper"); +module_param(reserved_bio_based_ios, uint, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(reserved_bio_based_ios, "Reserved IOs in bio-based mempools"); + module_param(reserved_rq_based_ios, uint, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(reserved_rq_based_ios, "Reserved IOs in request-based mempools"); diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 1539650..1d1ad7b 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -184,6 +184,7 @@ void dm_free_md_mempools(struct dm_md_mempools *pools); /* * Helpers that are used by DM core */ +unsigned dm_get_reserved_bio_based_ios(void); unsigned dm_get_reserved_rq_based_ios(void); static inline bool dm_message_test_buffer_overflow(char *result, unsigned maxlen) -- cgit v0.10.2 From 4a1132a023eb48cf10522d84c5908d43b612c041 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 23 Sep 2013 10:38:26 -0400 Subject: drm/radeon: disable tests/benchmarks if accel is disabled The tests are only usable if the acceleration engines have been successfully initialized. Based on an initial patch from: Alex Ivanov Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index e29faa7..841d0e0 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1320,13 +1320,22 @@ int radeon_device_init(struct radeon_device *rdev, return r; } if ((radeon_testing & 1)) { - radeon_test_moves(rdev); + if (rdev->accel_working) + radeon_test_moves(rdev); + else + DRM_INFO("radeon: acceleration disabled, skipping move tests\n"); } if ((radeon_testing & 2)) { - radeon_test_syncing(rdev); + if (rdev->accel_working) + radeon_test_syncing(rdev); + else + DRM_INFO("radeon: acceleration disabled, skipping sync tests\n"); } if (radeon_benchmarking) { - radeon_benchmark(rdev, radeon_benchmarking); + if (rdev->accel_working) + radeon_benchmark(rdev, radeon_benchmarking); + else + DRM_INFO("radeon: acceleration disabled, skipping benchmarks\n"); } return 0; } -- cgit v0.10.2 From 4b40e5921230beb1951f04d2b1b92c4c88fbad43 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Mon, 23 Sep 2013 09:42:32 +0200 Subject: drm/radeon/uvd: lower msg&fb buffer requirements on UVD3 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Starting with UVD3 message and feedback buffers have their own 256MB segment, so no need to force them into VRAM any more. Signed-off-by: Christian König Signed-off-by: Alex Deucher diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 80285e3..66c2228 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -85,8 +85,9 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) VRAM, also but everything into VRAM on AGP cards to avoid image corruptions */ if (p->ring == R600_RING_TYPE_UVD_INDEX && + p->rdev->family < CHIP_PALM && (i == 0 || drm_pci_device_is_agp(p->rdev->ddev))) { - /* TODO: is this still needed for NI+ ? */ + p->relocs[i].lobj.domain = RADEON_GEM_DOMAIN_VRAM; diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index 1a01bbf..a0f1185 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -476,8 +476,7 @@ static int radeon_uvd_cs_reloc(struct radeon_cs_parser *p, return -EINVAL; } - /* TODO: is this still necessary on NI+ ? */ - if ((cmd == 0 || cmd == 0x3) && + if (p->rdev->family < CHIP_PALM && (cmd == 0 || cmd == 0x3) && (start >> 28) != (p->rdev->uvd.gpu_addr >> 28)) { DRM_ERROR("msg/fb buffer %LX-%LX out of 256MB segment!\n", start, end); diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c index 7266805..3100fa9 100644 --- a/drivers/gpu/drm/radeon/uvd_v1_0.c +++ b/drivers/gpu/drm/radeon/uvd_v1_0.c @@ -212,8 +212,8 @@ int uvd_v1_0_start(struct radeon_device *rdev) /* enable VCPU clock */ WREG32(UVD_VCPU_CNTL, 1 << 9); - /* enable UMC */ - WREG32_P(UVD_LMI_CTRL2, 0, ~(1 << 8)); + /* enable UMC and NC0 */ + WREG32_P(UVD_LMI_CTRL2, 1 << 13, ~((1 << 8) | (1 << 13))); /* boot up the VCPU */ WREG32(UVD_SOFT_RESET, 0); -- cgit v0.10.2 From 284d20552461466b04d6bfeafeb1c47a8891b591 Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Thu, 5 Sep 2013 11:01:20 +0300 Subject: xhci: Fix oops happening after address device timeout When a command times out, the command ring is first aborted, and then stopped. If the command ring is empty when it is stopped the stop event will point to next command which is not yet set. xHCI tries to handle this next event often causing an oops. Don't handle command completion events on stopped cmd ring if ring is empty. This patch should be backported to kernels as old as 3.7, that contain the commit b92cc66c047ff7cf587b318fe377061a353c120f "xHCI: add aborting command ring function" Signed-off-by: Mathias Nyman Reported-by: Giovanni Signed-off-by: Sarah Sharp Cc: stable@vger.kernel.org diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 411da1f..aaa2906 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -1414,6 +1414,12 @@ static void handle_cmd_completion(struct xhci_hcd *xhci, inc_deq(xhci, xhci->cmd_ring); return; } + /* There is no command to handle if we get a stop event when the + * command ring is empty, event->cmd_trb points to the next + * unset command + */ + if (xhci->cmd_ring->dequeue == xhci->cmd_ring->enqueue) + return; } switch (le32_to_cpu(xhci->cmd_ring->dequeue->generic.field[3]) -- cgit v0.10.2 From ec7e43e2d98173483866fe2e4e690143626b659c Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Fri, 30 Aug 2013 18:25:49 +0300 Subject: xhci: Ensure a command structure points to the correct trb on the command ring If a command on the command ring needs to be cancelled before it is handled it can be turned to a no-op operation when the ring is stopped. We want to store the command ring enqueue pointer in the command structure when the command in enqueued for the cancellation case. Some commands used to store the command ring dequeue pointers instead of enqueue (these often worked because enqueue happends to equal dequeue quite often) Other commands correctly used the enqueue pointer but did not check if it pointed to a valid trb or a link trb, this caused for example stop endpoint command to timeout in xhci_stop_device() in about 2% of suspend/resume cases. This should also solve some weird behavior happening in command cancellation cases. This patch is based on a patch submitted by Sarah Sharp to linux-usb, but then forgotten: http://marc.info/?l=linux-usb&m=136269803207465&w=2 This patch should be backported to kernels as old as 3.7, that contain the commit b92cc66c047ff7cf587b318fe377061a353c120f "xHCI: add aborting command ring function" Signed-off-by: Mathias Nyman Signed-off-by: Sarah Sharp Cc: stable@vger.kernel.org diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index fae697e..ccf0a06 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -287,7 +287,7 @@ static int xhci_stop_device(struct xhci_hcd *xhci, int slot_id, int suspend) if (virt_dev->eps[i].ring && virt_dev->eps[i].ring->dequeue) xhci_queue_stop_endpoint(xhci, slot_id, i, suspend); } - cmd->command_trb = xhci->cmd_ring->enqueue; + cmd->command_trb = xhci_find_next_enqueue(xhci->cmd_ring); list_add_tail(&cmd->cmd_list, &virt_dev->cmd_list); xhci_queue_stop_endpoint(xhci, slot_id, 0, suspend); xhci_ring_cmd_db(xhci); diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index aaa2906..9ac9672 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -123,6 +123,16 @@ static int enqueue_is_link_trb(struct xhci_ring *ring) return TRB_TYPE_LINK_LE32(link->control); } +union xhci_trb *xhci_find_next_enqueue(struct xhci_ring *ring) +{ + /* Enqueue pointer can be left pointing to the link TRB, + * we must handle that + */ + if (TRB_TYPE_LINK_LE32(ring->enqueue->link.control)) + return ring->enq_seg->next->trbs; + return ring->enqueue; +} + /* Updates trb to point to the next TRB in the ring, and updates seg if the next * TRB is in a new segment. This does not skip over link TRBs, and it does not * effect the ring dequeue or enqueue pointers. diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 49b6edb..1e36dbb 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -2598,15 +2598,7 @@ static int xhci_configure_endpoint(struct xhci_hcd *xhci, if (command) { cmd_completion = command->completion; cmd_status = &command->status; - command->command_trb = xhci->cmd_ring->enqueue; - - /* Enqueue pointer can be left pointing to the link TRB, - * we must handle that - */ - if (TRB_TYPE_LINK_LE32(command->command_trb->link.control)) - command->command_trb = - xhci->cmd_ring->enq_seg->next->trbs; - + command->command_trb = xhci_find_next_enqueue(xhci->cmd_ring); list_add_tail(&command->cmd_list, &virt_dev->cmd_list); } else { cmd_completion = &virt_dev->cmd_completion; @@ -2614,7 +2606,7 @@ static int xhci_configure_endpoint(struct xhci_hcd *xhci, } init_completion(cmd_completion); - cmd_trb = xhci->cmd_ring->dequeue; + cmd_trb = xhci_find_next_enqueue(xhci->cmd_ring); if (!ctx_change) ret = xhci_queue_configure_endpoint(xhci, in_ctx->dma, udev->slot_id, must_succeed); @@ -3439,14 +3431,7 @@ int xhci_discover_or_reset_device(struct usb_hcd *hcd, struct usb_device *udev) /* Attempt to submit the Reset Device command to the command ring */ spin_lock_irqsave(&xhci->lock, flags); - reset_device_cmd->command_trb = xhci->cmd_ring->enqueue; - - /* Enqueue pointer can be left pointing to the link TRB, - * we must handle that - */ - if (TRB_TYPE_LINK_LE32(reset_device_cmd->command_trb->link.control)) - reset_device_cmd->command_trb = - xhci->cmd_ring->enq_seg->next->trbs; + reset_device_cmd->command_trb = xhci_find_next_enqueue(xhci->cmd_ring); list_add_tail(&reset_device_cmd->cmd_list, &virt_dev->cmd_list); ret = xhci_queue_reset_device(xhci, slot_id); @@ -3650,7 +3635,7 @@ int xhci_alloc_dev(struct usb_hcd *hcd, struct usb_device *udev) union xhci_trb *cmd_trb; spin_lock_irqsave(&xhci->lock, flags); - cmd_trb = xhci->cmd_ring->dequeue; + cmd_trb = xhci_find_next_enqueue(xhci->cmd_ring); ret = xhci_queue_slot_control(xhci, TRB_ENABLE_SLOT, 0); if (ret) { spin_unlock_irqrestore(&xhci->lock, flags); @@ -3785,7 +3770,7 @@ int xhci_address_device(struct usb_hcd *hcd, struct usb_device *udev) slot_ctx->dev_info >> 27); spin_lock_irqsave(&xhci->lock, flags); - cmd_trb = xhci->cmd_ring->dequeue; + cmd_trb = xhci_find_next_enqueue(xhci->cmd_ring); ret = xhci_queue_address_device(xhci, virt_dev->in_ctx->dma, udev->slot_id); if (ret) { diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 46aa148..f3e1020 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1840,6 +1840,7 @@ int xhci_cancel_cmd(struct xhci_hcd *xhci, struct xhci_command *command, union xhci_trb *cmd_trb); void xhci_ring_ep_doorbell(struct xhci_hcd *xhci, unsigned int slot_id, unsigned int ep_index, unsigned int stream_id); +union xhci_trb *xhci_find_next_enqueue(struct xhci_ring *ring); /* xHCI roothub code */ void xhci_set_link_state(struct xhci_hcd *xhci, __le32 __iomem **port_array, -- cgit v0.10.2 From 8b3d45705e54075cfb9d4212dbca9ea82c85c4b8 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Tue, 20 Aug 2013 08:12:12 -0700 Subject: usb: Fix xHCI host issues on remote wakeup. When a device signals remote wakeup on a roothub, and the suspend change bit is set, the host controller driver must not give control back to the USB core until the port goes back into the active state. EHCI accomplishes this by waiting in the get port status function until the PORT_RESUME bit is cleared: /* stop resume signaling */ temp &= ~(PORT_RWC_BITS | PORT_SUSPEND | PORT_RESUME); ehci_writel(ehci, temp, status_reg); clear_bit(wIndex, &ehci->resuming_ports); retval = ehci_handshake(ehci, status_reg, PORT_RESUME, 0, 2000 /* 2msec */); Similarly, the xHCI host should wait until the port goes into U0, before passing control up to the USB core. When the port transitions from the RExit state to U0, the xHCI driver will get a port status change event. We need to wait for that event before passing control up to the USB core. After the port transitions to the active state, the USB core should time a recovery interval before it talks to the device. The length of that recovery interval is TRSMRCY, 10 ms, mentioned in the USB 2.0 spec, section 7.1.7.7. The previous xHCI code (which did not wait for the port to go into U0) would cause the USB core to violate that recovery interval. This bug caused numerous USB device disconnects on remote wakeup under ChromeOS and a Lynx Point LP xHCI host that takes up to 20 ms to move from RExit to U0. ChromeOS is very aggressive about power savings, and sets the autosuspend_delay to 100 ms, and disables USB persist. I attempted to replicate this bug with Ubuntu 12.04, but could not. I used Ubuntu 12.04 on the same platform, with the same BIOS that the bug was triggered on ChromeOS with. I also changed the USB sysfs settings as described above, but still could not reproduce the bug under Ubuntu. It may be that ChromeOS userspace triggers this bug through additional settings. Signed-off-by: Sarah Sharp diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c index ccf0a06..773a6b2 100644 --- a/drivers/usb/host/xhci-hub.c +++ b/drivers/usb/host/xhci-hub.c @@ -552,11 +552,15 @@ void xhci_del_comp_mod_timer(struct xhci_hcd *xhci, u32 status, u16 wIndex) * - Mark a port as being done with device resume, * and ring the endpoint doorbells. * - Stop the Synopsys redriver Compliance Mode polling. + * - Drop and reacquire the xHCI lock, in order to wait for port resume. */ static u32 xhci_get_port_status(struct usb_hcd *hcd, struct xhci_bus_state *bus_state, __le32 __iomem **port_array, - u16 wIndex, u32 raw_port_status) + u16 wIndex, u32 raw_port_status, + unsigned long flags) + __releases(&xhci->lock) + __acquires(&xhci->lock) { struct xhci_hcd *xhci = hcd_to_xhci(hcd); u32 status = 0; @@ -591,21 +595,42 @@ static u32 xhci_get_port_status(struct usb_hcd *hcd, return 0xffffffff; if (time_after_eq(jiffies, bus_state->resume_done[wIndex])) { + int time_left; + xhci_dbg(xhci, "Resume USB2 port %d\n", wIndex + 1); bus_state->resume_done[wIndex] = 0; clear_bit(wIndex, &bus_state->resuming_ports); + + set_bit(wIndex, &bus_state->rexit_ports); xhci_set_link_state(xhci, port_array, wIndex, XDEV_U0); - xhci_dbg(xhci, "set port %d resume\n", - wIndex + 1); - slot_id = xhci_find_slot_id_by_port(hcd, xhci, - wIndex + 1); - if (!slot_id) { - xhci_dbg(xhci, "slot_id is zero\n"); - return 0xffffffff; + + spin_unlock_irqrestore(&xhci->lock, flags); + time_left = wait_for_completion_timeout( + &bus_state->rexit_done[wIndex], + msecs_to_jiffies( + XHCI_MAX_REXIT_TIMEOUT)); + spin_lock_irqsave(&xhci->lock, flags); + + if (time_left) { + slot_id = xhci_find_slot_id_by_port(hcd, + xhci, wIndex + 1); + if (!slot_id) { + xhci_dbg(xhci, "slot_id is zero\n"); + return 0xffffffff; + } + xhci_ring_device(xhci, slot_id); + } else { + int port_status = xhci_readl(xhci, + port_array[wIndex]); + xhci_warn(xhci, "Port resume took longer than %i msec, port status = 0x%x\n", + XHCI_MAX_REXIT_TIMEOUT, + port_status); + status |= USB_PORT_STAT_SUSPEND; + clear_bit(wIndex, &bus_state->rexit_ports); } - xhci_ring_device(xhci, slot_id); + bus_state->port_c_suspend |= 1 << wIndex; bus_state->suspended_ports &= ~(1 << wIndex); } else { @@ -728,7 +753,7 @@ int xhci_hub_control(struct usb_hcd *hcd, u16 typeReq, u16 wValue, break; } status = xhci_get_port_status(hcd, bus_state, port_array, - wIndex, temp); + wIndex, temp, flags); if (status == 0xffffffff) goto error; diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 53b972c..83bcd13 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -2428,6 +2428,8 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags) for (i = 0; i < USB_MAXCHILDREN; ++i) { xhci->bus_state[0].resume_done[i] = 0; xhci->bus_state[1].resume_done[i] = 0; + /* Only the USB 2.0 completions will ever be used. */ + init_completion(&xhci->bus_state[1].rexit_done[i]); } if (scratchpad_alloc(xhci, flags)) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 9ac9672..dd02402 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -1759,6 +1759,19 @@ static void handle_port_status(struct xhci_hcd *xhci, } } + /* + * Check to see if xhci-hub.c is waiting on RExit to U0 transition (or + * RExit to a disconnect state). If so, let the the driver know it's + * out of the RExit state. + */ + if (!DEV_SUPERSPEED(temp) && + test_and_clear_bit(faked_port_index, + &bus_state->rexit_ports)) { + complete(&bus_state->rexit_done[faked_port_index]); + bogus_port_status = true; + goto cleanup; + } + if (hcd->speed != HCD_USB3) xhci_test_and_clear_bit(xhci, port_array, faked_port_index, PORT_PLC); diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index f3e1020..289fbfb 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1412,8 +1412,18 @@ struct xhci_bus_state { unsigned long resume_done[USB_MAXCHILDREN]; /* which ports have started to resume */ unsigned long resuming_ports; + /* Which ports are waiting on RExit to U0 transition. */ + unsigned long rexit_ports; + struct completion rexit_done[USB_MAXCHILDREN]; }; + +/* + * It can take up to 20 ms to transition from RExit to U0 on the + * Intel Lynx Point LP xHCI host. + */ +#define XHCI_MAX_REXIT_TIMEOUT (20 * 1000) + static inline unsigned int hcd_index(struct usb_hcd *hcd) { if (hcd->speed == HCD_USB3) -- cgit v0.10.2 From 526867c3ca0caa2e3e846cb993b0f961c33c2abb Mon Sep 17 00:00:00 2001 From: Florian Wolter Date: Wed, 14 Aug 2013 10:33:16 +0200 Subject: xhci: Fix race between ep halt and URB cancellation The halted state of a endpoint cannot be cleared over CLEAR_HALT from a user process, because the stopped_td variable was overwritten in the handle_stopped_endpoint() function. So the xhci_endpoint_reset() function will refuse the reset and communication with device can not run over this endpoint. https://bugzilla.kernel.org/show_bug.cgi?id=60699 Signed-off-by: Florian Wolter Signed-off-by: Sarah Sharp diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index dd02402..6bfbd80 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -869,8 +869,12 @@ remove_finished_td: /* Otherwise ring the doorbell(s) to restart queued transfers */ ring_doorbell_for_active_rings(xhci, slot_id, ep_index); } - ep->stopped_td = NULL; - ep->stopped_trb = NULL; + + /* Clear stopped_td and stopped_trb if endpoint is not halted */ + if (!(ep->ep_state & EP_HALTED)) { + ep->stopped_td = NULL; + ep->stopped_trb = NULL; + } /* * Drop the lock and complete the URBs in the cancelled TD list. -- cgit v0.10.2 From 38d7f6885164b114fdfae84bc18397e8fced9b00 Mon Sep 17 00:00:00 2001 From: Xenia Ragiadakou Date: Wed, 4 Sep 2013 17:24:45 +0300 Subject: usbcore: check usb device's state before sending a Set SEL control transfer Set SEL control urbs cannot be sent to a device in unconfigured state. This patch adds a check in usb_req_set_sel() to ensure the usb device's state is USB_STATE_CONFIGURED. Signed-off-by: Xenia Ragiadakou Reported-by: Martin MOKREJS Suggested-by: Sarah Sharp Signed-off-by: Sarah Sharp diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index dde4c83..e6b682c 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -3426,6 +3426,9 @@ static int usb_req_set_sel(struct usb_device *udev, enum usb3_link_state state) unsigned long long u2_pel; int ret; + if (udev->state != USB_STATE_CONFIGURED) + return 0; + /* Convert SEL and PEL stored in ns to us */ u1_sel = DIV_ROUND_UP(udev->u1_params.sel, 1000); u1_pel = DIV_ROUND_UP(udev->u1_params.pel, 1000); -- cgit v0.10.2 From e29bb4ebbf000ff9ac081d29784a3331618f012e Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 20 Sep 2013 10:20:59 +0100 Subject: drm/i915: Use a temporary va_list for two-pass string handling In commit edc3d8848dc9fe2a470316363dab8ef211d77e01 Author: Mika Kuoppala Date: Thu May 23 13:55:35 2013 +0300 drm/i915: avoid big kmallocs on reading error state we introduce a two-pass mechanism for splitting long strings being formatted into the error-state. The first pass finds the length, and the second pass emits the right portion of the string into the accumulation buffer. Unfortunately we use the same va_list for both passes, resulting in the second pass reading garbage off the end of the argument list. As the two passes are only used for boundaries between read() calls, the corruption is only rarely seen. This fixes the root cause behind commit baf27f9b17bf2f369f3865e38c41d2163e8d815d Author: Chris Wilson Date: Sat Jun 29 23:26:50 2013 +0100 drm/i915: Break up the large vsnprintf() in print_error_buffers() Signed-off-by: Chris Wilson Cc: Mika Kuoppala Cc: Daniel Vetter Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index aba9d74..dae364f 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -143,8 +143,10 @@ static void i915_error_vprintf(struct drm_i915_error_state_buf *e, /* Seek the first printf which is hits start position */ if (e->pos < e->start) { - len = vsnprintf(NULL, 0, f, args); - if (!__i915_error_seek(e, len)) + va_list tmp; + + va_copy(tmp, args); + if (!__i915_error_seek(e, vsnprintf(NULL, 0, f, tmp))) return; } -- cgit v0.10.2 From becee6b8c7b2b4adc9a3e0bec633abecd591b9ef Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Sun, 22 Sep 2013 22:04:27 +0100 Subject: MIPS: cpu-features.h: s/MIPS53/MIPS64/ No support for MIPS53 processors yet. Signed-off-by: Maciej W. Rozycki Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/5876/ Signed-off-by: Ralf Baechle diff --git a/arch/mips/include/asm/cpu-features.h b/arch/mips/include/asm/cpu-features.h index 51680d1..d445d06 100644 --- a/arch/mips/include/asm/cpu-features.h +++ b/arch/mips/include/asm/cpu-features.h @@ -187,7 +187,7 @@ /* * MIPS32, MIPS64, VR5500, IDT32332, IDT32334 and maybe a few other - * pre-MIPS32/MIPS53 processors have CLO, CLZ. The IDT RC64574 is 64-bit and + * pre-MIPS32/MIPS64 processors have CLO, CLZ. The IDT RC64574 is 64-bit and * has CLO and CLZ but not DCLO nor DCLZ. For 64-bit kernels * cpu_has_clo_clz also indicates the availability of DCLO and DCLZ. */ -- cgit v0.10.2 From 69d75671d985ccdb291e4d19ddfdecd12440e857 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 12 Sep 2013 22:00:15 +0200 Subject: udf: Fortify LVID loading A user has reported an oops in udf_statfs() that was caused by numOfPartitions entry in LVID structure being corrupted. Fix the problem by verifying whether numOfPartitions makes sense at least to the extent that LVID fits into a single block as it should. Reported-by: Juergen Weigert Signed-off-by: Jan Kara diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index 7e5aae4..6eaf5ed 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -30,18 +30,17 @@ void udf_free_inode(struct inode *inode) { struct super_block *sb = inode->i_sb; struct udf_sb_info *sbi = UDF_SB(sb); + struct logicalVolIntegrityDescImpUse *lvidiu = udf_sb_lvidiu(sb); - mutex_lock(&sbi->s_alloc_mutex); - if (sbi->s_lvid_bh) { - struct logicalVolIntegrityDescImpUse *lvidiu = - udf_sb_lvidiu(sbi); + if (lvidiu) { + mutex_lock(&sbi->s_alloc_mutex); if (S_ISDIR(inode->i_mode)) le32_add_cpu(&lvidiu->numDirs, -1); else le32_add_cpu(&lvidiu->numFiles, -1); udf_updated_lvid(sb); + mutex_unlock(&sbi->s_alloc_mutex); } - mutex_unlock(&sbi->s_alloc_mutex); udf_free_blocks(sb, NULL, &UDF_I(inode)->i_location, 0, 1); } @@ -55,6 +54,7 @@ struct inode *udf_new_inode(struct inode *dir, umode_t mode, int *err) uint32_t start = UDF_I(dir)->i_location.logicalBlockNum; struct udf_inode_info *iinfo; struct udf_inode_info *dinfo = UDF_I(dir); + struct logicalVolIntegrityDescImpUse *lvidiu; inode = new_inode(sb); @@ -92,12 +92,10 @@ struct inode *udf_new_inode(struct inode *dir, umode_t mode, int *err) return NULL; } - if (sbi->s_lvid_bh) { - struct logicalVolIntegrityDescImpUse *lvidiu; - + lvidiu = udf_sb_lvidiu(sb); + if (lvidiu) { iinfo->i_unique = lvid_get_unique_id(sb); mutex_lock(&sbi->s_alloc_mutex); - lvidiu = udf_sb_lvidiu(sbi); if (S_ISDIR(mode)) le32_add_cpu(&lvidiu->numDirs, 1); else diff --git a/fs/udf/super.c b/fs/udf/super.c index 839a2ba..9121938 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -94,13 +94,25 @@ static unsigned int udf_count_free(struct super_block *); static int udf_statfs(struct dentry *, struct kstatfs *); static int udf_show_options(struct seq_file *, struct dentry *); -struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct udf_sb_info *sbi) +struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct super_block *sb) { - struct logicalVolIntegrityDesc *lvid = - (struct logicalVolIntegrityDesc *)sbi->s_lvid_bh->b_data; - __u32 number_of_partitions = le32_to_cpu(lvid->numOfPartitions); - __u32 offset = number_of_partitions * 2 * - sizeof(uint32_t)/sizeof(uint8_t); + struct logicalVolIntegrityDesc *lvid; + unsigned int partnum; + unsigned int offset; + + if (!UDF_SB(sb)->s_lvid_bh) + return NULL; + lvid = (struct logicalVolIntegrityDesc *)UDF_SB(sb)->s_lvid_bh->b_data; + partnum = le32_to_cpu(lvid->numOfPartitions); + if ((sb->s_blocksize - sizeof(struct logicalVolIntegrityDescImpUse) - + offsetof(struct logicalVolIntegrityDesc, impUse)) / + (2 * sizeof(uint32_t)) < partnum) { + udf_err(sb, "Logical volume integrity descriptor corrupted " + "(numOfPartitions = %u)!\n", partnum); + return NULL; + } + /* The offset is to skip freeSpaceTable and sizeTable arrays */ + offset = partnum * 2 * sizeof(uint32_t); return (struct logicalVolIntegrityDescImpUse *)&(lvid->impUse[offset]); } @@ -629,9 +641,10 @@ static int udf_remount_fs(struct super_block *sb, int *flags, char *options) struct udf_options uopt; struct udf_sb_info *sbi = UDF_SB(sb); int error = 0; + struct logicalVolIntegrityDescImpUse *lvidiu = udf_sb_lvidiu(sb); - if (sbi->s_lvid_bh) { - int write_rev = le16_to_cpu(udf_sb_lvidiu(sbi)->minUDFWriteRev); + if (lvidiu) { + int write_rev = le16_to_cpu(lvidiu->minUDFWriteRev); if (write_rev > UDF_MAX_WRITE_VERSION && !(*flags & MS_RDONLY)) return -EACCES; } @@ -1905,11 +1918,12 @@ static void udf_open_lvid(struct super_block *sb) if (!bh) return; - - mutex_lock(&sbi->s_alloc_mutex); lvid = (struct logicalVolIntegrityDesc *)bh->b_data; - lvidiu = udf_sb_lvidiu(sbi); + lvidiu = udf_sb_lvidiu(sb); + if (!lvidiu) + return; + mutex_lock(&sbi->s_alloc_mutex); lvidiu->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; lvidiu->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; udf_time_to_disk_stamp(&lvid->recordingDateAndTime, @@ -1937,10 +1951,12 @@ static void udf_close_lvid(struct super_block *sb) if (!bh) return; + lvid = (struct logicalVolIntegrityDesc *)bh->b_data; + lvidiu = udf_sb_lvidiu(sb); + if (!lvidiu) + return; mutex_lock(&sbi->s_alloc_mutex); - lvid = (struct logicalVolIntegrityDesc *)bh->b_data; - lvidiu = udf_sb_lvidiu(sbi); lvidiu->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; lvidiu->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; udf_time_to_disk_stamp(&lvid->recordingDateAndTime, CURRENT_TIME); @@ -2093,15 +2109,19 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) if (sbi->s_lvid_bh) { struct logicalVolIntegrityDescImpUse *lvidiu = - udf_sb_lvidiu(sbi); - uint16_t minUDFReadRev = le16_to_cpu(lvidiu->minUDFReadRev); - uint16_t minUDFWriteRev = le16_to_cpu(lvidiu->minUDFWriteRev); - /* uint16_t maxUDFWriteRev = - le16_to_cpu(lvidiu->maxUDFWriteRev); */ + udf_sb_lvidiu(sb); + uint16_t minUDFReadRev; + uint16_t minUDFWriteRev; + if (!lvidiu) { + ret = -EINVAL; + goto error_out; + } + minUDFReadRev = le16_to_cpu(lvidiu->minUDFReadRev); + minUDFWriteRev = le16_to_cpu(lvidiu->minUDFWriteRev); if (minUDFReadRev > UDF_MAX_READ_VERSION) { udf_err(sb, "minUDFReadRev=%x (max is %x)\n", - le16_to_cpu(lvidiu->minUDFReadRev), + minUDFReadRev, UDF_MAX_READ_VERSION); ret = -EINVAL; goto error_out; @@ -2265,11 +2285,7 @@ static int udf_statfs(struct dentry *dentry, struct kstatfs *buf) struct logicalVolIntegrityDescImpUse *lvidiu; u64 id = huge_encode_dev(sb->s_bdev->bd_dev); - if (sbi->s_lvid_bh != NULL) - lvidiu = udf_sb_lvidiu(sbi); - else - lvidiu = NULL; - + lvidiu = udf_sb_lvidiu(sb); buf->f_type = UDF_SUPER_MAGIC; buf->f_bsize = sb->s_blocksize; buf->f_blocks = sbi->s_partmaps[sbi->s_partition].s_partition_len; diff --git a/fs/udf/udf_sb.h b/fs/udf/udf_sb.h index ed401e9..1f32c7b 100644 --- a/fs/udf/udf_sb.h +++ b/fs/udf/udf_sb.h @@ -162,7 +162,7 @@ static inline struct udf_sb_info *UDF_SB(struct super_block *sb) return sb->s_fs_info; } -struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct udf_sb_info *sbi); +struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct super_block *sb); int udf_compute_nr_groups(struct super_block *sb, u32 partition); -- cgit v0.10.2 From 7bc9cc07ee5bbac58bab88e8a5d6e32785f8fd32 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Mon, 23 Sep 2013 16:50:28 -0400 Subject: reiserfs: remove useless flush_old_journal_lists Commit a3172027 introduced test_transaction as a requirement for flushing old lists -- but it can never return 1 unless the transaction has already been flushed. As a result, we have a routine that iterates the j_realblocks list but doesn't actually do anything. Since it's been this way since 2006 and the latency numbers were what Chris expected, let's just rip it out. Signed-off-by: Jeff Mahoney Signed-off-by: Jan Kara diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 73feacc4..7a37473 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -1163,21 +1163,6 @@ static struct reiserfs_journal_list *find_newer_jl_for_cn(struct return NULL; } -static int newer_jl_done(struct reiserfs_journal_cnode *cn) -{ - struct super_block *sb = cn->sb; - b_blocknr_t blocknr = cn->blocknr; - - cn = cn->hprev; - while (cn) { - if (cn->sb == sb && cn->blocknr == blocknr && cn->jlist && - atomic_read(&cn->jlist->j_commit_left) != 0) - return 0; - cn = cn->hprev; - } - return 1; -} - static void remove_journal_hash(struct super_block *, struct reiserfs_journal_cnode **, struct reiserfs_journal_list *, unsigned long, @@ -1593,31 +1578,6 @@ static int flush_journal_list(struct super_block *s, return err; } -static int test_transaction(struct super_block *s, - struct reiserfs_journal_list *jl) -{ - struct reiserfs_journal_cnode *cn; - - if (jl->j_len == 0 || atomic_read(&jl->j_nonzerolen) == 0) - return 1; - - cn = jl->j_realblock; - while (cn) { - /* if the blocknr == 0, this has been cleared from the hash, - ** skip it - */ - if (cn->blocknr == 0) { - goto next; - } - if (cn->bh && !newer_jl_done(cn)) - return 0; - next: - cn = cn->next; - cond_resched(); - } - return 0; -} - static int write_one_transaction(struct super_block *s, struct reiserfs_journal_list *jl, struct buffer_chunk *chunk) @@ -3868,27 +3828,6 @@ int reiserfs_prepare_for_journal(struct super_block *sb, return 1; } -static void flush_old_journal_lists(struct super_block *s) -{ - struct reiserfs_journal *journal = SB_JOURNAL(s); - struct reiserfs_journal_list *jl; - struct list_head *entry; - time_t now = get_seconds(); - - while (!list_empty(&journal->j_journal_list)) { - entry = journal->j_journal_list.next; - jl = JOURNAL_LIST_ENTRY(entry); - /* this check should always be run, to send old lists to disk */ - if (jl->j_timestamp < (now - (JOURNAL_MAX_TRANS_AGE * 4)) && - atomic_read(&jl->j_commit_left) == 0 && - test_transaction(s, jl)) { - flush_used_journal_lists(s, jl); - } else { - break; - } - } -} - /* ** long and ugly. If flush, will not return until all commit ** blocks and all real buffers in the trans are on disk. @@ -4232,7 +4171,6 @@ static int do_journal_end(struct reiserfs_transaction_handle *th, } } } - flush_old_journal_lists(sb); journal->j_current_jl->j_list_bitmap = get_list_bitmap(sb, journal->j_current_jl); -- cgit v0.10.2 From 721a769c034204e8e5f6ca4ecbff249e0225333b Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Mon, 23 Sep 2013 16:50:42 -0400 Subject: reiserfs: fix race with flush_used_journal_lists and flush_journal_list There are two locks involved in managing the journal lists. The general reiserfs_write_lock and the journal->j_flush_mutex. While flush_journal_list is sleeping to acquire the j_flush_mutex or to submit a block for write, it will drop the write lock. This allows another thread to acquire the write lock and ultimately call flush_used_journal_lists to traverse the list of journal lists and select one for flushing. It can select the journal_list that has just had flush_journal_list called on it in the original thread and call it again with the same journal_list. The second thread then drops the write lock to acquire j_flush_mutex and the first thread reacquires it and continues execution and eventually clears and frees the journal list before dropping j_flush_mutex and returning. The second thread acquires j_flush_mutex and ends up operating on a journal_list that has already been released. If the memory hasn't been reused, we'll soon after hit a BUG_ON because the transaction id has already been cleared. If it's been reused, we'll crash in other fun ways. Since flush_journal_list will synchronize on j_flush_mutex, we can fix the race by taking a proper reference in flush_used_journal_lists and checking to see if it's still valid after the mutex is taken. It's safe to iterate the list of journal lists and pick a list with just the write lock as long as a reference is taken on the journal list before we drop the lock. We already have code to handle whether a transaction has been flushed already so we can use that to handle the race and get rid of the trans_id BUG_ON. Signed-off-by: Jeff Mahoney Signed-off-by: Jan Kara diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 7a37473..fd77703 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -1338,7 +1338,6 @@ static int flush_journal_list(struct super_block *s, reiserfs_warning(s, "clm-2048", "called with wcount %d", atomic_read(&journal->j_wcount)); } - BUG_ON(jl->j_trans_id == 0); /* if flushall == 0, the lock is already held */ if (flushall) { @@ -1765,6 +1764,8 @@ static int flush_used_journal_lists(struct super_block *s, break; tjl = JOURNAL_LIST_ENTRY(tjl->j_list.next); } + get_journal_list(jl); + get_journal_list(flush_jl); /* try to find a group of blocks we can flush across all the ** transactions, but only bother if we've actually spanned ** across multiple lists @@ -1773,6 +1774,8 @@ static int flush_used_journal_lists(struct super_block *s, ret = kupdate_transactions(s, jl, &tjl, &trans_id, len, i); } flush_journal_list(s, flush_jl, 1); + put_journal_list(s, flush_jl); + put_journal_list(s, jl); return 0; } -- cgit v0.10.2 From 99d79aa2f3b7729e7290e8bda5d0dd8b0240ec62 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 23 Sep 2013 15:47:08 -0400 Subject: drm/radeon: add missing hdmi callbacks for rv6xx MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When dpm was merged, I added a new asic struct for rv6xx, but it never got properly updated when the hdmi callbacks were added due to the two patch sets being developed in parallel. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69729 Signed-off-by: Alex Deucher Reviewed-by: Christian König Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index 5003385..8f7e045 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -1004,6 +1004,8 @@ static struct radeon_asic rv6xx_asic = { .wait_for_vblank = &avivo_wait_for_vblank, .set_backlight_level = &atombios_set_backlight_level, .get_backlight_level = &atombios_get_backlight_level, + .hdmi_enable = &r600_hdmi_enable, + .hdmi_setmode = &r600_hdmi_setmode, }, .copy = { .blit = &r600_copy_cpdma, -- cgit v0.10.2 From 13c5bfdad758bddc199850c22246ddf26adcec1f Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 24 Sep 2013 10:56:55 -0400 Subject: drm/radeon/cik: fix overflow in vram fetch Missing ULL when calculating the amount of vram leads to an overflow when the amount of vram is >= 4G. Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index 5e6802d..d02fd1c 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -4454,8 +4454,8 @@ static int cik_mc_init(struct radeon_device *rdev) rdev->mc.aper_base = pci_resource_start(rdev->pdev, 0); rdev->mc.aper_size = pci_resource_len(rdev->pdev, 0); /* size in MB on si */ - rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024; - rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024; + rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024ULL * 1024ULL; + rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024ULL * 1024ULL; rdev->mc.visible_vram_size = rdev->mc.aper_size; si_vram_gtt_location(rdev, &rdev->mc); radeon_update_bandwidth_info(rdev); -- cgit v0.10.2 From 53dad6d3a8e5ac1af8bacc6ac2134ae1a8b085f1 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Mon, 23 Sep 2013 17:04:45 -0700 Subject: ipc: fix race with LSMs Currently, IPC mechanisms do security and auditing related checks under RCU. However, since security modules can free the security structure, for example, through selinux_[sem,msg_queue,shm]_free_security(), we can race if the structure is freed before other tasks are done with it, creating a use-after-free condition. Manfred illustrates this nicely, for instance with shared mem and selinux: -> do_shmat calls rcu_read_lock() -> do_shmat calls shm_object_check(). Checks that the object is still valid - but doesn't acquire any locks. Then it returns. -> do_shmat calls security_shm_shmat (e.g. selinux_shm_shmat) -> selinux_shm_shmat calls ipc_has_perm() -> ipc_has_perm accesses ipc_perms->security shm_close() -> shm_close acquires rw_mutex & shm_lock -> shm_close calls shm_destroy -> shm_destroy calls security_shm_free (e.g. selinux_shm_free_security) -> selinux_shm_free_security calls ipc_free_security(&shp->shm_perm) -> ipc_free_security calls kfree(ipc_perms->security) This patch delays the freeing of the security structures after all RCU readers are done. Furthermore it aligns the security life cycle with that of the rest of IPC - freeing them based on the reference counter. For situations where we need not free security, the current behavior is kept. Linus states: "... the old behavior was suspect for another reason too: having the security blob go away from under a user sounds like it could cause various other problems anyway, so I think the old code was at least _prone_ to bugs even if it didn't have catastrophic behavior." I have tested this patch with IPC testcases from LTP on both my quad-core laptop and on a 64 core NUMA server. In both cases selinux is enabled, and tests pass for both voluntary and forced preemption models. While the mentioned races are theoretical (at least no one as reported them), I wanted to make sure that this new logic doesn't break anything we weren't aware of. Suggested-by: Linus Torvalds Signed-off-by: Davidlohr Bueso Acked-by: Manfred Spraul Signed-off-by: Linus Torvalds diff --git a/ipc/msg.c b/ipc/msg.c index b0d541d4..9e4310c 100644 --- a/ipc/msg.c +++ b/ipc/msg.c @@ -165,6 +165,15 @@ static inline void msg_rmid(struct ipc_namespace *ns, struct msg_queue *s) ipc_rmid(&msg_ids(ns), &s->q_perm); } +static void msg_rcu_free(struct rcu_head *head) +{ + struct ipc_rcu *p = container_of(head, struct ipc_rcu, rcu); + struct msg_queue *msq = ipc_rcu_to_struct(p); + + security_msg_queue_free(msq); + ipc_rcu_free(head); +} + /** * newque - Create a new msg queue * @ns: namespace @@ -189,15 +198,14 @@ static int newque(struct ipc_namespace *ns, struct ipc_params *params) msq->q_perm.security = NULL; retval = security_msg_queue_alloc(msq); if (retval) { - ipc_rcu_putref(msq); + ipc_rcu_putref(msq, ipc_rcu_free); return retval; } /* ipc_addid() locks msq upon success. */ id = ipc_addid(&msg_ids(ns), &msq->q_perm, ns->msg_ctlmni); if (id < 0) { - security_msg_queue_free(msq); - ipc_rcu_putref(msq); + ipc_rcu_putref(msq, msg_rcu_free); return id; } @@ -276,8 +284,7 @@ static void freeque(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp) free_msg(msg); } atomic_sub(msq->q_cbytes, &ns->msg_bytes); - security_msg_queue_free(msq); - ipc_rcu_putref(msq); + ipc_rcu_putref(msq, msg_rcu_free); } /* @@ -717,7 +724,7 @@ long do_msgsnd(int msqid, long mtype, void __user *mtext, rcu_read_lock(); ipc_lock_object(&msq->q_perm); - ipc_rcu_putref(msq); + ipc_rcu_putref(msq, ipc_rcu_free); if (msq->q_perm.deleted) { err = -EIDRM; goto out_unlock0; diff --git a/ipc/sem.c b/ipc/sem.c index 69b6a21..19c8b98 100644 --- a/ipc/sem.c +++ b/ipc/sem.c @@ -243,6 +243,15 @@ static void merge_queues(struct sem_array *sma) } } +static void sem_rcu_free(struct rcu_head *head) +{ + struct ipc_rcu *p = container_of(head, struct ipc_rcu, rcu); + struct sem_array *sma = ipc_rcu_to_struct(p); + + security_sem_free(sma); + ipc_rcu_free(head); +} + /* * If the request contains only one semaphore operation, and there are * no complex transactions pending, lock only the semaphore involved. @@ -374,12 +383,7 @@ static inline struct sem_array *sem_obtain_object_check(struct ipc_namespace *ns static inline void sem_lock_and_putref(struct sem_array *sma) { sem_lock(sma, NULL, -1); - ipc_rcu_putref(sma); -} - -static inline void sem_putref(struct sem_array *sma) -{ - ipc_rcu_putref(sma); + ipc_rcu_putref(sma, ipc_rcu_free); } static inline void sem_rmid(struct ipc_namespace *ns, struct sem_array *s) @@ -458,14 +462,13 @@ static int newary(struct ipc_namespace *ns, struct ipc_params *params) sma->sem_perm.security = NULL; retval = security_sem_alloc(sma); if (retval) { - ipc_rcu_putref(sma); + ipc_rcu_putref(sma, ipc_rcu_free); return retval; } id = ipc_addid(&sem_ids(ns), &sma->sem_perm, ns->sc_semmni); if (id < 0) { - security_sem_free(sma); - ipc_rcu_putref(sma); + ipc_rcu_putref(sma, sem_rcu_free); return id; } ns->used_sems += nsems; @@ -1047,8 +1050,7 @@ static void freeary(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp) wake_up_sem_queue_do(&tasks); ns->used_sems -= sma->sem_nsems; - security_sem_free(sma); - ipc_rcu_putref(sma); + ipc_rcu_putref(sma, sem_rcu_free); } static unsigned long copy_semid_to_user(void __user *buf, struct semid64_ds *in, int version) @@ -1292,7 +1294,7 @@ static int semctl_main(struct ipc_namespace *ns, int semid, int semnum, rcu_read_unlock(); sem_io = ipc_alloc(sizeof(ushort)*nsems); if(sem_io == NULL) { - sem_putref(sma); + ipc_rcu_putref(sma, ipc_rcu_free); return -ENOMEM; } @@ -1328,20 +1330,20 @@ static int semctl_main(struct ipc_namespace *ns, int semid, int semnum, if(nsems > SEMMSL_FAST) { sem_io = ipc_alloc(sizeof(ushort)*nsems); if(sem_io == NULL) { - sem_putref(sma); + ipc_rcu_putref(sma, ipc_rcu_free); return -ENOMEM; } } if (copy_from_user (sem_io, p, nsems*sizeof(ushort))) { - sem_putref(sma); + ipc_rcu_putref(sma, ipc_rcu_free); err = -EFAULT; goto out_free; } for (i = 0; i < nsems; i++) { if (sem_io[i] > SEMVMX) { - sem_putref(sma); + ipc_rcu_putref(sma, ipc_rcu_free); err = -ERANGE; goto out_free; } @@ -1629,7 +1631,7 @@ static struct sem_undo *find_alloc_undo(struct ipc_namespace *ns, int semid) /* step 2: allocate new undo structure */ new = kzalloc(sizeof(struct sem_undo) + sizeof(short)*nsems, GFP_KERNEL); if (!new) { - sem_putref(sma); + ipc_rcu_putref(sma, ipc_rcu_free); return ERR_PTR(-ENOMEM); } diff --git a/ipc/shm.c b/ipc/shm.c index 2821cdf..d697396 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -167,6 +167,15 @@ static inline void shm_lock_by_ptr(struct shmid_kernel *ipcp) ipc_lock_object(&ipcp->shm_perm); } +static void shm_rcu_free(struct rcu_head *head) +{ + struct ipc_rcu *p = container_of(head, struct ipc_rcu, rcu); + struct shmid_kernel *shp = ipc_rcu_to_struct(p); + + security_shm_free(shp); + ipc_rcu_free(head); +} + static inline void shm_rmid(struct ipc_namespace *ns, struct shmid_kernel *s) { ipc_rmid(&shm_ids(ns), &s->shm_perm); @@ -208,8 +217,7 @@ static void shm_destroy(struct ipc_namespace *ns, struct shmid_kernel *shp) user_shm_unlock(file_inode(shp->shm_file)->i_size, shp->mlock_user); fput (shp->shm_file); - security_shm_free(shp); - ipc_rcu_putref(shp); + ipc_rcu_putref(shp, shm_rcu_free); } /* @@ -497,7 +505,7 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) shp->shm_perm.security = NULL; error = security_shm_alloc(shp); if (error) { - ipc_rcu_putref(shp); + ipc_rcu_putref(shp, ipc_rcu_free); return error; } @@ -566,8 +574,7 @@ no_id: user_shm_unlock(size, shp->mlock_user); fput(file); no_file: - security_shm_free(shp); - ipc_rcu_putref(shp); + ipc_rcu_putref(shp, shm_rcu_free); return error; } diff --git a/ipc/util.c b/ipc/util.c index e829da9..fdb8ae7 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -474,11 +474,6 @@ void ipc_free(void* ptr, int size) kfree(ptr); } -struct ipc_rcu { - struct rcu_head rcu; - atomic_t refcount; -} ____cacheline_aligned_in_smp; - /** * ipc_rcu_alloc - allocate ipc and rcu space * @size: size desired @@ -505,27 +500,24 @@ int ipc_rcu_getref(void *ptr) return atomic_inc_not_zero(&p->refcount); } -/** - * ipc_schedule_free - free ipc + rcu space - * @head: RCU callback structure for queued work - */ -static void ipc_schedule_free(struct rcu_head *head) -{ - vfree(container_of(head, struct ipc_rcu, rcu)); -} - -void ipc_rcu_putref(void *ptr) +void ipc_rcu_putref(void *ptr, void (*func)(struct rcu_head *head)) { struct ipc_rcu *p = ((struct ipc_rcu *)ptr) - 1; if (!atomic_dec_and_test(&p->refcount)) return; - if (is_vmalloc_addr(ptr)) { - call_rcu(&p->rcu, ipc_schedule_free); - } else { - kfree_rcu(p, rcu); - } + call_rcu(&p->rcu, func); +} + +void ipc_rcu_free(struct rcu_head *head) +{ + struct ipc_rcu *p = container_of(head, struct ipc_rcu, rcu); + + if (is_vmalloc_addr(p)) + vfree(p); + else + kfree(p); } /** diff --git a/ipc/util.h b/ipc/util.h index c5f3338b..f2f5036 100644 --- a/ipc/util.h +++ b/ipc/util.h @@ -47,6 +47,13 @@ static inline void msg_exit_ns(struct ipc_namespace *ns) { } static inline void shm_exit_ns(struct ipc_namespace *ns) { } #endif +struct ipc_rcu { + struct rcu_head rcu; + atomic_t refcount; +} ____cacheline_aligned_in_smp; + +#define ipc_rcu_to_struct(p) ((void *)(p+1)) + /* * Structure that holds the parameters needed by the ipc operations * (see after) @@ -120,7 +127,8 @@ void ipc_free(void* ptr, int size); */ void* ipc_rcu_alloc(int size); int ipc_rcu_getref(void *ptr); -void ipc_rcu_putref(void *ptr); +void ipc_rcu_putref(void *ptr, void (*func)(struct rcu_head *head)); +void ipc_rcu_free(struct rcu_head *head); struct kern_ipc_perm *ipc_lock(struct ipc_ids *, int); struct kern_ipc_perm *ipc_obtain_object(struct ipc_ids *ids, int id); -- cgit v0.10.2 From db6aaf4d55f95dcb6b162c3a59b56eb1e85ccdfe Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 24 Sep 2013 10:37:13 +0100 Subject: drm/i2c: tda998x: fix audio muting Fix a bug that was introduced in commit c4c11dd160a8 ("drm/i2c: tda998x: add video and audio input configuration") when Sebastian cleaned up my original patch. Without this being fixed, audio is muted when the display is turned off, never to be re-enabled. Signed-off-by: Russell King Cc: Sebastian Hesselbarth Cc: Darren Etheridge Cc: Dave Airlie Signed-off-by: Linus Torvalds diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index b1f8fc6..60e8404 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -707,8 +707,7 @@ tda998x_encoder_dpms(struct drm_encoder *encoder, int mode) reg_write(encoder, REG_VIP_CNTRL_2, priv->vip_cntrl_2); break; case DRM_MODE_DPMS_OFF: - /* disable audio and video ports */ - reg_write(encoder, REG_ENA_AP, 0x00); + /* disable video ports */ reg_write(encoder, REG_ENA_VP_0, 0x00); reg_write(encoder, REG_ENA_VP_1, 0x00); reg_write(encoder, REG_ENA_VP_2, 0x00); -- cgit v0.10.2 From 4921e320244e099bdf237fd10428594ce5f5b87d Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 12 Sep 2013 18:39:36 +0200 Subject: perf kmem: Make it work again on non NUMA machines The commit '2814eb0 perf kmem: Remove die() calls' disabled 'perf kmem' command for machines without numa support. It made the command fail if '/sys/devices/system/node' dir wasn't found. Skipping the numa based initialization in case the directory is not found and continue execution. Signed-off-by: Jiri Olsa Cc: Corey Ashford Cc: Frederic Weisbecker Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1379003976-5839-5-git-send-email-jolsa@redhat.