Uttered GianPiero Puccioni <gip@xxxxxx>, spake thus: > The machines installed are now twelve and on four of them I had to use > the no-hlt instruction. The problem is that two of the "no-hlt"ed ones > after a couple of days stop and I cannot find what happened. It seems to > be repeatable. So, again, what exactly is this "checking 'hlt' > instruction" and what means when installation stops there, is it a > symptom of something nasty in one of the CPU? Could be that what happens > is not connected to the no-hlt (I suppose it could be even if the odds > are quite low). Generally speaking is there anything to do about this? > besides changing CPUs. I believe the kernel uses the "hlt" instruction to reduce power consumption and system bus competition in its idle loop. For some reason, in interrupt doesn't seem to wake up some CPUs properly, so this option was added. Instead of simply hlt'ing with interrupts enabled, the kernel just spins in a tight loop waiting for an interrupt to happen. Try this: 1) Change the line "kernel.sysrq=0" to "kernel.sysrq=1" in the file "/etc/sysctl.conf". 2) As root: # echo 1 > /proc/sys/kernel/sysrq 3) Wait for the system to hang and then type Alt-SysRq-h to see if the system is seeing interrupts at all: no output is bad (interrupts not working), any output is OK (interrupts still work). If you have interrupts working, you may have: A) A system deadlock. Try Alt-SysRq-s and then Alt-Sysrq-b to safely reboot your system. Try disabling some of your modules. B) Broken hardware. C) A buggy BIOS that is incorrectly handling interrupt assignments or power management (try adding "apm=off" and "acpi=off" to the kernel boot arguments). D) Some other problem ;-) Hope this gives you a direction to proceed. Cheers!
Attachment:
pgp3mA9bviRnz.pgp
Description: PGP signature
