Created attachment 276549 [details] Output in the order of 4.16.15 error, lspci of device and pci slot, and 4.17.1 error Upgraded to kernel 4.17.1 yesterday on my Gentoo machine, from 4.16.15. On the 'old' 4.16.15 I encountered a repetitive error regarding iommu and my audigy 2 PCI sound card. Attached is a snippet of the errors and the lspci command to reference the product affecting it. This 4.16.15 error, by the way, did not affect performance of my system or audio card at all. On upgrading to 4.17.1, I no longer experienced this flood of errors in my dmesg, for which I was glad. However, once I used any program that output to the audigy sound card (pulseaudio, VLC through alsa, firefox built with alsa support, KDE's Juk, etc.) my machine would lock up for several minutes. Log files showed a different error from the 4.16 error. The error would appear only once in my log files for each time I tried to access any software that output to my audigy sound card. The system would temporarily lock up each time. I am running a linux host machine with, intermittently, a windows guest VM. I am passing a GFX card through to the VM with vfio, but NOT the audigy sound card. I would be glad to provide any other required info necessary, or test as needed. These results were with the "Gentoo-sources" kernel. If it is necessary that I re-test with vanilla kernels, I will need a bit more time for that. My kernel command line: "root=UUID=0084e716-7c34-4e80-a3ba-bd5a21922a69 intel_iommu=on vfio-pci.ids=10de:1b80,10de:10f0,8086:8cad enable_mtrr_cleanup mtrr_spare_reg_nr=1 default_hugepagesz=1GB hugepagesz=1GB hugepages=8 rootfstype=btrfs rootflags=compress=lzo ro quiet console=tty1 initrd=/boot/intel-uc.img"
Hello, The issue I described in the following report: <https://bugzilla.kernel.org/show_bug.cgi?id=199677> seems to be related to this. However this issue has been present for me for a very long time, not just in Kernel 4.17. I can remember this being a problem ever since I moved my system to the Ryzen platform. Apparently, from what I've read, the Audigy chip has a faulty DMA engine which only has an interface 31 bits wide as opposed to the intended 32 bits. Maybe the IOMMU issues could be related to that?
Hi, yes, I found a few bug reports in archives describing this fault with iommu and Audigy hardware. For lack of users with this combination, potential solutions seemed to go no-where. I've kept this *useful* sound card through multiple hardware upgrades and moves. When I explored KVM and iommu recently I began getting these errors spamming my logs. 4.17.1 was where it actually negatively impacted my system. Based on some generic advice across various forums, I (1) removed pulseaudio completely and re-built all my software to remove its support. (2) I switched over from gentoo's branded kernel sources to the provided vanilla kernel sources. Now I have been running vanilla-4.17.2, and I found some interesting messages. No warnings nor errors anymore. It seems https://github.com/torvalds/linux/commit/04f8773a3e980f60953e7aeb36ec6c2631e11f10 is addressing this bug. Though it seems to be implemented in 4.17, I am only seeing it work correctly in my current 4.17.2 build. The output is: Jul 01 12:59:33 [kernel] snd_emu10k1 0000:04:00.0: enabling device (0000 -> 0001) Jul 01 12:59:33 [kernel] snd_emu10k1 0000:04:00.0: non-passthrough IOMMU detected, widening DMA allocations
Yesterday, with a combination of Firefox and KDE's Juk open, under the 4.17.2 vanilla kernel, the system became temporarily unresponsive again. I received the same dmesg error as above, pasted here. Though the problem of iommu and Audigy hardware seems to be addressed with 4.17, I am getting continual interruptions to the desktop. [Thu Jul 5 15:07:14 2018] ------------[ cut here ]------------ [Thu Jul 5 15:07:14 2018] kernel BUG at drivers/iommu/intel-iommu.c:3670! [Thu Jul 5 15:07:14 2018] invalid opcode: 0000 [#4] SMP PTI [Thu Jul 5 15:07:14 2018] Modules linked in: joydev cmac rfcomm cfg80211 8021q bridge stp llc ipv6 bnep binfmt_misc snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi iTCO_wdt iTCO_vendor_support sidewinder it87 hwmon_vid vhost_net tun vhost tap vhba(O) btusb btrtl btbcm btintel bluetooth ecdh_generic coretemp rfkill hwmon snd_emu10k1 x86_pkg_temp_thermal input_leds kvm_intel hid_kensington uas snd_util_mem snd_ac97_codec kvm efi_pstore ac97_bus pcspkr snd_rawmidi efivars snd_seq_device firewire_ohci emu10k1_gp lpc_ich gameport mfd_core i2c_i801 snd_hda_intel snd_hda_codec snd_hda_core rtc_cmos snd_hwdep snd_pcm e1000e mei_me snd_timer mei ptp snd pps_core efivarfs virtio_pci virtio_blk virtio_balloon xts aes_x86_64 ecb cbc sha512_generic sha256_generic libiscsi scsi_transport_iscsi macvlan virtio_net [Thu Jul 5 15:07:14 2018] virtio_ring virtio fuse overlay xfs nfs lockd grace sunrpc jfs reiserfs ext4 jbd2 ext2 mbcache multipath linear raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx raid1 raid0 dm_snapshot dm_bufio dm_crypt dm_mirror dm_region_hash dm_log dm_mod dax hid_sony led_class hid_microsoft hid_logitech ff_memless sl811_hcd xhci_plat_hcd usb_storage aic94xx libsas scsi_transport_sas lpfc qla2xxx scsi_transport_fc megaraid_sas megaraid_mbox megaraid_mm megaraid aacraid 3w_9xxx 3w_xxxx atp870u dc395x qla1280 usbhid dmx3191d sym53c8xx gdth initio BusLogic arcmsr aic7xxx aic79xx scsi_transport_spi sr_mod cdrom sg ahci libahci firewire_core [Thu Jul 5 15:07:14 2018] CPU: 4 PID: 18187 Comm: juk Tainted: G D O 4.17.2-vaerc #2 [Thu Jul 5 15:07:14 2018] Hardware name: Gigabyte Technology Co., Ltd. Z97X-UD3H-BK/Z97X-UD3H-BK-CF, BIOS F7 04/21/2015 [Thu Jul 5 15:07:14 2018] RIP: 0010:intel_unmap+0xe5/0xf0 [Thu Jul 5 15:07:14 2018] RSP: 0018:ffff94c549507d30 EFLAGS: 00010246 [Thu Jul 5 15:07:14 2018] RAX: 0000000000000000 RBX: ffff90165cafe0a0 RCX: 00000000ffe10000 [Thu Jul 5 15:07:14 2018] RDX: 0000000000000000 RSI: 00000000ffe10000 RDI: ffff90165cafe0a0 [Thu Jul 5 15:07:14 2018] RBP: 00000000ffe10000 R08: 0000000000000000 R09: ffff90165cafe0a0 [Thu Jul 5 15:07:14 2018] R10: ffff9013936bafc0 R11: 0000000000000012 R12: 00000000ffe10000 [Thu Jul 5 15:07:14 2018] R13: ffff90128b440000 R14: 0000000000010000 R15: ffff901369cf1900 [Thu Jul 5 15:07:14 2018] FS: 00007f0f64373400(0000) GS:ffff90166fb00000(0000) knlGS:0000000000000000 [Thu Jul 5 15:07:14 2018] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [Thu Jul 5 15:07:14 2018] CR2: 00007f0f3010d528 CR3: 0000000028a58006 CR4: 00000000001606e0 [Thu Jul 5 15:07:14 2018] Call Trace: [Thu Jul 5 15:07:14 2018] intel_free_coherent+0x39/0x60 [Thu Jul 5 15:07:14 2018] snd_free_sgbuf_pages+0xbb/0x100 [snd_pcm] [Thu Jul 5 15:07:14 2018] ? snd_pcm_lib_free_pages+0x40/0x90 [snd_pcm] [Thu Jul 5 15:07:14 2018] snd_pcm_lib_free_pages+0x40/0x90 [snd_pcm] [Thu Jul 5 15:07:14 2018] snd_emu10k1_playback_hw_free+0xa4/0xb0 [snd_emu10k1] [Thu Jul 5 15:07:14 2018] snd_pcm_common_ioctl+0x735/0x9a0 [snd_pcm] [Thu Jul 5 15:07:14 2018] ? selinux_file_ioctl+0x114/0x1b0 [Thu Jul 5 15:07:14 2018] snd_pcm_ioctl+0x25/0x30 [snd_pcm] [Thu Jul 5 15:07:14 2018] do_vfs_ioctl+0x8a/0x5c0 [Thu Jul 5 15:07:14 2018] ? security_file_ioctl+0x27/0x40 [Thu Jul 5 15:07:14 2018] ksys_ioctl+0x6b/0x80 [Thu Jul 5 15:07:14 2018] __x64_sys_ioctl+0x11/0x20 [Thu Jul 5 15:07:14 2018] do_syscall_64+0x44/0xf0 [Thu Jul 5 15:07:14 2018] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [Thu Jul 5 15:07:14 2018] RIP: 0033:0x7f0f5e35a2c7 [Thu Jul 5 15:07:14 2018] RSP: 002b:00007fff9ecd9718 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [Thu Jul 5 15:07:14 2018] RAX: ffffffffffffffda RBX: 000056261d1102d0 RCX: 00007f0f5e35a2c7 [Thu Jul 5 15:07:14 2018] RDX: 000056261c343fd0 RSI: 0000000000004112 RDI: 0000000000000018 [Thu Jul 5 15:07:14 2018] RBP: 000056261c346600 R08: 0000000000000000 R09: 00000000000000a0 [Thu Jul 5 15:07:14 2018] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000018 [Thu Jul 5 15:07:14 2018] R13: 000056261c5019e0 R14: 000056261c57e148 R15: 000000000000001a [Thu Jul 5 15:07:14 2018] Code: 48 8d bb 18 03 00 00 48 89 ea 4c 89 ee e8 c4 88 ff ff 4c 89 f7 5b 5d 41 5c 41 5d 41 5e e9 a4 b9 ff ff 5b 5d 41 5c 41 5d 41 5e c3 <0f> 0b 66 0f 1f 84 00 00 00 00 00 41 56 41 55 41 54 49 89 fc 55 [Thu Jul 5 15:07:14 2018] RIP: intel_unmap+0xe5/0xf0 RSP: ffff94c549507d30 [Thu Jul 5 15:07:14 2018] ---[ end trace a5b45667e21a43b5 ]---