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index c2dff9c..9b5f077 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -101,7 +101,7 @@ static int setup_cpunode_map(void) dir1 = opendir(PATH_SYS_NODE); if (!dir1) - return -1; + return 0; while ((dent1 = readdir(dir1)) != NULL) { if (dent1->d_type != DT_DIR || -- cgit v0.10.2 From 384c671e33a7ddf905f2c60b433b9883f0e5a605 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Sun, 22 Sep 2013 19:44:58 -0600 Subject: perf trace: Add mmap2 handler 5c5e854b changed perf_event__synthesize_mmap_events to generate MMAP2 events. Since perf-trace does not have a handler for it it dies with a segfault when trying to process files: perf trace -i /tmp/perf.data Segmentation fault Signed-off-by: David Ahern Link: http://lkml.kernel.org/r/1379900700-5186-4-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index fd48534..71aa3e3 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -1055,6 +1055,7 @@ static int trace__replay(struct trace *trace) trace->tool.sample = trace__process_sample; trace->tool.mmap = perf_event__process_mmap; + trace->tool.mmap2 = perf_event__process_mmap2; trace->tool.comm = perf_event__process_comm; trace->tool.exit = perf_event__process_exit; trace->tool.fork = perf_event__process_fork; -- cgit v0.10.2 From 4885235806bb95c3fad5a5238e323cb746c3a49b Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 24 Sep 2013 16:01:13 +1000 Subject: xfs: lock the AIL before removing the buffer item Regression introduced by commit 46f9d2e ("xfs: aborted buf items can be in the AIL") which fails to lock the AIL before removing the item. Spinlock debugging throws a warning about this. Signed-off-by: Dave Chinner Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 88c5ea7..f1d85cf 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -628,6 +628,7 @@ xfs_buf_item_unlock( else if (aborted) { ASSERT(XFS_FORCED_SHUTDOWN(lip->li_mountp)); if (lip->li_flags & XFS_LI_IN_AIL) { + spin_lock(&lip->li_ailp->xa_lock); xfs_trans_ail_delete(lip->li_ailp, lip, SHUTDOWN_LOG_IO_ERROR); } -- cgit v0.10.2 From b313a5f1cb0bf3276d5457b52b9f75a940e7b5e9 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 24 Sep 2013 16:01:14 +1000 Subject: xfs: asserting lock not held during freeing not valid When we free an inode, we do so via RCU. As an RCU lookup can occur at any time before we free an inode, and that lookup takes the inode flags lock, we cannot safely assert that the flags lock is not held just before marking it dead and running call_rcu() to free the inode. We check on allocation of a new inode structre that the lock is not held, so we still have protection against locks being leaked and hence not correctly initialised when allocated out of the slab. Hence just remove the assert... Signed-off-by: Dave Chinner Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 193206b..474807a 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -119,11 +119,6 @@ xfs_inode_free( ip->i_itemp = NULL; } - /* asserts to verify all state is correct here */ - ASSERT(atomic_read(&ip->i_pincount) == 0); - ASSERT(!spin_is_locked(&ip->i_flags_lock)); - ASSERT(!xfs_isiflocked(ip)); - /* * Because we use RCU freeing we need to ensure the inode always * appears to be reclaimed with an invalid inode number when in the @@ -135,6 +130,10 @@ xfs_inode_free( ip->i_ino = 0; spin_unlock(&ip->i_flags_lock); + /* asserts to verify all state is correct here */ + ASSERT(atomic_read(&ip->i_pincount) == 0); + ASSERT(!xfs_isiflocked(ip)); + call_rcu(&VFS_I(ip)->i_rcu, xfs_inode_free_callback); } -- cgit v0.10.2 From b771af2fcb4a8a2df7e9b8784cd80e1774c34198 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 24 Sep 2013 16:01:15 +1000 Subject: xfs: fix XFS_IOC_FREE_EOFBLOCKS definition It uses a kernel internal structure in it's definition rather than the user visible structure that is passed to the ioctl. Signed-off-by: Dave Chinner Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index 1edb5cc..18272c7 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -515,7 +515,7 @@ typedef struct xfs_swapext /* XFS_IOC_GETBIOSIZE ---- deprecated 47 */ #define XFS_IOC_GETBMAPX _IOWR('X', 56, struct getbmap) #define XFS_IOC_ZERO_RANGE _IOW ('X', 57, struct xfs_flock64) -#define XFS_IOC_FREE_EOFBLOCKS _IOR ('X', 58, struct xfs_eofblocks) +#define XFS_IOC_FREE_EOFBLOCKS _IOR ('X', 58, struct xfs_fs_eofblocks) /* * ioctl commands that replace IRIX syssgi()'s -- cgit v0.10.2 From 566055d33a91ddddb1cb31220b01ac4abd2d2bdd Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 24 Sep 2013 16:01:16 +1000 Subject: xfs: log recovery lsn ordering needs uuid check After a fair number of xfstests runs, xfs/182 started to fail regularly with a corrupted directory - a directory read verifier was failing after recovery because it found a block with a XARM magic number (remote attribute block) rather than a directory data block. The first time I saw this repeated failure I did /something/ and the problem went away, so I was never able to find the underlying problem. Test xfs/182 failed again today, and I found the root cause before I did /something else/ that made it go away. Tracing indicated that the block in question was being correctly logged, the log was being flushed by sync, but the buffer was not being written back before the shutdown occurred. Tracing also indicated that log recovery was also reading the block, but then never writing it before log recovery invalidated the cache, indicating that it was not modified by log recovery. More detailed analysis of the corpse indicated that the filesystem had a uuid of "a4131074-1872-4cac-9323-2229adbcb886" but the XARM block had a uuid of "8f32f043-c3c9-e7f8-f947-4e7f989c05d3", which indicated it was a block from an older filesystem. The reason that log recovery didn't replay it was that the LSN in the XARM block was larger than the LSN of the transaction being replayed, and so the block was not overwritten by log recovery. Hence, log recovery cant blindly trust the magic number and LSN in the block - it must verify that it belongs to the filesystem being recovered before using the LSN. i.e. if the UUIDs don't match, we need to unconditionally recovery the change held in the log. This patch was first tested on a block device that was repeatedly causing xfs/182 to fail with the same failure on the same block with the same directory read corruption signature (i.e. XARM block). It did not fail, and hasn't failed since. Signed-off-by: Dave Chinner Reviewed-by: Ben Myers Signed-off-by: Ben Myers diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index dabda95..cc17987 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1970,6 +1970,13 @@ xlog_recover_do_inode_buffer( * magic number. If we don't recognise the magic number in the buffer, then * return a LSN of -1 so that the caller knows it was an unrecognised block and * so can recover the buffer. + * + * Note: we cannot rely solely on magic number matches to determine that the + * buffer has a valid LSN - we also need to verify that it belongs to this + * filesystem, so we need to extract the object's LSN and compare it to that + * which we read from the superblock. If the UUIDs don't match, then we've got a + * stale metadata block from an old filesystem instance that we need to recover + * over the top of. */ static xfs_lsn_t xlog_recover_get_buf_lsn( @@ -1980,6 +1987,8 @@ xlog_recover_get_buf_lsn( __uint16_t magic16; __uint16_t magicda; void *blk = bp->b_addr; + uuid_t *uuid; + xfs_lsn_t lsn = -1; /* v4 filesystems always recover immediately */ if (!xfs_sb_version_hascrc(&mp->m_sb)) @@ -1992,43 +2001,79 @@ xlog_recover_get_buf_lsn( case XFS_ABTB_MAGIC: case XFS_ABTC_MAGIC: case XFS_IBT_CRC_MAGIC: - case XFS_IBT_MAGIC: - return be64_to_cpu( - ((struct xfs_btree_block *)blk)->bb_u.s.bb_lsn); + case XFS_IBT_MAGIC: { + struct xfs_btree_block *btb = blk; + + lsn = be64_to_cpu(btb->bb_u.s.bb_lsn); + uuid = &btb->bb_u.s.bb_uuid; + break; + } case XFS_BMAP_CRC_MAGIC: - case XFS_BMAP_MAGIC: - return be64_to_cpu( - ((struct xfs_btree_block *)blk)->bb_u.l.bb_lsn); + case XFS_BMAP_MAGIC: { + struct xfs_btree_block *btb = blk; + + lsn = be64_to_cpu(btb->bb_u.l.bb_lsn); + uuid = &btb->bb_u.l.bb_uuid; + break; + } case XFS_AGF_MAGIC: - return be64_to_cpu(((struct xfs_agf *)blk)->agf_lsn); + lsn = be64_to_cpu(((struct xfs_agf *)blk)->agf_lsn); + uuid = &((struct xfs_agf *)blk)->agf_uuid; + break; case XFS_AGFL_MAGIC: - return be64_to_cpu(((struct xfs_agfl *)blk)->agfl_lsn); + lsn = be64_to_cpu(((struct xfs_agfl *)blk)->agfl_lsn); + uuid = &((struct xfs_agfl *)blk)->agfl_uuid; + break; case XFS_AGI_MAGIC: - return be64_to_cpu(((struct xfs_agi *)blk)->agi_lsn); + lsn = be64_to_cpu(((struct xfs_agi *)blk)->agi_lsn); + uuid = &((struct xfs_agi *)blk)->agi_uuid; + break; case XFS_SYMLINK_MAGIC: - return be64_to_cpu(((struct xfs_dsymlink_hdr *)blk)->sl_lsn); + lsn = be64_to_cpu(((struct xfs_dsymlink_hdr *)blk)->sl_lsn); + uuid = &((struct xfs_dsymlink_hdr *)blk)->sl_uuid; + break; case XFS_DIR3_BLOCK_MAGIC: case XFS_DIR3_DATA_MAGIC: case XFS_DIR3_FREE_MAGIC: - return be64_to_cpu(((struct xfs_dir3_blk_hdr *)blk)->lsn); + lsn = be64_to_cpu(((struct xfs_dir3_blk_hdr *)blk)->lsn); + uuid = &((struct xfs_dir3_blk_hdr *)blk)->uuid; + break; case XFS_ATTR3_RMT_MAGIC: - return be64_to_cpu(((struct xfs_attr3_rmt_hdr *)blk)->rm_lsn); + lsn = be64_to_cpu(((struct xfs_attr3_rmt_hdr *)blk)->rm_lsn); + uuid = &((struct xfs_attr3_rmt_hdr *)blk)->rm_uuid; + break; case XFS_SB_MAGIC: - return be64_to_cpu(((struct xfs_dsb *)blk)->sb_lsn); + lsn = be64_to_cpu(((struct xfs_dsb *)blk)->sb_lsn); + uuid = &((struct xfs_dsb *)blk)->sb_uuid; + break; default: break; } + if (lsn != (xfs_lsn_t)-1) { + if (!uuid_equal(&mp->m_sb.sb_uuid, uuid)) + goto recover_immediately; + return lsn; + } + magicda = be16_to_cpu(((struct xfs_da_blkinfo *)blk)->magic); switch (magicda) { case XFS_DIR3_LEAF1_MAGIC: case XFS_DIR3_LEAFN_MAGIC: case XFS_DA3_NODE_MAGIC: - return be64_to_cpu(((struct xfs_da3_blkinfo *)blk)->lsn); + lsn = be64_to_cpu(((struct xfs_da3_blkinfo *)blk)->lsn); + uuid = &((struct xfs_da3_blkinfo *)blk)->uuid; + break; default: break; } + if (lsn != (xfs_lsn_t)-1) { + if (!uuid_equal(&mp->m_sb.sb_uuid, uuid)) + goto recover_immediately; + return lsn; + } + /* * We do individual object checks on dquot and inode buffers as they * have their own individual LSN records. Also, we could have a stale -- cgit v0.10.2 From 8d16f258217f2f583af1fd57c5144aa4bbe73e48 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Fri, 20 Sep 2013 16:42:15 +0300 Subject: drm/i915/dp: increase i2c-over-aux retry interval on AUX DEFER There is no clear cut rules or specs for the retry interval, as there are many factors that affect overall response time. Increase the interval, and even more so on branch devices which may have limited i2c bit rates. Signed-off-by: Jani Nikula Reference: https://bugs.freedesktop.org/show_bug.cgi?id=60263 Tested-by: Nicolas Suzor Reviewed-by: Todd Previte Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 2151d13..79c14e2 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -588,7 +588,18 @@ intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode, DRM_DEBUG_KMS("aux_ch native nack\n"); return -EREMOTEIO; case AUX_NATIVE_REPLY_DEFER: - udelay(100); + /* + * For now, just give more slack to branch devices. We + * could check the DPCD for I2C bit rate capabilities, + * and if available, adjust the interval. We could also + * be more careful with DP-to-Legacy adapters where a + * long legacy cable may force very low I2C bit rates. + */ + if (intel_dp->dpcd[DP_DOWNSTREAMPORT_PRESENT] & + DP_DWN_STRM_PORT_PRESENT) + usleep_range(500, 600); + else + usleep_range(300, 400); continue; default: DRM_ERROR("aux_ch invalid native reply 0x%02x\n", -- cgit v0.10.2 From 1062b81598bc00e2f6620e6f3788f8f8df2f01e7 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 10 Sep 2013 11:44:30 +0200 Subject: drm/i915/tv: clear adjusted_mode.flags The native TV encoder has it's own flags to adjust sync modes and enabled interlaced modes which are totally irrelevant for the adjusted mode. This worked out nicely since the input modes used by both the load detect code and reported in the ->get_modes callbacks all have no flags set, and we also don't fill out any of them in the ->get_config callback. This changed with the additional sanitation done with commit 2960bc9cceecb5d556ce1c07656a6609e2f7e8b0 Author: Imre Deak Date: Tue Jul 30 13:36:32 2013 +0300 drm/i915: make user mode sync polarity setting explicit sinc now the "no flags at all" state wouldn't fit through core code any more. So fix this up again by explicitly clearing the flags in the ->compute_config callback. Aside: We have zero checking in place to make sure that the requested mode is indeed the right input mode we want for the selected TV mode. So we'll happily fall over if userspace tries to pull us. But that's definitely work for a different patch series. So just add a FIXME comment for now. Reported-by: Knut Petersen Cc: Knut Petersen Cc: Imre Deak Cc: Chris Wilson Tested-by: Knut Petersen Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c index f2c6d79..dd6f84b 100644 --- a/drivers/gpu/drm/i915/intel_tv.c +++ b/drivers/gpu/drm/i915/intel_tv.c @@ -916,6 +916,14 @@ intel_tv_compute_config(struct intel_encoder *encoder, DRM_DEBUG_KMS("forcing bpc to 8 for TV\n"); pipe_config->pipe_bpp = 8*3; + /* TV has it's own notion of sync and other mode flags, so clear them. */ + pipe_config->adjusted_mode.flags = 0; + + /* + * FIXME: We don't check whether the input mode is actually what we want + * or whether userspace is doing something stupid. + */ + return true; } -- cgit v0.10.2 From 67c72a12254101d4e8d9b9f3a02646ba0be84a2d Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 24 Sep 2013 11:46:14 +0200 Subject: drm/i915: preserve pipe A quirk in i9xx_set_pipeconf MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This regression has been introduced in commit 9f11a9e4e50006b615ba94722dfc33ced89664cf Author: Daniel Vetter Date: Thu Jun 13 00:54:58 2013 +0200 drm/i915: set up PIPECONF explicitly for i9xx/vlv platforms Ville brough up the idea that this is just the pipe A quirk gone wrong. Note that after resume the bios might or might not have enabled pipe A already. We have a bit of magic to make sure that on resume we set up a decent mode for pipe A, but I fear if I just smash pipe A to always on we'd enable it in a bogus state and hang the hw. Hence the readback. v2: Clarify the logic a bit as suggested by Chris. Also amend the commit message to clarify why we don't unconditionally enable the pipe. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=66462 References: https://lkml.org/lkml/2013/8/26/238 Cc: Meelis Roos Cc: Chris Wilson Cc: Ville Syrjälä Reviewed-by: Chris Wilson [danvet: Use |= instead of = as suggested by Chris.] Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index d8a1d98..e5822e7 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4775,6 +4775,10 @@ static void i9xx_set_pipeconf(struct intel_crtc *intel_crtc) pipeconf = 0; + if (dev_priv->quirks & QUIRK_PIPEA_FORCE && + I915_READ(PIPECONF(intel_crtc->pipe)) & PIPECONF_ENABLE) + pipeconf |= PIPECONF_ENABLE; + if (intel_crtc->pipe == 0 && INTEL_INFO(dev)->gen < 4) { /* Enable pixel doubling when the dot clock is > 90% of the (display) * core speed. -- cgit v0.10.2 From 56be88954bdb3d77619a786bf301af01efb2b057 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Mon, 23 Sep 2013 12:14:33 -0600 Subject: tpm: xen-tpmfront: Fix default durations All the default durations were being set to 10 minutes which is way too long for the timeouts. Normal values for the longest duration are around 5 mins, and short duration ar around .5s. Further, these are just the default, tpm_get_timeouts will set them to values from the TPM (or throw an error). Just remove them. Acked-by: Daniel De Graaf Cc: Konrad Rzeszutek Wilk Signed-off-by: Jason Gunthorpe Signed-off-by: Konrad Rzeszutek Wilk diff --git a/drivers/char/tpm/xen-tpmfront.c b/drivers/char/tpm/xen-tpmfront.c index 7a7929b..6f2fe2b 100644 --- a/drivers/char/tpm/xen-tpmfront.c +++ b/drivers/char/tpm/xen-tpmfront.c @@ -210,8 +210,6 @@ static struct attribute_group vtpm_attr_grp = { .attrs = vtpm_attrs, }; -#define TPM_LONG_TIMEOUT (10 * 60 * HZ) - static const struct tpm_vendor_specific tpm_vtpm = { .status = vtpm_status, .recv = vtpm_recv, @@ -224,11 +222,6 @@ static const struct tpm_vendor_specific tpm_vtpm = { .miscdev = { .fops = &vtpm_ops, }, - .duration = { - TPM_LONG_TIMEOUT, - TPM_LONG_TIMEOUT, - TPM_LONG_TIMEOUT, - }, }; static irqreturn_t tpmif_interrupt(int dummy, void *dev_id) -- cgit v0.10.2 From bf4a7c054be1306dcee5d4f5028eb66caf714037 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Mon, 23 Sep 2013 14:54:15 -0600 Subject: tpm: xen-tpmfront: Remove the locality sysfs attribute Upon deeper review it was agreed to remove the driver-unique 'locality' sysfs attribute before it is present in a released kernel. The attribute was introduced in e2683957fb268c6b29316fd9e7191e13239a30a5 during the 3.12 merge window, so this patch needs to go in before 3.12 is released. The hope is to have a well defined locality API that all the other locality aware drivers can use, perhaps in 3.13. Signed-off-by: Jason Gunthorpe Signed-off-by: Konrad Rzeszutek Wilk Acked-by: Daniel De Graaf diff --git a/drivers/char/tpm/xen-tpmfront.c b/drivers/char/tpm/xen-tpmfront.c index 6f2fe2b..06189e5 100644 --- a/drivers/char/tpm/xen-tpmfront.c +++ b/drivers/char/tpm/xen-tpmfront.c @@ -142,32 +142,6 @@ static int vtpm_recv(struct tpm_chip *chip, u8 *buf, size_t count) return length; } -ssize_t tpm_show_locality(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct tpm_chip *chip = dev_get_drvdata(dev); - struct tpm_private *priv = TPM_VPRIV(chip); - u8 locality = priv->shr->locality; - - return sprintf(buf, "%d\n", locality); -} - -ssize_t tpm_store_locality(struct device *dev, struct device_attribute *attr, - const char *buf, size_t len) -{ - struct tpm_chip *chip = dev_get_drvdata(dev); - struct tpm_private *priv = TPM_VPRIV(chip); - u8 val; - - int rv = kstrtou8(buf, 0, &val); - if (rv) - return rv; - - priv->shr->locality = val; - - return len; -} - static const struct file_operations vtpm_ops = { .owner = THIS_MODULE, .llseek = no_llseek, @@ -188,8 +162,6 @@ static DEVICE_ATTR(caps, S_IRUGO, tpm_show_caps, NULL); static DEVICE_ATTR(cancel, S_IWUSR | S_IWGRP, NULL, tpm_store_cancel); static DEVICE_ATTR(durations, S_IRUGO, tpm_show_durations, NULL); static DEVICE_ATTR(timeouts, S_IRUGO, tpm_show_timeouts, NULL); -static DEVICE_ATTR(locality, S_IRUGO | S_IWUSR, tpm_show_locality, - tpm_store_locality); static struct attribute *vtpm_attrs[] = { &dev_attr_pubek.attr, @@ -202,7 +174,6 @@ static struct attribute *vtpm_attrs[] = { &dev_attr_cancel.attr, &dev_attr_durations.attr, &dev_attr_timeouts.attr, - &dev_attr_locality.attr, NULL, }; -- cgit v0.10.2 From a945928ea2709bc0e8e8165d33aed855a0110279 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 12 Sep 2013 22:29:44 -0400 Subject: xen: Do not enable spinlocks before jump_label_init() has executed xen_init_spinlocks() currently calls static_key_slow_inc() before jump_label_init() is invoked. When CONFIG_JUMP_LABEL is set (which usually is the case) the effect of this static_key_slow_inc() is deferred until after jump_label_init(). This is different from when CONFIG_JUMP_LABEL is not set, in which case the key is set immediately. Thus, depending on the value of config option, we may observe different behavior. In addition, when we come to __jump_label_transform() from jump_label_init(), the key (paravirt_ticketlocks_enabled) is already enabled. On processors where ideal_nop is not the same as default_nop this will cause a BUG() since it is expected that before a key is enabled the latter is replaced by the former during initialization. To address this problem we need to move static_key_slow_inc(¶virt_ticketlocks_enabled) so that it is called after jump_label_init(). We also need to make sure that this is done before other cpus start to boot. early_initcall appears to be a good place to do so. (Note that we cannot move whole xen_init_spinlocks() there since pv_lock_ops need to be set before alternative_instructions() runs.) Signed-off-by: Konrad Rzeszutek Wilk [v2: Added extra comments in the code] Signed-off-by: Boris Ostrovsky Signed-off-by: Konrad Rzeszutek Wilk Reviewed-by: Steven Rostedt diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c index 253f63f..be6b860 100644 --- a/arch/x86/xen/spinlock.c +++ b/arch/x86/xen/spinlock.c @@ -259,6 +259,14 @@ void xen_uninit_lock_cpu(int cpu) } +/* + * Our init of PV spinlocks is split in two init functions due to us + * using paravirt patching and jump labels patching and having to do + * all of this before SMP code is invoked. + * + * The paravirt patching needs to be done _before_ the alternative asm code + * is started, otherwise we would not patch the core kernel code. + */ void __init xen_init_spinlocks(void) { @@ -267,12 +275,26 @@ void __init xen_init_spinlocks(void) return; } - static_key_slow_inc(¶virt_ticketlocks_enabled); - pv_lock_ops.lock_spinning = PV_CALLEE_SAVE(xen_lock_spinning); pv_lock_ops.unlock_kick = xen_unlock_kick; } +/* + * While the jump_label init code needs to happend _after_ the jump labels are + * enabled and before SMP is started. Hence we use pre-SMP initcall level + * init. We cannot do it in xen_init_spinlocks as that is done before + * jump labels are activated. + */ +static __init int xen_init_spinlocks_jump(void) +{ + if (!xen_pvspin) + return 0; + + static_key_slow_inc(¶virt_ticketlocks_enabled); + return 0; +} +early_initcall(xen_init_spinlocks_jump); + static __init int xen_parse_nopvspin(char *arg) { xen_pvspin = false; -- cgit v0.10.2 From 24f69373e212d4356268fbb1d01c5735423c300d Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Thu, 19 Sep 2013 17:14:53 +0100 Subject: xen/balloon: don't alloc page while non-preemptible get_balloon_scratch_page() disables preemption so we cannot call alloc_page() in between get/put_balloon_scratch_page(). Shuffle bits around in decrease_reservation() to avoid this. Signed-off-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index a50c6e3..b232908 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -398,8 +398,6 @@ static enum bp_state decrease_reservation(unsigned long nr_pages, gfp_t gfp) if (nr_pages > ARRAY_SIZE(frame_list)) nr_pages = ARRAY_SIZE(frame_list); - scratch_page = get_balloon_scratch_page(); - for (i = 0; i < nr_pages; i++) { page = alloc_page(gfp); if (page == NULL) { @@ -413,6 +411,12 @@ static enum bp_state decrease_reservation(unsigned long nr_pages, gfp_t gfp) scrub_page(page); + /* + * Ballooned out frames are effectively replaced with + * a scratch frame. Ensure direct mappings and the + * p2m are consistent. + */ + scratch_page = get_balloon_scratch_page(); #ifdef CONFIG_XEN_HAVE_PVMMU if (xen_pv_domain() && !PageHighMem(page)) { ret = HYPERVISOR_update_va_mapping( @@ -422,24 +426,19 @@ static enum bp_state decrease_reservation(unsigned long nr_pages, gfp_t gfp) BUG_ON(ret); } #endif - } - - /* Ensure that ballooned highmem pages don't have kmaps. */ - kmap_flush_unused(); - flush_tlb_all(); - - /* No more mappings: invalidate P2M and add to balloon. */ - for (i = 0; i < nr_pages; i++) { - pfn = mfn_to_pfn(frame_list[i]); if (!xen_feature(XENFEAT_auto_translated_physmap)) { unsigned long p; p = page_to_pfn(scratch_page); __set_phys_to_machine(pfn, pfn_to_mfn(p)); } + put_balloon_scratch_page(); + balloon_append(pfn_to_page(pfn)); } - put_balloon_scratch_page(); + /* Ensure that ballooned highmem pages don't have kmaps. */ + kmap_flush_unused(); + flush_tlb_all(); set_xen_guest_handle(reservation.extent_start, frame_list); reservation.nr_extents = nr_pages; -- cgit v0.10.2 From 2f6cf0de0281d210061ce976f2d42d246adc75bb Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:26 -0700 Subject: block: Fix bio_copy_data() The memcpy() in bio_copy_data() was using the wrong offset vars, leading to data corruption in weird unusual setups. Signed-off-by: Kent Overstreet Cc: Jens Axboe Cc: linux-stable # >= v3.9 Signed-off-by: Linus Torvalds diff --git a/fs/bio.c b/fs/bio.c index b3b20ed..ea5035d 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -917,8 +917,8 @@ void bio_copy_data(struct bio *dst, struct bio *src) src_p = kmap_atomic(src_bv->bv_page); dst_p = kmap_atomic(dst_bv->bv_page); - memcpy(dst_p + dst_bv->bv_offset, - src_p + src_bv->bv_offset, + memcpy(dst_p + dst_offset, + src_p + src_offset, bytes); kunmap_atomic(dst_p); -- cgit v0.10.2 From 6d9d21e35fbfa2934339e96934f862d118abac23 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:27 -0700 Subject: bcache: Fix a dumb journal discard bug That switch statement was obviously wrong, leading to some sort of weird spinning on rare occasion with discards enabled... Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c index ba95ab8..c0017ca 100644 --- a/drivers/md/bcache/journal.c +++ b/drivers/md/bcache/journal.c @@ -428,7 +428,7 @@ static void do_journal_discard(struct cache *ca) return; } - switch (atomic_read(&ja->discard_in_flight) == DISCARD_IN_FLIGHT) { + switch (atomic_read(&ja->discard_in_flight)) { case DISCARD_IN_FLIGHT: return; -- cgit v0.10.2 From aee6f1cfff3ce240eb4b43b41ca466b907acbd2e Mon Sep 17 00:00:00 2001 From: Gabriel de Perthuis Date: Mon, 23 Sep 2013 23:17:28 -0700 Subject: bcache: Strip endline when writing the label through sysfs sysfs attributes with unusual characters have crappy failure modes in Squeeze (udev 164); later versions of udev are unaffected. This should make these characters more unusual. Signed-off-by: Gabriel de Perthuis Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c index 4fe6ab2..924dcfd 100644 --- a/drivers/md/bcache/sysfs.c +++ b/drivers/md/bcache/sysfs.c @@ -223,8 +223,13 @@ STORE(__cached_dev) } if (attr == &sysfs_label) { - /* note: endlines are preserved */ - memcpy(dc->sb.label, buf, SB_LABEL_SIZE); + if (size > SB_LABEL_SIZE) + return -EINVAL; + memcpy(dc->sb.label, buf, size); + if (size < SB_LABEL_SIZE) + dc->sb.label[size] = '\0'; + if (size && dc->sb.label[size - 1] == '\n') + dc->sb.label[size - 1] = '\0'; bch_write_bdev_super(dc, NULL); if (dc->disk.c) { memcpy(dc->disk.c->uuids[dc->disk.id].label, -- cgit v0.10.2 From c426c4fd46f709ade2bddd51c5738729c7ae1db5 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:29 -0700 Subject: bcache: Fix for when no journal entries are found The journal replay code didn't handle this case, causing it to go into an infinite loop... Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c index c0017ca..f5203df 100644 --- a/drivers/md/bcache/journal.c +++ b/drivers/md/bcache/journal.c @@ -153,7 +153,8 @@ int bch_journal_read(struct cache_set *c, struct list_head *list, bitmap_zero(bitmap, SB_JOURNAL_BUCKETS); pr_debug("%u journal buckets", ca->sb.njournal_buckets); - /* Read journal buckets ordered by golden ratio hash to quickly + /* + * Read journal buckets ordered by golden ratio hash to quickly * find a sequence of buckets with valid journal entries */ for (i = 0; i < ca->sb.njournal_buckets; i++) { @@ -166,18 +167,20 @@ int bch_journal_read(struct cache_set *c, struct list_head *list, goto bsearch; } - /* If that fails, check all the buckets we haven't checked + /* + * If that fails, check all the buckets we haven't checked * already */ pr_debug("falling back to linear search"); - for (l = 0; l < ca->sb.njournal_buckets; l++) { - if (test_bit(l, bitmap)) - continue; - + for (l = find_first_zero_bit(bitmap, ca->sb.njournal_buckets); + l < ca->sb.njournal_buckets; + l = find_next_zero_bit(bitmap, ca->sb.njournal_buckets, l + 1)) if (read_bucket(l)) goto bsearch; - } + + if (list_empty(list)) + continue; bsearch: /* Binary search */ m = r = find_next_bit(bitmap, ca->sb.njournal_buckets, l + 1); @@ -197,10 +200,12 @@ bsearch: r = m; } - /* Read buckets in reverse order until we stop finding more + /* + * Read buckets in reverse order until we stop finding more * journal entries */ - pr_debug("finishing up"); + pr_debug("finishing up: m %u njournal_buckets %u", + m, ca->sb.njournal_buckets); l = m; while (1) { @@ -228,9 +233,10 @@ bsearch: } } - c->journal.seq = list_entry(list->prev, - struct journal_replay, - list)->j.seq; + if (!list_empty(list)) + c->journal.seq = list_entry(list->prev, + struct journal_replay, + list)->j.seq; return 0; #undef read_bucket -- cgit v0.10.2 From 61cbd250f867f98bb4738000afc6002d6f2b14bd Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 23 Sep 2013 23:17:30 -0700 Subject: bcache: Correct printf()-style format length modifier MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix drivers/md/bcache/btree.c: In function ‘bch_btree_node_read’: drivers/md/bcache/btree.c:259: warning: format ‘%lu’ expects type ‘long unsigned int’, but argument 3 has type ‘size_t’ Signed-off-by: Geert Uytterhoeven Signed-off-by: Kent Overstreet Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index f9764e6..8fad840 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -255,7 +255,7 @@ void bch_btree_node_read(struct btree *b) return; err: - bch_cache_set_error(b->c, "io error reading bucket %lu", + bch_cache_set_error(b->c, "io error reading bucket %zu", PTR_BUCKET_NR(b->c, &b->key, 0)); } -- cgit v0.10.2 From c2a4f3183a1248f615a695fbd8905da55ad11bba Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:31 -0700 Subject: bcache: Fix a writeback performance regression Background writeback works by scanning the btree for dirty data and adding those keys into a fixed size buffer, then for each dirty key in the keybuf writing it to the backing device. When read_dirty() finishes and it's time to scan for more dirty data, we need to wait for the outstanding writeback IO to finish - they still take up slots in the keybuf (so that foreground writes can check for them to avoid races) - without that wait, we'll continually rescan when we'll be able to add at most a key or two to the keybuf, and that takes locks that starves foreground IO. Doh. Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index b39f6f0..0f12382 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -498,7 +498,7 @@ struct cached_dev { */ atomic_t has_dirty; - struct ratelimit writeback_rate; + struct bch_ratelimit writeback_rate; struct delayed_work writeback_rate_update; /* @@ -507,10 +507,9 @@ struct cached_dev { */ sector_t last_read; - /* Number of writeback bios in flight */ - atomic_t in_flight; + /* Limit number of writeback bios in flight */ + struct semaphore in_flight; struct closure_with_timer writeback; - struct closure_waitlist writeback_wait; struct keybuf writeback_keys; diff --git a/drivers/md/bcache/util.c b/drivers/md/bcache/util.c index 98eb811..420dad5 100644 --- a/drivers/md/bcache/util.c +++ b/drivers/md/bcache/util.c @@ -190,7 +190,16 @@ void bch_time_stats_update(struct time_stats *stats, uint64_t start_time) stats->last = now ?: 1; } -unsigned bch_next_delay(struct ratelimit *d, uint64_t done) +/** + * bch_next_delay() - increment @d by the amount of work done, and return how + * long to delay until the next time to do some work. + * + * @d - the struct bch_ratelimit to update + * @done - the amount of work done, in arbitrary units + * + * Returns the amount of time to delay by, in jiffies + */ +uint64_t bch_next_delay(struct bch_ratelimit *d, uint64_t done) { uint64_t now = local_clock(); diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h index 1ae2a73..ea345c6 100644 --- a/drivers/md/bcache/util.h +++ b/drivers/md/bcache/util.h @@ -450,17 +450,23 @@ read_attribute(name ## _last_ ## frequency_units) (ewma) >> factor; \ }) -struct ratelimit { +struct bch_ratelimit { + /* Next time we want to do some work, in nanoseconds */ uint64_t next; + + /* + * Rate at which we want to do work, in units per nanosecond + * The units here correspond to the units passed to bch_next_delay() + */ unsigned rate; }; -static inline void ratelimit_reset(struct ratelimit *d) +static inline void bch_ratelimit_reset(struct bch_ratelimit *d) { d->next = local_clock(); } -unsigned bch_next_delay(struct ratelimit *d, uint64_t done); +uint64_t bch_next_delay(struct bch_ratelimit *d, uint64_t done); #define __DIV_SAFE(n, d, zero) \ ({ \ diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c index 22cbff5..27ac519 100644 --- a/drivers/md/bcache/writeback.c +++ b/drivers/md/bcache/writeback.c @@ -94,11 +94,15 @@ static void update_writeback_rate(struct work_struct *work) static unsigned writeback_delay(struct cached_dev *dc, unsigned sectors) { + uint64_t ret; + if (atomic_read(&dc->disk.detaching) || !dc->writeback_percent) return 0; - return bch_next_delay(&dc->writeback_rate, sectors * 10000000ULL); + ret = bch_next_delay(&dc->writeback_rate, sectors * 10000000ULL); + + return min_t(uint64_t, ret, HZ); } /* Background writeback */ @@ -208,7 +212,7 @@ normal_refill: up_write(&dc->writeback_lock); - ratelimit_reset(&dc->writeback_rate); + bch_ratelimit_reset(&dc->writeback_rate); /* Punt to workqueue only so we don't recurse and blow the stack */ continue_at(cl, read_dirty, dirty_wq); @@ -318,9 +322,7 @@ static void write_dirty_finish(struct closure *cl) } bch_keybuf_del(&dc->writeback_keys, w); - atomic_dec_bug(&dc->in_flight); - - closure_wake_up(&dc->writeback_wait); + up(&dc->in_flight); closure_return_with_destructor(cl, dirty_io_destructor); } @@ -349,7 +351,7 @@ static void write_dirty(struct closure *cl) closure_bio_submit(&io->bio, cl, &io->dc->disk); - continue_at(cl, write_dirty_finish, dirty_wq); + continue_at(cl, write_dirty_finish, system_wq); } static void read_dirty_endio(struct bio *bio, int error) @@ -369,7 +371,7 @@ static void read_dirty_submit(struct closure *cl) closure_bio_submit(&io->bio, cl, &io->dc->disk); - continue_at(cl, write_dirty, dirty_wq); + continue_at(cl, write_dirty, system_wq); } static void read_dirty(struct closure *cl) @@ -394,12 +396,9 @@ static void read_dirty(struct closure *cl) if (delay > 0 && (KEY_START(&w->key) != dc->last_read || - jiffies_to_msecs(delay) > 50)) { - w->private = NULL; - - closure_delay(&dc->writeback, delay); - continue_at(cl, read_dirty, dirty_wq); - } + jiffies_to_msecs(delay) > 50)) + while (delay) + delay = schedule_timeout(delay); dc->last_read = KEY_OFFSET(&w->key); @@ -424,15 +423,10 @@ static void read_dirty(struct closure *cl) trace_bcache_writeback(&w->key); - closure_call(&io->cl, read_dirty_submit, NULL, &dc->disk.cl); + down(&dc->in_flight); + closure_call(&io->cl, read_dirty_submit, NULL, cl); delay = writeback_delay(dc, KEY_SIZE(&w->key)); - - atomic_inc(&dc->in_flight); - - if (!closure_wait_event(&dc->writeback_wait, cl, - atomic_read(&dc->in_flight) < 64)) - continue_at(cl, read_dirty, dirty_wq); } if (0) { @@ -442,7 +436,11 @@ err: bch_keybuf_del(&dc->writeback_keys, w); } - refill_dirty(cl); + /* + * Wait for outstanding writeback IOs to finish (and keybuf slots to be + * freed) before refilling again + */ + continue_at(cl, refill_dirty, dirty_wq); } /* Init */ @@ -484,6 +482,7 @@ void bch_sectors_dirty_init(struct cached_dev *dc) void bch_cached_dev_writeback_init(struct cached_dev *dc) { + sema_init(&dc->in_flight, 64); closure_init_unlocked(&dc->writeback); init_rwsem(&dc->writeback_lock); @@ -513,7 +512,7 @@ void bch_writeback_exit(void) int __init bch_writeback_init(void) { - dirty_wq = create_singlethread_workqueue("bcache_writeback"); + dirty_wq = create_workqueue("bcache_writeback"); if (!dirty_wq) return -ENOMEM; -- cgit v0.10.2 From 1394d6761b6e9e15ee7c632a6d48791188727b40 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:32 -0700 Subject: bcache: Fix a flush/fua performance bug bch_journal_meta() was missing the flush to make the journal write actually go down (instead of waiting up to journal_delay_ms)... Whoops Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c index f5203df..8435f81 100644 --- a/drivers/md/bcache/journal.c +++ b/drivers/md/bcache/journal.c @@ -695,6 +695,7 @@ void bch_journal_meta(struct cache_set *c, struct closure *cl) if (cl) BUG_ON(!closure_wait(&w->wait, cl)); + closure_flush(&c->journal.io); __journal_try_write(c, true); } } -- cgit v0.10.2 From 79e3dab90d9f826ceca67c7890e048ac9169de49 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:33 -0700 Subject: bcache: Fix a dumb CPU spinning bug in writeback schedule_timeout() != schedule_timeout_uninterruptible() Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c index 27ac519..ba3ee48 100644 --- a/drivers/md/bcache/writeback.c +++ b/drivers/md/bcache/writeback.c @@ -397,8 +397,7 @@ static void read_dirty(struct closure *cl) if (delay > 0 && (KEY_START(&w->key) != dc->last_read || jiffies_to_msecs(delay) > 50)) - while (delay) - delay = schedule_timeout(delay); + delay = schedule_timeout_uninterruptible(delay); dc->last_read = KEY_OFFSET(&w->key); -- cgit v0.10.2 From a698e08c82dfb9771e0bac12c7337c706d729b6d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:34 -0700 Subject: bcache: Fix a shrinker deadlock GFP_NOIO means we could be getting called recursively - mca_alloc() -> mca_data_alloc() - definitely can't use mutex_lock(bucket_lock) then. Whoops. Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index 8fad840..f42fc7e 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -612,7 +612,7 @@ static unsigned long bch_mca_scan(struct shrinker *shrink, return SHRINK_STOP; /* Return -1 if we can't do anything right now */ - if (sc->gfp_mask & __GFP_WAIT) + if (sc->gfp_mask & __GFP_IO) mutex_lock(&c->bucket_lock); else if (!mutex_trylock(&c->bucket_lock)) return -1; -- cgit v0.10.2 From 84786438ed17978d72eeced580ab757e4da8830b Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:35 -0700 Subject: bcache: Fix for handling overlapping extents when reading in a btree node btree_sort_fixup() was overly clever, because it was trying to avoid pulling a key off the btree iterator in more than one place. This led to a really obscure bug where we'd break early from the loop in btree_sort_fixup() if the current key overlapped with keys in more than one older set, and the next key it overlapped with was zero size. Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/bset.c b/drivers/md/bcache/bset.c index 8010eed..22d1ae7 100644 --- a/drivers/md/bcache/bset.c +++ b/drivers/md/bcache/bset.c @@ -926,28 +926,45 @@ struct bkey *bch_next_recurse_key(struct btree *b, struct bkey *search) /* Mergesort */ +static void sort_key_next(struct btree_iter *iter, + struct btree_iter_set *i) +{ + i->k = bkey_next(i->k); + + if (i->k == i->end) + *i = iter->data[--iter->used]; +} + static void btree_sort_fixup(struct btree_iter *iter) { while (iter->used > 1) { struct btree_iter_set *top = iter->data, *i = top + 1; - struct bkey *k; if (iter->used > 2 && btree_iter_cmp(i[0], i[1])) i++; - for (k = i->k; - k != i->end && bkey_cmp(top->k, &START_KEY(k)) > 0; - k = bkey_next(k)) - if (top->k > i->k) - __bch_cut_front(top->k, k); - else if (KEY_SIZE(k)) - bch_cut_back(&START_KEY(k), top->k); - - if (top->k < i->k || k == i->k) + if (bkey_cmp(top->k, &START_KEY(i->k)) <= 0) break; - heap_sift(iter, i - top, btree_iter_cmp); + if (!KEY_SIZE(i->k)) { + sort_key_next(iter, i); + heap_sift(iter, i - top, btree_iter_cmp); + continue; + } + + if (top->k > i->k) { + if (bkey_cmp(top->k, i->k) >= 0) + sort_key_next(iter, i); + else + bch_cut_front(top->k, i->k); + + heap_sift(iter, i - top, btree_iter_cmp); + } else { + /* can't happen because of comparison func */ + BUG_ON(!bkey_cmp(&START_KEY(top->k), &START_KEY(i->k))); + bch_cut_back(&START_KEY(i->k), top->k); + } } } -- cgit v0.10.2 From c0f04d88e46d14de51f4baebb6efafb7d59e9f96 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 23 Sep 2013 23:17:36 -0700 Subject: bcache: Fix flushes in writeback mode In writeback mode, when we get a cache flush we need to make sure we issue a flush to the backing device. The code for sending down an extra flush was wrong - by cloning the bio we were probably getting flags that didn't make sense for a bare flush, and also the old code was firing for FUA bios, for which we don't need to send a flush to the backing device. This was causing data corruption somehow - the mechanism was never determined, but this patch fixes it for the users that were seeing it. Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Signed-off-by: Linus Torvalds diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index 786a1a4..71eb233 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -997,14 +997,17 @@ static void request_write(struct cached_dev *dc, struct search *s) } else { bch_writeback_add(dc); - if (s->op.flush_journal) { + if (bio->bi_rw & REQ_FLUSH) { /* Also need to send a flush to the backing device */ - s->op.cache_bio = bio_clone_bioset(bio, GFP_NOIO, - dc->disk.bio_split); + struct bio *flush = bio_alloc_bioset(0, GFP_NOIO, + dc->disk.bio_split); - bio->bi_size = 0; - bio->bi_vcnt = 0; - closure_bio_submit(bio, cl, s->d); + flush->bi_rw = WRITE_FLUSH; + flush->bi_bdev = bio->bi_bdev; + flush->bi_end_io = request_endio; + flush->bi_private = cl; + + closure_bio_submit(flush, cl, s->d); } else { s->op.cache_bio = bio; } -- cgit v0.10.2 From 359e6fab6600562073162348cd4c18c5958296d8 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Tue, 24 Sep 2013 15:27:29 -0700 Subject: watchdog: update watchdog attributes atomically proc_dowatchdog doesn't synchronize multiple callers which might lead to confusion when two parallel callers might confuse watchdog_enable_all_cpus resp watchdog_disable_all_cpus (eg watchdog gets enabled even if watchdog_thresh was set to 0 already). This patch adds a local mutex which synchronizes callers to the sysctl handler. Signed-off-by: Michal Hocko Cc: Frederic Weisbecker Acked-by: Don Zickus Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 51c4f34..ced7d06 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -520,13 +520,15 @@ int proc_dowatchdog(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { int err, old_thresh, old_enabled; + static DEFINE_MUTEX(watchdog_proc_mutex); + mutex_lock(&watchdog_proc_mutex); old_thresh = ACCESS_ONCE(watchdog_thresh); old_enabled = ACCESS_ONCE(watchdog_user_enabled); err = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (err || !write) - return err; + goto out; set_sample_period(); /* @@ -544,7 +546,8 @@ int proc_dowatchdog(struct ctl_table *table, int write, watchdog_thresh = old_thresh; watchdog_user_enabled = old_enabled; } - +out: + mutex_unlock(&watchdog_proc_mutex); return err; } #endif /* CONFIG_SYSCTL */ -- cgit v0.10.2 From 9809b18fcf6b8d8ec4d3643677345907e6b50eca Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Tue, 24 Sep 2013 15:27:30 -0700 Subject: watchdog: update watchdog_thresh properly watchdog_tresh controls how often nmi perf event counter checks per-cpu hrtimer_interrupts counter and blows up if the counter hasn't changed since the last check. The counter is updated by per-cpu watchdog_hrtimer hrtimer which is scheduled with 2/5 watchdog_thresh period which guarantees that hrtimer is scheduled 2 times per the main period. Both hrtimer and perf event are started together when the watchdog is enabled. So far so good. But... But what happens when watchdog_thresh is updated from sysctl handler? proc_dowatchdog will set a new sampling period and hrtimer callback (watchdog_timer_fn) will use the new value in the next round. The problem, however, is that nobody tells the perf event that the sampling period has changed so it is ticking with the period configured when it has been set up. This might result in an ear ripping dissonance between perf and hrtimer parts if the watchdog_thresh is increased. And even worse it might lead to KABOOM if the watchdog is configured to panic on such a spurious lockup. This patch fixes the issue by updating both nmi perf even counter and hrtimers if the threshold value has changed. The nmi one is disabled and then reinitialized from scratch. This has an unpleasant side effect that the allocation of the new event might fail theoretically so the hard lockup detector would be disabled for such cpus. On the other hand such a memory allocation failure is very unlikely because the original event is deallocated right before. It would be much nicer if we just changed perf event period but there doesn't seem to be any API to do that right now. It is also unfortunate that perf_event_alloc uses GFP_KERNEL allocation unconditionally so we cannot use on_each_cpu() and do the same thing from the per-cpu context. The update from the current CPU should be safe because perf_event_disable removes the event atomically before it clears the per-cpu watchdog_ev so it cannot change anything under running handler feet. The hrtimer is simply restarted (thanks to Don Zickus who has pointed this out) if it is queued because we cannot rely it will fire&adopt to the new sampling period before a new nmi event triggers (when the treshold is decreased). [akpm@linux-foundation.org: the UP version of __smp_call_function_single ended up in the wrong place] Signed-off-by: Michal Hocko Acked-by: Don Zickus Cc: Frederic Weisbecker Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Fabio Estevam Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/smp.h b/include/linux/smp.h index cfb7ca0..731f523 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -155,6 +155,12 @@ smp_call_function_any(const struct cpumask *mask, smp_call_func_t func, static inline void kick_all_cpus_sync(void) { } +static inline void __smp_call_function_single(int cpuid, + struct call_single_data *data, int wait) +{ + on_each_cpu(data->func, data->info, wait); +} + #endif /* !SMP */ /* diff --git a/kernel/watchdog.c b/kernel/watchdog.c index ced7d06..4431610 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -486,7 +486,52 @@ static struct smp_hotplug_thread watchdog_threads = { .unpark = watchdog_enable, }; -static int watchdog_enable_all_cpus(void) +static void restart_watchdog_hrtimer(void *info) +{ + struct hrtimer *hrtimer = &__raw_get_cpu_var(watchdog_hrtimer); + int ret; + + /* + * No need to cancel and restart hrtimer if it is currently executing + * because it will reprogram itself with the new period now. + * We should never see it unqueued here because we are running per-cpu + * with interrupts disabled. + */ + ret = hrtimer_try_to_cancel(hrtimer); + if (ret == 1) + hrtimer_start(hrtimer, ns_to_ktime(sample_period), + HRTIMER_MODE_REL_PINNED); +} + +static void update_timers(int cpu) +{ + struct call_single_data data = {.func = restart_watchdog_hrtimer}; + /* + * Make sure that perf event counter will adopt to a new + * sampling period. Updating the sampling period directly would + * be much nicer but we do not have an API for that now so + * let's use a big hammer. + * Hrtimer will adopt the new period on the next tick but this + * might be late already so we have to restart the timer as well. + */ + watchdog_nmi_disable(cpu); + __smp_call_function_single(cpu, &data, 1); + watchdog_nmi_enable(cpu); +} + +static void update_timers_all_cpus(void) +{ + int cpu; + + get_online_cpus(); + preempt_disable(); + for_each_online_cpu(cpu) + update_timers(cpu); + preempt_enable(); + put_online_cpus(); +} + +static int watchdog_enable_all_cpus(bool sample_period_changed) { int err = 0; @@ -496,6 +541,8 @@ static int watchdog_enable_all_cpus(void) pr_err("Failed to create watchdog threads, disabled\n"); else watchdog_running = 1; + } else if (sample_period_changed) { + update_timers_all_cpus(); } return err; @@ -537,7 +584,7 @@ int proc_dowatchdog(struct ctl_table *table, int write, * watchdog_*_all_cpus() function takes care of this. */ if (watchdog_user_enabled && watchdog_thresh) - err = watchdog_enable_all_cpus(); + err = watchdog_enable_all_cpus(old_thresh != watchdog_thresh); else watchdog_disable_all_cpus(); @@ -557,5 +604,5 @@ void __init lockup_detector_init(void) set_sample_period(); if (watchdog_user_enabled) - watchdog_enable_all_cpus(); + watchdog_enable_all_cpus(false); } -- cgit v0.10.2 From 99d7a8824a15cb6da046947e9b5a2335d52fbeb7 Mon Sep 17 00:00:00 2001 From: Goldwyn Rodrigues Date: Tue, 24 Sep 2013 15:27:32 -0700 Subject: fs/ocfs2/super.c: use a bigger nodestr in ocfs2_dismount_volume While printing 32-bit node numbers, an 8-byte string is not enough. Increase the size of the string to 12 chars. This got left out in commit 49fa8140e487 ("fs/ocfs2/super.c: Use bigger nodestr to accomodate 32-bit node numbers"). Signed-off-by: Goldwyn Rodrigues Cc: Joel Becker Cc: Mark Fasheh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 121da2d..d4e81e4 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1924,7 +1924,7 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) { int tmp, hangup_needed = 0; struct ocfs2_super *osb = NULL; - char nodestr[8]; + char nodestr[12]; trace_ocfs2_dismount_volume(sb); -- cgit v0.10.2 From 20ba27f52eded4d14e309ac57971c9c83dec46e4 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:33 -0700 Subject: revert "memcg, vmscan: do not fall into reclaim-all pass too quickly" Revert commit e975de998b96 ("memcg, vmscan: do not fall into reclaim-all pass too quickly") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/vmscan.c b/mm/vmscan.c index 8ed1b77..97b0ed1 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2176,11 +2176,10 @@ static inline bool should_continue_reclaim(struct zone *zone, } } -static int +static void __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) { unsigned long nr_reclaimed, nr_scanned; - int groups_scanned = 0; do { struct mem_cgroup *root = sc->target_mem_cgroup; @@ -2198,7 +2197,6 @@ __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) while ((memcg = mem_cgroup_iter_cond(root, memcg, &reclaim, filter))) { struct lruvec *lruvec; - groups_scanned++; lruvec = mem_cgroup_zone_lruvec(zone, memcg); shrink_lruvec(lruvec, sc); @@ -2226,8 +2224,6 @@ __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) } while (should_continue_reclaim(zone, sc->nr_reclaimed - nr_reclaimed, sc->nr_scanned - nr_scanned, sc)); - - return groups_scanned; } @@ -2235,19 +2231,8 @@ static void shrink_zone(struct zone *zone, struct scan_control *sc) { bool do_soft_reclaim = mem_cgroup_should_soft_reclaim(sc); unsigned long nr_scanned = sc->nr_scanned; - int scanned_groups; - scanned_groups = __shrink_zone(zone, sc, do_soft_reclaim); - /* - * memcg iterator might race with other reclaimer or start from - * a incomplete tree walk so the tree walk in __shrink_zone - * might have missed groups that are above the soft limit. Try - * another loop to catch up with others. Do it just once to - * prevent from reclaim latencies when other reclaimers always - * preempt this one. - */ - if (do_soft_reclaim && !scanned_groups) - __shrink_zone(zone, sc, do_soft_reclaim); + __shrink_zone(zone, sc, do_soft_reclaim); /* * No group is over the soft limit or those that are do not have -- cgit v0.10.2 From 8f939a9f4c769500c58c31ec81ce17297388f4bf Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:34 -0700 Subject: revert "memcg: track all children over limit in the root" Revert commit 1be171d60bdd ("memcg: track all children over limit in the root") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memcontrol.c b/mm/memcontrol.c index d5ff3ce..5ee0af4 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -851,15 +851,9 @@ static void mem_cgroup_update_soft_limit(struct mem_cgroup *memcg) /* * Necessary to update all ancestors when hierarchy is used * because their event counter is not touched. - * We track children even outside the hierarchy for the root - * cgroup because tree walk starting at root should visit - * all cgroups and we want to prevent from pointless tree - * walk if no children is below the limit. */ while (delta && (parent = parent_mem_cgroup(parent))) atomic_add(delta, &parent->children_in_excess); - if (memcg != root_mem_cgroup && !root_mem_cgroup->use_hierarchy) - atomic_add(delta, &root_mem_cgroup->children_in_excess); spin_unlock(&memcg->soft_lock); } @@ -6112,9 +6106,6 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css) if (memcg->soft_contributed) { while ((memcg = parent_mem_cgroup(memcg))) atomic_dec(&memcg->children_in_excess); - - if (memcg != root_mem_cgroup && !root_mem_cgroup->use_hierarchy) - atomic_dec(&root_mem_cgroup->children_in_excess); } mem_cgroup_destroy_all_caches(memcg); vmpressure_cleanup(&memcg->vmpressure); -- cgit v0.10.2 From 3120055e869f2a208480f238680d097eec8f0e02 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:35 -0700 Subject: revert "memcg, vmscan: do not attempt soft limit reclaim if it would not scan anything" Revert commit e839b6a1c8d0 ("memcg, vmscan: do not attempt soft limit reclaim if it would not scan anything") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 5ee0af4..5cf7726 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1872,11 +1872,7 @@ enum mem_cgroup_filter_t mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, struct mem_cgroup *root) { - struct mem_cgroup *parent; - - if (!memcg) - memcg = root_mem_cgroup; - parent = memcg; + struct mem_cgroup *parent = memcg; if (res_counter_soft_limit_excess(&memcg->res)) return VISIT; diff --git a/mm/vmscan.c b/mm/vmscan.c index 97b0ed1..cdd300b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -142,9 +142,7 @@ static bool global_reclaim(struct scan_control *sc) static bool mem_cgroup_should_soft_reclaim(struct scan_control *sc) { - struct mem_cgroup *root = sc->target_mem_cgroup; - return !mem_cgroup_disabled() && - mem_cgroup_soft_reclaim_eligible(root, root) != SKIP_TREE; + return !mem_cgroup_disabled(); } #else static bool global_reclaim(struct scan_control *sc) -- cgit v0.10.2 From 30361e51cae7a4df3fec89f935a450a6fe6f16fa Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:36 -0700 Subject: revert "memcg: track children in soft limit excess to improve soft limit" Revert commit 7d910c054be4 ("memcg: track children in soft limit excess to improve soft limit") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 5cf7726..916892c 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -124,7 +124,6 @@ static const char * const mem_cgroup_lru_names[] = { */ enum mem_cgroup_events_target { MEM_CGROUP_TARGET_THRESH, - MEM_CGROUP_TARGET_SOFTLIMIT, MEM_CGROUP_TARGET_NUMAINFO, MEM_CGROUP_NTARGETS, }; @@ -303,22 +302,6 @@ struct mem_cgroup { atomic_t numainfo_events; atomic_t numainfo_updating; #endif - /* - * Protects soft_contributed transitions. - * See mem_cgroup_update_soft_limit - */ - spinlock_t soft_lock; - - /* - * If true then this group has increased parents' children_in_excess - * when it got over the soft limit. - * When a group falls bellow the soft limit, parents' children_in_excess - * is decreased and soft_contributed changed to false. - */ - bool soft_contributed; - - /* Number of children that are in soft limit excess */ - atomic_t children_in_excess; struct mem_cgroup_per_node *nodeinfo[0]; /* WARNING: nodeinfo must be the last member here */ @@ -806,9 +789,6 @@ static bool mem_cgroup_event_ratelimit(struct mem_cgroup *memcg, case MEM_CGROUP_TARGET_THRESH: next = val + THRESHOLDS_EVENTS_TARGET; break; - case MEM_CGROUP_TARGET_SOFTLIMIT: - next = val + SOFTLIMIT_EVENTS_TARGET; - break; case MEM_CGROUP_TARGET_NUMAINFO: next = val + NUMAINFO_EVENTS_TARGET; break; @@ -822,42 +802,6 @@ static bool mem_cgroup_event_ratelimit(struct mem_cgroup *memcg, } /* - * Called from rate-limited memcg_check_events when enough - * MEM_CGROUP_TARGET_SOFTLIMIT events are accumulated and it makes sure - * that all the parents up the hierarchy will be notified that this group - * is in excess or that it is not in excess anymore. mmecg->soft_contributed - * makes the transition a single action whenever the state flips from one to - * the other. - */ -static void mem_cgroup_update_soft_limit(struct mem_cgroup *memcg) -{ - unsigned long long excess = res_counter_soft_limit_excess(&memcg->res); - struct mem_cgroup *parent = memcg; - int delta = 0; - - spin_lock(&memcg->soft_lock); - if (excess) { - if (!memcg->soft_contributed) { - delta = 1; - memcg->soft_contributed = true; - } - } else { - if (memcg->soft_contributed) { - delta = -1; - memcg->soft_contributed = false; - } - } - - /* - * Necessary to update all ancestors when hierarchy is used - * because their event counter is not touched. - */ - while (delta && (parent = parent_mem_cgroup(parent))) - atomic_add(delta, &parent->children_in_excess); - spin_unlock(&memcg->soft_lock); -} - -/* * Check events in order. * */ @@ -867,11 +811,8 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page) /* threshold event is triggered in finer grain than soft limit */ if (unlikely(mem_cgroup_event_ratelimit(memcg, MEM_CGROUP_TARGET_THRESH))) { - bool do_softlimit; bool do_numainfo __maybe_unused; - do_softlimit = mem_cgroup_event_ratelimit(memcg, - MEM_CGROUP_TARGET_SOFTLIMIT); #if MAX_NUMNODES > 1 do_numainfo = mem_cgroup_event_ratelimit(memcg, MEM_CGROUP_TARGET_NUMAINFO); @@ -879,8 +820,6 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page) preempt_enable(); mem_cgroup_threshold(memcg); - if (unlikely(do_softlimit)) - mem_cgroup_update_soft_limit(memcg); #if MAX_NUMNODES > 1 if (unlikely(do_numainfo)) atomic_inc(&memcg->numainfo_events); @@ -1864,9 +1803,6 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *memcg) * hierarchy if * a) it is over its soft limit * b) any parent up the hierarchy is over its soft limit - * - * If the given group doesn't have any children over the limit then it - * doesn't make any sense to iterate its subtree. */ enum mem_cgroup_filter_t mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, @@ -1888,8 +1824,6 @@ mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, break; } - if (!atomic_read(&memcg->children_in_excess)) - return SKIP_TREE; return SKIP; } @@ -6021,7 +5955,6 @@ mem_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) mutex_init(&memcg->thresholds_lock); spin_lock_init(&memcg->move_lock); vmpressure_init(&memcg->vmpressure); - spin_lock_init(&memcg->soft_lock); return &memcg->css; @@ -6099,10 +6032,6 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css) mem_cgroup_invalidate_reclaim_iterators(memcg); mem_cgroup_reparent_charges(memcg); - if (memcg->soft_contributed) { - while ((memcg = parent_mem_cgroup(memcg))) - atomic_dec(&memcg->children_in_excess); - } mem_cgroup_destroy_all_caches(memcg); vmpressure_cleanup(&memcg->vmpressure); } -- cgit v0.10.2 From 694fbc0fe78518d06efa63910bf4ecee660e7852 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:37 -0700 Subject: revert "memcg: enhance memcg iterator to support predicates" Revert commit de57780dc659 ("memcg: enhance memcg iterator to support predicates") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 60e9587..ef2b9bd 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -53,23 +53,6 @@ struct mem_cgroup_reclaim_cookie { unsigned int generation; }; -enum mem_cgroup_filter_t { - VISIT, /* visit current node */ - SKIP, /* skip the current node and continue traversal */ - SKIP_TREE, /* skip the whole subtree and continue traversal */ -}; - -/* - * mem_cgroup_filter_t predicate might instruct mem_cgroup_iter_cond how to - * iterate through the hierarchy tree. Each tree element is checked by the - * predicate before it is returned by the iterator. If a filter returns - * SKIP or SKIP_TREE then the iterator code continues traversal (with the - * next node down the hierarchy or the next node that doesn't belong under the - * memcg's subtree). - */ -typedef enum mem_cgroup_filter_t -(*mem_cgroup_iter_filter)(struct mem_cgroup *memcg, struct mem_cgroup *root); - #ifdef CONFIG_MEMCG /* * All "charge" functions with gfp_mask should use GFP_KERNEL or @@ -137,18 +120,9 @@ mem_cgroup_prepare_migration(struct page *page, struct page *newpage, extern void mem_cgroup_end_migration(struct mem_cgroup *memcg, struct page *oldpage, struct page *newpage, bool migration_ok); -struct mem_cgroup *mem_cgroup_iter_cond(struct mem_cgroup *root, - struct mem_cgroup *prev, - struct mem_cgroup_reclaim_cookie *reclaim, - mem_cgroup_iter_filter cond); - -static inline struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root, - struct mem_cgroup *prev, - struct mem_cgroup_reclaim_cookie *reclaim) -{ - return mem_cgroup_iter_cond(root, prev, reclaim, NULL); -} - +struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *, + struct mem_cgroup *, + struct mem_cgroup_reclaim_cookie *); void mem_cgroup_iter_break(struct mem_cgroup *, struct mem_cgroup *); /* @@ -260,8 +234,7 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, mem_cgroup_update_page_stat(page, idx, -1); } -enum mem_cgroup_filter_t -mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, +bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, struct mem_cgroup *root); void __mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx); @@ -376,15 +349,6 @@ static inline void mem_cgroup_end_migration(struct mem_cgroup *memcg, struct page *oldpage, struct page *newpage, bool migration_ok) { } -static inline struct mem_cgroup * -mem_cgroup_iter_cond(struct mem_cgroup *root, - struct mem_cgroup *prev, - struct mem_cgroup_reclaim_cookie *reclaim, - mem_cgroup_iter_filter cond) -{ - /* first call must return non-NULL, second return NULL */ - return (struct mem_cgroup *)(unsigned long)!prev; -} static inline struct mem_cgroup * mem_cgroup_iter(struct mem_cgroup *root, @@ -471,11 +435,10 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, } static inline -enum mem_cgroup_filter_t -mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, +bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, struct mem_cgroup *root) { - return VISIT; + return false; } static inline void mem_cgroup_split_huge_fixup(struct page *head) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 916892c..65e7bec 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -862,15 +862,6 @@ struct mem_cgroup *try_get_mem_cgroup_from_mm(struct mm_struct *mm) return memcg; } -static enum mem_cgroup_filter_t -mem_cgroup_filter(struct mem_cgroup *memcg, struct mem_cgroup *root, - mem_cgroup_iter_filter cond) -{ - if (!cond) - return VISIT; - return cond(memcg, root); -} - /* * Returns a next (in a pre-order walk) alive memcg (with elevated css * ref. count) or NULL if the whole root's subtree has been visited. @@ -878,7 +869,7 @@ mem_cgroup_filter(struct mem_cgroup *memcg, struct mem_cgroup *root, * helper function to be used by mem_cgroup_iter */ static struct mem_cgroup *__mem_cgroup_iter_next(struct mem_cgroup *root, - struct mem_cgroup *last_visited, mem_cgroup_iter_filter cond) + struct mem_cgroup *last_visited) { struct cgroup_subsys_state *prev_css, *next_css; @@ -896,31 +887,11 @@ skip_node: if (next_css) { struct mem_cgroup *mem = mem_cgroup_from_css(next_css); - switch (mem_cgroup_filter(mem, root, cond)) { - case SKIP: + if (css_tryget(&mem->css)) + return mem; + else { prev_css = next_css; goto skip_node; - case SKIP_TREE: - if (mem == root) - return NULL; - /* - * css_rightmost_descendant is not an optimal way to - * skip through a subtree (especially for imbalanced - * trees leaning to right) but that's what we have right - * now. More effective solution would be traversing - * right-up for first non-NULL without calling - * css_next_descendant_pre afterwards. - */ - prev_css = css_rightmost_descendant(next_css); - goto skip_node; - case VISIT: - if (css_tryget(&mem->css)) - return mem; - else { - prev_css = next_css; - goto skip_node; - } - break; } } @@ -984,7 +955,6 @@ static void mem_cgroup_iter_update(struct mem_cgroup_reclaim_iter *iter, * @root: hierarchy root * @prev: previously returned memcg, NULL on first invocation * @reclaim: cookie for shared reclaim walks, NULL for full walks - * @cond: filter for visited nodes, NULL for no filter * * Returns references to children of the hierarchy below @root, or * @root itself, or %NULL after a full round-trip. @@ -997,18 +967,15 @@ static void mem_cgroup_iter_update(struct mem_cgroup_reclaim_iter *iter, * divide up the memcgs in the hierarchy among all concurrent * reclaimers operating on the same zone and priority. */ -struct mem_cgroup *mem_cgroup_iter_cond(struct mem_cgroup *root, +struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root, struct mem_cgroup *prev, - struct mem_cgroup_reclaim_cookie *reclaim, - mem_cgroup_iter_filter cond) + struct mem_cgroup_reclaim_cookie *reclaim) { struct mem_cgroup *memcg = NULL; struct mem_cgroup *last_visited = NULL; - if (mem_cgroup_disabled()) { - /* first call must return non-NULL, second return NULL */ - return (struct mem_cgroup *)(unsigned long)!prev; - } + if (mem_cgroup_disabled()) + return NULL; if (!root) root = root_mem_cgroup; @@ -1019,9 +986,7 @@ struct mem_cgroup *mem_cgroup_iter_cond(struct mem_cgroup *root, if (!root->use_hierarchy && root != root_mem_cgroup) { if (prev) goto out_css_put; - if (mem_cgroup_filter(root, root, cond) == VISIT) - return root; - return NULL; + return root; } rcu_read_lock(); @@ -1044,7 +1009,7 @@ struct mem_cgroup *mem_cgroup_iter_cond(struct mem_cgroup *root, last_visited = mem_cgroup_iter_load(iter, root, &seq); } - memcg = __mem_cgroup_iter_next(root, last_visited, cond); + memcg = __mem_cgroup_iter_next(root, last_visited); if (reclaim) { mem_cgroup_iter_update(iter, last_visited, memcg, seq); @@ -1055,11 +1020,7 @@ struct mem_cgroup *mem_cgroup_iter_cond(struct mem_cgroup *root, reclaim->generation = iter->generation; } - /* - * We have finished the whole tree walk or no group has been - * visited because filter told us to skip the root node. - */ - if (!memcg && (prev || (cond && !last_visited))) + if (prev && !memcg) goto out_unlock; } out_unlock: @@ -1804,14 +1765,13 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *memcg) * a) it is over its soft limit * b) any parent up the hierarchy is over its soft limit */ -enum mem_cgroup_filter_t -mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, +bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, struct mem_cgroup *root) { struct mem_cgroup *parent = memcg; if (res_counter_soft_limit_excess(&memcg->res)) - return VISIT; + return true; /* * If any parent up to the root in the hierarchy is over its soft limit @@ -1819,12 +1779,12 @@ mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, */ while ((parent = parent_mem_cgroup(parent))) { if (res_counter_soft_limit_excess(&parent->res)) - return VISIT; + return true; if (parent == root) break; } - return SKIP; + return false; } static DEFINE_SPINLOCK(memcg_oom_lock); diff --git a/mm/vmscan.c b/mm/vmscan.c index cdd300b..17a7134 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2185,16 +2185,21 @@ __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) .zone = zone, .priority = sc->priority, }; - struct mem_cgroup *memcg = NULL; - mem_cgroup_iter_filter filter = (soft_reclaim) ? - mem_cgroup_soft_reclaim_eligible : NULL; + struct mem_cgroup *memcg; nr_reclaimed = sc->nr_reclaimed; nr_scanned = sc->nr_scanned; - while ((memcg = mem_cgroup_iter_cond(root, memcg, &reclaim, filter))) { + memcg = mem_cgroup_iter(root, NULL, &reclaim); + do { struct lruvec *lruvec; + if (soft_reclaim && + !mem_cgroup_soft_reclaim_eligible(memcg, root)) { + memcg = mem_cgroup_iter(root, memcg, &reclaim); + continue; + } + lruvec = mem_cgroup_zone_lruvec(zone, memcg); shrink_lruvec(lruvec, sc); @@ -2214,7 +2219,8 @@ __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) mem_cgroup_iter_break(root, memcg); break; } - } + memcg = mem_cgroup_iter(root, memcg, &reclaim); + } while (memcg); vmpressure(sc->gfp_mask, sc->target_mem_cgroup, sc->nr_scanned - nr_scanned, -- cgit v0.10.2 From b1aff7fcf86c88472b0a70f15d89d7a4adba07bb Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:38 -0700 Subject: revert "vmscan, memcg: do softlimit reclaim also for targeted reclaim" Revert commit a5b7c87f9207 ("vmscan, memcg: do softlimit reclaim also for targeted reclaim") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index ef2b9bd..6054c9f 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -234,8 +234,7 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, mem_cgroup_update_page_stat(page, idx, -1); } -bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, - struct mem_cgroup *root); +bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg); void __mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx); static inline void mem_cgroup_count_vm_event(struct mm_struct *mm, @@ -435,8 +434,7 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, } static inline -bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, - struct mem_cgroup *root) +bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg) { return false; } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 65e7bec..47cdc7e 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1760,13 +1760,11 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *memcg) #endif /* - * A group is eligible for the soft limit reclaim under the given root - * hierarchy if - * a) it is over its soft limit + * A group is eligible for the soft limit reclaim if + * a) it is over its soft limit * b) any parent up the hierarchy is over its soft limit */ -bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, - struct mem_cgroup *root) +bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg) { struct mem_cgroup *parent = memcg; @@ -1774,14 +1772,12 @@ bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg, return true; /* - * If any parent up to the root in the hierarchy is over its soft limit - * then we have to obey and reclaim from this group as well. + * If any parent up the hierarchy is over its soft limit then we + * have to obey and reclaim from this group as well. */ while ((parent = parent_mem_cgroup(parent))) { if (res_counter_soft_limit_excess(&parent->res)) return true; - if (parent == root) - break; } return false; diff --git a/mm/vmscan.c b/mm/vmscan.c index 17a7134..0e081ca 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -142,7 +142,7 @@ static bool global_reclaim(struct scan_control *sc) static bool mem_cgroup_should_soft_reclaim(struct scan_control *sc) { - return !mem_cgroup_disabled(); + return !mem_cgroup_disabled() && global_reclaim(sc); } #else static bool global_reclaim(struct scan_control *sc) @@ -2195,7 +2195,7 @@ __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) struct lruvec *lruvec; if (soft_reclaim && - !mem_cgroup_soft_reclaim_eligible(memcg, root)) { + !mem_cgroup_soft_reclaim_eligible(memcg)) { memcg = mem_cgroup_iter(root, memcg, &reclaim); continue; } -- cgit v0.10.2 From bb4cc1a8b5eaf3b9e5707d7c270400b05d11a2b7 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:40 -0700 Subject: revert "memcg: get rid of soft-limit tree infrastructure" Revert commit e883110aad71 ("memcg: get rid of soft-limit tree infrastructure") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 47cdc7e..852dbec 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include #include @@ -124,6 +125,7 @@ static const char * const mem_cgroup_lru_names[] = { */ enum mem_cgroup_events_target { MEM_CGROUP_TARGET_THRESH, + MEM_CGROUP_TARGET_SOFTLIMIT, MEM_CGROUP_TARGET_NUMAINFO, MEM_CGROUP_NTARGETS, }; @@ -159,6 +161,10 @@ struct mem_cgroup_per_zone { struct mem_cgroup_reclaim_iter reclaim_iter[DEF_PRIORITY + 1]; + struct rb_node tree_node; /* RB tree node */ + unsigned long long usage_in_excess;/* Set to the value by which */ + /* the soft limit is exceeded*/ + bool on_tree; struct mem_cgroup *memcg; /* Back pointer, we cannot */ /* use container_of */ }; @@ -167,6 +173,26 @@ struct mem_cgroup_per_node { struct mem_cgroup_per_zone zoneinfo[MAX_NR_ZONES]; }; +/* + * Cgroups above their limits are maintained in a RB-Tree, independent of + * their hierarchy representation + */ + +struct mem_cgroup_tree_per_zone { + struct rb_root rb_root; + spinlock_t lock; +}; + +struct mem_cgroup_tree_per_node { + struct mem_cgroup_tree_per_zone rb_tree_per_zone[MAX_NR_ZONES]; +}; + +struct mem_cgroup_tree { + struct mem_cgroup_tree_per_node *rb_tree_per_node[MAX_NUMNODES]; +}; + +static struct mem_cgroup_tree soft_limit_tree __read_mostly; + struct mem_cgroup_threshold { struct eventfd_ctx *eventfd; u64 threshold; @@ -405,6 +431,7 @@ static bool move_file(void) * limit reclaim to prevent infinite loops, if they ever occur. */ #define MEM_CGROUP_MAX_RECLAIM_LOOPS 100 +#define MEM_CGROUP_MAX_SOFT_LIMIT_RECLAIM_LOOPS 2 enum charge_type { MEM_CGROUP_CHARGE_TYPE_CACHE = 0, @@ -631,6 +658,164 @@ page_cgroup_zoneinfo(struct mem_cgroup *memcg, struct page *page) return mem_cgroup_zoneinfo(memcg, nid, zid); } +static struct mem_cgroup_tree_per_zone * +soft_limit_tree_node_zone(int nid, int zid) +{ + return &soft_limit_tree.rb_tree_per_node[nid]->rb_tree_per_zone[zid]; +} + +static struct mem_cgroup_tree_per_zone * +soft_limit_tree_from_page(struct page *page) +{ + int nid = page_to_nid(page); + int zid = page_zonenum(page); + + return &soft_limit_tree.rb_tree_per_node[nid]->rb_tree_per_zone[zid]; +} + +static void +__mem_cgroup_insert_exceeded(struct mem_cgroup *memcg, + struct mem_cgroup_per_zone *mz, + struct mem_cgroup_tree_per_zone *mctz, + unsigned long long new_usage_in_excess) +{ + struct rb_node **p = &mctz->rb_root.rb_node; + struct rb_node *parent = NULL; + struct mem_cgroup_per_zone *mz_node; + + if (mz->on_tree) + return; + + mz->usage_in_excess = new_usage_in_excess; + if (!mz->usage_in_excess) + return; + while (*p) { + parent = *p; + mz_node = rb_entry(parent, struct mem_cgroup_per_zone, + tree_node); + if (mz->usage_in_excess < mz_node->usage_in_excess) + p = &(*p)->rb_left; + /* + * We can't avoid mem cgroups that are over their soft + * limit by the same amount + */ + else if (mz->usage_in_excess >= mz_node->usage_in_excess) + p = &(*p)->rb_right; + } + rb_link_node(&mz->tree_node, parent, p); + rb_insert_color(&mz->tree_node, &mctz->rb_root); + mz->on_tree = true; +} + +static void +__mem_cgroup_remove_exceeded(struct mem_cgroup *memcg, + struct mem_cgroup_per_zone *mz, + struct mem_cgroup_tree_per_zone *mctz) +{ + if (!mz->on_tree) + return; + rb_erase(&mz->tree_node, &mctz->rb_root); + mz->on_tree = false; +} + +static void +mem_cgroup_remove_exceeded(struct mem_cgroup *memcg, + struct mem_cgroup_per_zone *mz, + struct mem_cgroup_tree_per_zone *mctz) +{ + spin_lock(&mctz->lock); + __mem_cgroup_remove_exceeded(memcg, mz, mctz); + spin_unlock(&mctz->lock); +} + + +static void mem_cgroup_update_tree(struct mem_cgroup *memcg, struct page *page) +{ + unsigned long long excess; + struct mem_cgroup_per_zone *mz; + struct mem_cgroup_tree_per_zone *mctz; + int nid = page_to_nid(page); + int zid = page_zonenum(page); + mctz = soft_limit_tree_from_page(page); + + /* + * Necessary to update all ancestors when hierarchy is used. + * because their event counter is not touched. + */ + for (; memcg; memcg = parent_mem_cgroup(memcg)) { + mz = mem_cgroup_zoneinfo(memcg, nid, zid); + excess = res_counter_soft_limit_excess(&memcg->res); + /* + * We have to update the tree if mz is on RB-tree or + * mem is over its softlimit. + */ + if (excess || mz->on_tree) { + spin_lock(&mctz->lock); + /* if on-tree, remove it */ + if (mz->on_tree) + __mem_cgroup_remove_exceeded(memcg, mz, mctz); + /* + * Insert again. mz->usage_in_excess will be updated. + * If excess is 0, no tree ops. + */ + __mem_cgroup_insert_exceeded(memcg, mz, mctz, excess); + spin_unlock(&mctz->lock); + } + } +} + +static void mem_cgroup_remove_from_trees(struct mem_cgroup *memcg) +{ + int node, zone; + struct mem_cgroup_per_zone *mz; + struct mem_cgroup_tree_per_zone *mctz; + + for_each_node(node) { + for (zone = 0; zone < MAX_NR_ZONES; zone++) { + mz = mem_cgroup_zoneinfo(memcg, node, zone); + mctz = soft_limit_tree_node_zone(node, zone); + mem_cgroup_remove_exceeded(memcg, mz, mctz); + } + } +} + +static struct mem_cgroup_per_zone * +__mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) +{ + struct rb_node *rightmost = NULL; + struct mem_cgroup_per_zone *mz; + +retry: + mz = NULL; + rightmost = rb_last(&mctz->rb_root); + if (!rightmost) + goto done; /* Nothing to reclaim from */ + + mz = rb_entry(rightmost, struct mem_cgroup_per_zone, tree_node); + /* + * Remove the node now but someone else can add it back, + * we will to add it back at the end of reclaim to its correct + * position in the tree. + */ + __mem_cgroup_remove_exceeded(mz->memcg, mz, mctz); + if (!res_counter_soft_limit_excess(&mz->memcg->res) || + !css_tryget(&mz->memcg->css)) + goto retry; +done: + return mz; +} + +static struct mem_cgroup_per_zone * +mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) +{ + struct mem_cgroup_per_zone *mz; + + spin_lock(&mctz->lock); + mz = __mem_cgroup_largest_soft_limit_node(mctz); + spin_unlock(&mctz->lock); + return mz; +} + /* * Implementation Note: reading percpu statistics for memcg. * @@ -789,6 +974,9 @@ static bool mem_cgroup_event_ratelimit(struct mem_cgroup *memcg, case MEM_CGROUP_TARGET_THRESH: next = val + THRESHOLDS_EVENTS_TARGET; break; + case MEM_CGROUP_TARGET_SOFTLIMIT: + next = val + SOFTLIMIT_EVENTS_TARGET; + break; case MEM_CGROUP_TARGET_NUMAINFO: next = val + NUMAINFO_EVENTS_TARGET; break; @@ -811,8 +999,11 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page) /* threshold event is triggered in finer grain than soft limit */ if (unlikely(mem_cgroup_event_ratelimit(memcg, MEM_CGROUP_TARGET_THRESH))) { + bool do_softlimit; bool do_numainfo __maybe_unused; + do_softlimit = mem_cgroup_event_ratelimit(memcg, + MEM_CGROUP_TARGET_SOFTLIMIT); #if MAX_NUMNODES > 1 do_numainfo = mem_cgroup_event_ratelimit(memcg, MEM_CGROUP_TARGET_NUMAINFO); @@ -820,6 +1011,8 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page) preempt_enable(); mem_cgroup_threshold(memcg); + if (unlikely(do_softlimit)) + mem_cgroup_update_tree(memcg, page); #if MAX_NUMNODES > 1 if (unlikely(do_numainfo)) atomic_inc(&memcg->numainfo_events); @@ -1661,7 +1854,6 @@ static unsigned long mem_cgroup_reclaim(struct mem_cgroup *memcg, return total; } -#if MAX_NUMNODES > 1 /** * test_mem_cgroup_node_reclaimable * @memcg: the target memcg @@ -1684,6 +1876,7 @@ static bool test_mem_cgroup_node_reclaimable(struct mem_cgroup *memcg, return false; } +#if MAX_NUMNODES > 1 /* * Always updating the nodemask is not very good - even if we have an empty @@ -1751,12 +1944,51 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *memcg) return node; } +/* + * Check all nodes whether it contains reclaimable pages or not. + * For quick scan, we make use of scan_nodes. This will allow us to skip + * unused nodes. But scan_nodes is lazily updated and may not cotain + * enough new information. We need to do double check. + */ +static bool mem_cgroup_reclaimable(struct mem_cgroup *memcg, bool noswap) +{ + int nid; + + /* + * quick check...making use of scan_node. + * We can skip unused nodes. + */ + if (!nodes_empty(memcg->scan_nodes)) { + for (nid = first_node(memcg->scan_nodes); + nid < MAX_NUMNODES; + nid = next_node(nid, memcg->scan_nodes)) { + + if (test_mem_cgroup_node_reclaimable(memcg, nid, noswap)) + return true; + } + } + /* + * Check rest of nodes. + */ + for_each_node_state(nid, N_MEMORY) { + if (node_isset(nid, memcg->scan_nodes)) + continue; + if (test_mem_cgroup_node_reclaimable(memcg, nid, noswap)) + return true; + } + return false; +} + #else int mem_cgroup_select_victim_node(struct mem_cgroup *memcg) { return 0; } +static bool mem_cgroup_reclaimable(struct mem_cgroup *memcg, bool noswap) +{ + return test_mem_cgroup_node_reclaimable(memcg, 0, noswap); +} #endif /* @@ -2692,7 +2924,9 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *memcg, unlock_page_cgroup(pc); /* - * "charge_statistics" updated event counter. + * "charge_statistics" updated event counter. Then, check it. + * Insert ancestor (and ancestor's ancestors), to softlimit RB-tree. + * if they exceeds softlimit. */ memcg_check_events(memcg, page); } @@ -5791,6 +6025,8 @@ static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node) for (zone = 0; zone < MAX_NR_ZONES; zone++) { mz = &pn->zoneinfo[zone]; lruvec_init(&mz->lruvec); + mz->usage_in_excess = 0; + mz->on_tree = false; mz->memcg = memcg; } memcg->nodeinfo[node] = pn; @@ -5846,6 +6082,7 @@ static void __mem_cgroup_free(struct mem_cgroup *memcg) int node; size_t size = memcg_size(); + mem_cgroup_remove_from_trees(memcg); free_css_id(&mem_cgroup_subsys, &memcg->css); for_each_node(node) @@ -5882,6 +6119,29 @@ struct mem_cgroup *parent_mem_cgroup(struct mem_cgroup *memcg) } EXPORT_SYMBOL(parent_mem_cgroup); +static void __init mem_cgroup_soft_limit_tree_init(void) +{ + struct mem_cgroup_tree_per_node *rtpn; + struct mem_cgroup_tree_per_zone *rtpz; + int tmp, node, zone; + + for_each_node(node) { + tmp = node; + if (!node_state(node, N_NORMAL_MEMORY)) + tmp = -1; + rtpn = kzalloc_node(sizeof(*rtpn), GFP_KERNEL, tmp); + BUG_ON(!rtpn); + + soft_limit_tree.rb_tree_per_node[node] = rtpn; + + for (zone = 0; zone < MAX_NR_ZONES; zone++) { + rtpz = &rtpn->rb_tree_per_zone[zone]; + rtpz->rb_root = RB_ROOT; + spin_lock_init(&rtpz->lock); + } + } +} + static struct cgroup_subsys_state * __ref mem_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) { @@ -6662,6 +6922,7 @@ static int __init mem_cgroup_init(void) { hotcpu_notifier(memcg_cpu_hotplug_callback, 0); enable_swap_cgroup(); + mem_cgroup_soft_limit_tree_init(); memcg_stock_init(); return 0; } -- cgit v0.10.2 From 0608f43da64a1f1c42507304b5f25bc8b1227aa4 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Sep 2013 15:27:41 -0700 Subject: revert "memcg, vmscan: integrate soft reclaim tighter with zone shrinking code" Revert commit 3b38722efd9f ("memcg, vmscan: integrate soft reclaim tighter with zone shrinking code") I merged this prematurely - Michal and Johannes still disagree about the overall design direction and the future remains unclear. Cc: Michal Hocko Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 