| Summary: | [NVE7] Nouveau read fault, locking up the gpu | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | xorg | Reporter: | exi+freedesktop | ||||||||||
| Component: | Driver/nouveau | Assignee: | Nouveau Project <nouveau> | ||||||||||
| Status: | RESOLVED DUPLICATE | QA Contact: | Xorg Project Team <xorg-team> | ||||||||||
| Severity: | normal | ||||||||||||
| Priority: | medium | CC: | davidmarch007, fountain.ge, jonrysh, rockowitz | ||||||||||
| Version: | unspecified | ||||||||||||
| Hardware: | x86-64 (AMD64) | ||||||||||||
| OS: | Linux (All) | ||||||||||||
| Whiteboard: | |||||||||||||
| i915 platform: | i915 features: | ||||||||||||
| Attachments: |
|
||||||||||||
Created attachment 116800 [details]
complete yournal output including dmesg/kernel messages
I just got this bug again freezing up my screen.
Just before restarting the computer via ssh I saw a "gpu lockup switching to..." message on the screen.
I hit this bug frequently, usually after some minor user action like closing a window. When it happens the screen freezes except for the cursor; when I look at a console via ctrl/f2 and wait for a while (about 2 minutes) I see messages like: [20235.667444] nouveau E[plasmashell][16831] fail set_domain [20235.667782] nouveau E[plasmashell][16831] validating bo list [20235.668235] nouveau E[plasmachell][16831] valicate: -22 As soon as these messages appear, the desktop becomes mostly usable, but often strange; the icons become weird and some system messages are corrupted. In a following comment (after I get the desktop back), I will attach a snapshot of the weird desktop and today's journal file. The system is difficult to use with this bug. System is Amd-64 4-processor hardware Nvidia 9500GT Graphics Card Fedora-22 with all updates KDE Frameworks 5.12.0 Xorg -version X.Org X Server 1.17.2 Release Date: 2015-06-16 X Protocol Version 11, Revision 0 Build Operating System: 4.0.4-202.fc21.x86_64 Current Operating System: Linux amito 4.1.4-200.fc22.x86_64 #1 SMP Tue Aug 4 03:22:33 UTC 2015 x86_64 Kernel command line: BOOT_IMAGE=/vmlinuz-4.1.4-200.fc22.x86_64 root=/dev/mapper/fedora00-root ro rd.lvm.lv=fedora00/swap rd.lvm.lv=fedora00/root rhgb quiet LANG=en_US.UTF-8 Build Date: 15 July 2015 08:16:41AM Build ID: xorg-x11-server 1.17.2-2.fc22 Current version of pixman: 0.32.6 Created attachment 117741 [details]
System Journal for 2015-08-17
Created attachment 117742 [details]
Desktop as it should be
This is a part of my desktop when displayed properly
Created attachment 117743 [details]
Corrupt desktop
This is the same part of my desktop when displayed when the graphics system is in its "weird" state"
Probable duplicate bug (though this current bug was filed earlier) with more information: https://bugs.freedesktop.org/show_bug.cgi?id=91598 Apparently per above "fail set_domain" means the it is running out of vram. I may have made some headway on this issue in bug 92504 (comment 20). Please see if the patch in that comment helps. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
This just happened to me after a suspend/resume cycle. I cannot reproduce the read fault, it happened right after opening a plasma context menu but i still wanted it documented: Kernel: Linux bugbox 4.0.5-1-ARCH #1 SMP PREEMPT Sat Jun 6 18:37:49 CEST 2015 x86_64 GNU/Linux Xorg -version: X.Org X Server 1.17.2 Release Date: 2015-06-16 X Protocol Version 11, Revision 0 Build Operating System: Linux 4.0.4-2-ARCH x86_64 Current Operating System: Linux bugbox 4.0.5-1-ARCH #1 SMP PREEMPT Sat Jun 6 18:37:49 CEST 2015 x86_64 Kernel command line: BOOT_IMAGE=/boot/vmlinuz-linux root=UUID=073feb51-816c-4a86-8de3-5e9ed3145571 rw resume=UUID=ba1b46a0-89f5-4ddd-adc0-5982d576f73c libahci.ignore_sss=1 swapaccount=1 quiet Build Date: 16 June 2015 05:24:27PM Current version of pixman: 0.32.6 Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. Mesa: mesa 10.6.0-1 Nouveau: xf86-video-nouveau 1.0.11-3 Dmesg error messages: Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] suspending console... