Re: bfq: BUG bfq_queue: Objects remaining in bfq_queue on __kmem_cache_shutdown() after rmmod

On 12/21/2017 03:53 PM, Paolo Valente wrote:

quoted

Il giorno 21 dic 2017, alle ore 08:08, Guoqing Jiang [off-list ref] ha scritto:

Hi,


On 12/08/2017 08:34 AM, Holger Hoffstätte wrote:

quoted

So plugging in a device on USB with BFQ as scheduler now works without
hiccup (probably thanks to Ming Lei's last patch), but of course I found
another problem. Unmounting the device after use, changing the scheduler
back to deadline or kyber and rmmod'ing the BFQ module reproducibly gives me:

kernel: =============================================================================
kernel: BUG bfq_queue (Tainted: G    B          ): Objects remaining in bfq_queue on __kmem_cache_shutdown()
kernel: -----------------------------------------------------------------------------
kernel:
kernel: INFO: Slab 0xffffea001601fc00 objects=37 used=3 fp=0xffff8805807f0360 flags=0x8000000000008100
kernel: CPU: 0 PID: 9967 Comm: rmmod Tainted: G    B           4.14.5 #1
kernel: Hardware name: Gigabyte Technology Co., Ltd. P67-DS3-B3/P67-DS3-B3, BIOS F1 05/06/2011
kernel: Call Trace:
kernel:  dump_stack+0x46/0x5e
kernel:  slab_err+0x9e/0xb0
kernel:  ? on_each_cpu_mask+0x35/0x40
kernel:  ? ksm_migrate_page+0x60/0x60
kernel:  ? __kmalloc+0x1c9/0x1d0
kernel:  __kmem_cache_shutdown+0x177/0x350
kernel:  shutdown_cache+0xf/0x130
kernel:  kmem_cache_destroy+0x19e/0x1b0
kernel:  SyS_delete_module+0x168/0x230
kernel:  ? exit_to_usermode_loop+0x39/0x80
kernel:  entry_SYSCALL_64_fastpath+0x13/0x94
kernel: RIP: 0033:0x7f53e4136b97
kernel: RSP: 002b:00007ffd660061d8 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0
kernel: RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f53e4136b97
kernel: RDX: 000000000000000a RSI: 0000000000000800 RDI: 00000000006247f8
kernel: RBP: 0000000000000000 R08: 00007ffd66005171 R09: 0000000000000000
kernel: R10: 00007f53e41acbc0 R11: 0000000000000206 R12: 0000000000624790
kernel: R13: 00007ffd660051d0 R14: 0000000000624790 R15: 0000000000623260
kernel: INFO: Object 0xffff8805807f0000 @offset=0
kernel: INFO: Object 0xffff8805807f01b0 @offset=432
kernel: INFO: Object 0xffff8805807f3cc0 @offset=15552
kernel: kmem_cache_destroy bfq_queue: Slab cache still has objects
kernel: CPU: 0 PID: 9967 Comm: rmmod Tainted: G    B           4.14.5 #1
kernel: Hardware name: Gigabyte Technology Co., Ltd. P67-DS3-B3/P67-DS3-B3, BIOS F1 05/06/2011
kernel: Call Trace:
kernel:  dump_stack+0x46/0x5e
kernel:  kmem_cache_destroy+0x191/0x1b0
kernel:  SyS_delete_module+0x168/0x230
kernel:  ? exit_to_usermode_loop+0x39/0x80
kernel:  entry_SYSCALL_64_fastpath+0x13/0x94
kernel: RIP: 0033:0x7f53e4136b97
kernel: RSP: 002b:00007ffd660061d8 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0
kernel: RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f53e4136b97
kernel: RDX: 000000000000000a RSI: 0000000000000800 RDI: 00000000006247f8
kernel: RBP: 0000000000000000 R08: 00007ffd66005171 R09: 0000000000000000
kernel: R10: 00007f53e41acbc0 R11: 0000000000000206 R12: 0000000000624790
kernel: R13: 00007ffd660051d0 R14: 0000000000624790 R15: 0000000000623260

I also encountered the similar bug, seems there are some async objects which are not freed with
BFQ_GROUP_IOSCHED build as "Y".

If revert part of commit e21b7a0b988772e82e7147e1c659a5afe2ae003c ("block, bfq: add full hierarchical
scheduling and cgroups support")like below, then I can't see the bug again with simple test.

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 889a8549d97f..c640e64e042f 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c

@@ -4710,6 +4710,7 @@ static void bfq_exit_queue(struct elevator_queue *e)
         spin_lock_irq(&bfqd->lock);
         list_for_each_entry_safe(bfqq, n, &bfqd->idle_list, bfqq_list)
                 bfq_deactivate_bfqq(bfqd, bfqq, false, false);
+       bfq_put_async_queues(bfqd, bfqd->root_group);
         spin_unlock_irq(&bfqd->lock);

         hrtimer_cancel(&bfqd->idle_slice_timer);

@@ -4718,7 +4719,6 @@ static void bfq_exit_queue(struct elevator_queue *e)
         blkcg_deactivate_policy(bfqd->queue, &blkcg_policy_bfq);
  #else
         spin_lock_irq(&bfqd->lock);
-       bfq_put_async_queues(bfqd, bfqd->root_group);
         kfree(bfqd->root_group);
         spin_unlock_irq(&bfqd->lock);

But perhaps we should do it inside blkcg_deactivate_policy, right?

Thank you very much for investigating this. And yes, I removed that bfq_put_async_queues invocation a long ago, because it had to be done in blkcg_deactivate_policy; and actually it was invoked as expected. We are trying to understand what is going wrong now.

Thanks for the in formations. I find another way, does it make sense?

diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index da1525ec4c87..0a070daf96c7 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c

@@ -474,7 +474,10 @@ static void bfq_pd_init(struct blkg_policy_data *pd)
  static void bfq_pd_free(struct blkg_policy_data *pd)
  {
         struct bfq_group *bfqg = pd_to_bfqg(pd);
+       struct bfq_data *bfqd = bfqg->bfqd;

+       if (bfqd)
+               bfq_put_async_queues(bfqd, bfqg);
         bfqg_stats_exit(&bfqg->stats);
         bfqg_put(bfqg);
  }

Regards,
Guoqing