lkml.org 
[lkml]   [2001]   [Apr]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: /proc/interrupts vs. /proc/stat (and reality)
On Sun, 29 Apr 2001, Michael Mauch wrote:

> Hi,
>
> my /proc/interrupts doesn't show all the interrupts that have been used,
> /proc/stat (and hence procinfo) does. All of the devices work without
> problems.
>
> % cat /proc/interrupts /proc/stat && procinfo
> CPU0
> 0: 476919 XT-PIC timer
> 1: 12333 XT-PIC keyboard
> 2: 0 XT-PIC cascade
> 3: 221 XT-PIC
> 5: 8388 XT-PIC soundblaster
> 7: 0 XT-PIC parport0
> 8: 67 XT-PIC rtc
> 10: 53 XT-PIC advansys, ncr53c8xx
> 11: 35704 XT-PIC HiSax
> 12: 49384 XT-PIC PS/2 Mouse
> 14: 29 XT-PIC ide0
> 15: 39616 XT-PIC ide1
> NMI: 0
> ERR: 0
> cpu 25315 461 35481 415662
> cpu0 25315 461 35481 415662
> page 331823 243999
> swap 2 0
> intr 622904 476919 12333 0 221 160 8388 30 0 67 0 53 35704 49384 0 29 39616 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> disk_io: (2,0):(34,34,37,0,0) (3,0):(5,5,10,0,0) (8,0):(8,8,16,0,0)
> ctxt 2581888
> btime 988535311
> processes 1291
> Linux 2.4.4-pre4 (root@elmicha) (gcc 2.95.3 20010315 ) #2 1CPU [elmicha.(none)]
>
> Memory: Total Used Free Shared Buffers Cached
> Mem: 384352 319980 64372 0 4816 137528
> Swap: 939752 260 939492
>
> Bootup: Sun Apr 29 11:08:31 2001 Load average: 0.05 0.13 0.10 6/103 1292
>
> user : 0:04:13.15 5.3% page in : 331823
> nice : 0:00:04.61 0.1% page out: 243999
> system: 0:05:54.82 7.4% swap in : 2
> idle : 1:09:16.62 87.2% swap out: 0
> uptime: 1:19:29.19 context : 2581892
>
> irq 0: 476920 timer irq 7: 0 parport0
> irq 1: 12333 keyboard irq 8: 67 rtc
> irq 2: 0 cascade [4] irq 10: 53 advansys, ncr53c8xx
> irq 3: 221 irq 11: 35704 HiSax
> irq 4: 160 irq 12: 49384 PS/2 Mouse
> irq 5: 8388 soundblaster irq 14: 29 ide0
> irq 6: 30 irq 15: 39616 ide1
>
>
> So IRQ 3 doesn't know it's getting serviced by the serial module, IRQ 4
> and IRQ 6 don't show in /proc/interrupts (and I've used both serial
> ports and the floppy before), only in /proc/stat.
>
> Have I done something wrong? Should I send more info?

I am too having the same thing appear, it started once i added my advansys
based Ultra Wide SCSI adapter, all the serial port identities dsiappeared
from the IRQ list but functionality is still there, i have COM1,2,3,4 on
IRQ's 4,3,7,9 respectivly (extra 2 are a ISA 2 x ST16650V2 fully
jumperable IO card), procinfo shows:

irq 0: 984034 timer irq 7: 22
irq 1: 2 keyboard irq 9: 3342
irq 2: 0 cascade [4] irq 11: 47697 advansys, eth0
irq 3: 1825934 irq 14: 23321 ide0
irq 5: 1 soundblaster irq 15: 13057 ide1
irq 6: 3

and /proc/interrupts show:

CPU0
0: 1048426 XT-PIC timer
1: 2 XT-PIC keyboard
2: 0 XT-PIC cascade
3: 1943828 XT-PIC
5: 1 XT-PIC soundblaster
7: 22 XT-PIC
9: 3342 XT-PIC
11: 56781 XT-PIC advansys, eth0
14: 28755 XT-PIC ide0
15: 14328 XT-PIC ide1
NMI: 0
ERR: 0

The above was taken from my system whist running kernel 2.4.4 (release)
it also appeared on 2.4.3 (release) after i added the advansys card, which
has appeared to have jumped on irq 11 along side the DEC Tulip ethernet
card. (The advansys replaced a sym53c810a card which used to sit on irq
10)

Maybe its something to do with IRQ sharing?

Regards, Jim.


-
name : Jim Woodward
www : http://www.jim.southcom.com.au
email : jim@jim.southcom.com.au

-
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 12:52    [W:0.033 / U:0.740 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site