Re: linux-next: Tree for May 16 (security/landlock/ruleset.c)

[Kees Cook <kees@kernel.org>]

On Tue, May 20, 2025 at 05:01:51PM +0300, Andy Shevchenko wrote:
> On Mon, May 19, 2025 at 12:15:30PM -0700, Kees Cook wrote:
> > On Mon, May 19, 2025 at 08:41:17PM +0200, Mickaël Salaün wrote:
> 
> ...
> 
> > >From 6fbf66fdfd0a7dac809b77faafdd72c60112bb8d Mon Sep 17 00:00:00 2001
> > From: Kees Cook <kees@kernel.org>
> > Date: Mon, 19 May 2025 11:52:06 -0700
> > Subject: [PATCH] string.h: Provide basic sanity checks for fallback memcpy()
> > MIME-Version: 1.0
> > Content-Type: text/plain; charset=UTF-8
> > Content-Transfer-Encoding: 8bit
> > 
> > Instead of defining memcpy() in terms of __builtin_memcpy() deep
> > in arch/x86/include/asm/string_32.h, notice that it is needed up in
> > the general string.h, as done with other common C String APIs. This
> > allows us to add basic sanity checking for pathological "size"
> > arguments to memcpy(). Besides the run-time checking benefit, this
> > avoids GCC trying to be very smart about value range tracking[1] when
> > CONFIG_PROFILE_ALL_BRANCHES=y but FORTIFY_SOURCE=n.
> > 
> > Link: https://lore.kernel.org/all/202505191117.C094A90F88@keescook/ [1]
> > Reported-by: kernel test robot <lkp@intel.com>
> > Closes: https://lore.kernel.org/all/202501040747.S3LYfvYq-lkp@intel.com/
> > Reported-by: Randy Dunlap <rdunlap@infradead.org>
> > Closes: https://lore.kernel.org/all/e3754f69-1dea-4542-8de0-a567a14fb95b@infradead.org/
> > Signed-off-by: Kees Cook <kees@kernel.org>
> > ---
> > Cc: "Mickaël Salaün" <mic@digikod.net>
> > Cc: Thomas Gleixner <tglx@linutronix.de>
> > Cc: Ingo Molnar <mingo@redhat.com>
> > Cc: Borislav Petkov <bp@alien8.de>
> > Cc: Dave Hansen <dave.hansen@linux.intel.com>
> > Cc: <x86@kernel.org>
> > Cc: "H. Peter Anvin" <hpa@zytor.com>
> > Cc: Andy Shevchenko <andy@kernel.org>
> > Cc: Uros Bizjak <ubizjak@gmail.com>
> > Cc: <linux-hardening@vger.kernel.org>
> > ---
> >  arch/x86/include/asm/string_32.h |  6 ------
> >  include/linux/string.h           | 13 +++++++++++++
> >  2 files changed, 13 insertions(+), 6 deletions(-)
> > 
> > diff --git a/arch/x86/include/asm/string_32.h b/arch/x86/include/asm/string_32.h
> > index e9cce169bb4c..74397c95fa37 100644
> > --- a/arch/x86/include/asm/string_32.h
> > +++ b/arch/x86/include/asm/string_32.h
> > @@ -145,12 +145,6 @@ static __always_inline void *__constant_memcpy(void *to, const void *from,
> >  #define __HAVE_ARCH_MEMCPY
> >  extern void *memcpy(void *, const void *, size_t);
> >  
> > -#ifndef CONFIG_FORTIFY_SOURCE
> > -
> > -#define memcpy(t, f, n) __builtin_memcpy(t, f, n)
> > -
> > -#endif /* !CONFIG_FORTIFY_SOURCE */
> > -
> >  #define __HAVE_ARCH_MEMMOVE
> >  void *memmove(void *dest, const void *src, size_t n);
> >  
> > diff --git a/include/linux/string.h b/include/linux/string.h
> > index 01621ad0f598..ffcee31a14f9 100644
> > --- a/include/linux/string.h
> > +++ b/include/linux/string.h
> > @@ -3,6 +3,7 @@
> >  #define _LINUX_STRING_H_
> >  
> >  #include <linux/args.h>
> > +#include <linux/bug.h>
> 
> In case you are go with this change, please keep the headers in order.
> 
> >  #include <linux/array_size.h>
> 
> (should be located here)

Oops, yes, that was my intent but I typoed my insert, it seems. Fixed
now; thanks!

-Kees

-- 
Kees Cook