Re: [PATCH] x86/fpu/signal: initialize sw_bytes in save_xstate_epilog()

From: Marco Elver
Date: Tue Nov 30 2021 - 18:29:58 EST


On Tue, 30 Nov 2021 at 22:39, Alexander Potapenko <glider@xxxxxxxxxx> wrote:
>
> On Tue, Nov 30, 2021 at 10:09 PM Dave Hansen <dave.hansen@xxxxxxxxx> wrote:
> >
> > On 11/26/21 4:47 AM, Alexander Potapenko wrote:
> > > save_sw_bytes() did not fully initialize sw_bytes, which caused KMSAN
> > > to report an infoleak (see below).
> > > Initialize sw_bytes explicitly to avoid this.
> > ...
> > > Reported-by: Alexander Potapenko <glider@xxxxxxxxxx>
> > > Signed-off-by: Marco Elver <elver@xxxxxxxxxx>
> > > Signed-off-by: Alexander Potapenko <glider@xxxxxxxxxx>
> > > Tested-by: Alexander Potapenko <glider@xxxxxxxxxx>
> > > Fixes: 53599b4d54b9b8dd ("x86/fpu/signal: Prepare for variable sigframe length")
> > > Link: https://lore.kernel.org/all/CAG_fn=V9T6OKPonSjsi9PmWB0hMHFC=yawozdft8i1-MSxrv=w@xxxxxxxxxxxxxx/
> >
> > Hi Alexander,
> >
> > Marco's SoB entry is before yours. Was this authored by you or Marco?
> > If it was Marco, it's customary to add a:
> >
> > From: Marco Elver <elver@xxxxxxxxxx>
> >
> > at the top of the changelog to make sure git gets the author right. I'm
> > happy to fix it up this time, I just need to know who wrote it.
>
> Hi Dave,
>
> Yes, it was authored by Marco. Thanks in advance for fixing this, I'll
> keep that in mind next time :)

I produced a diff, but Alex tested and wrote the commit message.

I'm happy to have Alex be the primary author here, he spent more time
on this. :-)
If you want it to be more precise, you could add a:

Co-developed-by: Marco Elver <elver@xxxxxxxxxx>

Thanks,
-- Marco