Hello, I finally got around to install the latest Kernel 4.17.9 on my Ubuntu 18.04 Ryzen machine and the fix you described seems to be fully working. No more errors in dmesg and glitch fre audio with IOMMU enabled. Are you still getting lockups on your desktop with this version? Another thing I had to do on my machine was to change the line load-module module-udev-detect to load-module module-udev-detect tsched=0 in /etc/pulse/default.pa to prevent certain applications from glitching the audio. This seems to be a pulseaudio issue in conjunction with the audigy driver. [ 0.742312] iommu: Adding device 0000:00:18.0 to group 11 [ 0.742326] iommu: Adding device 0000:00:18.1 to group 11 [ 0.742341] iommu: Adding device 0000:00:18.2 to group 11 [ 0.742354] iommu: Adding device 0000:00:18.3 to group 11 [ 0.742367] iommu: Adding device 0000:00:18.4 to group 11 [ 0.742379] iommu: Adding device 0000:00:18.5 to group 11 [ 0.742391] iommu: Adding device 0000:00:18.6 to group 11 [ 0.742404] iommu: Adding device 0000:00:18.7 to group 11 [ 0.742610] iommu: Adding device 0000:18:01.0 to group 12 [ 0.742619] iommu: Adding device 0000:18:01.1 to group 12 [ 0.742628] iommu: Adding device 0000:18:01.2 to group 12 [ 1.371728] firewire_ohci 0000:18:01.2: enabling device (0000 -> 0002) [ 1.432067] firewire_ohci 0000:18:01.2: added OHCI v1.10 device as card 0, 4 IR + 8 IT contexts, quirks 0x2 [ 1.948317] firewire_core 0000:18:01.2: created device fw0: GUID 00023c015112bde3, S400 [ 5.000560] Emu10k1_gameport 0000:18:01.1: enabling device (0000 -> 0001) [ 5.019063] gameport gameport0: EMU10K1 is pci0000:18:01.1/gameport0, io 0xe040, speed 411kHz [ 5.075601] snd_emu10k1 0000:18:01.0: enabling device (0000 -> 0001) [ 5.075671] snd_emu10k1 0000:18:01.0: non-passthrough IOMMU detected, widening DMA allocations [ 5.086655] snd_emu10k1 0000:18:01.0: Installing spdif_bug patch: SB Audigy 2 ZS [SB0350]
OK, I am glad that you're getting a positive result with that fix. I got around to building 4.17.9, upgrading without any relevant config changes. I tested various audio software (juk, vlc, audacity). The moment I opened audacity, the system froze, and dmesg spat out a similar error: [Thu Jul 26 08:06:53 2018] ------------[ cut here ]------------ [Thu Jul 26 08:06:53 2018] kernel BUG at drivers/iommu/intel-iommu.c:3670! [Thu Jul 26 08:06:53 2018] invalid opcode: 0000 [#1] SMP PTI [Thu Jul 26 08:06:53 2018] Modules linked in: cmac rfcomm cfg80211 8021q bridge stp llc ipv6 bnep binfmt_misc snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi sidewinder it87 hwmon_vid vhost_net tun vhost tap vhba(O) snd_seq_oss snd_emu10k1_synth snd_emux_synth snd_seq_midi_emul snd_seq_midi snd_seq_virmidi snd_seq_midi_event snd_seq coretemp hwmon iTCO_wdt iTCO_vendor_support btusb btrtl btbcm x86_pkg_temp_thermal btintel bluetooth ecdh_generic rfkill input_leds uas hid_kensington snd_emu10k1 kvm_intel snd_util_mem snd_hda_intel snd_ac97_codec snd_hda_codec kvm ac97_bus snd_hda_core efi_pstore snd_rawmidi snd_hwdep snd_seq_device snd_pcm e1000e pcspkr lpc_ich i2c_i801 efivars mei_me snd_timer mfd_core emu10k1_gp firewire_ohci gameport mei snd ptp pps_core rtc_cmos efivarfs virtio_pci virtio_blk virtio_balloon [Thu Jul 26 08:06:53 2018] xts aes_x86_64 ecb cbc sha512_generic sha256_generic libiscsi scsi_transport_iscsi macvlan virtio_net virtio_ring virtio fuse overlay xfs nfs lockd grace sunrpc jfs reiserfs ext4 jbd2 ext2 mbcache multipath linear raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx raid1 raid0 dm_snapshot dm_bufio dm_crypt dm_mirror dm_region_hash dm_log dm_mod dax hid_sony led_class hid_microsoft hid_logitech ff_memless sl811_hcd usbhid xhci_plat_hcd aic94xx libsas scsi_transport_sas lpfc qla2xxx scsi_transport_fc megaraid_sas megaraid_mbox megaraid_mm megaraid aacraid 3w_9xxx 3w_xxxx atp870u dc395x qla1280 dmx3191d sym53c8xx gdth initio BusLogic arcmsr aic7xxx aic79xx scsi_transport_spi sr_mod cdrom sg usb_storage ahci libahci firewire_core [Thu Jul 26 08:06:53 2018] CPU: 5 PID: 27730 Comm: audacity Tainted: P O 4.17.9-vaerc #2 [Thu Jul 26 08:06:53 2018] Hardware name: Gigabyte Technology Co., Ltd. Z97X-UD3H-BK/Z97X-UD3H-BK-CF, BIOS F7 04/21/2015 [Thu Jul 26 08:06:53 2018] RIP: 0010:intel_unmap+0xe5/0xf0 [Thu Jul 26 08:06:53 2018] RSP: 0018:ffffb27b60f3fd58 EFLAGS: 00010246 [Thu Jul 26 08:06:53 2018] RAX: 0000000000000000 RBX: 00000000ffe40000 RCX: 00000000ffe40000 [Thu Jul 26 08:06:53 2018] RDX: 0000000000000000 RSI: 00000000ffe40000 RDI: ffff9e251cb0c8a0 [Thu Jul 26 08:06:53 2018] RBP: ffff9e251cb0c8a0 R08: 0000000000000000 R09: ffff9e251cb0c8a0 [Thu Jul 26 08:06:53 2018] R10: fffff27e44e5ce40 R11: 00000000ffffffff R12: 0000000000020000 [Thu Jul 26 08:06:53 2018] R13: ffff9e21b7e00000 R14: 0000000000000000 R15: ffff9e25135a3378 [Thu Jul 26 08:06:53 2018] FS: 00007fb8c1597a00(0000) GS:ffff9e252fb40000(0000) knlGS:0000000000000000 [Thu Jul 26 08:06:53 2018] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [Thu Jul 26 08:06:53 2018] CR2: 00007fb8bab74d40 CR3: 00000001a50d8001 CR4: 00000000001606e0 [Thu Jul 26 08:06:53 2018] Call Trace: [Thu