[Bug 1086747] New: GPU HANG: ecode 4:0:0x7d65fafd, in Xorg [1618], reason: Hang on render ring, action: reset
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747 Bug ID: 1086747 Summary: GPU HANG: ecode 4:0:0x7d65fafd, in Xorg [1618], reason: Hang on render ring, action: reset Classification: openSUSE Product: openSUSE Distribution Version: Leap 42.3 Hardware: Other OS: Other Status: NEW Severity: Major Priority: P5 - None Component: X.Org Assignee: xorg-maintainer-bugs@forge.provo.novell.com Reporter: lnussel@suse.com QA Contact: xorg-maintainer-bugs@forge.provo.novell.com CC: tiwai@suse.com Found By: --- Blocker: --- After an upgrade from 42.2 to 42.3 the GPU locks up in X. Both with and without drm-kmp-default. See attached error dups from /sys/class/drm/card0/error # hwinfo --gfxcard 09: PCI 02.0: 0300 VGA compatible controller (VGA) [Created at pci.378] Unique ID: _Znp.2DyqmuGr1S3 SysFS ID: /devices/pci0000:00/0000:00:02.0 SysFS BusID: 0000:00:02.0 Hardware Class: graphics card Model: "Intel 965Q" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x2992 "965Q" SubVendor: pci 0x8086 "Intel Corporation" SubDevice: pci 0x4f43 Revision: 0x02 Driver: "i915" Driver Modules: "i915" Memory Range: 0xe0300000-0xe03fffff (rw,non-prefetchable) Memory Range: 0xd0000000-0xdfffffff (ro,non-prefetchable) I/O Ports: 0x2468-0x246f (rw) IRQ: 29 (77 events) I/O Ports: 0x3c0-0x3df (rw) Module Alias: "pci:v00008086d00002992sv00008086sd00004F43bc03sc00i00" Driver Info #0: XFree86 v4 Server Module: intel Driver Info #1: XFree86 v4 Server Module: intel 3D Support: yes Extensions: dri Config Status: cfg=new, avail=yes, need=no, active=unknown 10: PCI 02.1: 0380 Display controller [Created at pci.378] Unique ID: ruGf.ZfJTgkZVJ39 SysFS ID: /devices/pci0000:00/0000:00:02.1 SysFS BusID: 0000:00:02.1 Hardware Class: graphics card Model: "Intel 82Q963/Q965 Integrated Graphics Controller" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x2993 "82Q963/Q965 Integrated Graphics Controller" SubVendor: pci 0x8086 "Intel Corporation" SubDevice: pci 0x4f43 Revision: 0x02 Memory Range: 0xe0200000-0xe02fffff (rw,non-prefetchable) Module Alias: "pci:v00008086d00002993sv00008086sd00004F43bc03sc80i00" Config Status: cfg=new, avail=yes, need=no, active=unknown Primary display adapter: #9 kelvin:~ # hwinfo --gfxcard 09: PCI 02.0: 0300 VGA compatible controller (VGA) [Created at pci.378] Unique ID: _Znp.2DyqmuGr1S3 SysFS ID: /devices/pci0000:00/0000:00:02.0 SysFS BusID: 0000:00:02.0 Hardware Class: graphics card Model: "Intel 965Q" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x2992 "965Q" SubVendor: pci 0x8086 "Intel Corporation" SubDevice: pci 0x4f43 Revision: 0x02 Driver: "i915" Driver Modules: "i915" Memory Range: 0xe0300000-0xe03fffff (rw,non-prefetchable) Memory Range: 0xd0000000-0xdfffffff (ro,non-prefetchable) I/O Ports: 0x2468-0x246f (rw) IRQ: 29 (77 events) I/O Ports: 0x3c0-0x3df (rw) Module Alias: "pci:v00008086d00002992sv00008086sd00004F43bc03sc00i00" Driver Info #0: XFree86 v4 Server Module: intel Driver Info #1: XFree86 v4 Server Module: intel 3D Support: yes Extensions: dri Config Status: cfg=new, avail=yes, need=no, active=unknown 10: PCI 02.1: 0380 Display controller [Created at pci.378] Unique ID: ruGf.ZfJTgkZVJ39 SysFS ID: /devices/pci0000:00/0000:00:02.1 SysFS BusID: 0000:00:02.1 Hardware Class: graphics card Model: "Intel 82Q963/Q965 Integrated Graphics Controller" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x2993 "82Q963/Q965 Integrated Graphics Controller" SubVendor: pci 0x8086 "Intel Corporation" SubDevice: pci 0x4f43 Revision: 0x02 Memory Range: 0xe0200000-0xe02fffff (rw,non-prefetchable) Module Alias: "pci:v00008086d00002993sv00008086sd00004F43bc03sc80i00" Config Status: cfg=new, avail=yes, need=no, active=unknown Primary display adapter: #9 -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c1
--- Comment #1 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c2
--- Comment #2 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c3
Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c4
--- Comment #4 from Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c5
--- Comment #5 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c6
Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c7
--- Comment #7 from Stefan Dirsch
It's a Gen4 machine, and IIRC there was a point before which resetting an Intel GPU didn't really work due to a hardware bug or something.
(adding Stefan in case he remembers more)
Yes, GPU hardware resets never worked. ;-) -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c8
Stefan Dirsch
(In reply to Max Staudt from comment #6)
It's a Gen4 machine, and IIRC there was a point before which resetting an Intel GPU didn't really work due to a hardware bug or something.
(adding Stefan in case he remembers more)
Yes, GPU hardware resets never worked. ;-)
I mean GPU resets on older Intel GPUs. Ludwig, is this a desktop machine? Maybe a development machine even we received from Intel directly? It could be a machine with beta/alpha CPU/GPU revision still. In that case I would like to close this bugreport as INVALID. Seriously. ;-) -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c9
--- Comment #9 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c10
Stefan Dirsch
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c11
--- Comment #11 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c12
--- Comment #12 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c13
--- Comment #13 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c14
--- Comment #14 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
Stefan Dirsch
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
Stefan Dirsch
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c15
Stefan Dirsch
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c16
--- Comment #16 from Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c17
--- Comment #17 from Takashi Iwai
I've reproduced this, and both solutions that I've suggested resolve the issue when I try to start an X server manually.
It's good that we can reproduce locally. Did you test with intel driver, or modesetting? The original report seems with the former, so I wonder whether modesetting is stabler in this regard. And we know that old Intel chipsets had often problem even with SNA. In SLE11 era, we had to switch temporarily to UXA on Preload images for stabilization... -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c18
--- Comment #18 from Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c19
--- Comment #19 from Ludwig Nussel
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c20
--- Comment #20 from Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c23
Max Staudt
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747
http://bugzilla.opensuse.org/show_bug.cgi?id=1086747#c24
--- Comment #24 from Ludwig Nussel
participants (1)
-
bugzilla_noreply@novell.com