Nouveau works nicely on my laptop, including suspend and resume, until the power source changes (from AC to battery or from battery to AC). After that event, mouse pointer gets flaky and I see in /var/log/messages lots of lines like: Nov 12 01:01:12 localhost kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Nov 12 01:01:12 localhost kernel: [drm] nouveau 0000:01:00.0: no space while setting cursor image Nov 12 01:01:12 localhost kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Nov 12 01:01:14 localhost kernel: [drm] nouveau 0000:01:00.0: no space while setting cursor image Nov 12 01:01:14 localhost kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Card is: nVidia Corporation Quadro NVS 130M [10de:042a]
Is this 100% reproducible on your system, or a random occurance?
100% reproducible. The timing between power source change and the symptoms varies, but always within minutes I get to the issue. Right now the only way out seems a reboot, suspending or hibernating leads to a black screen on resume.
I have experienced this same problem on my desktop system a few days ago. I have not used suspend/resume. xorg-x11-server-Xorg-1.7.1-7.fc12.i686 xorg-x11-drv-nouveau-0.0.15-17.20091105gite1c2efd.fc12.i686 xorg-x11-drv-nouveau-debuginfo-0.0.15-17.20091105gite1c2efd.fc12.i686 It's only happened once, but when it happened, my machine got horribly slow. Restarting X "resolved" the issue. Xorg.0.log and /var/log/messages will be attached...
Created attachment 369809 [details] Xorg.0.log from the session where the cursor problem occured
Created attachment 369810 [details] Relevant /var/log/messages from the session where the cursor problem occured
Affects me too. 100% reproducible by doing the following: 1. Set close lid to blank screen 2. Close lid 3. Open lid It also seems to happen after 30-60 minutes of general usage. From that point on any cursor change (mouse over anything) creates the following messages and bursts of high CPU usage: Nov 22 09:58:44 localhost kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Nov 22 09:58:44 localhost kernel: [drm] nouveau 0000:01:00.0: no space while setting cursor image Nov 22 09:58:44 localhost kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Nov 22 09:58:45 localhost kernel: [drm] nouveau 0000:01:00.0: no space while setting cursor image Nov 22 09:58:45 localhost kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Card is: 01:00.0 VGA compatible controller: nVidia Corporation Quadro NVS 130M (rev a1) (prog-if 00 [VGA controller]) Subsystem: Toshiba America Info Systems Device 0001 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: 32 bytes Interrupt: pin A routed to IRQ 16 Region 0: Memory at fd000000 (32-bit, non-prefetchable) [size=16M] Region 1: Memory at e0000000 (64-bit, prefetchable) [size=256M] Region 3: Memory at fa000000 (64-bit, non-prefetchable) [size=32M] Region 5: I/O ports at cf00 [size=128] [virtual] Expansion ROM at fc000000 [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 <512ns, 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 <512ns, L1 <4us ClockPM- Surprise- LLActRep- BwNot- LnkCtl: ASPM L0s L1 Enabled; RCB 128 bytes Disabled- Retrain- CommClk+ ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt- LnkSta: Speed 2.5GT/s, Width x16, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt- Capabilities: [100] Virtual Channel <?> Capabilities: [128] Power Budgeting <?> Capabilities: [600] Vendor Specific Information <?> Kernel driver in use: nouveau Kernel modules: nouveau, nvidiafb Rebooting clears this, and doesn't happen with proprietary driver.
In my case suspend/resume works to clear it. I get a black screen with cursor on resume (animation at full speed which indicates problem cleared), switching to a console 2 and back to 1 gets the unlock screen. Once unlocked no cursor problems. So, a sort of workaround for now :) Also happens switching to battery for me, as the original reporter indicated.
This just happened to me as well - this is a desktop system so no suspend/resume or anything. The machine had been up for about four days (but unused for the last two) and just suddenly started spitting out those messages every time anything tried to change the cursor.
Tom, which graphic card are you running (use "lspci -nn" to find out)?
The card is: nVidia Corporation GeForce 8600 GT [10de:0402] (rev a1) I do have a slightly unusual setup in that I have two monitors attached, one of which is rotated. This has led to one other cursor related problem which is reported in bug #538866.
