Description of problem: BUG: at kernel/softirq.c:639 __tasklet_action() [<c0405fec>] dump_trace+0x5f/0x107 [<c04060ae>] show_trace_log_lvl+0x1a/0x2f [<c04066d0>] show_trace+0x12/0x14 [<c0406763>] dump_stack+0x16/0x18 [<c042c8f9>] __tasklet_action+0xe2/0x12a [<c042cd11>] tasklet_action+0x28/0x2b [<c042d12b>] ksoftirqd+0x142/0x21b [<c0439baf>] kthread+0xb3/0xdc [<c0405c03>] kernel_thread_helper+0x7/0x10 ======================= Version-Release number of selected component (if applicable): 2.6.21-2.el5rt How reproducible: Unknown at this time Steps to Reproduce: 1. Run the /kernel/stress/racer on 2.6.21-2.el5rt Additional info: This is the RHEL5 GA with 2.6.21-2.el5rt kernel
I have also seen this with 2.6.21-6.el5rt running the byte test on groucho. BUG: at kernel/softirq.c:639 __tasklet_action() [<c0405fec>] dump_trace+0x5f/0x107 [<c04060ae>] show_trace_log_lvl+0x1a/0x2f [<c04066d0>] show_trace+0x12/0x14 [<c0406763>] dump_stack+0x16/0x18 [<c042c8e9>] __tasklet_action+0xe2/0x12a [<c042cd08>] tasklet_action+0x28/0x2b [<c042d122>] ksoftirqd+0x142/0x21b [<c0439bb3>] kthread+0xb3/0xdc [<c0405c03>] kernel_thread_helper+0x7/0x10 ======================= BUG: at kernel/softirq.c:613 __tasklet_action() [<c0405fec>] dump_trace+0x5f/0x107 [<c04060ae>] show_trace_log_lvl+0x1a/0x2f [<c04066d0>] show_trace+0x12/0x14 [<c0406763>] dump_stack+0x16/0x18 [<c042c853>] __tasklet_action+0x4c/0x12a [<c042cd08>] tasklet_action+0x28/0x2b [<c042d122>] ksoftirqd+0x142/0x21b [<c0439bb3>] kthread+0xb3/0xdc [<c0405c03>] kernel_thread_helper+0x7/0x10 ======================= BUG: at kernel/softirq.c:639 __tasklet_action() [<c0405fec>] dump_trace+0x5f/0x107 [<c04060ae>] show_trace_log_lvl+0x1a/0x2f [<c04066d0>] show_trace+0x12/0x14 [<c0406763>] dump_stack+0x16/0x18 [<c042c8e9>] __tasklet_action+0xe2/0x12a [<c042cd08>] tasklet_action+0x28/0x2b [<c042d122>] ksoftirqd+0x142/0x21b [<c0439bb3>] kthread+0xb3/0xdc [<c0405c03>] kernel_thread_helper+0x7/0x10 http://rhts.lab.boston.redhat.com/cgi-bin/rhts/test_log.cgi?id=68138
We discussed this one in the weekly internal call. Putting notes from that discussion here: Ingo: - difficult to fix - occasionally sees this message, but it may be harmless noise. jburke said that he didn't see functional test failure, just presense of the BUG in log. - Paul McKinney looked into it but couldn't find the problem. tglx: - suggested to get Olaf Nester (sp?) to look into it - as he's good at fixing races - said he'd followup - so I'm reassigning this one to Thomas
See the following log: http://rhts.lab.boston.redhat.com/cgi-bin/rhts/test_log.cgi?id=77570 Not sure if it is cause and effect here, but shortly after seeing the message print out the system was running a test then went into a oom-kill death spiral. I am ot convinced that one has to do with the other just pointing it out. Jeff
*** Bug 242422 has been marked as a duplicate of this bug. ***
bug #242422 comment #19 is the proposed fix
Verified fixed in 2.6.21-31.el5rt per John Stultz jstultz.com https://bugzilla.redhat.com/show_bug.cgi?id=242422#c31