Skip to contentSkip to conversation

snd-intel-hda: screaming IRQ with 3.5.3 kernel (HDMI)

DISCUSSION Chris Rankin 2 participants 6 messages Sep 8, 2012 – Sep 10, 2012

lore
Share & tools
lore
#0 lore raw

Hi,

I have recently added a Radeon HD4670 to one of my older PCs, and have noticed that it has IRQ trouble when I try to enable the HDMI audio:

irq 17: nobody cared (try booting with the "irqpoll" option)
Pid: 0, comm: swapper/0 Not tainted 3.5.3 #2

Call Trace:
 [<c1063806>] ? __report_bad_irq+0x11/0x94
 [<c1063a2d>] ? note_interrupt+0x120/0x19b
 [<c1062325>] ? handle_irq_event_percpu+0xe8/0xfb
 [<c1062361>] ? handle_irq_event+0x29/0x40
 [<c10640fc>] ? handle_level_irq+0x93/0x93
 [<c1064163>] ? handle_fasteoi_irq+0x67/0x90
 <IRQ>  [<c1003625>] ? do_IRQ+0x2e/0x83
 [<c1229829>] ? common_interrupt+0x29/0x30
 [<c106007b>] ? audit_update_watch+0x59/0x285
 [<c1007479>] ? default_idle+0x23/0x3b
 [<c1007b38>] ? cpu_idle+0x4b/0x7e
 [<c1314877>] ? start_kernel+0x2a4/0x2a7

handlers:
[<f8c97b75>] azx_interrupt [snd_hda_intel]
Disabling IRQ #17

This is frequently followed by KMS trouble too:

irq 16: nobody cared (try booting with the "irqpoll" option)
Pid: 0, comm: swapper/0 Not tainted 3.5.3 #2

Call Trace:
 [<c1063806>] ? __report_bad_irq+0x11/0x94
 [<c1063a2d>] ? note_interrupt+0x120/0x19b
 [<c1062325>] ? handle_irq_event_percpu+0xe8/0xfb
 [<c1062361>] ? handle_irq_event+0x29/0x40
 [<c10640fc>] ? handle_level_irq+0x93/0x93
 [<c1064163>] ? handle_fasteoi_irq+0x67/0x90
 <IRQ>  [<c1003625>] ? do_IRQ+0x2e/0x83
 [<c1229829>] ? common_interrupt+0x29/0x30
 [<c106007b>] ? audit_update_watch+0x59/0x285
 [<c1007479>] ? default_idle+0x23/0x3b
 [<c1007b38>] ? cpu_idle+0x4b/0x7e
 [<c1314877>] ? start_kernel+0x2a4/0x2a7

handlers:
[<f818bfb3>] radeon_driver_irq_handler_kms [radeon] [<f81fe55f>] usb_hcd_irq [usbcore] Disabling IRQ #16

The PCI entries are:

01:00.1 Audio device: ATI Technologies Inc RV710/730
Subsystem: ATI Technologies Inc RV710/730
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 64 (2000ns min), Cache Line Size: 64 bytes
Interrupt: pin B routed to IRQ 17
Region 0: Memory at ff8ec000 (32-bit, non-prefetchable) [size=16K]
Capabilities: [50] Power Management version 3
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA
PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Kernel driver in use: snd_hda_intel

and:

01:00.0 VGA compatible controller: ATI Technologies Inc RV730 Pro AGP [Radeon HD 4600 Series] (prog-if 00 [VGA controller])
Subsystem: ATI Technologies Inc Device 0028
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR+ FastB2B- DisINTx-
Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx+

        Latency: 64 (2000ns min), Cache Line Size: 64 bytes
        Interrupt: pin A routed to IRQ 16
        Region 0: Memory at e0000000 (32-bit, prefetchable) [size=256M]
        Region 1: I/O ports at ec00 [size=256]
        Region 2: Memory at ff8f0000 (32-bit, non-prefetchable) [size=64K]
        Expansion ROM at ff800000 [disabled] [size=128K]
        Capabilities: [50] Power Management version 3
                Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA 

PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [58] AGP version 3.0
Status: RQ=256 Iso- ArqSz=0 Cal=0 SBA+ ITACoh- GART64- HTrans-
64bit- FW+ AGP3+ Rate=x4,x8
Command: RQ=32 ArqSz=2 Cal=0 SBA+ AGP+ GART64- 64bit- FW- Rate=x8
Kernel driver in use: radeon

Does anyone have any suggestions, please?

Thanks,
Chris

#1 lore raw

At Sat, 08 Sep 2012 14:59:09 +0100,

Quoted 89 lines · Chris Rankin

Hi,

I have recently added a Radeon HD4670 to one of my older PCs, and have noticed that it has IRQ trouble when I try to enable the HDMI audio:

irq 17: nobody cared (try booting with the "irqpoll" option)
Pid: 0, comm: swapper/0 Not tainted 3.5.3 #2

Call Trace:
  [<c1063806>] ? __report_bad_irq+0x11/0x94
  [<c1063a2d>] ? note_interrupt+0x120/0x19b
  [<c1062325>] ? handle_irq_event_percpu+0xe8/0xfb
  [<c1062361>] ? handle_irq_event+0x29/0x40
  [<c10640fc>] ? handle_level_irq+0x93/0x93
  [<c1064163>] ? handle_fasteoi_irq+0x67/0x90
  <IRQ>  [<c1003625>] ? do_IRQ+0x2e/0x83
  [<c1229829>] ? common_interrupt+0x29/0x30
  [<c106007b>] ? audit_update_watch+0x59/0x285
  [<c1007479>] ? default_idle+0x23/0x3b
  [<c1007b38>] ? cpu_idle+0x4b/0x7e
  [<c1314877>] ? start_kernel+0x2a4/0x2a7

handlers:
[<f8c97b75>] azx_interrupt [snd_hda_intel]
Disabling IRQ #17

This is frequently followed by KMS trouble too:

irq 16: nobody cared (try booting with the "irqpoll" option)
Pid: 0, comm: swapper/0 Not tainted 3.5.3 #2