I'm seeing this problem too, on a laptop, takes a while to occur, then my dmesg gets full of those no space while setting/unhiding cursor errors. lspci output: 01:00.0 VGA compatible controller [0300]: nVidia Corporation GeForce 8400M GS [10de:0427] (rev a1) dmesg output: [drm] nouveau 0000:01:00.0: no space while setting cursor image [drm] nouveau 0000:01:00.0: no space while unhiding cursor [drm] nouveau 0000:01:00.0: no space while setting cursor image [drm] nouveau 0000:01:00.0: no space while unhiding cursor
*** Bug 541628 has been marked as a duplicate of this bug. ***
*** Bug 548545 has been marked as a duplicate of this bug. ***
I'm seeing similar symptoms. 02:00.0 VGA compatible controller [0300]: nVidia Corporation GT200 [GeForce GTX 260] [10de:05e2] (rev a1) xorg-x11-drv-nouveau-0.0.15-17.20091105gite1c2efd.fc12.x86_64 kernel-2.6.31.6-166.fc12.x86_64 I have a dual screen setup and I'm seeing lots of the "no space while setting cursor image" notices. When moving the mouse from one monitor to the other, the cursor gets "stuck" on the previous monitor. This primarily happens when the cursor changes, ie, when moused over a border to resize a window. When this happens, the stuck cursor appears several hundred pixels higher on the middle border of the screen (in this example, with the "resize" cursor).
Are you guys able to update to kernel-2.6.31.9-174.fc12 (http://koji.fedoraproject.org/koji/buildinfo?buildID=147895) and report how things are.
I've been running on that kernel since last Wednesday, and no sign of any problems so far. It can sometimes take a week or more before it happens though so it may still appear again.
Hi Ben, I did not realize there were changes related to this in the latest kernel. I just tried unplugging the power cord and my cursor completely disappeared... However, after a reboot I was not able to reproduce the issue again, so I'm not really sure about the status of this bug. I will test a bit more then let you know how it goes. Thanks!
Ben, right it seems the original issue is gone for me. I think there's more to investigate (for instance, I experienced a black screen with only the mouse visible on a resume after suspend) but I guess I'll open other reports if I manage undestard which usage pattern triggers it.
Yup, I've re-tested and this issue seems fixed for me too.
Experiencing this issue right on Quadro NVS 140m (Thinkpad T61) - dual screen setup with kernel-2.6.31.12-174.2.3.fc12.x86_64 xorg-x11-drv-nouveau-0.0.15-19.20091105gite1c2efd.fc12.x86_64 Going for newest rawhide kernel, report back if issue won't dissapear From var log messages: ... Feb 9 09:31:59 kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Feb 9 09:31:59 kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Feb 9 09:32:24 kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Feb 9 09:32:24 kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Feb 9 09:32:24 kernel: [drm] nouveau 0000:01:00.0: no space while hiding cursor Feb 9 09:32:24 kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor Feb 9 09:32:27 kernel: [drm] nouveau 0000:01:00.0: no space while hiding cursor
Miroslav: is this an easily reproducible issue for you? I may have added a fix to nouveau upstream, which'll make it into rawhide soon that will hopefully fix this once and for all. I can't say for sure since none of my cards exhibit the problem for some reason. The current rawhide kernel has code which'll mostly avoid the issue, making it much harder to reproduce (in theory).
Ben, it isn't easily reproducible :/. My system was running for several days with kernel-2.6.31.12-174.2.3.fc12 and the bug appeared "from nowhere" there too. I will try the new rawhide kernel when it's build. When it will appear after that, I will report back. Thanks for the info.
