Skip to contentSkip to conversation

420356c350: WARNING:at_kernel/sched/core.c:#__might_sleep

peterz-queue:sched/core kernel test robot 3 participants 7 messages Aug 16, 2024 – Aug 27, 2024

lore
Share & tools
lore

kernel test robot

author build bot
#0 lore raw

Hello,

kernel test robot noticed "WARNING:at_kernel/sched/core.c:#__might_sleep" on:

commit: 420356c3504091f0f6021974389df7c58f365dad ("sched/fair: Implement delayed dequeue")
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/core

in testcase: rcutorture
version:
with following parameters:

	runtime: 300s
	test: cpuhotplug
	torture_type: tasks-rude

compiler: clang-18
test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 16G

(please refer to attached dmesg/kmsg for entire log/backtrace)

+-------------------------------------------------------------------------+------------+------------+
|                                                                         | 18fdefe603 | 420356c350 |
+-------------------------------------------------------------------------+------------+------------+
| WARNING:at_kernel/sched/core.c:#__might_sleep                           | 0          | 18         |
| RIP:__might_sleep                                                       | 0          | 18         |
+-------------------------------------------------------------------------+------------+------------+

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-lkp/[email protected]

[   86.252370][  T674] ------------[ cut here ]------------
[ 86.252945][ T674] do not call blocking ops when !TASK_RUNNING; state=1 set at kthread_worker_fn (kernel/kthread.c:?) 
[ 86.254001][ T674] WARNING: CPU: 1 PID: 674 at kernel/sched/core.c:8469 __might_sleep (kernel/sched/core.c:8465) 
[   86.255224][  T674] Modules linked in: rcutorture torture crct10dif_pclmul crc32c_intel polyval_clmulni polyval_generic tiny_power_button sha256_ssse3 rtc_cmos sha1_ssse3 button aesni_intel evdev ipmi_devintf ipmi_msghandler drm fuse loop efi_pstore drm_panel_orientation_quirks pstore qemu_fw_cfg ip_tables x_tables autofs4
[   86.260658][  T674] CPU: 1 UID: 0 PID: 674 Comm: erofs_worker/1 Tainted: G                T  6.11.0-rc1-00042-g420356c35040 #1 f37f1ad66c1ebf3b07abcf4fc3040d14b7156f8a
[   86.262530][  T674] Tainted: [T]=RANDSTRUCT
[   86.263000][  T674] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
[ 86.264783][ T674] RIP: 0010:__might_sleep (kernel/sched/core.c:8465) 
[ 86.265305][ T674] Code: 00 00 00 00 fc ff df 80 3c 08 00 74 08 4c 89 ff e8 e6 50 62 00 49 8b 0f 48 c7 c7 e0 b0 ee 84 44 89 e6 48 89 ca e8 c1 0e f5 ff <0f> 0b 48 c7 c7 b0 01 8c 86 be 01 00 00 00 31 d2 b9 01 00 00 00 eb
All code
========
   0:	00 00                	add    %al,(%rax)
   2:	00 00                	add    %al,(%rax)
   4:	fc                   	cld
   5:	ff                   	(bad)
   6:	df 80 3c 08 00 74    	filds  0x7400083c(%rax)
   c:	08 4c 89 ff          	or     %cl,-0x1(%rcx,%rcx,4)
  10:	e8 e6 50 62 00       	call   0x6250fb
  15:	49 8b 0f             	mov    (%r15),%rcx
  18:	48 c7 c7 e0 b0 ee 84 	mov    $0xffffffff84eeb0e0,%rdi
  1f:	44 89 e6             	mov    %r12d,%esi
  22:	48 89 ca             	mov    %rcx,%rdx
  25:	e8 c1 0e f5 ff       	call   0xfffffffffff50eeb
  2a:*	0f 0b                	ud2		<-- trapping instruction
  2c:	48 c7 c7 b0 01 8c 86 	mov    $0xffffffff868c01b0,%rdi
  33:	be 01 00 00 00       	mov    $0x1,%esi
  38:	31 d2                	xor    %edx,%edx
  3a:	b9 01 00 00 00       	mov    $0x1,%ecx

3f: eb .byte 0xeb

Code starting with the faulting instruction
===========================================
0: 0f 0b ud2

