Bug 740913

Summary:	PAPI testsuite hangs on POWER7 box
Product:	Red Hat Enterprise Linux 6	Reporter:	Petr Muller <pmuller>
Component:	kernel	Assignee:	Steve Best <sbest>
Status:	CLOSED WONTFIX	QA Contact:	Red Hat Kernel QE team <kernel-qe>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	6.2	CC:	arozansk, mnowak, ohudlick
Target Milestone:	rc
Target Release:	---
Hardware:	ppc64
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2012-11-30 17:29:25 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Bug Depends On:
Bug Blocks:	767187

Description Petr Muller 2011-09-23 18:37:55 UTC

Description of problem:
On POWER7 box I observe some testcases do not terminate even after running for tens of minutes. Those testcases usually spin the CPUs to 100%, and do not respond even to kill -9. 

Version-Release number of selected component (if applicable):
papi-4.1.3-3.el6.ppc64

How reproducible:
seems always

Steps to Reproduce:
1. either run /tools/papi/testsuite from Beaker, or install PAPI srpm and run the testsuite manually
  
Actual results:
see ctests/attach2 running for a long time 

Expected results:
testsuite terminates in tens of minutes

Additional info:
This is not a regression: old PAPI hangs too when it's testsuite is run on POWER7 box.

Comment 2 William Cohen 2011-09-27 16:51:03 UTC

Able to replicate this problem with the upstream papi from the cvs repository.

Comment 3 William Cohen 2011-09-30 01:03:57 UTC

Noticing that other ctests tests also hang such as:

byte_profile
calibrate
ipc

I suspect there is something going wrong in kernel space. One can not kill the hung tests. Also test program does not die when log out and then back in.

I did an strace of various tests and things seem to wrong after getting ioctl for performance monitoring like the something like the following:

ioctl(3, 0x20002400, 0)                 = 0


It doesn't look like perf perf works either on power7. The following also hangs:

perf stat ls

However, maybe this is to previous problems with the papi tests.

Comment 4 RHEL Program Management 2011-10-07 15:50:05 UTC

Since RHEL 6.2 External Beta has begun, and this bug remains
unresolved, it has been rejected as it is not proposed as
exception or blocker.

Red Hat invites you to ask your support representative to
propose this request, if appropriate and relevant, in the
next release of Red Hat Enterprise Linux.

Comment 5 Petr Muller 2011-10-10 13:01:21 UTC

Not a regression, proposing for 6.3

Comment 7 Steve Best 2012-02-17 20:42:54 UTC

Petr,

could you try this on 6.3 kernel 229 or later. I did a perf fix in that kernel and it should fix the hang issue.

-Steve

Comment 8 RHEL Program Management 2012-05-03 04:44:29 UTC

Since RHEL 6.3 External Beta has begun, and this bug remains
unresolved, it has been rejected as it is not proposed as
exception or blocker.

Red Hat invites you to ask your support representative to
propose this request, if appropriate and relevant, in the
next release of Red Hat Enterprise Linux.