public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
* 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32
       [not found] <4b3f02e0.6e69.17dd7a7ddcd.Coremail.canpool@163.com>
@ 2021-12-20 12:06 ` canpool
  2021-12-20 13:32   ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 canpool
  2021-12-20 18:51   ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Adhemerval Zanella
  0 siblings, 2 replies; 5+ messages in thread
From: canpool @ 2021-12-20 12:06 UTC (permalink / raw)
  To: libc-alpha

[-- Attachment #1: Type: text/plain, Size: 2143 bytes --]



branch:master


test:
1)Current
$ trace -r --trace=clock_gettime,clock_gettime64 ./a.out
CLOCK_MONOTONIC errno = Success (0)
CLOCK_MONOTONIC_COARSE errno = Success (0)
CLOCK_REALTIME errno = Success (0)
CLOCK_REALTIME_COARSE errno = Success (0)
    0.000000 clock_gettime(CLOCK_MONOTONIC_RAW, {tv_sec=22820, tv_nsec=304499935}) = 0
CLOCK_MONOTONIC_RAW errno = Success (0)
    0.012450 clock_gettime(CLOCK_BOOTTIME, {tv_sec=22820, tv_nsec=311351295}) = 0
CLOCK_BOOTTIME errno = Success (0)
    0.002625 clock_gettime(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0
    0.002625 clock_gettime64(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0(like this,I forgot the specific data)
CLOCK_PROCESS_CPUTIME_ID errno = Success (0)
    0.002314 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0
    0.002314 clock_gettime64(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0(like this,I forgot the specific data)
CLOCK_THREAD_CPUTIME_ID errno = Success (0)
    0.003358 clock_gettime(CLOCK_TAI, {tv_sec=1262326778, tv_nsec=790176712}) = 0
CLOCK_TAI errno = Success (0)
    0.007705 +++ exited with 0 +++


2)After applying the patch
$ trace -r --trace=clock_gettime,clock_gettime64 ./a.out
CLOCK_MONOTONIC errno = Success (0)
CLOCK_MONOTONIC_COARSE errno = Success (0)
CLOCK_REALTIME errno = Success (0)
CLOCK_REALTIME_COARSE errno = Success (0)
    0.000000 clock_gettime(CLOCK_MONOTONIC_RAW, {tv_sec=22820, tv_nsec=304499935}) = 0
CLOCK_MONOTONIC_RAW errno = Success (0)
    0.012450 clock_gettime(CLOCK_BOOTTIME, {tv_sec=22820, tv_nsec=311351295}) = 0
CLOCK_BOOTTIME errno = Success (0)
    0.002625 clock_gettime(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0
CLOCK_PROCESS_CPUTIME_ID errno = Success (0)
    0.002314 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0
CLOCK_THREAD_CPUTIME_ID errno = Success (0)
    0.003358 clock_gettime(CLOCK_TAI, {tv_sec=1262326778, tv_nsec=790176712}) = 0
CLOCK_TAI errno = Success (0)
    0.007705 +++ exited with 0 +++








| |
canpool
|
|
canpool@163.com
|
签名由网易邮箱大师定制

[-- Attachment #2: test.c --]
[-- Type: text/plain, Size: 1326 bytes --]

#include <time.h>
#include <stdio.h>
#include <errno.h>

//CLOCK_REALTIME
//CLOCK_REALTIME_COARSE
//CLOCK_MONOTONIC
//CLOCK_MONOTONIC_COARSE
//CLOCK_MONOTONIC_RAW
//CLOCK_BOOTTIME
//CLOCK_PROCESS_CPUTIME_ID
//CLOCK_THREAD_CPUTIME_ID
//CLOCK_TAI

int
main (void)
{
  struct timespec ts;
  errno = 0;
  clock_gettime (CLOCK_MONOTONIC, &ts);
  printf ("CLOCK_MONOTONIC errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_MONOTONIC_COARSE, &ts);
  printf ("CLOCK_MONOTONIC_COARSE errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_REALTIME, &ts);
  printf ("CLOCK_REALTIME errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_REALTIME_COARSE, &ts);
  printf ("CLOCK_REALTIME_COARSE errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_MONOTONIC_RAW, &ts);
  printf ("CLOCK_MONOTONIC_RAW errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_BOOTTIME, &ts);
  printf ("CLOCK_BOOTTIME errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_PROCESS_CPUTIME_ID, &ts);
  printf ("CLOCK_PROCESS_CPUTIME_ID errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_THREAD_CPUTIME_ID, &ts);
  printf ("CLOCK_THREAD_CPUTIME_ID errno = %m (%d)\n", errno);
  errno = 0;
  clock_gettime (CLOCK_TAI, &ts);
  printf ("CLOCK_TAI errno = %m (%d)\n", errno);

  return 0;
}


[-- Attachment #3: 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32.patch --]
[-- Type: application/octet-stream, Size: 1402 bytes --]

From 1d66f6b86670549165c9899e75594661a2841ee1 Mon Sep 17 00:00:00 2001
From: canpool <canpool@163.com>
Date: Mon, 20 Dec 2021 19:36:32 +0800
Subject: [PATCH] Linux: Fix 32-bit vDSO for clock_gettime on powerpc32

When the clock_id is CLOCK_PROCESS_CPUTIME_ID or CLOCK_THREAD_CPUTIME_ID,
on the 5.10 kernel powerpc 32-bit, the 32-bit vDSO is executed successfully (
because the __kernel_clock_gettime in arch/powerpc/kernel/vdso32/gettimeofday.S
does not support these two IDs, the 32-bit time_t syscall will be used),
but tp32.tv_sec is equal to 0, causing the 64-bit time_t syscall to continue to be used,
resulting in two system calls.

Fix commit 72e84d1db22203e01a43268de71ea8669eca2863.

Signed-off-by: maminjie <maminjie2@huawei.com>
---
 sysdeps/unix/sysv/linux/clock_gettime.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sysdeps/unix/sysv/linux/clock_gettime.c b/sysdeps/unix/sysv/linux/clock_gettime.c
index 91df6b3d96..9c7d907325 100644
--- a/sysdeps/unix/sysv/linux/clock_gettime.c
+++ b/sysdeps/unix/sysv/linux/clock_gettime.c
@@ -53,7 +53,7 @@ __clock_gettime64 (clockid_t clock_id, struct __timespec64 *tp)
     {
       struct timespec tp32;
       r = INTERNAL_VSYSCALL_CALL (vdso_time, 2, clock_id, &tp32);
-      if (r == 0 && tp32.tv_sec > 0)
+      if (r == 0 && tp32.tv_sec >= 0)
 	{
 	  *tp = valid_timespec_to_timespec64 (tp32);
 	  return 0;
-- 
2.17.1


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32
  2021-12-20 12:06 ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 canpool
@ 2021-12-20 13:32   ` canpool
  2021-12-20 18:51   ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Adhemerval Zanella
  1 sibling, 0 replies; 5+ messages in thread
From: canpool @ 2021-12-20 13:32 UTC (permalink / raw)
  To: libc-alpha

[-- Attachment #1: Type: text/plain, Size: 2311 bytes --]

new patch,change the author information


| |
canpool
|
|
canpool@163.com
|
签名由网易邮箱大师定制
On 12/20/2021 20:06,canpool<canpool@163.com> wrote:


