kernel.git/drivers/base/power/main.c, branch linux-4.19.y

PM: core: keep irq flags in device_pm_check_callbacks()

2022-04-15T12:14:54Z

[ Upstream commit 524bb1da785a7ae43dd413cd392b5071c6c367f8 ] The function device_pm_check_callbacks() can be called under the spin lock (in the reported case it happens from genpd_add_device() -> dev_pm_domain_set(), when the genpd uses spinlocks rather than mutexes. However this function uncoditionally uses spin_lock_irq() / spin_unlock_irq(), thus not preserving the CPU flags. Use the irqsave/irqrestore instead. The backtrace for the reference: [ 2.752010] ------------[ cut here ]------------ [ 2.756769] raw_local_irq_restore() called with IRQs enabled [ 2.762596] WARNING: CPU: 4 PID: 1 at kernel/locking/irqflag-debug.c:10 warn_bogus_irq_restore+0x34/0x50 [ 2.772338] Modules linked in: [ 2.775487] CPU: 4 PID: 1 Comm: swapper/0 Tainted: G S 5.17.0-rc6-00384-ge330d0d82eff-dirty #684 [ 2.781384] Freeing initrd memory: 46024K [ 2.785839] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 2.785841] pc : warn_bogus_irq_restore+0x34/0x50 [ 2.785844] lr : warn_bogus_irq_restore+0x34/0x50 [ 2.785846] sp : ffff80000805b7d0 [ 2.785847] x29: ffff80000805b7d0 x28: 0000000000000000 x27: 0000000000000002 [ 2.785850] x26: ffffd40e80930b18 x25: ffff7ee2329192b8 x24: ffff7edfc9f60800 [ 2.785853] x23: ffffd40e80930b18 x22: ffffd40e80930d30 x21: ffff7edfc0dffa00 [ 2.785856] x20: ffff7edfc09e3768 x19: 0000000000000000 x18: ffffffffffffffff [ 2.845775] x17: 6572206f74206465 x16: 6c696166203a3030 x15: ffff80008805b4f7 [ 2.853108] x14: 0000000000000000 x13: ffffd40e809550b0 x12: 00000000000003d8 [ 2.860441] x11: 0000000000000148 x10: ffffd40e809550b0 x9 : ffffd40e809550b0 [ 2.867774] x8 : 00000000ffffefff x7 : ffffd40e809ad0b0 x6 : ffffd40e809ad0b0 [ 2.875107] x5 : 000000000000bff4 x4 : 0000000000000000 x3 : 0000000000000000 [ 2.882440] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff7edfc03a8000 [ 2.889774] Call trace: [ 2.892290] warn_bogus_irq_restore+0x34/0x50 [ 2.896770] _raw_spin_unlock_irqrestore+0x94/0xa0 [ 2.901690] genpd_unlock_spin+0x20/0x30 [ 2.905724] genpd_add_device+0x100/0x2d0 [ 2.909850] __genpd_dev_pm_attach+0xa8/0x23c [ 2.914329] genpd_dev_pm_attach_by_id+0xc4/0x190 [ 2.919167] genpd_dev_pm_attach_by_name+0x3c/0xd0 [ 2.924086] dev_pm_domain_attach_by_name+0x24/0x30 [ 2.929102] psci_dt_attach_cpu+0x24/0x90 [ 2.933230] psci_cpuidle_probe+0x2d4/0x46c [ 2.937534] platform_probe+0x68/0xe0 [ 2.941304] really_probe.part.0+0x9c/0x2fc [ 2.945605] __driver_probe_device+0x98/0x144 [ 2.950085] driver_probe_device+0x44/0x15c [ 2.954385] __device_attach_driver+0xb8/0x120 [ 2.958950] bus_for_each_drv+0x78/0xd0 [ 2.962896] __device_attach+0xd8/0x180 [ 2.966843] device_initial_probe+0x14/0x20 [ 2.971144] bus_probe_device+0x9c/0xa4 [ 2.975092] device_add+0x380/0x88c [ 2.978679] platform_device_add+0x114/0x234 [ 2.983067] platform_device_register_full+0x100/0x190 [ 2.988344] psci_idle_init+0x6c/0xb0 [ 2.992113] do_one_initcall+0x74/0x3a0 [ 2.996060] kernel_init_freeable+0x2fc/0x384 [ 3.000543] kernel_init+0x28/0x130 [ 3.004132] ret_from_fork+0x10/0x20 [ 3.007817] irq event stamp: 319826 [ 3.011404] hardirqs last enabled at (319825): [] __up_console_sem+0x78/0x84 [ 3.020332] hardirqs last disabled at (319826): [] el1_dbg+0x24/0x8c [ 3.028458] softirqs last enabled at (318312): [] _stext+0x410/0x588 [ 3.036678] softirqs last disabled at (318299): [] __irq_exit_rcu+0x158/0x174 [ 3.045607] ---[ end trace 0000000000000000 ]--- Signed-off-by: Dmitry Baryshkov Signed-off-by: Rafael J. Wysocki Signed-off-by: Sasha Levin

