From da259465d7526804b21d274281fb4d60b4216c82 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 2 Apr 2013 01:25:09 +0200 Subject: USB / PM: Don't try to hide PM QoS flags from usb_port_device_release() Remove the call to dev_pm_qos_hide_flags(), added by commit 6e30d7cb "usb: Add driver/usb/core/(port.c,hub.h) files", from usb_port_device_release(), because (1) it is completely unnecessary (the flags have been removed already by the PM core during the unregistration of the device object) and (2) it triggers a NULL pointer dereference in sysfs_find_dirent() (dev->kobj.sd is NULL at this point). Signed-off-by: Rafael J. Wysocki Acked-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/port.c b/drivers/usb/core/port.c index 797f9d5..65d4e55 100644 --- a/drivers/usb/core/port.c +++ b/drivers/usb/core/port.c @@ -67,7 +67,6 @@ static void usb_port_device_release(struct device *dev) { struct usb_port *port_dev = to_usb_port(dev); - dev_pm_qos_hide_flags(dev); kfree(port_dev); } -- cgit v0.10.2 From 0f703069296664eb7c649c837cc8bb936c3ef07f Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 2 Apr 2013 01:25:24 +0200 Subject: PM / QoS: Avoid possible deadlock related to sysfs access Commit b81ea1b (PM / QoS: Fix concurrency issues and memory leaks in device PM QoS) put calls to pm_qos_sysfs_add_latency(), pm_qos_sysfs_add_flags(), pm_qos_sysfs_remove_latency(), and pm_qos_sysfs_remove_flags() under dev_pm_qos_mtx, which was a mistake, because it may lead to deadlocks in some situations. For example, if pm_qos_remote_wakeup_store() is run in parallel with dev_pm_qos_constraints_destroy(), they may deadlock in the following way: ====================================================== [ INFO: possible circular locking dependency detected ] 3.9.0-rc4-next-20130328-sasha-00014-g91a3267 #319 Tainted: G W ------------------------------------------------------- trinity-child6/12371 is trying to acquire lock: (s_active#54){++++.+}, at: [] sysfs_addrm_finish+0x31/0x60 but task is already holding lock: (dev_pm_qos_mtx){+.+.+.}, at: [] dev_pm_qos_constraints_destroy+0x23/0x250 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (dev_pm_qos_mtx){+.+.+.}: [] lock_acquire+0x1aa/0x240 [] __mutex_lock_common+0x59/0x5e0 [] mutex_lock_nested+0x3f/0x50 [] dev_pm_qos_update_flags+0x3f/0xc0 [] pm_qos_remote_wakeup_store+0x3f/0x70 [] dev_attr_store+0x13/0x20 [] sysfs_write_file+0xfa/0x150 [] __kernel_write+0x81/0x150 [] write_pipe_buf+0x4d/0x80 [] splice_from_pipe_feed+0x7c/0x120 [] __splice_from_pipe+0x45/0x80 [] splice_from_pipe+0x4c/0x70 [] default_file_splice_write+0x18/0x30 [] do_splice_from+0x83/0xb0 [] direct_splice_actor+0x1e/0x20 [] splice_direct_to_actor+0xe7/0x200 [] do_splice_direct+0x4c/0x70 [] do_sendfile+0x169/0x300 [] SyS_sendfile64+0x64/0xb0 [] tracesys+0xe1/0xe6 -> #0 (s_active#54){++++.+}: [] __lock_acquire+0x15bf/0x1e50 [] lock_acquire+0x1aa/0x240 [] sysfs_deactivate+0x122/0x1a0 [] sysfs_addrm_finish+0x31/0x60 [] sysfs_hash_and_remove+0x7f/0xb0 [] sysfs_unmerge_group+0x51/0x70 [] pm_qos_sysfs_remove_flags+0x14/0x20 [] __dev_pm_qos_hide_flags+0x30/0x70 [] dev_pm_qos_constraints_destroy+0x35/0x250 [] dpm_sysfs_remove+0x11/0x50 [] device_del+0x3f/0x1b0 [] device_unregister+0x48/0x60 [] usb_hub_remove_port_device+0x1c/0x20 [] hub_disconnect+0xdd/0x160 [] usb_unbind_interface+0x67/0x170 [] __device_release_driver+0x87/0xe0 [] device_release_driver+0x29/0x40 [] bus_remove_device+0x148/0x160 [] device_del+0x14f/0x1b0 [] usb_disable_device+0xf9/0x280 [] usb_set_configuration+0x268/0x840 [] usb_remove_store+0x4c/0x80 [] dev_attr_store+0x13/0x20 [] sysfs_write_file+0xfa/0x150 [] do_loop_readv_writev+0x4d/0x90 [] do_readv_writev+0xf9/0x1e0 [] vfs_writev+0x3a/0x60 [] SyS_writev+0x50/0xd0 [] tracesys+0xe1/0xe6 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(dev_pm_qos_mtx); lock(s_active#54); lock(dev_pm_qos_mtx); lock(s_active#54); *** DEADLOCK *** To avoid that, remove the calls to functions mentioned above from under dev_pm_qos_mtx and introduce a separate lock to prevent races between functions that add or remove device PM QoS sysfs attributes from happening. Reported-by: Sasha Levin Signed-off-by: Rafael J. Wysocki diff --git a/drivers/base/power/qos.c b/drivers/base/power/qos.c index 5f74587e..71671c4 100644 --- a/drivers/base/power/qos.c +++ b/drivers/base/power/qos.c @@ -46,6 +46,7 @@ #include "power.h" static DEFINE_MUTEX(dev_pm_qos_mtx); +static DEFINE_MUTEX(dev_pm_qos_sysfs_mtx); static BLOCKING_NOTIFIER_HEAD(dev_pm_notifiers); @@ -216,12 +217,17 @@ void dev_pm_qos_constraints_destroy(struct device *dev) struct pm_qos_constraints *c; struct pm_qos_flags *f; - mutex_lock(&dev_pm_qos_mtx); + mutex_lock(&dev_pm_qos_sysfs_mtx); /* * If the device's PM QoS resume latency limit or PM QoS flags have been * exposed to user space, they have to be hidden at this point. */ + pm_qos_sysfs_remove_latency(dev); + pm_qos_sysfs_remove_flags(dev); + + mutex_lock(&dev_pm_qos_mtx); + __dev_pm_qos_hide_latency_limit(dev); __dev_pm_qos_hide_flags(dev); @@ -254,6 +260,8 @@ void dev_pm_qos_constraints_destroy(struct device *dev) out: mutex_unlock(&dev_pm_qos_mtx); + + mutex_unlock(&dev_pm_qos_sysfs_mtx); } /** @@ -558,6 +566,14 @@ static void __dev_pm_qos_drop_user_request(struct device *dev, kfree(req); } +static void dev_pm_qos_drop_user_request(struct device *dev, + enum dev_pm_qos_req_type type) +{ + mutex_lock(&dev_pm_qos_mtx); + __dev_pm_qos_drop_user_request(dev, type); + mutex_unlock(&dev_pm_qos_mtx); +} + /** * dev_pm_qos_expose_latency_limit - Expose PM QoS latency limit to user space. * @dev: Device whose PM QoS latency limit is to be exposed to user space. @@ -581,6 +597,8 @@ int dev_pm_qos_expose_latency_limit(struct device *dev, s32 value) return ret; } + mutex_lock(&dev_pm_qos_sysfs_mtx); + mutex_lock(&dev_pm_qos_mtx); if (IS_ERR_OR_NULL(dev->power.qos)) @@ -591,26 +609,27 @@ int dev_pm_qos_expose_latency_limit(struct device *dev, s32 value) if (ret < 0) { __dev_pm_qos_remove_request(req); kfree(req); + mutex_unlock(&dev_pm_qos_mtx); goto out; } - dev->power.qos->latency_req = req; + + mutex_unlock(&dev_pm_qos_mtx); + ret = pm_qos_sysfs_add_latency(dev); if (ret) - __dev_pm_qos_drop_user_request(dev, DEV_PM_QOS_LATENCY); + dev_pm_qos_drop_user_request(dev, DEV_PM_QOS_LATENCY); out: - mutex_unlock(&dev_pm_qos_mtx); + mutex_unlock(&dev_pm_qos_sysfs_mtx); return ret; } EXPORT_SYMBOL_GPL(dev_pm_qos_expose_latency_limit); static void __dev_pm_qos_hide_latency_limit(struct device *dev) { - if (!IS_ERR_OR_NULL(dev->power.qos) && dev->power.qos->latency_req) { - pm_qos_sysfs_remove_latency(dev); + if (!IS_ERR_OR_NULL(dev->power.qos) && dev->power.qos->latency_req) __dev_pm_qos_drop_user_request(dev, DEV_PM_QOS_LATENCY); - } } /** @@ -619,9 +638,15 @@ static void __dev_pm_qos_hide_latency_limit(struct device *dev) */ void dev_pm_qos_hide_latency_limit(struct device *dev) { + mutex_lock(&dev_pm_qos_sysfs_mtx); + + pm_qos_sysfs_remove_latency(dev); + mutex_lock(&dev_pm_qos_mtx); __dev_pm_qos_hide_latency_limit(dev); mutex_unlock(&dev_pm_qos_mtx); + + mutex_unlock(&dev_pm_qos_sysfs_mtx); } EXPORT_SYMBOL_GPL(dev_pm_qos_hide_latency_limit); @@ -649,6 +674,8 @@ int dev_pm_qos_expose_flags(struct device *dev, s32 val) } pm_runtime_get_sync(dev); + mutex_lock(&dev_pm_qos_sysfs_mtx); + mutex_lock(&dev_pm_qos_mtx); if (IS_ERR_OR_NULL(dev->power.qos)) @@ -659,16 +686,19 @@ int dev_pm_qos_expose_flags(struct device *dev, s32 val) if (ret < 0) { __dev_pm_qos_remove_request(req); kfree(req); + mutex_unlock(&dev_pm_qos_mtx); goto out; } - dev->power.qos->flags_req = req; + + mutex_unlock(&dev_pm_qos_mtx); + ret = pm_qos_sysfs_add_flags(dev); if (ret) - __dev_pm_qos_drop_user_request(dev, DEV_PM_QOS_FLAGS); + dev_pm_qos_drop_user_request(dev, DEV_PM_QOS_FLAGS); out: - mutex_unlock(&dev_pm_qos_mtx); + mutex_unlock(&dev_pm_qos_sysfs_mtx); pm_runtime_put(dev); return ret; } @@ -676,10 +706,8 @@ EXPORT_SYMBOL_GPL(dev_pm_qos_expose_flags); static void __dev_pm_qos_hide_flags(struct device *dev) { - if (!IS_ERR_OR_NULL(dev->power.qos) && dev->power.qos->flags_req) { - pm_qos_sysfs_remove_flags(dev); + if (!IS_ERR_OR_NULL(dev->power.qos) && dev->power.qos->flags_req) __dev_pm_qos_drop_user_request(dev, DEV_PM_QOS_FLAGS); - } } /** @@ -689,9 +717,15 @@ static void __dev_pm_qos_hide_flags(struct device *dev) void dev_pm_qos_hide_flags(struct device *dev) { pm_runtime_get_sync(dev); + mutex_lock(&dev_pm_qos_sysfs_mtx); + + pm_qos_sysfs_remove_flags(dev); + mutex_lock(&dev_pm_qos_mtx); __dev_pm_qos_hide_flags(dev); mutex_unlock(&dev_pm_qos_mtx); + + mutex_unlock(&dev_pm_qos_sysfs_mtx); pm_runtime_put(dev); } EXPORT_SYMBOL_GPL(dev_pm_qos_hide_flags); -- cgit v0.10.2 From 5faaa035ddc16d3d2e31c0bfc1c4756b6869dbcc Mon Sep 17 00:00:00 2001 From: Rajagopal Venkat Date: Thu, 21 Mar 2013 13:28:25 +0000 Subject: PM / devfreq: Fix compiler warnings for CONFIG_PM_DEVFREQ unset Fix compiler warnings generated when devfreq is not enabled (CONFIG_PM_DEVFREQ is not set). Signed-off-by: Rajagopal Venkat Acked-by: MyungJoo Ham Signed-off-by: Rafael J. Wysocki diff --git a/include/linux/devfreq.h b/include/linux/devfreq.h index e83ef39..fe8c447 100644 --- a/include/linux/devfreq.h +++ b/include/linux/devfreq.h @@ -213,7 +213,7 @@ struct devfreq_simple_ondemand_data { #endif #else /* !CONFIG_PM_DEVFREQ */ -static struct devfreq *devfreq_add_device(struct device *dev, +static inline struct devfreq *devfreq_add_device(struct device *dev, struct devfreq_dev_profile *profile, const char *governor_name, void *data) @@ -221,34 +221,34 @@ static struct devfreq *devfreq_add_device(struct device *dev, return NULL; } -static int devfreq_remove_device(struct devfreq *devfreq) +static inline int devfreq_remove_device(struct devfreq *devfreq) { return 0; } -static int devfreq_suspend_device(struct devfreq *devfreq) +static inline int devfreq_suspend_device(struct devfreq *devfreq) { return 0; } -static int devfreq_resume_device(struct devfreq *devfreq) +static inline int devfreq_resume_device(struct devfreq *devfreq) { return 0; } -static struct opp *devfreq_recommended_opp(struct device *dev, +static inline struct opp *devfreq_recommended_opp(struct device *dev, unsigned long *freq, u32 flags) { - return -EINVAL; + return ERR_PTR(-EINVAL); } -static int devfreq_register_opp_notifier(struct device *dev, +static inline int devfreq_register_opp_notifier(struct device *dev, struct devfreq *devfreq) { return -EINVAL; } -static int devfreq_unregister_opp_notifier(struct device *dev, +static