public inbox for libc-hacker@sourceware.org
 help / color / mirror / Atom feed
* tst-getcpu failure
@ 2007-05-12  7:42 Andreas Jaeger
  2007-05-12 16:22 ` Ulrich Drepper
  0 siblings, 1 reply; 6+ messages in thread
From: Andreas Jaeger @ 2007-05-12  7:42 UTC (permalink / raw)
  To: GNU libc hackers

[-- Attachment #1: Type: text/plain, Size: 969 bytes --]


tst-getcpu fails for me on x86-64:
getcpu results 14 not possible

This is one a one CPU system running Linux 2.6.18.

Btw. the following patch fixes a warning about a missing getpid
prototype.  Ok to commit?

Andreas

2007-05-12  Andreas Jaeger  <aj@suse.de>

	* sysdeps/unix/sysv/linux/tst-getcpu.c: Include <unistd.h> for
	getpid.

============================================================
Index: sysdeps/unix/sysv/linux/tst-getcpu.c
--- sysdeps/unix/sysv/linux/tst-getcpu.c	11 May 2007 06:37:33 -0000	1.1
+++ sysdeps/unix/sysv/linux/tst-getcpu.c	12 May 2007 07:38:47 -0000
@@ -1,6 +1,7 @@
 #include <errno.h>
 #include <stdio.h>
 #include <sched.h>
+#include <unistd.h>
 
 
 static int

-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj/
  SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
   Maxfeldstr. 5, 90409 Nürnberg, Germany
    GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: tst-getcpu failure
  2007-05-12  7:42 tst-getcpu failure Andreas Jaeger
@ 2007-05-12 16:22 ` Ulrich Drepper
  2007-05-12 21:05   ` Andreas Jaeger
  0 siblings, 1 reply; 6+ messages in thread
From: Ulrich Drepper @ 2007-05-12 16:22 UTC (permalink / raw)
  To: Andreas Jaeger; +Cc: GNU libc hackers

Andreas Jaeger wrote:
> tst-getcpu fails for me on x86-64:
> getcpu results 14 not possible

AMD or Intel machine?



> Btw. the following patch fixes a warning about a missing getpid
> prototype.  Ok to commit?

Yes.

-- 
➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: tst-getcpu failure
  2007-05-12 16:22 ` Ulrich Drepper
@ 2007-05-12 21:05   ` Andreas Jaeger
  2007-05-12 21:09     ` Ulrich Drepper
  0 siblings, 1 reply; 6+ messages in thread
From: Andreas Jaeger @ 2007-05-12 21:05 UTC (permalink / raw)
  To: Ulrich Drepper; +Cc: GNU libc hackers

[-- Attachment #1: Type: text/plain, Size: 765 bytes --]

Ulrich Drepper <drepper@redhat.com> writes:

> Andreas Jaeger wrote:
>> tst-getcpu fails for me on x86-64:
>> getcpu results 14 not possible
>
> AMD or Intel machine?

AMD64.

My  4 CPU AMD64 machine gives:
getcpu results 14 not possible
getcpu results 0 not possible
getcpu results 0 not possible
getcpu results 0 not possible

both running 2.6.18.

>
>
>
>> Btw. the following patch fixes a warning about a missing getpid
>> prototype.  Ok to commit?
>
> Yes.

Thanks, will do now,
Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj/
  SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
   Maxfeldstr. 5, 90409 Nürnberg, Germany
    GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: tst-getcpu failure
  2007-05-12 21:05   ` Andreas Jaeger
@ 2007-05-12 21:09     ` Ulrich Drepper
  2007-05-13 18:15       ` Andreas Jaeger
  0 siblings, 1 reply; 6+ messages in thread
From: Ulrich Drepper @ 2007-05-12 21:09 UTC (permalink / raw)
  To: Andreas Jaeger; +Cc: GNU libc hackers

Andreas Jaeger wrote:
> My  4 CPU AMD64 machine gives:
> getcpu results 14 not possible
> getcpu results 0 not possible
> getcpu results 0 not possible
> getcpu results 0 not possible

So this is a 4 socket 4 core machine?

-- 
➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: tst-getcpu failure
  2007-05-12 21:09     ` Ulrich Drepper
@ 2007-05-13 18:15       ` Andreas Jaeger
  2007-05-13 18:19         ` Ulrich Drepper
  0 siblings, 1 reply; 6+ messages in thread
From: Andreas Jaeger @ 2007-05-13 18:15 UTC (permalink / raw)
  To: Ulrich Drepper; +Cc: Andreas Jaeger, GNU libc hackers

On Sat, May 12, 2007 at 02:09:19PM -0700, Ulrich Drepper wrote:
> Andreas Jaeger wrote:
> >My  4 CPU AMD64 machine gives:
> >getcpu results 14 not possible
> >getcpu results 0 not possible
> >getcpu results 0 not possible
> >getcpu results 0 not possible
> 
> So this is a 4 socket 4 core machine?

2 sockets with 2 cores each.

cpu4 reports as:
processor       : 3
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 33
model name      : Dual Core AMD Opteron(tm) Processor 875
stepping        : 0
cpu MHz         : 2200.000
cache size      : 1024 KB
physical id     : 1
siblings        : 2
core id         : 1
cpu cores       : 2


Note that my 1 CPU system fails as well - that was the original report,

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj/
  SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
   Maxfeldstr. 5, 90409 Nürnberg, Germany
    GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: tst-getcpu failure
  2007-05-13 18:15       ` Andreas Jaeger
@ 2007-05-13 18:19         ` Ulrich Drepper
  0 siblings, 0 replies; 6+ messages in thread
From: Ulrich Drepper @ 2007-05-13 18:19 UTC (permalink / raw)
  To: Andreas Jaeger; +Cc: GNU libc hackers

Andreas Jaeger wrote:
> 2 sockets with 2 cores each.

Then this...

>>> getcpu results 14 not possible

... is very wrong and should be reported as a kernel bug.


>>> getcpu results 0 not possible
>>> getcpu results 0 not possible
>>> getcpu results 0 not possible

This is probably the rdtscp instruction which doesn't need/use the cache 
and is not reset.

The problem can be ignored for now.  I think the kernel should adjust 
the information when it explicitly reschedules a thread.

-- 
➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2007-05-13 18:19 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-05-12  7:42 tst-getcpu failure Andreas Jaeger
2007-05-12 16:22 ` Ulrich Drepper
2007-05-12 21:05   ` Andreas Jaeger
2007-05-12 21:09     ` Ulrich Drepper
2007-05-13 18:15       ` Andreas Jaeger
2007-05-13 18:19         ` Ulrich Drepper

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).