![]() | ||||||||||
Messages in this thread |
Hi, Andrew Morton wrote: > I suspect I was fooled by the oprofile output, which showed tremendous > amounts of load in schedule() and switch_to(). The percentages which > opreport shows are the percentage of non-halted CPU time. So if you have a > function in the kernel which is using 1% of the total CPU, and the CPU is > halted for 95% of the time, it appears that the function is taking 20% of > CPU! > > The fix for that is to boot with the "idle=poll" boot parameter, to make > the CPU spin when it has nothing else to do. I'll test again the two_discs_bad situation after booting with that parameter. Thanks. > > I'm suspecting that your machine is just stuck in D state waiting for disk. > Did we have a sysrq-T trace? The amazing thing is that this doesn't happen! Every single cron jobs that keeps running (I intentionally said that before too) and never ends is in R state. By strace'ing the processes they just seem to be going *extremely* slow. I also changed the I/O elevator of hdb (the OS disk) to deadline from cfq, unfortunately with no results. That is why I 've been considering it a CPU scheduler issue. Dimitris - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | |||||||||
| Last update: 2007-08-06 18:31 [from the cache] ©2003-2008 | ||||||||||