inline int devfreq_unregister_opp_notifier(struct device *dev, struct devfreq *devfreq) { return -EINVAL; -- cgit v0.10.2 From f5c3ef21db149ba573a3f079df287f3d8b05ee13 Mon Sep 17 00:00:00 2001 From: Paolo Pisati Date: Thu, 28 Mar 2013 09:24:29 +0000 Subject: cpufreq: check OF node /cpus presence before dereferencing it Check for the presence of the '/cpus' OF node before dereferencing it blindly: [ 4.181793] Unable to handle kernel NULL pointer dereference at virtual address 0000001c [ 4.181793] pgd = c0004000 [ 4.181823] [0000001c] *pgd=00000000 [ 4.181823] Internal error: Oops: 5 [#1] SMP ARM [ 4.181823] Modules linked in: [ 4.181823] CPU: 1 Tainted: G W (3.8.0-15-generic #25~hbankD) [ 4.181854] PC is at of_get_next_child+0x64/0x70 [ 4.181854] LR is at of_get_next_child+0x24/0x70 [ 4.181854] pc : [] lr : [] psr: 60000113 [ 4.181854] sp : ed891ec0 ip : ed891ec0 fp : ed891ed4 [ 4.181884] r10: c04dafd0 r9 : c098690c r8 : c0936208 [ 4.181884] r7 : ed890000 r6 : c0a63d00 r5 : 00000000 r4 : 00000000 [ 4.181884] r3 : 00000000 r2 : 00000000 r1 : 00000000 r0 : c0b2acc8 [ 4.181884] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel [ 4.181884] Control: 10c5387d Table: adcb804a DAC: 00000015 [ 4.181915] Process swapper/0 (pid: 1, stack limit = 0xed890238) [ 4.181915] Stack: (0xed891ec0 to 0xed892000) [ 4.181915] 1ec0: c09b7b70 00000007 ed891efc ed891ed8 c04daff4 c04fd9c0 00000000 c09b7b70 [ 4.181915] 1ee0: 00000007 c0a63d00 ed890000 c0936208 ed891f54 ed891f00 c00088e0 c04dafdc [ 4.181945] 1f00: ed891f54 ed891f10 c006e940 00000000 00000000 00000007 00000007 c08a4914 [ 4.181945] 1f20: 00000000 c07dbd30 c0a63d00 c09b7b70 00000007 c0a63d00 000000bc c0936208 [ 4.181945] 1f40: c098690c c0986914 ed891f94 ed891f58 c0936a40 c00087bc 00000007 00000007 [ 4.181976] 1f60: c0936208 be8bda20 b6eea010 c0a63d00 c064547c 00000000 00000000 00000000 [ 4.181976] 1f80: 00000000 00000000 ed891fac ed891f98 c0645498 c09368c8 00000000 00000000 [ 4.181976] 1fa0: 00000000 ed891fb0 c0014658 c0645488 00000000 00000000 00000000 00000000 [ 4.182006] 1fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 [ 4.182006] 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000 00000000 00000000 [ 4.182037] [] (of_get_next_child+0x64/0x70) from [] (cpu0_cpufreq_driver_init+0x24/0x284) [ 4.182067] [] (cpu0_cpufreq_driver_init+0x24/0x284) from [] (do_one_initcall+0x130/0x1b0) [ 4.182067] [] (do_one_initcall+0x130/0x1b0) from [] (kernel_init_freeable+0x184/0x24c) [ 4.182098] [] (kernel_init_freeable+0x184/0x24c) from [] (kernel_init+0x1c/0xf4) [ 4.182128] [] (kernel_init+0x1c/0xf4) from [] (ret_from_fork+0x14/0x20) [ 4.182128] Code: f57ff04f e320f004 e89da830 e89da830 (e595001c) [ 4.182128] ---[ end trace 634903a22e8609cb ]--- [ 4.182189] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b [ 4.182189] [ 4.642395] CPU0: stopping [rjw: Changelog] Signed-off-by: Paolo Pisati Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki diff --git a/drivers/cpufreq/cpufreq-cpu0.c b/drivers/cpufreq/cpufreq-cpu0.c index 4e5b7fb..37d23a0 100644 --- a/drivers/cpufreq/cpufreq-cpu0.c +++ b/drivers/cpufreq/cpufreq-cpu0.c @@ -178,10 +178,16 @@ static struct cpufreq_driver cpu0_cpufreq_driver = { static int cpu0_cpufreq_probe(struct platform_device *pdev) { - struct device_node *np; + struct device_node *np, *parent; int ret; - for_each_child_of_node(of_find_node_by_path("/cpus"), np) { + parent = of_find_node_by_path("/cpus"); + if (!parent) { + pr_err("failed to find OF /cpus\n"); + return -ENOENT; + } + + for_each_child_of_node(parent, np) { if (of_get_property(np, "operating-points", NULL)) break; } -- cgit v0.10.2 From 29896178cff0cd6e4ee78edc26158cc197a6b504 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 1 Apr 2013 00:21:08 +0000 Subject: ACPI / SPI: Use parent's ACPI_HANDLE() in acpi_register_spi_devices() The ACPI handle of struct spi_master's dev member should not be set, because this causes that struct spi_master to be associated with the ACPI device node corresponding to its parent as the second "physical_device", which is incorrect (this happens during the registration of struct spi_master). Consequently, acpi_register_spi_devices() should use the ACPI handle of the parent of the struct spi_master it is called for rather than that struct spi_master's ACPI handle (which should be NULL). Make that happen and modify the spi-pxa2xx driver, which currently is the only driver for ACPI-enumerated SPI controller chips, not to set the ACPI handle for the struct spi_master it creates. Signed-off-by: Rafael J. Wysocki Acked-by: Mika Westerberg Acked-by: Mark Brown diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c index 90b27a3..8104138 100644 --- a/drivers/spi/spi-pxa2xx.c +++ b/drivers/spi/spi-pxa2xx.c @@ -1168,7 +1168,6 @@ static int pxa2xx_spi_probe(struct platform_device *pdev) master->dev.parent = &pdev->dev; master->dev.of_node = pdev->dev.of_node; - ACPI_HANDLE_SET(&master->dev, ACPI_HANDLE(&pdev->dev)); /* the spi->mode bits understood by this driver: */ master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH | SPI_LOOP; diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index f996c60..5e073d8 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -984,7 +984,7 @@ static void acpi_register_spi_devices(struct spi_master *master) acpi_status status; acpi_handle handle; - handle = ACPI_HANDLE(&master->dev); + handle = ACPI_HANDLE(master->dev.parent); if (!handle) return; -- cgit v0.10.2 From beb0ff390e208e71668a1389e3db02631a88cff5 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Tue, 2 Apr 2013 12:26:15 +0000 Subject: cpufreq: Correct header guards typo It should be "governor". Signed-off-by: Borislav Petkov Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki diff --git a/drivers/cpufreq/cpufreq_governor.h b/drivers/cpufreq/cpufreq_governor.h index 46bde01..cc4bd2f 100644 --- a/drivers/cpufreq/cpufreq_governor.h +++ b/drivers/cpufreq/cpufreq_governor.h @@ -14,8 +14,8 @@ * published by the Free Software Foundation. */ -#ifndef _CPUFREQ_GOVERNER_H -#define _CPUFREQ_GOVERNER_H +#ifndef _CPUFREQ_GOVERNOR_H +#define _CPUFREQ_GOVERNOR_H #include #include @@ -175,4 +175,4 @@ bool need_load_eval(struct cpu_dbs_common_info *cdbs, unsigned int sampling_rate); int cpufreq_governor_dbs(struct dbs_data *dbs_data, struct cpufreq_policy *policy, unsigned int event); -#endif /* _CPUFREQ_GOVERNER_H */ +#endif /* _CPUFREQ_GOVERNOR_H */ -- cgit v0.10.2 From b34bb1ee71158d5b0f9028fb98afd026202bcfe9 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 1 Apr 2013 00:25:03 +0000 Subject: ACPI / I2C: Use parent's ACPI_HANDLE() in acpi_i2c_register_devices() The ACPI handle of struct i2c_adapter's dev member should not be set, because this causes that struct i2c_adapter to be associated with the ACPI device node corresponding to its parent as the second "physical_device", which is incorrect (this happens during the registration of struct i2c_adapter). Consequently, acpi_i2c_register_devices() should use the ACPI handle of the