block, bfq: NULL out the bic when it's no longer valid
Subsystems (2)
PATCH 1/1 block, bfq: NULL out the bic when it's no longer valid
In reboot tests on several devices we were seeing a "use after free" when slub_debug or KASAN was enabled. The kernel complained about:
Unable to handle kernel paging request at virtual address 6b6b6c2b
...which is a classic sign of use after free under slub_debug. The stack crawl in kgdb looked like:
0 test_bit (addr=<optimized out>, nr=<optimized out>) 1 bfq_bfqq_busy (bfqq=<optimized out>) 2 bfq_select_queue (bfqd=<optimized out>) 3 __bfq_dispatch_request (hctx=<optimized out>) 4 bfq_dispatch_request (hctx=<optimized out>) 5 0xc056ef00 in blk_mq_do_dispatch_sched (hctx=0xed249440) 6 0xc056f728 in blk_mq_sched_dispatch_requests (hctx=0xed249440) 7 0xc0568d24 in __blk_mq_run_hw_queue (hctx=0xed249440) 8 0xc0568d94 in blk_mq_run_work_fn (work=<optimized out>) 9 0xc024c5c4 in process_one_work (worker=0xec6d4640, work=0xed249480)
10 0xc024cff4 in worker_thread (__worker=0xec6d4640)
Digging in kgdb, it could be found that, though bfqq looked fine, bfqq->bic had been freed.
Through further digging, I postulated that perhaps it is illegal to access a "bic" (AKA an "icq") after bfq_exit_icq() had been called because the "bic" can be freed at some point in time after this call is made. I confirmed that there certainly were cases where the exact crashing code path would access the "bic" after bfq_exit_icq() had been called. Sspecifically I set the "bfqq->bic" to (void *)0x7 and saw that the bic was 0x7 at the time of the crash.
To understand a bit more about why this crash was fairly uncommon (I saw it only once in a few hundred reboots), you can see that much of the time bfq_exit_icq_fbqq() fully frees the bfqq and thus it can't access the ->bic anymore. The only case it doesn't is if bfq_put_queue() sees a reference still held.
However, even in the case when bfqq isn't freed, the crash is still rare. Why? I tracked what happened to the "bic" after the exit routine. It doesn't get freed right away. Rather, put_io_context_active() eventually called put_io_context() which queued up freeing on a workqueue. The freeing then actually happened later than that through call_rcu(). Despite all these delays, some extra debugging showed that all the hoops could be jumped through in time and the memory could be freed causing the original crash. Phew!
To make a long story short, assuming it truly is illegal to access an icq after the "exit_icq" callback is finished, this patch is needed.
---
Most of the testing of this was done on the Chrome OS 4.19 kernel with BFQ backported (thanks to Paolo's help). I did manage to reproduce a crash on mainline Linux (v5.2-rc6) though.
To see some of the techniques used to debug this, see
<https://crrev.com/c/1679134> and <https://crrev.com/c/1681258/1>.
I'll also note that on linuxnext (next-20190627) I saw some other use-after-frees that seemed related to BFQ but haven't had time to debug. They seemed unrelated.
block/bfq-iosched.c | 1 + 1 file changed, 1 insertion(+)
block/bfq-iosched.c +1 -0
@@ -4584,6 +4584,7 @@ static void bfq_exit_icq_bfqq(struct bfq_io_cq *bic, bool is_sync)unsigned long flags;spin_lock_irqsave (&bfqd->lock, flags); + bfqq->bic =NULL ;bfq_exit_bfqq (bfqd, bfqq);bic_set_bfqq (bic,NULL , is_sync);spin_unlock_irqrestore (&bfqd->lock, flags);
Signature · 1 line
2.22.0.410.gd8fdbe21b5-goog
Quoted 23 lines · Douglas Anderson
In reboot tests on several devices we were seeing a "use after free" when
slub_debugor KASAN was enabled. The kernel complained about:Unable to handle kernel paging request at virtual address 6b6b6c2b...which is a classic sign of use after free under
slub_debug. The stack crawl in kgdb looked like:0 test_bit (addr=<optimized out>, nr=<optimized out>) 1 bfq_bfqq_busy (bfqq=<optimized out>) 2 bfq_select_queue (bfqd=<optimized out>) 3 __bfq_dispatch_request (hctx=<optimized out>) 4 bfq_dispatch_request (hctx=<optimized out>) 5 0xc056ef00 in blk_mq_do_dispatch_sched (hctx=0xed249440) 6 0xc056f728 in blk_mq_sched_dispatch_requests (hctx=0xed249440) 7 0xc0568d24 in __blk_mq_run_hw_queue (hctx=0xed249440) 8 0xc0568d94 in blk_mq_run_work_fn (work=<optimized out>) 9 0xc024c5c4 in process_one_work (worker=0xec6d4640, work=0xed249480)10
0xc024cff4inworker_thread(__worker=0xec6d4640)Digging in kgdb, it could be found that, though bfqq looked fine,
bfqq->bichad been freed.