   2:	48 c7 c7 b0 01 8c 86 	mov    $0xffffffff868c01b0,%rdi
   9:	be 01 00 00 00       	mov    $0x1,%esi
   e:	31 d2                	xor    %edx,%edx
  10:	b9 01 00 00 00       	mov    $0x1,%ecx
  15:	eb                   	.byte 0xeb
[   86.266970][  T674] RSP: 0000:ffffc90000f1fdb8 EFLAGS: 00010246
[   86.267710][  T674] RAX: 93ef2e1fd3a87a00 RBX: 0000000000000001 RCX: 0000000000000000
[   86.268961][  T674] RDX: dffffc0000000000 RSI: 0000000000000000 RDI: ffffffff868af248
[   86.269995][  T674] RBP: ffffc90000f1fdf0 R08: ffffc90000f1fbf7 R09: 1ffff920001e3f7e
[   86.270937][  T674] R10: dffffc0000000000 R11: fffff520001e3f7f R12: 0000000000000001
[   86.272231][  T674] R13: 0000000000000001 R14: 0000000000000001 R15: ffff88817b340080
[   86.273485][  T674] FS:  0000000000000000(0000) GS:ffff8883aef00000(0000) knlGS:0000000000000000
[   86.274487][  T674] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   86.275275][  T674] CR2: 0000000000000000 CR3: 0000000005815000 CR4: 00000000000406f0
[   86.276454][  T674] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   86.277565][  T674] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[   86.278443][  T674] Call Trace:
[   86.279070][  T674]  <TASK>
[ 86.279553][ T674] ? show_regs (arch/x86/kernel/dumpstack.c:479) 
[ 86.279942][ T674] ? __warn (kernel/panic.c:735) 
[ 86.280703][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.281360][ T674] ? report_bug (lib/bug.c:?) 
[ 86.282183][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[   86.282816][  T674]  ? __wake_up_klogd
[ 86.283398][ T674] ? handle_bug (arch/x86/kernel/traps.c:239) 
[ 86.283995][ T674] ? exc_invalid_op (arch/x86/kernel/traps.c:260) 
[ 86.284787][ T674] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621) 
[ 86.285682][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.286380][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.287116][ T674] kthread_worker_fn (include/linux/kernel.h:73 include/linux/freezer.h:53 kernel/kthread.c:851) 
[ 86.287701][ T674] ? kthread_worker_fn (kernel/kthread.c:?) 
[ 86.288138][ T674] kthread (kernel/kthread.c:391) 
[ 86.288482][ T674] ? __cfi_kthread_worker_fn (kernel/kthread.c:803) 
[ 86.288951][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.289560][ T674] ret_from_fork (arch/x86/kernel/process.c:153) 
[ 86.290162][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.291465][ T674] ret_from_fork_asm (arch/x86/entry/entry_64.S:254) 
[   86.292080][  T674]  </TASK>
[   86.292468][  T674] irq event stamp: 579
[ 86.292872][ T674] hardirqs last enabled at (589): console_unlock (arch/x86/include/asm/irqflags.h:42 arch/x86/include/asm/irqflags.h:97 arch/x86/include/asm/irqflags.h:155 kernel/printk/printk.c:341 kernel/printk/printk.c:2801 kernel/printk/printk.c:3120) 
[ 86.293852][ T674] hardirqs last disabled at (598): console_unlock (kernel/printk/printk.c:339 kernel/printk/printk.c:2801 kernel/printk/printk.c:3120) 
[ 86.294901][ T674] softirqs last enabled at (0): copy_process (include/linux/rcupdate.h:326 include/linux/rcupdate.h:838 kernel/fork.c:2243) 
[ 86.296048][ T674] softirqs last disabled at (0): 0x0 
[   86.296931][  T674] ---[ end trace 0000000000000000 ]---

The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20240816/[email protected]

Signature · 2 lines
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
#1 lore raw
Quoted 9 lines · kernel test robot

kernel test robot noticed "WARNING:at_kernel/sched/core.c:#__might_sleep" on:

commit: 420356c3504091f0f6021974389df7c58f365dad ("sched/fair: Implement delayed dequeue")
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/core

