In the Linux kernel, the following vulnerability has been resolved:
LoongArch: Fix warnings during S3 suspend
The enablegpewakeup() function calls acpienableallwakeupgpes(), and the later one may call the preemptschedulecommon() function, resulting in a thread switch and causing the CPU to be in an interrupt enabled state after the enablegpewakeup() function returns, leading to the warnings as follow.
[ C0] WARNING: ... at kernel/time/timekeeping.c:845 ktimeget+0xbc/0xc8 [ C0] ... [ C0] Call Trace: [ C0] [<90000000002243b4>] showstack+0x64/0x188 [ C0] [<900000000164673c>] dumpstacklvl+0x60/0x88 [ C0] [<90000000002687e4>] _warn+0x8c/0x148 [ C0] [<90000000015e9978>] reportbug+0x1c0/0x2b0 [ C0] [<90000000016478e4>] dobp+0x204/0x3b8 [ C0] [<90000000025b1924>] exceptionhandlers+0x1924/0x10000 [ C0] [<9000000000343bbc>] ktimeget+0xbc/0xc8 [ C0] [<9000000000354c08>] tickschedtimer+0x30/0xb0 [ C0] [<90000000003408e0>] _hrtimerrunqueues+0x160/0x378 [ C0] [<9000000000341f14>] hrtimerinterrupt+0x144/0x388 [ C0] [<9000000000228348>] constanttimerinterrupt+0x38/0x48 [ C0] [<90000000002feba4>] _handleirqeventpercpu+0x64/0x1e8 [ C0] [<90000000002fed48>] handleirqeventpercpu+0x20/0x80 [ C0] [<9000000000306b9c>] handlepercpuirq+0x5c/0x98 [ C0] [<90000000002fd4a0>] generichandledomainirq+0x30/0x48 [ C0] [<9000000000d0c7b0>] handlecpuirq+0x70/0xa8 [ C0] [<9000000001646b30>] handleloongarchirq+0x30/0x48 [ C0] [<9000000001646bc8>] dovint+0x80/0xe0 [ C0] [<90000000002aea1c>] finishtaskswitch.isra.0+0x8c/0x2a8 [ C0] [<900000000164e34c>] _schedule+0x314/0xa48 [ C0] [<900000000164ead8>] schedule+0x58/0xf0 [ C0] [<9000000000294a2c>] workerthread+0x224/0x498 [ C0] [<900000000029d2f0>] kthread+0xf8/0x108 [ C0] [<9000000000221f28>] retfromkernel_thread+0xc/0xa4 [ C0] [ C0] ---[ end trace 0000000000000000 ]---
The root cause is acpienableallwakeupgpes() uses a mutex to protect acpihwenableallwakeupgpes(), and acpiutacquiremutex() may cause a thread switch. Since there is no longer concurrent execution during loongarchacpisuspend(), we can call acpihwenableallwakeupgpes() directly in enablegpe_wakeup().
The solution is similar to commit 22db06337f590d01 ("ACPI: sleep: Avoid breaking S3 wakeup due to might_sleep()").