Subject : [2.6.27-rc7-git1] (Fedora) - ACPI instabilities and IRQs being disabled Submitter : Shawn Starr <shawn.starr@rogers.com> Date : 2008-09-27 22:40 References : http://marc.info/?l=linux-kernel&m=122255527412178&w=4 This entry is being used for tracking a regression from 2.6.26. Please don't close it until the problem is fixed in the mainline. Reportedly caused by: commit de85422b94ddb23c021126815ea49414047c13dc Author: Stefan Becker <Stefan.Becker@nokia.com> Date: Tue Jul 1 19:19:22 2008 +0300 USB: fix interrupt disabling for HCDs with shared interrupt handlers Signed-off-by: Stefan Becker <stefan.becker@nokia.com> Cc: stable <stable@kernel.org> Acked-by: David Brownell <david-b@pacbell.net> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
re-worded the summary, as this doesn't appear to be an ACPI bug. I do agree that this regression is a bug -- PCI level IRQ sharing should work even in old XT-PIC mode. However, you should be aware that a T42 has an IOAPIC and that if you build it into the kernel, it is likely that the additional IRQs will allow running the system with less IRQ sharing.
Len - it *does* look like an ACPI bug. First we get an ACPI interrupt (irq#9) that is pretty clearly the result of some ACPI activity (the stack trace has that "acpi_ex_system_io_space_handler" being the last thing before the interrupt happens). But ACPI apparently claims isn't for it at all, so we shut it off due to an interrupt storm. Then, the very next thing that happens is that ACPI says "I didn't see an interrupt, so I'm not going to use interrupt mode". If that isn't an ACPI bug, I don't know what it would be... Everything that is going on shouts ACPI.
Hi, Shawn Will you please confirm whether the problem is fixed if the commit is reverted? >commit de85422b94ddb23c021126815ea49414047c13dc >Author: Stefan Becker <Stefan.Becker@nokia.com> >Date: Tue Jul 1 19:19:22 2008 +0300 >USB: fix interrupt disabling for HCDs with shared interrupt handlers From the /proc/interrupts it seems that the IRQ 9 is shared by ACPI, USB, Yenta, e1000. And when IRQ 9 is disabled, the number of IRQ 9 is still very small. How this happens? Thanks.
How about not load devices like USB, yanta which share irq 9 with ACPI? This could help narrow down if this is an ACPI issue.
I reverted the code, no change, I even disabled usb 'nousb' in boot up, still same result. I spoke to the radeon drm maintainer it might be a bug in the radeon drm driver. Since we've eliminated the yenta and usb devices, the last thing is e1000. [ 69.948958] agpgart-intel 0000:00:00.0: AGP 2.0 bridge [ 69.949002] agpgart-intel 0000:00:00.0: putting AGP V2 device into 4x mode [ 69.949093] pci 0000:01:00.0: putting AGP V2 device into 4x mode [ 70.632272] [drm] Setting GART location based on new memory map [ 70.632302] [drm] Loading R300 Microcode [ 70.632365] [drm] Num pipes: 1 [ 70.632378] [drm] writeback test succeeded in 1 usecs [ 169.435597] irq 9: nobody cared (try booting with the "irqpoll" option) [ 169.435620] Pid: 2106, comm: X Not tainted 2.6.27-0.398.rc9.fc10.i686 #1 [ 