PM: sleep: core: Fix the handling of pending runtime resume requests

2020-09-03T09:24:29Z

commit e3eb6e8fba65094328b8dca635d00de74ba75b45 upstream. It has been reported that system-wide suspend may be aborted in the absence of any wakeup events due to unforseen interactions of it with the runtume PM framework. One failing scenario is when there are multiple devices sharing an ACPI power resource and runtime-resume needs to be carried out for one of them during system-wide suspend (for example, because it needs to be reconfigured before the whole system goes to sleep). In that case, the runtime-resume of that device involves turning the ACPI power resource "on" which in turn causes runtime-resume requests to be queued up for all of the other devices sharing it. Those requests go to the runtime PM workqueue which is frozen during system-wide suspend, so they are not actually taken care of until the resume of the whole system, but the pm_runtime_barrier() call in __device_suspend() sees them and triggers system wakeup events for them which then cause the system-wide suspend to be aborted if wakeup source objects are in active use. Of course, the logic that leads to triggering those wakeup events is questionable in the first place, because clearly there are cases in which a pending runtime resume request for a device is not connected to any real wakeup events in any way (like the one above). Moreover, it is racy, because the device may be resuming already by the time the pm_runtime_barrier() runs and so if the driver doesn't take care of signaling the wakeup event as appropriate, it will be lost. However, if the driver does take care of that, the extra pm_wakeup_event() call in the core is redundant. Accordingly, drop the conditional pm_wakeup_event() call fron __device_suspend() and make the latter call pm_runtime_barrier() alone. Also modify the comment next to that call to reflect the new code and extend it to mention the need to avoid unwanted interactions between runtime PM and system-wide device suspend callbacks. Fixes: 1e2ef05bb8cf8 ("PM: Limit race conditions between runtime PM and system sleep (v2)") Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern Reported-by: Utkarsh H Patel Tested-by: Utkarsh H Patel Tested-by: Pengfei Xu Cc: All applicable Signed-off-by: Greg Kroah-Hartman

PM: core: Fix handling of devices deleted during system-wide resume

2020-02-11T12:34:03Z

commit 0552e05fdfea191a2cf3a0abd33574b5ef9ca818 upstream. If a device is deleted by one of its system-wide resume callbacks (for example, because it does not appear to be present or accessible any more) along with its children, the resume of the children may continue leading to use-after-free errors and other issues (potentially). Namely, if the device's children are resumed asynchronously, their resume may have been scheduled already before the device's callback runs and so the device may be deleted while dpm_wait_for_superior() is being executed for them. The memory taken up by the parent device object may be freed then while dpm_wait() is waiting for the parent's resume callback to complete, which leads to a use-after-free. Moreover, the resume of the children is really not expected to continue after they have been unregistered, so it must be terminated right away in that case. To address this problem, modify dpm_wait_for_superior() to check if the target device is still there in the system-wide PM list of devices and if so, to increment its parent's reference counter, both under dpm_list_mtx which prevents device_del() running for the child from dropping the parent's reference counter prematurely. If the device is not present in the system-wide PM list of devices any more, the resume of it cannot continue, so check that again after dpm_wait() returns, which means that the parent's callback has been completed, and pass the result of that check to the caller of dpm_wait_for_superior() to allow it to abort the device's resume if it is not there any more. Link: https://lore.kernel.org/linux-pm/1579568452-27253-1-git-send-email-chanho.min@lge.com Reported-by: Chanho Min Cc: All applicable Signed-off-by: Rafael J. Wysocki Acked-by: Greg Kroah-Hartman Signed-off-by: Greg Kroah-Hartman