6054c9f..ecc82b3 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -234,7 +234,9 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, mem_cgroup_update_page_stat(page, idx, -1); } -bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg); +unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, + gfp_t gfp_mask, + unsigned long *total_scanned); void __mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx); static inline void mem_cgroup_count_vm_event(struct mm_struct *mm, @@ -434,9 +436,11 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, } static inline -bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg) +unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, + gfp_t gfp_mask, + unsigned long *total_scanned) { - return false; + return 0; } static inline void mem_cgroup_split_huge_fixup(struct page *head) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 852dbec..1c52ddb 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1991,28 +1991,57 @@ static bool mem_cgroup_reclaimable(struct mem_cgroup *memcg, bool noswap) } #endif -/* - * A group is eligible for the soft limit reclaim if - * a) it is over its soft limit - * b) any parent up the hierarchy is over its soft limit - */ -bool mem_cgroup_soft_reclaim_eligible(struct mem_cgroup *memcg) -{ - struct mem_cgroup *parent = memcg; - - if (res_counter_soft_limit_excess(&memcg->res)) - return true; - - /* - * If any parent up the hierarchy is over its soft limit then we - * have to obey and reclaim from this group as well. - */ - while ((parent = parent_mem_cgroup(parent))) { - if (res_counter_soft_limit_excess(&parent->res)) - return true; +static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg, + struct zone *zone, + gfp_t gfp_mask, + unsigned long *total_scanned) +{ + struct mem_cgroup *victim = NULL; + int total = 0; + int loop = 0; + unsigned long excess; + unsigned long nr_scanned; + struct mem_cgroup_reclaim_cookie reclaim = { + .zone = zone, + .priority = 0, + }; + + excess = res_counter_soft_limit_excess(&root_memcg->res) >> PAGE_SHIFT; + + while (1) { + victim = mem_cgroup_iter(root_memcg, victim, &reclaim); + if (!victim) { + loop++; + if (loop >= 2) { + /* + * If we have not been able to reclaim + * anything, it might because there are + * no reclaimable pages under this hierarchy + */ + if (!total) + break; + /* + * We want to do more targeted reclaim. + * excess >> 2 is not to excessive so as to + * reclaim too much, nor too less that we keep + * coming back to reclaim from this cgroup + */ + if (total >= (excess >> 2) || + (loop > MEM_CGROUP_MAX_RECLAIM_LOOPS)) + break; + } + continue; + } + if (!mem_cgroup_reclaimable(victim, false)) + continue; + total += mem_cgroup_shrink_node_zone(victim, gfp_mask, false, + zone, &nr_scanned); + *total_scanned += nr_scanned; + if (!res_counter_soft_limit_excess(&root_memcg->res)) + break; } - - return false; + mem_cgroup_iter_break(root_memcg, victim); + return total; } static DEFINE_SPINLOCK(memcg_oom_lock); @@ -4761,6 +4790,98 @@ static int mem_cgroup_resize_memsw_limit(struct mem_cgroup *memcg, return ret; } +unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, + gfp_t gfp_mask, + unsigned long *total_scanned) +{ + unsigned long nr_reclaimed = 0; + struct mem_cgroup_per_zone *mz, *next_mz = NULL; + unsigned long reclaimed; + int loop = 0; + struct mem_cgroup_tree_per_zone *mctz; + unsigned long long excess; + unsigned long nr_scanned; + + if (order > 0) + return 0; + + mctz = soft_limit_tree_node_zone(zone_to_nid(zone), zone_idx(zone)); + /* + * This loop can run a while, specially if mem_cgroup's continuously + * keep exceeding their soft limit and putting the system under + * pressure + */ + do { + if (next_mz) + mz = next_mz; + else + mz = mem_cgroup_largest_soft_limit_node(mctz); + if (!mz) + break; + + nr_scanned = 0; + reclaimed = mem_cgroup_soft_reclaim(mz->memcg, zone, + gfp_mask, &nr_scanned); + nr_reclaimed += reclaimed; + *total_scanned += nr_scanned; + spin_lock(&mctz->lock); + + /* + * If we failed to reclaim anything from this memory cgroup + * it is time to move on to the next cgroup + */ + next_mz = NULL; + if (!reclaimed) { + do { + /* + * Loop until we find yet another one. + * + * By the time we get the soft_limit lock + * again, someone might have aded the + * group back on the RB tree. Iterate to + * make sure we get a different mem. + * mem_cgroup_largest_soft_limit_node returns + * NULL if no other cgroup is present on + * the tree + */ + next_mz = + __mem_cgroup_largest_soft_limit_node(mctz); + if (next_mz == mz) + css_put(&next_mz->memcg->css); + else /* next_mz == NULL or other memcg */ + break; + } while (1); + } + __mem_cgroup_remove_exceeded(mz->memcg, mz, mctz); + excess = res_counter_soft_limit_excess(&mz->memcg->res); + /* + * One school of thought says that we should not add + * back the node to the tree if reclaim returns 0. + * But our reclaim could return 0, simply because due + * to priority we are exposing a smaller subset of + * memory to reclaim from. Consider this as a longer + * term TODO. + */ + /* If excess == 0, no tree ops */ + __mem_cgroup_insert_exceeded(mz->memcg, mz, mctz, excess); + spin_unlock(&mctz->lock); + css_put(&mz->memcg->css); + loop++; + /* + * Could not reclaim anything and there are no more + * mem cgroups to try or we seem to be looping without + * reclaiming anything. + */ + if (!nr_reclaimed && + (next_mz == NULL || + loop > MEM_CGROUP_MAX_SOFT_LIMIT_RECLAIM_LOOPS)) + break; + } while (!nr_reclaimed); + if (next_mz) + css_put(&next_mz->memcg->css); + return nr_reclaimed; +} + /** * mem_cgroup_force_empty_list - clears LRU of a group * @memcg: group to clear diff --git a/mm/vmscan.c b/mm/vmscan.c index 0e081ca..beb3577 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -139,21 +139,11 @@ static bool global_reclaim(struct scan_control *sc) { return !sc->target_mem_cgroup; } - -static bool mem_cgroup_should_soft_reclaim(struct scan_control *sc) -{ - return !mem_cgroup_disabled() && global_reclaim(sc); -} #else static bool global_reclaim(struct scan_control *sc) { return true; } - -static bool mem_cgroup_should_soft_reclaim(struct scan_control *sc) -{ - return false; -} #endif unsigned long zone_reclaimable_pages(struct zone *zone) @@ -2174,8 +2164,7 @@ static inline bool should_continue_reclaim(struct zone *zone, } } -static void -__shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) +static void shrink_zone(struct zone *zone, struct scan_control *sc) { unsigned long nr_reclaimed, nr_scanned; @@ -2194,12 +2183,6 @@ __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) do { struct lruvec *lruvec; - if (soft_reclaim && - !mem_cgroup_soft_reclaim_eligible(memcg)) { - memcg = mem_cgroup_iter(root, memcg, &reclaim); - continue; - } - lruvec = mem_cgroup_zone_lruvec(zone, memcg); shrink_lruvec(lruvec, sc); @@ -2230,24 +2213,6 @@ __shrink_zone(struct zone *zone, struct scan_control *sc, bool soft_reclaim) sc->nr_scanned - nr_scanned, sc)); } - -static void shrink_zone(struct zone *zone, struct scan_control *sc) -{ - bool do_soft_reclaim = mem_cgroup_should_soft_reclaim(sc); - unsigned long nr_scanned = sc->nr_scanned; - - __shrink_zone(zone, sc, do_soft_reclaim); - - /* - * No group is over the soft limit or those that are do not have - * pages in the zone we are reclaiming so we have to reclaim everybody - */ - if (do_soft_reclaim && (sc->nr_scanned == nr_scanned)) { - __shrink_zone(zone, sc, false); - return; - } -} - /* Returns true if compaction should go ahead for a high-order request */ static inline bool compaction_ready(struct zone *zone, struct scan_control *sc) { @@ -2309,6 +2274,8 @@ static bool shrink_zones(struct zonelist *zonelist, struct scan_control *sc) { struct zoneref *z; struct zone *zone; + unsigned long nr_soft_reclaimed; + unsigned long nr_soft_scanned; bool aborted_reclaim = false; /* @@ -2348,6 +2315,18 @@ static bool shrink_zones(struct zonelist *zonelist, struct scan_control *sc) continue; } } + /* + * This steals pages from memory cgroups over softlimit + * and returns the number of reclaimed pages and + * scanned pages. This works for global memory pressure + * and balancing, not for a memcg's limit. + */ + nr_soft_scanned = 0; + nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone, + sc->order, sc->gfp_mask, + &nr_soft_scanned); + sc->nr_reclaimed += nr_soft_reclaimed; + sc->nr_scanned += nr_soft_scanned; /* need some check for avoid more shrink_zone() */ } @@ -2941,6 +2920,8 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order, { int i; int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */ + unsigned long nr_soft_reclaimed; + unsigned long nr_soft_scanned; struct scan_control sc = { .gfp_mask = GFP_KERNEL, .priority = DEF_PRIORITY, @@ -3055,6 +3036,15 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order, sc.nr_scanned = 0; + nr_soft_scanned = 0; + /* + * Call soft limit reclaim before calling shrink_zone. + */ + nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone, + order, sc.gfp_mask, + &nr_soft_scanned); + sc.nr_reclaimed += nr_soft_reclaimed; + /* * There should be no need to raise the scanning * priority if enough pages are already being scanned -- cgit v0.10.2 From 8ac1c8d5deba65513b6a82c35e89e73996c8e0d6 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Tue, 24 Sep 2013 15:27:42 -0700 Subject: audit: fix endless wait in audit_log_start() After commit 829199197a43 ("kernel/audit.c: avoid negative sleep durations") audit emitters will block forever if userspace daemon cannot handle backlog. After the timeout the waiting loop turns into busy loop and runs until daemon dies or returns back to work. This is a minimal patch for that bug. Signed-off-by: Konstantin Khlebnikov Cc: Luiz Capitulino Cc: Richard Guy Briggs Cc: Eric Paris Cc: Chuck Anderson Cc: Dan Duval Cc: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/audit.c b/kernel/audit.c index 91e53d0..7b0e23a 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -1117,9 +1117,10 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, sleep_time = timeout_start + audit_backlog_wait_time - jiffies; - if ((long)sleep_time > 0) + if ((long)sleep_time > 0) { wait_for_auditd(sleep_time); - continue; + continue; + } } if (audit_rate_check() && printk_ratelimit()) printk(KERN_WARNING -- cgit v0.10.2 From e2f0b88e84eed9cf9797f0a88c8012ee0b885a6d Mon Sep 17 00:00:00 2001 From: Chuansheng Liu Date: Tue, 24 Sep 2013 15:27:43 -0700 Subject: kernel/reboot.c: re-enable the function of variable reboot_default Commit 1b3a5d02ee07 ("reboot: move arch/x86 reboot= handling to generic kernel") did some cleanup for reboot= command line, but it made the reboot_default inoperative. The default value of variable reboot_default should be 1, and if command line reboot= is not set, system will use the default reboot mode. [akpm@linux-foundation.org: fix comment layout] Signed-off-by: Li Fei Signed-off-by: liu chuansheng Acked-by: Robin Holt Cc: [3.11.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/reboot.c b/kernel/reboot.c index 269ed93..f813b34 100644 --- a/kernel/reboot.c +++ b/kernel/reboot.c @@ -32,7 +32,14 @@ EXPORT_SYMBOL(cad_pid); #endif enum reboot_mode reboot_mode DEFAULT_REBOOT_MODE; -int reboot_default; +/* + * This variable is used privately to keep track of whether or not + * reboot_type is still set to its default value (i.e., reboot= hasn't + * been set on the command line). This is needed so that we can + * suppress DMI scanning for reboot quirks. Without it, it's + * impossible to override a faulty reboot quirk without recompiling. + */ +int reboot_default = 1; int reboot_cpu; enum reboot_type reboot_type = BOOT_ACPI; int reboot_force; -- cgit v0.10.2 From 627aad1c01da6f881e7f98d71fd928ca0c316b1a Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 24 Sep 2013 15:27:44 -0700 Subject: cpqarray: fix info leak in ida_locked_ioctl() The pciinfo struct has a two byte hole after ->dev_fn so stack information could be leaked to the user. This was assigned CVE-2013-2147. Signed-off-by: Dan Carpenter Acked-by: Mike Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index 639d26b..2b94403 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -1193,6 +1193,7 @@ out_passthru: ida_pci_info_struct pciinfo; if (!arg) return -EINVAL; + memset(&pciinfo, 0, sizeof(pciinfo)); pciinfo.bus = host->pci_dev->bus->number; pciinfo.dev_fn = host->pci_dev->devfn; pciinfo.board_id = host->board_id; -- cgit v0.10.2 From 58f09e00ae095e46ef9edfcf3a5fd9ccdfad065e Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 24 Sep 2013 15:27:45 -0700 Subject: cciss: fix info leak in cciss_ioctl32_passthru() The arg64 struct has a hole after ->buf_size which isn't cleared. Or if any of the calls to copy_from_user() fail then that would cause an information leak as well. This was assigned CVE-2013-2147. Signed-off-by: Dan Carpenter Acked-by: Mike Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index d2d95ff..edfa251 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -1189,6 +1189,7 @@ static int cciss_ioctl32_passthru(struct block_device *bdev, fmode_t mode, int err; u32 cp; + memset(&arg64, 0, sizeof(arg64)); err = 0; err |= copy_from_user(&arg64.LUN_info, &arg32->LUN_info, -- cgit v0.10.2 From d1d85780dd30e137d8ff505c1c2e79eaf729853d Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 24 Sep 2013 15:27:46 -0700 Subject: checkpatch: make extern in .h prototypes quieter The use of extern in .h files is a bit contentious. Make the warning be emitted only when --strict is used on the command line. Signed-off-by: Joe Perches Cc: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 47016c3..66cad50 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -3975,8 +3975,8 @@ sub string_find_replace { # check for new externs in .h files. if ($realfile =~ /\.h$/ && $line =~ /^\+\s*(extern\s+)$Type\s*$Ident\s*\(/s) { - if (WARN("AVOID_EXTERNS", - "extern prototypes should be avoided in .h files\n" . $herecurr) && + if (CHK("AVOID_EXTERNS", + "extern prototypes should be avoided in .h files\n" . $herecurr) && $fix) { $fixed[$linenr - 1] =~ s/(.*)\bextern\b\s*(.*)/$1$2/; } -- cgit v0.10.2 From 497a045d13dcd7a00f5535ded1ebb49313d4a211 Mon Sep 17 00:00:00 2001 From: Christian Daudt Date: Tue, 24 Sep 2013 15:27:47 -0700 Subject: MAINTAINERS: update mach-bcm related email address Update email address on mach-bcm + drivers for Broadcom mobile SoCs. Signed-off-by: Christian Daudt Cc: Olof Johansson Cc: Arnd Bergmann Cc: Stephen Warren Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index e61c2e8..b2f857c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1812,7 +1812,8 @@ S: Supported F: drivers/net/ethernet/broadcom/bnx2x/ BROADCOM BCM281XX/BCM11XXX ARM ARCHITECTURE -M: Christian Daudt +M: Christian Daudt +L: bcm-kernel-feedback-list@broadcom.com T: git git://git.github.com/broadcom/bcm11351 S: Maintained F: arch/arm/mach-bcm/ -- cgit v0.10.2 From 06a8566bcf5cf7db9843a82cde7a33c7bf3947d9 Mon Sep 17 00:00:00 2001 From: Lv Zheng Date: Fri, 13 Sep 2013 13:13:23 +0800 Subject: ACPI / IPMI: Fix atomic context requirement of ipmi_msg_handler() This patch fixes the issues indicated by the test results that ipmi_msg_handler() is invoked in atomic context. BUG: scheduling while atomic: kipmi0/18933/0x10000100 Modules linked in: ipmi_si acpi_ipmi ... CPU: 3 PID: 18933 Comm: kipmi0 Tainted: G AW 3.10.0-rc7+ #2 Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS QSSC-S4R.QCI.01.00.0027.070120100606 07/01/2010 ffff8838245eea00 ffff88103fc63c98 ffffffff814c4a1e ffff88103fc63ca8 ffffffff814bfbab ffff88103fc63d28 ffffffff814c73e0 ffff88103933cbd4 0000000000000096 ffff88103fc63ce8 ffff88102f618000 ffff881035c01fd8 Call Trace: [] dump_stack+0x19/0x1b [] __schedule_bug+0x46/0x54 [] __schedule+0x83/0x59c [] __cond_resched+0x22/0x2d [] _cond_resched+0x14/0x1d [] mutex_lock+0x11/0x32 [] ? __default_send_IPI_dest_field.constprop.0+0x53/0x58 [] ipmi_msg_handler+0x23/0x166 [ipmi_si] [] deliver_response+0x55/0x5a [] handle_new_recv_msgs+0xb67/0xc65 [] ? read_tsc+0x9/0x19 [] ? _raw_spin_lock_irq+0xa/0xc [] ipmi_thread+0x5c/0x146 [ipmi_si] ... Also Tony Camuso says: We were getting occasional "Scheduling while atomic" call traces during boot on some systems. Problem was first seen on a Cisco C210 but we were able to reproduce it on a Cisco c220m3. Setting CONFIG_LOCKDEP and LOCKDEP_SUPPORT to 'y' exposed a lockdep around tx_msg_lock in acpi_ipmi.c struct acpi_ipmi_device. ================================= [ INFO: inconsistent lock state ] 2.6.32-415.el6.x86_64-debug-splck #1 --------------------------------- inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage. ksoftirqd/3/17 [HC0[0]:SC1[1]:HE1:SE0] takes: (&ipmi_device->tx_msg_lock){+.?...}, at: [] ipmi_msg_handler+0x71/0x126 {SOFTIRQ-ON-W} state was registered at: [] __lock_acquire+0x63c/0x1570 [] lock_acquire+0xa4/0x120 [] __mutex_lock_common+0x4c/0x400 [] mutex_lock_nested+0x4a/0x60 [] acpi_ipmi_space_handler+0x11b/0x234 [] acpi_ev_address_space_dispatch+0x170/0x1be The fix implemented by this change has been tested by Tony: Tested the patch in a boot loop with lockdep debug enabled and never saw the problem in over 400 reboots. Reported-and-tested-by: Tony Camuso Signed-off-by: Lv Zheng Reviewed-by: Huang Ying Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/acpi_ipmi.c b/drivers/acpi/acpi_ipmi.c index f40acef..a6977e1 100644 --- a/drivers/acpi/acpi_ipmi.c +++ b/drivers/acpi/acpi_ipmi.c @@ -39,6 +39,7 @@ #include #include #include +#include MODULE_AUTHOR("Zhao Yakui"); MODULE_DESCRIPTION("ACPI IPMI Opregion driver"); @@ -57,7 +58,7 @@ struct acpi_ipmi_device { struct list_head head; /* the IPMI request message list */ struct list_head tx_msg_list; - struct mutex tx_msg_lock; + spinlock_t tx_msg_lock; acpi_handle handle; struct pnp_dev *pnp_dev; ipmi_user_t user_interface; @@ -147,6 +148,7 @@ static void acpi_format_ipmi_msg(struct acpi_ipmi_msg *tx_msg, struct kernel_ipmi_msg *msg; struct acpi_ipmi_buffer *buffer; struct acpi_ipmi_device *device; + unsigned long flags; msg = &tx_msg->tx_message; /* @@ -177,10 +179,10 @@ static void acpi_format_ipmi_msg(struct acpi_ipmi_msg *tx_msg, /* Get the msgid */ device = tx_msg->device; - mutex_lock(&device->tx_msg_lock); + spin_lock_irqsave(&device->tx_msg_lock, flags); device->curr_msgid++; tx_msg->tx_msgid = device->curr_msgid; - mutex_unlock(&device->tx_msg_lock); + spin_unlock_irqrestore(&device->tx_msg_lock, flags); } static void acpi_format_ipmi_response(struct acpi_ipmi_msg *msg, @@ -242,6 +244,7 @@ static void ipmi_msg_handler(struct ipmi_recv_msg *msg, void *user_msg_data) int msg_found = 0; struct acpi_ipmi_msg *tx_msg; struct pnp_dev *pnp_dev = ipmi_device->pnp_dev; + unsigned long flags; if (msg->user != ipmi_device->user_interface) { dev_warn(&pnp_dev->dev, "Unexpected response is returned. " @@ -250,7 +253,7 @@ static void ipmi_msg_handler(struct ipmi_recv_msg *msg, void *user_msg_data) ipmi_free_recv_msg(msg); return; } - mutex_lock(&ipmi_device->tx_msg_lock); + spin_lock_irqsave(&ipmi_device->tx_msg_lock, flags); list_for_each_entry(tx_msg, &ipmi_device->tx_msg_list, head) { if (msg->msgid == tx_msg->tx_msgid) { msg_found = 1; @@ -258,7 +261,7 @@ static void ipmi_msg_handler(struct ipmi_recv_msg *msg, void *user_msg_data) } } - mutex_unlock(&ipmi_device->tx_msg_lock); + spin_unlock_irqrestore(&ipmi_device->tx_msg_lock, flags); if (!msg_found) { dev_warn(&pnp_dev->dev, "Unexpected response (msg id %ld) is " "returned.\n", msg->msgid); @@ -378,6 +381,7 @@ acpi_ipmi_space_handler(u32 function, acpi_physical_address address, struct acpi_ipmi_device *ipmi_device = handler_context; int err, rem_time; acpi_status status; + unsigned long flags; /* * IPMI opregion message. * IPMI message is firstly written to the BMC and system software @@ -395,9 +399,9 @@ acpi_ipmi_space_handler(u32 function, acpi_physical_address address, return AE_NO_MEMORY; acpi_format_ipmi_msg(tx_msg, address, value); - mutex_lock(&ipmi_device->tx_msg_lock); + spin_lock_irqsave(&ipmi_device->tx_msg_lock, flags); list_add_tail(&tx_msg->head, &ipmi_device->tx_msg_list); - mutex_unlock(&ipmi_device->tx_msg_lock); + spin_unlock_irqrestore(&ipmi_device->tx_msg_lock, flags); err = ipmi_request_settime(ipmi_device->user_interface, &tx_msg->addr, tx_msg->tx_msgid, @@ -413,9 +417,9 @@ acpi_ipmi_space_handler(u32 function, acpi_physical_address address, status = AE_OK; end_label: - mutex_lock(&ipmi_device->tx_msg_lock); + spin_lock_irqsave(&ipmi_device->tx_msg_lock, flags); list_del(&tx_msg->head); - mutex_unlock(&ipmi_device->tx_msg_lock); + spin_unlock_irqrestore(&ipmi_device->tx_msg_lock, flags); kfree(tx_msg); return status; } @@ -457,7 +461,7 @@ static void acpi_add_ipmi_device(struct acpi_ipmi_device *ipmi_device) INIT_LIST_HEAD(&ipmi_device->head); - mutex_init(&ipmi_device->tx_msg_lock); + spin_lock_init(&ipmi_device->tx_msg_lock); INIT_LIST_HEAD(&ipmi_device->tx_msg_list); ipmi_install_space_handler(ipmi_device); -- cgit v0.10.2 From 8a61e12e84597b5f8155ac91b44dea866ccfaac2 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 20 Sep 2013 10:43:56 -0700 Subject: acpi-cpufreq: skip loading acpi_cpufreq after intel_pstate If the hw supports intel_pstate and acpi_cpufreq, intel_pstate will get loaded first. acpi_cpufreq_init() will call acpi_cpufreq_early_init() and that will allocate perf data and init those perf data in ACPI core, (that will cover all CPUs). But later it will free them as cpufreq_register_driver(acpi_cpufreq) will fail as intel_pstate is already registered Use cpufreq_get_current_driver() to check if we can skip the acpi_cpufreq loading. Signed-off-by: Yinghai Lu Signed-off-by: Rafael J. Wysocki diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index a1260b4..d2c3253 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -986,6 +986,10 @@ static int __init acpi_cpufreq_init(void) { int ret; + /* don't keep reloading if cpufreq_driver exists */ + if (cpufreq_get_current_driver()) + return 0; + if (acpi_disabled) return 0; -- cgit v0.10.2 From 26ca8694344af4c833d22590c5b77d6b9cff0722 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Fri, 20 Sep 2013 22:37:31 +0530 Subject: cpufreq: check cpufreq driver is valid and cpufreq isn't disabled in cpufreq_get() cpufreq_get() can be called from external drivers which might not be aware if cpufreq driver is registered or not. And so we should actually check if cpufreq driver is registered or not and also if cpufreq is active or disabled, at the beginning of cpufreq_get(). Otherwise call to lock_policy_rwsem_read() might hit BUG_ON(!policy). Reported-and-tested-by: Linus Walleij Signed-off-by: Viresh Kumar Reviewed-by: Srivatsa S. Bhat Signed-off-by: Rafael J. Wysocki diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 89b3c52..04548f7 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1460,6 +1460,9 @@ unsigned int cpufreq_get(unsigned int cpu) { unsigned int ret_freq = 0; + if (cpufreq_disabled() || !cpufreq_driver) + return -ENOENT; + if (!down_read_trylock(&cpufreq_rwsem)) return 0; -- cgit v0.10.2 From 116decb7e48b46a8c171bd47f377fa527067f788 Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Wed, 18 Sep 2013 10:44:53 +0530 Subject: cpufreq: exynos5440: Fix potential NULL pointer dereference If 'dvfs_info' is NULL (due to devm_kzalloc failure) the failure error message would try to dereference it. Use 'pdev' instead. Signed-off-by: Sachin Kamat Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki diff --git a/drivers/cpufreq/exynos5440-cpufreq.c b/drivers/cpufreq/exynos5440-cpufreq.c index d514c15..be5380e 100644 --- a/drivers/cpufreq/exynos5440-cpufreq.c +++ b/drivers/cpufreq/exynos5440-cpufreq.c @@ -457,7 +457,7 @@ err_free_table: opp_free_cpufreq_table(dvfs_info->dev, &dvfs_info->freq_table); err_put_node: of_node_put(np); - dev_err(dvfs_info->dev, "%s: failed initialization\n", __func__); + dev_err(&pdev->dev, "%s: failed initialization\n", __func__); return ret; } -- cgit v0.10.2 From ede2033c405bfbd6b74b113cf69c6f54cef35902 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Fri, 6 Sep 2013 16:49:18 -0500 Subject: openrisc: clean-up prom.h Clean-up some copy/paste declarations that are not necessary. All the functions either don't exist or are already declared in other headers. This is needed in preparation of of_irq.h clean-up. Signed-off-by: Rob Herring Cc: Jonas Bonn Cc: linux@lists.openrisc.net diff --git a/arch/openrisc/include/asm/prom.h b/arch/openrisc/include/asm/prom.h index eb59bfe..93c9980 100644 --- a/arch/openrisc/include/asm/prom.h +++ b/arch/openrisc/include/asm/prom.h @@ -14,53 +14,9 @@ * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. */ - -#include /* linux/of.h gets to determine #include ordering */ - #ifndef _ASM_OPENRISC_PROM_H #define _ASM_OPENRISC_PROM_H -#ifdef __KERNEL__ -#ifndef __ASSEMBLY__ -#include -#include -#include -#include -#include -#include -#include -#include -#include #define HAVE_ARCH_DEVTREE_FIXUPS -/* Other Prototypes */ -extern int early_uartlite_console(void); - -/* Parse the ibm,dma-window property of an OF node into the busno, phys and - * size parameters. - */ -void of_parse_dma_window(struct device_node *dn, const void *dma_window_prop, - unsigned long *busno, unsigned long *phys, unsigned long *size); - -extern void kdump_move_device_tree(void); - -/* Get the MAC address */ -extern const void *of_get_mac_address(struct device_node *np); - -/** - * of_irq_map_pci - Resolve the interrupt for a PCI device - * @pdev: the device whose interrupt is to be resolved - * @out_irq: structure of_irq filled by this function - * - * This function resolves the PCI interrupt for a given PCI device. If a - * device-node exists for a given pci_dev, it will use normal OF tree - * walking. If not, it will implement standard swizzling and walk up the - * PCI tree until an device-node is found, at which point it will finish - * resolving using the OF tree walking. - */ -struct pci_dev; -extern int of_irq_map_pci(struct pci_dev *pdev, struct of_irq *out_irq); - -#endif /* __ASSEMBLY__ */ -#endif /* __KERNEL__ */ #endif /* _ASM_OPENRISC_PROM_H */ -- cgit v0.10.2 From b0b8c960ffcc5bfc82d1a09ad931673e3a36c15a Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Wed, 4 Sep 2013 10:52:57 -0500 Subject: of: clean-up ifdefs in of_irq.h Much of of_irq.h is needlessly ifdef'ed. Clean this up and minimize the amount ifdef'ed code. This fixes some build warnings when CONFIG_OF is not enabled (seen on i386 and x86_64): include/linux/of_irq.h:82:7: warning: 'struct device_node' declared inside parameter list [enabled by default] include/linux/of_irq.h:82:7: warning: its scope is only this definition or declaration, which is probably not what you want [enabled by default] include/linux/of_irq.h:87:47: warning: 'struct device_node' declared inside parameter list [enabled by default] Compile tested on i386, sparc and arm. Reported-by: Randy Dunlap Cc: Grant Likely Signed-off-by: Rob Herring diff --git a/include/linux/of_irq.h b/include/linux/of_irq.h index 535cecf..fcd63ba 100644 --- a/include/linux/of_irq.h +++ b/include/linux/of_irq.h @@ -1,8 +1,6 @@ #ifndef __OF_IRQ_H #define __OF_IRQ_H -#if defined(CONFIG_OF) -struct of_irq; #include #include #include @@ -10,14 +8,6 @@ struct of_irq; #include #include -/* - * irq_of_parse_and_map() is used by all OF enabled platforms; but SPARC - * implements it differently. However, the prototype is the same for all, - * so declare it here regardless of the CONFIG_OF_IRQ setting. - */ -extern unsigned int irq_of_parse_and_map(struct device_node *node, int index); - -#if defined(CONFIG_OF_IRQ) /** * of_irq - container for device_node/irq_specifier pair for an irq controller * @controller: pointer to interrupt controller device tree node @@ -71,11 +61,17 @@ extern int of_irq_to_resource(struct device_node *dev, int index, extern int of_irq_count(struct device_node *dev); extern int of_irq_to_resource_table(struct device_node *dev, struct resource *res, int nr_irqs); -extern struct device_node *of_irq_find_parent(struct device_node *child); extern void of_irq_init(const struct of_device_id *matches); -#endif /* CONFIG_OF_IRQ */ +#if defined(CONFIG_OF) +/* + * irq_of_parse_and_map() is used by all OF enabled platforms; but SPARC + * implements it differently. However, the prototype is the same for all, + * so declare it here regardless of the CONFIG_OF_IRQ setting. + */ +extern unsigned int irq_of_parse_and_map(struct device_node *node, int index); +extern struct device_node *of_irq_find_parent(struct device_node *child); #else /* !CONFIG_OF */ static inline unsigned int irq_of_parse_and_map(struct device_node *dev, -- cgit v0.10.2 From 53126a2f679382f3eac370a012290f19c672f088 Mon Sep 17 00:00:00 2001 From: Dinh Nguyen Date: Mon, 16 Sep 2013 15:57:48 -0500 Subject: dts: Fix misspelling of Synopsys s/Synopsis/Synopsys s/synopsis/synopsys Signed-off-by: Dinh Nguyen Cc: Pavel Machek CC: Arnd Bergmann CC: Olof Johansson Cc: Rob Herring Cc: Pawel Moll Cc: Mark Rutland Cc: Stephen Warren Cc: Ian Campbell Cc: Chris Ball Cc: Jaehoon Chung Cc: Seungwon Jeon Cc: Tomasz Figa Cc: devicetree@vger.kernel.org Cc: linux-mmc@vger.kernel.org CC: linux-arm-kernel@lists.infradead.org Acked-by: Olof Johansson Signed-off-by: Rob Herring diff --git a/Documentation/devicetree/bindings/mmc/exynos-dw-mshc.txt b/Documentation/devicetree/bindings/mmc/exynos-dw-mshc.txt index 6d1c098..c67b975 100644 --- a/Documentation/devicetree/bindings/mmc/exynos-dw-mshc.txt +++ b/Documentation/devicetree/bindings/mmc/exynos-dw-mshc.txt @@ -1,11 +1,11 @@ -* Samsung Exynos specific extensions to the Synopsis Designware Mobile +* Samsung Exynos specific extensions to the Synopsys Designware Mobile Storage Host Controller -The Synopsis designware mobile storage host controller is used to interface +The Synopsys designware mobile storage host controller is used to interface a SoC with storage medium such as eMMC or SD/MMC cards. This file documents -differences between the core Synopsis dw mshc controller properties described -by synopsis-dw-mshc.txt and the properties used by the Samsung Exynos specific -extensions to the Synopsis Designware Mobile Storage Host Controller. +differences between the core Synopsys dw mshc controller properties described +by synopsys-dw-mshc.txt and the properties used by the Samsung Exynos specific +extensions to the Synopsys Designware Mobile Storage Host Controller. Required Properties: diff --git a/Documentation/devicetree/bindings/mmc/rockchip-dw-mshc.txt b/Documentation/devicetree/bindings/mmc/rockchip-dw-mshc.txt index 8a3d91d..c559f3f 100644 --- a/Documentation/devicetree/bindings/mmc/rockchip-dw-mshc.txt +++ b/Documentation/devicetree/bindings/mmc/rockchip-dw-mshc.txt @@ -1,11 +1,11 @@ -* Rockchip specific extensions to the Synopsis Designware Mobile +* Rockchip specific extensions to the Synopsys Designware Mobile Storage Host Controller -The Synopsis designware mobile storage host controller is used to interface +The Synopsys designware mobile storage host controller is used to interface a SoC with storage medium such as eMMC or SD/MMC cards. This file documents -differences between the core Synopsis dw mshc controller properties described -by synopsis-dw-mshc.txt and the properties used by the Rockchip specific -extensions to the Synopsis Designware Mobile Storage Host Controller. +differences between the core Synopsys dw mshc controller properties described +by synopsys-dw-mshc.txt and the properties used by the Rockchip specific +extensions to the Synopsys Designware Mobile Storage Host Controller. Required Properties: diff --git a/Documentation/devicetree/bindings/mmc/synopsis-dw-mshc.txt b/Documentation/devicetree/bindings/mmc/synopsis-dw-mshc.txt deleted file mode 100644 index cdcebea..0000000 --- a/Documentation/devicetree/bindings/mmc/synopsis-dw-mshc.txt +++ /dev/null @@ -1,107 +0,0 @@ -* Synopsis Designware Mobile Storage Host Controller - -The Synopsis designware mobile storage host controller is used to interface -a SoC with storage medium such as eMMC or SD/MMC cards. This file documents -differences between the core mmc properties described by mmc.txt and the -properties used by the Synopsis Designware Mobile Storage Host Controller. - -Required Properties: - -* compatible: should be - - snps,dw-mshc: for controllers compliant with synopsis dw-mshc. -* #address-cells: should be 1. -* #size-cells: should be 0. - -# Slots: The slot specific information are contained within child-nodes with - each child-node representing a supported slot. There should be atleast one - child node representing a card slot. The name of the child node representing - the slot is recommended to be slot@n where n is the unique number of the slot - connnected to the controller. The following are optional properties which - can be included in the slot child node. - - * reg: specifies the physical slot number. The valid values of this - property is 0 to (num-slots -1), where num-slots is the value - specified by the num-slots property. - - * bus-width: as documented in mmc core bindings. - - * wp-gpios: specifies the write protect gpio line. The format of the - gpio specifier depends on the gpio controller. If a GPIO is not used - for write-protect, this property is optional. - - * disable-wp: If the wp-gpios property isn't present then (by default) - we'd assume that the write protect is hooked up directly to the - controller's special purpose write protect line (accessible via - the WRTPRT register). However, it's possible that we simply don't - want write protect. In that case specify 'disable-wp'. - NOTE: This property is not required for slots known to always - connect to eMMC or SDIO cards. - -Optional properties: - -* clocks: from common clock binding: handle to biu and ciu clocks for the - bus interface unit clock and the card interface unit clock. - -* clock-names: from common clock binding: Shall be "biu" and "ciu". - If the biu clock is missing we'll simply skip enabling it. If the - ciu clock is missing we'll just assume that the clock is running at - clock-frequency. It is an error to omit both the ciu clock and the - clock-frequency. - -* clock-frequency: should be the frequency (in Hz) of the ciu clock. If this - is specified and the ciu clock is specified then we'll try to set the ciu - clock to this at probe time. - -* num-slots: specifies the number of slots supported by the controller. - The number of physical slots actually used could be equal or less than the - value specified by num-slots. If this property is not specified, the value - of num-slot property is assumed to be 1. - -* fifo-depth: The maximum size of the tx/rx fifo's. If this property is not - specified, the default value of the fifo size is determined from the - controller registers. - -* card-detect-delay: Delay in milli-seconds before detecting card after card - insert event. The default value is 0. - -* supports-highspeed: Enables support for high speed cards (up to 50MHz) - -* broken-cd: as documented in mmc core bindings. - -* vmmc-supply: The phandle to the regulator to use for vmmc. If this is - specified we'll defer probe until we can find this regulator. - -Aliases: - -- All the MSHC controller nodes should be represented in the aliases node using - the following format 'mshc{n}' where n is a unique number for the alias. - -Example: - -The MSHC controller node can be split into two portions, SoC specific and -board specific portions as listed below. - - dwmmc0@12200000 { - compatible = "snps,dw-mshc"; - clocks = <&clock 351>, <&clock 132>; - clock-names = "biu", "ciu"; - reg = <0x12200000 0x1000>; - interrupts = <0 75 0>; - #address-cells = <1>; - #size-cells = <0>; - }; - - dwmmc0@12200000 { - clock-frequency = <400000000>; - num-slots = <1>; - supports-highspeed; - broken-cd; - fifo-depth = <0x80>; - card-detect-delay = <200>; - vmmc-supply = <&buck8>; - - slot@0 { - reg = <0>; - bus-width = <8>; - }; - }; diff --git a/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt b/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt new file mode 100644 index 0000000..066a78b --- /dev/null +++ b/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt @@ -0,0 +1,107 @@ +* Synopsys Designware Mobile Storage Host Controller + +The Synopsys designware mobile storage host controller is used to interface +a SoC with storage medium such as eMMC or SD/MMC cards. This file documents +differences between the core mmc properties described by mmc.txt and the +properties used by the Synopsys Designware Mobile Storage Host Controller. + +Required Properties: + +* compatible: should be + - snps,dw-mshc: for controllers compliant with synopsys dw-mshc. +* #address-cells: should be 1. +* #size-cells: should be 0. + +# Slots: The slot specific information are contained within child-nodes with + each child-node representing a supported slot. There should be atleast one + child node representing a card slot. The name of the child node representing + the slot is recommended to be slot@n where n is the unique number of the slot + connnected to the controller. The following are optional properties which + can be included in the slot child node. + + * reg: specifies the physical slot number. The valid values of this + property is 0 to (num-slots -1), where num-slots is the value + specified by the num-slots property. + + * bus-width: as documented in mmc core bindings. + + * wp-gpios: specifies the write protect gpio line. The format of the + gpio specifier depends on the gpio controller. If a GPIO is not used + for write-protect, this property is optional. + + * disable-wp: If the wp-gpios property isn't present then (by default) + we'd assume that the write protect is hooked up directly to the + controller's special purpose write protect line (accessible via + the WRTPRT register). However, it's possible that we simply don't + want write protect. In that case specify 'disable-wp'. + NOTE: This property is not required for slots known to always + connect to eMMC or SDIO cards. + +Optional properties: + +* clocks: from common clock binding: handle to biu and ciu clocks for the + bus interface unit clock and the card interface unit clock. + +* clock-names: from common clock binding: Shall be "biu" and "ciu". + If the biu clock is missing we'll simply skip enabling it. If the + ciu clock is missing we'll just assume that the clock is running at + clock-frequency. It is an error to omit both the ciu clock and the + clock-frequency. + +* clock-frequency: should be the frequency (in Hz) of the ciu clock. If this + is specified and the ciu clock is specified then we'll try to set the ciu + clock to this at probe time. + +* num-slots: specifies the number of slots supported by the controller. + The number of physical slots actually used could be equal or less than the + value specified by num-slots. If this property is not specified, the value + of num-slot property is assumed to be 1. + +* fifo-depth: The maximum size of the tx/rx fifo's. If this property is not + specified, the default value of the fifo size is determined from the + controller registers. + +* card-detect-delay: Delay in milli-seconds before detecting card after card + insert event. The default value is 0. + +* supports-highspeed: Enables support for high speed cards (up to 50MHz) + +* broken-cd: as documented in mmc core bindings. + +* vmmc-supply: The phandle to the regulator to use for vmmc. If this is + specified we'll defer probe until we can find this regulator. + +Aliases: + +- All the MSHC controller nodes should be represented in the aliases node using + the following format 'mshc{n}' where n is a unique number for the alias. + +Example: + +The MSHC controller node can be split into two portions, SoC specific and +board specific portions as listed below. + + dwmmc0@12200000 { + compatible = "snps,dw-mshc"; + clocks = <&clock 351>, <&clock 132>; + clock-names = "biu", "ciu"; + reg = <0x12200000 0x1000>; + interrupts = <0 75 0>; + #address-cells = <1>; + #size-cells = <0>; + }; + + dwmmc0@12200000 { + clock-frequency = <400000000>; + num-slots = <1>; + supports-highspeed; + broken-cd; + fifo-depth = <0x80>; + card-detect-delay = <200>; + vmmc-supply = <&buck8>; + + slot@0 { + reg = <0>; + bus-width = <8>; + }; + }; diff --git a/Documentation/devicetree/bindings/pci/designware-pcie.txt b/Documentation/devicetree/bindings/pci/designware-pcie.txt index eabcb4b..e216af3 100644 --- a/Documentation/devicetree/bindings/pci/designware-pcie.txt +++ b/Documentation/devicetree/bindings/pci/designware-pcie.txt @@ -1,4 +1,4 @@ -* Synopsis Designware PCIe interface +* Synopsys Designware PCIe interface Required properties: - compatible: should contain "snps,dw-pcie" to identify the -- cgit v0.10.2 From 22356f447ceb8d97a4885792e7d9e4607f712e1b Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 24 Sep 2013 18:29:11 -0700 Subject: mm: Place preemption point in do_mlockall() loop There is a loop in do_mlockall() that lacks a preemption point, which means that the following can happen on non-preemptible builds of the kernel. Dave Jones reports: "My fuzz tester keeps hitting this. Every instance shows the non-irq stack came in from mlockall. I'm only seeing this on one box, but that has more ram (8gb) than my other machines, which might explain it. INFO: rcu_preempt self-detected stall on CPU { 3} (t=6500 jiffies g=470344 c=470343 q=0) sending NMI to all CPUs: NMI backtrace for cpu 3 CPU: 3 PID: 29664 Comm: trinity-child2 Not tainted 3.11.0-rc1+ #32 Call Trace: lru_add_drain_all+0x15/0x20 SyS_mlockall+0xa5/0x1a0 tracesys+0xdd/0xe2" This commit addresses this problem by inserting the required preemption point. Reported-by: Dave Jones Signed-off-by: Paul E. McKenney Cc: KOSAKI Motohiro Cc: Michel Lespinasse Cc: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/mlock.c b/mm/mlock.c index d638026..67ba6da 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -736,6 +736,7 @@ static int do_mlockall(int flags) /* Ignore errors */ mlock_fixup(vma, &prev, vma->vm_start, vma->vm_end, newflags); + cond_resched(); } out: return 0; -- cgit v0.10.2 From 0366a1c70b89efed4f9d590216bb004a16effbed Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 23 Sep 2013 14:29:11 +1000 Subject: powerpc/irq: Run softirqs off the top of the irq stack Nowadays, irq_exit() calls __do_softirq() pretty much directly instead of calling do_softirq() which switches to the decicated softirq stack. This has lead to observed stack overflows on powerpc since we call irq_enter() and irq_exit() outside of the scope that switches to the irq stack. This fixes it by moving the stack switching up a level, making irq_enter() and irq_exit() run off the irq stack. Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/include/asm/irq.h b/arch/powerpc/include/asm/irq.h index 0e40843..41f13ce 100644 --- a/arch/powerpc/include/asm/irq.h +++ b/arch/powerpc/include/asm/irq.h @@ -69,9 +69,9 @@ extern struct thread_info *softirq_ctx[NR_CPUS]; extern void irq_ctx_init(void); extern void call_do_softirq(struct thread_info *tp); -extern int call_handle_irq(int irq, void *p1, - struct thread_info *tp, void *func); +extern void call_do_irq(struct pt_regs *regs, struct thread_info *tp); extern void do_IRQ(struct pt_regs *regs); +extern void __do_irq(struct pt_regs *regs); int irq_choose_cpu(const struct cpumask *mask); diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index c69440c..2234a12 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -441,50 +441,6 @@ void migrate_irqs(void) } #endif -static inline void handle_one_irq(unsigned int irq) -{ - struct thread_info *curtp, *irqtp; - unsigned long saved_sp_limit; - struct irq_desc *desc; - - desc = irq_to_desc(irq); - if (!desc) - return; - - /* Switch to the irq stack to handle this */ - curtp = current_thread_info(); - irqtp = hardirq_ctx[smp_processor_id()]; - - if (curtp == irqtp) { - /* We're already on the irq stack, just handle it */ - desc->handle_irq(irq, desc); - return; - } - - saved_sp_limit = current->thread.ksp_limit; - - irqtp->task = curtp->task; - irqtp->flags = 0; - - /* Copy the softirq bits in preempt_count so that the - * softirq checks work in the hardirq context. */ - irqtp->preempt_count = (irqtp->preempt_count & ~SOFTIRQ_MASK) | - (curtp->preempt_count & SOFTIRQ_MASK); - - current->thread.ksp_limit = (unsigned long)irqtp + - _ALIGN_UP(sizeof(struct thread_info), 16); - - call_handle_irq(irq, desc, irqtp, desc->handle_irq); - current->thread.ksp_limit = saved_sp_limit; - irqtp->task = NULL; - - /* Set any flag that may have been set on the - * alternate stack - */ - if (irqtp->flags) - set_bits(irqtp->flags, &curtp->flags); -} - static inline void check_stack_overflow(void) { #ifdef CONFIG_DEBUG_STACKOVERFLOW @@ -501,9 +457,9 @@ static inline void check_stack_overflow(void) #endif } -void do_IRQ(struct pt_regs *regs) +void __do_irq(struct pt_regs *regs) { - struct pt_regs *old_regs = set_irq_regs(regs); + struct irq_desc *desc; unsigned int irq; irq_enter(); @@ -519,18 +475,64 @@ void do_IRQ(struct pt_regs *regs) */ irq = ppc_md.get_irq(); - /* We can hard enable interrupts now */ + /* We can hard enable interrupts now to allow perf interrupts */ may_hard_irq_enable(); /* And finally process it */ - if (irq != NO_IRQ) - handle_one_irq(irq); - else + if (unlikely(irq == NO_IRQ)) __get_cpu_var(irq_stat).spurious_irqs++; + else { + desc = irq_to_desc(irq); + if (likely(desc)) + desc->handle_irq(irq, desc); + } trace_irq_exit(regs); irq_exit(); +} + +void do_IRQ(struct pt_regs *regs) +{ + struct pt_regs *old_regs = set_irq_regs(regs); + struct thread_info *curtp, *irqtp; + unsigned long saved_sp_limit; + + /* Switch to the irq stack to handle this */ + curtp = current_thread_info(); + irqtp = hardirq_ctx[raw_smp_processor_id()]; + + /* Already there ? */ + if (unlikely(curtp == irqtp)) { + __do_irq(regs); + set_irq_regs(old_regs); + return; + } + + /* Adjust the stack limit */ + saved_sp_limit = current->thread.ksp_limit; + current->thread.ksp_limit = (unsigned long)irqtp + + _ALIGN_UP(sizeof(struct thread_info), 16); + + + /* Prepare the thread_info in the irq stack */ + irqtp->task = curtp->task; + irqtp->flags = 0; + + /* Copy the preempt_count so that the [soft]irq checks work. */ + irqtp->preempt_count = curtp->preempt_count; + + /* Switch stack and call */ + call_do_irq(regs, irqtp); + + /* Restore stack limit */ + current->thread.ksp_limit = saved_sp_limit; + irqtp->task = NULL; + + /* Copy back updates to the thread_info */ + if (irqtp->flags) + set_bits(irqtp->flags, &curtp->flags); + set_irq_regs(old_regs); } @@ -592,12 +594,10 @@ void irq_ctx_init(void) memset((void *)softirq_ctx[i], 0, THREAD_SIZE); tp = softirq_ctx[i]; tp->cpu = i; - tp->preempt_count = 0; memset((void *)hardirq_ctx[i], 0, THREAD_SIZE); tp = hardirq_ctx[i]; tp->cpu = i; - tp->preempt_count = HARDIRQ_OFFSET; } } diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S index 777d999..7da3882 100644 --- a/arch/powerpc/kernel/misc_32.S +++ b/arch/powerpc/kernel/misc_32.S @@ -47,13 +47,12 @@ _GLOBAL(call_do_softirq) mtlr r0 blr -_GLOBAL(call_handle_irq) +_GLOBAL(call_do_irq) mflr r0 stw r0,4(r1) - mtctr r6 - stwu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r5) - mr r1,r5 - bctrl + stwu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r4) + mr r1,r4 + bl __do_irq lwz r1,0(r1) lwz r0,4(r1) mtlr r0 diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S index 971d7e7..e59caf8 100644 --- a/arch/powerpc/kernel/misc_64.S +++ b/arch/powerpc/kernel/misc_64.S @@ -40,14 +40,12 @@ _GLOBAL(call_do_softirq) mtlr r0 blr -_GLOBAL(call_handle_irq) - ld r8,0(r6) +_GLOBAL(call_do_irq) mflr r0 std r0,16(r1) - mtctr r8 - stdu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r5) - mr r1,r5 - bctrl + stdu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r4) + mr r1,r4 + bl .__do_irq ld r1,0(r1) ld r0,16(r1) mtlr r0 -- cgit v0.10.2 From cbc9565ee82694dec31d8137dec975b83175183b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 24 Sep 2013 15:17:21 +1000 Subject: powerpc: Remove ksp_limit on ppc64 We've been keeping that field in thread_struct for a while, it contains the "limit" of the current stack pointer and is meant to be used for detecting stack overflows. It has a few problems however: - First, it was never actually *used* on 64-bit. Set and updated but not actually exploited - When switching stack to/from irq and softirq stacks, it's update is racy unless we hard disable interrupts, which is costly. This is fine on 32-bit as we don't soft-disable there but not on 64-bit. Thus rather than fixing 2 in order to implement 1 in some hypothetical future, let's remove the code completely from 64-bit. In order to avoid a clutter of ifdef's, we remove the updates from C code completely during interrupt stack switching, and instead maintain it from the asm helper that is used to do the stack switching in the first place. Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h index e378ccc..ce4de5a 100644 --- a/arch/powerpc/include/asm/processor.h +++ b/arch/powerpc/include/asm/processor.h @@ -149,8 +149,6 @@ typedef struct { struct thread_struct { unsigned long ksp; /* Kernel stack pointer */ - unsigned long ksp_limit; /* if ksp <= ksp_limit stack overflow */ - #ifdef CONFIG_PPC64 unsigned long ksp_vsid; #endif @@ -162,6 +160,7 @@ struct thread_struct { #endif #ifdef CONFIG_PPC32 void *pgdir; /* root of page-table tree */ + unsigned long ksp_limit; /* if ksp <= ksp_limit stack overflow */ #endif #ifdef CONFIG_PPC_ADV_DEBUG_REGS /* @@ -321,7 +320,6 @@ struct thread_struct { #else #define INIT_THREAD { \ .ksp = INIT_SP, \ - .ksp_limit = INIT_SP_LIMIT, \ .regs = (struct pt_regs *)INIT_SP - 1, /* XXX bogus, I think */ \ .fs = KERNEL_DS, \ .fpr = {{0}}, \ diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index d8958be..502c7a4 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -80,10 +80,11 @@ int main(void) DEFINE(TASKTHREADPPR, offsetof(struct task_struct, thread.ppr)); #else DEFINE(THREAD_INFO, offsetof(struct task_struct, stack)); + DEFINE(THREAD_INFO_GAP, _ALIGN_UP(sizeof(struct thread_info), 16)); + DEFINE(KSP_LIMIT, offsetof(struct thread_struct, ksp_limit)); #endif /* CONFIG_PPC64 */ DEFINE(KSP, offsetof(struct thread_struct, ksp)); - DEFINE(KSP_LIMIT, offsetof(struct thread_struct, ksp_limit)); DEFINE(PT_REGS, offsetof(struct thread_struct, regs)); #ifdef CONFIG_BOOKE DEFINE(THREAD_NORMSAVES, offsetof(struct thread_struct, normsave[0])); diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 2234a12..57d286a 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -496,7 +496,6 @@ void do_IRQ(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); struct thread_info *curtp, *irqtp; - unsigned long saved_sp_limit; /* Switch to the irq stack to handle this */ curtp = current_thread_info(); @@ -509,12 +508,6 @@ void do_IRQ(struct pt_regs *regs) return; } - /* Adjust the stack limit */ - saved_sp_limit = current->thread.ksp_limit; - current->thread.ksp_limit = (unsigned long)irqtp + - _ALIGN_UP(sizeof(struct thread_info), 16); - - /* Prepare the thread_info in the irq stack */ irqtp->task = curtp->task; irqtp->flags = 0; @@ -526,7 +519,6 @@ void do_IRQ(struct pt_regs *regs) call_do_irq(regs, irqtp); /* Restore stack limit */ - current->thread.ksp_limit = saved_sp_limit; irqtp->task = NULL; /* Copy back updates to the thread_info */ @@ -604,16 +596,12 @@ void irq_ctx_init(void) static inline void do_softirq_onstack(void) { struct thread_info *curtp, *irqtp; - unsigned long saved_sp_limit = current->thread.ksp_limit; curtp = current_thread_info(); irqtp = softirq_ctx[smp_processor_id()]; irqtp->task = curtp->task; irqtp->flags = 0; - current->thread.ksp_limit = (unsigned long)irqtp + - _ALIGN_UP(sizeof(struct thread_info), 16); call_do_softirq(irqtp); - current->thread.ksp_limit = saved_sp_limit; irqtp->task = NULL; /* Set any flag that may have been set on the diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S index 7da3882..2b0ad98 100644 --- a/arch/powerpc/kernel/misc_32.S +++ b/arch/powerpc/kernel/misc_32.S @@ -36,25 +36,41 @@ .text +/* + * We store the saved ksp_limit in the unused part + * of the STACK_FRAME_OVERHEAD + */ _GLOBAL(call_do_softirq) mflr r0 stw r0,4(r1) + lwz r10,THREAD+KSP_LIMIT(r2) + addi r11,r3,THREAD_INFO_GAP stwu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r3) mr r1,r3 + stw r10,8(r1) + stw r11,THREAD+KSP_LIMIT(r2) bl __do_softirq + lwz r10,8(r1) lwz r1,0(r1) lwz r0,4(r1) + stw r10,THREAD+KSP_LIMIT(r2) mtlr r0 blr _GLOBAL(call_do_irq) mflr r0 stw r0,4(r1) + lwz r10,THREAD+KSP_LIMIT(r2) + addi r11,r3,THREAD_INFO_GAP stwu r1,THREAD_SIZE-STACK_FRAME_OVERHEAD(r4) mr r1,r4 + stw r10,8(r1) + stw r11,THREAD+KSP_LIMIT(r2) bl __do_irq + lwz r10,8(r1) lwz r1,0(r1) lwz r0,4(r1) + stw r10,THREAD+KSP_LIMIT(r2) mtlr r0 blr diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 6f428da..96d2fdf 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -1000,9 +1000,10 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, kregs = (struct pt_regs *) sp; sp -= STACK_FRAME_OVERHEAD; p->thread.ksp = sp; +#ifdef CONFIG_PPC32 p->thread.ksp_limit = (unsigned long)task_stack_page(p) + _ALIGN_UP(sizeof(struct thread_info), 16); - +#endif #ifdef CONFIG_HAVE_HW_BREAKPOINT p->thread.ptrace_bps[0] = NULL; #endif diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c index a7ee978..b1faa15 100644 --- a/arch/powerpc/lib/sstep.c +++ b/arch/powerpc/lib/sstep.c @@ -1505,6 +1505,7 @@ int __kprobes emulate_step(struct pt_regs *regs, unsigned int instr) */ if ((ra == 1) && !(regs->msr & MSR_PR) \ && (val3 >= (regs->gpr[1] - STACK_INT_FRAME_SIZE))) { +#ifdef CONFIG_PPC32 /* * Check if we will touch kernel sack overflow */ @@ -1513,7 +1514,7 @@ int __kprobes emulate_step(struct pt_regs *regs, unsigned int instr) err = -EINVAL; break; } - +#endif /* CONFIG_PPC32 */ /* * Check if we already set since that means we'll * lose the previous value. -- cgit v0.10.2 From 0c9fa29149d3726e14262aeb0c8461a948cc9d56 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 24 Sep 2013 16:10:38 +1000 Subject: powerpc/zImage: make the "OF" wrapper support ePAPR boot This makes the "OF" zImage wrapper (zImage.pseries, zImage.pmac, zImage.maple) work if booted via a flat device-tree (ePAPR boot mode), and thus potentially usable with kexec. Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 6a15c96..15ca225 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -74,7 +74,7 @@ src-wlib-$(CONFIG_8xx) += mpc8xx.c planetcore.c src-wlib-$(CONFIG_PPC_82xx) += pq2.c fsl-soc.c planetcore.c src-wlib-$(CONFIG_EMBEDDED6xx) += mv64x60.c mv64x60_i2c.c ugecon.c -src-plat-y := of.c +src-plat-y := of.c epapr.c src-plat-$(CONFIG_40x) += fixed-head.S ep405.c cuboot-hotfoot.c \ treeboot-walnut.c cuboot-acadia.c \ cuboot-kilauea.c simpleboot.c \ @@ -97,7 +97,7 @@ src-plat-$(CONFIG_EMBEDDED6xx) += cuboot-pq2.c cuboot-mpc7448hpc2.c \ prpmc2800.c src-plat-$(CONFIG_AMIGAONE) += cuboot-amigaone.c src-plat-$(CONFIG_PPC_PS3) += ps3-head.S ps3-hvcall.S ps3.c -src-plat-$(CONFIG_EPAPR_BOOT) += epapr.c +src-plat-$(CONFIG_EPAPR_BOOT) += epapr.c epapr-wrapper.c src-wlib := $(sort $(src-wlib-y)) src-plat := $(sort $(src-plat-y)) diff --git a/arch/powerpc/boot/epapr-wrapper.c b/arch/powerpc/boot/epapr-wrapper.c new file mode 100644 index 0000000..c101910 --- /dev/null +++ b/arch/powerpc/boot/epapr-wrapper.c @@ -0,0 +1,9 @@ +extern void epapr_platform_init(unsigned long r3, unsigned long r4, + unsigned long r5, unsigned long r6, + unsigned long r7); + +void platform_init(unsigned long r3, unsigned long r4, unsigned long r5, + unsigned long r6, unsigned long r7) +{ + epapr_platform_init(r3, r4, r5, r6, r7); +} diff --git a/arch/powerpc/boot/epapr.c b/arch/powerpc/boot/epapr.c index 06c1961..02e91aa 100644 --- a/arch/powerpc/boot/epapr.c +++ b/arch/powerpc/boot/epapr.c @@ -48,8 +48,8 @@ static void platform_fixups(void) fdt_addr, fdt_totalsize((void *)fdt_addr), ima_size); } -void platform_init(unsigned long r3, unsigned long r4, unsigned long r5, - unsigned long r6, unsigned long r7) +void epapr_platform_init(unsigned long r3, unsigned long r4, unsigned long r5, + unsigned long r6, unsigned long r7) { epapr_magic = r6; ima_size = r7; diff --git a/arch/powerpc/boot/of.c b/arch/powerpc/boot/of.c index 61d9899..62e2f43 100644 --- a/arch/powerpc/boot/of.c +++ b/arch/powerpc/boot/of.c @@ -26,6 +26,9 @@ static unsigned long claim_base; +void epapr_platform_init(unsigned long r3, unsigned long r4, unsigned long r5, + unsigned long r6, unsigned long r7); + static void *of_try_claim(unsigned long size) { unsigned long addr = 0; @@ -61,7 +64,7 @@ static void of_image_hdr(const void *hdr) } } -void platform_init(unsigned long a1, unsigned long a2, void *promptr) +static void of_platform_init(unsigned long a1, unsigned long a2, void *promptr) { platform_ops.image_hdr = of_image_hdr; platform_ops.malloc = of_try_claim; @@ -81,3 +84,14 @@ void platform_init(unsigned long a1, unsigned long a2, void *promptr) loader_info.initrd_size = a2; } } + +void platform_init(unsigned long r3, unsigned long r4, unsigned long r5, + unsigned long r6, unsigned long r7) +{ + /* Detect OF vs. ePAPR boot */ + if (r5) + of_platform_init(r3, r4, (void *)r5); + else + epapr_platform_init(r3, r4, r5, r6, r7); +} + diff --git a/arch/powerpc/boot/wrapper b/arch/powerpc/boot/wrapper index 6761c74..cd7af84 100755 --- a/arch/powerpc/boot/wrapper +++ b/arch/powerpc/boot/wrapper @@ -148,18 +148,18 @@ make_space=y case "$platform" in pseries) - platformo=$object/of.o + platformo="$object/of.o $object/epapr.o" link_address='0x4000000' ;; maple) - platformo=$object/of.o + platformo="$object/of.o $object/epapr.o" link_address='0x400000' ;; pmac|chrp) - platformo=$object/of.o + platformo="$object/of.o $object/epapr.o" ;; coff) - platformo="$object/crt0.o $object/of.o" + platformo="$object/crt0.o $object/of.o $object/epapr.o" lds=$object/zImage.coff.lds link_address='0x500000' pie= @@ -253,6 +253,7 @@ treeboot-iss4xx-mpic) platformo="$object/treeboot-iss4xx.o" ;; epapr) + platformo="$object/epapr.o $object/epapr-wrapper.o" link_address='0x20000000' pie=-pie ;; -- cgit v0.10.2 From dbe78b40118636f2d5d276144239dd4bfd5f04f9 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 25 Sep 2013 14:02:50 +1000 Subject: powerpc/pseries: Do not start secondaries in Open Firmware Starting secondary CPUs early on from Open Firmware and placing them in a holding spin loop slows down the boot process significantly under some hypervisors such as KVM. This is also unnecessary when RTAS supports querying the CPU state So let's not do it. Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 12e656f..5fe2842 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -196,6 +196,8 @@ static int __initdata mem_reserve_cnt; static cell_t __initdata regbuf[1024]; +static bool rtas_has_query_cpu_stopped; + /* * Error results ... some OF calls will return "-1" on error, some @@ -1574,6 +1576,11 @@ static void __init prom_instantiate_rtas(void) prom_setprop(rtas_node, "/rtas", "linux,rtas-entry", &val, sizeof(val)); + /* Check if it supports "query-cpu-stopped-state" */ + if (prom_getprop(rtas_node, "query-cpu-stopped-state", + &val, sizeof(val)) != PROM_ERROR) + rtas_has_query_cpu_stopped = true; + #if defined(CONFIG_PPC_POWERNV) && defined(__BIG_ENDIAN__) /* PowerVN takeover hack */ prom_rtas_data = base; @@ -1815,6 +1822,18 @@ static void __init prom_hold_cpus(void) = (void *) LOW_ADDR(__secondary_hold_acknowledge); unsigned long secondary_hold = LOW_ADDR(__secondary_hold); + /* + * On pseries, if RTAS supports "query-cpu-stopped-state", + * we skip this stage, the CPUs will be started by the + * kernel using RTAS. + */ + if ((of_platform == PLATFORM_PSERIES || + of_platform == PLATFORM_PSERIES_LPAR) && + rtas_has_query_cpu_stopped) { + prom_printf("prom_hold_cpus: skipped\n"); + return; + } + prom_debug("prom_hold_cpus: start...\n"); prom_debug(" 1) spinloop = 0x%x\n", (unsigned long)spinloop); prom_debug(" 1) *spinloop = 0x%x\n", *spinloop); @@ -3011,6 +3030,8 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, * On non-powermacs, put all CPUs in spin-loops. * * PowerMacs use a different mechanism to spin CPUs + * + * (This must be done after instanciating RTAS) */ if (of_platform != PLATFORM_POWERMAC && of_platform != PLATFORM_OPAL) diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c index 1c1771a..24f58cb 100644 --- a/arch/powerpc/platforms/pseries/smp.c +++ b/arch/powerpc/platforms/pseries/smp.c @@ -233,18 +233,24 @@ static void __init smp_init_pseries(void) alloc_bootmem_cpumask_var(&of_spin_mask); - /* Mark threads which are still spinning in hold loops. */ - if (cpu_has_feature(CPU_FTR_SMT)) { - for_each_present_cpu(i) { - if (cpu_thread_in_core(i) == 0) - cpumask_set_cpu(i, of_spin_mask); - } - } else { - cpumask_copy(of_spin_mask, cpu_present_mask); + /* + * Mark threads which are still spinning in hold loops + * + * We know prom_init will not have started them if RTAS supports + * query-cpu-stopped-state. + */ + if (rtas_token("query-cpu-stopped-state") == RTAS_UNKNOWN_SERVICE) { + if (cpu_has_feature(CPU_FTR_SMT)) { + for_each_present_cpu(i) { + if (cpu_thread_in_core(i) == 0) + cpumask_set_cpu(i, of_spin_mask); + } + } else + cpumask_copy(of_spin_mask, cpu_present_mask); + + cpumask_clear_cpu(boot_cpuid, of_spin_mask); } - cpumask_clear_cpu(boot_cpuid, of_spin_mask); - /* Non-lpar has additional take/give timebase */ if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) { smp_ops->give_timebase = rtas_give_timebase; -- cgit v0.10.2 From 7a20c2fad61aa3624e83c671d36dbd36b2661476 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Tue, 24 Sep 2013 20:13:44 -0400 Subject: x86/reboot: Fix apparent cut-n-paste mistake in Dell reboot workaround This seems to have been copied from the Optiplex 990 entry above, but somoene forgot to change the ident text. Signed-off-by: Dave Jones Link: http://lkml.kernel.org/r/20130925001344.GA13554@redhat.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 5f4ad27..e643e74 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -352,7 +352,7 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { }, { /* Handle problems with rebooting on the Precision M6600. */ .callback = set_pci_reboot, - .ident = "Dell OptiPlex 990", + .ident = "Dell Precision M6600", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), DMI_MATCH(DMI_PRODUCT_NAME, "Precision M6600"), -- cgit v0.10.2 From 0160676bba69523e8b0ac83f306cce7d342ed7c8 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Fri, 13 Sep 2013 15:13:30 +0100 Subject: xen/p2m: check MFN is in range before using the m2p table On hosts with more than 168 GB of memory, a 32-bit guest may attempt to grant map an MFN that is error cannot lookup in its mapping of the m2p table. There is an m2p lookup as part of m2p_add_override() and m2p_remove_override(). The lookup falls off the end of the mapped portion of the m2p and (because the mapping is at the highest virtual address) wraps around and the lookup causes a fault on what appears to be a user space address. do_page_fault() (thinking it's a fault to a userspace address), tries to lock mm->mmap_sem. If the gntdev device is used for the grant map, m2p_add_override() is called from from gnttab_mmap() with mm->mmap_sem already locked. do_page_fault() then deadlocks. The deadlock would most commonly occur when a 64-bit guest is started and xenconsoled attempts to grant map its console ring. Introduce mfn_to_pfn_no_overrides() which checks the MFN is within the mapped portion of the m2p table before accessing the table and use this in m2p_add_override(), m2p_remove_override(), and mfn_to_pfn() (which already had the correct range check). All faults caused by accessing the non-existant parts of the m2p are thus within the kernel address space and exception_fixup() is called without trying to lock mm->mmap_sem. This means that for MFNs that are outside the mapped range of the m2p then mfn_to_pfn() will always look in the m2p overrides. This is correct because it must be a foreign MFN (and the PFN in the m2p in this case is only relevant for the other domain). Signed-off-by: David Vrabel Cc: Stefano Stabellini Cc: Jan Beulich -- v3: check for auto_translated_physmap in mfn_to_pfn_no_overrides() v2: in mfn_to_pfn() look in m2p_overrides if the MFN is out of range as it's probably foreign. Signed-off-by: Konrad Rzeszutek Wilk Acked-by: Stefano Stabellini diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index 6aef9fb..b913915 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -79,30 +79,38 @@ static inline int phys_to_machine_mapping_valid(unsigned long pfn) return get_phys_to_machine(pfn) != INVALID_P2M_ENTRY; } -static inline unsigned long mfn_to_pfn(unsigned long mfn) +static inline unsigned long mfn_to_pfn_no_overrides(unsigned long mfn) { unsigned long pfn; - int ret = 0; + int ret; if (xen_feature(XENFEAT_auto_translated_physmap)) return mfn; - if (unlikely(mfn >= machine_to_phys_nr)) { - pfn = ~0; - goto try_override; - } - pfn = 0; + if (unlikely(mfn >= machine_to_phys_nr)) + return ~0; + /* * The array access can fail (e.g., device space beyond end of RAM). * In such cases it doesn't matter what we return (we return garbage), * but we must handle the fault without crashing! */ ret = __get_user(pfn, &machine_to_phys_mapping[mfn]); -try_override: - /* ret might be < 0 if there are no entries in the m2p for mfn */ if (ret < 0) - pfn = ~0; - else if (get_phys_to_machine(pfn) != mfn) + return ~0; + + return pfn; +} + +static inline unsigned long mfn_to_pfn(unsigned long mfn) +{ + unsigned long pfn; + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return mfn; + + pfn = mfn_to_pfn_no_overrides(mfn); + if (get_phys_to_machine(pfn) != mfn) { /* * If this appears to be a foreign mfn (because the pfn * doesn't map back to the mfn), then check the local override @@ -111,6 +119,7 @@ try_override: * m2p_find_override_pfn returns ~0 if it doesn't find anything. */ pfn = m2p_find_override_pfn(mfn, ~0); + } /* * pfn is ~0 if there are no entries in the m2p for mfn or if the diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c index 8b901e8..a61c7d5 100644 --- a/arch/x86/xen/p2m.c +++ b/arch/x86/xen/p2m.c @@ -879,7 +879,6 @@ int m2p_add_override(unsigned long mfn, struct page *page, unsigned long uninitialized_var(address); unsigned level; pte_t *ptep = NULL; - int ret = 0; pfn = page_to_pfn(page); if (!PageHighMem(page)) { @@ -926,8 +925,8 @@ int m2p_add_override(unsigned long mfn, struct page *page, * frontend pages while they are being shared with the backend, * because mfn_to_pfn (that ends up being called by GUPF) will * return the backend pfn rather than the frontend pfn. */ - ret = __get_user(pfn, &machine_to_phys_mapping[mfn]); - if (ret == 0 && get_phys_to_machine(pfn) == mfn) + pfn = mfn_to_pfn_no_overrides(mfn); + if (get_phys_to_machine(pfn) == mfn) set_phys_to_machine(pfn, FOREIGN_FRAME(mfn)); return 0; @@ -942,7 +941,6 @@ int m2p_remove_override(struct page *page, unsigned long uninitialized_var(address); unsigned level; pte_t *ptep = NULL; - int ret = 0; pfn = page_to_pfn(page); mfn = get_phys_to_machine(pfn); @@ -1029,8 +1027,8 @@ int m2p_remove_override(struct page *page, * the original pfn causes mfn_to_pfn(mfn) to return the frontend * pfn again. */ mfn &= ~FOREIGN_FRAME_BIT; - ret = __get_user(pfn, &machine_to_phys_mapping[mfn]); - if (ret == 0 && get_phys_to_machine(pfn) == FOREIGN_FRAME(mfn) && + pfn = mfn_to_pfn_no_overrides(mfn); + if (get_phys_to_machine(pfn) == FOREIGN_FRAME(mfn) && m2p_find_override(mfn) == NULL) set_phys_to_machine(pfn, mfn); -- cgit v0.10.2 From 15a3eac0784edb73cb2ec0f0c97705ba4bd39b3f Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Wed, 25 Sep 2013 10:07:20 -0400 Subject: xen/spinlock: Document the xen_nopvspin parameter. Which disables in the ticketlock slowpath the Xen PV optimization's. Useful for diagnosing issues and comparing benchmarks in over-commit CPU scenarios. Signed-off-by: Konrad Rzeszutek Wilk diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 7f9d4f5..b958912 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -3412,6 +3412,10 @@ bytes respectively. Such letter suffixes can also be entirely omitted. the unplug protocol never -- do not unplug even if version check succeeds + xen_nopvspin [X86,XEN] + Disables the ticketlock slowpath using Xen PV + optimizations. + xirc2ps_cs= [NET,PCMCIA] Format: ,,,,,[,[,[,]]] -- cgit v0.10.2 From 55c25c2f14496badefd780a9f179442756216b67 Mon Sep 17 00:00:00 2001 From: Jayachandran C Date: Wed, 25 Sep 2013 18:31:05 +0530 Subject: MIPS: mm: Move some checks out of 'for' loop in DMA operations The check cpu_needs_post_dma_flush() in mips_dma_sync_sg_for_cpu() and the check !plat_device_is_coherent() in mips_dma_sync_sg_for_device() can be moved outside the for loop. As a side effect, this also avoids a GCC bug that caused kernel compile to fail with the error: arch/mips/mm/dma-default.c: In function 'mips_dma_sync_sg_for_cpu': arch/mips/mm/dma-default.c:316:1: internal compiler error: in add_insn_before, at emit-rtl.c:3852 This gcc failure is seen in Code Sourcery toolchains [e.g. gcc version 4.7.2 (Sourcery CodeBench Lite 2012.09-99)] after commit "MIPS: Optimize current_cpu_type() for better code." Signed-off-by: Jayachandran C Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/5907/ Reviewed-by: Markos Chandras Tested-by: Markos Chandras Signed-off-by: Ralf Baechle diff --git a/arch/mips/mm/dma-default.c b/arch/mips/mm/dma-default.c index f25a7e9..5f8b955 100644 --- a/arch/mips/mm/dma-default.c +++ b/arch/mips/mm/dma-default.c @@ -308,12 +308,10 @@ static void mips_dma_sync_sg_for_cpu(struct device *dev, { int i; - /* Make sure that gcc doesn't leave the empty loop body. */ - for (i = 0; i < nelems; i++, sg++) { - if (cpu_needs_post_dma_flush(dev)) + if (cpu_needs_post_dma_flush(dev)) + for (i = 0; i < nelems; i++, sg++) __dma_sync(sg_page(sg), sg->offset, sg->length, direction); - } } static void mips_dma_sync_sg_for_device(struct device *dev, @@ -321,12 +319,10 @@ static void mips_dma_sync_sg_for_device(struct device *dev, { int i; - /* Make sure that gcc doesn't leave the empty loop body. */ - for (i = 0; i < nelems; i++, sg++) { - if (!plat_device_is_coherent(dev)) + if (!plat_device_is_coherent(dev)) + for (i = 0; i < nelems; i++, sg++) __dma_sync(sg_page(sg), sg->offset, sg->length, direction); - } } int mips_dma_mapping_error(struct device *dev, dma_addr_t dma_addr) -- cgit v0.10.2 From 576b523721b7fbbf44a411ff0b3d958906c41653 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Wed, 25 Sep 2013 22:16:16 +0900 Subject: perf probe: Fix probing symbols with optimization suffix Fix perf probe to probe on some symbols which have some optimzation suffixes, e.g. ".part", ".isra", and ".constprop". To fix this issue, instead of using the DIE name, perf probe uses the symbol name found by dwfl_module_addrsym(). This also involves a perf probe --vars operation update which now shows the symbol name instead of the DIE name. Without this patch, putting a probe on an inlined function which was compiled with a suffixed symbol will fail like this: $ perf probe -v getname_flags probe-definition(0): getname_flags symbol:getname_flags file:(null) line:0 offset:0 return:0 lazy:(null) 0 arguments Looking at the vmlinux_path (6 entries long) Using /lib/modules/3.11.0+/build/vmlinux for symbols found inline addr: 0xffffffff8119bb70 Probe point found: getname_flags+0 found inline addr: 0xffffffff8119bcb6 Probe point found: getname+6 found inline addr: 0xffffffff811a06a6 Probe point found: user_path_at_empty+6 find 3 probe_trace_events. Opening /sys/kernel/debug//tracing/kprobe_events write=1 Added new events: Writing event: p:probe/getname_flags getname_flags+0 Failed to write event: No such file or directory Error: Failed to add events. (-1) Because the debuginfo knows only the original (non suffix) symbol name, it uses the original symbol for probe address but the kernel (kallsyms) knows only suffixed symbol. Then, the kernel rejects that original symbol. This patch uses dwfl_module_addrsym() to get the correct (suffixed) symbol from symtab when a probe point is found. Reported-by: Arnaldo Carvalho de Melo Signed-off-by: Masami Hiramatsu Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20130925131616.31632.46658.stgit@udc4-manage.rcp.hitachi.co.jp Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 20c7299..371476c 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -118,7 +118,6 @@ static const Dwfl_Callbacks offline_callbacks = { static int debuginfo__init_offline_dwarf(struct debuginfo *self, const char *path) { - Dwfl_Module *mod; int fd; fd = open(path, O_RDONLY); @@ -129,11 +128,11 @@ static int debuginfo__init_offline_dwarf(struct debuginfo *self, if (!self->dwfl) goto error; - mod = dwfl_report_offline(self->dwfl, "", "", fd); - if (!mod) + self->mod = dwfl_report_offline(self->dwfl, "", "", fd); + if (!self->mod) goto error; - self->dbg = dwfl_module_getdwarf(mod, &self->bias); + self->dbg = dwfl_module_getdwarf(self->mod, &self->bias); if (!self->dbg) goto error; @@ -676,37 +675,42 @@ static int find_variable(Dwarf_Die *sc_die, struct probe_finder *pf) } /* Convert subprogram DIE to trace point */ -static int convert_to_trace_point(Dwarf_Die *sp_die, Dwarf_Addr paddr, - bool retprobe, struct probe_trace_point *tp) +static int convert_to_trace_point(Dwarf_Die *sp_die, Dwfl_Module *mod, + Dwarf_Addr paddr, bool retprobe, + struct probe_trace_point *tp) { Dwarf_Addr eaddr, highaddr; - const char *name; - - /* Copy the name of probe point */ - name = dwarf_diename(sp_die); - if (name) { - if (dwarf_entrypc(sp_die, &eaddr) != 0) { - pr_warning("Failed to get entry address of %s\n", - dwarf_diename(sp_die)); - return -ENOENT; - } - if (dwarf_highpc(sp_die, &highaddr) != 0) { - pr_warning("Failed to get end address of %s\n", - dwarf_diename(sp_die)); - return -ENOENT; - } - if (paddr > highaddr) { - pr_warning("Offset specified is greater than size of %s\n", - dwarf_diename(sp_die)); - return -EINVAL; - } - tp->symbol = strdup(name); - if (tp->symbol == NULL) - return -ENOMEM; - tp->offset = (unsigned long)(paddr - eaddr); - } else - /* This function has no name. */ - tp->offset = (unsigned long)paddr; + GElf_Sym sym; + const char *symbol; + + /* Verify the address is correct */ + if (dwarf_entrypc(sp_die, &eaddr) != 0) { + pr_warning("Failed to get entry address of %s\n", + dwarf_diename(sp_die)); + return -ENOENT; + } + if (dwarf_highpc(sp_die, &highaddr) != 0) { + pr_warning("Failed to get end address of %s\n", + dwarf_diename(sp_die)); + return -ENOENT; + } + if (paddr > highaddr) { + pr_warning("Offset specified is greater than size of %s\n", + dwarf_diename(sp_die)); + return -EINVAL; + } + + /* Get an appropriate symbol from symtab */ + symbol = dwfl_module_addrsym(mod, paddr, &sym, NULL); + if (!symbol) { + pr_warning("Failed to find symbol at 0x%lx\n", + (unsigned long)paddr); + return -ENOENT; + } + tp->offset = (unsigned long)(paddr - sym.st_value); + tp->symbol = strdup(symbol); + if (!tp->symbol) + return -ENOMEM; /* Return probe must be on the head of a subprogram */ if (retprobe) { @@ -1149,7 +1153,7 @@ static int add_probe_trace_event(Dwarf_Die *sc_die, struct probe_finder *pf) tev = &tf->tevs[tf->ntevs++]; /* Trace point should be converted from subprogram DIE */ - ret = convert_to_trace_point(&pf->sp_die, pf->addr, + ret = convert_to_trace_point(&pf->sp_die, tf->mod, pf->addr, pf->pev->point.retprobe, &tev->point); if (ret < 0) return ret; @@ -1181,7 +1185,7 @@ int debuginfo__find_trace_events(struct debuginfo *self, { struct trace_event_finder tf = { .pf = {.pev = pev, .callback = add_probe_trace_event}, - .max_tevs = max_tevs}; + .mod = self->mod, .max_tevs = max_tevs}; int ret; /* Allocate result tevs array */ @@ -1250,7 +1254,7 @@ static int add_available_vars(Dwarf_Die *sc_die, struct probe_finder *pf) vl = &af->vls[af->nvls++]; /* Trace point should be converted from subprogram DIE */ - ret = convert_to_trace_point(&pf->sp_die, pf->addr, + ret = convert_to_trace_point(&pf->sp_die, af->mod, pf->addr, pf->pev->point.retprobe, &vl->point); if (ret < 0) return ret; @@ -1289,6 +1293,7 @@ int debuginfo__find_available_vars_at(struct debuginfo *self, { struct available_var_finder af = { .pf = {.pev = pev, .callback = add_available_vars}, + .mod = self->mod, .max_vls = max_vls, .externs = externs}; int ret; diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h index 17e94d0..3b7d630 100644 --- a/tools/perf/util/probe-finder.h +++ b/tools/perf/util/probe-finder.h @@ -23,6 +23,7 @@ static inline int is_c_varname(const char *name) /* debug information structure */ struct debuginfo { Dwarf *dbg; + Dwfl_Module *mod; Dwfl *dwfl; Dwarf_Addr bias; }; @@ -77,6 +78,7 @@ struct probe_finder { struct trace_event_finder { struct probe_finder pf; + Dwfl_Module *mod; /* For solving symbols */ struct probe_trace_event *tevs; /* Found trace events */ int ntevs; /* Number of trace events */ int max_tevs; /* Max number of trace events */ @@ -84,6 +86,7 @@ struct trace_event_finder { struct available_var_finder { struct probe_finder pf; + Dwfl_Module *mod; /* For solving symbols */ struct variable_list *vls; /* Found variable lists */ int nvls; /* Number of variable lists */ int max_vls; /* Max no. of variable lists */ -- cgit v0.10.2 From 6d19912c9bc0ae0ba5f7dd32c714ea7cc3339c81 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Sun, 22 Sep 2013 19:44:57 -0600 Subject: perf tools: Explicitly add libdl dependency Fixes compile failure on Fedora 12. Signed-off-by: David Ahern Link: http://lkml.kernel.org/r/1379900700-5186-3-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile index 346ee92..5f6f9b3 100644 --- a/tools/perf/config/Makefile +++ b/tools/perf/config/Makefile @@ -87,7 +87,7 @@ CFLAGS += -Wall CFLAGS += -Wextra CFLAGS += -std=gnu99 -EXTLIBS = -lelf -lpthread -lrt -lm +EXTLIBS = -lelf -lpthread -lrt -lm -ldl ifeq ($(call try-cc,$(SOURCE_HELLO),$(CFLAGS) -Werror -fstack-protector-all,-fstack-protector-all),y) CFLAGS += -fstack-protector-all -- cgit v0.10.2 From aa7fe3b0c499fb7987245ac40295af03546f2bd2 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Sun, 22 Sep 2013 13:22:09 +0300 Subject: perf machine: Fix path unpopulated in machine__create_modules() In machine__create_modules() the 'path' char array was used in a call to symbol__restricted_filename() without always being populated. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1379845338-29637-2-git-send-email-adrian.hunter@intel.com [ Split patch removing unrelated conversion of sprintf to snprintf to perf/core ] Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 933d14f..6188d28 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -792,7 +792,7 @@ static int machine__create_modules(struct machine *machine) modules = path; } - if (symbol__restricted_filename(path, "/proc/modules")) + if (symbol__restricted_filename(modules, "/proc/modules")) return -1; file = fopen(modules, "r"); -- cgit v0.10.2 From de95ab53645a2f0015e0f68ee723f18dce2b8b51 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 12 Sep 2013 08:16:38 -0700 Subject: perf symbols: Demangle cloned functions The libbfd C++ demangler doesn't seem to deal with cloned functions, like symbol.clone.NUM. Just strip the dot part before demangling and add it back later. Signed-off-by: Andi Kleen Link: http://lkml.kernel.org/r/1378998998-10802-1-git-send-email-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index a9c829b..d2a888e 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -928,8 +928,33 @@ int dso__load_sym(struct dso *dso, struct map *map, * to it... */ if (symbol_conf.demangle) { - demangled = bfd_demangle(NULL, elf_name, + /* + * The demangler doesn't deal with cloned functions. + * XXXX.clone.NUM or similar + * Strip the dot part and readd it later. + */ + char *p = (char *)elf_name, *dot; + dot = strchr(elf_name, '.'); + if (dot) { + p = strdup(elf_name); + if (!p) + goto new_symbol; + dot = strchr(p, '.'); + *dot = 0; + } + + demangled = bfd_demangle(NULL, p, DMGL_PARAMS | DMGL_ANSI); + if (dot) + *dot = '.'; + if (demangled && dot) { + demangled = realloc(demangled, strlen(demangled) + strlen(dot) + 1); + if (!demangled) + goto new_symbol; + strcpy(demangled + (dot - p), dot); + } + if (p != elf_name) + free(p); if (demangled != NULL) elf_name = demangled; } -- cgit v0.10.2 From 58d327da9721f7a0f6e46c8dfa5cc5546fd7078a Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 25 Sep 2013 12:04:37 -0400 Subject: drm/radeon: fix hdmi audio on DCE3.0/3.1 asics These asics seem to use a mix of the DCE2.x and DCE3.2 audio interfaces despite what the register spec says. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=69729 https://bugs.freedesktop.org/show_bug.cgi?id=69671 Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c index f443010..b0fa600 100644 --- a/drivers/gpu/drm/radeon/r600_hdmi.c +++ b/drivers/gpu/drm/radeon/r600_hdmi.c @@ -257,10 +257,7 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock) * number (coefficient of two integer numbers. DCCG_AUDIO_DTOx_PHASE * is the numerator, DCCG_AUDIO_DTOx_MODULE is the denominator */ - if (ASIC_IS_DCE3(rdev)) { - /* according to the reg specs, this should DCE3.2 only, but in - * practice it seems to cover DCE3.0 as well. - */ + if (ASIC_IS_DCE32(rdev)) { if (dig->dig_encoder == 0) { dto_cntl = RREG32(DCCG_AUDIO_DTO0_CNTL) & ~DCCG_AUDIO_DTO_WALLCLOCK_RATIO_MASK; dto_cntl |= DCCG_AUDIO_DTO_WALLCLOCK_RATIO(wallclock_ratio); @@ -276,8 +273,21 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock) WREG32(DCCG_AUDIO_DTO1_MODULE, dto_modulo); WREG32(DCCG_AUDIO_DTO_SELECT, 1); /* select DTO1 */ } + } else if (ASIC_IS_DCE3(rdev)) { + /* according to the reg specs, this should DCE3.2 only, but in + * practice it seems to cover DCE3.0/3.1 as well. + */ + if (dig->dig_encoder == 0) { + WREG32(DCCG_AUDIO_DTO0_PHASE, base_rate * 100); + WREG32(DCCG_AUDIO_DTO0_MODULE, clock * 100); + WREG32(DCCG_AUDIO_DTO_SELECT, 0); /* select DTO0 */ + } else { + WREG32(DCCG_AUDIO_DTO1_PHASE, base_rate * 100); + WREG32(DCCG_AUDIO_DTO1_MODULE, clock * 100); + WREG32(DCCG_AUDIO_DTO_SELECT, 1); /* select DTO1 */ + } } else { - /* according to the reg specs, this should be DCE2.0 and DCE3.0 */ + /* according to the reg specs, this should be DCE2.0 and DCE3.0/3.1 */ WREG32(AUDIO_DTO, AUDIO_DTO_PHASE(base_rate / 10) | AUDIO_DTO_MODULE(clock / 10)); } -- cgit v0.10.2 From b27b14cebfdb687e6d281d2cf82858a30796dc2a Mon Sep 17 00:00:00 2001 From: Hanjun Guo Date: Sun, 22 Sep 2013 15:42:41 +0800 Subject: ACPI / scan: fix typo in comments of acpi_bus_unregister_driver() "APIC" should be "ACPI" here. Signed-off-by: Hanjun Guo Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index fbdb82e..611ce90 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -1121,7 +1121,7 @@ int acpi_bus_register_driver(struct acpi_driver *driver) EXPORT_SYMBOL(acpi_bus_register_driver); /** - * acpi_bus_unregister_driver - unregisters a driver with the APIC bus + * acpi_bus_unregister_driver - unregisters a driver with the ACPI bus * @driver: driver to unregister * * Unregisters a driver with the ACPI bus. Searches the namespace for all -- cgit v0.10.2 From 743a7ecbe8e28d487e2573ec2b7fb1179a443717 Mon Sep 17 00:00:00 2001 From: Mikael Pettersson Date: Wed, 25 Sep 2013 22:21:55 +0200 Subject: update contact information for Mikael Pettersson My old @it.uu.se email address is going away, so update relevant files to point to my @gmail.com address instead. In sata_promise.c just delete the address, people can get it from MAINTAINERS. Signed-off-by: Mikael Pettersson Signed-off-by: Linus Torvalds diff --git a/CREDITS b/CREDITS index 9416a9a..0640e16 100644 --- a/CREDITS +++ b/CREDITS @@ -2808,8 +2808,7 @@ S: Ottawa, Ontario S: Canada K2P 0X8 N: Mikael Pettersson -E: mikpe@it.uu.se -W: http://user.it.uu.se/~mikpe/linux/ +E: mikpelinux@gmail.com D: Miscellaneous fixes N: Reed H. Petty diff --git a/MAINTAINERS b/MAINTAINERS index b2f857c..c53fe95 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6596,7 +6596,7 @@ S: Obsolete F: drivers/net/wireless/prism54/ PROMISE SATA TX2/TX4 CONTROLLER LIBATA DRIVER -M: Mikael Pettersson +M: Mikael Pettersson L: linux-ide@vger.kernel.org S: Maintained F: drivers/ata/sata_promise.* diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c index 958ba2a..97f4acb 100644 --- a/drivers/ata/sata_promise.c +++ b/drivers/ata/sata_promise.c @@ -2,7 +2,7 @@ * sata_promise.c - Promise SATA * * Maintained by: Tejun Heo - * Mikael Pettersson + * Mikael Pettersson * Please ALWAYS copy linux-ide@vger.kernel.org * on emails. * -- cgit v0.10.2 From d32270460fee83e22ee9e6b1bfd7b486263eeb1d Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Wed, 25 Sep 2013 14:00:02 +0200 Subject: drm/i915: Fix up usage of SHRINK_STOP In commit 81e49f811404f428a9d9a63295a0c267e802fa12 Author: Glauber Costa Date: Wed Aug 28 10:18:13 2013 +1000 i915: bail out earlier when shrinker cannot acquire mutex SHRINK_STOP was added to tell the core shrinker code to bail out and go to the next shrinker since the i915 shrinker couldn't acquire required locks. But the SHRINK_STOP return code was added to the ->count_objects callback and not the ->scan_objects callback as it should have been, resulting in tons of dmesg noise like shrink_slab: i915_gem_inactive_scan+0x0/0x9c negative objects to delete nr=-xxxxxxxxx Fix discusssed with Dave Chinner. References: http://www.spinics.net/lists/intel-gfx/msg33597.html Reported-by: Knut Petersen Cc: Knut Petersen Cc: Dave Chinner Cc: Glauber Costa Cc: Glauber Costa Cc: Andrew Morton Cc: Rik van Riel Cc: Mel Gorman Cc: Johannes Weiner Cc: Michal Hocko Acked-by: Dave Chinner Signed-off-by: Daniel Vetter diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index df9253d..cdfb9da 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -4800,10 +4800,10 @@ i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc) if (!mutex_trylock(&dev->struct_mutex)) { if (!mutex_is_locked_by(&dev->struct_mutex, current)) - return SHRINK_STOP; + return 0; if (dev_priv->mm.shrinker_no_lock_stealing) - return SHRINK_STOP; + return 0; unlock = false; } @@ -4901,10 +4901,10 @@ i915_gem_inactive_scan(struct shrinker *shrinker, struct shrink_control *sc) if (!mutex_trylock(&dev->struct_mutex)) { if (!mutex_is_locked_by(&dev->struct_mutex, current)) - return 0; + return SHRINK_STOP; if (dev_priv->mm.shrinker_no_lock_stealing) - return 0; + return SHRINK_STOP; unlock = false; } -- cgit v0.10.2 From 430b849a5e0ad5875da5bc5e10b072844dcf600f Mon Sep 17 00:00:00 2001 From: Thomas Meyer Date: Thu, 19 Sep 2013 23:45:46 +0200 Subject: staging: lustre: Cocci spatch "noderef" sizeof when applied to a pointer typed expression gives the size of the pointer. Found by coccinelle spatch "misc/noderef.cocci" Signed-off-by: Thomas Meyer Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/lustre/lustre/obdecho/echo_client.c b/drivers/staging/lustre/lustre/obdecho/echo_client.c index 2644edf..c8b4344 100644 --- a/drivers/staging/lustre/lustre/obdecho/echo_client.c +++ b/drivers/staging/lustre/lustre/obdecho/echo_client.c @@ -1387,7 +1387,7 @@ echo_copyout_lsm (struct lov_stripe_md *lsm, void *_ulsm, int ulsm_nob) if (nob > ulsm_nob) return (-EINVAL); - if (copy_to_user (ulsm, lsm, sizeof(ulsm))) + if (copy_to_user (ulsm, lsm, sizeof(*ulsm))) return (-EFAULT); for (i = 0; i < lsm->lsm_stripe_count; i++) { -- cgit v0.10.2 From 599aa6975e7a26ba1fee6a1eee21f46c2730890e Mon Sep 17 00:00:00 2001 From: Lidza Louina Date: Mon, 23 Sep 2013 16:13:15 -0400 Subject: MAINTAINERS: staging: dgnc and dgap drivers: add maintainer This patch adds the staging/dgnc [DIGI NEO AND CLASSIC PCI PRODUCTS] and staging/dgap [DIGI EPCA PCI PRODUCTS] drivers to the MAINTAINERS file. I am listed as the maintainer and the driverdev-devel list is the mailing list for these drivers. Signed-off-by: Lidza Louina Signed-off-by: Greg Kroah-Hartman diff --git a/MAINTAINERS b/MAINTAINERS index e61c2e8..7a426ba 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2639,6 +2639,18 @@ F: include/linux/device-mapper.h F: include/linux/dm-*.h F: include/uapi/linux/dm-*.h +DIGI NEO AND CLASSIC PCI PRODUCTS +M: Lidza Louina +L: driverdev-devel@linuxdriverproject.org +S: Maintained +F: drivers/staging/dgnc/ + +DIGI EPCA PCI PRODUCTS +M: Lidza Louina +L: driverdev-devel@linuxdriverproject.org +S: Maintained +F: drivers/staging/dgap/ + DIOLAN U2C-12 I2C DRIVER M: Guenter Roeck L: linux-i2c@vger.kernel.org -- cgit v0.10.2 From f6853448667b3b3234331b22d34d7158b12237fd Mon Sep 17 00:00:00 2001 From: Thomas Meyer Date: Thu, 19 Sep 2013 23:45:46 +0200 Subject: staging: r8188eu: Add files for new drive: Cocci spatch "noderef" sizeof when applied to a pointer typed expression gives the size of the pointer. Found by coccinelle spatch "misc/noderef.cocci" Signed-off-by: Thomas Meyer Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/rtl8188eu/hal/rtl8188e_dm.c b/drivers/staging/rtl8188eu/hal/rtl8188e_dm.c index 9c2e7a2..ec0028d 100644 --- a/drivers/staging/rtl8188eu/hal/rtl8188e_dm.c +++ b/drivers/staging/rtl8188eu/hal/rtl8188e_dm.c @@ -57,7 +57,7 @@ static void Init_ODM_ComInfo_88E(struct adapter *Adapter) u8 cut_ver, fab_ver; /* Init Value */ - _rtw_memset(dm_odm, 0, sizeof(dm_odm)); + _rtw_memset(dm_odm, 0, sizeof(*dm_odm)); dm_odm->Adapter = Adapter; -- cgit v0.10.2 From 0a69bb4691662054596f48b9a1cab8f9323442c8 Mon Sep 17 00:00:00 2001 From: Thomas Meyer Date: Thu, 19 Sep 2013 23:45:46 +0200 Subject: staging: octeon-usb: Cocci spatch "noderef" sizeof when applied to a pointer typed expression gives the size of the pointer. Found by coccinelle spatch "misc/noderef.cocci" Signed-off-by: Thomas Meyer Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/octeon-usb/cvmx-usb.c b/drivers/staging/octeon-usb/cvmx-usb.c index d7b3c82..45dfe94 100644 --- a/drivers/staging/octeon-usb/cvmx-usb.c +++ b/drivers/staging/octeon-usb/cvmx-usb.c @@ -604,7 +604,7 @@ int cvmx_usb_initialize(struct cvmx_usb_state *state, int usb_port_number, } } - memset(usb, 0, sizeof(usb)); + memset(usb, 0, sizeof(*usb)); usb->init_flags = flags; /* Initialize the USB state structure */ -- cgit v0.10.2 From 5a2d8292f1051c1d1c08fc1732c74ac12ef61452 Mon Sep 17 00:00:00 2001 From: Thomas Meyer Date: Thu, 19 Sep 2013 23:45:46 +0200 Subject: staging: r8188eu: Cocci spatch "noderef" sizeof when applied to a pointer typed expression gives the size of the pointer. Found by coccinelle spatch "misc/noderef.cocci" Signed-off-by: Thomas Meyer Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c b/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c index cd4100f..95953eb 100644 --- a/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c +++ b/drivers/staging/rtl8188eu/os_dep/ioctl_linux.c @@ -6973,7 +6973,7 @@ static int rtw_mp_ctx(struct net_device *dev, stop = strncmp(extra, "stop", 4); sscanf(extra, "count =%d, pkt", &count); - _rtw_memset(extra, '\0', sizeof(extra)); + _rtw_memset(extra, '\0', sizeof(*extra)); if (stop == 0) { bStartTest = 0; /* To set Stop */ -- cgit v0.10.2 From ffceff44e45c29943286428d246516e8e9d093d9 Mon Sep 17 00:00:00 2001 From: Thomas Meyer Date: Thu, 19 Sep 2013 23:45:46 +0200 Subject: staging: r8188eu: Add files for new drive: Cocci spatch "noderef" sizeof when applied to a pointer typed expression gives the size of the pointer. Found by coccinelle spatch "misc/noderef.cocci" Signed-off-by: Thomas Meyer Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/rtl8188eu/core/rtw_mp.c b/drivers/staging/rtl8188eu/core/rtw_mp.c index c7ff2e4..9832dcb 100644 --- a/drivers/staging/rtl8188eu/core/rtw_mp.c +++ b/drivers/staging/rtl8188eu/core/rtw_mp.c @@ -907,7 +907,7 @@ u32 mp_query_psd(struct adapter *pAdapter, u8 *data) sscanf(data, "pts =%d, start =%d, stop =%d", &psd_pts, &psd_start, &psd_stop); } - _rtw_memset(data, '\0', sizeof(data)); + _rtw_memset(data, '\0', sizeof(*data)); i = psd_start; while (i < psd_stop) { -- cgit v0.10.2 From 3ff4afe80e50ff4e84b4a59a38a6a657d1e7fdd1 Mon Sep 17 00:00:00 2001 From: Tobias Polzer Date: Wed, 18 Sep 2013 18:04:59 +0200 Subject: staging: usbip: Orphan usbip The domain of Matt Mooneey's email doesn't exist anymore. Setting usbip to Oprhan. Signed-off-by: Tobias Polzer Signed-off-by: Dominik Paulus Signed-off-by: Greg Kroah-Hartman diff --git a/MAINTAINERS b/MAINTAINERS index 7a426ba..fd4f583 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8736,9 +8736,8 @@ F: Documentation/hid/hiddev.txt F: drivers/hid/usbhid/ USB/IP DRIVERS -M: Matt Mooney L: linux-usb@vger.kernel.org -S: Maintained +S: Orphan F: drivers/staging/usbip/ USB ISP116X DRIVER -- cgit v0.10.2 From c3aed2312fd3c74e77e049b795e092689d414474 Mon Sep 17 00:00:00 2001 From: Iker Pedrosa Date: Fri, 20 Sep 2013 17:11:27 +0200 Subject: Staging: rtl8192u: r819xU_cmdpkt: checking NULL value after doing dev_alloc_skb Checking the return of dev_alloc_skb as stated in the following bug: https://bugzilla.kernel.org/show_bug.cgi?id=60411 Signed-off-by: Iker Pedrosa Reported-by: RUC_Soft_Sec rucsoftsec@gmail.com Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/rtl8192u/r819xU_cmdpkt.c b/drivers/staging/rtl8192u/r819xU_cmdpkt.c index 5bc361b..5614401 100644 --- a/drivers/staging/rtl8192u/r819xU_cmdpkt.c +++ b/drivers/staging/rtl8192u/r819xU_cmdpkt.c @@ -37,6 +37,8 @@ rt_status SendTxCommandPacket(struct net_device *dev, void *pData, u32 DataLen) /* Get TCB and local buffer from common pool. (It is shared by CmdQ, MgntQ, and USB coalesce DataQ) */ skb = dev_alloc_skb(USB_HWDESC_HEADER_LEN + DataLen + 4); + if (!skb) + return RT_STATUS_FAILURE; memcpy((unsigned char *)(skb->cb), &dev, sizeof(dev)); tcb_desc = (cb_desc *)(skb->cb + MAX_DEV_ADDR_SIZE); tcb_desc->queue_index = TXCMD_QUEUE; -- cgit v0.10.2 From 18e35e081e1bb129c6dc5a5874c260a613cd13d7 Mon Sep 17 00:00:00 2001 From: Malcolm Priestley Date: Sun, 22 Sep 2013 20:15:58 +0100 Subject: staging: vt6656: rxtx.c [BUG] s_vGetFreeContext dead lock on null apTD. There seems to be race condition that the device is ndo_start_xmit at a point where the device is closing and apTD is NULL resulting in dead lock. Add a NULL check to apTD and return NULL to calling functions. This is more likely on 64 bit systems. Signed-off-by: Malcolm Priestley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/vt6656/rxtx.c b/drivers/staging/vt6656/rxtx.c index fb743a8..14f3e85 100644 --- a/drivers/staging/vt6656/rxtx.c +++ b/drivers/staging/vt6656/rxtx.c @@ -148,6 +148,8 @@ static void *s_vGetFreeContext(struct vnt_private *pDevice) DBG_PRT(MSG_LEVEL_DEBUG, KERN_INFO"GetFreeContext()\n"); for (ii = 0; ii < pDevice->cbTD; ii++) { + if (!pDevice->apTD[ii]) + return NULL; pContext = pDevice->apTD[ii]; if (pContext->bBoolInUse == false) { pContext->bBoolInUse = true; -- cgit v0.10.2 From e3eb270fab7734427dd8171a93e4946fe28674bc Mon Sep 17 00:00:00 2001 From: Malcolm Priestley Date: Sun, 22 Sep 2013 19:48:54 +0100 Subject: staging: vt6656: [BUG] main_usb.c oops on device_close move flag earlier. The vt6656 is prone to resetting on the usb bus. It seems there is a race condition and wpa supplicant is trying to open the device via iw_handlers before its actually closed at a stage that the buffers are being removed. The device is longer considered open when the buffers are being removed. So move ~DEVICE_FLAGS_OPENED flag to before freeing the device buffers. Signed-off-by: Malcolm Priestley Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/vt6656/main_usb.c b/drivers/staging/vt6656/main_usb.c index 5369717..6f9d281 100644 --- a/drivers/staging/vt6656/main_usb.c +++ b/drivers/staging/vt6656/main_usb.c @@ -1098,6 +1098,8 @@ static int device_close(struct net_device *dev) memset(pMgmt->abyCurrBSSID, 0, 6); pMgmt->eCurrState = WMAC_STATE_IDLE; + pDevice->flags &= ~DEVICE_FLAGS_OPENED; + device_free_tx_bufs(pDevice); device_free_rx_bufs(pDevice); device_free_int_bufs(pDevice); @@ -1109,7 +1111,6 @@ static int device_close(struct net_device *dev) usb_free_urb(pDevice->pInterruptURB); BSSvClearNodeDBTable(pDevice, 0); - pDevice->flags &=(~DEVICE_FLAGS_OPENED); DBG_PRT(MSG_LEVEL_DEBUG, KERN_INFO "device_close2 \n"); -- cgit v0.10.2 From 5e8c3d3e41b0bf241e830a1ee0752405adecc050 Mon Sep 17 00:00:00 2001 From: Malcolm Priestley Date: Mon, 23 Sep 2013 20:30:42 +0100 Subject: staging: vt6656: [BUG] iwctl_siwencodeext return if device not open Don't allow entry to iwctl_siwencodeext if device not open. This fixes a race condition where wpa supplicant/network manager enters the function when the device is already closed. Signed-off-by: Malcolm Priestley Cc: stable@vger.kernel.org # 3.8+ Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/vt6656/iwctl.c b/drivers/staging/vt6656/iwctl.c index d0cf7d8..8872e0f 100644 --- a/drivers/staging/vt6656/iwctl.c +++ b/drivers/staging/vt6656/iwctl.c @@ -1634,6 +1634,9 @@ int iwctl_siwencodeext(struct net_device *dev, struct iw_request_info *info, if (pMgmt == NULL) return -EFAULT; + if (!(pDevice->flags & DEVICE_FLAGS_OPENED)) + return -ENODEV; + buf = kzalloc(sizeof(struct viawget_wpa_param), GFP_KERNEL); if (buf == NULL) return -ENOMEM; -- cgit v0.10.2 From 099326d8cde12b4aa056fbb3ee05d78e5629df02 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Tue, 17 Sep 2013 14:53:23 -0300 Subject: staging: imx-drm: Fix probe failure Since commit b5dc0d10 (drm/imx: kill firstopen callback) the following probe failure is seen: [drm] Supports vblank timestamp caching Rev 1 (10.10.2010). [drm] No driver support for vblank timestamp query. [drm] Initialized imx-drm 1.0.0 20120507 on minor 0 imx-ldb ldb.10: adding encoder failed with -16 imx-ldb: probe of ldb.10 failed with error -16 imx-ipuv3 2400000.ipu: IPUv3H probed imx-ipuv3 2800000.ipu: IPUv3H probed imx-ipuv3-crtc imx-ipuv3-crtc.0: adding crtc failed with -16. imx-ipuv3-crtc: probe of imx-ipuv3-crtc.0 failed with error -16 imx-ipuv3-crtc imx-ipuv3-crtc.1: adding crtc failed with -16. imx-ipuv3-crtc: probe of imx-ipuv3-crtc.1 failed with error -16 imx-ipuv3-crtc imx-ipuv3-crtc.2: adding crtc failed with -16. imx-ipuv3-crtc: probe of imx-ipuv3-crtc.2 failed with error -16 imx-ipuv3-crtc imx-ipuv3-crtc.3: adding crtc failed with -16. imx-ipuv3-crtc: probe of imx-ipuv3-crtc.3 failed with error -16 The reason for the probe failure is that now 'imxdrm->references' is incremented early in imx_drm_driver_load(), so the following checks in imx_drm_add_crtc() and imx_drm_add_encoder(): if (imxdrm->references) { ret = -EBUSY; goto err_busy; } ,will always fail. Instead of manually keeping the references in the imx-drm driver, let's use drm->open_count. After this patch, lvds panel is functional on a mx6qsabrelite board. Signed-off-by: Fabio Estevam Acked-by: Sascha Hauer Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/imx-drm/imx-drm-core.c b/drivers/staging/imx-drm/imx-drm-core.c index 47c58884..a2e52a0 100644 --- a/drivers/staging/imx-drm/imx-drm-core.c +++ b/drivers/staging/imx-drm/imx-drm-core.c @@ -41,7 +41,6 @@ struct imx_drm_device { struct list_head encoder_list; struct list_head connector_list; struct mutex mutex; - int references; int pipes; struct drm_fbdev_cma *fbhelper; }; @@ -241,8 +240,6 @@ struct drm_device *imx_drm_device_get(void) } } - imxdrm->references++; - return imxdrm->drm; unwind_crtc: @@ -280,8 +277,6 @@ void imx_drm_device_put(void) list_for_each_entry(enc, &imxdrm->encoder_list, list) module_put(enc->owner); - imxdrm->references--; - mutex_unlock(&imxdrm->mutex); } EXPORT_SYMBOL_GPL(imx_drm_device_put); @@ -485,7 +480,7 @@ int imx_drm_add_crtc(struct drm_crtc *crtc, mutex_lock(&imxdrm->mutex); - if (imxdrm->references) { + if (imxdrm->drm->open_count) { ret = -EBUSY; goto err_busy; } @@ -564,7 +559,7 @@ int imx_drm_add_encoder(struct drm_encoder *encoder, mutex_lock(&imxdrm->mutex); - if (imxdrm->references) { + if (imxdrm->drm->open_count) { ret = -EBUSY; goto err_busy; } @@ -709,7 +704,7 @@ int imx_drm_add_connector(struct drm_connector *connector, mutex_lock(&imxdrm->mutex); - if (imxdrm->references) { + if (imxdrm->drm->open_count) { ret = -EBUSY; goto err_busy; } -- cgit v0.10.2 From 8937669fd637e2edff05973cb1fdee5f910cbcc7 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 24 Sep 2013 15:48:05 -0400 Subject: USB: iMX21: accept very late isochronous URBs Commit 24f531371de1 (USB: EHCI: accept very late isochronous URBs) changed the isochronous API provided by ehci-hcd. URBs submitted too late, so that the time slots for all their packets have already expired, are no longer rejected outright. Instead the submission is accepted, and the URB completes normally with a -EXDEV error for each packet. This is what client drivers expect. The same policy should be implemented in imx21-hcd, but I don't know enough about the hardware to do it. As a second-best substitute, this patch treats very late isochronous submissions as though the URB_ISO_ASAP flag were set. I don't have any way to test this change, unfortunately. Signed-off-by: Alan Stern CC: Sascha Hauer CC: Martin Fuzzey Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/imx21-hcd.c b/drivers/usb/host/imx21-hcd.c index 60a5de5..adb01d9 100644 --- a/drivers/usb/host/imx21-hcd.c +++ b/drivers/usb/host/imx21-hcd.c @@ -824,13 +824,13 @@ static int imx21_hc_urb_enqueue_isoc(struct usb_hcd *hcd, i = DIV_ROUND_UP(wrap_frame( cur_frame - urb->start_frame), urb->interval); - if (urb->transfer_flags & URB_ISO_ASAP) { + + /* Treat underruns as if URB_ISO_ASAP was set */ + if ((urb->transfer_flags & URB_ISO_ASAP) || + i >= urb->number_of_packets) { urb->start_frame = wrap_frame(urb->start_frame + i * urb->interval); i = 0; - } else if (i >= urb->number_of_packets) { - ret = -EXDEV; - goto alloc_dmem_failed; } } } -- cgit v0.10.2 From bef073b067a7b1874a6b381e0035bb0516d71a77 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 24 Sep 2013 15:47:20 -0400 Subject: USB: UHCI: accept very late isochronous URBs Commit 24f531371de1 (USB: EHCI: accept very late isochronous URBs) changed the isochronous API provided by ehci-hcd. URBs submitted too late, so that the time slots for all their packets have already expired, are no longer rejected outright. Instead the submission is accepted, and the URB completes normally with a -EXDEV error for each packet. This is what client drivers expect. This patch implements the same policy in uhci-hcd. It should be applied to all kernels containing commit c44b225077bb (UHCI: implement new semantics for URB_ISO_ASAP). Signed-off-by: Alan Stern CC: Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/uhci-q.c b/drivers/usb/host/uhci-q.c index 041c6dd..da6f56d 100644 --- a/drivers/usb/host/uhci-q.c +++ b/drivers/usb/host/uhci-q.c @@ -1303,7 +1303,7 @@ static int uhci_submit_isochronous(struct uhci_hcd *uhci, struct urb *urb, } /* Fell behind? */ - if (uhci_frame_before_eq(frame, next)) { + if (!uhci_frame_before_eq(next, frame)) { /* USB_ISO_ASAP: Round up to the first available slot */ if (urb->transfer_flags & URB_ISO_ASAP) @@ -1311,13 +1311,17 @@ static int uhci_submit_isochronous(struct uhci_hcd *uhci, struct urb *urb, -qh->period; /* - * Not ASAP: Use the next slot in the stream. If - * the entire URB falls before the threshold, fail. + * Not ASAP: Use the next slot in the stream, + * no matter what. */ else if (!uhci_frame_before_eq(next, frame + (urb->number_of_packets - 1) * qh->period)) - return -EXDEV; + dev_dbg(uhci_dev(uhci), "iso underrun %p (%u+%u < %u)\n", + urb, frame, + (urb->number_of_packets - 1) * + qh->period, + next); } } -- cgit v0.10.2 From a8693424c751b8247ee19bd8b857f1d4f432b972 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 24 Sep 2013 15:46:45 -0400 Subject: USB: OHCI: accept very late isochronous URBs Commit 24f531371de1 (USB: EHCI: accept very late isochronous URBs) changed the isochronous API provided by ehci-hcd. URBs submitted too late, so that the time slots for all their packets have already expired, are no longer rejected outright. Instead the submission is accepted, and the URB completes normally with a -EXDEV error for each packet. This is what client drivers expect. This patch implements the same policy in ohci-hcd. The change is more complicated than it was in ehci-hcd, because ohci-hcd doesn't scan for isochronous completions in the same way as ehci-hcd does. Rather, it depends on the hardware adding completed TDs to a "done queue". Some OHCI controller don't handle this properly when a TD's time slot has already expired, so we have to avoid adding such TDs to the schedule in the first place. As a result, if the URB was submitted too late then none of its TDs will get put on the schedule, so none of them will end up on the done queue, so the driver will never realize that the URB should be completed. To solve this problem, the patch adds one to urb_priv->td_cnt for such URBs, making it larger than urb_priv->length (td_cnt already gets set to the number of TD's that had to be skipped because their slots have expired). Each time an URB is given back, the finish_urb() routine looks to see if urb_priv->td_cnt for the next URB on the same endpoint is marked in this way. If so, it gives back the next URB right away. This should be applied to all kernels containing commit 815fa7b91761 (USB: OHCI: fix logic for scheduling isochronous URBs). Signed-off-by: Alan Stern CC: Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c index 8f6b695..604cad1 100644 --- a/drivers/usb/host/ohci-hcd.c +++ b/drivers/usb/host/ohci-hcd.c @@ -216,31 +216,26 @@ static int ohci_urb_enqueue ( frame &= ~(ed->interval - 1); frame |= ed->branch; urb->start_frame = frame; + ed->last_iso = frame + ed->interval * (size - 1); } } else if (ed->type == PIPE_ISOCHRONOUS) { u16 next = ohci_frame_no(ohci) + 1; u16 frame = ed->last_iso + ed->interval; + u16 length = ed->interval * (size - 1); /* Behind the scheduling threshold? */ if (unlikely(tick_before(frame, next))) { - /* USB_ISO_ASAP: Round up to the first available slot */ + /* URB_ISO_ASAP: Round up to the first available slot */ if (urb->transfer_flags & URB_ISO_ASAP) { frame += (next - frame + ed->interval - 1) & -ed->interval; /* - * Not ASAP: Use the next slot in the stream. If - * the entire URB falls before the threshold, fail. + * Not ASAP: Use the next slot in the stream, + * no matter what. */ } else { - if (tick_before(frame + ed->interval * - (urb->number_of_packets - 1), next)) { - retval = -EXDEV; - usb_hcd_unlink_urb_from_ep(hcd, urb); - goto fail; - } - /* * Some OHCI hardware doesn't handle late TDs * correctly. After retiring them it proceeds @@ -251,9 +246,16 @@ static int ohci_urb_enqueue ( urb_priv->td_cnt = DIV_ROUND_UP( (u16) (next - frame), ed->interval); + if (urb_priv->td_cnt >= urb_priv->length) { + ++urb_priv->td_cnt; /* Mark it */ + ohci_dbg(ohci, "iso underrun %p (%u+%u < %u)\n", + urb, frame, length, + next); + } } } urb->start_frame = frame; + ed->last_iso = frame + length; } /* fill the TDs and link them to the ed; and diff --git a/drivers/usb/host/ohci-q.c b/drivers/usb/host/ohci-q.c index df4a670..e7f577e 100644 --- a/drivers/usb/host/ohci-q.c +++ b/drivers/usb/host/ohci-q.c @@ -41,9 +41,13 @@ finish_urb(struct ohci_hcd *ohci, struct urb *urb, int status) __releases(ohci->lock) __acquires(ohci->lock) { - struct device *dev = ohci_to_hcd(ohci)->self.controller; + struct device *dev = ohci_to_hcd(ohci)->self.controller; + struct usb_host_endpoint *ep = urb->ep; + struct urb_priv *urb_priv; + // ASSERT (urb->hcpriv != 0); + restart: urb_free_priv (ohci, urb->hcpriv); urb->hcpriv = NULL; if (likely(status == -EINPROGRESS)) @@ -80,6 +84,21 @@ __acquires(ohci->lock) ohci->hc_control &= ~(OHCI_CTRL_PLE|OHCI_CTRL_IE); ohci_writel (ohci, ohci->hc_control, &ohci->regs->control); } + + /* + * An isochronous URB that is sumitted too late won't have any TDs + * (marked by the fact that the td_cnt value is larger than the + * actual number of TDs). If the next URB on this endpoint is like + * that, give it back now. + */ + if (!list_empty(&ep->urb_list)) { + urb = list_first_entry(&ep->urb_list, struct urb, urb_list); + urb_priv = urb->hcpriv; + if (urb_priv->td_cnt > urb_priv->length) { + status = 0; + goto restart; + } + } } @@ -546,7 +565,6 @@ td_fill (struct ohci_hcd *ohci, u32 info, td->hwCBP = cpu_to_hc32 (ohci, data & 0xFFFFF000); *ohci_hwPSWp(ohci, td, 0) = cpu_to_hc16 (ohci, (data & 0x0FFF) | 0xE000); - td->ed->last_iso = info & 0xffff; } else { td->hwCBP = cpu_to_hc32 (ohci, data); } @@ -996,7 +1014,7 @@ rescan_this: urb_priv->td_cnt++; /* if URB is done, clean up */ - if (urb_priv->td_cnt == urb_priv->length) { + if (urb_priv->td_cnt >= urb_priv->length) { modified = completed = 1; finish_urb(ohci, urb, 0); } @@ -1086,7 +1104,7 @@ static void takeback_td(struct ohci_hcd *ohci, struct td *td) urb_priv->td_cnt++; /* If all this urb's TDs are done, call complete() */ - if (urb_priv->td_cnt == urb_priv->length) + if (urb_priv->td_cnt >= urb_priv->length) finish_urb(ohci, urb, status); /* clean schedule: unlink EDs that are no longer busy */ -- cgit v0.10.2 From f875fdbf344b9fde207f66b392c40845dd7e5aa6 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 24 Sep 2013 15:45:25 -0400 Subject: USB: fix PM config symbol in uhci-hcd, ehci-hcd, and xhci-hcd Since uhci-hcd, ehci-hcd, and xhci-hcd support runtime PM, the .pm field in their pci_driver structures should be protected by CONFIG_PM rather than CONFIG_PM_SLEEP. The corresponding change has already been made for ohci-hcd. Without this change, controllers won't do runtime suspend if system suspend or hibernation isn't enabled. Signed-off-by: Alan Stern CC: Sarah Sharp CC: Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c index 6bd299e..854c2ec 100644 --- a/drivers/usb/host/ehci-pci.c +++ b/drivers/usb/host/ehci-pci.c @@ -361,7 +361,7 @@ static struct pci_driver ehci_pci_driver = { .remove = usb_hcd_pci_remove, .shutdown = usb_hcd_pci_shutdown, -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_PM .driver = { .pm = &usb_hcd_pci_pm_ops }, diff --git a/drivers/usb/host/uhci-pci.c b/drivers/usb/host/uhci-pci.c index c300bd2f7..0f228c4 100644 --- a/drivers/usb/host/uhci-pci.c +++ b/drivers/usb/host/uhci-pci.c @@ -293,7 +293,7 @@ static struct pci_driver uhci_pci_driver = { .remove = usb_hcd_pci_remove, .shutdown = uhci_shutdown, -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_PM .driver = { .pm = &usb_hcd_pci_pm_ops }, diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index c2d4950..236c3aa 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -351,7 +351,7 @@ static struct pci_driver xhci_pci_driver = { /* suspend and resume implemented later */ .shutdown = usb_hcd_pci_shutdown, -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_PM .driver = { .pm = &usb_hcd_pci_pm_ops }, -- cgit v0.10.2 From 7b0dd72a4401d999dac70377f5d423b8e6d44f4b Mon Sep 17 00:00:00 2001 From: Jean Pihet Date: Wed, 18 Sep 2013 20:55:09 +0200 Subject: arm: clocksource: mvebu: Use the main timer as clock source from DT This commit: 573145f08c2b92c45498468afbbba909f6ce6135 clocksource: armada-370-xp: Use CLOCKSOURCE_OF_DECLARE replaced a call to the driver's timer initialization by a call to clocksource_of_init(). However, it failed to select CONFIG_CLKSRC_OF. Fix this by selecting CONFIG_CLKSRC_OF for Armada370/XP machines. Without this change the kernel is stuck at: 'Calibrating delay loop...'. Signed-off-by: Jean Pihet Acked-by: Ezequiel Garcia Signed-off-by: Daniel Lezcano diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig index 41c6946..971d796 100644 --- a/drivers/clocksource/Kconfig +++ b/drivers/clocksource/Kconfig @@ -26,6 +26,7 @@ config DW_APB_TIMER_OF config ARMADA_370_XP_TIMER bool + select CLKSRC_OF config ORION_TIMER select CLKSRC_OF -- cgit v0.10.2 From 5df718d84679936454e815451d748ccca0e9edad Mon Sep 17 00:00:00 2001 From: Tomasz Figa Date: Wed, 25 Sep 2013 12:00:59 +0200 Subject: clocksource: exynos_mct: Set IRQ affinity when the CPU goes online Some variants of Exynos MCT, namely exynos4210-mct at the moment, use normal, shared interrupts for local timers. This means that each interrupt must have correct affinity set to fire only on CPU corresponding to given local timer. However after recent conversion of clocksource drivers to not use the local timer API for local timer initialization any more, the point of time when local timers get initialized changed and irq_set_affinity() fails because the CPU is not marked as online yet. This patch fixes this by moving the call to irq_set_affinity() to CPU_ONLINE notification, so the affinity is being set when the CPU goes online. This fixes a regression introduced by commit ee98d27df6 ARM: EXYNOS4: Divorce mct from local timer API which rendered all Exynos4210 based boards unbootable due to failing irq_set_affinity() making local timers inoperatible. Signed-off-by: Tomasz Figa Signed-off-by: Kyungmin Park Acked-by: Stephen Boyd Signed-off-by: Daniel Lezcano diff --git a/drivers/clocksource/exynos_mct.c b/drivers/clocksource/exynos_mct.c index 5b34768..62b0de6 100644 --- a/drivers/clocksource/exynos_mct.c +++ b/drivers/clocksource/exynos_mct.c @@ -428,7 +428,6 @@ static int exynos4_local_timer_setup(struct clock_event_device *evt) evt->irq); return -EIO; } - irq_set_affinity(evt->irq, cpumask_of(cpu)); } else { enable_percpu_irq(mct_irqs[MCT_L0_IRQ], 0); } @@ -449,6 +448,7 @@ static int exynos4_mct_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { struct mct_clock_event_device *mevt; + unsigned int cpu; /* * Grab cpu pointer in each case to avoid spurious @@ -459,6 +459,12 @@ static int exynos4_mct_cpu_notify(struct notifier_block *self, mevt = this_cpu_ptr(&percpu_mct_tick); exynos4_local_timer_setup(&mevt->evt); break; + case CPU_ONLINE: + cpu = (unsigned long)hcpu; + if (mct_int_type == MCT_INT_SPI) + irq_set_affinity(mct_irqs[MCT_L0_IRQ + cpu], + cpumask_of(cpu)); + break; case CPU_DYING: mevt = this_cpu_ptr(&percpu_mct_tick); exynos4_local_timer_stop(&mevt->evt); @@ -500,6 +506,8 @@ static void __init exynos4_timer_resources(struct device_node *np, void __iomem &percpu_mct_tick); WARN(err, "MCT: can't request IRQ %d (%d)\n", mct_irqs[MCT_L0_IRQ], err); + } else { + irq_set_affinity(mct_irqs[MCT_L0_IRQ], cpumask_of(0)); } err = register_cpu_notifier(&exynos4_mct_cpu_nb); -- cgit v0.10.2 From eeb93d02c5d8819c74f6d4da08325cff8c6a9376 Mon Sep 17 00:00:00 2001 From: Sebastian Hesselbarth Date: Thu, 26 Sep 2013 02:08:17 +0200 Subject: clocksource: of: Respect device tree node status Clocksource devices provided by DT can be disabled (status != "okay"). Instead of registering clocksource drivers for disabled nodes, respect the device's status by skiping disabled nodes. Signed-off-by: Sebastian Hesselbarth Signed-off-by: Daniel Lezcano diff --git a/drivers/clocksource/clksrc-of.c b/drivers/clocksource/clksrc-of.c index 37f5325..b9ddd9e 100644 --- a/drivers/clocksource/clksrc-of.c +++ b/drivers/clocksource/clksrc-of.c @@ -30,6 +30,9 @@ void __init clocksource_of_init(void) clocksource_of_init_fn init_func; for_each_matching_node_and_match(np, __clksrc_of_table, &match) { + if (!of_device_is_available(np)) + continue; + init_func = match->data; init_func(np); } -- cgit v0.10.2 From f84839daa793f36ec7f8795d7510f0d61fa8875a Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Tue, 17 Sep 2013 12:37:20 +0800 Subject: usb: chipidea: udc: fix the oops after rmmod gadget When we rmmod gadget, the ci->driver needs to be cleared. Otherwise, when we plug in usb cable again, the driver will consider gadget is there, and go to enumeration procedure, but in fact, it was removed. ci_hdrc ci_hdrc.0: Connected to host Unable to handle kernel paging request at virtual address 7f02a42c pgd = 80004000 [7f02a42c] *pgd=3f13d811, *pte=00000000, *ppte=00000000 Internal error: Oops: 7 [#1] SMP ARM Modules linked in: usb_f_acm u_serial libcomposite configfs [last unloaded: g_serial] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.10.0+ #42 task: 807dba88 ti: 807d0000 task.ti: 807d0000 PC is at udc_irq+0x8fc/0xea4 LR is at l2x0_cache_sync+0x5c/0x6c pc : [<803de7f4>] lr : [<8001d0f0>] psr: 20000193 sp : 807d1d98 ip : 807d1d80 fp : 807d1df4 r10: af809900 r9 : 808184d4 r8 : 00080001 r7 : 00082001 r6 : afb711f8 r5 : afb71010 r4 : ffffffea r3 : 7f02a41c r2 : afb71010 r1 : 807d1dc0 r0 : afb71068 Flags: nzCv IRQs off FIQs on Mode SVC_32 ISA ARM Segment kernel Control: 10c53c7d Table: 3f01804a DAC: 00000017 Process swapper/0 (pid: 0, stack limit = 0x807d0238) Stack: (0x807d1d98 to 0x807d2000) 1d80: 00000000 afb71014 1da0: 000040f6 00000000 00000001 00000000 00007530 00000000 afb71010 001dcd65 1dc0: 01000680 00400000 807d1e2c afb71010 0000004e 00000000 00000000 0000004b 1de0: 808184d4 af809900 807d1e0c 807d1df8 803dbc24 803ddf04 afba75c0 0000004e 1e00: 807d1e44 807d1e10 8007a19c 803dbb9c 8108e7e0 8108e7e0 9ceddce0 af809900 1e20: 0000004e 807d0000 0000004b 00000000 00000010 00000000 807d1e5c 807d1e48 1e40: 8007a334 8007a154 af809900 0000004e 807d1e74 807d1e60 8007d3b4 8007a2f0 1e60: 0000004b 807cce3c 807d1e8c 807d1e78 80079b08 8007d300 00000180 807d8ba0 1e80: 807d1eb4 807d1e90 8000eef4 80079aec 00000000 f400010c 807d8ce4 807d1ed8 1ea0: f4000100 96d5c75d 807d1ed4 807d1eb8 80008600 8000eeac 8042699c 60000013 1ec0: ffffffff 807d1f0c 807d1f54 807d1ed8 8000e180 800085dc 807d1f20 00000046 1ee0: 9cedd275 00000010 8108f080 807de294 00000001 807de248 96d5c75d 00000010 1f00: 00000000 807d1f54 00000000 807d1f20 8005ff54 8042699c 60000013 ffffffff 1f20: 9cedd275 00000010 00000005 8108f080 8108f080 00000001 807de248 8086bd00 1f40: 807d0000 00000001 807d1f7c 807d1f58 80426af0 80426950 807d0000 00000000 1f60: 808184c0 808184c0 807d8954 805b886c 807d1f8c 807d1f80 8000f294 80426a44 1f80: 807d1fac 807d1f90 8005f110 8000f288 807d1fac 807d8908 805b4748 807dc86c 1fa0: 807d1fbc 807d1fb0 805aa58c 8005f068 807d1ff4 807d1fc0 8077c860 805aa530 1fc0: ffffffff ffffffff 8077c330 00000000 00000000 807bef88 00000000 10c53c7d 1fe0: 807d88d0 807bef84 00000000 807d1ff8 10008074 8077c594 00000000 00000000 Backtrace: [<803ddef8>] (udc_irq+0x0/0xea4) from [<803dbc24>] (ci_irq+0x94/0x14c) [<803dbb90>] (ci_irq+0x0/0x14c) from [<8007a19c>] (handle_irq_event_percpu+0x54/0x19c) r5:0000004e r4:afba75c0 [<8007a148>] (handle_irq_event_percpu+0x0/0x19c) from [<8007a334>] (handle_irq_event+0x50/0x70) [<8007a2e4>] (handle_irq_event+0x0/0x70) from [<8007d3b4>] (handle_fasteoi_irq+0xc0/0x16c) r5:0000004e r4:af809900 [<8007d2f4>] (handle_fasteoi_irq+0x0/0x16c) from [<80079b08>] (generic_handle_irq+0x28/0x38) r5:807cce3c r4:0000004b [<80079ae0>] (generic_handle_irq+0x0/0x38) from [<8000eef4>] (handle_IRQ+0x54/0xb4) r4:807d8ba0 r3:00000180 [<8000eea0>] (handle_IRQ+0x0/0xb4) from [<80008600>] (gic_handle_irq+0x30/0x64) r8:96d5c75d r7:f4000100 r6:807d1ed8 r5:807d8ce4 r4:f400010c r3:00000000 [<800085d0>] (gic_handle_irq+0x0/0x64) from [<8000e180>] (__irq_svc+0x40/0x54) Exception stack(0x807d1ed8 to 0x807d1f20) 1ec0: 807d1f20 00000046 1ee0: 9cedd275 00000010 8108f080 807de294 00000001 807de248 96d5c75d 00000010 1f00: 00000000 807d1f54 00000000 807d1f20 8005ff54 8042699c 60000013 ffffffff r7:807d1f0c r6:ffffffff r5:60000013 r4:8042699c [<80426944>] (cpuidle_enter_state+0x0/0xf4) from [<80426af0>] (cpuidle_idle_call+0xb8/0x174) r9:00000001 r8:807d0000 r7:8086bd00 r6:807de248 r5:00000001 r4:8108f080 [<80426a38>] (cpuidle_idle_call+0x0/0x174) from [<8000f294>] (arch_cpu_idle+0x18/0x5c) [<8000f27c>] (arch_cpu_idle+0x0/0x5c) from [<8005f110>] (cpu_startup_entry+0xb4/0x148) [<8005f05c>] (cpu_startup_entry+0x0/0x148) from [<805aa58c>] (rest_init+0x68/0x80) r7:807dc86c [<805aa524>] (rest_init+0x0/0x80) from [<8077c860>] (start_kernel+0x2d8/0x334) [<8077c588>] (start_kernel+0x0/0x334) from [<10008074>] (0x10008074) Code: e59031e0 e51b203c e24b1034 e2820058 (e5933010) ---[ end trace f874b2c5533c04bc ]--- Kernel panic - not syncing: Fatal exception in interrupt Tested-by: Marek Vasut Acked-by: Shawn Guo Signed-off-by: Peter Chen Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/chipidea/udc.c b/drivers/usb/chipidea/udc.c index 6b4c2f2..5280454 100644 --- a/drivers/usb/chipidea/udc.c +++ b/drivers/usb/chipidea/udc.c @@ -1667,13 +1667,13 @@ static int ci_udc_stop(struct usb_gadget *gadget, if (ci->platdata->notify_event) ci->platdata->notify_event(ci, CI_HDRC_CONTROLLER_STOPPED_EVENT); - ci->driver = NULL; spin_unlock_irqrestore(&ci->lock, flags); _gadget_stop_activity(&ci->gadget); spin_lock_irqsave(&ci->lock, flags); pm_runtime_put(&ci->gadget.dev); } + ci->driver = NULL; spin_unlock_irqrestore(&ci->lock, flags); return 0; -- cgit v0.10.2 From 222bed9b2dff53e4402d4819d6fca09e26b765fb Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Tue, 17 Sep 2013 12:37:21 +0800 Subject: usb: chipidea: Fix memleak for ci->hw_bank.regmap when removal It needs to free ci->hw_bank.regmap explicitly since it is not managed resource. Signed-off-by: Peter Chen Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/chipidea/core.c b/drivers/usb/chipidea/core.c index 9462640..23763dc 100644 --- a/drivers/usb/chipidea/core.c +++ b/drivers/usb/chipidea/core.c @@ -605,6 +605,7 @@ static int ci_hdrc_remove(struct platform_device *pdev) dbg_remove_files(ci); free_irq(ci->irq, ci); ci_role_destroy(ci); + kfree(ci->hw_bank.regmap); return 0; } -- cgit v0.10.2 From 3a254fea70f402859b92a9cd4299ee5de3bbc2f6 Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Tue, 17 Sep 2013 12:37:23 +0800 Subject: usb: chipidea: imx: Add usb_phy_shutdown at probe's error path If not, the PHY will be active even the controller is not in use. We find this issue due to the PHY's clock refcount is not correct due to -EPROBE_DEFER return after phy's init. Signed-off-by: Peter Chen Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/chipidea/ci_hdrc_imx.c b/drivers/usb/chipidea/ci_hdrc_imx.c index 74d998d..be822a2 100644 --- a/drivers/usb/chipidea/ci_hdrc_imx.c +++ b/drivers/usb/chipidea/ci_hdrc_imx.c @@ -131,7 +131,7 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev) if (ret) { dev_err(&pdev->dev, "usbmisc init failed, ret=%d\n", ret); - goto err_clk; + goto err_phy; } } @@ -143,7 +143,7 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev) dev_err(&pdev->dev, "Can't register ci_hdrc platform device, err=%d\n", ret); - goto err_clk; + goto err_phy; } if (data->usbmisc_data) { @@ -164,6 +164,9 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev) disable_device: ci_hdrc_remove_device(data->ci_pdev); +err_phy: + if (data->phy) + usb_phy_shutdown(data->phy); err_clk: clk_disable_unprepare(data->clk); return ret; -- cgit v0.10.2 From e7ef5265b158ac1975b1556511e2b11bad5f5522 Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Tue, 17 Sep 2013 12:37:24 +0800 Subject: usb: chipidea: udc: free pending TD at removal procedure There is a pending TD which is not freed after request finishes, we do this due to a controller bug. This TD needs to be freed when the driver is removed. It prints below error message when unload chipidea driver at current code: "ci_hdrc ci_hdrc.0: dma_pool_destroy ci_hw_td, b0001000 busy" It indicates the buffer at dma pool are still in use. This commit will free the pending TD at driver's removal procedure, it can fix the problem described above. Acked-by: Michael Grzeschik Signed-off-by: Peter Chen Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/chipidea/udc.c b/drivers/usb/chipidea/udc.c index 5280454..9333083 100644 --- a/drivers/usb/chipidea/udc.c +++ b/drivers/usb/chipidea/udc.c @@ -1600,6 +1600,8 @@ static void destroy_eps(struct ci_hdrc *ci) for (i = 0; i < ci->hw_ep_max; i++) { struct ci_hw_ep *hwep = &ci->ci_hw_ep[i]; + if (hwep->pending_td) + free_pending_td(hwep); dma_pool_free(ci->qh_pool, hwep->qh.ptr, hwep->qh.dma); } } -- cgit v0.10.2 From 2c740336159d9785ae81e8f9d6ba1a6b922723c9 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 17 Sep 2013 12:37:25 +0800 Subject: usb: chipidea: USB_CHIPIDEA should depend on HAS_DMA If NO_DMA=y: drivers/built-in.o: In function `dma_set_coherent_mask': include/linux/dma-mapping.h:93: undefined reference to `dma_supported' Reviewed-and-tested-by: Peter Chen Signed-off-by: Geert Uytterhoeven Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/chipidea/Kconfig b/drivers/usb/chipidea/Kconfig index 4a851e1..77b47d8 100644 --- a/drivers/usb/chipidea/Kconfig +++ b/drivers/usb/chipidea/Kconfig @@ -1,6 +1,6 @@ config USB_CHIPIDEA tristate "ChipIdea Highspeed Dual Role Controller" - depends on (USB_EHCI_HCD && USB_GADGET) || (USB_EHCI_HCD && !USB_GADGET) || (!USB_EHCI_HCD && USB_GADGET) + depends on ((USB_EHCI_HCD && USB_GADGET) || (USB_EHCI_HCD && !USB_GADGET) || (!USB_EHCI_HCD && USB_GADGET)) && HAS_DMA help Say Y here if your system has a dual role high speed USB controller based on ChipIdea silicon IP. Currently, only the -- cgit v0.10.2 From 831abf76643555a99b80a3b54adfa7e4fa0a3259 Mon Sep 17 00:00:00 2001 From: Kurt Garloff Date: Tue, 24 Sep 2013 14:13:48 +0200 Subject: usb/core/devio.c: Don't reject control message to endpoint with wrong direction bit Trying to read data from the Pegasus Technologies NoteTaker (0e20:0101) [1] with the Windows App (EasyNote) works natively but fails when Windows is running under KVM (and the USB device handed to KVM). The reason is a USB control message usb 4-2.2: control urb: bRequestType=22 bRequest=09 wValue=0200 wIndex=0001 wLength=0008 This goes to endpoint address 0x01 (wIndex); however, endpoint address 0x01 does not exist. There is an endpoint 0x81 though (same number, but other direction); the app may have meant that endpoint instead. The kernel thus rejects the IO and thus we see the failure. Apparently, Linux is more strict here than Windows ... we can't change the Win app easily, so that's a problem. It seems that the Win app/driver is buggy here and the driver does not behave fully according to the USB HID class spec that it claims to belong to. The device seems to happily deal with that though (and seems to not really care about this value much). So the question is whether the Linux kernel should filter here. Rejecting has the risk that somewhat non-compliant userspace apps/ drivers (most likely in a virtual machine) are prevented from working. Not rejecting has the risk of confusing an overly sensitive device with such a transfer. Given the fact that Windows does not filter it makes this risk rather small though. The patch makes the kernel more tolerant: If the endpoint address in wIndex does not exist, but an endpoint with toggled direction bit does, it will let the transfer through. (It does NOT change the message.) With attached patch, the app in Windows in KVM works. usb 4-2.2: check_ctrlrecip: process 13073 (qemu-kvm) requesting ep 01 but needs 81 I suspect this will mostly affect apps in virtual environments; as on Linux the apps would have been adapted to the stricter handling of the kernel. I have done that for mine[2]. [1] http://www.pegatech.com/ [2] https://sourceforge.net/projects/notetakerpen/ Signed-off-by: Kurt Garloff Acked-by: Alan Stern Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c index 737e3c1..71dc5d7 100644 --- a/drivers/usb/core/devio.c +++ b/drivers/usb/core/devio.c @@ -742,6 +742,22 @@ static int check_ctrlrecip(struct dev_state *ps, unsigned int requesttype, if ((index & ~USB_DIR_IN) == 0) return 0; ret = findintfep(ps->dev, index); + if (ret < 0) { + /* + * Some not fully compliant Win apps seem to get + * index wrong and have the endpoint number here + * rather than the endpoint address (with the + * correct direction). Win does let this through, + * so we'll not reject it here but leave it to + * the device to not break KVM. But we warn. + */ + ret = findintfep(ps->dev, index ^ 0x80); + if (ret >= 0) + dev_info(&ps->dev->dev, + "%s: process %i (%s) requesting ep %02x but needs %02x\n", + __func__, task_pid_nr(current), + current->comm, index, index ^ 0x80); + } if (ret >= 0) ret = checkintf(ps, ret); break; -- cgit v0.10.2 From 2199a5574b6d94b9ca26c6345356f45ec60fef8b Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 18 Sep 2013 15:01:16 -0500 Subject: clocksource: em_sti: Set cpu_possible_mask to fix SMP broadcast Update the STI driver by setting cpu_possible_mask to make EMEV2 SMP work as expected together with the ARM broadcast timer. This breakage was introduced by: f7db706 ARM: 7674/1: smp: Avoid dummy clockevent being preferred over real hardware clock-event Without this fix SMP operation is broken on EMEV2 since no broadcast timer interrupts trigger on the secondary CPU cores. Signed-off-by: Magnus Damm Tested-by: Simon Horman Reviewed-by: Stephen Boyd Signed-off-by: Simon Horman Signed-off-by: Daniel Lezcano diff --git a/drivers/clocksource/em_sti.c b/drivers/clocksource/em_sti.c index b9c81b7..3a5909c 100644 --- a/drivers/clocksource/em_sti.c +++ b/drivers/clocksource/em_sti.c @@ -301,7 +301,7 @@ static void em_sti_register_clockevent(struct em_sti_priv *p) ced->name = dev_name(&p->pdev->dev); ced->features = CLOCK_EVT_FEAT_ONESHOT; ced->rating = 200; - ced->cpumask = cpumask_of(0); + ced->cpumask = cpu_possible_mask; ced->set_next_event = em_sti_clock_event_next; ced->set_mode = em_sti_clock_event_mode; -- cgit v0.10.2 From 5cec7bf699c61d14f0538345076480bb8c8ebfbb Mon Sep 17 00:00:00 2001 From: Peter Hurley Date: Wed, 25 Sep 2013 20:13:04 -0400 Subject: tty: Fix SIGTTOU not sent with tcflush() Commit 'e7f3880cd9b98c5bf9391ae7acdec82b75403776' tty: Fix recursive deadlock in tty_perform_flush() introduced a regression where tcflush() does not generate SIGTTOU for background process groups. Make sure ioctl(TCFLSH) calls tty_check_change() when invoked from the line discipline. Cc: stable@vger.kernel.org # v3.10+ Reported-by: Oleg Nesterov Signed-off-by: Peter Hurley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/tty/tty_ioctl.c b/drivers/tty/tty_ioctl.c index 03ba081..6fd60fe 100644 --- a/drivers/tty/tty_ioctl.c +++ b/drivers/tty/tty_ioctl.c @@ -1201,6 +1201,9 @@ int n_tty_ioctl_helper(struct tty_struct *tty, struct file *file, } return 0; case TCFLSH: + retval = tty_check_change(tty); + if (retval) + return retval; return __tty_perform_flush(tty, arg); default: /* Try the mode commands */ -- cgit v0.10.2 From b26d19e44adfd10b691bf4ffd50ed411c1be9317 Mon Sep 17 00:00:00 2001 From: Liam Girdwood Date: Fri, 13 Sep 2013 17:43:16 +0100 Subject: ALSA: compress: Make sure we trigger STOP before closing the stream. Currently we assume that userspace will shut down the compressed stream correctly. However, if userspcae dies (e.g. cplay & ctrl-C) we dont stop the stream before freeing it. This now checks that the stream is stopped before freeing. Signed-off-by: Liam Girdwood Signed-off-by: Takashi Iwai diff --git a/sound/core/compress_offload.c b/sound/core/compress_offload.c index 5863ba6..bea523a 100644 --- a/sound/core/compress_offload.c +++ b/sound/core/compress_offload.c @@ -139,6 +139,18 @@ static int snd_compr_open(struct inode *inode, struct file *f) static int snd_compr_free(struct inode *inode, struct file *f) { struct snd_compr_file *data = f->private_data; + struct snd_compr_runtime *runtime = data->stream.runtime; + + switch (runtime->state) { + case SNDRV_PCM_STATE_RUNNING: + case SNDRV_PCM_STATE_DRAINING: + case SNDRV_PCM_STATE_PAUSED: + data->stream.ops->trigger(&data->stream, SNDRV_PCM_TRIGGER_STOP); + break; + default: + break; + } + data->stream.ops->free(&data->stream); kfree(data->stream.runtime->buffer); kfree(data->stream.runtime); -- cgit v0.10.2 From f82d7d16aee5eb4c2315ba11a90f2f3c662d45b8 Mon Sep 17 00:00:00 2001 From: Mengdong Lin Date: Sat, 21 Sep 2013 20:34:45 -0400 Subject: ALSA : hda - not use assigned converters for all unused pins BIOS can mark a pin as "no physical connection" if the port is used by an integrated display which is not audio capable. And audio driver will overlook such pins. On Haswell, such a disconneted pin will keep muted and connected to the 1st converter by default. But if the 1st convertor is assigned to a connected pin for audio streaming. The muted disconnected pin can make the connected pin no sound output. So this patch avoids using assigned converters for all unused pins for Haswell, including the disconected pins. Signed-off-by: Mengdong Lin Reviewed-by: David Henningsson Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_hdmi.c b/sound/pci/hda/patch_hdmi.c index 3d8cd044..7ea0245 100644 --- a/sound/pci/hda/patch_hdmi.c +++ b/sound/pci/hda/patch_hdmi.c @@ -1149,32 +1149,43 @@ static int hdmi_choose_cvt(struct hda_codec *codec, } static void haswell_config_cvts(struct hda_codec *codec, - int pin_id, int mux_id) + hda_nid_t pin_nid, int mux_idx) { struct hdmi_spec *spec = codec->spec; - struct hdmi_spec_per_pin *per_pin; - int pin_idx, mux_idx; - int curr; - int err; + hda_nid_t nid, end_nid; + int cvt_idx, curr; + struct hdmi_spec_per_cvt *per_cvt; - for (pin_idx = 0; pin_idx < spec->num_pins; pin_idx++) { - per_pin = get_pin(spec, pin_idx); + /* configure all pins, including "no physical connection" ones */ + end_nid = codec->start_nid + codec->num_nodes; + for (nid = codec->start_nid; nid < end_nid; nid++) { + unsigned int wid_caps = get_wcaps(codec, nid); + unsigned int wid_type = get_wcaps_type(wid_caps); - if (pin_idx == pin_id) + if (wid_type != AC_WID_PIN) continue; - curr = snd_hda_codec_read(codec, per_pin->pin_nid, 0, + if (nid == pin_nid) + continue; + + curr = snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_CONNECT_SEL, 0); + if (curr != mux_idx) + continue; - /* Choose another unused converter */ - if (curr == mux_id) { - err = hdmi_choose_cvt(codec, pin_idx, NULL, &mux_idx); - if (err < 0) - return; - snd_printdd("HDMI: choose converter %d for pin %d\n", mux_idx, pin_idx); - snd_hda_codec_write_cache(codec, per_pin->pin_nid, 0, + /* choose an unassigned converter. The conveters in the + * connection list are in the same order as in the codec. + */ + for (cvt_idx = 0; cvt_idx < spec->num_cvts; cvt_idx++) { + per_cvt = get_cvt(spec, cvt_idx); + if (!per_cvt->assigned) { + snd_printdd("choose cvt %d for pin nid %d\n", + cvt_idx, nid); + snd_hda_codec_write_cache(codec, nid, 0, AC_VERB_SET_CONNECT_SEL, - mux_idx); + cvt_idx); + break; + } } } } @@ -1216,7 +1227,7 @@ static int hdmi_pcm_open(struct hda_pcm_stream *hinfo, /* configure unused pins to choose other converters */ if (is_haswell(codec)) - haswell_config_cvts(codec, pin_idx, mux_idx); + haswell_config_cvts(codec, per_pin->pin_nid, mux_idx); snd_hda_spdif_ctls_assign(codec, pin_idx, per_cvt->cvt_nid); -- cgit v0.10.2 From 633358e9ff19145fde81903a8310ecc252dc8bca Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Wed, 25 Sep 2013 12:31:01 +0100 Subject: OMAPDSS: Add missing dependency on backlight for DSI-CM panel drier The DSI-CM driver uses the backlight class so needs to build depend on it. Signed-off-by: Mark Brown Reviewed-by: Jingoo Han Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/omap2/displays-new/Kconfig b/drivers/video/omap2/displays-new/Kconfig index 6c90885..10b25e7 100644 --- a/drivers/video/omap2/displays-new/Kconfig +++ b/drivers/video/omap2/displays-new/Kconfig @@ -35,6 +35,7 @@ config DISPLAY_PANEL_DPI config DISPLAY_PANEL_DSI_CM tristate "Generic DSI Command Mode Panel" + depends on BACKLIGHT_CLASS_DEVICE help Driver for generic DSI command mode panels. -- cgit v0.10.2 From fb8a15079a579f4cd0e01c9d37e868f5c2b4994c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Mon, 23 Sep 2013 18:13:10 +0200 Subject: video: mmp: drop needless devm cleanup MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The nice thing about devm_* is that the driver doesn't need to free the resources but the driver core takes care about that. This also simplifies the error path quite a bit and removes the wrong check for a clock pointer being NULL. Russell King - ARM Linux : "And this patch also fixes the above: disabling/unpreparing _after_ putting the thing - which was quite silly... :)" Signed-off-by: Uwe Kleine-König Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/mmp/hw/mmp_ctrl.c b/drivers/video/mmp/hw/mmp_ctrl.c index 75dca19..6ac7552 100644 --- a/drivers/video/mmp/hw/mmp_ctrl.c +++ b/drivers/video/mmp/hw/mmp_ctrl.c @@ -514,7 +514,7 @@ static int mmphw_probe(struct platform_device *pdev) if (IS_ERR(ctrl->clk)) { dev_err(ctrl->dev, "unable to get clk %s\n", mi->clk_name); ret = -ENOENT; - goto failed_get_clk; + goto failed; } clk_prepare_enable(ctrl->clk); @@ -551,21 +551,8 @@ failed_path_init: path_deinit(path_plat); } - if (ctrl->clk) { - devm_clk_put(ctrl->dev, ctrl->clk); - clk_disable_unprepare(ctrl->clk); - } -failed_get_clk: - devm_free_irq(ctrl->dev, ctrl->irq, ctrl); + clk_disable_unprepare(ctrl->clk); failed: - if (ctrl) { - if (ctrl->reg_base) - devm_iounmap(ctrl->dev, ctrl->reg_base); - devm_release_mem_region(ctrl->dev, res->start, - resource_size(res)); - devm_kfree(ctrl->dev, ctrl); - } - dev_err(&pdev->dev, "device init failed\n"); return ret; -- cgit v0.10.2 From a2a6fc5f182fdb179a4f5262ee77c2c2684fe0cd Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Mon, 23 Sep 2013 23:00:15 +0800 Subject: s3fb: fix error return code in s3_pci_probe() Fix to return -EINVAL when virtual vertical size smaller than real instead of 0, as done elsewhere in this function. Also remove dup code. Signed-off-by: Wei Yongjun Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/s3fb.c b/drivers/video/s3fb.c index 47ca86c..d838ba8 100644 --- a/drivers/video/s3fb.c +++ b/drivers/video/s3fb.c @@ -1336,14 +1336,7 @@ static int s3_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) (info->var.bits_per_pixel * info->var.xres_virtual); if (info->var.yres_virtual < info->var.yres) { dev_err(info->device, "virtual vertical size smaller than real\n"); - goto err_find_mode; - } - - /* maximize virtual vertical size for fast scrolling */ - info->var.yres_virtual = info->fix.smem_len * 8 / - (info->var.bits_per_pixel * info->var.xres_virtual); - if (info->var.yres_virtual < info->var.yres) { - dev_err(info->device, "virtual vertical size smaller than real\n"); + rc = -EINVAL; goto err_find_mode; } -- cgit v0.10.2 From 8c42cf75490a90f6016a027765a4f053d7ba2f37 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Mon, 23 Sep 2013 23:00:34 +0800 Subject: neofb: fix error return code in neofb_probe() Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function. Signed-off-by: Wei Yongjun Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c index 7ef079c..c172a52 100644 --- a/drivers/video/neofb.c +++ b/drivers/video/neofb.c @@ -2075,6 +2075,7 @@ static int neofb_probe(struct pci_dev *dev, const struct pci_device_id *id) if (!fb_find_mode(&info->var, info, mode_option, NULL, 0, info->monspecs.modedb, 16)) { printk(KERN_ERR "neofb: Unable to find usable video mode.\n"); + err = -EINVAL; goto err_map_video; } @@ -2097,7 +2098,8 @@ static int neofb_probe(struct pci_dev *dev, const struct pci_device_id *id) info->fix.smem_len >> 10, info->var.xres, info->var.yres, h_sync / 1000, h_sync % 1000, v_sync); - if (fb_alloc_cmap(&info->cmap, 256, 0) < 0) + err = fb_alloc_cmap(&info->cmap, 256, 0); + if (err < 0) goto err_map_video; err = register_framebuffer(info); -- cgit v0.10.2 From 60119a1d686702504fa4cbe666cbcb96744b175c Mon Sep 17 00:00:00 2001 From: Andrzej Hajda Date: Wed, 25 Sep 2013 13:51:31 +0200 Subject: video: of: display_timing: correct display-timings node finding of_get_display_timing(s) use of_find_node_by_name to get child node, this is incorrect, of_get_child_by_name should be used instead. The patch fixes it. Small typo is also corrected. Signed-off-by: Andrzej Hajda Signed-off-by: Kyungmin Park Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/of_display_timing.c b/drivers/video/of_display_timing.c index 171821d..ba5b40f 100644 --- a/drivers/video/of_display_timing.c +++ b/drivers/video/of_display_timing.c @@ -120,7 +120,7 @@ int of_get_display_timing(struct device_node *np, const char *name, return -EINVAL; } - timing_np = of_find_node_by_name(np, name); + timing_np = of_get_child_by_name(np, name); if (!timing_np) { pr_err("%s: could not find node '%s'\n", of_node_full_name(np), name); @@ -143,11 +143,11 @@ struct display_timings *of_get_display_timings(struct device_node *np) struct display_timings *disp; if (!np) { - pr_err("%s: no devicenode given\n", of_node_full_name(np)); + pr_err("%s: no device node given\n", of_node_full_name(np)); return NULL; } - timings_np = of_find_node_by_name(np, "display-timings"); + timings_np = of_get_child_by_name(np, "display-timings"); if (!timings_np) { pr_err("%s: could not find display-timings node\n", of_node_full_name(np)); -- cgit v0.10.2 From 5f4513864304672e6ea9eac60583eeac32e679f2 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Thu, 26 Sep 2013 08:33:16 +0200 Subject: hwmon: (applesmc) Check key count before proceeding After reports from Chris and Josh Boyer of a rare crash in applesmc, Guenter pointed at the initialization problem fixed below. The patch has not been verified to fix the crash, but should be applied regardless. Reported-by: Suggested-by: Guenter Roeck Signed-off-by: Henrik Rydberg Cc: stable@vger.kernel.org Signed-off-by: Guenter Roeck diff --git a/drivers/hwmon/applesmc.c b/drivers/hwmon/applesmc.c index 62c2e32..98814d1 100644 --- a/drivers/hwmon/applesmc.c +++ b/drivers/hwmon/applesmc.c @@ -525,16 +525,25 @@ static int applesmc_init_smcreg_try(void) { struct applesmc_registers *s = &smcreg; bool left_light_sensor, right_light_sensor; + unsigned int count; u8 tmp[1]; int ret; if (s->init_complete) return 0; - ret = read_register_count(&s->key_count); + ret = read_register_count(&count); if (ret) return ret; + if (s->cache && s->key_count != count) { + pr_warn("key count changed from %d to %d\n", + s->key_count, count); + kfree(s->cache); + s->cache = NULL; + } + s->key_count = count; + if (!s->cache) s->cache = kcalloc(s->key_count, sizeof(*s->cache), GFP_KERNEL); if (!s->cache) -- cgit v0.10.2 From 6d0bb818041a02be682abadb3ba35ff608f7d60a Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Thu, 26 Sep 2013 13:13:58 +0200 Subject: video: mxsfb: Add missing break Add missing break into the restore function. Signed-off-by: Marek Vasut Cc: Fabio Estevam Cc: Shawn Guo Cc: Tomi Valkeinen Signed-off-by: Tomi Valkeinen diff --git a/drivers/video/mxsfb.c b/drivers/video/mxsfb.c index d250ed0..27197a8 100644 --- a/drivers/video/mxsfb.c +++ b/drivers/video/mxsfb.c @@ -620,6 +620,7 @@ static int mxsfb_restore_mode(struct mxsfb_info *host) break; case 3: bits_per_pixel = 32; + break; case 1: default: return -EINVAL; -- cgit v0.10.2 From 997def25e4b9cee3b01609e18a52f926bca8bd2b Mon Sep 17 00:00:00 2001 From: Mark Tinguely Date: Mon, 23 Sep 2013 12:18:58 -0500 Subject: xfs: fix node forward in xfs_node_toosmall Commit f5ea1100 cleans up the disk to host conversions for node directory entries, but because a variable is reused in xfs_node_toosmall() the next node is not correctly found. If the original node is small enough (<= 3/8 of the node size), this change may incorrectly cause a node collapse when it should not. That will cause an assert in xfstest generic/319: Assertion failed: first <= last && last < BBTOB(bp->b_length), file: /root/newest/xfs/fs/xfs/xfs_trans_buf.c, line: 569 Keep the original node header to get the correct forward node. (When a node is considered for a merge with a sibling, it overwrites the sibling pointers of the original incore nodehdr with the sibling's pointers. This leads to loop considering the original node as a merge candidate with itself in the second pass, and so it incorrectly determines a merge should occur.) Signed-off-by: Mark Tinguely Reviewed-by: Ben Myers Signed-off-by: Ben Myers [v3: added Dave Chinner's (slightly modified) suggestion to the commit header, cleaned up whitespace. -bpm] diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 069537c..20bf8e8 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -1224,6 +1224,7 @@ xfs_da3_node_toosmall( /* start with smaller blk num */ forward = nodehdr.forw < nodehdr.back; for (i = 0; i < 2; forward = !forward, i++) { + struct xfs_da3_icnode_hdr thdr; if (forward) blkno = nodehdr.forw; else @@ -1236,10 +1237,10 @@ xfs_da3_node_toosmall( return(error); node = bp->b_addr; - xfs_da3_node_hdr_from_disk(&nodehdr, node); + xfs_da3_node_hdr_from_disk(&thdr, node); xfs_trans_brelse(state->args->trans, bp); - if (count - nodehdr.count >= 0) + if (count - thdr.count >= 0) break; /* fits with at least 25% to spare */ } if (i >= 2) { -- cgit v0.10.2 From ad1260e9fbf768d6bed227d9604ebee76a84aae3 Mon Sep 17 00:00:00 2001 From: Ramneek Mehresh Date: Mon, 16 Sep 2013 15:11:33 +0530 Subject: fsl/usb: Resolve PHY_CLK_VLD instability issue for ULPI phy For controller versions greater than 1.6, setting ULPI_PHY_CLK_SEL bit when USB_EN bit is already set causes instability issues with PHY_CLK_VLD bit. So USB_EN is set only for IP controller version below 1.6 before setting ULPI_PHY_CLK_SEL bit Signed-off-by: Ramneek Mehresh Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ehci-fsl.c b/drivers/usb/host/ehci-fsl.c index 4449f56..d3166e6 100644 --- a/drivers/usb/host/ehci-fsl.c +++ b/drivers/usb/host/ehci-fsl.c @@ -130,7 +130,7 @@ static int usb_hcd_fsl_probe(const struct hc_driver *driver, } /* Enable USB controller, 83xx or 8536 */ - if (pdata->have_sysif_regs) + if (pdata->have_sysif_regs && pdata->controller_ver < FSL_USB_VER_1_6) setbits32(hcd->regs + FSL_SOC_USB_CTRL, 0x4); /* Don't need to set host mode here. It will be done by tdi_reset() */ @@ -232,15 +232,9 @@ static int ehci_fsl_setup_phy(struct usb_hcd *hcd, case FSL_USB2_PHY_ULPI: if (pdata->have_sysif_regs && pdata->controller_ver) { /* controller version 1.6 or above */ + clrbits32(non_ehci + FSL_SOC_USB_CTRL, UTMI_PHY_EN); setbits32(non_ehci + FSL_SOC_USB_CTRL, - ULPI_PHY_CLK_SEL); - /* - * Due to controller issue of PHY_CLK_VALID in ULPI - * mode, we set USB_CTRL_USB_EN before checking - * PHY_CLK_VALID, otherwise PHY_CLK_VALID doesn't work. - */ - clrsetbits_be32(non_ehci + FSL_SOC_USB_CTRL, - UTMI_PHY_EN, USB_CTRL_USB_EN); + ULPI_PHY_CLK_SEL | USB_CTRL_USB_EN); } portsc |= PORT_PTS_ULPI; break; -- cgit v0.10.2 From 2332575ef10a1bcf5ca485a6bf043831be11c7a0 Mon Sep 17 00:00:00 2001 From: Gabor Juhos Date: Sat, 31 Aug 2013 09:15:44 +0200 Subject: tty: ar933x_uart: move devicetree binding documentation Commit 'tty: ar933x_uart: add device tree support and binding documentation' introduced a new doc in bindins/tty/serial. According to a recent thread [1] on the linux-serial list, the binding documentation of serial drivers should be added into bindings/serial. Move the documentation of qca,ar9330-uart to the correct place. 1. http://marc.info/?l=linux-serial&m=137771295411517 Cc: devicetree@vger.kernel.org Signed-off-by: Gabor Juhos Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/devicetree/bindings/serial/qca,ar9330-uart.txt b/Documentation/devicetree/bindings/serial/qca,ar9330-uart.txt new file mode 100644 index 0000000..c5e032c --- /dev/null +++ b/Documentation/devicetree/bindings/serial/qca,ar9330-uart.txt @@ -0,0 +1,34 @@ +* Qualcomm Atheros AR9330 High-Speed UART + +Required properties: + +- compatible: Must be "qca,ar9330-uart" + +- reg: Specifies the physical base address of the controller and + the length of the memory mapped region. + +- interrupt-parent: The phandle for the interrupt controller that + services interrupts for this device. + +- interrupts: Specifies the interrupt source of the parent interrupt + controller. The format of the interrupt specifier depends on the + parent interrupt controller. + +Additional requirements: + + Each UART port must have an alias correctly numbered in "aliases" + node. + +Example: + + aliases { + serial0 = &uart0; + }; + + uart0: uart@18020000 { + compatible = "qca,ar9330-uart"; + reg = <0x18020000 0x14>; + + interrupt-parent = <&intc>; + interrupts = <3>; + }; diff --git a/Documentation/devicetree/bindings/tty/serial/qca,ar9330-uart.txt b/Documentation/devicetree/bindings/tty/serial/qca,ar9330-uart.txt deleted file mode 100644 index c5e032c..0000000 --- a/Documentation/devicetree/bindings/tty/serial/qca,ar9330-uart.txt +++ /dev/null @@ -1,34 +0,0 @@ -* Qualcomm Atheros AR9330 High-Speed UART - -Required properties: - -- compatible: Must be "qca,ar9330-uart" - -- reg: Specifies the physical base address of the controller and - the length of the memory mapped region. - -- interrupt-parent: The phandle for the interrupt controller that - services interrupts for this device. - -- interrupts: Specifies the interrupt source of the parent interrupt - controller. The format of the interrupt specifier depends on the - parent interrupt controller. - -Additional requirements: - - Each UART port must have an alias correctly numbered in "aliases" - node. - -Example: - - aliases { - serial0 = &uart0; - }; - - uart0: uart@18020000 { - compatible = "qca,ar9330-uart"; - reg = <0x18020000 0x14>; - - interrupt-parent = <&intc>; - interrupts = <3>; - }; -- cgit v0.10.2 From 1aee351a739153529fbb98ee461777b2abd5e1c9 Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Mon, 2 Sep 2013 13:29:45 +0300 Subject: mei: make me client counters less error prone 1. u8 counters are prone to hard to detect overflow: make them unsigned long to match bit_ functions argument type 2. don't check me_clients_num for negativity, it is unsigned. 3. init all the me client counters from one place Cc: # 3.9+ Signed-off-by: Tomas Winkler Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/misc/mei/hbm.c b/drivers/misc/mei/hbm.c index 6127ab6..0a044832 100644 --- a/drivers/misc/mei/hbm.c +++ b/drivers/misc/mei/hbm.c @@ -35,11 +35,15 @@ static void mei_hbm_me_cl_allocate(struct mei_device *dev) struct mei_me_client *clients; int b; + dev->me_clients_num = 0; + dev->me_client_presentation_num = 0; + dev->me_client_index = 0; + /* count how many ME clients we have */ for_each_set_bit(b, dev->me_clients_map, MEI_CLIENTS_MAX) dev->me_clients_num++; - if (dev->me_clients_num <= 0) + if (dev->me_clients_num == 0) return; kfree(dev->me_clients); @@ -221,7 +225,7 @@ static int mei_hbm_prop_req(struct mei_device *dev) struct hbm_props_request *prop_req; const size_t len = sizeof(struct hbm_props_request); unsigned long next_client_index; - u8 client_num; + unsigned long client_num; client_num = dev->me_client_presentation_num; @@ -677,8 +681,6 @@ void mei_hbm_dispatch(struct mei_device *dev, struct mei_msg_hdr *hdr) if (dev->dev_state == MEI_DEV_INIT_CLIENTS && dev->hbm_state == MEI_HBM_ENUM_CLIENTS) { dev->init_clients_timer = 0; - dev->me_client_presentation_num = 0; - dev->me_client_index = 0; mei_hbm_me_cl_allocate(dev); dev->hbm_state = MEI_HBM_CLIENT_PROPERTIES; diff --git a/drivers/misc/mei/mei_dev.h b/drivers/misc/mei/mei_dev.h index 7b918b2..456b322 100644 --- a/drivers/misc/mei/mei_dev.h +++ b/drivers/misc/mei/mei_dev.h @@ -396,9 +396,9 @@ struct mei_device { struct mei_me_client *me_clients; /* Note: memory has to be allocated */ DECLARE_BITMAP(me_clients_map, MEI_CLIENTS_MAX); DECLARE_BITMAP(host_clients_map, MEI_CLIENTS_MAX); - u8 me_clients_num; - u8 me_client_presentation_num; - u8 me_client_index; + unsigned long me_clients_num; + unsigned long me_client_presentation_num; + unsigned long me_client_index; struct mei_cl wd_cl; enum mei_wd_states wd_state; -- cgit v0.10.2 From e2b31644e999e8bfe3efce880fb32840299abf41 Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Mon, 2 Sep 2013 13:29:46 +0300 Subject: mei: bus: stop wait for read during cl state transition Bus layer omitted check for client state transition while waiting for read completion The client state transition may occur for example as result of firmware initiated reset Add mei_cl_is_transitioning wrapper to reduce the code repetition.: Cc: # 3.9+ Signed-off-by: Tomas Winkler Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/misc/mei/bus.c b/drivers/misc/mei/bus.c index 6d0282c..cd2033c 100644 --- a/drivers/misc/mei/bus.c +++ b/drivers/misc/mei/bus.c @@ -297,10 +297,13 @@ int __mei_cl_recv(struct mei_cl *cl, u8 *buf, size_t length) if (cl->reading_state != MEI_READ_COMPLETE && !waitqueue_active(&cl->rx_wait)) { + mutex_unlock(&dev->device_lock); if (wait_event_interruptible(cl->rx_wait, - (MEI_READ_COMPLETE == cl->reading_state))) { + cl->reading_state == MEI_READ_COMPLETE || + mei_cl_is_transitioning(cl))) { + if (signal_pending(current)) return -EINTR; return -ERESTARTSYS; diff --git a/drivers/misc/mei/client.h b/drivers/misc/mei/client.h index 9eb031e..892cc42 100644 --- a/drivers/misc/mei/client.h +++ b/drivers/misc/mei/client.h @@ -90,6 +90,12 @@ static inline bool mei_cl_is_connected(struct mei_cl *cl) cl->dev->dev_state == MEI_DEV_ENABLED && cl->state == MEI_FILE_CONNECTED); } +static inline bool mei_cl_is_transitioning(struct mei_cl *cl) +{ + return (MEI_FILE_INITIALIZING == cl->state || + MEI_FILE_DISCONNECTED == cl->state || + MEI_FILE_DISCONNECTING == cl->state); +} bool mei_cl_is_other_connecting(struct mei_cl *cl); int mei_cl_disconnect(struct mei_cl *cl); diff --git a/drivers/misc/mei/main.c b/drivers/misc/mei/main.c index 173ff09..cabeddd 100644 --- a/drivers/misc/mei/main.c +++ b/drivers/misc/mei/main.c @@ -249,19 +249,16 @@ static ssize_t mei_read(struct file *file, char __user *ubuf, mutex_unlock(&dev->device_lock); if (wait_event_interruptible(cl->rx_wait, - (MEI_READ_COMPLETE == cl->reading_state || - MEI_FILE_INITIALIZING == cl->state || - MEI_FILE_DISCONNECTED == cl->state || - MEI_FILE_DISCONNECTING == cl->state))) { + MEI_READ_COMPLETE == cl->reading_state || + mei_cl_is_transitioning(cl))) { + if (signal_pending(current)) return -EINTR; return -ERESTARTSYS; } mutex_lock(&dev->device_lock); - if (MEI_FILE_INITIALIZING == cl->state || - MEI_FILE_DISCONNECTED == cl->state || - MEI_FILE_DISCONNECTING == cl->state) { + if (mei_cl_is_transitioning(cl)) { rets = -EBUSY; goto out; } -- cgit v0.10.2 From 4a704575cc1afb3b848f096778fa9b8d7b3d5813 Mon Sep 17 00:00:00 2001 From: Alexander Usyskin Date: Mon, 2 Sep 2013 13:29:47 +0300 Subject: mei: cancel stall timers in mei_reset Unset init_clients_timer and amthif_stall_timers in mei_reset in order to cancel timer ticking and hence avoid recursive reset calls. Cc: # 3.9+ Signed-off-by: Alexander Usyskin Signed-off-by: Tomas Winkler Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/misc/mei/amthif.c b/drivers/misc/mei/amthif.c index d0fdc13..f6ff711 100644 --- a/drivers/misc/mei/amthif.c +++ b/drivers/misc/mei/amthif.c @@ -57,6 +57,7 @@ void mei_amthif_reset_params(struct mei_device *dev) dev->iamthif_ioctl = false; dev->iamthif_state = MEI_IAMTHIF_IDLE; dev->iamthif_timer = 0; + dev->iamthif_stall_timer = 0; } /** diff --git a/drivers/misc/mei/init.c b/drivers/misc/mei/init.c index 92c7311..6197018 100644 --- a/drivers/misc/mei/init.c +++ b/drivers/misc/mei/init.c @@ -175,6 +175,9 @@ void mei_reset(struct mei_device *dev, int interrupts_enabled) memset(&dev->wr_ext_msg, 0, sizeof(dev->wr_ext_msg)); } + /* we're already in reset, cancel the init timer */ + dev->init_clients_timer = 0; + dev->me_clients_num = 0; dev->rd_msg_hdr = 0; dev->wd_pending = false; -- cgit v0.10.2 From 3a4916050ba2e0f1d114ef540abdf02b2b173e61 Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Fri, 6 Sep 2013 11:49:56 -0700 Subject: Drivers: hv: util: Correctly support ws2008R2 and earlier The current code does not correctly negotiate the version numbers for the util driver when hosted on earlier hosts. The version numbers presented by this driver were not compatible with the version numbers supported by Windows Server 2008. Fix this problem. I would like to thank Olaf Hering (ohering@suse.com) for identifying the problem. Reported-by: Olaf Hering Signed-off-by: K. Y. Srinivasan Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/hv/hv_kvp.c b/drivers/hv/hv_kvp.c index 28b0332..09988b2 100644 --- a/drivers/hv/hv_kvp.c +++ b/drivers/hv/hv_kvp.c @@ -32,13 +32,17 @@ /* * Pre win8 version numbers used in ws2008 and ws 2008 r2 (win7) */ +#define WS2008_SRV_MAJOR 1 +#define WS2008_SRV_MINOR 0 +#define WS2008_SRV_VERSION (WS2008_SRV_MAJOR << 16 | WS2008_SRV_MINOR) + #define WIN7_SRV_MAJOR 3 #define WIN7_SRV_MINOR 0 -#define WIN7_SRV_MAJOR_MINOR (WIN7_SRV_MAJOR << 16 | WIN7_SRV_MINOR) +#define WIN7_SRV_VERSION (WIN7_SRV_MAJOR << 16 | WIN7_SRV_MINOR) #define WIN8_SRV_MAJOR 4 #define WIN8_SRV_MINOR 0 -#define WIN8_SRV_MAJOR_MINOR (WIN8_SRV_MAJOR << 16 | WIN8_SRV_MINOR) +#define WIN8_SRV_VERSION (WIN8_SRV_MAJOR << 16 | WIN8_SRV_MINOR) /* * Global state maintained for transaction that is being processed. @@ -587,6 +591,8 @@ void hv_kvp_onchannelcallback(void *context) struct icmsg_hdr *icmsghdrp; struct icmsg_negotiate *negop = NULL; + int util_fw_version; + int kvp_srv_version; if (kvp_transaction.active) { /* @@ -606,17 +612,26 @@ void hv_kvp_onchannelcallback(void *context) if (icmsghdrp->icmsgtype == ICMSGTYPE_NEGOTIATE) { /* - * We start with win8 version and if the host cannot - * support that we use the previous version. + * Based on the host, select appropriate + * framework and service versions we will + * negotiate. */ - if (vmbus_prep_negotiate_resp(icmsghdrp, negop, - recv_buffer, UTIL_FW_MAJOR_MINOR, - WIN8_SRV_MAJOR_MINOR)) - goto done; - + switch (vmbus_proto_version) { + case (VERSION_WS2008): + util_fw_version = UTIL_WS2K8_FW_VERSION; + kvp_srv_version = WS2008_SRV_VERSION; + break; + case (VERSION_WIN7): + util_fw_version = UTIL_FW_VERSION; + kvp_srv_version = WIN7_SRV_VERSION; + break; + default: + util_fw_version = UTIL_FW_VERSION; + kvp_srv_version = WIN8_SRV_VERSION; + } vmbus_prep_negotiate_resp(icmsghdrp, negop, - recv_buffer, UTIL_FW_MAJOR_MINOR, - WIN7_SRV_MAJOR_MINOR); + recv_buffer, util_fw_version, + kvp_srv_version); } else { kvp_msg = (struct hv_kvp_msg *)&recv_buffer[ @@ -649,7 +664,6 @@ void hv_kvp_onchannelcallback(void *context) return; } -done: icmsghdrp->icflags = ICMSGHDRFLAG_TRANSACTION | ICMSGHDRFLAG_RESPONSE; diff --git a/drivers/hv/hv_snapshot.c b/drivers/hv/hv_snapshot.c index e4572f3..0c35462 100644 --- a/drivers/hv/hv_snapshot.c +++ b/drivers/hv/hv_snapshot.c @@ -26,7 +26,7 @@ #define VSS_MAJOR 5 #define VSS_MINOR 0 -#define VSS_MAJOR_MINOR (VSS_MAJOR << 16 | VSS_MINOR) +#define VSS_VERSION (VSS_MAJOR << 16 | VSS_MINOR) @@ -190,8 +190,8 @@ void hv_vss_onchannelcallback(void *context) if (icmsghdrp->icmsgtype == ICMSGTYPE_NEGOTIATE) { vmbus_prep_negotiate_resp(icmsghdrp, negop, - recv_buffer, UTIL_FW_MAJOR_MINOR, - VSS_MAJOR_MINOR); + recv_buffer, UTIL_FW_VERSION, + VSS_VERSION); } else { vss_msg = (struct hv_vss_msg *)&recv_buffer[ sizeof(struct vmbuspipe_hdr) + diff --git a/drivers/hv/hv_util.c b/drivers/hv/hv_util.c index cb82233..273e3dd 100644 --- a/drivers/hv/hv_util.c +++ b/drivers/hv/hv_util.c @@ -28,17 +28,32 @@ #include #include -#define SHUTDOWN_MAJOR 3 -#define SHUTDOWN_MINOR 0 -#define SHUTDOWN_MAJOR_MINOR (SHUTDOWN_MAJOR << 16 | SHUTDOWN_MINOR) -#define TIMESYNCH_MAJOR 3 -#define TIMESYNCH_MINOR 0 -#define TIMESYNCH_MAJOR_MINOR (TIMESYNCH_MAJOR << 16 | TIMESYNCH_MINOR) +#define SD_MAJOR 3 +#define SD_MINOR 0 +#define SD_VERSION (SD_MAJOR << 16 | SD_MINOR) -#define HEARTBEAT_MAJOR 3 -#define HEARTBEAT_MINOR 0 -#define HEARTBEAT_MAJOR_MINOR (HEARTBEAT_MAJOR << 16 | HEARTBEAT_MINOR) +#define SD_WS2008_MAJOR 1 +#define SD_WS2008_VERSION (SD_WS2008_MAJOR << 16 | SD_MINOR) + +#define TS_MAJOR 3 +#define TS_MINOR 0 +#define TS_VERSION (TS_MAJOR << 16 | TS_MINOR) + +#define TS_WS2008_MAJOR 1 +#define TS_WS2008_VERSION (TS_WS2008_MAJOR << 16 | TS_MINOR) + +#define HB_MAJOR 3 +#define HB_MINOR 0 +#define HB_VERSION (HB_MAJOR << 16 | HB_MINOR) + +#define HB_WS2008_MAJOR 1 +#define HB_WS2008_VERSION (HB_WS2008_MAJOR << 16 | HB_MINOR) + +static int sd_srv_version; +static int ts_srv_version; +static int hb_srv_version; +static int util_fw_version; static void shutdown_onchannelcallback(void *context); static struct hv_util_service util_shutdown = { @@ -99,8 +114,8 @@ static void shutdown_onchannelcallback(void *context) if (icmsghdrp->icmsgtype == ICMSGTYPE_NEGOTIATE) { vmbus_prep_negotiate_resp(icmsghdrp, negop, - shut_txf_buf, UTIL_FW_MAJOR_MINOR, - SHUTDOWN_MAJOR_MINOR); + shut_txf_buf, util_fw_version, + sd_srv_version); } else { shutdown_msg = (struct shutdown_msg_data *)&shut_txf_buf[ @@ -216,6 +231,7 @@ static void timesync_onchannelcallback(void *context) struct icmsg_hdr *icmsghdrp; struct ictimesync_data *timedatap; u8 *time_txf_buf = util_timesynch.recv_buffer; + struct icmsg_negotiate *negop = NULL; vmbus_recvpacket(channel, time_txf_buf, PAGE_SIZE, &recvlen, &requestid); @@ -225,9 +241,10 @@ static void timesync_onchannelcallback(void *context) sizeof(struct vmbuspipe_hdr)]; if (icmsghdrp->icmsgtype == ICMSGTYPE_NEGOTIATE) { - vmbus_prep_negotiate_resp(icmsghdrp, NULL, time_txf_buf, - UTIL_FW_MAJOR_MINOR, - TIMESYNCH_MAJOR_MINOR); + vmbus_prep_negotiate_resp(icmsghdrp, negop, + time_txf_buf, + util_fw_version, + ts_srv_version); } else { timedatap = (struct ictimesync_data *)&time_txf_buf[ sizeof(struct vmbuspipe_hdr) + @@ -257,6 +274,7 @@ static void heartbeat_onchannelcallback(void *context) struct icmsg_hdr *icmsghdrp; struct heartbeat_msg_data *heartbeat_msg; u8 *hbeat_txf_buf = util_heartbeat.recv_buffer; + struct icmsg_negotiate *negop = NULL; vmbus_recvpacket(channel, hbeat_txf_buf, PAGE_SIZE, &recvlen, &requestid); @@ -266,9 +284,9 @@ static void heartbeat_onchannelcallback(void *context) sizeof(struct vmbuspipe_hdr)]; if (icmsghdrp->icmsgtype == ICMSGTYPE_NEGOTIATE) { - vmbus_prep_negotiate_resp(icmsghdrp, NULL, - hbeat_txf_buf, UTIL_FW_MAJOR_MINOR, - HEARTBEAT_MAJOR_MINOR); + vmbus_prep_negotiate_resp(icmsghdrp, negop, + hbeat_txf_buf, util_fw_version, + hb_srv_version); } else { heartbeat_msg = (struct heartbeat_msg_data *)&hbeat_txf_buf[ @@ -321,6 +339,25 @@ static int util_probe(struct hv_device *dev, goto error; hv_set_drvdata(dev, srv); + /* + * Based on the host; initialize the framework and + * service version numbers we will negotiate. + */ + switch (vmbus_proto_version) { + case (VERSION_WS2008): + util_fw_version = UTIL_WS2K8_FW_VERSION; + sd_srv_version = SD_WS2008_VERSION; + ts_srv_version = TS_WS2008_VERSION; + hb_srv_version = HB_WS2008_VERSION; + break; + + default: + util_fw_version = UTIL_FW_VERSION; + sd_srv_version = SD_VERSION; + ts_srv_version = TS_VERSION; + hb_srv_version = HB_VERSION; + } + return 0; error: diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index a3b8b2e..d98503b 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -30,10 +30,13 @@ /* * Framework version for util services. */ +#define UTIL_FW_MINOR 0 + +#define UTIL_WS2K8_FW_MAJOR 1 +#define UTIL_WS2K8_FW_VERSION (UTIL_WS2K8_FW_MAJOR << 16 | UTIL_FW_MINOR) #define UTIL_FW_MAJOR 3 -#define UTIL_FW_MINOR 0 -#define UTIL_FW_MAJOR_MINOR (UTIL_FW_MAJOR << 16 | UTIL_FW_MINOR) +#define UTIL_FW_VERSION (UTIL_FW_MAJOR << 16 | UTIL_FW_MINOR) /* -- cgit v0.10.2 From 8bbf9f440f196ee08cc5ab1c19e62b1c5a736ff0 Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Wed, 4 Sep 2013 15:41:58 -0700 Subject: Drivers: hv: vmbus: Terminate vmbus version negotiation on timeout commit 666b9adc801ef012612c4e43e0f44b2cdc1979cf terminated vmbus version negotiation incorrectly. We need to terminate the version negotiation only if the current negotiation were to timeout. Signed-off-by: K. Y. Srinivasan Cc: Olaf Hering Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c index 8f4743a..936093e 100644 --- a/drivers/hv/connection.c +++ b/drivers/hv/connection.c @@ -195,7 +195,7 @@ int vmbus_connect(void) do { ret = vmbus_negotiate_version(msginfo, version); - if (ret) + if (ret == -ETIMEDOUT) goto cleanup; if (vmbus_connection.conn_state == CONNECTED) -- cgit v0.10.2 From 667b4102b3e63856ca7770521ee74b1c44629df1 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 23 Sep 2013 14:41:17 -0700 Subject: sysfs: Allow mounting without CONFIG_NET In kobj_ns_current_may_mount the default should be to allow the mount. The test is only for a single kobj_ns_type at a time, and unless there is a reason to prevent it the mounting sysfs should be allowed. Subsystems that are not registered can't have are not involved so can't have a reason to prevent mounting sysfs. This is a bug-fix to: commit 7dc5dbc879bd0779924b5132a48b731a0bc04a1e Author: Eric W. Biederman Date: Mon Mar 25 20:07:01 2013 -0700 sysfs: Restrict mounting sysfs Don't allow mounting sysfs unless the caller has CAP_SYS_ADMIN rights over the net namespace. The principle here is if you create or have capabilities over it you can mount it, otherwise you get to live with what other people have mounted. Instead of testing this with a straight forward ns_capable call, perform this check the long and torturous way with kobject helpers, this keeps direct knowledge of namespaces out of sysfs, and preserves the existing sysfs abstractions. Acked-by: Greg Kroah-Hartman Signed-off-by: "Eric W. Biederman" That came in via the userns tree during the 3.12 merge window. Reported-by: James Hogan Signed-off-by: "Eric W. Biederman" Signed-off-by: Greg Kroah-Hartman diff --git a/lib/kobject.c b/lib/kobject.c index 9621751..669bf19 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -933,10 +933,7 @@ const struct kobj_ns_type_operations *kobj_ns_ops(struct kobject *kobj) bool kobj_ns_current_may_mount(enum kobj_ns_type type) { - bool may_mount = false; - - if (type == KOBJ_NS_TYPE_NONE) - return true; + bool may_mount = true; spin_lock(&kobj_ns_type_lock); if ((type > KOBJ_NS_TYPE_NONE) && (type < KOBJ_NS_TYPES) && -- cgit v0.10.2 From f123db8e9d6c84c863cb3c44d17e61995dc984fb Mon Sep 17 00:00:00 2001 From: Benson Leung Date: Tue, 24 Sep 2013 20:05:11 -0700 Subject: driver core : Fix use after free of dev->parent in device_shutdown The put_device(dev) at the bottom of the loop of device_shutdown may result in the dev being cleaned up. In device_create_release, the dev is kfreed. However, device_shutdown attempts to use the dev pointer again after put_device by referring to dev->parent. Copy the parent pointer instead to avoid this condition. This bug was found on Chromium OS's chromeos-3.8, which is based on v3.8.11. See bug report : https://code.google.com/p/chromium/issues/detail?id=297842 This can easily be reproduced when shutting down with hidraw devices that report battery condition. Two examples are the HP Bluetooth Mouse X4000b and the Apple Magic Mouse. For example, with the magic mouse : The dev in question is "hidraw0" dev->parent is "magicmouse" In the course of the shutdown for this device, the input event cleanup calls a put on hidraw0, decrementing its reference count. When we finally get to put_device(dev) in device_shutdown, kobject_cleanup is called and device_create_release does kfree(dev). dev->parent is no longer valid, and we may crash in put_device(dev->parent). This change should be applied on any kernel with this change : d1c6c030fcec6f860d9bb6c632a3ebe62e28440b Cc: stable@vger.kernel.org Signed-off-by: Benson Leung Reviewed-by: Ming Lei Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/core.c b/drivers/base/core.c index c7cfadc..34abf4d 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -2017,7 +2017,7 @@ EXPORT_SYMBOL_GPL(device_move); */ void device_shutdown(void) { - struct device *dev; + struct device *dev, *parent; spin_lock(&devices_kset->list_lock); /* @@ -2034,7 +2034,7 @@ void device_shutdown(void) * prevent it from being freed because parent's * lock is to be held */ - get_device(dev->parent); + parent = get_device(dev->parent); get_device(dev); /* * Make sure the device is off the kset list, in the @@ -2044,8 +2044,8 @@ void device_shutdown(void) spin_unlock(&devices_kset->list_lock); /* hold lock to avoid race with probe/release */ - if (dev->parent) - device_lock(dev->parent); + if (parent) + device_lock(parent); device_lock(dev); /* Don't allow any more runtime suspends */ @@ -2063,11 +2063,11 @@ void device_shutdown(void) } device_unlock(dev); - if (dev->parent) - device_unlock(dev->parent); + if (parent) + device_unlock(parent); put_device(dev); - put_device(dev->parent); + put_device(parent); spin_lock(&devices_kset->list_lock); } -- cgit v0.10.2 From 2606b28aabd7dea1766c23a105e1124c95409c96 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 20 Sep 2013 17:14:21 +0100 Subject: USB: Fix breakage in ffs_fs_mount() There's a bunch of failure exits in ffs_fs_mount() with seriously broken recovery logics. Most of that appears to stem from misunderstanding of the ->kill_sb() semantics; unlike ->put_super() it is called for *all* superblocks of given type, no matter how (in)complete the setup had been. ->put_super() is called only if ->s_root is not NULL; any failure prior to setting ->s_root will have the call of ->put_super() skipped. ->kill_sb(), OTOH, awaits every superblock that has come from sget(). Current behaviour of ffs_fs_mount(): We have struct ffs_sb_fill_data data on stack there. We do ffs_dev = functionfs_acquire_dev_callback(dev_name); and store that in data.private_data. Then we call mount_nodev(), passing it ffs_sb_fill() as a callback. That will either fail outright, or manage to call ffs_sb_fill(). There we allocate an instance of struct ffs_data, slap the value of ffs_dev (picked from data.private_data) into ffs->private_data and overwrite data.private_data by storing ffs into an overlapping member (data.ffs_data). Then we store ffs into sb->s_fs_info and attempt to set the rest of the things up (root inode, root dentry, then create /ep0 there). Any of those might fail. Should that happen, we get ffs_fs_kill_sb() called before mount_nodev() returns. If mount_nodev() fails for any reason whatsoever, we proceed to functionfs_release_dev_callback(data.ffs_data); That's broken in a lot of ways. Suppose the thing has failed in allocation of e.g. root inode or dentry. We have functionfs_release_dev_callback(ffs); ffs_data_put(ffs); done by ffs_fs_kill_sb() (ffs accessed via sb->s_fs_info), followed by functionfs_release_dev_callback(ffs); from ffs_fs_mount() (via data.ffs_data). Note that the second functionfs_release_dev_callback() has every chance to be done to freed memory. Suppose we fail *before* root inode allocation. What happens then? ffs_fs_kill_sb() doesn't do anything to ffs (it's either not called at all, or it doesn't have a pointer to ffs stored in sb->s_fs_info). And functionfs_release_dev_callback(data.ffs_data); is called by ffs_fs_mount(), but here we are in nasal daemon country - we are reading from a member of union we'd never stored into. In practice, we'll get what we used to store into the overlapping field, i.e. ffs_dev. And then we get screwed, since we treat it (struct gfs_ffs_obj * in disguise, returned by functionfs_acquire_dev_callback()) as struct ffs_data *, pick what would've been ffs_data ->private_data from it (*well* past the actual end of the struct gfs_ffs_obj - struct ffs_data is much bigger) and poke in whatever it points to. FWIW, there's a minor leak on top of all that in case if ffs_sb_fill() fails on kstrdup() - ffs is obviously forgotten. The thing is, there is no point in playing all those games with union. Just allocate and initialize ffs_data *before* calling mount_nodev() and pass a pointer to it via data.ffs_data. And once it's stored in sb->s_fs_info, clear data.ffs_data, so that ffs_fs_mount() knows that it doesn't need to kill the sucker manually - from that point on we'll have it done by ->kill_sb(). Signed-off-by: Al Viro Acked-by: Michal Nazarewicz Cc: stable # 3.3+ Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c index 1a66c5b..0658908 100644 --- a/drivers/usb/gadget/f_fs.c +++ b/drivers/usb/gadget/f_fs.c @@ -1034,37 +1034,19 @@ struct ffs_sb_fill_data { struct ffs_file_perms perms; umode_t root_mode; const char *dev_name; - union { - /* set by ffs_fs_mount(), read by ffs_sb_fill() */ - void *private_data; - /* set by ffs_sb_fill(), read by ffs_fs_mount */ - struct ffs_data *ffs_data; - }; + struct ffs_data *ffs_data; }; static int ffs_sb_fill(struct super_block *sb, void *_data, int silent) { struct ffs_sb_fill_data *data = _data; struct inode *inode; - struct ffs_data *ffs; + struct ffs_data *ffs = data->ffs_data; ENTER(); - /* Initialise data */ - ffs = ffs_data_new(); - if (unlikely(!ffs)) - goto Enomem; - ffs->sb = sb; - ffs->dev_name = kstrdup(data->dev_name, GFP_KERNEL); - if (unlikely(!ffs->dev_name)) - goto Enomem; - ffs->file_perms = data->perms; - ffs->private_data = data->private_data; - - /* used by the caller of this function */ - data->ffs_data = ffs; - + data->ffs_data = NULL; sb->s_fs_info = ffs; sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; @@ -1080,17 +1062,14 @@ static int ffs_sb_fill(struct super_block *sb, void *_data, int silent) &data->perms); sb->s_root = d_make_root(inode); if (unlikely(!sb->s_root)) - goto Enomem; + return -ENOMEM; /* EP0 file */ if (unlikely(!ffs_sb_create_file(sb, "ep0", ffs, &ffs_ep0_operations, NULL))) - goto Enomem; + return -ENOMEM; return 0; - -Enomem: - return -ENOMEM; } static int ffs_fs_parse_opts(struct ffs_sb_fill_data *data, char *opts) @@ -1193,6 +1172,7 @@ ffs_fs_mount(struct file_system_type *t, int flags, struct dentry *rv; int ret; void *ffs_dev; + struct ffs_data *ffs; ENTER(); @@ -1200,18 +1180,30 @@ ffs_fs_mount(struct file_system_type *t, int flags, if (unlikely(ret < 0)) return ERR_PTR(ret); + ffs = ffs_data_new(); + if (unlikely(!ffs)) + return ERR_PTR(-ENOMEM); + ffs->file_perms = data.perms; + + ffs->dev_name = kstrdup(dev_name, GFP_KERNEL); + if (unlikely(!ffs->dev_name)) { + ffs_data_put(ffs); + return ERR_PTR(-ENOMEM); + } + ffs_dev = functionfs_acquire_dev_callback(dev_name); - if (IS_ERR(ffs_dev)) - return ffs_dev; + if (IS_ERR(ffs_dev)) { + ffs_data_put(ffs); + return ERR_CAST(ffs_dev); + } + ffs->private_data = ffs_dev; + data.ffs_data = ffs; - data.dev_name = dev_name; - data.private_data = ffs_dev; rv = mount_nodev(t, flags, &data, ffs_sb_fill); - - /* data.ffs_data is set by ffs_sb_fill */ - if (IS_ERR(rv)) + if (IS_ERR(rv) && data.ffs_data) { functionfs_release_dev_callback(data.ffs_data); - + ffs_data_put(data.ffs_data); + } return rv; } -- cgit v0.10.2 From eee41b49b80420e3e8c118d18dfacb7da43c1caa Mon Sep 17 00:00:00 2001 From: Shengzhou Liu Date: Mon, 2 Sep 2013 13:25:52 +0800 Subject: USB: fsl/ehci: fix failure of checking PHY_CLK_VALID during reinitialization In case of usb phy reinitialization: e.g. insmod usb-module(usb works well) -> rmmod usb-module -> insmod usb-module It found the PHY_CLK_VALID bit didn't work if it's not with the power-on reset. So we just check PHY_CLK_VALID bit during the stage with POR, this can be met by the tricky of checking FSL_SOC_USB_PRICTRL register. Signed-off-by: Shengzhou Liu Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ehci-fsl.c b/drivers/usb/host/ehci-fsl.c index d3166e6..f2407b2 100644 --- a/drivers/usb/host/ehci-fsl.c +++ b/drivers/usb/host/ehci-fsl.c @@ -264,8 +264,9 @@ static int ehci_fsl_setup_phy(struct usb_hcd *hcd, if (pdata->have_sysif_regs && pdata->controller_ver && (phy_mode == FSL_USB2_PHY_ULPI)) { /* check PHY_CLK_VALID to get phy clk valid */ - if (!spin_event_timeout(in_be32(non_ehci + FSL_SOC_USB_CTRL) & - PHY_CLK_VALID, FSL_USB_PHY_CLK_TIMEOUT, 0)) { + if (!