*** Bug 562496 has been marked as a duplicate of this bug. ***
This problem still happens for me, using the latest upstream git of everything (linux, xorg xorg drivers, etc) as of Feb 18 2010, however it has "improved" a little since earlier. My system chip is Quadro NVS 135M (rev a1). This kludge to the nouveau driver in linux stops the problems happening for me altogether: ------------------------------------------------------------ --- a/drivers/gpu/drm/nouveau/nv50_cursor.c +++ b/drivers/gpu/drm/nouveau/nv50_cursor.c @@ -41,6 +41,14 @@ nv50_cursor_show(struct nouveau_crtc *nv_crtc, bool update) struct drm_device *dev = nv_crtc->base.dev; int ret; + static int count = 0; + + if (count > 3) { + return; + } else { + count++; + } + NV_DEBUG_KMS(dev, "\n"); if (update && nv_crtc->cursor.visible) @@ -76,6 +84,8 @@ nv50_cursor_hide(struct nouveau_crtc *nv_crtc, bool update) struct drm_device *dev = nv_crtc->base.dev; int ret; + return; + NV_DEBUG_KMS(dev, "\n"); if (update && !nv_crtc->cursor.visible) ------------------------------------------------------------ This raises a lot of issues - what was happening is that READ_GET() in drivers/gpu/drm/nouveau/nouveau_dma.c returned EBUSY. It looks to me that the performance problems are because of a spin lock - if I try increasing the timeout int READ_GET() from 100000 to 1000000, the performance problems get 10 times worse. Secondly, why are nv50_cursor_show() & nv50_cursor_hide() in /drivers/gpu/drm/nouveau/nv50_cursor.c being called so often? _show only needs to be called once per attached monitor, and then _hide only when it really does need to be hidden, ie. when playing a movie or switching VT (ie. with the above kludge the mouse cursor stays visible on the VT's), not every time the mouse cursor changes. It could be even my WM (IceWM) making all those calls, I haven't looked into this further. I don't have time to look into this further at all.
Show/Hide will get called every time the cursor image is updated, however, they'll actually do nothing unless it needs to change the state, which isn't very often. The reason it's returning EBUSY is because the display engine is hanging for some, yet unknown, reason. Unfortunately none of my systems appear to display the behaviour so I've not been able to track down what's going on yet.
The same is also happening for me with Fedora 13 Alpha (fully updated). Hardware: 01:00.0 VGA compatible controller: nVidia Corporation Quadro NVS 290 (rev a1) Intel(R) Core(TM)2 Quad CPU Q9400 HP dc7900 (IC10 chipset) Software: kernel-2.6.33-1.fc13.x86_64 with no non-Fedora modules xorg-x11-drv-nouveau-0.0.16-2.20100218git2964702.fc13.x86_64 I have been running Fedora 13 Alpha since the release day, and the problem only happened today after about 4 hours uptime and 2 hours of active work. Something is different today than the days before: I've mounted the debugfs at /sys/kernel/debug, which I have not done before today. It would be a strange connection if this was the culprit, though.
I'm running a normal GNOME desktop. It happens when I move the cursor from Firefox to Emacs or vice versa. It does not happen when the cursor moves between Firefox/xterm or Emacs/xterm.
Sorry for spamming, but the only messages I get are "no space while hiding cursor", no unhiding or setting cursor image.
One more thing - it doesn't have to do with Firefox/emacs/whatever. It happened when I moved the mouse cursor from the left screen to the right one and vice versa. I have two 24" displays at 1920x1200 attached via DVI. I've never had this problem with Fedora 12 which I've been running extensively until a few days ago.
Created attachment 402070 [details] log excerpt It happened again today with the same kernel and nouveau versions still. And xorg-x11-server-Xorg-1.7.99.901-8.20100223.fc13.x86_64 (forgot to mention this in comment 26). When it happened, I tried logging out of the GNOME session which caused both monitors to power down until reboot. I'm attaching an excerpt from /var/log/messages at the time it happened and also the Xorg output, while I'm at it.
Created attachment 402071 [details] xorg server log
Ben, I'm sorry to say this issue is showing itself again, I'm now running the latest F12 bits, kernel-2.6.32.10-90.fc12.x86_64, xorg-x11-drv-nouveau-0.0.15-21.20091105gite1c2efd.fc12.x86_64. I'm probably going to move on F13 soon so I will be able to say if the same happens there.
Albeit, I'm noticing a difference: this time only one kind of messages is appearing in /var/log/messages: Apr 5 01:29:41 localhost kernel: [drm] nouveau 0000:01:00.0: no space while unhiding cursor
*** Bug 593695 has been marked as a duplicate of this bug. ***
I originally filed 548545, which was marked as a duplicate of this bug. I just had the same symptom happen: the mouse pointer is no longer drawn when on the left monitor. No messages in /var/log/Xorg.0.log, just this one line in /var/log/messages: May 28 09:49:10 localhost kernel: [drm] nouveau 0000:03:00.0: EvoCh 0 Mthd 0x0000 Data 0x00000400 (0x0002 0x01) The mouse pointer is still drawn on the right monitor. This machine is now running F-13, updated this morning. Possibly relevant version numbers: kernel-2.6.33.4-95.fc13.x86_64 mesa-dri-drivers-7.8.1-6.fc13.x86_64 xorg-x11-drivers-7.3-14.fc13.x86_64 xorg-x11-drv-nouveau-0.0.16-6.20100423git13c1043.fc13.x86_64 Is this really the same bug?