PM / core: Propagate dev->power.wakeup_path when no callbacks

2019-05-31T13:46:23Z

[ Upstream commit dc351d4c5f4fe4d0f274d6d660227be0c3a03317 ] The dev->power.direct_complete flag may become set in device_prepare() in case the device don't have any PM callbacks (dev->power.no_pm_callbacks is set). This leads to a broken behaviour, when there is child having wakeup enabled and relies on its parent to be used in the wakeup path. More precisely, when the direct complete path becomes selected for the child in __device_suspend(), the propagation of the dev->power.wakeup_path becomes skipped as well. Let's address this problem, by checking if the device is a part the wakeup path or has wakeup enabled, then prevent the direct complete path from being used. Reported-by: Loic Pallardy Signed-off-by: Ulf Hansson [ rjw: Comment cleanup ] Signed-off-by: Rafael J. Wysocki Signed-off-by: Sasha Levin

PM / core: Clear the direct_complete flag on errors

2018-10-04T17:39:31Z

If __device_suspend() runs asynchronously (in which case the device passed to it is in dpm_suspended_list at that point) and it returns early on an error or pending wakeup, and the power.direct_complete flag has been set for the device already, the subsequent device_resume() will be confused by that and it will call pm_runtime_enable() incorrectly, as runtime PM has not been disabled for the device by __device_suspend(). To avoid that, clear power.direct_complete if __device_suspend() is not going to disable runtime PM for the device before returning. Fixes: aae4518b3124 (PM / sleep: Mechanism to avoid resuming runtime-suspended devices unnecessarily) Reported-by: Al Cooper Tested-by: Al Cooper Reviewed-by: Ulf Hansson Cc: 3.16+ # 3.16+ Signed-off-by: Rafael J. Wysocki

Merge back PM core material for v4.18.

2018-05-25T08:39:39Z

PM / core: Fix direct_complete handling for devices with no callbacks

2018-05-22T12:50:11Z

Commit 08810a4119aa (PM / core: Add NEVER_SKIP and SMART_PREPARE driver flags) inadvertently prevented the power.direct_complete flag from being set for devices without PM callbacks and with disabled runtime PM which also prevents power.direct_complete from being set for their parents. That led to problems including a resume crash on HP ZBook 14u. Restore the previous behavior by causing power.direct_complete to be set for those devices again, but do that in a more direct way to avoid overlooking that case in the future. Link: https://bugzilla.kernel.org/show_bug.cgi?id=199693 Fixes: 08810a4119aa (PM / core: Add NEVER_SKIP and SMART_PREPARE driver flags) Reported-by: Thomas Martitz Tested-by: Thomas Martitz Cc: 4.15+ # 4.15+ Signed-off-by: Rafael J. Wysocki Reviewed-by: Ulf Hansson Reviewed-by: Johan Hovold

PM / core: Use dev_printk() and symbols in suspend/resume diagnostics

2018-05-10T09:31:45Z

When we print diagnostic messages about suspend/resume, we have a device pointer, so use dev_printk() to match other device-related things. Add the function name, similar to initcall_debug output. E.g., - calling 0000:01:00.0+ @ 998, parent: 0000:00:1c.0 + pci 0000:01:00.0: calling @ 998, parent: 0000:00:1c.0 I wondered if this would break scripts/bootgraph.pl, but I don't think it will because bootgraph.pl doesn't add any timing information to $start{} after it sees "Write protecting the" or "Freeing unused kernel memory". Signed-off-by: Bjorn Helgaas Signed-off-by: Rafael J. Wysocki

PM / core: Simplify initcall_debug_report() timing

2018-05-10T09:31:45Z

initcall_debug_report() always called ktime_get(), even if we didn't need the result. Change it so we only call it when we're going to use the result, and change initcall_debug_start() to follow the same style. Signed-off-by: Bjorn Helgaas Signed-off-by: Rafael J. Wysocki

PM / core: Remove unused initcall_debug_report() arguments

2018-05-10T09:31:45Z

Commit e8bca479c3f2 (PM / sleep: trace events for device PM callbacks) removed the only uses of "state" and "info" from initcall_debug_report(). Remove the now-unused arguments completely. Signed-off-by: Bjorn Helgaas Signed-off-by: Rafael J. Wysocki