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] suspending display... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] evicting buffers... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] waiting for kernel channels to go idle... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] suspending client object trees... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] suspending kernel object tree... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] re-enabling device... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] resuming kernel object tree... Jun 27 13:19:35 bugbox kernel: nouveau [ VBIOS][0000:01:00.0] running init tables Jun 27 13:19:35 bugbox kernel: nouveau [ VOLT][0000:01:00.0] GPU voltage: 875000uv Jun 27 13:19:35 bugbox kernel: nouveau [ PTHERM][0000:01:00.0] fan management: automatic Jun 27 13:19:35 bugbox kernel: nouveau [ CLK][0000:01:00.0] --: core 405 MHz memory 648 MHz Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] resuming client object trees... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] resuming display... Jun 27 13:19:35 bugbox kernel: nouveau [ DRM] resuming console... Jun 27 13:19:35 bugbox kernel: Modules linked in: xt_multiport ipt_REJECT nf_reject_ipv4 xt_comment hid_lenovo rfcomm ecb bnep hid_apple nouveau mxm_wmi ttm drm_kms_helper drm i2c_algo_bit veth tun fuse ctr ccm btrfs xor ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_nat_ipv4 nf_nat xt_tcpudp nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack iptable_filter snd_hda_codec_hdmi bridge stp llc raid6_pq snd_hda_codec_realtek joydev mousedev snd_hda_codec_generic psmouse serio_raw nls_iso8859_1 atkbd nls_cp437 iTCO_wdt hid_generic iTCO_vendor_support libps2 vfat fat intel_rapl iosf_mbi uvcvideo x86_pkg_temp_thermal arc4 videobuf2_vmalloc intel_powerclamp usbhid videobuf2_memops btusb coretemp videobuf2_core v4l2_common videodev hid kvm_intel bluetooth media rtl8192ce rtl_pci kvm rtl8192c_common crct10dif_pclmul Jun 27 13:19:38 bugbox kernel: nouveau E[plasmashell[11934]] fail set_domain Jun 27 13:19:38 bugbox kernel: nouveau E[plasmashell[11934]] validating bo list Jun 27 13:19:38 bugbox kernel: nouveau E[plasmashell[11934]] validate: -22 Jun 27 13:19:38 bugbox kernel: nouveau E[plasmashell[11934]] fail set_domain Jun 27 13:19:38 bugbox kernel: nouveau E[plasmashell[11934]] validating bo list Jun 27 13:19:38 bugbox kernel: nouveau E[plasmashell[11934]] validate: -22 Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] fail set_domain Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] validating bo list Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] validate: -22 Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] fail set_domain Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] validating bo list Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] validate: -22 Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] fail set_domain Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] validating bo list Jun 27 13:31:13 bugbox kernel: nouveau E[plasmashell[11934]] validate: -22 Jun 27 13:31:13 bugbox kernel: nouveau E[ PFIFO][0000:01:00.0] read fault at 0x0000ae1000 [PTE] from GR/GPC0/T1_0 on channel 0x007f74b000 [plasmashell[11934]] Jun 27 13:31:13 bugbox kernel: nouveau E[ PFIFO][0000:01:00.0] PGR engine fault on channel 8, recovering... Jun 27 13:31:13 bugbox kernel: nouveau E[ PGR][0000:01:00.0] TRAP ch 8 [0x007f74b000 plasmashell[11934]] Jun 27 13:31:13 bugbox kernel: nouveau E[ PGR][0000:01:00.0] GPC0/TPC0/TEX: 0x80000049 Jun 27 13:32:57 bugbox kernel: nouveau E[plasmashell[11934]] failed to idle channel 0xcccc0000 [plasmashell[11934]] Jun 27 13:33:12 bugbox kernel: nouveau E[plasmashell[11934]] failed to idle channel 0xcccc0000 [plasmashell[11934]] The "validate bo list" errors happen on every suspend/resume, so i can reproduce them. The read fault and "failed to idle channel" error happened the first time and froze my screen.