[   86.252370][  T674] ------------[ cut here ]------------
[ 86.252945][ T674] do not call blocking ops when !TASK_RUNNING; state=1 set at kthread_worker_fn (kernel/kthread.c:?) 
[ 86.254001][ T674] WARNING: CPU: 1 PID: 674 at kernel/sched/core.c:8469 __might_sleep (kernel/sched/core.c:8465) 
Quoted 13 lines
[ 86.283398][ T674] ? handle_bug (arch/x86/kernel/traps.c:239) 
[ 86.283995][ T674] ? exc_invalid_op (arch/x86/kernel/traps.c:260) 
[ 86.284787][ T674] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621) 
[ 86.285682][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.286380][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.287116][ T674] kthread_worker_fn (include/linux/kernel.h:73 include/linux/freezer.h:53 kernel/kthread.c:851) 
[ 86.287701][ T674] ? kthread_worker_fn (kernel/kthread.c:?) 
[ 86.288138][ T674] kthread (kernel/kthread.c:391) 
[ 86.288482][ T674] ? __cfi_kthread_worker_fn (kernel/kthread.c:803) 
[ 86.288951][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.289560][ T674] ret_from_fork (arch/x86/kernel/process.c:153) 
[ 86.290162][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.291465][ T674] ret_from_fork_asm (arch/x86/entry/entry_64.S:254) 

AFAICT this is a pre-existing issue. Notably that all transcribes to:

kthread_worker_fn()
...
repeat:
set_current_state(TASK_INTERRUPTIBLE);
...
if (work) { // false

    __set_current_state(TASK_RUNNING);
    ...
  } else if (!freezing(current)) // false -- we are freezing
    schedule();

// so state really is still TASK_INTERRUPTIBLE here try_to_freeze()
might_sleep() <--- boom, per the above.

#2 lore raw
Quoted 32 lines · Peter Zijlstra
Quoted 9 lines · kernel test robot

kernel test robot noticed "WARNING:at_kernel/sched/core.c:#__might_sleep" on:

commit: 420356c3504091f0f6021974389df7c58f365dad ("sched/fair: Implement delayed dequeue")
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/core

[   86.252370][  T674] ------------[ cut here ]------------
[ 86.252945][ T674] do not call blocking ops when !TASK_RUNNING; state=1 set at kthread_worker_fn (kernel/kthread.c:?) 
[ 86.254001][ T674] WARNING: CPU: 1 PID: 674 at kernel/sched/core.c:8469 __might_sleep (kernel/sched/core.c:8465) 
Quoted 13 lines
[ 86.283398][ T674] ? handle_bug (arch/x86/kernel/traps.c:239) 
[ 86.283995][ T674] ? exc_invalid_op (arch/x86/kernel/traps.c:260) 
[ 86.284787][ T674] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621) 
[ 86.285682][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.286380][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.287116][ T674] kthread_worker_fn (include/linux/kernel.h:73 include/linux/freezer.h:53 kernel/kthread.c:851) 
[ 86.287701][ T674] ? kthread_worker_fn (kernel/kthread.c:?) 
[ 86.288138][ T674] kthread (kernel/kthread.c:391) 
[ 86.288482][ T674] ? __cfi_kthread_worker_fn (kernel/kthread.c:803) 
[ 86.288951][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.289560][ T674] ret_from_fork (arch/x86/kernel/process.c:153) 
[ 86.290162][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.291465][ T674] ret_from_fork_asm (arch/x86/entry/entry_64.S:254) 

AFAICT this is a pre-existing issue. Notably that all transcribes to:

kthread_worker_fn()
...
repeat:
set_current_state(TASK_INTERRUPTIBLE);
...
if (work) { // false

Quoted 9 lines
    __set_current_state(TASK_RUNNING);
    ...
  } else if (!freezing(current)) // false -- we are freezing
    schedule();

// so state really is still TASK_INTERRUPTIBLE here try_to_freeze()
might_sleep() <--- boom, per the above.

