linux-kernel - Re: [PATCH 3/3] net/ceph/messenger: add empty check to ceph_con_get_out

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] [day] [month] [year] [list]

Message-ID: <CAOi1vP-FVKUUvQQT7=EuHij9uerqRvdrTQegMch4_JYQp64Qvg@mail.gmail.com>
Date: Thu, 9 Oct 2025 15:01:31 +0200
From: Ilya Dryomov <idryomov@...il.com>
To: Max Kellermann <max.kellermann@...os.com>
Cc: xiubli@...hat.com, amarkuze@...hat.com, ceph-devel@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/3] net/ceph/messenger: add empty check to ceph_con_get_out_msg()

On Thu, Oct 9, 2025 at 1:47 PM Max Kellermann <max.kellermann@...os.com> wrote:
>
> On Thu, Oct 9, 2025 at 1:18 PM Ilya Dryomov <idryomov@...il.com> wrote:
> > I made a change to net/ceph/messenger_v1.c hunks of this patch to
> > follow what is done for msgr2 where ceph_con_get_out_msg() is called
> > outside of the prepare helper and the new message is passed in.
> > prepare_write_message() doesn't need to return a bool anymore.
>
> But ... why?
> Your change is not bad, but I don't believe it belongs in this patch,
> because it is out of this patch's scope. It would have been a good
> follow-up patch.

Changing a function to return a bool only to immediately undo that
change in a follow-up patch (both touching the same 10-15 lines of
code) seemed like pointless churn to me.

>
> There are lots of unnecessary (and sometimes confusing) differences
> between the v1 and v2 messengers, but unifying these is out of the
> scope of my patch. All my patch does is remove visibility of a
> messenger.c implementation detail from the v1/v2 specializations.

ceph_con_get_out_msg() is a common helper and given that this series
changed its signature and how it's supposed to be used, I wouldn't say
that adjusting the specializations to do the same thing with it is out
of scope.  This isn't unifying some completely unrelated aspect of v1
vs v2 and the resulting patch actually ended up being smaller.

Thanks,

                Ilya