It appears that on my GeForce 7600 GS (G73), nouveau is doing DMA to/from unexpected addresses. This gives a lot of warnings, and eventually a system freeze. This is similar to https://bugs.freedesktop.org/show_bug.cgi?id=27063 / https://bugzilla.redhat.com/show_bug.cgi?id=561267. This was fixed for nv50 in http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=4eb3033c72099fab3536ed8ac54a5dc99f0832d7 but with the G73 I am seeing this problem in a 3.10 kernel. [ 1.499379] [drm] Initialized drm 1.1.0 20060810 [ 1.504059] MXM: GUID detected in BIOS [ 1.508111] nouveau [ DEVICE][0000:01:00.0] BOOT0 : 0x04b200a2 [ 1.514207] nouveau [ DEVICE][0000:01:00.0] Chipset: G73 (NV4B) [ 1.520301] nouveau [ DEVICE][0000:01:00.0] Family : NV40 [ 1.527097] nouveau [ VBIOS][0000:01:00.0] checking PRAMIN for image... [ 1.575254] nouveau [ VBIOS][0000:01:00.0] ... appears to be valid [ 1.581696] nouveau [ VBIOS][0000:01:00.0] using image from PRAMIN [ 1.588216] nouveau [ VBIOS][0000:01:00.0] BIT signature found [ 1.594305] nouveau [ VBIOS][0000:01:00.0] version 05.73.22.16.02 [ 1.600858] nouveau [ PFB][0000:01:00.0] RAM type: DDR2 [ 1.606516] nouveau [ PFB][0000:01:00.0] RAM size: 256 MiB [ 1.612436] nouveau [ PFB][0000:01:00.0] ZCOMP: 379904 tags [ 1.652901] nouveau [ PTHERM][0000:01:00.0] FAN control: toggle [ 1.658996] nouveau [ PTHERM][0000:01:00.0] fan management: disabled [ 1.665521] nouveau [ PTHERM][0000:01:00.0] internal sensor: yes [ 1.691750] [TTM] Zone kernel: Available graphics memory: 16478742 kiB [ 1.698366] [TTM] Zone dma32: Available graphics memory: 2097152 kiB [ 1.704892] [TTM] Initializing pool allocator [ 1.709254] [TTM] Initializing DMA pool allocator [ 1.713964] mtrr: type mismatch for 80000000,10000000 old: write-back new: write-combining [ 1.722226] nouveau [ DRM] VRAM: 251 MiB [ 1.726584] nouveau [ DRM] GART: 512 MiB [ 1.730945] nouveau [ DRM] TMDS table version 1.1 [ 1.736082] nouveau W[ DRM] TMDS table script pointers not stubbed [ 1.742608] nouveau [ DRM] DCB version 3.0 [ 1.747143] nouveau [ DRM] DCB outp 00: 01000300 00000028 [ 1.752974] nouveau [ DRM] DCB outp 01: 03000302 00000000 [ 1.758807] nouveau [ DRM] DCB outp 02: 04011310 00000028 [ 1.764641] nouveau [ DRM] DCB outp 03: 04011312 00c00000 [ 1.770471] nouveau [ DRM] DCB outp 04: 020223f1 0000c080 [ 1.776304] nouveau [ DRM] DCB conn 00: 1030 [ 1.781037] nouveau [ DRM] DCB conn 01: 2130 [ 1.785765] nouveau [ DRM] DCB conn 02: 0210 [ 1.790492] nouveau [ DRM] DCB conn 03: 0211 [ 1.795224] nouveau [ DRM] DCB conn 04: 0213 [ 1.801402] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010). [ 1.808018] [drm] No driver support for vblank timestamp query. [ 1.813978] nouveau [ DRM] 0xD1A1: Parsing digital output script table [ 1.870947] nouveau [ DRM] 0xD1F1: Parsing digital output script table [ 1.928298] nouveau [ DRM] 1 available performance level(s) [ 1.934308] nouveau [ DRM] 0: core 400MHz shader 400MHz memory 400MHz voltage 1050mV fanspeed 100% [ 1.943694] nouveau [ DRM] c: core 400MHz shader 400MHz memory 405MHz fanspeed 100% [ 1.955321] nouveau [ DRM] MM: using M2MF for buffer copies [ 1.961340] nouveau [ DRM] Setting dpms mode 3 on TV encoder (output 4) [ 2.060713] nouveau [ DRM] allocated 1920x1200 fb: 0x9000, bo ffff8808528bf800 [ 2.068437] fbcon: nouveaufb (fb0) is primary device [ 2.078765] nouveau [ DRM] 0xD1A1: Parsing digital output script table [ 2.139034] nouveau [ DRM] 0xD1F1: Parsing digital output script table [ 2.189610] dmar: DRHD: handling fault status reg 2 [ 2.189612] dmar: DMAR:[DMA Read] Request device [01:00.0] fault addr 0 [ 2.189612] DMAR:[fault reason 06] PTE Read access is not set [ 2.190017] dmar: DRHD: handling fault status reg 102 [ 2.190019] dmar: DMAR:[DMA Read] Request device [01:00.0] fault addr 0 [ 2.190019] DMAR:[fault reason 06] PTE Read access is not set [ 2.190120] Console: switching to colour frame buffer device 240x75 [ 2.190254] dmar: DRHD: handling fault status reg 202 [ 2.190255] dmar: DMAR:[DMA Read] Request device [01:00.0] fault addr 0 [ 2.190255] DMAR:[fault reason 06] PTE Read access is not set [ 2.190672] dmar: DRHD: handling fault status reg 302 [ 2.190674] dmar: DMAR:[DMA Read] Request device [01:00.0] fault addr 0 [ 2.190674] DMAR:[fault reason 06] PTE Read access is not set [ 2.191081] dmar: DRHD: handling fault status reg 402 [ 2.191083] dmar: DMAR:[DMA Read] Request device [01:00.0] fault addr 0 [ 2.191083] DMAR:[fault reason 06] PTE Read access is not set [ 2.203182] dmar: DRHD: handling fault status reg 502 [ 2.203184] dmar: DMAR:[DMA Read] Request device [01:00.0] fault