linux-hardening - Re: [PATCH] uml: Replace all non-returning strlcpy with strscpy

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]

Date:   Wed, 31 May 2023 10:48:45 -0400
From:   Azeem Shaikh <azeemshaikh38@...il.com>
To:     Geert Uytterhoeven <geert@...ux-m68k.org>
Cc:     Richard Weinberger <richard@....at>, Kees Cook <kees@...nel.org>,
        kernel test robot <lkp@...el.com>,
        Maxim Krasnyansky <maxk@....qualcomm.com>,
        oe-kbuild-all <oe-kbuild-all@...ts.linux.dev>,
        linux-hardening@...r.kernel.org,
        linux-kernel <linux-kernel@...r.kernel.org>,
        anton ivanov <anton.ivanov@...bridgegreys.com>,
        Johannes Berg <johannes@...solutions.net>,
        linux-um <linux-um@...ts.infradead.org>
Subject: Re: [PATCH] uml: Replace all non-returning strlcpy with strscpy

Thanks Geert and Richard for the review.

On Wed, May 31, 2023 at 4:05 AM Geert Uytterhoeven <geert@...ux-m68k.org> wrote:
>
> On Wed, May 31, 2023 at 8:23 AM Richard Weinberger <richard@....at> wrote:
> > ----- Ursprüngliche Mail -----
> > >>>> arch/um/os-Linux/drivers/tuntap_user.c:149:17: error: implicit declaration of
> > >>>> function 'strscpy'; did you mean 'strncpy'?
> > >>>> [-Werror=implicit-function-declaration]
> > >>     149 |                 strscpy(ifr.ifr_name, pri->dev_name,
> > >>     sizeof(ifr.ifr_name));
> > >>         |                 ^~~~~~~
> > >>         |                 strncpy
> > >>   cc1: some warnings being treated as errors
> > >
> > > Ah, yeah, this is another "not actually in the kernel" cases. Let's ignore this
> > > strlcpy for now.
> >
> > Well, actually it's another case of "not even compile tested". :-(

Argh, my test script wasn't cross-compiling for um. Sorry about that :(

> > Well, strlcpy() isn't part of libc either, so all this would need is to
> > add it to user.h just like strlcpy() is now?
>
> I think so.
> Azeem, can you please test your changes with this fixup applied?
>
> diff --git a/arch/um/include/shared/user.h b/arch/um/include/shared/user.h
> index bda66e5a9d4e3..e5d3fbbafe4d2 100644
> --- a/arch/um/include/shared/user.h
> +++ b/arch/um/include/shared/user.h
> @@ -52,6 +52,7 @@ static inline int printk(const char *fmt, ...)
>  extern int in_aton(char *str);
>  extern size_t strlcpy(char *, const char *, size_t);
>  extern size_t strlcat(char *, const char *, size_t);
> +extern ssize_t strscpy(char *, const char *, size_t);
>
>  /* Copied from linux/compiler-gcc.h since we can't include it directly */
>  #define barrier() __asm__ __volatile__("": : :"memory")
>

Tested with this fixup, and it builds successfully on ARCH=um
SUBARCH=i386. Let me know if I need to test with any other
cross-compilation options before I send out v2.

>
> Ah, yeah, this is another "not actually in the kernel" cases. Let's ignore this strlcpy for now.
>
> -Keed

Planning to send out v2 with the fixup from Richard applied. Let me
know if that's ok.