From mboxrd@z Thu Jan  1 00:00:00 1970
From: Andreas Gruenbacher <agruenba@redhat.com>
Date: Thu, 18 Jun 2020 18:17:18 +0200
Subject: [Cluster-devel] [GFS2 PATCH] gfs2: simplify code with nq_init
	functions
In-Reply-To: <1194682176.34134614.1592327973401.JavaMail.zimbra@redhat.com>
References: <850442235.34134581.1592327923140.JavaMail.zimbra@redhat.com>
	<1194682176.34134614.1592327973401.JavaMail.zimbra@redhat.com>
Message-ID: <CAHc6FU7aFDYXGp9Ng5c9XT7j6WUAtGWVTaN9EnYaiguiZ+9uFA@mail.gmail.com>
List-Id: <cluster-devel.redhat.com>
To: cluster-devel.redhat.com
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit

Hi Bob,

On Tue, Jun 16, 2020 at 7:19 PM Bob Peterson <rpeterso@redhat.com> wrote:
> This patch simply cleans up some code to make it more readable by
> using helper functions gfs2_glock_nq_init and dq_uninit instead of
> doing things manually.

the more complicated call sequence [gfs2_holder_init ... gfs2_glock_nq
... gfs2_glock_dq ... gfs2_holder_uninit] produces better code than
just [gfs2_glock_nq_init ... gfs2_glock_dq_uninit]. I wonder if
there's a way to implement those primitives so that the compiler will
automatically produce the better code?

Andreas