169.435640] [<c0466a2e>] __report_bad_irq+0x33/0x74 [ 169.435665] [<c0466c42>] note_interrupt+0x1d3/0x225 [ 169.435678] [<c0466159>] ? handle_IRQ_event+0x61/0x69 [ 169.435692] [<c04672fe>] handle_level_irq+0x8d/0xc3 [ 169.435704] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 169.435717] [<c0406fc1>] do_IRQ+0x9f/0xc9 [ 169.435732] [<c0405710>] common_interrupt+0x28/0x30 [ 169.435748] [<c0431e28>] ? __do_softirq+0x6b/0x10f [ 169.435765] [<c0431dbd>] ? __do_softirq+0x0/0x10f [ 169.435779] [<c0407067>] do_softirq+0x7c/0xdd [ 169.435789] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 169.435802] [<c0431a80>] irq_exit+0x49/0x88 [ 169.435813] [<c0406fd5>] do_IRQ+0xb3/0xc9 [ 169.435824] [<c0405710>] common_interrupt+0x28/0x30 [ 169.435839] ======================= [ 169.435844] handlers: [ 169.435848] [<c055131c>] (acpi_irq+0x0/0x28) [ 169.435861] [<f8939006>] (e1000_intr+0x0/0x13f [e1000]) [ 169.435911] Disabling IRQ #9 [ 170.895981] BUG: unable to handle kernel NULL pointer dereference at 00000004 [ 170.895990] IP: [<c052eef2>] __list_add+0xa/0x5c [ 170.896000] *pde = 22ee6067 *pte = 00000000 [ 170.896012] Oops: 0000 [#1] SMP [ 170.896012] Modules linked in: bridge stp bnep l2cap bluetooth autofs4 fuse sunrpc ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 cpufreq_ondemand acpi_cpufreq dm_multipath snd_intel8x0 snd_intel8x0m snd_ac97_codec snd_seq_dummy ac97_bus snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd ppdev pcspkr soundcore iTCO_wdt joydev iTCO_vendor_support snd_page_alloc e1000 i2c_i801 video parport_pc output parport pata_acpi ata_generic sha256_generic cbc aes_i586 dm_crypt radeon drm i2c_algo_bit i2c_core [last unloaded: microcode] [ 170.896012] [ 170.896012] Pid: 2641, comm: glxgears Not tainted (2.6.27-0.398.rc9.fc10.i686 #1) [ 170.896012] EIP: 0060:[<c052eef2>] EFLAGS: 00010046 CPU: 0 [ 170.896012] EIP is at __list_add+0xa/0x5c [ 170.896012] EAX: e2ea9f10 EBX: e2ea9f04 ECX: 00000000 EDX: f78230ec [ 170.896012] ESI: f78230ec EDI: e2ea9f10 EBP: e2ea9ee4 ESP: e2ea9ed8 [ 170.896012] DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 [ 170.896012] Process glxgears (pid: 2641, ti=e2ea9000 task=e2e7b300 task.ti=e2ea9000) [ 170.896012] Stack: e2ea9f04 f78230e8 00000246 e2ea9ef8 c043eddf f7823000 f78230e8 fffe1367 [ 170.896012] e2ea9f24 f88e399f 00000003 00000000 e2e7b300 c0427600 00000000 00000000 [ 170.896012] fffffff4 f6f59fa0 40046457 e2ea9f48 f886f6dc f69e1120 f6e09000 f88e3910 [ 170.896012] Call Trace: [ 170.896012] [<c043eddf>] ? add_wait_queue+0x27/0x35 [ 170.896012] [<f88e399f>] ? radeon_irq_wait+0x8f/0x102 [radeon] [ 170.896012] [<c0427600>] ? default_wake_function+0x0/0xd [ 170.896012] [<f886f6dc>] ? drm_ioctl+0x1b2/0x227 [drm] [ 170.896012] [<f88e3910>] ? radeon_irq_wait+0x0/0x102 [radeon] [ 170.896012] [<c04a0789>] ? vfs_ioctl+0x55/0x6e [ 170.896012] [<c04a09f1>] ? do_vfs_ioctl+0x24f/0x262 [ 170.896012] [<c06d987a>] ? _cond_resched+0x8/0x32 [ 170.896012] [<c052c1d2>] ? copy_to_user+0x40/0x110 [ 170.896012] [<c04a0a49>] ? sys_ioctl+0x45/0x60 [ 170.896012] [<c0404d32>] ? syscall_call+0x7/0xb [ 