Bug 8817
Summary: | "irq 17: nobody cared" with ICH5/ALC650F audio | ||
---|---|---|---|
Product: | ACPI | Reporter: | David Campbell (david) |
Component: | Config-Interrupts | Assignee: | Len Brown (lenb) |
Status: | REJECTED INSUFFICIENT_DATA | ||
Severity: | normal | CC: | akpm, alan, lenb, rui.zhang, snecklifter, tiwai, yakui.zhao |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | 2.6.27 | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: |
dmesg from centos5 with noapic
dmesg from centos5 with pci=noacpi /proc/interrupts from centos5 with noapic /proc/interrupts from centos5 with pci=noacpi output of lspci -vxxx from centos 5 |
Description
David Campbell
2007-07-27 09:11:21 UTC
> Most recent kernel where this bug did not occur: unknown
that's a shame. It would really help if you could find a kernel which worked OK.
Then we can look at the dmesg for that kernel and compare it with the dmesg for
2/6/22 and compare them, which is useful info.
I don't know if it has ever worked in previous versions. I've only recently installed linux for the first time on my computer. A similar problem still happening in fedora 8 with kernel as below... Linux version 2.6.23.9-85.fc8 (kojibuilder@xenbuilder4.fedora.phx.redhat.com) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-33)) #1 SMP Fri Dec 7 15:49:59 EST 2007 --- irq 17: nobody cared (try booting with the "irqpoll" option) [<c045b16a>] __report_bad_irq+0x36/0x75 [<c045b380>] note_interrupt+0x1d7/0x213 [<c045a807>] handle_IRQ_event+0x23/0x51 [<c045bb0b>] handle_fasteoi_irq+0x86/0xa6 [<c045ba85>] handle_fasteoi_irq+0x0/0xa6 [<c04074c3>] do_IRQ+0x8c/0xb9 [<c04074d6>] do_IRQ+0x9f/0xb9 [<c0403ddf>] default_idle+0x0/0x55 [<c0405b6f>] common_interrupt+0x23/0x28 [<c0403ddf>] default_idle+0x0/0x55 [<c042007b>] save_v86_state+0x19/0x12b [<c0421f78>] native_safe_halt+0x2/0x3 [<c0403e18>] default_idle+0x39/0x55 [<c040340b>] cpu_idle+0xab/0xcc ======================= handlers: [<f8aacf8e>] (snd_intel8x0_interrupt+0x0/0x1e2 [snd_intel8x0]) [<f8a7bf9e>] (snd_intel8x0_interrupt+0x0/0x1a3 [snd_intel8x0m]) Disabling IRQ #17 Same deal with Fedora 10, 2.6.27.5-117.fc10.i686 Nov 27 10:29:03 localhost kernel: [<c0463984>] __report_bad_irq+0x2e/0x6f Nov 27 10:29:03 localhost kernel: [<c0463b93>] note_interrupt+0x1ce/0x223 Nov 27 10:29:03 localhost kernel: [<c0463104>] ? handle_IRQ_event+0x5c/0x64 Nov 27 10:29:03 localhost kernel: [<c0464154>] handle_fasteoi_irq+0x99/0xc0 Nov 27 10:29:03 localhost kernel: [<c04640bb>] ? handle_fasteoi_irq+0x0/0xc0 Nov 27 10:29:03 localhost kernel: [<c0405e3e>] do_IRQ+0xc7/0xfe Nov 27 10:29:03 localhost kernel: [<c0404654>] common_interrupt+0x28/0x30 Nov 27 10:29:03 localhost kernel: [<c04198d7>] ? native_safe_halt+0x5/0x7 Nov 27 10:29:03 localhost kernel: [<c040912d>] default_idle+0x38/0x6a Nov 27 10:29:03 localhost kernel: [<c0402c4d>] cpu_idle+0x101/0x134 Nov 27 10:29:03 localhost kernel: [<c0697406>] rest_init+0x4e/0x50 Nov 27 10:29:03 localhost kernel: ======================= Nov 27 10:29:03 localhost kernel: handlers: Nov 27 10:29:03 localhost kernel: [<f93ffe4e>] (snd_intel8x0_interrupt+0x0/0x1e3 [snd_intel8x0]) Nov 27 10:29:03 localhost kernel: [<f93e53d4>] (snd_intel8x0_interrupt+0x0/0x1b6 [snd_intel8x0m]) Nov 27 10:29:03 localhost kernel: Disabling IRQ #17 Nov 27 00:29:09 localhost kernel: [drm] LVDS-9: set mode a Nov 27 00:29:09 localhost kernel: [drm] bios LVDS_GEN_CNTL: 0xff24 Nov 27 00:29:10 localhost acpid: client connected from 2976[0:0] Nov 27 00:29:10 localhost kernel: mtrr: base(0xe1ca7000) is not aligned on a size(0xac5000) boundary IRQ routing so moving this to platform, although it could be ACPI and its often bogus BIOS provided data Can you see if it works if you boot with acpi=off and if that makes no difference if it works if you boot with the option "irqpoll" (To add options to your kernel for booting hit a key during the grub prompt and follow the instructions it provides to add these to the end of the kernel line) Alan if you boot with acpi=off or irqpoll then you don't see this problem, though I don't really know how those options make the problem go away someone recently changed the kernel version to "2.6.27-vmware" but this happens with or without vmware kernel modules loaded acpi=off changes which BIOS data is used for interrupt routing (and the fact it fixes it means we have a much clearer idea of where to look) irqpoll tells the kernel to assume the interrupt data on the PC is wrong and take alternative measures (at a small performance hit) and is the emergency sledgehammer for dealing with systems where there isn't a Linux problem and the vendor isn't updating their BIOS. As this goes away without ACPI I'll reassign this to the ACPI people (and fix the -vmware tag). It could equally be a buggy BIOS however but we shall see first please try the latest upstream kernel to see if the problem still exists. If yes, please attach the acpidump output using the latest pmtools at http://www.lesswatts.org/projects/acpi/utilities.php Hi, David Will you please try the following boot options and attach the output of dmesg, /proc/interrupts? a. noapic b. pci=noacpi Will you please also attach the output of lspci -vxxx? Thanks. ping Alan I no longer have fedora 7 installed, and Fedora 10 broke the video driver, so I'm attaching the requested detail from Centos5 on the same hardware. Created attachment 21225 [details]
dmesg from centos5 with noapic
Created attachment 21226 [details]
dmesg from centos5 with pci=noacpi
Created attachment 21227 [details]
/proc/interrupts from centos5 with noapic
Created attachment 21228 [details]
/proc/interrupts from centos5 with pci=noacpi
Created attachment 21229 [details]
output of lspci -vxxx from centos 5
hi, David Please also attach the output of acpidump. Thanks. The acpidump command doesn't seem to be present by default on centos5, and a yum search for it showed up nothing on centos5. I don't know where to go from here to get the output you request. You can download the acpidump tool(pmtools-20071116) from the following website and attach the output of acpidump. http://www.lesswatts.org/projects/acpi/utilities.php Thanks. Hi, Rui Can we reject this bug as there is no response for one month? Thanks. |