* [PATCH] objtool: Permit __kasan_check_{read,write} under UACCESS
@ 2020-09-28 22:49 Jann Horn
2020-09-29 7:05 ` Peter Zijlstra
0 siblings, 1 reply; 2+ messages in thread
From: Jann Horn @ 2020-09-28 22:49 UTC (permalink / raw)
To: Josh Poimboeuf, Peter Zijlstra
Cc: linux-kernel, x86, Andrey Ryabinin, Alexander Potapenko,
Dmitry Vyukov, kasan-dev, Dan Williams, Tony Luck, Vishal Verma
Building linux-next with JUMP_LABEL=n and KASAN=y, I got this objtool
warning:
arch/x86/lib/copy_mc.o: warning: objtool: copy_mc_to_user()+0x22: call to
__kasan_check_read() with UACCESS enabled
What happens here is that copy_mc_to_user() branches on a static key in a
UACCESS region:
__uaccess_begin();
if (static_branch_unlikely(©_mc_fragile_key))
ret = copy_mc_fragile(to, from, len);
ret = copy_mc_generic(to, from, len);
__uaccess_end();
and the !CONFIG_JUMP_LABEL version of static_branch_unlikely() uses
static_key_enabled(), which uses static_key_count(), which uses
atomic_read(), which calls instrument_atomic_read(), which uses
kasan_check_read(), which is __kasan_check_read().
Let's permit these KASAN helpers in UACCESS regions - static keys should
probably work under UACCESS, I think.
Signed-off-by: Jann Horn <jannh@google.com>
---
Calling atomic_read() on a global under UACCESS should probably be fine,
right? The alternative to this patch would be to change
copy_mc_to_user()...
Note that copy_mc_to_user() does not exist in the tip tree yet; it
appeared in commit 0a78de3d4b7b1b80e5c1eead24ce11c4b3cc8791 in the
nvdimm tree.
tools/objtool/check.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index a88fb05242d5..1141a8e26c1e 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -583,6 +583,8 @@ static const char *uaccess_safe_builtin[] = {
"__asan_store4_noabort",
"__asan_store8_noabort",
"__asan_store16_noabort",
+ "__kasan_check_read",
+ "__kasan_check_write",
/* KASAN in-line */
"__asan_report_load_n_noabort",
"__asan_report_load1_noabort",
base-commit: 0248dedd12d43035bf53c326633f0610a49d7134
--
2.28.0.709.gb0816b6eb0-goog
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] objtool: Permit __kasan_check_{read,write} under UACCESS
2020-09-28 22:49 [PATCH] objtool: Permit __kasan_check_{read,write} under UACCESS Jann Horn
@ 2020-09-29 7:05 ` Peter Zijlstra
0 siblings, 0 replies; 2+ messages in thread
From: Peter Zijlstra @ 2020-09-29 7:05 UTC (permalink / raw)
To: Jann Horn
Cc: Josh Poimboeuf, linux-kernel, x86, Andrey Ryabinin,
Alexander Potapenko, Dmitry Vyukov, kasan-dev, Dan Williams,
Tony Luck, Vishal Verma
On Tue, Sep 29, 2020 at 12:49:16AM +0200, Jann Horn wrote:
> Building linux-next with JUMP_LABEL=n and KASAN=y, I got this objtool
> warning:
>
> arch/x86/lib/copy_mc.o: warning: objtool: copy_mc_to_user()+0x22: call to
> __kasan_check_read() with UACCESS enabled
>
> What happens here is that copy_mc_to_user() branches on a static key in a
> UACCESS region:
>
> __uaccess_begin();
> if (static_branch_unlikely(©_mc_fragile_key))
> ret = copy_mc_fragile(to, from, len);
> ret = copy_mc_generic(to, from, len);
> __uaccess_end();
>
> and the !CONFIG_JUMP_LABEL version of static_branch_unlikely() uses
> static_key_enabled(), which uses static_key_count(), which uses
> atomic_read(), which calls instrument_atomic_read(), which uses
> kasan_check_read(), which is __kasan_check_read().
>
> Let's permit these KASAN helpers in UACCESS regions - static keys should
> probably work under UACCESS, I think.
It's not a matter of permitting, it's a matter of being safe and
correct. In this case it is, because it's a thin wrapper around
check_memory_region() which was already marked safe.
check_memory_region() is correct because the only thing it ends up
calling is kasa_report() and that is also marked safe because that is
annotated with user_access_save/restore() before it does anything else.
On top of that, all of KASAN is noinstr, so nothing in here will end up
in tracing and/or call schedule() before the user_access_save().
> Signed-off-by: Jann Horn <jannh@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> ---
> Calling atomic_read() on a global under UACCESS should probably be fine,
> right?
Yes, per the above.
> tools/objtool/check.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/tools/objtool/check.c b/tools/objtool/check.c
> index a88fb05242d5..1141a8e26c1e 100644
> --- a/tools/objtool/check.c
> +++ b/tools/objtool/check.c
> @@ -583,6 +583,8 @@ static const char *uaccess_safe_builtin[] = {
> "__asan_store4_noabort",
> "__asan_store8_noabort",
> "__asan_store16_noabort",
> + "__kasan_check_read",
> + "__kasan_check_write",
> /* KASAN in-line */
> "__asan_report_load_n_noabort",
> "__asan_report_load1_noabort",
>
> base-commit: 0248dedd12d43035bf53c326633f0610a49d7134
> --
> 2.28.0.709.gb0816b6eb0-goog
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2020-09-29 7:05 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-09-28 22:49 [PATCH] objtool: Permit __kasan_check_{read,write} under UACCESS Jann Horn
2020-09-29 7:05 ` Peter Zijlstra
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox