Those definitely look bad. Do you have a way to get a stack trace for one of these traps? --Brian
Rodent of Unusual Size wrote: >These showed up in my kernel log from a build of HEAD >on RH Linux 6.2 on Alpha, using GCC egcs-2.91.66. (This >is the first build I've done on this platform for a >long time, so I don't think this is a new issue.) > >Dec 13 17:10:03 kernel: httpd(21879): unaligned trap at 00000200012dcd00: >000003ffff9fdadc 2d 0 >Dec 13 17:10:05 kernel: httpd(21922): unaligned trap at 00000200012dcd00: >000003ffff9fdadc 2d 0 >Dec 13 17:10:07 kernel: httpd(21927): unaligned trap at 00000200012dcd00: >000003ffff9fdadc 2d 0 >Dec 13 17:10:09 kernel: httpd(21932): unaligned trap at 00000200012dcd00: >000003ffff9fdadc 2d 0 >Dec 13 17:23:53 kernel: httpd(24031): unaligned trap at 00000200012dcd00: >000000011fffd44c 2d 0 >Dec 13 17:23:55 kernel: httpd(24152): unaligned trap at 00000200012dcd00: >000000011fffd3bc 2d 0 >Dec 13 17:23:57 kernel: httpd(24153): unaligned trap at 00000200012dcd00: >000000011fffd3bc 2d 0 >Dec 13 17:23:59 kernel: httpd(24154): unaligned trap at 00000200012dcd00: >000000011fffd3bc 2d 0 > >I'm pretty sure these are wicked expensive in terms of things >like the instruction cache.. Dean, OtherBrian? >
