Re: [RFC PATCH v1] powerpc: Enable KFENCE for PPC32

[Marco Elver]

On Thu, 4 Mar 2021 at 15:08, Christophe Leroy
<christophe.leroy@xxxxxxxxxx> wrote:
>
>
>
> Le 04/03/2021 à 13:48, Marco Elver a écrit :
> >  From d118080eb9552073f5dcf1f86198f3d86d5ea850 Mon Sep 17 00:00:00 2001
> > From: Marco Elver <elver@xxxxxxxxxx>
> > Date: Thu, 4 Mar 2021 13:15:51 +0100
> > Subject: [PATCH] kfence: fix reports if constant function prefixes exist
> >
> > Some architectures prefix all functions with a constant string ('.' on
> > ppc64). Add ARCH_FUNC_PREFIX, which may optionally be defined in
> > <asm/kfence.h>, so that get_stack_skipnr() can work properly.
>
>
> It works, thanks.
>
> >
> > Link: https://lkml.kernel.org/r/f036c53d-7e81-763c-47f4-6024c6c5f058@xxxxxxxxxx
> > Reported-by: Christophe Leroy <christophe.leroy@xxxxxxxxxx>
> > Signed-off-by: Marco Elver <elver@xxxxxxxxxx>
>
> Tested-by: Christophe Leroy <christophe.leroy@xxxxxxxxxx>

Thanks, I'll send this to Andrew for inclusion in -mm, since this is
not a strict dependency (it'll work without the patch, just the stack
traces aren't that pretty but still useful). If the ppc patches and
this make it into the next merge window, everything should be good for
5.13.

> > ---
> >   mm/kfence/report.c | 18 ++++++++++++------
> >   1 file changed, 12 insertions(+), 6 deletions(-)
> >
> > diff --git a/mm/kfence/report.c b/mm/kfence/report.c
> > index 519f037720f5..e3f71451ad9e 100644
> > --- a/mm/kfence/report.c
> > +++ b/mm/kfence/report.c
> > @@ -20,6 +20,11 @@
> >
> >   #include "kfence.h"
> >
> > +/* May be overridden by <asm/kfence.h>. */
> > +#ifndef ARCH_FUNC_PREFIX
> > +#define ARCH_FUNC_PREFIX ""
> > +#endif
> > +
> >   extern bool no_hash_pointers;
> >
> >   /* Helper function to either print to a seq_file or to console. */
> > @@ -67,8 +72,9 @@ static int get_stack_skipnr(const unsigned long stack_entries[], int num_entries
> >       for (skipnr = 0; skipnr < num_entries; skipnr++) {
> >               int len = scnprintf(buf, sizeof(buf), "%ps", (void *)stack_entries[skipnr]);
> >
> > -             if (str_has_prefix(buf, "kfence_") || str_has_prefix(buf, "__kfence_") ||
> > -                 !strncmp(buf, "__slab_free", len)) {
> > +             if (str_has_prefix(buf, ARCH_FUNC_PREFIX "kfence_") ||
> > +                 str_has_prefix(buf, ARCH_FUNC_PREFIX "__kfence_") ||
> > +                 !strncmp(buf, ARCH_FUNC_PREFIX "__slab_free", len)) {
> >                       /*
> >                        * In case of tail calls from any of the below
> >                        * to any of the above.
> > @@ -77,10 +83,10 @@ static int get_stack_skipnr(const unsigned long stack_entries[], int num_entries
> >               }
> >
> >               /* Also the *_bulk() variants by only checking prefixes. */
> > -             if (str_has_prefix(buf, "kfree") ||
> > -                 str_has_prefix(buf, "kmem_cache_free") ||
> > -                 str_has_prefix(buf, "__kmalloc") ||
> > -                 str_has_prefix(buf, "kmem_cache_alloc"))
> > +             if (str_has_prefix(buf, ARCH_FUNC_PREFIX "kfree") ||
> > +                 str_has_prefix(buf, ARCH_FUNC_PREFIX "kmem_cache_free") ||
> > +                 str_has_prefix(buf, ARCH_FUNC_PREFIX "__kmalloc") ||
> > +                 str_has_prefix(buf, ARCH_FUNC_PREFIX "kmem_cache_alloc"))
> >                       goto found;
> >       }
> >       if (fallback < num_entries)
> >