From 32bdfac5462d777f35b00838893c4f87baf23efe Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 24 May 2009 21:15:07 +0200 Subject: PM: Do not hold dpm_list_mtx while disabling/enabling nonboot CPUs We shouldn't hold dpm_list_mtx while executing [disable|enable]_nonboot_cpus(), because theoretically this may lead to a deadlock as shown by the following example (provided by Johannes Berg): CPU 3 CPU 2 CPU 1 suspend/hibernate something: rtnl_lock() device_pm_lock() -> mutex_lock(&dpm_list_mtx) mutex_lock(&dpm_list_mtx) linkwatch_work -> rtnl_lock() disable_nonboot_cpus() -> flush CPU 3 workqueue Fortunately, device drivers are supposed to stop any activities that might lead to the registration of new device objects way before disable_nonboot_cpus() is called, so it shouldn't be necessary to hold dpm_list_mtx over the entire late part of device suspend and early part of device resume. Thus, during the late suspend and the early resume of devices acquire dpm_list_mtx only when dpm_list is going to be traversed and release it right after that. This patch is reported to fix the regressions tracked as http://bugzilla.kernel.org/show_bug.cgi?id=13245. Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern Reported-by: Miles Lane Tested-by: Ming Lei --- kernel/kexec.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel/kexec.c') diff --git a/kernel/kexec.c b/kernel/kexec.c index 5a758c6e495..e4983770913 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1451,7 +1451,6 @@ int kernel_kexec(void) error = device_suspend(PMSG_FREEZE); if (error) goto Resume_console; - device_pm_lock(); /* At this point, device_suspend() has been called, * but *not* device_power_down(). We *must* * device_power_down() now. Otherwise, drivers for @@ -1489,7 +1488,6 @@ int kernel_kexec(void) enable_nonboot_cpus(); device_power_up(PMSG_RESTORE); Resume_devices: - device_pm_unlock(); device_resume(PMSG_RESTORE); Resume_console: resume_console(); -- cgit From e39a71ef80877f4e30d808af9acceec80f4d2f7c Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Fri, 15 May 2009 00:53:26 +0200 Subject: PM: Rename device_power_down/up() Rename the functions performing "_noirq" dev_pm_ops operations from device_power_down() and device_power_up() to device_suspend_noirq() and device_resume_noirq(). The new function names are chosen to show that the functions are responsible for calling the _noirq() versions to finalize the suspend/resume operation. The current function names do not perform power down/up anymore so the names may be misleading. Global function renames: - device_power_down() -> device_suspend_noirq() - device_power_up() -> device_resume_noirq() Static function renames: - suspend_device_noirq() -> __device_suspend_noirq() - resume_device_noirq() -> __device_resume_noirq() Signed-off-by: Magnus Damm Acked-by: Greg Kroah-Hartman Acked-by: Len Brown Signed-off-by: Rafael J. Wysocki --- kernel/kexec.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel/kexec.c') diff --git a/kernel/kexec.c b/kernel/kexec.c index e4983770913..5a3da87adae 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1452,13 +1452,13 @@ int kernel_kexec(void) if (error) goto Resume_console; /* At this point, device_suspend() has been called, - * but *not* device_power_down(). We *must* - * device_power_down() now. Otherwise, drivers for + * but *not* device_suspend_noirq(). We *must* call + * device_suspend_noirq() now. Otherwise, drivers for * some devices (e.g. interrupt controllers) become * desynchronized with the actual state of the * hardware at resume time, and evil weirdness ensues. */ - error = device_power_down(PMSG_FREEZE); + error = device_suspend_noirq(PMSG_FREEZE); if (error) goto Resume_devices; error = disable_nonboot_cpus(); @@ -1486,7 +1486,7 @@ int kernel_kexec(void) local_irq_enable(); Enable_cpus: enable_nonboot_cpus(); - device_power_up(PMSG_RESTORE); + device_resume_noirq(PMSG_RESTORE); Resume_devices: device_resume(PMSG_RESTORE); Resume_console: -- cgit From d161630297a20802d01c55847bfcba85d2118a9f Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sun, 24 May 2009 22:05:42 +0200 Subject: PM core: rename suspend and resume functions This patch (as1241) renames a bunch of functions in the PM core. Rather than go through a boring list of name changes, suffice it to say that in the end we have a bunch of pairs of functions: device_resume_noirq dpm_resume_noirq device_resume dpm_resume device_complete dpm_complete device_suspend_noirq dpm_suspend_noirq device_suspend dpm_suspend device_prepare dpm_prepare in which device_X does the X operation on a single device and dpm_X invokes device_X for all devices in the dpm_list. In addition, the old dpm_power_up and device_resume_noirq have been combined into a single function (dpm_resume_noirq). Lastly, dpm_suspend_start and dpm_resume_end are the renamed versions of the former top-level device_suspend and device_resume routines. Signed-off-by: Alan Stern Acked-by: Magnus Damm Signed-off-by: Rafael J. Wysocki --- kernel/kexec.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel/kexec.c') diff --git a/kernel/kexec.c b/kernel/kexec.c index 5a3da87adae..ae1c35201cc 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1448,17 +1448,17 @@ int kernel_kexec(void) goto Restore_console; } suspend_console(); - error = device_suspend(PMSG_FREEZE); + error = dpm_suspend_start(PMSG_FREEZE); if (error) goto Resume_console; - /* At this point, device_suspend() has been called, - * but *not* device_suspend_noirq(). We *must* call - * device_suspend_noirq() now. Otherwise, drivers for + /* At this point, dpm_suspend_start() has been called, + * but *not* dpm_suspend_noirq(). We *must* call + * dpm_suspend_noirq() now. Otherwise, drivers for * some devices (e.g. interrupt controllers) become * desynchronized with the actual state of the * hardware at resume time, and evil weirdness ensues. */ - error = device_suspend_noirq(PMSG_FREEZE); + error = dpm_suspend_noirq(PMSG_FREEZE); if (error) goto Resume_devices; error = disable_nonboot_cpus(); @@ -1486,9 +1486,9 @@ int kernel_kexec(void) local_irq_enable(); Enable_cpus: enable_nonboot_cpus(); - device_resume_noirq(PMSG_RESTORE); + dpm_resume_noirq(PMSG_RESTORE); Resume_devices: - device_resume(PMSG_RESTORE); + dpm_resume_end(PMSG_RESTORE); Resume_console: resume_console(); thaw_processes(); -- cgit