170.896012] [<c06d007b>] ? agp_via_probe+0xbe/0x114 [ 170.896012] ======================= [ 170.896012] Code: ef ff 83 c4 14 8b 13 8b 43 04 89 42 04 89 10 c7 43 04 00 02 20 00 c7 03 00 01 10 00 8b 5d fc c9 c3 55 89 e5 57 89 c7 56 89 d6 53 <8b> 41 04 89 cb 39 d0 74 17 51 50 52 68 31 05 7b c0 6a 1a 68 e6 [ 170.896012] EIP: [<c052eef2>] __list_add+0xa/0x5c SS:ESP 0068:e2ea9ed8 [ 170.896012] ---[ end trace 6ad7a78f97ed0bde ]--- [ 172.069062] ACPI: EC: missing confirmations, switch off interrupt mode. [ 205.789278] e1000 0000:02:01.0: PCI INT A disabled [ 208.094695] Intel(R) PRO/1000 Network Driver - version 7.3.20-k3-NAPI [ 208.094706] Copyright (c) 1999-2006 Intel Corporation. [ 208.094806] e1000 0000:02:01.0: PCI INT A -> Link[LNKA] -> GSI 9 (level, low) -> IRQ 9 [ 208.343652] e1000: 0000:02:01.0: e1000_probe: (PCI:33MHz:32-bit) 00:0d:60:ca:c1:97 [ 208.373842] e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network Connection [ 214.163399] irq 9: nobody cared (try booting with the "irqpoll" option) [ 214.163416] Pid: 2047, comm: NetworkManager Tainted: G D 2.6.27-0.398.rc9.fc10.i686 #1 [ 214.163426] [<c0466a2e>] __report_bad_irq+0x33/0x74 [ 214.163445] [<c0466c42>] note_interrupt+0x1d3/0x225 [ 214.163457] [<c0466159>] ? handle_IRQ_event+0x61/0x69 [ 214.163470] [<c04672fe>] handle_level_irq+0x8d/0xc3 [ 214.163481] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 214.163494] [<c0406fc1>] do_IRQ+0x9f/0xc9 [ 214.163505] [<c0405710>] common_interrupt+0x28/0x30 [ 214.163519] [<c0431e28>] ? __do_softirq+0x6b/0x10f [ 214.163535] [<c0431dbd>] ? __do_softirq+0x0/0x10f [ 214.163548] [<c0407067>] do_softirq+0x7c/0xdd [ 214.163558] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 214.163570] [<c0431a80>] irq_exit+0x49/0x88 [ 214.163581] [<c0406fd5>] do_IRQ+0xb3/0xc9 [ 214.163591] [<c0405710>] common_interrupt+0x28/0x30 [ 214.163603] [<c046007b>] ? audit_receive_filter+0x700/0x875 [ 214.163619] [<c06daec4>] ? _spin_unlock_irqrestore+0x32/0x3d [ 214.163634] [<c0466837>] setup_irq+0x1cf/0x232 [ 214.163645] [<f8937006>] ? e1000_intr+0x0/0x13f [e1000] [ 214.163679] [<c046698d>] request_irq+0xf3/0x111 [ 214.163689] [<f8937006>] ? e1000_intr+0x0/0x13f [e1000] [ 214.163718] [<f8936d5e>] e1000_request_irq+0x6d/0xb5 [e1000] [ 214.163746] [<f89375f9>] e1000_open+0xa6/0x143 [e1000] [ 214.163773] [<c0643fed>] dev_open+0x7b/0xb9 [ 214.163785] [<c0643a27>] dev_change_flags+0xa1/0x154 [ 214.163796] [<c064a6ab>] ? rtnl_setlink+0x0/0xdb [ 214.163810] [<c064a5f9>] do_setlink+0x232/0x2e4 [ 214.163822] [<c064a280>] ? nlmsg_parse+0x3c/0x4a [ 214.163835] [<c064a6ab>] ? rtnl_setlink+0x0/0xdb [ 214.163847] [<c064a782>] rtnl_setlink+0xd7/0xdb [ 214.163860] [<c064b524>] rtnetlink_rcv_msg+0x19b/0x1b5 [ 214.163873] [<c064b389>] ? rtnetlink_rcv_msg+0x0/0x1b5 [ 214.163886] [<c0658de7>] netlink_rcv_skb+0x35/0x7a [ 214.163899] [<c064b381>] rtnetlink_rcv+0x21/0x29 [ 214.163911] [<c06589e4>] netlink_unicast+0xf3/0x149 [ 214.163923] [<c0658c6e>] netlink_sendmsg+0x234/0x241 [ 214.163935] [<c0637fb3>] __sock_sendmsg+0x4a/0x53 [ 214.163947] [<c06387d1>] sock_sendmsg+0xbd/0xd3 [ 