addr 0 [ 2.203184] DMAR:[fault reason 06] PTE Read access is not set [ 2.321167] nouveau 0000:01:00.0: fb0: nouveaufb frame buffer device [ 2.327527] nouveau 0000:01:00.0: registered panic notifier [ 2.333109] [drm] Initialized nouveau 1.1.1 20120801 for 0000:01:00.0 on minor 0 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation G73 [GeForce 7600 GS] [10de:0392] (rev a1) (prog-if 00 [VGA controller]) Subsystem: Micro-Star International Co., Ltd. NX7600GS-T2D256EH [1462:0622] Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort+ >SERR- <PERR- INTx- Latency: 0, Cache Line Size: 64 bytes Interrupt: pin A routed to IRQ 32 Region 0: Memory at 91000000 (32-bit, non-prefetchable) [size=16M] Region 1: Memory at 80000000 (64-bit, prefetchable) [size=256M] Region 3: Memory at 90000000 (64-bit, non-prefetchable) [size=16M] Region 5: I/O ports at e000 [size=128] Expansion ROM at 92000000 [disabled] [size=128K] Capabilities: [60] Power Management version 2 Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-) Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME- Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+ Address: 0000000000000000 Data: 0000 Capabilities: [78] Express (v1) Endpoint, MSI 00 DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <256ns, L1 <4us ExtTag- AttnBtn- AttnInd- PwrInd- RBE- FLReset- DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported- RlxdOrd- ExtTag- PhantFunc- AuxPwr- NoSnoop+ MaxPayload 128 bytes, MaxReadReq 512 bytes DevSta: CorrErr- UncorrErr- FatalErr- UnsuppReq- AuxPwr- TransPend- LnkCap: Port #0, Speed 2.5GT/s, Width x16, ASPM L0s L1, Latency L0 <256ns, L1 <4us ClockPM- Surprise- LLActRep- BwNot- LnkCtl: ASPM Disabled; RCB 128 bytes Disabled- Retrain- CommClk+ ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt- LnkSta: Speed 2.5GT/s, Width x8, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt- Capabilities: [100 v1] Virtual Channel Caps: LPEVC=0 RefClk=100ns PATEntryBits=1 Arb: Fixed- WRR32- WRR64- WRR128- Ctrl: ArbSelect=Fixed Status: InProgress- VC0: Caps: PATOffset=00 MaxTimeSlots=1 RejSnoopTrans- Arb: Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256- Ctrl: Enable+ ID=0 ArbSelect=Fixed TC/VC=01 Status: NegoPending- InProgress- Capabilities: [128 v1] Power Budgeting <?> Kernel driver in use: nouveau Kernel modules: nvidia
Looking at that redhat bug report, someone added comments about similar issues on NV43 a while after it was closed. Since you have a NV4B, I expect that the issue is the same. Has this always been happening, or is this a new issue? (Guessing the former...) If it's a new issue, doing a kernel bisect would narrow down the commit that broke things. If it has always been like that, does this issue occur with the nvidia binary driver? If not, doing an mmiotrace would be helpful (guide at https://wiki.ubuntu.com/X/MMIOTracing). The register written to by 4eb3033c only exists on NV50+ according to envytools. Are all the errors the same (i.e. trying to read from address 0) or does it vary?
(In reply to comment #1) > Has this always been happening, or is this a new issue? (Guessing the > former...) I can't tell, as I have been using nvidia.ko until now. > If it has always been like that, does this issue occur with the nvidia > binary driver? If not, doing an mmiotrace would be helpful (guide at > https://wiki.ubuntu.com/X/MMIOTracing). The register written to by 4eb3033c > only exists on NV50+ according to envytools. It does not occur with nvidia.ko. MMIOTrace output @ http://stewie.be.tintel.eu/mmiotrace.log > Are all the errors the same (i.e. trying to read from address 0) or does it > vary? Grep on /var/log messages shows only address 0.
Forgot to actually start X, updated the log @ http://stewie.be.tintel.eu/mmiotrace.log :-)
Created attachment 82309 [details] [review] Attempt at adding some writes that the blob does
Please give this patch a shot. Just saw a few things that stood out in your mmiotrace, I have no idea what they actually do :) Perhaps totally unrelated. If it causes new problems, try the new writes individually.
After correcting the typos (mv_wr32 -> nv_wr32), I was able to build the kernel. Unfortunately I still see the same DMAR errors.
Still something actually happening with newer kernels (3.17/3.18)?
(In reply to Tobias Klausmann from comment #7) > Still something actually happening with newer kernels (3.17/3.18)? Unfortunately the card died so I am unable to test anything.
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-nouveau/issues/46.
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.