[PerfSuite-users] Fwd: [Ptools-perfapi] potential kernel crashes with perf_event substrate

[Rick K. <rk...@nc...>]

Greetings all,

Normally I would not forward messages from another mailing list to this one, but I think the below posting from the PAPI mailing list is important enough to go ahead and forward as it concerns an issue that affected us (and the PAPI team helped sort it out) and may also affect anyone who uses similar Linux distributions as we do.

The PerfSuite test suite triggers "Bug 2", described below, and can render the system useless - unresponsive, and in need of a hard reboot.  We ran into this  with the recent release of Ubuntu 11.04, and found that upgrading to a mainline kernel >= 2.6.38-4 was required to set things right again.

Rick

----- Forwarded Message -----
From: "Vince Weaver" <vwe...@ee...>
To: pto...@cs..., per...@cs...
Sent: Thursday, May 12, 2011 4:32:52 PM
Subject: [Ptools-perfapi] potential kernel crashes with perf_event substrate

Hello

Recent Linux kernels have bugs with the perf_event subsystem which can be 
triggered by PAPI.  These bugs can lead to kernel panics and system crashes.  
The bugs have been fixed upstream but the fixes are only gradually 
getting into vendor kernels.  If your code causes a crash, you 
should upgrade to a kernel with the fixes applied.


Bug 1 involves "inherit" support.  perf_event inherit support was only 
added to PAPI as of the recent 4.1.3 release.  This bug causes a memory
leak in the kernel that will eventually out-of-memory your machine with
unfreeable memory.  It is triggered by using the perf_event substrate
to measure at least two events with the "inherit" option set, then running
a multithreaded workload.

The fix for this will be in the forthcoming 2.6.39 Linux kernel, and the 
patch has been included in the relevant stable series trees (2.6.38.2, 
etc.).  The changeset for the fix is: 
   38b435b16c36b0d863efcf3f07b34a6fac9873fd


Bug 2 involves multiplexing support (PAPI will only use kernel 
multiplexing on kernels 2.6.33 and newer).  If you use multiplexing
on multithreaded workloads you can eventually see a kernel panic.

The fix for this will be in the forthcoming 2.6.39 Linux kernel, and
the patch has been included in the relevant stable trees (2.6.38.4, etc.).
The changeset for the fix is:
   ab711fe08297de1485fff0a366e6db8828cafd6a



Vince
vwe...@ee...
_______________________________________________
Ptools-perfapi mailing list
Pto...@ee...
http://lists.eecs.utk.edu/mailman/listinfo/ptools-perfapi