The Linux Kernel Mailing List
 help / color / mirror / Atom feed
* Disabling IRQ #11 - uhci || e100 || r128-dri problem
@ 2005-06-02 16:15 Meelis Roos
  2005-06-04  5:31 ` Dave Airlie
  0 siblings, 1 reply; 2+ messages in thread
From: Meelis Roos @ 2005-06-02 16:15 UTC (permalink / raw)
  To: Linux Kernel list

In recent 2.6.12 prereleases a bug has appeared. Whenever I log out of 
KDE, IRQ 11 is disabled on X server restart (kdm restarts it). It always 
results with the dmesg below and sometimes the new X loses keyboard 
input (ps2 keyboard, usb mouse). USB mouse works well. X works (sans 
keyboard sometimes). Network is dead.

irq 11: nobody cared!
  [<c0136144>] __report_bad_irq+0x24/0x90
  [<c0136241>] note_interrupt+0x61/0x90
  [<c0135c1b>] __do_IRQ+0x13b/0x150
  [<c0105607>] do_IRQ+0x47/0x70
  =======================
  [<c0103b2a>] common_interrupt+0x1a/0x20
  [<c011dfce>] __do_softirq+0x2e/0x80
  [<c0105711>] do_softirq+0x41/0x50
  =======================
  [<c011e0e5>] irq_exit+0x35/0x40
  [<c010560e>] do_IRQ+0x4e/0x70
  [<c0103b2a>] common_interrupt+0x1a/0x20
handlers:
[<e09622b0>] (usb_hcd_irq+0x0/0x60 [usbcore])
[<c02aabb0>] (e100_intr+0x0/0x150)

Disabling IRQ #11
Celeron 900, Intel ICH2 chipset with integrated UHCI and onboard e100 
NIC, running Debian unstable with XFree 4.3. 2.6.12-rc2 + XFree of that 
time was OK, didn't try rc3 and rc4, rc5 probably also not tested but 
rc5+git snapshots are the broken ones (together with current XFree). ATI 
rage128 graphics with DRI enabled; I get the following dmesg lines on 
X server start and at least the agpgart lines on each X server restart 
too. I have seen this bug some time earlier too, probably in 2.6.11 
timeframe, or in early 2.6.12 timeframe.

[drm] Initialized r128 2.5.0 20030725 on minor 0: ATI Technologies Inc Rage 128 RF/SG AGP
agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
agpgart: Putting AGP V2 device at 0000:00:00.0 into 2x mode
agpgart: Putting AGP V2 device at 0000:02:00.0 into 2x mode

/proc/interrupts during normal work:
            CPU0
   0:   13006046          XT-PIC  timer
   1:      26607          XT-PIC  i8042
   2:          0          XT-PIC  cascade
   7:          1          XT-PIC  parport0
   8:          4          XT-PIC  rtc
   9:          1          XT-PIC  acpi, Intel 82801BA-ICH2
  10:     336628          XT-PIC  uhci_hcd:usb2
  11:    1161748          XT-PIC  uhci_hcd:usb1, eth0, r128@PCI:2:0:0
  14:      48258          XT-PIC  ide0
  15:     194377          XT-PIC  ide1
NMI:       2680
LOC:   13007697
ERR:          0
MIS:          0


-- 
Meelis Roos (mroos@linux.ee)

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Disabling IRQ #11 - uhci || e100 || r128-dri problem
  2005-06-02 16:15 Disabling IRQ #11 - uhci || e100 || r128-dri problem Meelis Roos
@ 2005-06-04  5:31 ` Dave Airlie
  0 siblings, 0 replies; 2+ messages in thread
From: Dave Airlie @ 2005-06-04  5:31 UTC (permalink / raw)
  To: Meelis Roos; +Cc: Linux Kernel list

> In recent 2.6.12 prereleases a bug has appeared. Whenever I log out of
> KDE, IRQ 11 is disabled on X server restart (kdm restarts it). It always
> results with the dmesg below and sometimes the new X loses keyboard
> input (ps2 keyboard, usb mouse). USB mouse works well. X works (sans
> keyboard sometimes). Network is dead.

It looks like X or something is causing the r128 to emit an IRQ after
we've stopped listening for them,...

Can you see if you get any info with drm debug on?

echo 1 > /sys/modules/drm/parameters/debug
and restart X...
Dave.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2005-06-04  5:38 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-06-02 16:15 Disabling IRQ #11 - uhci || e100 || r128-dri problem Meelis Roos
2005-06-04  5:31 ` Dave Airlie

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox