linux-kernel - Re: [PATCH] kconfig: WERROR unmet symbol dependency

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] [thread-next>] [day] [month] [year] [list]

Message-ID: <20231222025701.GA3568309@google.com>
Date: Fri, 22 Dec 2023 11:57:01 +0900
From: Sergey Senozhatsky <senozhatsky@...omium.org>
To: Masahiro Yamada <masahiroy@...nel.org>
Cc: Sergey Senozhatsky <senozhatsky@...omium.org>,
	Patrick Georgi <pgeorgi@...gle.com>, linux-kbuild@...r.kernel.org,
	linux-kernel@...r.kernel.org, Stefan Reinauer <reinauer@...gle.com>
Subject: Re: [PATCH] kconfig: WERROR unmet symbol dependency

On (23/12/01 00:42), Masahiro Yamada wrote:
> On Wed, Nov 29, 2023 at 1:13 PM Sergey Senozhatsky
> <senozhatsky@...omium.org> wrote:
> >
> > On (23/11/28 23:19), Masahiro Yamada wrote:
> >
> > [..]
> >
> > > KCONFIG_WERROR is meant to turn all warnings
> > > to errors.
> > > I do not see getenv("KCONFIG_WERROR")
> > > sprinkled everywhere in Kconfig.
> > > One more thing, you cannot directly exit(1)
> > > from sym_calc_value().
> >
> > We do exit(1) for KCONFIG_WARN_UNKNOWN_SYMBOLS in conf_read().
> >
> > I can introduce two new helpers that will tell if confdata.c and symbol.c
> > triggered any warnings and if KCONFIG_WERROR is set. And then different
> > code paths can call them and handle exit gracefully, depending on the
> > context (ncurses, menu, etc.).
> >
> > Something like this
> 
> 
> I do not want to patch warnings one by one.
> 
> 
> I will take some time to think about it.

Gentle ping on this.

We are not concerned with every possible warning at the moment, however,
we do want the critical ones from CI and (semi)automated continuous uprev
PoV to be covered by WERROR. We do experience real life problems with
"missing direct dependency" not being a terminal condition under WERROR.