linux-kernel - Re: [PATCH 1/4] modpost: shorten warning messages in report_sec

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]

Date:   Thu, 4 Aug 2022 01:35:48 +0900
From:   Masahiro Yamada <masahiroy@...nel.org>
To:     Nick Desaulniers <ndesaulniers@...gle.com>
Cc:     Linux Kbuild mailing list <linux-kbuild@...r.kernel.org>,
        Michal Marek <michal.lkml@...kovi.net>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/4] modpost: shorten warning messages in report_sec_mismatch()

On Wed, Aug 3, 2022 at 3:13 AM Nick Desaulniers <ndesaulniers@...gle.com> wrote:
>
> On Mon, Aug 1, 2022 at 2:41 AM Masahiro Yamada <masahiroy@...nel.org> wrote:
> >
> > Each section mismatch results in long warning messages. Too much.
>
> :(
>
> Yes; they are too verbose.  That said, I have found the
> recommendations about annotations for function attributes handy in the
> past and would be sad to see them go.  They remind me of "note"
> diagnostics from the compiler that add additional context to
> "warning"/"error" diagnostics on what the recommended next steps are
> for fixing them.
>
> Is there a "happy middle ground" here?

I do not know.
modpost became painfully ugly.


Moreover, the current hint is not necessarily precise.
("lacks a __initdata" in the following)



[sample code]
int dummy __initdata;
void set_dummy(void) { dummy = 1; }


[warning]
WARNING: modpost: vmlinux.o(.text+0x194412): Section mismatch in
reference from the function set_dummy() to the variable
.init.data:dummy
The function set_dummy() references
the variable __initdata dummy.
This is often because set_dummy lacks a __initdata
annotation or the annotation of dummy is wrong.




--
Best Regards


Masahiro Yamada