Jul 26 08:06:53 2018] intel_free_coherent+0x39/0x60 [Thu Jul 26 08:06:53 2018] snd_free_sgbuf_pages+0xac/0x100 [snd_pcm] [Thu Jul 26 08:06:53 2018] ? snd_pcm_lib_free_pages+0x42/0x90 [snd_pcm] [Thu Jul 26 08:06:53 2018] snd_pcm_lib_free_pages+0x42/0x90 [snd_pcm] [Thu Jul 26 08:06:53 2018] snd_emu10k1_playback_hw_free+0xa4/0xb0 [snd_emu10k1] [Thu Jul 26 08:06:53 2018] snd_pcm_release_substream.part.21+0x3f/0xa0 [snd_pcm] [Thu Jul 26 08:06:53 2018] snd_pcm_release+0x9c/0xa0 [snd_pcm] [Thu Jul 26 08:06:53 2018] __fput+0xa9/0x220 [Thu Jul 26 08:06:53 2018] task_work_run+0x79/0xa0 [Thu Jul 26 08:06:53 2018] exit_to_usermode_loop+0x85/0x90 [Thu Jul 26 08:06:53 2018] do_syscall_64+0xe2/0x100 [Thu Jul 26 08:06:53 2018] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [Thu Jul 26 08:06:53 2018] RIP: 0033:0x7fb8b9664ad7 [Thu Jul 26 08:06:53 2018] RSP: 002b:00007ffead5502a8 EFLAGS: 00000202 ORIG_RAX: 000000000000000b [Thu Jul 26 08:06:53 2018] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 00007fb8b9664ad7 [Thu Jul 26 08:06:53 2018] RDX: 0000000000000010 RSI: 0000000000001000 RDI: 00007fb8c14c1000 [Thu Jul 26 08:06:53 2018] RBP: 000056270106c780 R08: 0000000000080000 R09: 0000000000000002 [Thu Jul 26 08:06:53 2018] R10: 000056270103f220 R11: 0000000000000202 R12: 000056270106c780 [Thu Jul 26 08:06:53 2018] R13: 000056270106c780 R14: 0000000000000000 R15: 0000000000000000 [Thu Jul 26 08:06:53 2018] Code: 00 00 45 0f b6 c0 e8 db d2 ff ff 4c 89 ff 48 89 da 4c 89 f6 e8 dd 89 ff ff 4c 89 e7 5b 5d 41 5c 41 5d 41 5e 41 5f e9 7b ba ff ff <0f> 0b 66 0f 1f 84 00 00 00 00 00 41 56 41 55 41 54 49 89 fc 55 [Thu Jul 26 08:06:53 2018] RIP: intel_unmap+0xe5/0xf0 RSP: ffffb27b60f3fd58 [Thu Jul 26 08:06:53 2018] ---[ end trace 045288933ed40a51 ]--- For months, I have been pulseaudio free (rebuilt my system software to work directly with alsa. No negative consequences of this so far), so your udev solution won't work in this case. Currently as a workaround I am dual-booting (yuck!) to enable and disable IOMMU as I need it, so it will not interfere with my work when I am not using the VM, which kind of defeats the purpose of the VM. All my sessions with the IOMMU disabled are, of course, bug-free and smooth. Audacity opens without errors, and I can use it as normal.
Hello, I was troubled in a similar way as you are right now, when I couldn't use my system for IOMMU GPU passthrough at the same time as using my Audigy... luckily this seems to be on the way of getting fixed. I was honestly surprised to see a fix for this at all. Audacity works perfectly fine for me, even when recording directly through the card's ALSA (It defaults to the ALSA pulse device in my case.) What version of Audacity are you running? If I recall correctly, there used to be a bug in the program that would render the card completely useless on launch. Or rather a bug in the Audigy driver that was triggered by Audacity probing all devices on startup. One of them (I think the P16V device of the Audigy card) was/is just straight out dysfunctional and would glitch the audio chip until reboot. https://forum.audacityteam.org/viewtopic.php?f=48&t=72361 I'm running Audacity 2.2.1 , but guessing from the fact you're on gentoo you are probably running this version too, or higher. I haven't seen any IOMMU errors anymore ever since I upgraded so far (~a week) so maybe this could be platform related? Are you on the latest bios and is your intel processor microcode being updated by gentoo? Cheers
Right, my BIOS and processor are all getting their updates. My Audacity is 2.1.3. I recall having problems with that particular software a long time ago, but that was due to conflicts with JACK and early pulseaudio. Just to prove that this isn't isolated to Audacity, I ran Firefox 61 (built with ALSA support) and navigated to websites with audio. Amazon.com's music sampler worked fine, so did news sites, but when I went to a flash-based website, the system froze again with: [Thu Jul 26 15:11:18 2018] ------------[ cut here ]------------ [Thu Jul 26 15:11:18 2018] kernel BUG at drivers/iommu/intel-iommu.c:3670! [Thu Jul 26 15:11:18 2018] invalid opcode: 0000 [#1] SMP PTI [Thu Jul 26 15:11:18 2018] Modules linked in: cmac rfcomm cfg80211 8021q bridge stp llc ipv6 bnep binfmt_misc snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi iTCO_wdt iTCO_vendor_support sidewinder it87 hwmon_vid vhost_net tun vhost tap vhba(O) snd_seq_oss snd_emu10k1_synth snd_emux_synth snd_seq_midi_emul