Bug 6082

I think it would be neccesary for people to have a dmesg and /proc/interrupts
output from your machine

/proc/interrupts

           CPU0
  0:    1061648    IO-APIC-edge  timer
  1:      15411    IO-APIC-edge  i8042
  8:          4    IO-APIC-edge  rtc
  9:          2   IO-APIC-level  acpi
 12:        114    IO-APIC-edge  i8042
 14:      24644    IO-APIC-edge  libata
 15:      10583    IO-APIC-edge  libata
169:          2   IO-APIC-level  ehci_hcd:usb1, uhci_hcd:usb2, Intel ICH6, eth0
177:          1   IO-APIC-level  uhci_hcd:usb5, yenta
209:     171405   IO-APIC-level  uhci_hcd:usb3, Intel ICH6 Modem, ipw2200
217:      72378   IO-APIC-level  uhci_hcd:usb4
NMI:          0
LOC:     125846
ERR:          0
MIS:          0

with the DRM module loaded i get this /proc/interrupts:

           CPU0
  0:      98559    IO-APIC-edge  timer
  1:       1939    IO-APIC-edge  i8042
  8:          4    IO-APIC-edge  rtc
  9:          2   IO-APIC-level  acpi
 12:        114    IO-APIC-edge  i8042
 14:      10739    IO-APIC-edge  libata
 15:        967    IO-APIC-edge  libata
169:      19046   IO-APIC-level  ehci_hcd:usb1, uhci_hcd:usb2, Intel ICH6,
radeon@pci:0000:01:00.0, eth0
177:          1   IO-APIC-level  uhci_hcd:usb5, yenta
209:      16065   IO-APIC-level  uhci_hcd:usb3, ipw2200, Intel ICH6 Modem
217:       2188   IO-APIC-level  uhci_hcd:usb4
NMI:          0
LOC:      21519
ERR:          0
MIS:          0

sorry i made a mistake, it is the wired interface the one that doesn't work.
Actually it seems that everything that is sharing the same interrupt is not
working (the sound doesn't work either).

I'm chaging the summary of the bug to indicate this.

Created attachment 7367 [details]
output of kernel.logs for this kernel (runing with the radeon module)

Hi,      
     
I have the same with r300 and realtek r8186 (onboard) network card.     
Here is dmesg after X having started (probed drm and radeon modules):     
     
agpgart: Found an AGP 3.0 compliant device at 0000:00:00.0.     
agpgart: X tried to set rate=x12. Setting to AGP3 x8 mode.     
agpgart: Putting AGP V3 device at 0000:00:00.0 into 8x mode     
agpgart: Putting AGP V3 device at 0000:01:00.0 into 8x mode     
[drm] Loading R300 Microcode     
irq 16: nobody cared (try booting with the "irqpoll" option)     
     
Call Trace: <IRQ> <ffffffff8014ec70>{__report_bad_irq+48}      
<ffffffff8014ef0e>{note_interrupt+593}     
       <ffffffff8014e74c>{__do_IRQ+206} <ffffffff80110eec>{do_IRQ+45}     
       <ffffffff8010ed1c>{ret_from_intr+0}  <EOI>      
<ffffffff803ee07b>{thread_return+0}     
       <ffffffff8010cb47>{default_idle+0} <ffffffff8010cb7b>{default_idle+52}     
       <ffffffff8010ccc5>{cpu_idle+63} <ffffffff8054a78a>{start_kernel+465}     
       <ffffffff8054a24c>{_sinittext+588}     
handlers:     
[<ffffffff80310ba0>] (rtl8169_interrupt+0x0/0x2af)     
Disabling IRQ #16     
     
What helps is pci=routeirq as boot parameter. Problem exisit only with cvs drm    
module, one from kernel does not cause this problem. System is amd64 runing    
natively, with 2.6.15 kernel and x11r7. gcc is 3.4.5    
   

I find very weird that a system uses IRQ 169, CC'ing the ACPI guys who may know
more about this.

Created attachment 7381 [details]
patch to invoke pci_enable_device() when using old PCI detect

If pci=routeirq helps, this means that the driver did not call
pci_enable_device(), and is therefore using a bogus IRQ number (recent kernels
perform IRQ routing in pci_enable_device(), and pdev->irq does not have the
correct value until that function is called).

The drm core currently does not call pci_enable_device() when a framebuffer
driver (including vesafb) is active.  If a proper framebuffer driver (e.g.,
radeonfb) is used, drm still works fine, because the framebuffer driver calls
pci_enable_device() itself, and drm gets proper pdev->irq.  However, when drm
is used together with vesafb, pci_enable_device() is not performed, and the IRQ
problem shows up.

This patch makes the drm core call pci_enable_device() even if a framebuffer
driver was loaded before; this fixes the IRQ problem.

---

drm: fix IRQ problem when used together with vesafb

Even if drm uses old PCI probing (because a framebuffer driver is active), it
still needs to call pci_enable_device() to get correct IRQ.

Signed-off-by: Sergey Vlasov <vsu@altlinux.ru>

i just tried using pci=routeirq and it works

this isn't a kernel bug if you are using drivers from DRM CVS.. if you are using
DRM CVS drivers please close this bug, I've check an always enavble device into
DRM CVS.

Hi, I'm having a similar issue, even with latest drm, except my problem is with
a CDRW device. It gets disconnected each time a 3d program runs.

I've got Slackware current, kernel 2.6.15.4, XOrg 6.9 (both self-compiled),
latest drm and mesa.

