linux-kernel - Re: [PATCH] blk-mq: remove needless goto from blk_mq_get_driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 23 Oct 2019 21:32:52 -0600
From:   Jens Axboe <axboe@...nel.dk>
To:     Bart Van Assche <bvanassche@....org>,
        André Almeida <andrealmeid@...labora.com>,
        linux-block@...r.kernel.org, linux-kernel@...r.kernel.org
Cc:     kernel@...labora.com, krisman@...labora.com
Subject: Re: [PATCH] blk-mq: remove needless goto from blk_mq_get_driver_tag

On 10/23/19 7:34 PM, Bart Van Assche wrote:
> On 2019-10-22 10:41, André Almeida wrote:
>> The only usage of the label "done" is when (rq->tag != -1) at the
>> begging of the function. Rather than jumping to label, we can just
>> remove this label and execute the code at the "if". Besides that,
>> the code that would be executed after the label "done" is the return of
>> the logical expression (rq->tag != -1) but since we are already inside
>> the if, we now that this is true. Remove the label and replace the goto
>> with the proper result of the label.
>>
>> Signed-off-by: André Almeida <andrealmeid@...labora.com>
>> ---
>> Hello,
>>
>> I've used `blktest` to check if this change add any regression. I have
>> used `./check block` and I got the same results with and without this
>> patch (a bunch of "passed" and three "not run" because of the virtual
>> scsi capabilities). Please let me know if there would be a better way to
>> test changes at block stack.
>>
>> This commit was rebase at linux-block/for-5.5/block.
>>
>> Thanks,
>> 	André
>> ---
>>   block/blk-mq.c | 3 +--
>>   1 file changed, 1 insertion(+), 2 deletions(-)
>>
>> diff --git a/block/blk-mq.c b/block/blk-mq.c
>> index 8538dc415499..1e067b78ab97 100644
>> --- a/block/blk-mq.c
>> +++ b/block/blk-mq.c
>> @@ -1036,7 +1036,7 @@ bool blk_mq_get_driver_tag(struct request *rq)
>>   	bool shared;
>>   
>>   	if (rq->tag != -1)
>> -		goto done;
>> +		return true;
>>   
>>   	if (blk_mq_tag_is_reserved(data.hctx->sched_tags, rq->internal_tag))
>>   		data.flags |= BLK_MQ_REQ_RESERVED;
>> @@ -1051,7 +1051,6 @@ bool blk_mq_get_driver_tag(struct request *rq)
>>   		data.hctx->tags->rqs[rq->tag] = rq;
>>   	}
>>   
>> -done:
>>   	return rq->tag != -1;
>>   }
> 
> Do we really need code changes like the above? I'm not aware of any text
> in the Documentation/ directory that forbids the use of goto statements.

Agree, it looks fine as-is. It's also a fast path, so I'd never get rid
of that without looking at the generated code.

-- 
Jens Axboe