snd_seq_midi snd_seq_virmidi snd_seq_midi_event snd_seq coretemp hwmon x86_pkg_temp_thermal btusb kvm_intel btrtl btbcm btintel kvm bluetooth ecdh_generic rfkill input_leds efi_pstore uas hid_kensington snd_emu10k1 pcspkr i2c_i801 efivars snd_util_mem snd_hda_intel snd_ac97_codec snd_hda_codec ac97_bus rtc_cmos snd_hda_core snd_rawmidi lpc_ich snd_seq_device snd_hwdep mfd_core snd_pcm e1000e firewire_ohci emu10k1_gp gameport snd_timer mei_me ptp snd mei pps_core efivarfs virtio_pci virtio_blk virtio_balloon [Thu Jul 26 15:11:18 2018] xts aes_x86_64 ecb cbc sha512_generic sha256_generic libiscsi scsi_transport_iscsi macvlan virtio_net virtio_ring virtio fuse overlay xfs nfs lockd grace sunrpc jfs reiserfs ext4 jbd2 ext2 mbcache multipath linear raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx raid1 raid0 dm_snapshot dm_bufio dm_crypt dm_mirror dm_region_hash dm_log dm_mod dax hid_sony led_class hid_microsoft hid_logitech ff_memless sl811_hcd xhci_plat_hcd aic94xx libsas scsi_transport_sas lpfc qla2xxx scsi_transport_fc megaraid_sas megaraid_mbox megaraid_mm megaraid aacraid 3w_9xxx 3w_xxxx atp870u dc395x qla1280 dmx3191d sym53c8xx gdth usbhid initio BusLogic arcmsr aic7xxx aic79xx scsi_transport_spi sr_mod cdrom sg usb_storage ahci libahci firewire_core [Thu Jul 26 15:11:18 2018] CPU: 5 PID: 14157 Comm: plugin-containe Tainted: P O 4.17.9-vaerc #2 [Thu Jul 26 15:11:18 2018] Hardware name: Gigabyte Technology Co., Ltd. Z97X-UD3H-BK/Z97X-UD3H-BK-CF, BIOS F7 04/21/2015 [Thu Jul 26 15:11:18 2018] RIP: 0010:intel_unmap+0xe5/0xf0 [Thu Jul 26 15:11:18 2018] RSP: 0018:ffffac6241afbd28 EFLAGS: 00010246 [Thu Jul 26 15:11:18 2018] RAX: 0000000000000000 RBX: 00000000ffe40000 RCX: 00000000ffe40000 [Thu Jul 26 15:11:18 2018] RDX: 0000000000000000 RSI: 00000000ffe40000 RDI: ffffa0721cab78a0 [Thu Jul 26 15:11:18 2018] RBP: ffffa0721cab78a0 R08: 0000000000000000 R09: ffffa0721cab78a0 [Thu Jul 26 15:11:18 2018] R10: ffffa06eb031d880 R11: 00000000ffffffff R12: 0000000000020000 [Thu Jul 26 15:11:18 2018] R13: ffffa06e97720000 R14: 0000000000000000 R15: ffffa06e97569800 [Thu Jul 26 15:11:18 2018] FS: 00007f5611d02d40(0000) GS:ffffa0722fb40000(0000) knlGS:0000000000000000 [Thu Jul 26 15:11:18 2018] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [Thu Jul 26 15:11:18 2018] CR2: 00007ff8d73ce000 CR3: 00000000976d0004 CR4: 00000000001606e0 [Thu Jul 26 15:11:18 2018] Call Trace: [Thu Jul 26 15:11:18 2018] intel_free_coherent+0x39/0x60 [Thu Jul 26 15:11:18 2018] snd_free_sgbuf_pages+0xac/0x100 [snd_pcm] [Thu Jul 26 15:11:18 2018] ? snd_pcm_lib_free_pages+0x42/0x90 [snd_pcm] [Thu Jul 26 15:11:18 2018] snd_pcm_lib_free_pages+0x42/0x90 [snd_pcm] [Thu Jul 26 15:11:18 2018] snd_emu10k1_playback_hw_free+0xa4/0xb0 [snd_emu10k1] [Thu Jul 26 15:11:18 2018] snd_pcm_common_ioctl+0x72a/0x9b0 [snd_pcm] [Thu Jul 26 15:11:18 2018] ? selinux_file_ioctl+0x112/0x1b0 [Thu Jul 26 15:11:18 2018] snd_pcm_ioctl+0x25/0x30 [snd_pcm] [Thu Jul 26 15:11:18 2018] do_vfs_ioctl+0x8b/0x5e0 [Thu Jul 26 15:11:18 2018] ? security_file_ioctl+0x2d/0x50 [Thu Jul 26 15:11:18 2018] ksys_ioctl+0x6b/0x80 [Thu Jul 26 15:11:18 2018] __x64_sys_ioctl+0x11/0x20 [Thu Jul 26 15:11:18 2018] do_syscall_64+0x4a/0x100 [Thu Jul 26 15:11:18 2018] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [Thu Jul 26 15:11:18 2018] RIP: 0033:0x7f560a9e12c7 [Thu Jul 26 15:11:18 2018] RSP: 002b:00007ffdaf6b4d88 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [Thu Jul 26 15:11:18 2018] RAX: ffffffffffffffda RBX: 00007f55fee70dd0 RCX: 00007f560a9e12c7 [Thu Jul 26 15:11:18 2018] RDX: 00007f55fee0d3d0 RSI: 0000000000004112 RDI: 0000000000000015 [Thu Jul 26 15:11:18 2018] RBP: 00007f55fee0d000 R08: 0000000000000000 R09: 00007f55f0f3a9d0 [Thu Jul 26 15:11:18 2018] R10: 0000000000000529 R11: 0000000000000246 R12: 0000000000000015 [Thu Jul 26 15:11:18 2018] R13: 00007f56118ef6e0 R14: 0000000000000001 R15: 0000219934991000 [Thu Jul 26 15:11:18 2018] Code: 00 00 45 0f b6 c0 e8 db d2 ff ff 4c 89 ff 48 89 da 4c 89 f6 e8 dd 89 ff ff 4c 89 e7 5b 5d 41 5c 41 5d 41 5e 41 5f e9 7b ba ff ff <0f> 0b 66 0f 1f 84 00 00 00 00 00 41 56 41 55 41 54 49 89 fc 55 [Thu Jul 26 15:11:18 2018] RIP: intel_unmap+0xe5/0xf0 RSP: ffffac6241afbd28 [Thu Jul 26 15:11:18 2018] ---[ end trace 0c2b269566110aa6 ]--- Again, Audacity was NOT running. This might just mean that my flash plug-in (adobe-flash 30) is behaving similarly to Audacity, but that is a large leap in logic from my perspective. Thank you Johann, for your feedback!
