* Dual Xeon, MTRR problem still there?
@ 2000-11-06 9:40 Dr. David Gilbert
2000-11-06 9:51 ` Tigran Aivazian
0 siblings, 1 reply; 2+ messages in thread
From: Dr. David Gilbert @ 2000-11-06 9:40 UTC (permalink / raw)
To: linux-kernel
Hi,
While my friendly dual Xeon machine now appears to be pretty fast, I
notice that the NMI counter is still incrementing like topsy; this
presumably means that there is still something dodgy going on.
A hdparm -t seems to be giving respectable values, so I am not quite sure
where the time/performance is going:
/proc/mtrr shows the following
reg00: base=0xf9f00000 (3999MB), size= 1MB: uncachable, count=1
reg01: base=0xfa000000 (4000MB), size= 32MB: uncachable, count=1
reg02: base=0xfc000000 (4032MB), size= 64MB: uncachable, count=1
reg03: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
reg04: base=0x100000000 (4096MB), size= 64MB: write-back, count=1
reg05: base=0x104000000 (4160MB), size= 32MB: write-back, count=1
reg06: base=0x106000000 (4192MB), size= 1MB: write-back, count=1
Here are two /proc/interrupts straight after each other:
CPU0 CPU1
0: 14047275 17655700 IO-APIC-edge timer
1: 73 37 IO-APIC-edge keyboard
2: 0 0 XT-PIC cascade
9: 0 0 IO-APIC-edge acpi
12: 1196 928 IO-APIC-edge PS/2 Mouse
13: 0 0 XT-PIC fpu
20: 226485 240634 IO-APIC-level eth0
56: 14 16 IO-APIC-level sym53c8xx
57: 52554 55135 IO-APIC-level sym53c8xx
58: 15 12 IO-APIC-level sym53c8xx
NMI: 31702796 31702796
LOC: 31702834 31702833
ERR: 0
CPU0 CPU1
0: 14047328 17655766 IO-APIC-edge timer
1: 73 37 IO-APIC-edge keyboard
2: 0 0 XT-PIC cascade
9: 0 0 IO-APIC-edge acpi
12: 1196 928 IO-APIC-edge PS/2 Mouse
13: 0 0 XT-PIC fpu
20: 226496 240646 IO-APIC-level eth0
56: 14 16 IO-APIC-level sym53c8xx
57: 52554 55135 IO-APIC-level sym53c8xx
58: 15 12 IO-APIC-level sym53c8xx
NMI: 31702915 31702915
LOC: 31702953 31702952
ERR: 0
Thats 2.4.0-test10 with last weeks patch.
Dave
--
/------------------------------------------------------------------\
| Dr. David Alan Gilbert | Work:dg@px.uk.com +44-161-286-2000 Ex258|
| -------- G7FHJ --------|---------------------------------------- |
| Home: dave@treblig.org http://www.treblig.org |
\------------------------------------------------------------------/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Dual Xeon, MTRR problem still there?
2000-11-06 9:40 Dual Xeon, MTRR problem still there? Dr. David Gilbert
@ 2000-11-06 9:51 ` Tigran Aivazian
0 siblings, 0 replies; 2+ messages in thread
From: Tigran Aivazian @ 2000-11-06 9:51 UTC (permalink / raw)
To: Dr. David Gilbert; +Cc: linux-kernel
No, there is nothing dodgy going on -- NMI counter should be incrementing
fast -- this is normal. See the NMI watchdog routine in the source.
However, you can do even more optimization to your machine by tweaking the
memory timings settings (it will cause data corruption sometimes but I
assume it's not a production machine anyway so it's worth to corrupt data
for satisfaction of curiosity).
Regards,
Tigran
On Mon, 6 Nov 2000, Dr. David Gilbert wrote:
> Hi,
> While my friendly dual Xeon machine now appears to be pretty fast, I
> notice that the NMI counter is still incrementing like topsy; this
> presumably means that there is still something dodgy going on.
> A hdparm -t seems to be giving respectable values, so I am not quite sure
> where the time/performance is going:
>
> /proc/mtrr shows the following
>
> reg00: base=0xf9f00000 (3999MB), size= 1MB: uncachable, count=1
> reg01: base=0xfa000000 (4000MB), size= 32MB: uncachable, count=1
> reg02: base=0xfc000000 (4032MB), size= 64MB: uncachable, count=1
> reg03: base=0x00000000 ( 0MB), size=4096MB: write-back, count=1
> reg04: base=0x100000000 (4096MB), size= 64MB: write-back, count=1
> reg05: base=0x104000000 (4160MB), size= 32MB: write-back, count=1
> reg06: base=0x106000000 (4192MB), size= 1MB: write-back, count=1
>
> Here are two /proc/interrupts straight after each other:
>
> CPU0 CPU1
> 0: 14047275 17655700 IO-APIC-edge timer
> 1: 73 37 IO-APIC-edge keyboard
> 2: 0 0 XT-PIC cascade
> 9: 0 0 IO-APIC-edge acpi
> 12: 1196 928 IO-APIC-edge PS/2 Mouse
> 13: 0 0 XT-PIC fpu
> 20: 226485 240634 IO-APIC-level eth0
> 56: 14 16 IO-APIC-level sym53c8xx
> 57: 52554 55135 IO-APIC-level sym53c8xx
> 58: 15 12 IO-APIC-level sym53c8xx
> NMI: 31702796 31702796
> LOC: 31702834 31702833
> ERR: 0
>
> CPU0 CPU1
> 0: 14047328 17655766 IO-APIC-edge timer
> 1: 73 37 IO-APIC-edge keyboard
> 2: 0 0 XT-PIC cascade
> 9: 0 0 IO-APIC-edge acpi
> 12: 1196 928 IO-APIC-edge PS/2 Mouse
> 13: 0 0 XT-PIC fpu
> 20: 226496 240646 IO-APIC-level eth0
> 56: 14 16 IO-APIC-level sym53c8xx
> 57: 52554 55135 IO-APIC-level sym53c8xx
> 58: 15 12 IO-APIC-level sym53c8xx
> NMI: 31702915 31702915
> LOC: 31702953 31702952
> ERR: 0
>
> Thats 2.4.0-test10 with last weeks patch.
>
> Dave
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2000-11-06 9:50 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2000-11-06 9:40 Dual Xeon, MTRR problem still there? Dr. David Gilbert
2000-11-06 9:51 ` Tigran Aivazian
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).