Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
Date:   Thu, 7 Apr 2022 23:03:49 -0700
From:   Christoph Hellwig <hch@infradead.org>
To:     Logan Gunthorpe <logang@deltatee.com>
Cc:     linux-kernel@vger.kernel.org, linux-raid@vger.kernel.org,
        Song Liu <song@kernel.org>, Shaohua Li <shli@kernel.org>,
        Guoqing Jiang <guoqing.jiang@linux.dev>,
        Stephen Bates <sbates@raithlin.com>,
        Martin Oliveira <Martin.Oliveira@eideticom.com>,
        David Sloan <David.Sloan@eideticom.com>
Subject: Re: [PATCH v1 4/8] md/raid5: Make common label for schedule/retry in
 raid5_make_request()
Message-ID: <Yk/QRWBOUhglpFn6@infradead.org>
References: <20220407164511.8472-1-logang@deltatee.com>
 <20220407164511.8472-5-logang@deltatee.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220407164511.8472-5-logang@deltatee.com>
Precedence: bulk

On Thu, Apr 07, 2022 at 10:45:07AM -0600, Logan Gunthorpe wrote:
> Cleanup the code to make a common label for the schedule,
> prepare_to_wait() and retry path. This drops the do_prepare boolean.
> 
> This requires moveing the prepare_to_wait() above the
> read_seqcount_begin() call on the retry path. However there's no
> appearant requirement for ordering between these two calls.
> 
> This should hopefully be easier to read rather than following the
> extra do_prepare boolean, but it will also be used in a subsequent
> patch to add more code common to all schedule() calls.

Hm.  Maybe it is marginally better, but I always hate gotos inside
of loop bodys.  What prevents us from factoring most of the loop
body into a helper function?