Would the following fix make sense?

kernel/kthread.c +6 -0
index f7be976ff88a..09850b2109c9 100644
@@ -848,6 +848,12 @@ int kthread_worker_fn(void *worker_ptr)
 	} else if (!freezing(current))
 		schedule();
 
+	/*
+	 * Explictly set the running state in case we are being frozen
+	 * and skip the schedule() above. try_to_freeze() expects the
+	 * current task to be in running state.
+	 */
+	__set_current_state(TASK_RUNNING);
 	try_to_freeze();
 	cond_resched();
 	goto repeat;
Signature · 7 lines
2.25.1

Hi Oliver,
Could you please help check if above change would make the warning go away?

thanks,
Chenyu
#3 lore raw

hi, Chen Yu,

Quoted 67 lines · Chen Yu
Quoted 32 lines · Peter Zijlstra
Quoted 9 lines · kernel test robot

kernel test robot noticed "WARNING:at_kernel/sched/core.c:#__might_sleep" on:

commit: 420356c3504091f0f6021974389df7c58f365dad ("sched/fair: Implement delayed dequeue")
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/core

[   86.252370][  T674] ------------[ cut here ]------------
[ 86.252945][ T674] do not call blocking ops when !TASK_RUNNING; state=1 set at kthread_worker_fn (kernel/kthread.c:?) 
[ 86.254001][ T674] WARNING: CPU: 1 PID: 674 at kernel/sched/core.c:8469 __might_sleep (kernel/sched/core.c:8465) 
Quoted 13 lines
[ 86.283398][ T674] ? handle_bug (arch/x86/kernel/traps.c:239) 
[ 86.283995][ T674] ? exc_invalid_op (arch/x86/kernel/traps.c:260) 
[ 86.284787][ T674] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621) 
[ 86.285682][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.286380][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.287116][ T674] kthread_worker_fn (include/linux/kernel.h:73 include/linux/freezer.h:53 kernel/kthread.c:851) 
[ 86.287701][ T674] ? kthread_worker_fn (kernel/kthread.c:?) 
[ 86.288138][ T674] kthread (kernel/kthread.c:391) 
[ 86.288482][ T674] ? __cfi_kthread_worker_fn (kernel/kthread.c:803) 
[ 86.288951][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.289560][ T674] ret_from_fork (arch/x86/kernel/process.c:153) 
[ 86.290162][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.291465][ T674] ret_from_fork_asm (arch/x86/entry/entry_64.S:254) 

AFAICT this is a pre-existing issue. Notably that all transcribes to:

kthread_worker_fn()
...
repeat:
set_current_state(TASK_INTERRUPTIBLE);
...
if (work) { // false

Quoted 9 lines
    __set_current_state(TASK_RUNNING);
    ...
  } else if (!freezing(current)) // false -- we are freezing
    schedule();

// so state really is still TASK_INTERRUPTIBLE here try_to_freeze()
might_sleep() <--- boom, per the above.

Would the following fix make sense?

kernel/kthread.c +6 -0
index f7be976ff88a..09850b2109c9 100644
@@ -848,6 +848,12 @@ int kthread_worker_fn(void *worker_ptr)
 	} else if (!freezing(current))
 		schedule();
 
+	/*
+	 * Explictly set the running state in case we are being frozen
+	 * and skip the schedule() above. try_to_freeze() expects the
+	 * current task to be in running state.
+	 */
+	__set_current_state(TASK_RUNNING);
 	try_to_freeze();
 	cond_resched();
 	goto repeat;
Signature · 4 lines
2.25.1

Hi Oliver,
Could you please help check if above change would make the warning go away?

confirmed this patch could resolve the WARNING. thanks

Quoted 3 lines

thanks,
Chenyu

#4 lore raw
In reply to Chen Yu
Quoted 38 lines · Chen Yu
Quoted 37 lines · Peter Zijlstra
Quoted 9 lines · kernel test robot

kernel test robot noticed "WARNING:at_kernel/sched/core.c:#__might_sleep" on:

commit: 420356c3504091f0f6021974389df7c58f365dad ("sched/fair: Implement delayed dequeue")
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/core

[   86.252370][  T674] ------------[ cut here ]------------
[ 86.252945][ T674] do not call blocking ops when !TASK_RUNNING; state=1 set at kthread_worker_fn (kernel/kthread.c:?) 
[ 86.254001][ T674] WARNING: CPU: 1 PID: 674 at kernel/sched/core.c:8469 __might_sleep (kernel/sched/core.c:8465) 
Quoted 13 lines
[ 86.283398][ T674] ? handle_bug (arch/x86/kernel/traps.c:239) 
[ 86.283995][ T674] ? exc_invalid_op (arch/x86/kernel/traps.c:260) 
[ 86.284787][ T674] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621) 
[ 86.285682][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.286380][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.287116][ T674] kthread_worker_fn (include/linux/kernel.h:73 include/linux/freezer.h:53 kernel/kthread.c:851) 
[ 86.287701][ T674] ? kthread_worker_fn (kernel/kthread.c:?) 
[ 86.288138][ T674] kthread (kernel/kthread.c:391) 
[ 86.288482][ T674] ? __cfi_kthread_worker_fn (kernel/kthread.c:803) 
[ 86.288951][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.289560][ T674] ret_from_fork (arch/x86/kernel/process.c:153) 
[ 86.290162][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.291465][ T674] ret_from_fork_asm (arch/x86/entry/entry_64.S:254) 

