diff mbox series

[block/for-next] iocost: Fix incorrect operation order during iocg free

Message ID 20190910161525.GT2263813@devbig004.ftw2.facebook.com (mailing list archive)
State New, archived
Headers show
Series [block/for-next] iocost: Fix incorrect operation order during iocg free | expand

Commit Message

Tejun Heo Sept. 10, 2019, 4:15 p.m. UTC
ioc_pd_free() first cancels the hrtimers and then deactivates the
iocg.  However, the iocg timer can run inbetween and reschedule the
hrtimers which will end up running after the iocg is freed leading to
crashes like the following.

  general protection fault: 0000 [#1] SMP
  ...
  RIP: 0010:iocg_kick_delay+0xbe/0x1b0
  RSP: 0018:ffffc90003598ea0 EFLAGS: 00010046
  RAX: 1cee00fd69512b54 RBX: ffff8881bba48400 RCX: 00000000000003e8
  RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff8881bba48400
  RBP: 0000000000004e20 R08: 0000000000000002 R09: 00000000000003e8
  R10: 0000000000000000 R11: 0000000000000000 R12: ffffc90003598ef0
  R13: 00979f3810ad461f R14: ffff8881bba4b400 R15: 25439f950d26e1d1
  FS:  0000000000000000(0000) GS:ffff88885f800000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007f64328c7e40 CR3: 0000000002409005 CR4: 00000000003606e0
  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
  DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
  Call Trace:
   <IRQ>
   iocg_delay_timer_fn+0x3d/0x60
   __hrtimer_run_queues+0xfe/0x270
   hrtimer_interrupt+0xf4/0x210
   smp_apic_timer_interrupt+0x5e/0x120
   apic_timer_interrupt+0xf/0x20
   </IRQ>

Fix it by canceling hrtimers after deactivating the iocg.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Dave Jones <davej@codemonkey.org.uk>
Fixes: 7caa47151ab2 ("blkcg: implement blk-iocost")
---
 block/blk-iocost.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

Comments

Jens Axboe Sept. 10, 2019, 6:21 p.m. UTC | #1
On 9/10/19 10:15 AM, Tejun Heo wrote:
> ioc_pd_free() first cancels the hrtimers and then deactivates the
> iocg.  However, the iocg timer can run inbetween and reschedule the
> hrtimers which will end up running after the iocg is freed leading to
> crashes like the following.
> 
>    general protection fault: 0000 [#1] SMP
>    ...
>    RIP: 0010:iocg_kick_delay+0xbe/0x1b0
>    RSP: 0018:ffffc90003598ea0 EFLAGS: 00010046
>    RAX: 1cee00fd69512b54 RBX: ffff8881bba48400 RCX: 00000000000003e8
>    RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff8881bba48400
>    RBP: 0000000000004e20 R08: 0000000000000002 R09: 00000000000003e8
>    R10: 0000000000000000 R11: 0000000000000000 R12: ffffc90003598ef0
>    R13: 00979f3810ad461f R14: ffff8881bba4b400 R15: 25439f950d26e1d1
>    FS:  0000000000000000(0000) GS:ffff88885f800000(0000) knlGS:0000000000000000
>    CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>    CR2: 00007f64328c7e40 CR3: 0000000002409005 CR4: 00000000003606e0
>    DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>    DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>    Call Trace:
>     <IRQ>
>     iocg_delay_timer_fn+0x3d/0x60
>     __hrtimer_run_queues+0xfe/0x270
>     hrtimer_interrupt+0xf4/0x210
>     smp_apic_timer_interrupt+0x5e/0x120
>     apic_timer_interrupt+0xf/0x20
>     </IRQ>
> 
> Fix it by canceling hrtimers after deactivating the iocg.

Applied, thanks.
diff mbox series

Patch

diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index 2aae8ec391ef..7af350293c2f 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -1957,15 +1957,15 @@  static void ioc_pd_free(struct blkg_policy_data *pd)
 	struct ioc *ioc = iocg->ioc;
 
 	if (ioc) {
-		hrtimer_cancel(&iocg->waitq_timer);
-		hrtimer_cancel(&iocg->delay_timer);
-
 		spin_lock(&ioc->lock);
 		if (!list_empty(&iocg->active_list)) {
 			propagate_active_weight(iocg, 0, 0);
 			list_del_init(&iocg->active_list);
 		}
 		spin_unlock(&ioc->lock);
+
+		hrtimer_cancel(&iocg->waitq_timer);
+		hrtimer_cancel(&iocg->delay_timer);
 	}
 	kfree(iocg);
 }