lkml.org 
[lkml]   [2001]   [Aug]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: Dual 1.7 GHz Xeon -- slow, interrupts not balance, etc
Hi Alan,

Thanks for the response!

On Thu, Aug 16, 2001 at 11:10:18PM +0100, Alan Cox wrote:
> If you look at the figures from the various benchmarking sites that doesnt
> sound unexpected. If you can recode your applications (eg if its a complex
> analysis problem) to use SSE2 and the like you may well get big speed ups

I noticed those web sites, too, which is why I'm confused as to why
I'm achieving such dismal performance.

I don't think we can take advantage of the SSE2 instruction since
we're mostly using these boxes as fast build/compile boxes.

One of the benchmark sites that I browsed through showed a 30 second
difference in kernel 2.4.2 compile times between a 1 GHz PIII a 1.7GHz
Xeon. I realize that such differences are application dependent, but
we only noticed a 10 second difference in our application (ACE/TAO).
Sometimes the PIII box was faster with the compiles, too.

I wasn't expecting a 70% increase in speed but I did expect a
noticeable performance boost rather than a equal or lesser
performance. Was I expecting too much?

> > Any ideas why CPU1 isn't getting any interrupts? Incidentally,
> > interrupts appear to be fairly well balanced on our PIII box.
>
> Do the boot messages ay anothing about this ?

To be honest, I'm not sure what I should be looking for (sorry... I'm
not a kernel hacker yet). Should I just send the boot messages (about
a 15K file)?

I do get the following from 2.4.7 boot (2.4.8 has the same messages):

IO APIC #2......
.... register #00: 02000000
....... : physical APIC id: 02
.... register #01: 00178020
....... : max redirection entries: 0017
....... : IO APIC version: 0020
WARNING: unexpected IO-APIC, please mail
to linux-smp@vger.kernel.org
WARNING: unexpected IO-APIC, please mail
to linux-smp@vger.kernel.org
.... register #02: 00000000
....... : arbitration: 00
.... IRQ redirection table:
NR Log Phy Mask Trig IRR Pol Stat Dest Deli Vect:
00 000 00 1 0 0 0 0 0 0 00
01 003 03 0 0 0 0 0 1 1 39
02 003 03 0 0 0 0 0 1 1 31
03 003 03 0 0 0 0 0 1 1 41
04 003 03 0 0 0 0 0 1 1 49
05 003 03 0 0 0 0 0 1 1 51
06 003 03 0 0 0 0 0 1 1 59
07 003 03 0 0 0 0 0 1 1 61
08 003 03 0 0 0 0 0 1 1 69
09 003 03 0 0 0 0 0 1 1 71
0a 003 03 0 0 0 0 0 1 1 79
0b 003 03 0 0 0 0 0 1 1 81
0c 003 03 0 0 0 0 0 1 1 89
0d 000 00 1 0 0 0 0 0 0 00
0e 003 03 0 0 0 0 0 1 1 91
0f 003 03 0 0 0 0 0 1 1 99
10 003 03 1 1 0 1 0 1 1 A1
11 003 03 1 1 0 1 0 1 1 A9
12 003 03 1 1 0 1 0 1 1 B1
13 003 03 1 1 0 1 0 1 1 B9
14 003 03 1 1 0 1 0 1 1 C1
15 003 03 1 1 0 1 0 1 1 C9
16 003 03 1 1 0 1 0 1 1 D1
17 003 03 1 1 0 1 0 1 1 D9

I'll send this stuff off to linux-smp, as stated by the message. I
also get the following (probably unrelated) messages:

PCI: Probing PCI hardware
Unknown bridge resource 0: assuming transparent
Unknown bridge resource 0: assuming transparent
Unknown bridge resource 2: assuming transparent
Unknown bridge resource 0: assuming transparent
Unknown bridge resource 1: assuming transparent
Unknown bridge resource 2: assuming transparent
Unknown bridge resource 2: assuming transparent

agpgart: Unsupported Intel chipset (device id: 2531), you might want
to try agp_try_unsupported=1.
agpgart: no supported devices found.


BTW, if this mailing list isn't the appropriate forum for these
issues, I'd be glad to move it to another forum. I certainly don't
want to both the community with noise.

Thanks again,
-Ossama
--
Ossama Othman <ossama@ece.uci.edu>
Distributed Object Computing Laboratory, Univ. of California at Irvine
1024D/F7A394A8 - 84ED AA0B 1203 99E4 1068 70E6 5EB7 5E71 F7A3 94A8
-
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 12:57    [W:0.045 / U:0.684 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site