linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] selftests/powerpc: fix exec benchmark
@ 2018-05-12  3:35 Nicholas Piggin
  2018-05-12  8:54 ` Mathieu Malaterre
  2018-05-16 13:38 ` Michael Ellerman
  0 siblings, 2 replies; 4+ messages in thread
From: Nicholas Piggin @ 2018-05-12  3:35 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin

The exec_target binary could segfault calling _exit(2) because r13
is not set up properly (and libc looks at that when performing a
syscall). Call SYS_exit using syscall(2) which doesn't seem to
have this problem.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 tools/testing/selftests/powerpc/benchmarks/exec_target.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/powerpc/benchmarks/exec_target.c b/tools/testing/selftests/powerpc/benchmarks/exec_target.c
index 3c9c144192be..c14b0fc1edde 100644
--- a/tools/testing/selftests/powerpc/benchmarks/exec_target.c
+++ b/tools/testing/selftests/powerpc/benchmarks/exec_target.c
@@ -6,8 +6,11 @@
  * Copyright 2018, Anton Blanchard, IBM Corp.
  */
 
-void _exit(int);
+#define _GNU_SOURCE
+#include <unistd.h>
+#include <sys/syscall.h>
+
 void _start(void)
 {
-	_exit(0);
+	syscall(SYS_exit, 0);
 }
-- 
2.17.0

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] selftests/powerpc: fix exec benchmark
  2018-05-12  3:35 [PATCH] selftests/powerpc: fix exec benchmark Nicholas Piggin
@ 2018-05-12  8:54 ` Mathieu Malaterre
  2018-05-12 16:35   ` Mathieu Malaterre
  2018-05-16 13:38 ` Michael Ellerman
  1 sibling, 1 reply; 4+ messages in thread
From: Mathieu Malaterre @ 2018-05-12  8:54 UTC (permalink / raw)
  To: Nicholas Piggin; +Cc: linuxppc-dev, Segher Boessenkool

Nick,

On Sat, May 12, 2018 at 5:35 AM, Nicholas Piggin <npiggin@gmail.com> wrote:
> The exec_target binary could segfault calling _exit(2) because r13
> is not set up properly (and libc looks at that when performing a
> syscall). Call SYS_exit using syscall(2) which doesn't seem to
> have this problem.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
>  tools/testing/selftests/powerpc/benchmarks/exec_target.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/tools/testing/selftests/powerpc/benchmarks/exec_target.c b/tools/testing/selftests/powerpc/benchmarks/exec_target.c
> index 3c9c144192be..c14b0fc1edde 100644
> --- a/tools/testing/selftests/powerpc/benchmarks/exec_target.c
> +++ b/tools/testing/selftests/powerpc/benchmarks/exec_target.c
> @@ -6,8 +6,11 @@
>   * Copyright 2018, Anton Blanchard, IBM Corp.
>   */
>
> -void _exit(int);
> +#define _GNU_SOURCE
> +#include <unistd.h>
> +#include <sys/syscall.h>
> +
>  void _start(void)
>  {
> -       _exit(0);
> +       syscall(SYS_exit, 0);
>  }
> --
> 2.17.0
>

Could you please apply the same patch to :

./tools/testing/selftests/size/get_size.c

It segfault on ppc32 with default cross compiler from Debian.

Thanks

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] selftests/powerpc: fix exec benchmark
  2018-05-12  8:54 ` Mathieu Malaterre
@ 2018-05-12 16:35   ` Mathieu Malaterre
  0 siblings, 0 replies; 4+ messages in thread
From: Mathieu Malaterre @ 2018-05-12 16:35 UTC (permalink / raw)
  To: Nicholas Piggin; +Cc: linuxppc-dev, Segher Boessenkool

On Sat, May 12, 2018 at 10:54 AM, Mathieu Malaterre <malat@debian.org> wrote:
> Nick,
>
> On Sat, May 12, 2018 at 5:35 AM, Nicholas Piggin <npiggin@gmail.com> wrote:
>> The exec_target binary could segfault calling _exit(2) because r13
>> is not set up properly (and libc looks at that when performing a
>> syscall). Call SYS_exit using syscall(2) which doesn't seem to
>> have this problem.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>>  tools/testing/selftests/powerpc/benchmarks/exec_target.c | 7 +++++--
>>  1 file changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/testing/selftests/powerpc/benchmarks/exec_target.c b/tools/testing/selftests/powerpc/benchmarks/exec_target.c
>> index 3c9c144192be..c14b0fc1edde 100644
>> --- a/tools/testing/selftests/powerpc/benchmarks/exec_target.c
>> +++ b/tools/testing/selftests/powerpc/benchmarks/exec_target.c
>> @@ -6,8 +6,11 @@
>>   * Copyright 2018, Anton Blanchard, IBM Corp.
>>   */
>>
>> -void _exit(int);
>> +#define _GNU_SOURCE
>> +#include <unistd.h>
>> +#include <sys/syscall.h>
>> +
>>  void _start(void)
>>  {
>> -       _exit(0);
>> +       syscall(SYS_exit, 0);
>>  }
>> --
>> 2.17.0
>>
>
> Could you please apply the same patch to :
>
> ./tools/testing/selftests/size/get_size.c
>
> It segfault on ppc32 with default cross compiler from Debian.

Nevermind, I did not tested and this seems not to be sufficient to
make the executable run properly.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: selftests/powerpc: fix exec benchmark
  2018-05-12  3:35 [PATCH] selftests/powerpc: fix exec benchmark Nicholas Piggin
  2018-05-12  8:54 ` Mathieu Malaterre
@ 2018-05-16 13:38 ` Michael Ellerman
  1 sibling, 0 replies; 4+ messages in thread
From: Michael Ellerman @ 2018-05-16 13:38 UTC (permalink / raw)
  To: Nicholas Piggin, linuxppc-dev; +Cc: Nicholas Piggin

On Sat, 2018-05-12 at 03:35:24 UTC, Nicholas Piggin wrote:
> The exec_target binary could segfault calling _exit(2) because r13
> is not set up properly (and libc looks at that when performing a
> syscall). Call SYS_exit using syscall(2) which doesn't seem to
> have this problem.
> 
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>

Applied to powerpc next, thanks.

https://git.kernel.org/powerpc/c/c906d2c74e0bb9059fbf43e21f8a96

cheers

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2018-05-16 13:38 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-05-12  3:35 [PATCH] selftests/powerpc: fix exec benchmark Nicholas Piggin
2018-05-12  8:54 ` Mathieu Malaterre
2018-05-12 16:35   ` Mathieu Malaterre
2018-05-16 13:38 ` Michael Ellerman

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).