lkml.org 
[lkml]   [2004]   [Apr]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRE: [PATCH] 2.6.5- es7000 subarch update
Date
From
Hi Len,

>please test it on your ES7000. I'd love to see
>the full dmesg from an ES7000 if you can grow your
>CONFIG_LOG_BUF_SHIFT to capture it all.

The patch worked great, my system came up beautifully. Len, it was so well done, and it was fast, too - thanks! I forgot to increase this parameter, but still collected a full trace with some extra IO-APIC snapshots. For now, I am attaching a serial console trace, and later will provide you full dmesg.

>I expect that if the Linux IRQ convention was working
>on the ES7000 before, this patch will maintain
>compatibility with that. However, if things were
>not working as expected before, I just wanted to
>point out that another convention is possible --
>particularly with a sub-architecture.

Yes, the IRQ schema you described was used on the ES7000 previously on 2.6. For example, the way interrupts looked with the old interrupt code (with irq_balance off, and my tweak for the IDE):
0: 1566205 0 0 0 IO-APIC-edge timer
1: 12 0 0 0 IO-APIC-edge i8042
2: 0 0 0 0 XT-PIC cascade
4: 20 0 0 0 IO-APIC-edge serial
8: 3 0 0 0 IO-APIC-edge rtc
12: 399 0 0 0 IO-APIC-edge i8042
15: 37 0 0 0 IO-APIC-edge ide1
17: 8083 0 0 0 IO-APIC-level megaraid
20: 20141 0 0 0 IO-APIC-level eth0
23: 0 0 0 0 IO-APIC-level acpi
NMI: 0 0 0 0
LOC: 1562738 1562605 1562608 1562607
ERR: 0
MIS: 0

With your patch, it looks like this:

0: 21266 79032 0 0 IO-APIC-edge timer
1: 14 0 0 0 IO-APIC-edge i8042
2: 0 0 0 0 XT-PIC cascade
4: 19 0 0 0 IO-APIC-edge serial
8: 2 0 0 0 IO-APIC-edge rtc
12: 503 0 0 0 IO-APIC-edge i8042
15: 39 0 0 1 IO-APIC-edge ide1
17: 1762 0 0 0 IO-APIC-level megaraid
20: 868 0 0 0 IO-APIC-level eth0
23: 0 0 0 0 IO-APIC-level acpi
NMI: 0 0 0 0
LOC: 94472 94484 94483 94482
ERR: 0
MIS: 0

There were other strange schemas that I've used before in 2.4, mostly due to ongoing inconsistencies in the BIOS and ACPI. Those were eventually cleaned up and now I guess a general convention can be applied, even though BIOS-clean IRQ schema still looks pretty exotic, as you noticed... With this patch, it looks like everything's taken care off.

Let me know if you need particular debug done with this patch. I'll send you a dmesg you've requested with increased CONFIG_LOG_BUF_SHIFT shortly (as soon as the system becomes available today).

Thanks,
--Natalie
\
 
 \ /
  Last update: 2005-03-22 14:02    [W:0.031 / U:0.272 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site