[PATCH] x86/signals: Fix lower/upper bound reporting in compat siginfo

[PATCH] x86/signals: Fix lower/upper bound reporting in compat siginfo

[Joerg Roedel]

From: Joerg Roedel <jroedel@suse.de>

Put the right values from the original siginfo into the
userspace compat-siginfo.

This fixes the 32-bit mpx tabletest on a 64-bit kernel.

Fixes: a4455082dc6f0 ('x86/signals: Add missing signal_compat code for x86 features')
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
---
 arch/x86/kernel/signal_compat.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/signal_compat.c b/arch/x86/kernel/signal_compat.c
index ec1f756..71beb28 100644
--- a/arch/x86/kernel/signal_compat.c
+++ b/arch/x86/kernel/signal_compat.c
@@ -151,8 +151,8 @@ int __copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from,
 
 				if (from->si_signo == SIGSEGV) {
 					if (from->si_code == SEGV_BNDERR) {
-						compat_uptr_t lower = (unsigned long)&to->si_lower;
-						compat_uptr_t upper = (unsigned long)&to->si_upper;
+						compat_uptr_t lower = (unsigned long)from->si_lower;
+						compat_uptr_t upper = (unsigned long)from->si_upper;
 						put_user_ex(lower, &to->si_lower);
 						put_user_ex(upper, &to->si_upper);
 					}
-- 
1.9.1

Re: [PATCH] x86/signals: Fix lower/upper bound reporting in compat siginfo

[Dave Hansen]

On 04/04/2017 09:15 AM, Joerg Roedel wrote:
> Put the right values from the original siginfo into the
> userspace compat-siginfo.
> 
> This fixes the 32-bit mpx tabletest on a 64-bit kernel.

Ugh, thanks for finding that.

Acked-by: Dave Hansen <dave.hansen@linux.intel.com>

Re: [PATCH] x86/signals: Fix lower/upper bound reporting in compat siginfo

[Joerg Roedel]

On Tue, Apr 04, 2017 at 09:56:45AM -0700, Dave Hansen wrote:
> On 04/04/2017 09:15 AM, Joerg Roedel wrote:
> > Put the right values from the original siginfo into the
> > userspace compat-siginfo.
> > 
> > This fixes the 32-bit mpx tabletest on a 64-bit kernel.
> 
> Ugh, thanks for finding that.

Yeah, I was also looking at the vaddrexhaust test, but is that supposed
to work? It does pretty weird things at least, on 64 bit the function
effectivly does nothing because nr_to_fill evaluates to 0 there, and on
32 bit it gets a (probably #BF) exception which is not caught, so that
the test fails.

I am a bit confused, but maybe I just didn't stare long enough at it :-)
Hope you can sched some light on it.


	Joerg

[tip:x86/urgent] x86/signals: Fix lower/upper bound reporting in compat siginfo

[tip-bot for Joerg Roedel]

Commit-ID:  cfac6dfa42bddfa9711b20d486e521d1a41ab09f
Gitweb:     http://git.kernel.org/tip/cfac6dfa42bddfa9711b20d486e521d1a41ab09f
Author:     Joerg Roedel <jroedel@suse.de>
AuthorDate: Tue, 4 Apr 2017 18:15:01 +0200
Committer:  Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 5 Apr 2017 10:16:43 +0200

x86/signals: Fix lower/upper bound reporting in compat siginfo

Put the right values from the original siginfo into the
userspace compat-siginfo.

This fixes the 32-bit MPX "tabletest" testcase on 64-bit kernels.

Signed-off-by: Joerg Roedel <jroedel@suse.de>
Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: <stable@vger.kernel.org> # v4.8+
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Borislav Petkov <bp@suse.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Dmitry Safonov <0x7f454c46@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: a4455082dc6f0 ('x86/signals: Add missing signal_compat code for x86 features')
Link: http://lkml.kernel.org/r/1491322501-5054-1-git-send-email-joro@8bytes.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
 arch/x86/kernel/signal_compat.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/signal_compat.c b/arch/x86/kernel/signal_compat.c
index ec1f756..71beb28 100644
--- a/arch/x86/kernel/signal_compat.c
+++ b/arch/x86/kernel/signal_compat.c
@@ -151,8 +151,8 @@ int __copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from,
 
 				if (from->si_signo == SIGSEGV) {
 					if (from->si_code == SEGV_BNDERR) {
-						compat_uptr_t lower = (unsigned long)&to->si_lower;
-						compat_uptr_t upper = (unsigned long)&to->si_upper;
+						compat_uptr_t lower = (unsigned long)from->si_lower;
+						compat_uptr_t upper = (unsigned long)from->si_upper;
 						put_user_ex(lower, &to->si_lower);
 						put_user_ex(upper, &to->si_upper);
 					}

Re: [PATCH] x86/signals: Fix lower/upper bound reporting in compat siginfo

[Dave Hansen]

On 04/04/2017 02:53 PM, Joerg Roedel wrote:
> On Tue, Apr 04, 2017 at 09:56:45AM -0700, Dave Hansen wrote:
>> On 04/04/2017 09:15 AM, Joerg Roedel wrote:
>>> Put the right values from the original siginfo into the
>>> userspace compat-siginfo.
>>>
>>> This fixes the 32-bit mpx tabletest on a 64-bit kernel.
>>
>> Ugh, thanks for finding that.
> 
> Yeah, I was also looking at the vaddrexhaust test, but is that supposed
> to work? It does pretty weird things at least, on 64 bit the function
> effectivly does nothing because nr_to_fill evaluates to 0 there, and on
> 32 bit it gets a (probably #BF) exception which is not caught, so that
> the test fails.
> 
> I am a bit confused, but maybe I just didn't stare long enough at it :-)
> Hope you can sched some light on it.

That code can probably just get removed from the tests.  I haven't run
it in a while and it's probably not worth refreshing.