No worries, Paul! :) The last thing I'd try out in your case would be to try a fresh install of a fairly recent distro with the most up to date kernel. (Maybe Fedora or Manjaro) (Of course on another hard disk if you have one spare) to rule out a hardware fault, because honestly I've never ever seen crashes this serious in my case. Hopefully it's just something related to your OS configuration... I'm not too familiar with Gentoo and how up to date its software is compared to bleeding edge, sadly. What specific type of Audigy do you have? Mine's an Audigy 2 ZS with the front panel, I'm not sure if it makes any difference, but yours seems to be a standard 2, not the ZS, right? I happen to have a spare one like this in another computer, so I could try tomorrow if I can reproduce this behavior with it.
That's a good idea. I'll give Fedora a shot soon with a spare 80Gig drive I've got here. I learned on that distro back when it was plain Red Hat. My Audigy's the ZS, same as yours, front panel and all. I can't speak for Gentoo's version progress relative to other distros.
Update: could not get Fedora to install properly, so installed Ubuntu. After install, I upgraded it to kernel 4.17.9. On reboot, I issued the intel_iommu command to the kernel boot line. Since pulseaudio starts immediately, I instantly got this similar error: [ +8.621028] ------------[ cut here ]------------ [ +0.000002] kernel BUG at drivers/iommu/intel-iommu.c:3670! [ +0.000007] invalid opcode: 0000 [#1] SMP PTI [ +0.000028] Modules linked in: rfcomm cmac bnep btrfs zstd_compress libcrc32c xor nls_iso8859_1 raid6_pq intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul gpio_ich btusb btrtl btbcm btintel ghash_clmulni_intel pcbc snd_soc_rt5640 aesni_intel snd_soc_rl6231 snd_soc_core aes_x86_64 bluetooth snd_compress crypto_simd cryptd snd_hda_codec_hdmi snd_emu10k1_synth snd_emux_synth snd_seq_midi_emul snd_seq_virmidi snd_hda_codec_realtek snd_hda_codec_generic snd_hda_intel snd_emu10k1 snd_hda_codec snd_pcm_dmaengine snd_util_mem snd_hda_core snd_ac97_codec glue_helper ecdh_generic snd_seq_midi snd_seq_midi_event snd_hwdep ac97_bus snd_rawmidi snd_pcm snd_seq joydev snd_seq_device intel_cstate input_leds emu10k1_gp intel_rapl_perf gameport hid_kensington [ +0.000295] snd_timer snd mei_me soundcore lpc_ich shpchp mei mac_hid acpi_pad sch_fq_codel parport_pc ppdev lp parport ip_tables x_tables autofs4 hid_logitech_hidpp uas usb_storage hid_logitech_dj hid_generic usbhid hid nouveau i915 mxm_wmi ttm i2c_algo_bit drm_kms_helper firewire_ohci syscopyarea sysfillrect sysimgblt firewire_core fb_sys_fops crc_itu_t ahci drm e1000e libahci wmi video [ +0.000160] CPU: 3 PID: 1449 Comm: alsa-sink-emu10 Not tainted 4.17.9-041709-generic #201807221110 [ +0.000039] Hardware name: Gigabyte Technology Co., Ltd. Z97X-UD3H-BK/Z97X-UD3H-BK-CF, BIOS F7 04/21/2015 [ +0.000052] RIP: 0010:intel_unmap+0x12a/0x140 [ +0.000023] RSP: 0018:ffff9b0ac3bc7ca8 EFLAGS: 00010246 [ +0.000028] RAX: 0000000000000000 RBX: 00000000ffe40000 RCX: 00000000ffe40000 [ +0.000034] RDX: 0000000000000000 RSI: 00000000ffe40000 RDI: ffff8f2f1c98e0a0 [ +0.000033] RBP: ffff9b0ac3bc7cd8 R08: 0000000000000000 R09: ffff8f2f1c98e0a0 [ +0.000033] R10: fffff36cd05158c0 R11: 0000000000000000 R12: ffff8f2f1c98e0a0 [ +0.000033] R13: 0000000000020000 R14: ffff8f2bab960000 R15: 0000000000000000 [ +0.000034] FS: 00007f55b2ffd700(0000) GS:ffff8f2f2fac0000(0000) knlGS:0000000000000000 [ +0.000039] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ +0.000028] CR2: 00005603f0c061b0 CR3: 000000041abf0004 CR4: 00000000001606e0 [ +0.000033] Call Trace: [ +0.000021] intel_free_coherent+0x63/0x70 [ +0.000033] snd_dma_free_pages+0xc8/0xf0 [snd_pcm] [ +0.000028] snd_free_sgbuf_pages+0xb9/0x110 [snd_pcm] [ +0.000030] ? __snd_util_mem_free+0x41/0x50 [snd_util_mem] [ +0.000032] snd_dma_free_pages+0x1c/0xf0 [snd_pcm] [ +0.000027] snd_pcm_lib_free_pages+0x4e/0xa0 [snd_pcm] [ +0.000037] snd_emu10k1_playback_hw_free+0xae/0xc0 [snd_emu10k1] [ +0.000033] snd_pcm_common_ioctl+0x5ec/0xb40 [snd_pcm] [ +0.000030] snd_pcm_ioctl+0x27/0x40 [snd_pcm] [ +0.000028] do_vfs_ioctl+0xa8/0x610 [ +0.000023] ? handle_mm_fault+0xe1/0x210 [ +0.000024] ksys_ioctl+0x67/0x90 [ +0.000021] __x64_sys_ioctl+0x1a/0x20 [ +0.000024] do_syscall_64+0x5a/0x110 [ +0.000024] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ +0.000025] RIP: 0033:0x7f55c22d35d7 [ +0.000023] RSP: 002b:00007f55b2ffcbe8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ +0.000035] RAX: ffffffffffffffda RBX: 00005603f09b4190 RCX: 00007f55c22d35d7 [ +0.000032] RDX: 00005603f0b9b7a0 RSI: 0000000000004112 RDI: 0000000000000034 [ +0.000032] RBP: 00005603f0c3d760 R08: 0000000000000002 R09: 00005603f0af9710 [ +0.000032] R10: 0000000000003fff R11: 0000000000000246 R12: 0000000000000034 [ +0.000033] R13: 0000000000000000 R14: 0000000000000000 R15: 00005603f0c56b40 [ +0.000033] Code: 84 e8 db 34 e8 ff e9 6f ff ff ff 4c 89 e1 48 89 da 4c 89 f6 4c 89 ef e8 f5 07 ff ff 48 83 c4 08 5b 41 5c 41 5d 41 5e 41 5f 5d c3 <0f> 0b 49 8b 54 24 10 eb b9 66 66 2e 0f 1f 84 00 00 00 00 00 66 [ +0.000101] RIP: intel_unmap+0x12a/0x140 RSP: ffff9b0ac3bc7ca8 [ +0.016826] ---[ end trace fb83226ca8f3efba ]--- Now, installing Audacity and running it provided the Audacity splash, but the program exits immediately. Same kernel error occurs. I won't waste room with another copy/paste unless anyone needs to see it again. Running Audacity from the console gives me a few GTK warnings, but nothing useful. Same errors occur. Before anyone suggests changing the pci slot the sound card is in, the motherboard has ONLY one pci slot. I doubt there is anything more I can do on my end unless a developer issues a change that I can test.
Hi, sorry to hear that this is still affecting you. Kernel 4.17.10 and 4.17.11 came out in the last couple of days, and both of them seem to have changes to the IOMMU subsystem in their changelogs. I can only encourage you to try out those newer ones, and hopefully there's some change in behaviour. On another note, your motherboard seems to have a certain PCIe to PCI chip from ITE to provide the legacy PCI slot, as PCI isn't provided by any modern chipset. Are there any settings inside the bios that you could play around with?
Well I took your advice to heart, dug into my BIOS, and changed one thing at a time between reboots - things only having to do with PCI. I even disabled the onboard audio just to be safe. The error still gets reported. I went as far as setting BIOS to defaults. No luck there. I then updated to vanilla kernel 4.17.11. As I was re-compiling some software, I had Juk playing music, then a system sound came on. As that happened the system froze. Once I could regain my desktop I checked dmesg. Sure enough I get: kernel BUG at drivers/iommu/intel-iommu.c:3669! in all its 70something lines of glory.
UPDATE I configured with my usual config options and installed kernel 4.18.4 Running audio is much stabler. I have a majority of functionality back with Audacity. I'm able to open, edit, and export audio properly. The ONLY time I got an error was when I tried to change the recording input source. The system slowed as before and this similar error came up in dmesg: [Thu Aug 23 16:29:45 2018] ------------[ cut here ]------------ [Thu Aug 23 16:29:45 2018] kernel BUG at drivers/iommu/intel-iommu.c:3701! [Thu Aug 23 16:29:45 2018] invalid opcode: 0000 [#1] SMP PTI [Thu Aug 23 16:29:45 2018] CPU: 2 PID: 12666 Comm: audacity Tainted: P O 4.18.4-vaerc #2 [Thu Aug 23 16:29:45 2018] Hardware name: Gigabyte Technology Co., Ltd. Z97X-UD3H-BK/Z97X-UD3H-BK-CF, BIOS F7 04/21/2015 [Thu Aug 23 16:29:45 2018] RIP: 0010:intel_unmap+0xe5/0xf0 [Thu Aug 23 16:29:45 2018] Code: 00 45 0f b6 c0 e8 1b d1 ff ff 4c 89 ff 48 89 da 4c 89 f6 e8 7d 86 ff ff 4c 89 e7 5b 5d 41 5c 41 5d 41 5e 41 5f e9 1b b8 ff ff <0f> 0b 66 0f 1f 84 00 00 00 00 00 41 56 41 55 41 54 49 89 fc 55 53 [Thu Aug 23 16:29:45 2018] RSP: 0018:ffffb76020877d10 EFLAGS: 00010246 [Thu Aug 23 16:29:45 2018] RAX: 0000000000000000 RBX: 00000000ffe40000 RCX: 00000000ffe40000 [Thu Aug 23 16:29:45 2018] RDX: 0000000000000000 RSI: 00000000ffe40000 RDI: ffff92439cef88a0 [Thu Aug 23 16:29:45 2018] RBP: ffff92439cef88a0 R08: 0000000000000000 R09: ffff92439cef88a0 [Thu Aug 23 16:29:45 2018] R10: fffff944c4492f80 R11: 00000000ffffffff R12: 0000000000020000 [Thu Aug 23 16:29:45 2018] R13: ffff9240924be440 R14: ffff924391ca0000 R15: ffff92439b7f8e20 [Thu Aug 23 16:29:45 2018] FS: 00007f8befb66a00(0000) GS:ffff9243afa80000(0000) knlGS:0000000000000000 [Thu Aug 23 16:29:45 2018] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [Thu Aug 23 16:29:45 2018] CR2: 00007f8befbc6008 CR3: 0000000127644006 CR4: 00000000001606e0 [Thu Aug 23 16:29:45 2018] Call Trace: [Thu Aug 23 16:29:45 2018] intel_free_coherent+0x53/0x70 [Thu Aug 23 16:29:45 2018] snd_free_sgbuf_pages+0xbb/0x120 [snd_pcm] [Thu Aug 23 16:29:45 2018] ? snd_pcm_lib_free_pages+0x42/0x90 [snd_pcm] [Thu Aug 23 16:29:45 2018] snd_pcm_lib_free_pages+0x42/0x90 [snd_pcm] [Thu Aug 23 16:29:45 2018] snd_emu10k1_efx_playback_hw_free+0xa3/0xb0 [snd_emu10k1] [Thu Aug 23 16:29:45 2018] snd_pcm_common_ioctl+0x230/0x9d0 [snd_pcm] [Thu Aug 23 16:29:45 2018] snd_pcm_ioctl+0x25/0x30 [snd_pcm] [Thu Aug 23 16:29:45 2018] do_vfs_ioctl+0x9f/0x620 [Thu Aug 23 16:29:45 2018] ksys_ioctl+0x6b/0x80 [Thu Aug 23 16:29:45 2018] __x64_sys_ioctl+0x11/0x20 [Thu Aug 23 16:29:45 2018] do_syscall_64+0x4a/0x100 [Thu Aug 23 16:29:45 2018] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [Thu Aug 23 16:29:45 