branch:master


test:
1)Current
$ trace -r --trace=clock_gettime,clock_gettime64 ./a.out
CLOCK_MONOTONIC errno = Success (0)
CLOCK_MONOTONIC_COARSE errno = Success (0)
CLOCK_REALTIME errno = Success (0)
CLOCK_REALTIME_COARSE errno = Success (0)
    0.000000 clock_gettime(CLOCK_MONOTONIC_RAW, {tv_sec=22820, tv_nsec=304499935}) = 0
CLOCK_MONOTONIC_RAW errno = Success (0)
    0.012450 clock_gettime(CLOCK_BOOTTIME, {tv_sec=22820, tv_nsec=311351295}) = 0
CLOCK_BOOTTIME errno = Success (0)
    0.002625 clock_gettime(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0
    0.002625 clock_gettime64(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0(like this,I forgot the specific data)
CLOCK_PROCESS_CPUTIME_ID errno = Success (0)
    0.002314 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0
    0.002314 clock_gettime64(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0(like this,I forgot the specific data)
CLOCK_THREAD_CPUTIME_ID errno = Success (0)
    0.003358 clock_gettime(CLOCK_TAI, {tv_sec=1262326778, tv_nsec=790176712}) = 0
CLOCK_TAI errno = Success (0)
    0.007705 +++ exited with 0 +++


2)After applying the patch
$ trace -r --trace=clock_gettime,clock_gettime64 ./a.out
CLOCK_MONOTONIC errno = Success (0)
CLOCK_MONOTONIC_COARSE errno = Success (0)
CLOCK_REALTIME errno = Success (0)
CLOCK_REALTIME_COARSE errno = Success (0)
    0.000000 clock_gettime(CLOCK_MONOTONIC_RAW, {tv_sec=22820, tv_nsec=304499935}) = 0
CLOCK_MONOTONIC_RAW errno = Success (0)
    0.012450 clock_gettime(CLOCK_BOOTTIME, {tv_sec=22820, tv_nsec=311351295}) = 0
CLOCK_BOOTTIME errno = Success (0)
    0.002625 clock_gettime(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0
CLOCK_PROCESS_CPUTIME_ID errno = Success (0)
    0.002314 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0
CLOCK_THREAD_CPUTIME_ID errno = Success (0)
    0.003358 clock_gettime(CLOCK_TAI, {tv_sec=1262326778, tv_nsec=790176712}) = 0
CLOCK_TAI errno = Success (0)
    0.007705 +++ exited with 0 +++








| |
canpool
|
|
canpool@163.com
|
签名由网易邮箱大师定制

[-- Attachment #2: 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32.patch --]
[-- Type: application/octet-stream, Size: 1409 bytes --]

From 1d66f6b86670549165c9899e75594661a2841ee1 Mon Sep 17 00:00:00 2001
From: maminjie <maminjie2@huawei.com> 
Date: Mon, 20 Dec 2021 19:36:32 +0800
Subject: [PATCH] Linux: Fix 32-bit vDSO for clock_gettime on powerpc32

When the clock_id is CLOCK_PROCESS_CPUTIME_ID or CLOCK_THREAD_CPUTIME_ID,
on the 5.10 kernel powerpc 32-bit, the 32-bit vDSO is executed successfully (
because the __kernel_clock_gettime in arch/powerpc/kernel/vdso32/gettimeofday.S
does not support these two IDs, the 32-bit time_t syscall will be used),
but tp32.tv_sec is equal to 0, causing the 64-bit time_t syscall to continue to be used,
resulting in two system calls.

Fix commit 72e84d1db22203e01a43268de71ea8669eca2863.

Signed-off-by: maminjie <maminjie2@huawei.com>
---
 sysdeps/unix/sysv/linux/clock_gettime.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sysdeps/unix/sysv/linux/clock_gettime.c b/sysdeps/unix/sysv/linux/clock_gettime.c
index 91df6b3d96..9c7d907325 100644
--- a/sysdeps/unix/sysv/linux/clock_gettime.c
+++ b/sysdeps/unix/sysv/linux/clock_gettime.c
@@ -53,7 +53,7 @@ __clock_gettime64 (clockid_t clock_id, struct __timespec64 *tp)
     {
       struct timespec tp32;
       r = INTERNAL_VSYSCALL_CALL (vdso_time, 2, clock_id, &tp32);
-      if (r == 0 && tp32.tv_sec > 0)
+      if (r == 0 && tp32.tv_sec >= 0)
 	{
 	  *tp = valid_timespec_to_timespec64 (tp32);
 	  return 0;
-- 
2.17.1


