On 4/10/18 8:05 PM, Ming Lei wrote:
> On Tue, Apr 10, 2018 at 02:30:35PM -0600, Bart Van Assche wrote:
>> Because blkcg_exit_queue() is now called from inside blk_cleanup_queue()
>> it is no longer safe to access cgroup information during or after the
>> blk_cleanup_queue() call. Hence protect the generic_make_request_checks()
>> call with blk_queue_enter() / blk_queue_exit().
>>
>> Reported-by: Ming Lei <ming.lei@xxxxxxxxxx>
>> Fixes: a063057d7c73 ("block: Fix a race between request queue removal and the block cgroup controller")
>> Signed-off-by: Bart Van Assche <bart.vanassche@xxxxxxx>
>> Cc: Ming Lei <ming.lei@xxxxxxxxxx>
>> Cc: Joseph Qi <joseph.qi@xxxxxxxxxxxxxxxxx>
>> ---
>> block/blk-core.c | 32 ++++++++++++++++++++++++++------
>> 1 file changed, 26 insertions(+), 6 deletions(-)
>>
>> diff --git a/block/blk-core.c b/block/blk-core.c
>> index 34e2f2227fd9..a330cd2829e1 100644
>> --- a/block/blk-core.c
>> +++ b/block/blk-core.c
>> @@ -2386,8 +2386,19 @@ blk_qc_t generic_make_request(struct bio *bio)
>> * yet.
>> */
>> struct bio_list bio_list_on_stack[2];
>> + blk_mq_req_flags_t flags = bio->bi_opf & REQ_NOWAIT ?
>> + BLK_MQ_REQ_NOWAIT : 0;
>> + struct request_queue *q = bio->bi_disk->queue;
>> blk_qc_t ret = BLK_QC_T_NONE;
>>
>> + if (blk_queue_enter(q, flags) < 0) {
>
> As I mentioned last time, the queue pointer has to be checked before
> calling blk_queue_enter(), since it isn't difficult to trigger the
> failure log of 'generic_make_request: Trying to access nonexistent
> block-device'.
That's a good point, the NULL check needs to be included first. Bart,
do you want to send an update? I can just rebase the for-linus branch.
--
Jens Axboe
