Gentoo Archives: gentoo-amd64

From: Peter Humphrey <prh@××××××××××.uk>
To: gentoo-amd64@l.g.o
Subject: [gentoo-amd64] Identifying CPUs in the kernel
Date: Fri, 22 Jun 2007 09:33:20
Message-Id: 200706221030.26924.prh@gotadsl.co.uk
1 I've been having trouble with the task scheduler in the kernel. I'm running
2 BOINC from a manual installation (because portage builds a useless version,
3 but that's another story) and as I have two CPUs I've told it to use them
4 both. This used to work well on my previous motherboard, now defunct, but
5 it doesn't on this Supermicro H8DCE.
6
7 I'm running gkrellm to show me what's happening in the system, including
8 processor loads. Nice time is shown separately from user and system time,
9 so I can easily see what BOINC's up to.
10
11 This is what happens: when BOINC starts up it starts two processes, which it
12 thinks are going to occupy up to 100% of each processor's time. But both
13 gkrellm and top show both processes running at 50% on CPU1, always that
14 one, with CPU0 idling. Then, if I start an emerge or something, that
15 divides its time more-or-less equally between the two processors with the
16 BOINC processes still confined to CPU1.
17
18 Even more confusingly, sometimes top even disagrees with itself about the
19 processor loadings, the heading lines showing one CPU loaded and the task
20 lines showing the other.
21
22 Just occasionally, BOINC will start its processes properly, each using 100%
23 of a CPU, but after a while it reverts spontaneously to its usual
24 behaviour. I can't find anything in any log to coincide with the reversion.
25
26 I've tried all the versions of BOINC I can find, and I've tried all the
27 available kernels, including vanilla-sources, with no change. I've also
28 tried running folding@home instead of BOINC, and that behaves in the same
29 way. I've talked to the BOINC people, who say they haven't seen this
30 behaviour anywhere else and that it sounds like a problem with my
31 particular configuration of components. I'm trying an installation of
32 Kubuntu to see if that's any different, but it's a long process getting to
33 an equivalent state so I can't report a result yet.
34
35 I'm beginning to think there must be a problem with my motherboard. Can
36 anyone suggest something else for me to check?
37
38 --
39 Rgds
40 Peter Humphrey
41 Linux Counter 5290, Aug 93
42 --
43 gentoo-amd64@g.o mailing list

Replies