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32
  2021-12-20 12:06 ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 canpool
  2021-12-20 13:32   ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 canpool
@ 2021-12-20 18:51   ` Adhemerval Zanella
  2021-12-20 20:01     ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Florian Weimer
  1 sibling, 1 reply; 5+ messages in thread
From: Adhemerval Zanella @ 2021-12-20 18:51 UTC (permalink / raw)
  To: canpool, libc-alpha



On 20/12/2021 09:06, canpool via Libc-alpha wrote:
> 
> 
> branch:master
> 
> 
> test:
> 1)Current
> $ trace -r --trace=clock_gettime,clock_gettime64 ./a.out
> CLOCK_MONOTONIC errno = Success (0)
> CLOCK_MONOTONIC_COARSE errno = Success (0)
> CLOCK_REALTIME errno = Success (0)
> CLOCK_REALTIME_COARSE errno = Success (0)
>     0.000000 clock_gettime(CLOCK_MONOTONIC_RAW, {tv_sec=22820, tv_nsec=304499935}) = 0
> CLOCK_MONOTONIC_RAW errno = Success (0)
>     0.012450 clock_gettime(CLOCK_BOOTTIME, {tv_sec=22820, tv_nsec=311351295}) = 0
> CLOCK_BOOTTIME errno = Success (0)
>     0.002625 clock_gettime(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0
>     0.002625 clock_gettime64(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0(like this,I forgot the specific data)
> CLOCK_PROCESS_CPUTIME_ID errno = Success (0)
>     0.002314 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0
>     0.002314 clock_gettime64(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0(like this,I forgot the specific data)
> CLOCK_THREAD_CPUTIME_ID errno = Success (0)
>     0.003358 clock_gettime(CLOCK_TAI, {tv_sec=1262326778, tv_nsec=790176712}) = 0
> CLOCK_TAI errno = Success (0)
>     0.007705 +++ exited with 0 +++
> 
> 
> 2)After applying the patch
> $ trace -r --trace=clock_gettime,clock_gettime64 ./a.out
> CLOCK_MONOTONIC errno = Success (0)
> CLOCK_MONOTONIC_COARSE errno = Success (0)
> CLOCK_REALTIME errno = Success (0)
> CLOCK_REALTIME_COARSE errno = Success (0)
>     0.000000 clock_gettime(CLOCK_MONOTONIC_RAW, {tv_sec=22820, tv_nsec=304499935}) = 0
> CLOCK_MONOTONIC_RAW errno = Success (0)
>     0.012450 clock_gettime(CLOCK_BOOTTIME, {tv_sec=22820, tv_nsec=311351295}) = 0
> CLOCK_BOOTTIME errno = Success (0)
>     0.002625 clock_gettime(CLOCK_PROCESS_CPUTIME_ID, {tv_sec=0, tv_nsec=57978964}) = 0
> CLOCK_PROCESS_CPUTIME_ID errno = Success (0)
>     0.002314 clock_gettime(CLOCK_THREAD_CPUTIME_ID, {tv_sec=0, tv_nsec=58799688}) = 0
> CLOCK_THREAD_CPUTIME_ID errno = Success (0)
>     0.003358 clock_gettime(CLOCK_TAI, {tv_sec=1262326778, tv_nsec=790176712}) = 0
> CLOCK_TAI errno = Success (0)
>     0.007705 +++ exited with 0 +++
> 
> From 1d66f6b86670549165c9899e75594661a2841ee1 Mon Sep 17 00:00:00 2001
> From: canpool <canpool@163.com>
> Date: Mon, 20 Dec 2021 19:36:32 +0800
> Subject: [PATCH] Linux: Fix 32-bit vDSO for clock_gettime on powerpc32
> 
> When the clock_id is CLOCK_PROCESS_CPUTIME_ID or CLOCK_THREAD_CPUTIME_ID,
> on the 5.10 kernel powerpc 32-bit, the 32-bit vDSO is executed successfully (
> because the __kernel_clock_gettime in arch/powerpc/kernel/vdso32/gettimeofday.S
> does not support these two IDs, the 32-bit time_t syscall will be used),
> but tp32.tv_sec is equal to 0, causing the 64-bit time_t syscall to continue to be used,
> resulting in two system calls.
> 
> Fix commit 72e84d1db22203e01a43268de71ea8669eca2863.
> 
> Signed-off-by: maminjie <maminjie2@huawei.com>