Call Trace:
  [<c1063806>] ? __report_bad_irq+0x11/0x94
  [<c1063a2d>] ? note_interrupt+0x120/0x19b
  [<c1062325>] ? handle_irq_event_percpu+0xe8/0xfb
  [<c1062361>] ? handle_irq_event+0x29/0x40
  [<c10640fc>] ? handle_level_irq+0x93/0x93
  [<c1064163>] ? handle_fasteoi_irq+0x67/0x90
  <IRQ>  [<c1003625>] ? do_IRQ+0x2e/0x83
  [<c1229829>] ? common_interrupt+0x29/0x30
  [<c106007b>] ? audit_update_watch+0x59/0x285
  [<c1007479>] ? default_idle+0x23/0x3b
  [<c1007b38>] ? cpu_idle+0x4b/0x7e
  [<c1314877>] ? start_kernel+0x2a4/0x2a7

handlers:
[<f818bfb3>] radeon_driver_irq_handler_kms [radeon] [<f81fe55f>] usb_hcd_irq [usbcore] Disabling IRQ #16

The PCI entries are:

01:00.1 Audio device: ATI Technologies Inc RV710/730
Subsystem: ATI Technologies Inc RV710/730
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 64 (2000ns min), Cache Line Size: 64 bytes
Interrupt: pin B routed to IRQ 17
Region 0: Memory at ff8ec000 (32-bit, non-prefetchable) [size=16K]
Capabilities: [50] Power Management version 3
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA
PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Kernel driver in use: snd_hda_intel

and:

01:00.0 VGA compatible controller: ATI Technologies Inc RV730 Pro AGP [Radeon HD 4600 Series] (prog-if 00 [VGA controller])
Subsystem: ATI Technologies Inc Device 0028
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR+ FastB2B- DisINTx-
Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx+

         Latency: 64 (2000ns min), Cache Line Size: 64 bytes
         Interrupt: pin A routed to IRQ 16
         Region 0: Memory at e0000000 (32-bit, prefetchable) [size=256M]
         Region 1: I/O ports at ec00 [size=256]
         Region 2: Memory at ff8f0000 (32-bit, non-prefetchable) [size=64K]
         Expansion ROM at ff800000 [disabled] [size=128K]
         Capabilities: [50] Power Management version 3
                 Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA 

PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [58] AGP version 3.0
Status: RQ=256 Iso- ArqSz=0 Cal=0 SBA+ ITACoh- GART64- HTrans-
64bit- FW+ AGP3+ Rate=x4,x8
Command: RQ=32 ArqSz=2 Cal=0 SBA+ AGP+ GART64- 64bit- FW- Rate=x8
Kernel driver in use: radeon

Quoted 1 line

Does anyone have any suggestions, please?

Did you try any workaround boot options such as irqpoll, pci=nocrs or whatever?

Takashi

#2 lore raw

----- Original Message -----

Quoted 1 line

Did you try any workaround boot options such as irqpoll, pci=nocrs or whatever?

No, because wouldn't "irqpoll" have forced polling for every IRQ on the box? I need to avoid polling on the video card's IRQ because it noticeably impacts the responsiveness. (I assume that the kernel switches to polling the device instead when it disables the IRQ anyway).

I didn't try "pci=nocrs" because I didn't know about it. I can try that tonight.

Cheers,
Chris

#3 lore raw

At Mon, 10 Sep 2012 03:28:58 -0700 (PDT),

Quoted 6 lines · Chris Rankin

----- Original Message -----

Quoted 1 line

Did you try any workaround boot options such as irqpoll, pci=nocrs or whatever?

No, because wouldn't "irqpoll" have forced polling for every IRQ on the box?

No, you misunderstand.

Takashi

#4 lore raw

----- Original Message -----

Quoted 1 line

Did you try any workaround boot options such as irqpoll, pci=nocrs or whatever?

OK, the "irqpoll=1" option does help here. Although this message is obviously scary:

Misrouted IRQ fixup and polling support enabled
This may significantly impact system performance

And the "pci=nocrs" option does Really Bad Things to the display at boot-time.

So what is "irqpoll" actually doing, please? Is it just a temporary workaround for a driver bug?

Thanks,
Chris

#5 lore raw
In reply to Takashi Iwai

----- Original Message -----

Quoted 1 line

  Did you try any workaround boot options such as irqpoll, pci=nocrs or whatever?

Actually, it turns out that booting with "irqpoll=1" simply means that this problem takes longer to happen:

irq 17: nobody cared (try booting with the "irqpoll" option)
Pid: 3347, comm: wineserver Not tainted 3.5.3 #2

Call Trace:
 [<c1063806>] ? __report_bad_irq+0x11/0x94
 [<c1063a2d>] ? note_interrupt+0x120/0x19b
 [<c1062325>] ? handle_irq_event_percpu+0xe8/0xfb
 [<c1062361>] ? handle_irq_event+0x29/0x40
 [<c1064163>] ? handle_fasteoi_irq+0x67/0x90
 [<c1003957>] ? handle_irq+0x63/0x70
 [<c1003625>] ? do_IRQ+0x2e/0x83
 [<c1229829>] ? common_interrupt+0x29/0x30

handlers:
[<f8d09b75>] azx_interrupt [snd_hda_intel]
Disabling IRQ #17

Cheers,
Chris