AFAICT this is a pre-existing issue. Notably that all transcribes to:

kthread_worker_fn()
...
repeat:
set_current_state(TASK_INTERRUPTIBLE);
...
if (work) { // false

    __set_current_state(TASK_RUNNING);
    ...
  } else if (!freezing(current)) // false -- we are freezing
    schedule();
Quoted 6 lines
Quoted 4 lines

// so state really is still TASK_INTERRUPTIBLE here try_to_freeze()
might_sleep() <--- boom, per the above.

Would the following fix make sense?

Yeah, that looks fine. Could you write it up as a proper patch please?

Quoted 26 lines
kernel/kthread.c +6 -0
index f7be976ff88a..09850b2109c9 100644
@@ -848,6 +848,12 @@ int kthread_worker_fn(void *worker_ptr)
 	} else if (!freezing(current))
 		schedule();
 
+	/*
+	 * Explictly set the running state in case we are being frozen
+	 * and skip the schedule() above. try_to_freeze() expects the
+	 * current task to be in running state.
+	 */
+	__set_current_state(TASK_RUNNING);
 	try_to_freeze();
 	cond_resched();
 	goto repeat;
Signature · 7 lines
2.25.1

Hi Oliver,
Could you please help check if above change would make the warning go away?

thanks,
Chenyu
#5 lore raw
Quoted 39 lines · Peter Zijlstra
Quoted 38 lines · Chen Yu
Quoted 37 lines · Peter Zijlstra
Quoted 9 lines · kernel test robot

kernel test robot noticed "WARNING:at_kernel/sched/core.c:#__might_sleep" on:

commit: 420356c3504091f0f6021974389df7c58f365dad ("sched/fair: Implement delayed dequeue")
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/core

[   86.252370][  T674] ------------[ cut here ]------------
[ 86.252945][ T674] do not call blocking ops when !TASK_RUNNING; state=1 set at kthread_worker_fn (kernel/kthread.c:?) 
[ 86.254001][ T674] WARNING: CPU: 1 PID: 674 at kernel/sched/core.c:8469 __might_sleep (kernel/sched/core.c:8465) 
Quoted 13 lines
[ 86.283398][ T674] ? handle_bug (arch/x86/kernel/traps.c:239) 
[ 86.283995][ T674] ? exc_invalid_op (arch/x86/kernel/traps.c:260) 
[ 86.284787][ T674] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621) 
[ 86.285682][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.286380][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.287116][ T674] kthread_worker_fn (include/linux/kernel.h:73 include/linux/freezer.h:53 kernel/kthread.c:851) 
[ 86.287701][ T674] ? kthread_worker_fn (kernel/kthread.c:?) 
[ 86.288138][ T674] kthread (kernel/kthread.c:391) 
[ 86.288482][ T674] ? __cfi_kthread_worker_fn (kernel/kthread.c:803) 
[ 86.288951][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.289560][ T674] ret_from_fork (arch/x86/kernel/process.c:153) 
[ 86.290162][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.291465][ T674] ret_from_fork_asm (arch/x86/entry/entry_64.S:254) 

AFAICT this is a pre-existing issue. Notably that all transcribes to:

kthread_worker_fn()
...
repeat:
set_current_state(TASK_INTERRUPTIBLE);
...
if (work) { // false

    __set_current_state(TASK_RUNNING);
    ...
  } else if (!freezing(current)) // false -- we are freezing
    schedule();
Quoted 9 lines
Quoted 6 lines
Quoted 4 lines

// so state really is still TASK_INTERRUPTIBLE here try_to_freeze()
might_sleep() <--- boom, per the above.

Would the following fix make sense?

Yeah, that looks fine. Could you write it up as a proper patch please?

Yes, it should be a race condition in theory and I've sent a patch here:
https://lore.kernel.org/lkml/[email protected]/
And Andrew has given some comments on it.