parent of the struct i2c_adapter it is called for rather than the struct i2c_adapter's ACPI handle (which should be NULL). Make that happen and modify the i2c-designware-platdrv driver, which currently is the only driver for ACPI-enumerated I2C controller chips, not to set the ACPI handle for the struct i2c_adapter it creates. Signed-off-by: Rafael J. Wysocki Reviewed-by: Aaron Lu Acked-by: Mika Westerberg Acked-by: Wolfram Sang diff --git a/drivers/acpi/acpi_i2c.c b/drivers/acpi/acpi_i2c.c index 82045e3..a82c762 100644 --- a/drivers/acpi/acpi_i2c.c +++ b/drivers/acpi/acpi_i2c.c @@ -90,7 +90,7 @@ void acpi_i2c_register_devices(struct i2c_adapter *adapter) acpi_handle handle; acpi_status status; - handle = ACPI_HANDLE(&adapter->dev); + handle = ACPI_HANDLE(adapter->dev.parent); if (!handle) return; diff --git a/drivers/i2c/busses/i2c-designware-platdrv.c b/drivers/i2c/busses/i2c-designware-platdrv.c index 0ceb6e1..e3085c4 100644 --- a/drivers/i2c/busses/i2c-designware-platdrv.c +++ b/drivers/i2c/busses/i2c-designware-platdrv.c @@ -182,7 +182,6 @@ static int dw_i2c_probe(struct platform_device *pdev) adap->algo = &i2c_dw_algo; adap->dev.parent = &pdev->dev; adap->dev.of_node = pdev->dev.of_node; - ACPI_HANDLE_SET(&adap->dev, ACPI_HANDLE(&pdev->dev)); r = i2c_add_numbered_adapter(adap); if (r) { -- cgit v0.10.2 From 6240a10dc55a954ce05a86d2f4bc27e473b6352c Mon Sep 17 00:00:00 2001 From: Alex Shi Date: Tue, 2 Apr 2013 01:56:54 +0200 Subject: cpuidle / ACPI: recover percpu ACPI processor cstate Commit ac3ebafa81af76d6 "ACPI / idle: remove usage of the statedata" changed the percpu processor cstate to a unified cstate in ACPI idle. That caused all our NHM boxes to boot hang or panic. 2178751 Task dump for CPU 1: 2178752 swapper/1 R running task 6736 0 1 0x00000000 2178753 ffff8801e8029dc8 ffffffff8101cf96 ffff8801e8029e28 ffffffff813d294b 2178754 0000000000000f99 0000000000000003 00000000003cf654 0000000025c17d03 2178755 ffff8801e8029e38 ffff8801e74fc000 00000002590dc5c4 ffffffff8163cdb0 2178756 Call Trace: 2178757 [] ? acpi_processor_ffh_cstate_enter+0x2d/0x2f 2178758 [] acpi_idle_enter_bm+0x1b1/0x236 2178759 [] ? disable_cpuidle+0x10/0x10 2178760 [] cpuidle_enter+0x12/0x14 2178761 [] cpuidle_wrap_enter+0x2f/0x6d 2178762 [] cpuidle_enter_tk+0x10/0x12 2178763 [] cpuidle_enter_state+0x12/0x3a 2178764 [] cpuidle_idle_call+0xe8/0x161 2178765 [] cpu_idle+0x5e/0xa4 2178766 [] start_secondary+0x1a9/0x1ad 2178767 Task dump for CPU 2: In fact, the ACPI idle is based on the assumption of difference percpu cstate structures that are necessary for the implementation to work cprrectly. A unique acpi_processor_cx is not sifficient by far. This patch is just a quick fix re-introducing the percpu cstates. If someone really wants to unify the ACPI cstates, please make sure that the whole software infrastructure is changed and take hardware as well as many different kinds of BIOS settings into account. [rjw: Changelog] Reported-by: LKP project Reported-by: Xie ChanglongX Tested-by: Yinghai Lu Signed-off-by: Alex Shi Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index fc95308..ee255c6 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -66,7 +66,8 @@ module_param(latency_factor, uint, 0644); static DEFINE_PER_CPU(struct cpuidle_device *, acpi_cpuidle_device); -static struct acpi_processor_cx *acpi_cstate[CPUIDLE_STATE_MAX]; +static DEFINE_PER_CPU(struct acpi_processor_cx * [CPUIDLE_STATE_MAX], + acpi_cstate); static int disabled_by_idle_boot_param(void) { @@ -722,7 +723,7 @@ static int acpi_idle_enter_c1(struct cpuidle_device *dev, struct cpuidle_driver *drv, int index) { struct acpi_processor *pr; - struct acpi_processor_cx *cx = acpi_cstate[index]; + struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu); pr = __this_cpu_read(processors); @@ -745,7 +746,7 @@ static int acpi_idle_enter_c1(struct cpuidle_device *dev, */ static int acpi_idle_play_dead(struct cpuidle_device *dev, int index) { - struct acpi_processor_cx *cx = acpi_cstate[index]; + struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu); ACPI_FLUSH_CPU_CACHE(); @@ -775,7 +776,7 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev, struct cpuidle_driver *drv, int index) { struct acpi_processor *pr; - struct acpi_processor_cx *cx = acpi_cstate[index]; + struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu); pr = __this_cpu_read(processors); @@ -833,7 +834,7 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev, struct cpuidle_driver *drv, int index) { struct acpi_processor *pr; - struct acpi_processor_cx *cx = acpi_cstate[index]; + struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu); pr = __this_cpu_read(processors); @@ -960,7 +961,7 @@ static int acpi_processor_setup_cpuidle_cx(struct acpi_processor *pr, !(acpi_gbl_FADT.flags & ACPI_FADT_C2_MP_SUPPORTED)) continue; #endif - acpi_cstate[count] = cx; + per_cpu(acpi_cstate[count], dev->cpu) = cx; count++; if (count == CPUIDLE_STATE_MAX) -- cgit v0.10.2 From e66cd5372d09d58002b2779a3bcdc564d6348883 Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Wed, 3 Apr 2013 13:17:20 +0200 Subject: ACPI / BGRT: Don't let users configure BGRT on non X86 systems Fengguang Wu's 0-Day kernel build testing backend found the following build error for an allmodconfig build on ia64: drivers/built-in.o: In function `show_yoffset': >> bgrt.c:(.text+0xe5a71): undefined reference to `bgrt_tab' >> bgrt.c:(.text+0xe5a91): undefined reference to `bgrt_tab' drivers/built-in.o: In function `show_xoffset': >> bgrt.c:(.text+0xe5b51): undefined reference to `bgrt_tab' >> bgrt.c:(.text+0xe5b71): undefined reference to `bgrt_tab' drivers/built-in.o: In function `show_type': >> bgrt.c:(.text+0xe5c31): undefined reference to `bgrt_tab' drivers/built-in.o:bgrt.c:(.text+0xe5c51): more undefined references to `bgrt_tab' follow drivers/built-in.o: In function `bgrt_init': bgrt.c:(.init.text+0x8931): undefined reference to `bgrt_image' bgrt.c:(.init.text+0x8932): undefined reference to `bgrt_image_size' bgrt.c:(.init.text+0x8950): undefined reference to `bgrt_image' bgrt.c:(.init.text+0x8960): undefined reference to `bgrt_image_size' The problem is that all these undefined names are provided by arch/x86/platform/efi/efi-bgrt.c - which is obviously not available to the ia64 build. It doesn't seem useful to provide the BGRT support for Itanium (many systems are headless and have no graphics at all). So just don't let users configure this driver on non-X86 machines. Reported-by: Fengguang Wu Acked-by: H. Peter Anvin Signed-off-by: Tony Luck Signed-off-by: Rafael J. Wysocki diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index 92ed969..4bf68c8 100644 --- a/drivers/acpi/Kconfig +++ b/drivers/acpi/Kconfig @@ -396,7 +396,7 @@ config ACPI_CUSTOM_METHOD config ACPI_BGRT bool "Boottime Graphics Resource Table support" - depends on EFI + depends on EFI && X86 help This driver adds support for exposing the ACPI Boottime Graphics Resource Table, which allows the operating system to obtain -- cgit v0.10.2