linux-kernel - Re: [PATCH net v3] selftest:net: Fix uninit return values

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: <CAJE-K+AeEYkAN8wX3FbBCbQMGTDsueA-YiC4w_qi+TZgUzkS-w@mail.gmail.com>
Date: Tue, 30 Sep 2025 15:17:02 +0530
From: Sidharth Seela <sidharthseela@...il.com>
To: Simon Horman <horms@...nel.org>
Cc: antonio@...nvpn.net, sd@...asysnail.net, edumazet@...gle.com, 
	kuba@...nel.org, pabeni@...hat.com, shuah@...nel.org, 
	willemdebruijn.kernel@...il.com, kernelxing@...cent.com, nathan@...nel.org, 
	nick.desaulniers+lkml@...il.com, morbo@...gle.com, justinstitt@...gle.com, 
	netdev@...r.kernel.org, linux-kselftest@...r.kernel.org, 
	linux-kernel@...r.kernel.org, llvm@...ts.linux.dev
Subject: Re: [PATCH net v3] selftest:net: Fix uninit return values

On Tue, Sep 30, 2025 at 2:39 PM Simon Horman <horms@...nel.org> wrote:
> Hi,
>
> I don't want to block progress.
> But there are some format problems with the commit message.
>
> Locally, git truncates the commit message at the line above ('--').
> Which, omits a lot of useful information.
> Most critically your Signed-off-by line.
>
> There is also another '--' below. Just above the fixes tag.
> Which would cause a similar problem.
>
> And the v2/v3 information should go below the scissors ('---'),
> below your signed-off by line.
>
> Maybe the maintainers can fix this when applying,
> given how close we are to the pull for v6.18-rc1.
> And that I believe there has already been some
> discussion of this patch with the maintainers.
>
> > ovpn-cli.c:1587:6: warning: variable 'ret' is used uninitialized whenever 'if' condition is true [-Wsometimes-uninitialized]
> >  1587 |         if (!sock) {
> >       |             ^~~~~
> > ovpn-cli.c:1635:9: note: uninitialized use occurs here
> >  1635 |         return ret;
> >       |                ^~~
> > ovpn-cli.c:1587:2: note: remove the 'if' if its condition is always false
> >  1587 |         if (!sock) {
> >       |         ^~~~~~~~~~~~
> >  1588 |                 fprintf(stderr, "cannot allocate netlink socket\n");
> >       |                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >  1589 |                 goto err_free;
> >       |                 ~~~~~~~~~~~~~~
> >  1590 |         }
> >       |         ~
> > ovpn-cli.c:1584:15: note: initialize the variable 'ret' to silence this warning
> >  1584 |         int mcid, ret;
> >       |                      ^
> >       |                       = 0
> > ovpn-cli.c:2107:7: warning: variable 'ret' is used uninitialized whenever switch case is taken [-Wsometimes-uninitialized]
> >  2107 |         case CMD_INVALID:
> >       |              ^~~~~~~~~~~
> > ovpn-cli.c:2111:9: note: uninitialized use occurs here
> >  2111 |         return ret;
> >       |                ^~~
> > ovpn-cli.c:1939:12: note: initialize the variable 'ret' to silence this warning
> >  1939 |         int n, ret;
> >       |                   ^
> >       |
> > --
> > Fixes: 959bc330a439 ("testing/selftests: add test tool and scripts for ovpn module")
> > ovpn module")
> >
> > v3:
> >       - Use prefix net.
> >       - Remove so_txtime fix as default case calls error().
> >       - Changelog before sign-off.
> >       - Three dashes after sign-off
> >
> > v2:
> >       - Use subsystem name "net".
> >       - Add fixes tags.
> >       - Remove txtimestamp fix as default case calls error.
> >       - Assign constant error string instead of NULL.
> >
> > Signed-off-by: Sidharth Seela <sidharthseela@...il.com>
> > ---
> >
>
> This is where the v2/v3 information should go.
>
> ...

Thankyou Simon, I didn't know that double hyphen would cause
an issue. Although I need a logical separator between commit message
and warning log, may I ask what could be used instead?

-- 
Thanks,
Sidharth Seela
www.realtimedesign.org