[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <x497fss196b.fsf@segfault.boston.devel.redhat.com>
Date: Fri, 01 May 2015 13:47:56 -0400
From: Jeff Moyer <jmoyer@...hat.com>
To: Christoph Hellwig <hch@....de>
Cc: Shaohua Li <shli@...com>, linux-kernel@...r.kernel.org,
axboe@...com, neilb@...e.de
Subject: Re: [PATCH 3/5] blk-mq: fix plugging in blk_sq_make_request
Christoph Hellwig <hch@....de> writes:
>> - if (use_plug && !blk_queue_nomerges(q) &&
>> + if (likely(!is_flush_fua) && !blk_queue_nomerges(q) &&
>
> Please don't sprinkle likely annotations for no go reason. Especially
> on metadata write heavy workloads (e.g. an NFS server) it might be very
> likely.
I have no problem getting rid of the likely, but I was mimicing the mq
case:
blk_mq_make_request:
if (unlikely(is_flush_fua)) {
blk_mq_bio_to_request(rq, bio);
blk_insert_flush(rq);
goto run_queue;
}
So would you also want that unlikely removed?
Cheers,
Jeff
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists