Re: SMP broken on Dell PowerEdge 4100/200 under 2.6.0-testxx?

From: William Park
Date: Sat Dec 06 2003 - 12:39:45 EST


On Sat, Dec 06, 2003 at 10:59:01AM +0000, William Lee Irwin III wrote:
> This leads to a similar conclusion to Stian Jordet's case. It's not
> mistaking you for HT, it's the lack of an internal distinction between
> the cases that need and don't need irq balancing.

I have VP6 (Apollo Pro 133A, 82c694X/686B) dual-P3 (800MHz/133MHz). I'm
currently using MPS 1.1, because USB doesn't work with MPS 1.4 (or, it
does but only with 'noapic'). And, I can report the same finding as
others.

Before:
-------
CPU0 CPU1
0: 79365 63 IO-APIC-edge timer
1: 215 1 IO-APIC-edge i8042
2: 0 0 XT-PIC cascade
4: 8654 0 IO-APIC-edge serial
8: 1 1 IO-APIC-edge rtc
12: 52 1 IO-APIC-edge i8042
14: 2155 0 IO-APIC-edge ide0
15: 2 0 IO-APIC-edge ide1
17: 0 0 IO-APIC-level eth0
NMI: 0 0
LOC: 79231 79266
ERR: 0
MIS: 0

After 'noirqbalance':
---------------------
CPU0 CPU1
0: 15039 16025 IO-APIC-edge timer
1: 47 75 IO-APIC-edge i8042
2: 0 0 XT-PIC cascade
4: 21 43 IO-APIC-edge serial
8: 2 0 IO-APIC-edge rtc
12: 21 32 IO-APIC-edge i8042
14: 828 410 IO-APIC-edge ide0
15: 2 0 IO-APIC-edge ide1
17: 0 0 IO-APIC-level eth0
NMI: 0 0
LOC: 30865 30900
ERR: 0
MIS: 0

--
William Park, Open Geometry Consulting, <opengeometry@xxxxxxxx>
Linux solution for data management and processing.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/