214.163958] [<c043ec03>] ? autoremove_wake_function+0x0/0x38 [ 214.163973] [<c043ec03>] ? autoremove_wake_function+0x0/0x38 [ 214.163987] [<c041ec7e>] ? need_resched+0x18/0x22 [ 214.164001] [<c052c0ab>] ? copy_from_user+0x3e/0x125 [ 214.164014] [<c063ed7a>] ? verify_iovec+0x45/0x73 [ 214.164014] [<c063892b>] sys_sendmsg+0x144/0x197 [ 214.164014] [<c0639420>] ? sys_recvmsg+0x173/0x180 [ 214.164014] [<c044123e>] ? enqueue_hrtimer+0xc6/0xd1 [ 214.164014] [<c06daeb9>] ? _spin_unlock_irqrestore+0x27/0x3d [ 214.164014] [<c0441b1e>] ? hrtimer_start+0x12f/0x149 [ 214.164014] [<c0441d10>] ? ktime_get+0x18/0x34 [ 214.164014] [<c0421eca>] ? hrtick_start_fair+0x11c/0x153 [ 214.164014] [<c042a410>] ? finish_task_switch+0x2f/0xb0 [ 214.164014] [<c06d9787>] ? schedule+0x6ee/0x70d [ 214.164014] [<c0639943>] sys_socketcall+0x16d/0x1a3 [ 214.164014] [<c0406fd5>] ? do_IRQ+0xb3/0xc9 [ 214.164014] [<c0404d32>] syscall_call+0x7/0xb [ 214.164014] ======================= [ 214.164014] handlers: [ 214.164014] [<c055131c>] (acpi_irq+0x0/0x28) [ 214.164014] [<f8937006>] (e1000_intr+0x0/0x13f [e1000]) [ 214.164014] Disabling IRQ #9 [ 214.234392] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 223.380226] e1000 0000:02:01.0: PCI INT A disabled [ 227.812789] Intel(R) PRO/1000 Network Driver - version 7.3.20-k3-NAPI [ 227.812801] Copyright (c) 1999-2006 Intel Corporation. [ 227.812896] e1000 0000:02:01.0: PCI INT A -> Link[LNKA] -> GSI 9 (level, low) -> IRQ 9 [ 228.061653] e1000: 0000:02:01.0: e1000_probe: (PCI:33MHz:32-bit) 00:0d:60:ca:c1:97 [ 228.097070] e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network Connection [ 233.874792] irq 9: nobody cared (try booting with the "irqpoll" option) [ 233.874809] Pid: 2047, comm: NetworkManager Tainted: G D 2.6.27-0.398.rc9.fc10.i686 #1 [ 233.874819] [<c0466a2e>] __report_bad_irq+0x33/0x74 [ 233.874838] [<c0466c42>] note_interrupt+0x1d3/0x225 [ 233.874850] [<c0466159>] ? handle_IRQ_event+0x61/0x69 [ 233.874863] [<c04672fe>] handle_level_irq+0x8d/0xc3 [ 233.874875] [<c0406fd0>] do_IRQ+0xae/0xc9 [ 233.874885] [<c0405710>] common_interrupt+0x28/0x30 [ 233.874899] [<c06d007b>] ? agp_via_probe+0xbe/0x114 [ 233.874914] [<c0466118>] ? handle_IRQ_event+0x20/0x69 [ 233.874926] [<c04672ea>] handle_level_irq+0x79/0xc3 [ 233.874937] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 233.874950] [<c0406fc1>] do_IRQ+0x9f/0xc9 [ 233.874960] [<c0405710>] common_interrupt+0x28/0x30 [ 233.874972] [<c0431e28>] ? __do_softirq+0x6b/0x10f [ 233.874989] [<c0431dbd>] ? __do_softirq+0x0/0x10f [ 233.875002] [<c0407067>] do_softirq+0x7c/0xdd [ 233.875012] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 233.875019] [<c0431a80>] irq_exit+0x49/0x88 [ 233.875019] [<c0406fd5>] do_IRQ+0xb3/0xc9 [ 233.875019] [<c0405710>] common_interrupt+0x28/0x30 [ 233.875019] [<c046007b>] ? audit_receive_filter+0x700/0x875 [ 233.875019] [<c06daec4>] ? _spin_unlock_irqrestore+0x32/0x3d [ 233.875019] [<c0466837>] setup_irq+0x1cf/0x232 [ 233.875019] [<f8937006>] ? e1000_intr+0x0/0x13f [e1000] [ 233.875019] [<c046698d>] request_irq+0xf3/0x111 [ 233.875019] [<f8937006>] ? e1000_intr+0x0/0x13f [e1000] [ 233.875019] [<f8936d5e>] e1000_request_irq+0x6d/0xb5 [e1000] [ 233.875019] [<f89375f9>] e1000_open+0xa6/0x143 [e1000] [ 233.875019] [<c0643fed>] dev_open+0x7b/0xb9 [ 233.875019] [<c0643a27>] dev_change_flags+0xa1/0x154 [ 233.875019] [<c064a6ab>] ? rtnl_setlink+0x0/0xdb [ 233.875019] [<c064a5f9>] do_setlink+0x232/0x2e4 [ 233.875019] [<c064a280>] ? nlmsg_parse+0x3c/0x4a [ 233.875019] [<c064a6ab>] ? rtnl_setlink+0x0/0xdb [ 233.875019] [<c064a782>] rtnl_setlink+0xd7/0xdb [ 233.875019] [<c064b524>] rtnetlink_rcv_msg+0x19b/0x1b5 [ 233.875019] [<c064b389>] ? rtnetlink_rcv_msg+0x0/0x1b5 [ 233.875019] [<c0658de7>] netlink_rcv_skb+0x35/0x7a [ 233.875019] [<c064b381>] rtnetlink_rcv+0x21/0x29 [ 233.875019] [<c06589e4>] netlink_unicast+0xf3/0x149 [ 233.875019] [<c0658c6e>] netlink_sendmsg+0x234/0x241 [ 233.875019] [<c0637fb3>] __sock_sendmsg+0x4a/0x53 [ 233.875019] [<c06387d1>] sock_sendmsg+0xbd/0xd3 [ 233.875019] [<c043ec03>] ? autoremove_wake_function+0x0/0x38 [ 233.875019] [<c043ec03>] ? autoremove_wake_function+0x0/0x38 [ 233.875019] [<c041ec7e>] ? need_resched+0x18/0x22 [ 233.875019] [<c052c0ab>] ? copy_from_user+0x3e/0x125 [ 233.875019] [<c063ed7a>] ? verify_iovec+0x45/0x73 [ 233.875019] [<c063892b>] sys_sendmsg+0x144/0x197 [ 233.875019] [<c0639420>] ? sys_recvmsg+0x173/0x180 [ 233.875019] [<c044123e>] ? enqueue_hrtimer+0xc6/0xd1 [ 233.875019] [<c06daeb9>] ? _spin_unlock_irqrestore+0x27/0x3d [ 233.875019] [<c0441b1e>] ? hrtimer_start+0x12f/0x149 [ 233.875019] [<c0441d10>] ? ktime_get+0x18/0x34 [ 233.875019] [<c0421eca>] ? hrtick_start_fair+0x11c/0x153 [ 233.875019] [<c042a410>] ? finish_task_switch+0x2f/0xb0 [ 233.875019] [<c06d9787>] ? schedule+0x6ee/0x70d [ 233.875019] [<c0639943>] sys_socketcall+0x16d/0x1a3 [ 233.875019] [<c0406fd5>] ? do_IRQ+0xb3/0xc9 [ 233.875019] [<c0404d32>] syscall_call+0x7/0xb [ 233.875019] ======================= [ 233.875019] handlers: [ 233.875019] [<c055131c>] (acpi_irq+0x0/0x28) [ 233.875019] [<f8937006>] (e1000_intr+0x0/0x13f [e1000]) [ 233.875019] Disabling IRQ #9 [ 233.949746] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 245.228855] [drm] Num pipes: 1 [ 245.408888] [drm:drm_release] *ERROR* Device busy: 1 [ 254.550941] [drm] Module unloaded [ 259.184939] e1000 0000:02:01.0: PCI INT A disabled [ 262.211114] Intel(R) PRO/1000 Network Driver - version 7.3.20-k3-NAPI [ 262.211163] Copyright (c) 1999-2006 Intel Corporation. [ 262.211284] e1000 0000:02:01.0: PCI INT A -> Link[LNKA] -> GSI 9 (level, low) -> IRQ 9 [ 262.460755] e1000: 0000:02:01.0: e1000_probe: (PCI:33MHz:32-bit) 00:0d:60:ca:c1:97 [ 262.490498] e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network Connection [ 266.560440] e1000: eth0: e1000_watchdog: NIC Link is Up 100 Mbps Full Duplex, Flow Control: RX/TX [ 266.561918] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 266.563835] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready [ 273.526951] [drm] Initialized drm 1.1.0 20060810 [ 273.595648] [drm] Initialized radeon 1.29.0 20080528 on minor 0 [ 275.065220] agpgart-intel 0000:00:00.0: AGP 2.0 bridge [ 275.065260] agpgart-intel 0000:00:00.0: putting AGP V2 device into 4x mode [ 275.065325] pci 0000:01:00.0: putting AGP V2 device into 4x mode [ 275.617217] [drm] Setting GART location based on new memory map [ 275.617245] [drm] Loading R300 Microcode [ 275.617311] [drm] Num pipes: 1 [ 275.617324] [drm] writeback test succeeded in 2 usecs [ 277.309063] irq 9: nobody cared (try booting with the "irqpoll" option) [ 277.309078] Pid: 3526, comm: X Tainted: G D 2.6.27-0.398.rc9.fc10.i686 #1 [ 277.309088] [<c0466a2e>] __report_bad_irq+0x33/0x74 [ 277.309106] [<c0466c42>] note_interrupt+0x1d3/0x225 [ 277.309118] [<c0466159>] ? handle_IRQ_event+0x61/0x69 [ 277.309131] [<c04672fe>] handle_level_irq+0x8d/0xc3 [ 277.309142] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 277.309155] [<c0406fc1>] do_IRQ+0x9f/0xc9 [ 277.309166] [<c0405710>] common_interrupt+0x28/0x30 [ 277.309179] [<c042007b>] ? sched_group_rt_runtime+0x7/0x50 [ 277.309194] [<c0431e28>] ? __do_softirq+0x6b/0x10f [ 277.309210] [<c0431dbd>] ? __do_softirq+0x0/0x10f [ 277.309223] [<c0407067>] do_softirq+0x7c/0xdd [ 277.309233] [<c0467271>] ? handle_level_irq+0x0/0xc3 [ 277.309245] [<c0431a80>] irq_exit+0x49/0x88 [ 277.309256] [<c0406fd5>] do_IRQ+0xb3/0xc9 [ 277.309266] [<c0405710>] common_interrupt+0x28/0x30 [ 277.309278] [<c04a007b>] ? __kill_fasync+0x50/0x52 [ 277.309292] [<c04a07a1>] ? vfs_ioctl+0x6d/0x6e [ 277.309305] [<c04a09f1>] ? do_vfs_ioctl+0x24f/0x262 [ 277.309318] [<c04733d3>] ? filemap_fdatawait+0x4c/0x50 [ 277.309333] [<c04b2c64>] ? do_fsync+0x63/0x78 [ 277.309347] [<c04a0a49>] sys_ioctl+0x45/0x60 [ 277.309359] [<c0404d32>] syscall_call+0x7/0xb [ 277.309370] ======================= [ 277.309375] handlers: [ 277.309380] [<c055131c>] (acpi_irq+0x0/0x28) [ 277.309393] [<f88dc006>] (e1000_intr+0x0/0x13f [e1000]) [ 277.309428] Disabling IRQ #9 [ 277.309491] eth0: no IPv6 routers present [ 348.756422] e1000: eth0: e1000_watchdog: NIC Link is Up 100 Mbps Full Duplex, Flow Control: RX/TX
We aren't sure if it is radeon drm or not. So this remains a problem.
Ignore this noise and bug: (II) RADEON(0): [drm] failure adding irq handler, there is a device already using that irq [drm] falling back to irq-free operation It is a radeon bug, sorry for the noise. But upstream has now confirmed they reproduced it also.
> there is a device already using that irq though it remains a bug (that radeon provokes interrupts w/ handling them) you'll likely be able to work around this by enabling the IOAPIC and seeing if radeon gets its own IRQ that way. > it *does* look like an ACPI bug Apparently ACPI turned out to be the innocent bystander on this one... I don't recall why I quickly asserted it was not, because i agree that the IO port write in the stack trace above suggests that ACPI could have provoked an interrupt. Perhaps it was some additional info on the list that isn't in this report...
The radeon issue was fixed by Dave Airlie. The drm driver was not handling interrupts and this is now fixed as of 2.6.27-3.fc10.i686. I tried enabling ioapic but that didn't work. But this is all moot now.