However, after I did some further investigation, this warning seems to not be directly related to task freeze, but has connection with the delay dequeue. I'm planning to add debug patch and investigate the symptom in 0day's environment, will send the finding later.

thanks,
Chenyu

Quoted 26 lines
Quoted 26 lines
kernel/kthread.c +6 -0
index f7be976ff88a..09850b2109c9 100644
@@ -848,6 +848,12 @@ int kthread_worker_fn(void *worker_ptr)
 	} else if (!freezing(current))
 		schedule();
 
+	/*
+	 * Explictly set the running state in case we are being frozen
+	 * and skip the schedule() above. try_to_freeze() expects the
+	 * current task to be in running state.
+	 */
+	__set_current_state(TASK_RUNNING);
 	try_to_freeze();
 	cond_resched();
 	goto repeat;
Signature · 7 lines
2.25.1

Hi Oliver,
Could you please help check if above change would make the warning go away?

thanks,
Chenyu
#6 lore raw
Quoted 50 lines · Chen Yu
Quoted 39 lines · Peter Zijlstra
Quoted 38 lines · Chen Yu
Quoted 37 lines · Peter Zijlstra
Quoted 9 lines · kernel test robot

kernel test robot noticed "WARNING:at_kernel/sched/core.c:#__might_sleep" on:

commit: 420356c3504091f0f6021974389df7c58f365dad ("sched/fair: Implement delayed dequeue")
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git sched/core

[   86.252370][  T674] ------------[ cut here ]------------
[ 86.252945][ T674] do not call blocking ops when !TASK_RUNNING; state=1 set at kthread_worker_fn (kernel/kthread.c:?) 
[ 86.254001][ T674] WARNING: CPU: 1 PID: 674 at kernel/sched/core.c:8469 __might_sleep (kernel/sched/core.c:8465) 
Quoted 13 lines
[ 86.283398][ T674] ? handle_bug (arch/x86/kernel/traps.c:239) 
[ 86.283995][ T674] ? exc_invalid_op (arch/x86/kernel/traps.c:260) 
[ 86.284787][ T674] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621) 
[ 86.285682][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.286380][ T674] ? __might_sleep (kernel/sched/core.c:8465) 
[ 86.287116][ T674] kthread_worker_fn (include/linux/kernel.h:73 include/linux/freezer.h:53 kernel/kthread.c:851) 
[ 86.287701][ T674] ? kthread_worker_fn (kernel/kthread.c:?) 
[ 86.288138][ T674] kthread (kernel/kthread.c:391) 
[ 86.288482][ T674] ? __cfi_kthread_worker_fn (kernel/kthread.c:803) 
[ 86.288951][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.289560][ T674] ret_from_fork (arch/x86/kernel/process.c:153) 
[ 86.290162][ T674] ? __cfi_kthread (kernel/kthread.c:342) 
[ 86.291465][ T674] ret_from_fork_asm (arch/x86/entry/entry_64.S:254) 

AFAICT this is a pre-existing issue. Notably that all transcribes to:

kthread_worker_fn()
...
repeat:
set_current_state(TASK_INTERRUPTIBLE);
...
if (work) { // false

    __set_current_state(TASK_RUNNING);
    ...
  } else if (!freezing(current)) // false -- we are freezing
    schedule();
Quoted 9 lines
Quoted 6 lines
Quoted 4 lines

// so state really is still TASK_INTERRUPTIBLE here try_to_freeze()
might_sleep() <--- boom, per the above.

Would the following fix make sense?

Yeah, that looks fine. Could you write it up as a proper patch please?

Quoted 9 lines

Yes, it should be a race condition in theory and I've sent a patch here:
https://lore.kernel.org/lkml/[email protected]/
And Andrew has given some comments on it.

However, after I did some further investigation, this warning seems to not be directly related to task freeze, but has connection with the delay dequeue. I'm planning to add debug patch and investigate the symptom in 0day's environment, will send the finding later.

The root cause is replied to the delayed dequeue patch set here:
https://lore.kernel.org/lkml/Zs2ZoAcUsZMX2B%2FI@chenyu5-mobl2/

And since the race condition mentioned previously is real (although not the root cause of the warning reported in this thread), I'll send a v2 patch to get that fixed.

thanks,
Chenyu