Hardware Environment:
Athlon XP 2500+, 1 GB RAM
video card: ATI Radeon R350

lspci output:
00:00.0 Host bridge: nVidia Corporation nForce2 AGP (different version?) (rev c1)
00:00.1 RAM memory: nVidia Corporation nForce2 Memory Controller 1 (rev c1)
00:00.2 RAM memory: nVidia Corporation nForce2 Memory Controller 4 (rev c1)
00:00.3 RAM memory: nVidia Corporation nForce2 Memory Controller 3 (rev c1)
00:00.4 RAM memory: nVidia Corporation nForce2 Memory Controller 2 (rev c1)
00:00.5 RAM memory: nVidia Corporation nForce2 Memory Controller 5 (rev c1)
00:01.0 ISA bridge: nVidia Corporation nForce2 ISA Bridge (rev a4)
00:01.1 SMBus: nVidia Corporation nForce2 SMBus (MCP) (rev a2)
00:02.0 USB Controller: nVidia Corporation nForce2 USB Controller (rev a4)
00:02.1 USB Controller: nVidia Corporation nForce2 USB Controller (rev a4)
00:02.2 USB Controller: nVidia Corporation nForce2 USB Controller (rev a4)
00:04.0 Ethernet controller: nVidia Corporation nForce2 Ethernet Controller (rev a1)
00:05.0 Multimedia audio controller: nVidia Corporation nForce Audio Processing
Unit (rev a2)
00:06.0 Multimedia audio controller: nVidia Corporation nForce2 AC97 Audio
Controler (MCP) (rev a1)
00:08.0 PCI bridge: nVidia Corporation nForce2 External PCI Bridge (rev a3)
00:09.0 IDE interface: nVidia Corporation nForce2 IDE (rev a2)
00:0d.0 FireWire (IEEE 1394): nVidia Corporation nForce2 FireWire (IEEE 1394)
Controller (rev a3)
00:1e.0 PCI bridge: nVidia Corporation nForce2 AGP (rev c1)
01:0b.0 Ethernet controller: Intel Corporation 82540EM Gigabit Ethernet
Controller (rev 02)
01:0c.0 RAID bus controller: Integrated Technology Express, Inc. IT/ITE8212 Dual
channel ATA RAID controller (PCI version seems to be IT8212, embedded seems (rev 10)
01:0d.0 Mass storage controller: Silicon Image, Inc. SiI 3112
[SATALink/SATARaid] Serial ATA Controller (rev 02)
03:00.0 VGA compatible controller: ATI Technologies Inc Radeon R350 [Radeon 9800
Pro]
03:00.1 Display controller: ATI Technologies Inc Radeon R350 [Radeon 9800 Pro]
(Secondary)

My /proc/interrupts with drm loaded:
           CPU0
  0:   74255552          XT-PIC  timer
  1:     110676          XT-PIC  i8042
  2:          0          XT-PIC  cascade
  5:    7360005          XT-PIC  libata, ohci_hcd:usb2, radeon@pci:0000:03:00.0
  7:          3          XT-PIC  parport0
  8:          0          XT-PIC  rtc
  9:     323178          XT-PIC  acpi, ehci_hcd:usb1, eth0, NVidia nForce2
 11:   15840418          XT-PIC  ide2, ide3, ohci_hcd:usb3, eth1
 12:        102          XT-PIC  i8042
 14:     223958          XT-PIC  ide0
 15:      37171          XT-PIC  ide1
NMI:          0
ERR:          0

If I mount a cdrom, then start a 3d program, the CDRW read light starts flashing.
dmesg output (repeated a couple of times):
hda: status error: status=0x7f { DriveReady DeviceFault SeekComplete DataRequest
CorrectedError Index Error }
hda: status error: error=0x7f { IllegalLengthIndication EndOfMedia
AbortedCommand MediaChangeRequested LastFailedSense=0x07 }
ide: failed opcode was: unknown
hda: drive not ready for command
hda: ATAPI reset complete
hda: status error: status=0x7f { DriveReady DeviceFault SeekComplete DataRequest
CorrectedError Index Error }
hda: status error: error=0x7f { IllegalLengthIndication EndOfMedia
AbortedCommand MediaChangeRequested LastFailedSense=0x07 }
ide: failed opcode was: unknown
hda: drive not ready for command
hda: ATAPI reset complete
hda: status error: status=0x7f { DriveReady DeviceFault SeekComplete DataRequest
CorrectedError Index Error }
hda: status error: error=0x7f { IllegalLengthIndication EndOfMedia
AbortedCommand MediaChangeRequested LastFailedSense=0x07 }
ide: failed opcode was: unknown
hda: drive not ready for command

and after it:
ide: failed opcode was: unknown
hda: drive not ready for command
hda: drive not ready for command
VFS: busy inodes on changed media.
VFS: busy inodes on changed media.

dmesg output when I exit the program and try to access the CDRW:
hda: lost interrupt
VFS: busy inodes on changed media.

and after a couple of seconds, the CDRW starts working normally again, until the
next time I activate a 3d program.

Is this bug fixed? shouldn't it be closed?

Boris, in your setup the video card is not sharing the irq with the ide device
and you are using a nvidia binary module aswell, it looks different from this
bug. If the problem persist (without the nvidia driver loaded) using the latest
stable kernel, report this issue in a new bug

this has been fixed in git as far as I'm aware.