(spin_event_timeout(in_be32(non_ehci + FSL_SOC_USB_CTRL) & + PHY_CLK_VALID, FSL_USB_PHY_CLK_TIMEOUT, 0) || + in_be32(non_ehci + FSL_SOC_USB_PRICTRL))) { printk(KERN_WARNING "fsl-ehci: USB PHY clock invalid\n"); return -EINVAL; } -- cgit v0.10.2 From 85601f8cf67c56a561a6dd5e130e65fdc179047d Mon Sep 17 00:00:00 2001 From: David Cohen Date: Thu, 26 Sep 2013 13:01:44 -0700 Subject: usb: dwc3: add support for Merrifield Add PCI id for Intel Merrifield Signed-off-by: David Cohen Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/dwc3/dwc3-pci.c b/drivers/usb/dwc3/dwc3-pci.c index 997ebe4..2e252aa 100644 --- a/drivers/usb/dwc3/dwc3-pci.c +++ b/drivers/usb/dwc3/dwc3-pci.c @@ -29,6 +29,7 @@ #define PCI_VENDOR_ID_SYNOPSYS 0x16c3 #define PCI_DEVICE_ID_SYNOPSYS_HAPSUSB3 0xabcd #define PCI_DEVICE_ID_INTEL_BYT 0x0f37 +#define PCI_DEVICE_ID_INTEL_MRFLD 0x119e struct dwc3_pci { struct device *dev; @@ -189,6 +190,7 @@ static DEFINE_PCI_DEVICE_TABLE(dwc3_pci_id_table) = { PCI_DEVICE_ID_SYNOPSYS_HAPSUSB3), }, { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_BYT), }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MRFLD), }, { } /* Terminating Entry */ }; MODULE_DEVICE_TABLE(pci, dwc3_pci_id_table); -- cgit v0.10.2 From df3f4edc6ca6e7768376d05eb99536bcf92362f0 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Thu, 26 Sep 2013 13:07:31 -0500 Subject: staging: r8188eu: Add new device ID The DLink DWA-125 Rev D1 also uses this driver. Signed-off-by: Larry Finger Reported-by: Sergey Kostanbaev Tested-by: Sergey Kostanbaev Cc: Sergey Kostanbaev Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/rtl8188eu/os_dep/usb_intf.c b/drivers/staging/rtl8188eu/os_dep/usb_intf.c index d3078d2..9ca3180 100644 --- a/drivers/staging/rtl8188eu/os_dep/usb_intf.c +++ b/drivers/staging/rtl8188eu/os_dep/usb_intf.c @@ -54,6 +54,7 @@ static struct usb_device_id rtw_usb_id_tbl[] = { /*=== Customer ID ===*/ /****** 8188EUS ********/ {USB_DEVICE(0x8179, 0x07B8)}, /* Abocom - Abocom */ + {USB_DEVICE(0x2001, 0x330F)}, /* DLink DWA-125 REV D1 */ {} /* Terminating entry */ }; -- cgit v0.10.2 From accd1e823ed1d5980106dd522a4c535084400830 Mon Sep 17 00:00:00 2001 From: Suravee Suthikulpanit Date: Wed, 29 Sep 2010 19:27:12 -0500 Subject: x86/microcode/AMD: Fix patch level reporting for family 15h On AMD family 14h, applying microcode patch on the a core (core0) would also affect the other core (core1) in the same compute unit. The driver would skip applying the patch on core1, but it still need to update kernel structures to reflect the proper patch level. The current logic is not updating the struct ucode_cpu_info.cpu_sig.rev of the skipped core. This causes the /sys/devices/system/cpu/cpu1/microcode/version to report incorrect patch level as shown below: $ grep . cpu?/microcode/version cpu0/microcode/version:0x600063d cpu1/microcode/version:0x6000626 cpu2/microcode/version:0x600063d cpu3/microcode/version:0x6000626 cpu4/microcode/version:0x600063d Signed-off-by: Suravee Suthikulpanit Acked-by: Borislav Petkov Cc: Cc: Cc: Link: http://lkml.kernel.org/r/1285806432-1995-1-git-send-email-suravee.suthikulpanit@amd.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index 7123b5d..af99f71 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -216,6 +216,7 @@ int apply_microcode_amd(int cpu) /* need to apply patch? */ if (rev >= mc_amd->hdr.patch_id) { c->microcode = rev; + uci->cpu_sig.rev = rev; return 0; } -- cgit v0.10.2 From b5bf0a929d7ca35b9ccfc24647a397899d307659 Mon Sep 17 00:00:00 2001 From: Ben Whitten Date: Fri, 27 Sep 2013 10:13:51 +0200 Subject: ALSA: hda - Add fixup for MacBook Air 6,1 and 6,2 with CS4208 codec This patch adds the default pin configuration and some init verbs for setting COEFs, in addition to the correction of input pin AMP caps for MacBook Air 6,1 and 6,2. With these changes, the headphone jack detection starts working properly. [trivial space fixes by tiwai] Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=60811 Signed-off-by: Ben Whitten Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_cirrus.c b/sound/pci/hda/patch_cirrus.c index b524f89..18d9725 100644 --- a/sound/pci/hda/patch_cirrus.c +++ b/sound/pci/hda/patch_cirrus.c @@ -111,6 +111,9 @@ enum { /* 0x0009 - 0x0014 -> 12 test regs */ /* 0x0015 - visibility reg */ +/* Cirrus Logic CS4208 */ +#define CS4208_VENDOR_NID 0x24 + /* * Cirrus Logic CS4210 * @@ -223,6 +226,16 @@ static const struct hda_verb cs_coef_init_verbs[] = { {} /* terminator */ }; +static const struct hda_verb cs4208_coef_init_verbs[] = { + {0x01, AC_VERB_SET_POWER_STATE, 0x00}, /* AFG: D0 */ + {0x24, AC_VERB_SET_PROC_STATE, 0x01}, /* VPW: processing on */ + {0x24, AC_VERB_SET_COEF_INDEX, 0x0033}, + {0x24, AC_VERB_SET_PROC_COEF, 0x0001}, /* A1 ICS */ + {0x24, AC_VERB_SET_COEF_INDEX, 0x0034}, + {0x24, AC_VERB_SET_PROC_COEF, 0x1C01}, /* A1 Enable, A Thresh = 300mV */ + {} /* terminator */ +}; + /* Errata: CS4207 rev C0/C1/C2 Silicon * * http://www.cirrus.com/en/pubs/errata/ER880C3.pdf @@ -295,6 +308,8 @@ static int cs_init(struct hda_codec *codec) /* init_verb sequence for C0/C1/C2 errata*/ snd_hda_sequence_write(codec, cs_errata_init_verbs); snd_hda_sequence_write(codec, cs_coef_init_verbs); + } else if (spec->vendor_nid == CS4208_VENDOR_NID) { + snd_hda_sequence_write(codec, cs4208_coef_init_verbs); } snd_hda_gen_init(codec); @@ -434,6 +449,29 @@ static const struct hda_pintbl mba42_pincfgs[] = { {} /* terminator */ }; +static const struct hda_pintbl mba6_pincfgs[] = { + { 0x10, 0x032120f0 }, /* HP */ + { 0x11, 0x500000f0 }, + { 0x12, 0x90100010 }, /* Speaker */ + { 0x13, 0x500000f0 }, + { 0x14, 0x500000f0 }, + { 0x15, 0x770000f0 }, + { 0x16, 0x770000f0 }, + { 0x17, 0x430000f0 }, + { 0x18, 0x43ab9030 }, /* Mic */ + { 0x19, 0x770000f0 }, + { 0x1a, 0x770000f0 }, + { 0x1b, 0x770000f0 }, + { 0x1c, 0x90a00090 }, + { 0x1d, 0x500000f0 }, + { 0x1e, 0x500000f0 }, + { 0x1f, 0x500000f0 }, + { 0x20, 0x500000f0 }, + { 0x21, 0x430000f0 }, + { 0x22, 0x430000f0 }, + {} /* terminator */ +}; + static void cs420x_fixup_gpio_13(struct hda_codec *codec, const struct hda_fixup *fix, int action) { @@ -556,22 +594,23 @@ static int patch_cs420x(struct hda_codec *codec) /* * CS4208 support: - * Its layout is no longer compatible with CS4206/CS4207, and the generic - * parser seems working fairly well, except for trivial fixups. + * Its layout is no longer compatible with CS4206/CS4207 */ enum { + CS4208_MBA6, CS4208_GPIO0, }; static const struct hda_model_fixup cs4208_models[] = { { .id = CS4208_GPIO0, .name = "gpio0" }, + { .id = CS4208_MBA6, .name = "mba6" }, {} }; static const struct snd_pci_quirk cs4208_fixup_tbl[] = { /* codec SSID */ - SND_PCI_QUIRK(0x106b, 0x7100, "MacBookPro 6,1", CS4208_GPIO0), - SND_PCI_QUIRK(0x106b, 0x7200, "MacBookPro 6,2", CS4208_GPIO0), + SND_PCI_QUIRK(0x106b, 0x7100, "MacBookAir 6,1", CS4208_MBA6), + SND_PCI_QUIRK(0x106b, 0x7200, "MacBookAir 6,2", CS4208_MBA6), {} /* terminator */ }; @@ -588,18 +627,35 @@ static void cs4208_fixup_gpio0(struct hda_codec *codec, } static const struct hda_fixup cs4208_fixups[] = { + [CS4208_MBA6] = { + .type = HDA_FIXUP_PINS, + .v.pins = mba6_pincfgs, + .chained = true, + .chain_id = CS4208_GPIO0, + }, [CS4208_GPIO0] = { .type = HDA_FIXUP_FUNC, .v.func = cs4208_fixup_gpio0, }, }; +/* correct the 0dB offset of input pins */ +static void cs4208_fix_amp_caps(struct hda_codec *codec, hda_nid_t adc) +{ + unsigned int caps; + + caps = query_amp_caps(codec, adc, HDA_INPUT); + caps &= ~(AC_AMPCAP_OFFSET); + caps |= 0x02; + snd_hda_override_amp_caps(codec, adc, HDA_INPUT, caps); +} + static int patch_cs4208(struct hda_codec *codec) { struct cs_spec *spec; int err; - spec = cs_alloc_spec(codec, 0); /* no specific w/a */ + spec = cs_alloc_spec(codec, CS4208_VENDOR_NID); if (!spec) return -ENOMEM; @@ -609,6 +665,12 @@ static int patch_cs4208(struct hda_codec *codec) cs4208_fixups); snd_hda_apply_fixup(codec, HDA_FIXUP_ACT_PRE_PROBE); + snd_hda_override_wcaps(codec, 0x18, + get_wcaps(codec, 0x18) | AC_WCAP_STEREO); + cs4208_fix_amp_caps(codec, 0x18); + cs4208_fix_amp_caps(codec, 0x1b); + cs4208_fix_amp_caps(codec, 0x1c); + err = cs_parse_auto_config(codec); if (err < 0) goto error; -- cgit v0.10.2 From a0760527f51db86a13daff345fe4f206cc45e6bf Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 27 Sep 2013 10:19:31 +0200 Subject: ALSA: hda - Add documentation for CS4208 fixups Signed-off-by: Takashi Iwai diff --git a/Documentation/sound/alsa/HD-Audio-Models.txt b/Documentation/sound/alsa/HD-Audio-Models.txt index a46ddb8..f911e36 100644 --- a/Documentation/sound/alsa/HD-Audio-Models.txt +++ b/Documentation/sound/alsa/HD-Audio-Models.txt @@ -296,6 +296,12 @@ Cirrus Logic CS4206/4207 imac27 IMac 27 Inch auto BIOS setup (default) +Cirrus Logic CS4208 +=================== + mba6 MacBook Air 6,1 and 6,2 + gpio0 Enable GPIO 0 amp + auto BIOS setup (default) + VIA VT17xx/VT18xx/VT20xx ======================== auto BIOS setup (default) -- cgit v0.10.2 From 068fd3a0811c0aace07ff9c3e1d17bbd1802f4b9 Mon Sep 17 00:00:00 2001 From: Kailang Yang Date: Fri, 27 Sep 2013 14:02:28 +0200 Subject: ALSA: hda - Fix Internal Mic boost can't control with ALC283 ALC283 pin control for Line1 default control by hidden register. Use line1 as internal Mic will not get sound when boost value up. Set control by verb for hidden register will solve this issue. 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 bc07d36..41f18f4 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -3439,6 +3439,9 @@ static void alc283_fixup_chromebook(struct hda_codec *codec, /* Set to manual mode */ val = alc_read_coef_idx(codec, 0x06); alc_write_coef_idx(codec, 0x06, val & ~0x000c); + /* Enable Line1 input control by verb */ + val = alc_read_coef_idx(codec, 0x1a); + alc_write_coef_idx(codec, 0x1a, val | (1 << 4)); break; } } -- cgit v0.10.2 From 1bb3e062d4006bda733e8a0023b83f90700f663f Mon Sep 17 00:00:00 2001 From: Kailang Yang Date: Fri, 27 Sep 2013 13:10:25 +0200 Subject: ALSA: hda - Enable internal mic on a Thinkpad machine with ALC283 More thorough testing showed that these verbs were necessary to improve quality of the internal mic. Patch originally from Realtek. BugLink: https://bugs.launchpad.net/bugs/1231931 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 41f18f4..0e303b9 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -3534,6 +3534,7 @@ enum { ALC269VB_FIXUP_ORDISSIMO_EVE2, ALC283_FIXUP_CHROME_BOOK, ALC282_FIXUP_ASUS_TX300, + ALC283_FIXUP_INT_MIC, }; static const struct hda_fixup alc269_fixups[] = { @@ -3793,6 +3794,16 @@ static const struct hda_fixup alc269_fixups[] = { .type = HDA_FIXUP_FUNC, .v.func = alc282_fixup_asus_tx300, }, + [ALC283_FIXUP_INT_MIC] = { + .type = HDA_FIXUP_VERBS, + .v.verbs = (const struct hda_verb[]) { + {0x20, AC_VERB_SET_COEF_INDEX, 0x1a}, + {0x20, AC_VERB_SET_PROC_COEF, 0x0011}, + { } + }, + .chained = true, + .chain_id = ALC269_FIXUP_LIMIT_INT_MIC_BOOST + }, }; static const struct snd_pci_quirk alc269_fixup_tbl[] = { @@ -3877,7 +3888,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x2214, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x17aa, 0x2215, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x17aa, 0x5013, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), - SND_PCI_QUIRK(0x17aa, 0x501a, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), + SND_PCI_QUIRK(0x17aa, 0x501a, "Thinkpad", ALC283_FIXUP_INT_MIC), SND_PCI_QUIRK(0x17aa, 0x5026, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x17aa, 0x5109, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x17aa, 0x3bf8, "Quanta FL1", ALC269_FIXUP_PCM_44K), -- cgit v0.10.2 From 0c06a5d4b13cd66c833805a0d1db76b977944aac Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 10 Sep 2013 00:54:17 +0200 Subject: arm: Fix build error with context tracking calls ad65782fba50 (context_tracking: Optimize main APIs off case with static key) converted context tracking main APIs to inline function and left ARM asm callers behind. This can be easily fixed by making ARM calling the post static keys context tracking function. We just need to replicate the static key checks there. We'll remove these later when ARM will support the context tracking static keys. Reported-by: Guenter Roeck Reported-by: Russell King Signed-off-by: Frederic Weisbecker Tested-by: Kevin Hilman Cc: Nicolas Pitre Cc: Anil Kumar Cc: Tony Lindgren Cc: Benoit Cousson Cc: Guenter Roeck Cc: Russell King Cc: Kevin Hilman diff --git a/arch/arm/kernel/entry-header.S b/arch/arm/kernel/entry-header.S index de23a9b..39f89fb 100644 --- a/arch/arm/kernel/entry-header.S +++ b/arch/arm/kernel/entry-header.S @@ -329,10 +329,10 @@ #ifdef CONFIG_CONTEXT_TRACKING .if \save stmdb sp!, {r0-r3, ip, lr} - bl user_exit + bl context_tracking_user_exit ldmia sp!, {r0-r3, ip, lr} .else - bl user_exit + bl context_tracking_user_exit .endif #endif .endm @@ -341,10 +341,10 @@ #ifdef CONFIG_CONTEXT_TRACKING .if \save stmdb sp!, {r0-r3, ip, lr} - bl user_enter + bl context_tracking_user_enter ldmia sp!, {r0-r3, ip, lr} .else - bl user_enter + bl context_tracking_user_enter .endif #endif .endm diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 247091b..859c8df 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -51,6 +51,15 @@ void context_tracking_user_enter(void) unsigned long flags; /* + * Repeat the user_enter() check here because some archs may be calling + * this from asm and if no CPU needs context tracking, they shouldn't + * go further. Repeat the check here until they support the static key + * check. + */ + if (!static_key_false(&context_tracking_enabled)) + return; + + /* * Some contexts may involve an exception occuring in an irq, * leading to that nesting: * rcu_irq_enter() rcu_user_exit() rcu_user_exit() rcu_irq_exit() @@ -151,6 +160,9 @@ void context_tracking_user_exit(void) { unsigned long flags; + if (!static_key_false(&context_tracking_enabled)) + return; + if (in_interrupt()) return; -- cgit v0.10.2 From 153369139a06462c310b607e612093074c73b382 Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Tue, 17 Sep 2013 11:31:53 -0500 Subject: i2c: s3c2410: fix clk_disable/clk_unprepare WARNings commit d16933b33914a6dff38a4ecbe8edce44a17898e8 "i2c: s3c2410: Move location of clk_prepare_enable() call in probe function" refactored clk_enable and clk_disable calls yet neglected to remove the clk_disable_unprepare call in the module's remove(). It helps remove warnings on an arndale during unbind: echo 12c90000.i2c > /sys/bus/platform/devices/12c90000.i2c/driver/unbind ------------[ cut here ]------------ WARNING: CPU: 0 PID: 2548 at drivers/clk/clk.c:842 clk_disable+0x18/0x24() Modules linked in: CPU: 0 PID: 2548 Comm: bash Not tainted 3.11.0-next-20130916-00003-gf4bddbc #6 [] (unwind_backtrace+0x0/0xf8) from [] (show_stack+0x10/0x14) [] (show_stack+0x10/0x14) from [] (dump_stack+0x6c/0xac) [] (dump_stack+0x6c/0xac) from [] (warn_slowpath_common+0x64/0x88) [] (warn_slowpath_common+0x64/0x88) from [] (warn_slowpath_null+0x1c/0x24) [] (warn_slowpath_null+0x1c/0x24) from [] (clk_disable+0x18/0x24) [] (clk_disable+0x18/0x24) from [] (s3c24xx_i2c_remove+0x28/0x70) [] (s3c24xx_i2c_remove+0x28/0x70) from [] (platform_drv_remove+0x18/0x1c) [] (platform_drv_remove+0x18/0x1c) from [] (__device_release_driver+0x58/0xb4) [] (__device_release_driver+0x58/0xb4) from [] (device_release_driver+0x1c/0x28) [] (device_release_driver+0x1c/0x28) from [] (unbind_store+0x58/0x90) [] (unbind_store+0x58/0x90) from [] (drv_attr_store+0x20/0x2c) [] (drv_attr_store+0x20/0x2c) from [] (sysfs_write_file+0x168/0x198) [] (sysfs_write_file+0x168/0x198) from [] (vfs_write+0xb0/0x194) [] (vfs_write+0xb0/0x194) from [] (SyS_write+0x3c/0x70) [] (SyS_write+0x3c/0x70) from [] (ret_fast_syscall+0x0/0x30) ---[ end trace 4c9f9403066f57a6 ]--- ------------[ cut here ]------------ WARNING: CPU: 0 PID: 2548 at drivers/clk/clk.c:751 clk_unprepare+0x14/0x1c() Modules linked in: CPU: 0 PID: 2548 Comm: bash Tainted: G W 3.11.0-next-20130916-00003-gf4bddbc #6 [] (unwind_backtrace+0x0/0xf8) from [] (show_stack+0x10/0x14) [] (show_stack+0x10/0x14) from [] (dump_stack+0x6c/0xac) [] (dump_stack+0x6c/0xac) from [] (warn_slowpath_common+0x64/0x88) [] (warn_slowpath_common+0x64/0x88) from [] (warn_slowpath_null+0x1c/0x24) [] (warn_slowpath_null+0x1c/0x24) from [] (clk_unprepare+0x14/0x1c) [] (clk_unprepare+0x14/0x1c) from [] (s3c24xx_i2c_remove+0x30/0x70) [] (s3c24xx_i2c_remove+0x30/0x70) from [] (platform_drv_remove+0x18/0x1c) [] (platform_drv_remove+0x18/0x1c) from [] (__device_release_driver+0x58/0xb4) [] (__device_release_driver+0x58/0xb4) from [] (device_release_driver+0x1c/0x28) [] (device_release_driver+0x1c/0x28) from [] (unbind_store+0x58/0x90) [] (unbind_store+0x58/0x90) from [] (drv_attr_store+0x20/0x2c) [] (drv_attr_store+0x20/0x2c) from [] (sysfs_write_file+0x168/0x198) [] (sysfs_write_file+0x168/0x198) from [] (vfs_write+0xb0/0x194) [] (vfs_write+0xb0/0x194) from [] (SyS_write+0x3c/0x70) [] (SyS_write+0x3c/0x70) from [] (ret_fast_syscall+0x0/0x30) ---[ end trace 4c9f9403066f57a7 ]--- Signed-off-by: Kim Phillips Signed-off-by: Wolfram Sang diff --git a/drivers/i2c/busses/i2c-s3c2410.c b/drivers/i2c/busses/i2c-s3c2410.c index 3535f3c..3747b9b 100644 --- a/drivers/i2c/busses/i2c-s3c2410.c +++ b/drivers/i2c/busses/i2c-s3c2410.c @@ -1178,8 +1178,6 @@ static int s3c24xx_i2c_remove(struct platform_device *pdev) i2c_del_adapter(&i2c->adap); - clk_disable_unprepare(i2c->clk); - if (pdev->dev.of_node && IS_ERR(i2c->pctrl)) s3c24xx_i2c_dt_gpio_free(i2c); -- cgit v0.10.2 From c1a99467804767400752a4087d3e96af2759a046 Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Wed, 18 Sep 2013 14:50:52 +0200 Subject: i2c: mv64xxx: Fix some build warnings Some functions and variables are only used if the configuration selects HAVE_CLK. Protect them with a corresponding #ifdef CONFIG_HAVE_CLK block to avoid compiler warnings. Signed-off-by: Thierry Reding [wsa: added marker to #endif] Signed-off-by: Wolfram Sang diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c index 7f3a474..52c6943 100644 --- a/drivers/i2c/busses/i2c-mv64xxx.c +++ b/drivers/i2c/busses/i2c-mv64xxx.c @@ -697,6 +697,7 @@ static const struct of_device_id mv64xxx_i2c_of_match_table[] = { MODULE_DEVICE_TABLE(of, mv64xxx_i2c_of_match_table); #ifdef CONFIG_OF +#ifdef CONFIG_HAVE_CLK static int mv64xxx_calc_freq(const int tclk, const int n, const int m) { @@ -726,16 +727,12 @@ mv64xxx_find_baud_factors(const u32 req_freq, const u32 tclk, u32 *best_n, return false; return true; } +#endif /* CONFIG_HAVE_CLK */ static int mv64xxx_of_config(struct mv64xxx_i2c_data *drv_data, struct device *dev) { - const struct of_device_id *device; - struct device_node *np = dev->of_node; - u32 bus_freq, tclk; - int rc = 0; - /* CLK is mandatory when using DT to describe the i2c bus. We * need to know tclk in order to calculate bus clock * factors. @@ -744,6 +741,11 @@ mv64xxx_of_config(struct mv64xxx_i2c_data *drv_data, /* Have OF but no CLK */ return -ENODEV; #else + const struct of_device_id *device; + struct device_node *np = dev->of_node; + u32 bus_freq, tclk; + int rc = 0; + if (IS_ERR(drv_data->clk)) { rc = -ENODEV; goto out; -- cgit v0.10.2 From 85b3a9356e84f683dd27fe8b73ad15608b4fc2c5 Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Wed, 18 Sep 2013 14:51:40 +0200 Subject: i2c: mv64xxx: Do not use writel_relaxed() The driver is used on PowerPC which don't provide writel_relaxed(). This breaks the c2k and prpmc2800 default configurations. To fix the build, turn the calls to writel_relaxed() into writel(). The impacts for ARM should be minimal. Signed-off-by: Thierry Reding Signed-off-by: Wolfram Sang diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c index 52c6943..d3e9cc3 100644 --- a/drivers/i2c/busses/i2c-mv64xxx.c +++ b/drivers/i2c/busses/i2c-mv64xxx.c @@ -234,9 +234,9 @@ static int mv64xxx_i2c_offload_msg(struct mv64xxx_i2c_data *drv_data) ctrl_reg |= MV64XXX_I2C_BRIDGE_CONTROL_WR | (msg->len - 1) << MV64XXX_I2C_BRIDGE_CONTROL_TX_SIZE_SHIFT; - writel_relaxed(data_reg_lo, + writel(data_reg_lo, drv_data->reg_base + MV64XXX_I2C_REG_TX_DATA_LO); - writel_relaxed(data_reg_hi, + writel(data_reg_hi, drv_data->reg_base + MV64XXX_I2C_REG_TX_DATA_HI); } else { -- cgit v0.10.2 From bd63ace4dc4290165bbf3bf546eba50453d0aa9d Mon Sep 17 00:00:00 2001 From: "Chew, Chiau Ee" Date: Fri, 27 Sep 2013 02:57:35 +0800 Subject: i2c: designware: 10-bit addressing mode enabling if I2C_DYNAMIC_TAR_UPDATE is set According to Designware I2C spec, if I2C_DYNAMIC_TAR_UPDATE is set to 1, the 10-bit addressing mode is controlled by IC_10BITADDR_MASTER bit of IC_TAR register instead of IC_CON register. The IC_10BITADDR_MASTER in IC_CON register becomes read-only copy. Since I2C_DYNAMIC_TAR_UPDATE value can't be detected from hardware register, so we will always set the IC_10BITADDR_MASTER bit in both IC_CON and IC_TAR register whenever 10-bit addresing mode is requested by user application. Signed-off-by: Chew, Chiau Ee Reviewed-by: Mika Westerberg Signed-off-by: Wolfram Sang diff --git a/drivers/i2c/busses/i2c-designware-core.c b/drivers/i2c/busses/i2c-designware-core.c index dbecf08..5888fee 100644 --- a/drivers/i2c/busses/i2c-designware-core.c +++ b/drivers/i2c/busses/i2c-designware-core.c @@ -98,6 +98,8 @@ #define DW_IC_ERR_TX_ABRT 0x1 +#define DW_IC_TAR_10BITADDR_MASTER BIT(12) + /* * status codes */ @@ -388,22 +390,34 @@ static int i2c_dw_wait_bus_not_busy(struct dw_i2c_dev *dev) static void i2c_dw_xfer_init(struct dw_i2c_dev *dev) { struct i2c_msg *msgs = dev->msgs; - u32 ic_con; + u32 ic_con, ic_tar = 0; /* Disable the adapter */ __i2c_dw_enable(dev, false); - /* set the slave (target) address */ - dw_writel(dev, msgs[dev->msg_write_idx].addr, DW_IC_TAR); - /* if the slave address is ten bit address, enable 10BITADDR */ ic_con = dw_readl(dev, DW_IC_CON); - if (msgs[dev->msg_write_idx].flags & I2C_M_TEN) + if (msgs[dev->msg_write_idx].flags & I2C_M_TEN) { ic_con |= DW_IC_CON_10BITADDR_MASTER; - else + /* + * If I2C_DYNAMIC_TAR_UPDATE is set, the 10-bit addressing + * mode has to be enabled via bit 12 of IC_TAR register. + * We set it always as I2C_DYNAMIC_TAR_UPDATE can't be + * detected from registers. + */ + ic_tar = DW_IC_TAR_10BITADDR_MASTER; + } else { ic_con &= ~DW_IC_CON_10BITADDR_MASTER; + } + dw_writel(dev, ic_con, DW_IC_CON); + /* + * Set the slave (target) address and enable 10-bit addressing mode + * if applicable. + */ + dw_writel(dev, msgs[dev->msg_write_idx].addr | ic_tar, DW_IC_TAR); + /* Enable the adapter */ __i2c_dw_enable(dev, true); -- cgit v0.10.2 From d2212b4dce596fee83e5c523400bf084f4cc816c Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 26 Sep 2013 17:27:00 +0100 Subject: lockref: allow relaxed cmpxchg64 variant for lockless updates The 64-bit cmpxchg operation on the lockref is ordered by virtue of hazarding between the cmpxchg operation and the reference count manipulation. On weakly ordered memory architectures (such as ARM), it can be of great benefit to omit the barrier instructions where they are not needed. This patch moves the lockless lockref code over to a cmpxchg64_relaxed operation, which doesn't provide barrier semantics. If the operation isn't defined, we simply #define it as the usual 64-bit cmpxchg macro. Cc: Waiman Long Signed-off-by: Will Deacon Signed-off-by: Linus Torvalds diff --git a/lib/lockref.c b/lib/lockref.c index 677d036c..e294ae4 100644 --- a/lib/lockref.c +++ b/lib/lockref.c @@ -4,6 +4,14 @@ #ifdef CONFIG_CMPXCHG_LOCKREF /* + * Allow weakly-ordered memory architectures to provide barrier-less + * cmpxchg semantics for lockref updates. + */ +#ifndef cmpxchg64_relaxed +# define cmpxchg64_relaxed cmpxchg64 +#endif + +/* * Note that the "cmpxchg()" reloads the "old" value for the * failure case. */ @@ -14,8 +22,9 @@ while (likely(arch_spin_value_unlocked(old.lock.rlock.raw_lock))) { \ struct lockref new = old, prev = old; \ CODE \ - old.lock_count = cmpxchg64(&lockref->lock_count, \ - old.lock_count, new.lock_count); \ + old.lock_count = cmpxchg64_relaxed(&lockref->lock_count, \ + old.lock_count, \ + new.lock_count); \ if (likely(old.lock_count == prev.lock_count)) { \ SUCCESS; \ } \ -- cgit v0.10.2 From 730d7d339884f20da28b59bf6f0a16af6400a113 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 23 Sep 2013 14:41:17 -0700 Subject: sysfs: Allow mounting without CONFIG_NET In kobj_ns_current_may_mount the default should be to allow the mount. The test is only for a single kobj_ns_type at a time, and unless there is a reason to prevent it the mounting sysfs should be allowed. Subsystems that are not registered can't have are not involved so can't have a reason to prevent mounting sysfs. This is a bug-fix to commit 7dc5dbc879bd ("sysfs: Restrict mounting sysfs") that came in via the userns tree during the 3.12 merge window. Reported-and-tested-by: James Hogan Signed-off-by: "Eric W. Biederman" Signed-off-by: Linus Torvalds diff --git a/lib/kobject.c b/lib/kobject.c index 9621751..669bf19 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -933,10 +933,7 @@ const struct kobj_ns_type_operations *kobj_ns_ops(struct kobject *kobj) bool kobj_ns_current_may_mount(enum kobj_ns_type type) { - bool may_mount = false; - - if (type == KOBJ_NS_TYPE_NONE) - return true; + bool may_mount = true; spin_lock(&kobj_ns_type_lock); if ((type > KOBJ_NS_TYPE_NONE) && (type < KOBJ_NS_TYPES) && -- cgit v0.10.2 From 083986e8248d978b6c961d3da6beb0c921c68220 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sat, 28 Sep 2013 11:23:59 +0200 Subject: mutex: replace CONFIG_HAVE_ARCH_MUTEX_CPU_RELAX with simple ifdef Linus suggested to replace #ifndef CONFIG_HAVE_ARCH_MUTEX_CPU_RELAX #define arch_mutex_cpu_relax() cpu_relax() #endif with just a simple #ifndef arch_mutex_cpu_relax # define arch_mutex_cpu_relax() cpu_relax() #endif to get rid of CONFIG_HAVE_CPU_RELAX_SIMPLE. So architectures can simply define arch_mutex_cpu_relax if they want an architecture specific function instead of having to add a select statement in their Kconfig in addition. Suggested-by: Linus Torvalds Signed-off-by: Heiko Carstens diff --git a/arch/Kconfig b/arch/Kconfig index 1feb169..af2cc6e 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -286,9 +286,6 @@ config HAVE_PERF_USER_STACK_DUMP config HAVE_ARCH_JUMP_LABEL bool -config HAVE_ARCH_MUTEX_CPU_RELAX - bool - config HAVE_RCU_TABLE_FREE bool diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index dcc6ac2..d3fa840 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -102,7 +102,6 @@ config S390 select GENERIC_TIME_VSYSCALL_OLD select HAVE_ALIGNED_STRUCT_PAGE if SLUB select HAVE_ARCH_JUMP_LABEL if !MARCH_G5 - select HAVE_ARCH_MUTEX_CPU_RELAX select HAVE_ARCH_SECCOMP_FILTER select HAVE_ARCH_TRACEHOOK select HAVE_ARCH_TRANSPARENT_HUGEPAGE if 64BIT diff --git a/arch/s390/include/asm/mutex.h b/arch/s390/include/asm/mutex.h index 688271f..458c1f7 100644 --- a/arch/s390/include/asm/mutex.h +++ b/arch/s390/include/asm/mutex.h @@ -7,5 +7,3 @@ */ #include - -#define arch_mutex_cpu_relax() barrier() diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 0eb3750..ca7821f 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -198,6 +198,8 @@ static inline void cpu_relax(void) barrier(); } +#define arch_mutex_cpu_relax() barrier() + static inline void psw_set_key(unsigned int key) { asm volatile("spka 0(%0)" : : "d" (key)); diff --git a/include/linux/mutex.h b/include/linux/mutex.h index ccd4260..bab49da 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -15,8 +15,8 @@ #include #include #include - #include +#include /* * Simple, straightforward mutexes with strict semantics: @@ -175,8 +175,8 @@ extern void mutex_unlock(struct mutex *lock); extern int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock); -#ifndef CONFIG_HAVE_ARCH_MUTEX_CPU_RELAX -#define arch_mutex_cpu_relax() cpu_relax() +#ifndef arch_mutex_cpu_relax +# define arch_mutex_cpu_relax() cpu_relax() #endif #endif -- cgit v0.10.2 From 491f6f8e5fd9a57aaf03b6d6e3e153f1c27d8a46 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 23 Sep 2013 12:59:56 +0200 Subject: lockref: use arch_mutex_cpu_relax() in CMPXCHG_LOOP() Make use of arch_mutex_cpu_relax() so architectures can override the default cpu_relax() semantics. This is especially useful for s390, where cpu_relax() means that we yield() the current (virtual) cpu and therefore is very expensive, and would contradict the whole purpose of the lockless cmpxchg loop. Signed-off-by: Heiko Carstens diff --git a/lib/lockref.c b/lib/lockref.c index e294ae4..6f9d434 100644 --- a/lib/lockref.c +++ b/lib/lockref.c @@ -12,6 +12,14 @@ #endif /* + * Allow architectures to override the default cpu_relax() within CMPXCHG_LOOP. + * This is useful for architectures with an expensive cpu_relax(). + */ +#ifndef arch_mutex_cpu_relax +# define arch_mutex_cpu_relax() cpu_relax() +#endif + +/* * Note that the "cmpxchg()" reloads the "old" value for the * failure case. */ @@ -28,7 +36,7 @@ if (likely(old.lock_count == prev.lock_count)) { \ SUCCESS; \ } \ - cpu_relax(); \ + arch_mutex_cpu_relax(); \ } \ } while (0) -- cgit v0.10.2 From efc1d23b3d9af8cbff9f26677d67fb9c1b9cb792 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 5 Sep 2013 13:26:17 +0200 Subject: s390: enable ARCH_USE_CMPXCHG_LOCKREF Enable ARCH_USE_CMPXCHG_LOCKREF since it shows performance improvements with Linus' simple stat() test case of up to 50% on a 30 cpu system. Signed-off-by: Heiko Carstens diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index d3fa840..7143793 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -93,6 +93,7 @@ config S390 select ARCH_INLINE_WRITE_UNLOCK_IRQ select ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE select ARCH_SAVE_PAGE_KEYS if HIBERNATION + select ARCH_USE_CMPXCHG_LOCKREF select ARCH_WANT_IPC_PARSE_VERSION select BUILDTIME_EXTABLE_SORT select CLONE_BACKWARDS2 diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 701fe8c..83e5d216 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -44,6 +44,11 @@ extern void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); extern int arch_spin_trylock_retry(arch_spinlock_t *); extern void arch_spin_relax(arch_spinlock_t *lock); +static inline int arch_spin_value_unlocked(arch_spinlock_t lock) +{ + return lock.owner_cpu == 0; +} + static inline void arch_spin_lock(arch_spinlock_t *lp) { int old; -- cgit v0.10.2 From e4826a94c7444698725572f3def3686fb3f38d6d Mon Sep 17 00:00:00 2001 From: Thomas Meyer Date: Mon, 16 Sep 2013 23:19:54 +0200 Subject: drm/msm: Odd PTR_ERR usage The variable priv->kms is not initialized yet. Found by "scripts/coccinelle/tests/odd_ptr_err.cocci". PTR_ERR should access the value just tested by IS_ERR. Signed-off-by: Thomas Meyer diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 008d772..fe03f9c 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -199,7 +199,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags) * imx drm driver on iMX5 */ dev_err(dev->dev, "failed to load kms\n"); - ret = PTR_ERR(priv->kms); + ret = PTR_ERR(kms); goto fail; } -- cgit v0.10.2 From c55d1c41df68afaaf717850d735916ce3b355b77 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Wed, 25 Sep 2013 16:49:40 +0200 Subject: drm/msm: Remove iommu include from mdp4_kms.c The include file has been removed and the file does not need it anyway, so remove it. Fixes a compile error. Signed-off-by: Joerg Roedel Signed-off-by: Rob Clark diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/mdp4/mdp4_kms.c index 5db5bba..bc7fd11 100644 --- a/drivers/gpu/drm/msm/mdp4/mdp4_kms.c +++ b/drivers/gpu/drm/msm/mdp4/mdp4_kms.c @@ -19,8 +19,6 @@ #include "msm_drv.h" #include "mdp4_kms.h" -#include - static struct mdp4_platform_config *mdp4_get_config(struct platform_device *dev); static int mdp4_hw_init(struct msm_kms *kms) -- cgit v0.10.2 From 8a3da6c7d0031fcb6a0d17f9c7a68b0e01f52855 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 28 Sep 2013 15:48:48 +0200 Subject: perf/x86: Fix PMU detection printout when no PMU is detected Ran into this cryptic PMU bootup log recently: [ 0.124047] Performance Events: [ 0.125000] smpboot: ... Turns out we print this if no PMU is detected. Fall back to the right condition so that the following is printed: [ 0.122381] Performance Events: no PMU driver, software events only. Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Link: http://lkml.kernel.org/n/tip-u2fwaUffakjp0qkpRfqljgsn@git.kernel.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index a9c606b..897783b 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1506,7 +1506,7 @@ static int __init init_hw_perf_events(void) err = amd_pmu_init(); break; default: - return 0; + err = -ENOTSUPP; } if (err != 0) { pr_cont("no PMU driver, software events only.\n"); -- cgit v0.10.2 From 33b559630cf4345f2eeda849efd9b8548b192706 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Sat, 28 Sep 2013 10:07:06 -0400 Subject: drm/msm: deal with mach/iommu.h removal We still need an API exported by msm iommu driver (but not visible in any public header anymore). For now, just declare the prototype ourselves, but when msm iommu driver provides a better option, use that instead. Signed-off-by: Rob Clark diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index fe03f9c..9c871b5 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -18,8 +18,6 @@ #include "msm_drv.h" #include "msm_gpu.h" -#include - static void msm_fb_output_poll_changed(struct drm_device *dev) { struct msm_drm_private *priv = dev->dev_private; @@ -62,6 +60,8 @@ int msm_iommu_attach(struct drm_device *dev, struct iommu_domain *iommu, int i, ret; for (i = 0; i < cnt; i++) { + /* TODO maybe some day msm iommu won't require this hack: */ + struct device *msm_iommu_get_ctx(const char *ctx_name); struct device *ctx = msm_iommu_get_ctx(names[i]); if (!ctx) continue; -- cgit v0.10.2 From 30600a9092dc44694fbb4d0c3fa796e977fd5c96 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Sat, 28 Sep 2013 10:13:04 -0400 Subject: drm/msm: use drm_gem_dumb_destroy helper Signed-off-by: Rob Clark diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 9c871b5..b3a2f16 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -697,7 +697,7 @@ static struct drm_driver msm_driver = { .gem_vm_ops = &vm_ops, .dumb_create = msm_gem_dumb_create, .dumb_map_offset = msm_gem_dumb_map_offset, - .dumb_destroy = msm_gem_dumb_destroy, + .dumb_destroy = drm_gem_dumb_destroy, #ifdef CONFIG_DEBUG_FS .debugfs_init = msm_debugfs_init, .debugfs_cleanup = msm_debugfs_cleanup, diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 29eacfa..2bae46c 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -319,13 +319,6 @@ int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev, MSM_BO_SCANOUT | MSM_BO_WC, &args->handle); } -int msm_gem_dumb_destroy(struct drm_file *file, struct drm_device *dev, - uint32_t handle) -{ - /* No special work needed, drop the reference and see what falls out */ - return drm_gem_handle_delete(file, handle); -} - int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev, uint32_t handle, uint64_t *offset) { -- cgit v0.10.2 From bf4169100c909667ede6af67668b3ecce6928343 Mon Sep 17 00:00:00 2001 From: James Ralston Date: Tue, 24 Sep 2013 16:47:55 -0700 Subject: i2c: ismt: initialize DMA buffer This patch adds code to initialize the DMA buffer to compensate for possible hardware data corruption. Signed-off-by: James Ralston [wsa: changed to use 'sizeof'] Signed-off-by: Wolfram Sang diff --git a/drivers/i2c/busses/i2c-ismt.c b/drivers/i2c/busses/i2c-ismt.c index 8ed79a0..1672eff 100644 --- a/drivers/i2c/busses/i2c-ismt.c +++ b/drivers/i2c/busses/i2c-ismt.c @@ -393,6 +393,9 @@ static int ismt_access(struct i2c_adapter *adap, u16 addr, desc = &priv->hw[priv->head]; + /* Initialize the DMA buffer */ + memset(priv->dma_buffer, 0, sizeof(priv->dma_buffer)); + /* Initialize the descriptor */ memset(desc, 0, sizeof(struct ismt_desc)); desc->tgtaddr_rw = ISMT_DESC_ADDR_RW(addr, read_write); -- cgit v0.10.2 From 3a126f85e015701e56240884f27f97543580d5f7 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Fri, 27 Sep 2013 13:17:39 -0700 Subject: kernel/params: fix handling of signed integer types Commit 6072ddc8520b ("kernel: replace strict_strto*() with kstrto*()") broke the handling of signed integer types, fix it. Signed-off-by: Jean Delvare Reported-by: Christian Kujau Tested-by: Christian Kujau Cc: Jingoo Han Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/params.c b/kernel/params.c index 81c4e78..c00d5b5 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -254,11 +254,11 @@ int parse_args(const char *doing, STANDARD_PARAM_DEF(byte, unsigned char, "%hhu", unsigned long, kstrtoul); -STANDARD_PARAM_DEF(short, short, "%hi", long, kstrtoul); +STANDARD_PARAM_DEF(short, short, "%hi", long, kstrtol); STANDARD_PARAM_DEF(ushort, unsigned short, "%hu", unsigned long, kstrtoul); -STANDARD_PARAM_DEF(int, int, "%i", long, kstrtoul); +STANDARD_PARAM_DEF(int, int, "%i", long, kstrtol); STANDARD_PARAM_DEF(uint, unsigned int, "%u", unsigned long, kstrtoul); -STANDARD_PARAM_DEF(long, long, "%li", long, kstrtoul); +STANDARD_PARAM_DEF(long, long, "%li", long, kstrtol); STANDARD_PARAM_DEF(ulong, unsigned long, "%lu", unsigned long, kstrtoul); int param_set_charp(const char *val, const struct kernel_param *kp) -- cgit v0.10.2 From f41f064cf4352e6a7fd982f1de8a690897702513 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sat, 28 Sep 2013 13:13:07 -0700 Subject: PCI: Workaround missing pci_set_master in pci drivers Ben Herrenschmidt found that commit 928bea964827 ("PCI: Delay enabling bridges until they're needed") breaks PCI in some powerpc environments. The reason is that the PCIe port driver will call pci_enable_device() on the bridge, so the device is enabled, but skips pci_set_master because pcie_port_auto and no acpi on powerpc. Because of that, pci_enable_bridge() later on (called as a result of the child device driver doing pci_enable_device) will see the bridge as already enabled and will not call pci_set_master() on it. Fixed by add checking in pci_enable_bridge, and call pci_set_master if driver skip that. That will make the code more robot and wade off problem for missing pci_set_master in drivers. Reported-by: Benjamin Herrenschmidt Signed-off-by: Yinghai Lu Signed-off-by: Linus Torvalds diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index e8ccf6c..bdd64b1 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1155,8 +1155,14 @@ static void pci_enable_bridge(struct pci_dev *dev) pci_enable_bridge(dev->bus->self); - if (pci_is_enabled(dev)) + if (pci_is_enabled(dev)) { + if (!dev->is_busmaster) { + dev_warn(&dev->dev, "driver skip pci_set_master, fix it!\n"); + pci_set_master(dev); + } return; + } + retval = pci_enable_device(dev); if (retval) dev_err(&dev->dev, "Error enabling bridge (%d), continuing\n", -- cgit v0.10.2 From 14951f22f1cc8375ead345d2ca08455e91f2152b Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 29 Sep 2013 16:12:54 +0200 Subject: Revert "perf symbols: Demangle cloned functions" This reverts commit de95ab53645a2f0015e0f68ee723f18dce2b8b51. Markus Trippelsdorf reported that this commit broke 'perf top': > I just see a gray screen with no text at all. Sometimes the > following error messages are printed: > > *** Error in `perf': invalid fastbin entry (free): 0x00000000029b18c0 > *** > *** Error in `perf': malloc(): memory corruption (fast): 0x0000000000ee0b10 *** While this code is fixable, the commit itself fails on several levels: - it should have been a separate helper function - why the heck does it do strchr() twice - it casts a const char * over into char * - sloppy style - it's not even a regression fix! So lets revert it and re-try the patch in v3.13. Reported-by: Markus Trippelsdorf Cc: Andi Kleen Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Linus Torvalds Signed-off-by: Ingo Molnar diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index d2a888e..a9c829b 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -928,33 +928,8 @@ int dso__load_sym(struct dso *dso, struct map *map, * to it... */ if (symbol_conf.demangle) { - /* - * The demangler doesn't deal with cloned functions. - * XXXX.clone.NUM or similar - * Strip the dot part and readd it later. - */ - char *p = (char *)elf_name, *dot; - dot = strchr(elf_name, '.'); - if (dot) { - p = strdup(elf_name); - if (!p) - goto new_symbol; - dot = strchr(p, '.'); - *dot = 0; - } - - demangled = bfd_demangle(NULL, p, + demangled = bfd_demangle(NULL, elf_name, DMGL_PARAMS | DMGL_ANSI); - if (dot) - *dot = '.'; - if (demangled && dot) { - demangled = realloc(demangled, strlen(demangled) + strlen(dot) + 1); - if (!demangled) - goto new_symbol; - strcpy(demangled + (dot - p), dot); - } - if (p != elf_name) - free(p); if (demangled != NULL) elf_name = demangled; } -- cgit v0.10.2 From 15c03dd4859ab16f9212238f29dd315654aa94f6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 29 Sep 2013 15:02:38 -0700 Subject: Linux 3.12-rc3 diff --git a/Makefile b/Makefile index 8d0668f..2ae108d 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 3 PATCHLEVEL = 12 SUBLEVEL = 0 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc3 NAME = One Giant Leap for Frogkind # *DOCUMENTATION* -- cgit v0.10.2