Bug 727324 - Able to softlockup kernel-PAE-2.6.35.13-92.fc14.i686 when running with multiple CPUs (smp)
Summary: Able to softlockup kernel-PAE-2.6.35.13-92.fc14.i686 when running with multip...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 14
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2011-08-01 20:02 UTC by David Shaw
Modified: 2011-10-05 19:19 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2011-10-05 19:19:10 UTC
Type: ---

Attachments (Terms of Use)
Program to softlockup the kernel (1.23 KB, application/octet-stream)
2011-08-01 20:02 UTC, David Shaw 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Description David Shaw 2011-08-01 20:02:15 UTC 
Created attachment 516206 [details]
Program to softlockup the kernel

Description of problem:

Running a program that does a lot of fork/waitpids from multiple threads can cause the kernel to softlockup.  It does not recover afterwards.  This seems to only happen if there is multiple CPUs (I tested with four).

Version-Release number of selected component (if applicable):

kernel-PAE-2.6.35.13-92.fc14.i686

How reproducible:

Every time

Steps to Reproduce:
1. Run the attached program
  
Actual results:

If the kernel is set to panic on softlockup (kernel.softlockup_panic=1) then there is a panic.  Otherwise, the system just hits softlockup over and over again.  The system is not usable (not pingable, etc).

Expected results:

Nothing should happen.

Additional info:

I also tried this on kernel-PAE-2.6.35.12-88.fc14.i686.rpm and kernel-PAE-2.6.35.13-91.fc14.i686.rpm and saw the same problem.

However, I could *not* reproduce this using kernel-PAE-2.6.35.6-45.fc14.i686.rpm (the kernel that shipped with F14).
Comment 1 David Shaw 2011-10-05 19:19:10 UTC 
This bug seems to have been resolved (intentionally or not) in kernel-PAE-2.6.35.14-96.fc14.i686
Note You need to log in before you can comment on or make changes to this bug.