From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 4.0) Description of problem: We've seen the kernel hang twice while running stress tests on a two node IBM x455. In both cases a series of "floating-point assist fault" messages had been previously written to the console. It is not clear whether the hang is associated with the floating-point errors. The following messages are from the most recent failure. float_exp_log(17776): floating-point assist fault at ip 200000000008f7b2 float_exp_log(17780): floating-point assist fault at ip 200000000008f7b2 float_exp_log(17769): floating-point assist fault at ip 200000000008f7b2 float_exp_log(17771): floating-point assist fault at ip 200000000008f7b2 float_exp_log(17771): floating-point assist fault at ip 200000000008f7b2 float_exp_log(17781): floating-point assist fault at ip 200000000008f7b2 float_exp_log(17783): floating-point assist fault at ip 200000000008f7b2 float_exp_log(17785): floating-point assist fault at ip 200000000008f7b2 Version-Release number of selected component (if applicable): kernel-2.4.21-18.EL How reproducible: Sometimes Steps to Reproduce: 1. We encountered the problem while subjecting the system to heavy test load. 2. 3. Actual Results: kernel hang Expected Results: no kernel hang Additional info:
hmmm, there isn't much to go on here. Is it possible to get an alt-sysrq-t at the time of the hang?
Gary, are you going to follow up on this at some point?
Ernie, thanks for pinging me on this. I somehow missed Jason's 6 month old request for more information. One reason I neglected to respond was probably due to the lack of subsequent problem occurrences that would have provided reminders. Since the problem was not reproducible this bug can be closed.