linux-kernel - Re: [PATCH RFC 2/2] wcslen() prototype in string.h

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: <Z-PCCCAPS4uvL3jZ@smile.fi.intel.com>
Date: Wed, 26 Mar 2025 10:59:52 +0200
From: Andy Shevchenko <andy@...nel.org>
To: Nathan Chancellor <nathan@...nel.org>
Cc: Kees Cook <kees@...nel.org>,
	Nick Desaulniers <nick.desaulniers+lkml@...il.com>,
	Bill Wendling <morbo@...gle.com>,
	Justin Stitt <justinstitt@...gle.com>,
	Ard Biesheuvel <ardb@...nel.org>, linux-kernel@...r.kernel.org,
	linux-hardening@...r.kernel.org, llvm@...ts.linux.dev,
	linux-efi@...r.kernel.org
Subject: Re: [PATCH RFC 2/2] wcslen() prototype in string.h

On Tue, Mar 25, 2025 at 05:33:03PM -0700, Nathan Chancellor wrote:
> On Tue, Mar 25, 2025 at 02:45:21PM -0700, Nathan Chancellor wrote:

...

> > +#include <linux/nls.h>		/* for wchar_t */
> 
> Good thing I waited :) This include makes s390 unhappy:
> 
> https://lore.kernel.org/202503260611.MDurOUhF-lkp@intel.com/
> 
> It is possible that should be fixed by adding -Wno-pointer-sign to
> KBUILD_CFLAGS_DECOMPRESSOR so that arch/s390/boot matches the rest of
> the kernel but...

Ah, yes, you beat me up to commenting on this, the string.h and string.c made
in a way that they may be and are used in early boot code, i.e. it must not be
dirtyfied with the kernel code.

...

> >  #include <linux/errno.h>
> >  #include <linux/limits.h>
> >  #include <linux/linkage.h>
> > +#include <linux/nls.h>
> >  #include <linux/stddef.h>
> >  #include <linux/string.h>
> >  #include <linux/types.h>
> 
> I wonder if would be better to do something like the below patch in lieu
> of the EFI change above (since there is no chance for a collision) then
> change both of the includes for wchar_t in this diff to nls_types.h? I
> have no strong opinion but this seems like it would be cleaner for the
> sake of backports while not being a bad solution upstream?

>  #define _LINUX_NLS_H
>  
>  #include <linux/init.h>

As I just replied to your previous mail, consider fixing this list as well
by adding module.h and types.h.

...

Overall, can you browse the Ingo's series [1] for the stuff related to this,
if any?

I would avoid doing double efforts or different approaches if we already have
something ready.

[1]: https://lore.kernel.org/linux-kernel/YjBr10JXLGHfEFfi@gmail.com/

-- 
With Best Regards,
Andy Shevchenko