2018] RIP: 0033:0x7f8be79dc2c7 [Thu Aug 23 16:29:45 2018] Code: 00 00 00 75 0c 48 c7 c0 ff ff ff ff 48 83 c4 18 c3 e8 9d db 01 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 91 4b 2c 00 f7 d8 64 89 01 48 [Thu Aug 23 16:29:45 2018] RSP: 002b:00007ffcbee51828 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [Thu Aug 23 16:29:45 2018] RAX: ffffffffffffffda RBX: 0000561f71ccf920 RCX: 00007f8be79dc2c7 [Thu Aug 23 16:29:45 2018] RDX: 00007f8be7ca1c80 RSI: 0000000000004112 RDI: 000000000000000a [Thu Aug 23 16:29:45 2018] RBP: 0000561f71919000 R08: 0000000000080000 R09: 0000000000000010 [Thu Aug 23 16:29:45 2018] R10: 0000561f71c912e0 R11: 0000000000000246 R12: 000000000000000a [Thu Aug 23 16:29:45 2018] R13: 0000000000000010 R14: 0000561f71b20ee0 R15: 0000000000000001 [Thu Aug 23 16:29:45 2018] Modules linked in: cmac rfcomm cfg80211 8021q bridge stp llc ipv6 bnep binfmt_misc hid_logitech_hidpp sidewinder it87 hwmon_vid vhost_net tun vhost tap vhba(O) snd_seq_oss snd_emu10k1_synth snd_emux_synth snd_seq_midi_emul snd_seq_midi snd_seq_virmidi snd_seq_midi_event joydev snd_seq iTCO_wdt iTCO_vendor_support coretemp hwmon x86_pkg_temp_thermal btusb btrtl btbcm btintel bluetooth ecdh_generic kvm_intel rfkill input_leds hid_kensington efi_pstore kvm snd_emu10k1 snd_hda_codec_hdmi hid_logitech_dj xpad snd_util_mem snd_ac97_codec emu10k1_gp ac97_bus snd_rawmidi pcspkr gameport firewire_ohci efivars snd_seq_device i2c_i801 snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep mei_me snd_pcm e1000e mei lpc_ich snd_timer mfd_core ptp snd pps_core rtc_cmos efivarfs virtio_pci virtio_blk virtio_balloon [Thu Aug 23 16:29:45 2018] xts aes_x86_64 ecb cbc sha512_generic sha256_generic libiscsi scsi_transport_iscsi macvlan virtio_net net_failover failover virtio_ring virtio fuse overlay xfs nfs lockd grace sunrpc jfs reiserfs ext4 jbd2 ext2 mbcache multipath linear raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx raid1 raid0 dm_snapshot dm_bufio dm_crypt dm_mirror dm_region_hash dm_log dm_mod dax hid_sony hid_microsoft hid_logitech sl811_hcd xhci_plat_hcd usb_storage aic94xx libsas scsi_transport_sas lpfc qla2xxx scsi_transport_fc megaraid_sas megaraid_mbox megaraid_mm megaraid aacraid 3w_9xxx 3w_xxxx atp870u dc395x qla1280 dmx3191d sym53c8xx gdth initio BusLogic arcmsr aic7xxx aic79xx scsi_transport_spi sr_mod cdrom sg ff_memless led_class usbhid ahci libahci firewire_core [Thu Aug 23 16:29:45 2018] ---[ end trace dd0337d2c9ac51c4 ]--- [Thu Aug 23 16:30:45 2018] RIP: 0010:intel_unmap+0xe5/0xf0 [Thu Aug 23 16:30:45 2018] Code: 00 45 0f b6 c0 e8 1b d1 ff ff 4c 89 ff 48 89 da 4c 89 f6 e8 7d 86 ff ff 4c 89 e7 5b 5d 41 5c 41 5d 41 5e 41 5f e9 1b b8 ff ff <0f> 0b 66 0f 1f 84 00 00 00 00 00 41 56 41 55 41 54 49 89 fc 55 53 [Thu Aug 23 16:30:45 2018] RSP: 0018:ffffb76020877d10 EFLAGS: 00010246 [Thu Aug 23 16:30:45 2018] RAX: 0000000000000000 RBX: 00000000ffe40000 RCX: 00000000ffe40000 [Thu Aug 23 16:30:45 2018] RDX: 0000000000000000 RSI: 00000000ffe40000 RDI: ffff92439cef88a0 [Thu Aug 23 16:30:45 2018] RBP: ffff92439cef88a0 R08: 0000000000000000 R09: ffff92439cef88a0 [Thu Aug 23 16:30:45 2018] R10: fffff944c4492f80 R11: 00000000ffffffff R12: 0000000000020000 [Thu Aug 23 16:30:45 2018] R13: ffff9240924be440 R14: ffff924391ca0000 R15: ffff92439b7f8e20 [Thu Aug 23 16:30:45 2018] FS: 00007f8befb66a00(0000) GS:ffff9243afa40000(0000) knlGS:0000000000000000 [Thu Aug 23 16:30:45 2018] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [Thu Aug 23 16:30:45 2018] CR2: 0000561943eb8da8 CR3: 0000000127644004 CR4: 00000000001606e0 Again, much stabler, and it works for what I currently need, but it feels like this is an "almost but not quite" scenario, meaning if I tried to cause the error, or if I inadvertently issued a command through audacity, I could end up with this same error. Paul