Re: [PATCH 1/7] blk-mq: factor out a blk_rq_init_flush helper

From: Christoph Hellwig <hch@lst.de>
To: Damien Le Moal <dlemoal@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>, Jens Axboe <axboe@kernel.dk>,
	Bart Van Assche <bvanassche@acm.org>,
	linux-block@vger.kernel.org
Subject: Re: [PATCH 1/7] blk-mq: factor out a blk_rq_init_flush helper
Date: Mon, 17 Apr 2023 08:24:08 +0200	[thread overview]
Message-ID: <20230417062408.GA2566@lst.de> (raw)
In-Reply-To: <3947a26e-ceb6-4117-7b45-fd7578710120@kernel.org>

On Mon, Apr 17, 2023 at 03:22:21PM +0900, Damien Le Moal wrote:
> > +static void blk_rq_init_flush(struct request *rq)
> > +{
> > +	memset(&rq->flush, 0, sizeof(rq->flush));
> > +	INIT_LIST_HEAD(&rq->flush.list);
> > +	rq->rq_flags |= RQF_FLUSH_SEQ;
> > +	rq->flush.saved_end_io = rq->end_io; /* Usually NULL */
> > +	rq->end_io = mq_flush_data_end_io;
> > +}
> 
> struct flush is:
> 
> struct {
> 	unsigned int		seq;
> 	struct list_head	list;
> 	rq_end_io_fn		*saved_end_io;
> } flush;
> 
> So given that list and saved_end_io are initialized here, we could remove the
> memset() by initializing seq "manually":

Yes.  And seq is always initialized in the callers, and I think I can also
removed saved_end_io entirely.  So maybe we can drop this patch in the
end and just do some other changes.