![]() | |||||||||||||
Messages in this thread |
Re: Haakon Riiser >I think I've been able to create a simple test program that >demonstrates the bug I encountered with Qmail. I've compiled and tested the code you posted, i thought that i would add that renicing the LISTENER processes shows some interesting effects. at nice 0 the writer spits out about 40 writes under .01 ms, then 3 to 5 writes at 97+ ms, after about 3 minutes its starts to get more random. at nice -1 the writer spits out about 40 writes under .01 ms, then about 40 writes at 97+ ms, kept this cycle for over 10 minutes. at nice 1 the writer spits out a few hundred writes under .01 ms, and occasionally one at 97+ ms. same effect if elevator=deadline This thread caught my intrest because I run qmail on this box. Peter Maas Pentium II - 400MHz 128MB PC100 kernel-2.6.1-1.126 from http://people.redhat.com/arjanv/2.6/ (not subscribed - reads archives) - 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: 2005-03-22 13:00 [from the cache] ©2003-2008 | |||||||||||||