LGTM, thanks.  However I am confused who is the author here, since 'canpool <canpool@163.com>'
is different than the signed-off-by and I think we can't really use alias as authors.

> ---
>  sysdeps/unix/sysv/linux/clock_gettime.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/sysdeps/unix/sysv/linux/clock_gettime.c b/sysdeps/unix/sysv/linux/clock_gettime.c
> index 91df6b3d96..9c7d907325 100644
> --- a/sysdeps/unix/sysv/linux/clock_gettime.c
> +++ b/sysdeps/unix/sysv/linux/clock_gettime.c
> @@ -53,7 +53,7 @@ __clock_gettime64 (clockid_t clock_id, struct __timespec64 *tp)
>      {
>        struct timespec tp32;
>        r = INTERNAL_VSYSCALL_CALL (vdso_time, 2, clock_id, &tp32);
> -      if (r == 0 && tp32.tv_sec > 0)
> +      if (r == 0 && tp32.tv_sec >= 0)
>  	{
>  	  *tp = valid_timespec_to_timespec64 (tp32);
>  	  return 0;
> -- 
> 2.17.1





^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32
  2021-12-20 18:51   ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Adhemerval Zanella
@ 2021-12-20 20:01     ` Florian Weimer
  2021-12-20 20:03       ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Adhemerval Zanella
  0 siblings, 1 reply; 5+ messages in thread
From: Florian Weimer @ 2021-12-20 20:01 UTC (permalink / raw)
  To: Adhemerval Zanella via Libc-alpha; +Cc: canpool, Adhemerval Zanella

* Adhemerval Zanella via Libc-alpha:

> LGTM, thanks.  However I am confused who is the author here, since
> 'canpool <canpool@163.com>' is different than the signed-off-by and I
> think we can't really use alias as authors.

The FSF accepts contributions under aliases.  Given that the change is
so small, I think we can attribute it to the submitter in any way they
want.

Thanks,
Florian


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32
  2021-12-20 20:01     ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Florian Weimer
@ 2021-12-20 20:03       ` Adhemerval Zanella
  0 siblings, 0 replies; 5+ messages in thread
From: Adhemerval Zanella @ 2021-12-20 20:03 UTC (permalink / raw)
  To: Florian Weimer, Adhemerval Zanella via Libc-alpha; +Cc: canpool



On 20/12/2021 17:01, Florian Weimer wrote:
> * Adhemerval Zanella via Libc-alpha:
> 
>> LGTM, thanks.  However I am confused who is the author here, since
>> 'canpool <canpool@163.com>' is different than the signed-off-by and I
>> think we can't really use alias as authors.
> 
> The FSF accepts contributions under aliases.  Given that the change is
> so small, I think we can attribute it to the submitter in any way they
> want.

Alright, lgtm then.

Reviewed-by: Adhemerval Zanella  <adhemerval.zanella@linaro.org>

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2021-12-20 20:03 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <4b3f02e0.6e69.17dd7a7ddcd.Coremail.canpool@163.com>
2021-12-20 12:06 ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 canpool
2021-12-20 13:32   ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 canpool
2021-12-20 18:51   ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Adhemerval Zanella
2021-12-20 20:01     ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Florian Weimer
2021-12-20 20:03       ` 0001-Linux-Fix-32-bit-vDSO-for-clock_gettime-on-powerpc32 Adhemerval Zanella

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 read-only IMAP folder(s) and NNTP newsgroup(s).