Quoted 6 lines
Through further digging, I postulated that perhaps it is illegal to access a "bic" (AKA an "icq") after
bfq_exit_icq()had been called because the "bic" can be freed at some point in time after this call is made. I confirmed that there certainly were cases where the exact crashing code path would access the "bic" afterbfq_exit_icq()had been called. Sspecifically I set the "bfqq->bic" to (void *)0x7 and
^^^
Quoted 17 lines
saw that the bic was
0x7at the time of the crash.To understand a bit more about why this crash was fairly uncommon (I saw it only once in a few hundred reboots), you can see that much of the time
bfq_exit_icq_fbqq()fully frees the bfqq and thus it can't access the ->bic anymore. The only case it doesn't is ifbfq_put_queue()sees a reference still held.However, even in the case when bfqq isn't freed, the crash is still rare. Why? I tracked what happened to the "bic" after the exit routine. It doesn't get freed right away. Rather,
put_io_context_active()eventually calledput_io_context()which queued up freeing on a workqueue. The freeing then actually happened later than that throughcall_rcu(). Despite all these delays, some extra debugging showed that all the hoops could be jumped through in time and the memory could be freed causing the original crash. Phew!
Quoted 4 lines
To make a long story short, assuming it truly is illegal to access an icq after the "
exit_icq" callback is finished, this patch is needed.Signed-off-by Douglas Anderson <[email protected]>
Nicely done ... thanks!
Quoted 30 lines
---
Most of the testing of this was done on the Chrome OS 4.19 kernel with BFQ backported (thanks to Paolo's help). I did manage to reproduce a crash on mainline Linux (v5.2-rc6) though.To see some of the techniques used to debug this, see
<https://crrev.com/c/1679134> and <https://crrev.com/c/1681258/1>.I'll also note that on linuxnext (next-20190627) I saw some other use-after-frees that seemed related to BFQ but haven't had time to debug. They seemed unrelated.
block/bfq-iosched.c | 1 + 1 file changed, 1 insertion(+)block/bfq-iosched.c +1
-0@@ -4584,6 +4584,7 @@ static void bfq_exit_icq_bfqq(struct bfq_io_cq *bic, bool is_sync)unsigned long flags;spin_lock_irqsave (&bfqd->lock, flags); + bfqq->bic =NULL ;bfq_exit_bfqq (bfqd, bfqq);bic_set_bfqq (bic,NULL , is_sync);spin_unlock_irqrestore (&bfqd->lock, flags);Signature · 1 line
2.22.0.410.gd8fdbe21b5-goog
Paolo Valente
Reviewed-byQuoted 28 lines
Il giorno 28 giu 2019, alle ore 06:57, Guenter Roeck <[email protected]> ha scritto:
Quoted 23 lines · Douglas Anderson
In reboot tests on several devices we were seeing a "use after free" when
slub_debugor KASAN was enabled. The kernel complained about:Unable to handle kernel paging request at virtual address 6b6b6c2b...which is a classic sign of use after free under
slub_debug. The stack crawl in kgdb looked like:0 test_bit (addr=<optimized out>, nr=<optimized out>) 1 bfq_bfqq_busy (bfqq=<optimized out>) 2 bfq_select_queue (bfqd=<optimized out>) 3 __bfq_dispatch_request (hctx=<optimized out>) 4 bfq_dispatch_request (hctx=<optimized out>) 5 0xc056ef00 in blk_mq_do_dispatch_sched (hctx=0xed249440) 6 0xc056f728 in blk_mq_sched_dispatch_requests (hctx=0xed249440) 7 0xc0568d24 in __blk_mq_run_hw_queue (hctx=0xed249440) 8 0xc0568d94 in blk_mq_run_work_fn (work=<optimized out>) 9 0xc024c5c4 in process_one_work (worker=0xec6d4640, work=0xed249480)10
0xc024cff4inworker_thread(__worker=0xec6d4640)Digging in kgdb, it could be found that, though bfqq looked fine,
bfqq->bichad been freed.
Quoted 28 lines
Quoted 6 lines
Through further digging, I postulated that perhaps it is illegal to access a "bic" (AKA an "icq") after
bfq_exit_icq()had been called because the "bic" can be freed at some point in time after this call is made. I confirmed that there certainly were cases where the exact crashing code path would access the "bic" afterbfq_exit_icq()had been called. Sspecifically I set the "bfqq->bic" to (void *)0x7 and^^^Quoted 20 lines
saw that the bic was
0x7at the time of the crash.To understand a bit more about why this crash was fairly uncommon (I saw it only once in a few hundred reboots), you can see that much of the time
bfq_exit_icq_fbqq()fully frees the bfqq and thus it can't access the ->bic anymore. The only case it doesn't is ifbfq_put_queue()sees a reference still held.However, even in the case when bfqq isn't freed, the crash is still rare. Why? I tracked what happened to the "bic" after the exit routine. It doesn't get freed right away. Rather,
put_io_context_active()eventually calledput_io_context()which queued up freeing on a workqueue. The freeing then actually happened later than that throughcall_rcu(). Despite all these delays, some extra debugging showed that all the hoops could be jumped through in time and the memory could be freed causing the original crash. Phew!To make a long story short, assuming it truly is illegal to access an icq after the "
exit_icq" callback is finished, this patch is needed.
