All of lore.kernel.org
 help / color / mirror / Atom feed
From: Bart Van Assche <bvanassche@acm.org>
To: Ming Lei <ming.lei@redhat.com>
Cc: Jens Axboe <axboe@kernel.dk>,
	linux-block@vger.kernel.org, Khazhy Kumykov <khazhy@google.com>,
	Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>,
	Hannes Reinecke <hare@suse.de>,
	John Garry <john.garry@huawei.com>,
	David Jeffery <djeffery@redhat.com>
Subject: Re: [PATCH V2 2/3] blk-mq: complete request locally if the completion is from tagset iterator
Date: Tue, 27 Apr 2021 07:53:09 -0700	[thread overview]
Message-ID: <023a5045-bc1d-ecf4-784d-6de9adda85ba@acm.org> (raw)
In-Reply-To: <YIe4CzfUDX4yCCNO@T590>

On 4/27/21 12:06 AM, Ming Lei wrote:
> On Mon, Apr 26, 2021 at 07:30:51PM -0700, Bart Van Assche wrote:
>> On 4/26/21 6:45 PM, Ming Lei wrote:
>>> diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
>>> index 100fa44d52a6..773aea4db90c 100644
>>> --- a/block/blk-mq-tag.c
>>> +++ b/block/blk-mq-tag.c
>>> @@ -284,8 +284,11 @@ static bool bt_tags_iter(struct sbitmap *bitmap, unsigned int bitnr, void *data)
>>>  	if ((iter_data->flags & BT_TAG_ITER_STARTED) &&
>>>  	    !blk_mq_request_started(rq))
>>>  		ret = true;
>>> -	else
>>> +	else {
>>> +		rq->rq_flags |= RQF_ITERATING;
>>>  		ret = iter_data->fn(rq, iter_data->data, reserved);
>>> +		rq->rq_flags &= ~RQF_ITERATING;
>>> +	}
>>>  	if (!iter_static_rqs)
>>>  		blk_mq_put_rq_ref(rq);
>>>  	return ret;
>>
>> All existing rq->rq_flags modifications are serialized. The above change
>> adds code that may change rq_flags concurrently with regular request
>> processing. I think that counts as a race condition.
> 
> Good catch, but we still can change .rq_flags via atomic op, such as:
> 
> 	do {
> 		old = rq->rq_flags;
> 		new = old | RQF_ITERATING;
> 	} while (cmpxchg(&rq->rq_flags, old, new) != old);

That's not sufficient because the above would not work correctly in
combination with statements like the following:

	rq->rq_flags &= ~RQF_MQ_INFLIGHT;
	req->rq_flags |= RQF_TIMED_OUT;

How about setting a flag in 'current', just like the memalloc_noio_*()
functions set or clear PF_MEMALLOC_NOIO to indicate whether or not
GFP_NOIO should be used? That should work fine in thread context and
also in interrupt context.

>> Additionally, the
>> RQF_ITERATING flag won't be set correctly in the (unlikely) case that
>> two concurrent bt_tags_iter() calls examine the same request at the same
>> time.
> 
> If the driver completes request from two concurrent bt_tags_iter(), there has
> been big trouble of double completion, so I'd rather not to consider this case.

bt_tags_iter() may be used for other purposes than completing requests.
Here is an example of a blk_mq_tagset_busy_iter() call (from debugfs)
that may run concurrently with other calls of that function:

static int hctx_busy_show(void *data, struct seq_file *m)
{
	struct blk_mq_hw_ctx *hctx = data;
	struct show_busy_params params = { .m = m, .hctx = hctx };

	blk_mq_tagset_busy_iter(hctx->queue->tag_set, hctx_show_busy_rq,
				&params);

	return 0;
}

Bart.

  parent reply	other threads:[~2021-04-27 14:53 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-27  1:45 [PATCH V2 0/3] blk-mq: fix request UAF related with iterating over tagset requests Ming Lei
2021-04-27  1:45 ` [PATCH V2 1/3] blk-mq: grab rq->refcount before calling ->fn in blk_mq_tagset_busy_iter Ming Lei
2021-04-27  2:28   ` Bart Van Assche
2021-04-27  2:45     ` Ming Lei
2021-04-27  1:45 ` [PATCH V2 2/3] blk-mq: complete request locally if the completion is from tagset iterator Ming Lei
2021-04-27  2:30   ` Bart Van Assche
2021-04-27  7:06     ` Ming Lei
2021-04-27  9:00       ` Ming Lei
2021-04-27 14:53       ` Bart Van Assche [this message]
2021-04-27  1:45 ` [PATCH V2 3/3] blk-mq: clear stale request in tags->rq[] before freeing one request pool Ming Lei
2021-04-27  2:34   ` Bart Van Assche

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=023a5045-bc1d-ecf4-784d-6de9adda85ba@acm.org \
    --to=bvanassche@acm.org \
    --cc=axboe@kernel.dk \
    --cc=djeffery@redhat.com \
    --cc=hare@suse.de \
    --cc=john.garry@huawei.com \
    --cc=khazhy@google.com \
    --cc=linux-block@vger.kernel.org \
    --cc=ming.lei@redhat.com \
    --cc=shinichiro.kawasaki@wdc.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.