Re: [PATCH v2 05/17] vdso: Avoid call to memset() by getrandom

[Segher Boessenkool <segher@kernel.crashing.org>]

On Wed, Aug 28, 2024 at 02:26:08PM +0200, Jason A. Donenfeld wrote:
> On Wed, Aug 28, 2024 at 2:24 PM Arnd Bergmann <arnd@arndb.de> wrote:
> >
> > On Wed, Aug 28, 2024, at 11:18, Jason A. Donenfeld wrote:
> > > On Tue, Aug 27, 2024 at 05:53:30PM -0500, Segher Boessenkool wrote:
> > >> On Tue, Aug 27, 2024 at 11:08:19AM -0700, Eric Biggers wrote:
> > >> >
> > >> > Is there a compiler flag that could be used to disable the generation of calls
> > >> > to memset?
> > >>
> > >> -fno-tree-loop-distribute-patterns .  But, as always, read up on it, see
> > >> what it actually does (and how it avoids your problem, and mostly: learn
> > >> what the actual problem *was*!)
> > >
> > > This might help with various loops, but it doesn't help with the matter
> > > that this patch fixes, which is struct initialization. I just tried it
> > > with the arm64 patch to no avail.
> >
> > Maybe -ffreestanding can help here? That should cause the vdso to be built
> > with the assumption that there is no libc, so it would neither add nor
> > remove standard library calls. Not sure if that causes other problems,
> > e.g. if the calling conventions are different.
> 
> >From https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90701:
> 
> | You need -ffreestanding but that is documented to emit memset and
> memcpy still.

Yeah.

'-nostdlib'
     Do not use the standard system startup files or libraries when
     linking.

This won't help a bit, the compiler will still optimise your
initialisation loop to a memset() call, it just won't link to libgcc.a
and crt*.o and its ilk (which is not where mem* are implemented in the
first place!)


Segher