Nice work! I think this bug has been there since when BFQ was born ...
Thanks,
Paolo
Quoted 35 lines
Quoted 1 line
Signed-off-by Douglas Anderson <[email protected]>Nicely done ... thanks!
Reviewed-by Guenter Roeck <[email protected]>Quoted 29 lines
---
Most of the testing of this was done on the Chrome OS 4.19 kernel with BFQ backported (thanks to Paolo's help). I did manage to reproduce a crash on mainline Linux (v5.2-rc6) though.To see some of the techniques used to debug this, see
<https://crrev.com/c/1679134> and <https://crrev.com/c/1681258/1>.I'll also note that on linuxnext (next-20190627) I saw some other use-after-frees that seemed related to BFQ but haven't had time to debug. They seemed unrelated.
block/bfq-iosched.c| 1 +1 file changed, 1 insertion(+)block/bfq-iosched.c +1
-0@@ -4584,6 +4584,7 @@ static void bfq_exit_icq_bfqq(struct bfq_io_cq *bic, bool is_sync)unsigned long flags;spin_lock_irqsave (&bfqd->lock, flags); + bfqq->bic =NULL ;bfq_exit_bfqq (bfqd, bfqq);bic_set_bfqq (bic,NULL , is_sync);spin_unlock_irqrestore (&bfqd->lock, flags);Signature · 1 line
2.22.0.410.gd8fdbe21b5-goog
Jens Axboe
maintainerQuoted 55 lines · Douglas Anderson
In reboot tests on several devices we were seeing a "use after free" when
slub_debugor KASAN was enabled. The kernel complained about:Unable to handle kernel paging request at virtual address 6b6b6c2b...which is a classic sign of use after free under
slub_debug. The stack crawl in kgdb looked like:0 test_bit (addr=<optimized out>, nr=<optimized out>) 1 bfq_bfqq_busy (bfqq=<optimized out>) 2 bfq_select_queue (bfqd=<optimized out>) 3 __bfq_dispatch_request (hctx=<optimized out>) 4 bfq_dispatch_request (hctx=<optimized out>) 5 0xc056ef00 in blk_mq_do_dispatch_sched (hctx=0xed249440) 6 0xc056f728 in blk_mq_sched_dispatch_requests (hctx=0xed249440) 7 0xc0568d24 in __blk_mq_run_hw_queue (hctx=0xed249440) 8 0xc0568d94 in blk_mq_run_work_fn (work=<optimized out>) 9 0xc024c5c4 in process_one_work (worker=0xec6d4640, work=0xed249480)10
0xc024cff4inworker_thread(__worker=0xec6d4640)Digging in kgdb, it could be found that, though bfqq looked fine,
bfqq->bichad been freed.Through further digging, I postulated that perhaps it is illegal to access a "bic" (AKA an "icq") after
bfq_exit_icq()had been called because the "bic" can be freed at some point in time after this call is made. I confirmed that there certainly were cases where the exact crashing code path would access the "bic" afterbfq_exit_icq()had been called. Sspecifically I set the "bfqq->bic" to (void *)0x7 and saw that the bic was0x7at the time of the crash.To understand a bit more about why this crash was fairly uncommon (I saw it only once in a few hundred reboots), you can see that much of the time
bfq_exit_icq_fbqq()fully frees the bfqq and thus it can't access the ->bic anymore. The only case it doesn't is ifbfq_put_queue()sees a reference still held.However, even in the case when bfqq isn't freed, the crash is still rare. Why? I tracked what happened to the "bic" after the exit routine. It doesn't get freed right away. Rather,
put_io_context_active()eventually calledput_io_context()which queued up freeing on a workqueue. The freeing then actually happened later than that throughcall_rcu(). Despite all these delays, some extra debugging showed that all the hoops could be jumped through in time and the memory could be freed causing the original crash. Phew!To make a long story short, assuming it truly is illegal to access an icq after the "
exit_icq" callback is finished, this patch is needed.Signed-off-by Douglas Anderson <[email protected]>---
Most of the testing of this was done on the Chrome OS 4.19 kernel with BFQ backported (thanks to Paolo's help). I did manage to reproduce a crash on mainline Linux (v5.2-rc6) though.
Quoted 6 lines
To see some of the techniques used to debug this, see
<https://crrev.com/c/1679134> and <https://crrev.com/c/1681258/1>.I'll also note that on linuxnext (next-20190627) I saw some other use-after-frees that seemed related to BFQ but haven't had time to debug. They seemed unrelated.
Applied for 5.3, but I marked it for stable as well.
Signature · 1 line
Jens Axboe