Something similar seems to have recurred in F-13 [drm] nouveau 0000:01:00.0: EvoCh 0 Mthd 0x0000 Data 0x00000400 (0x0002 0x01)
I get the same message as David in comment #36, then the stream of no space messages with jerky window and cursor behaviour. Fully up-to-date F13, kernel-2.6.33.5-124.fc13.x86_64 xorg-x11-drv-nouveau-0.0.16-6.20100423git13c1043.fc13.x86_64 xorg-x11-server-Xorg-1.8.0-17.fc13.x86_64 nVidia Corporation G84M [Quadro NVS 140M] (rev a1)
I started getting these messages yesterday, I suspect it was when I experienced a sudden very high load and non-responsiveness - I was running a windows machine inside virtualbox so that I could run gotomeeting, and I was given control of the meeting. When I would click on something in the vm nothing would happen until I moved the mouse out of the virtualbox. But I've continued to get the messages at a lower rate even after suspending the windows machine. kernel= 2.6.32.14-127.fc12.x86_64
I had this problem a while ago and rolled back nouveau and the problem went away (well it didn't happen again - but of course that's no certainty that the bug was not present in the earlier versions) The old nouveau that appears to be OK is: xorg-x11-drv-nouveau-0.0.15-17.20091105gite1c2ef (that's the one I rolled back to when the mouse problem first showed itself) 5 days ago I did a full update that included: kernel-2.6.32.16-141.fc12.x86_64 and xorg-x11-drv-nouveau-0.0.15-21.20091105gite1c2efd.fc12.x86_64 and the problem has happened again today (for the first time since the rollback) Xorg.0.log info: First nVidia message is: nVidia Corporation GeForce 8600 GTS rev 161, Mem @ 0xf6000000/16777216, 0xe0000000/268435456, 0xf4000000/33554432, I/O @ 0x0000b000/128, BIOS @ 0x????????/131072 I use a USB mouse, log says: Microsoft Microsoft 5-Button Mouse with IntelliEye(TM) I also have a USB Bamboo that the log finds BEFORE the mouse: Wacom BambooFun 6x8 (I don't use the pen) The rest of the log looks like status dumps without any error messages. In my case it's a dual display and acts even stranger when the problem occurs (randomly?): The left display shows the mouse correctly (but jerky) In the right display (which has the menu bar) the mouse works but it is invisible More strange: the mouse image is partially still shown on the edge of the left screen while using the right screen When the screen saver comes on the mouse is still visible in the left screen It doesn't fix the problem when the screen comes back (thought I'd mention that because I've seen it said elsewhere) It maybe happens when mplayer is playing fullscreen (so a fullscreen app may be needed to activate the bug) Of course when the problem wasn't there I was using the current mplayer version constantly also The first dmesg error was: [drm] nouveau 0000:01:00.0: EvoCh 0 Mthd 0x0000 Data 0x00000400 (0x0002 0x01) followed by lots of the cursor messages (both 'hiding' and 'unhiding')
Minor updates on previous post: I only boot runlevel 3 and thus always use startx (as user "root") Once the problem has occurred: X menu System->"Log Out root..." fails to go back to the console (the screen clears to just showing the backgrounds and then stops) OS is of course unaffected by this - ssh login works fine (I use that to "shutdown -r now") I've rolled back to xorg-x11-drv-nouveau-0.0.15-17.20091105gite1c2ef again (nothing else changed) and will report if it happens again with this version Logout messages of Xorg.0.log (before logout stops): (II) AT Translated Set 2 keyboard: Close (II) UnloadModule: "evdev" (II) Macintosh mouse button emulation: Close (II) UnloadModule: "evdev" (II) UnloadModule: "wacom" (II) UnloadModule: "wacom" (II) UnloadModule: "wacom" (II) Wacom BambooFun 6x8: removing automatically added devices. (II) UnloadModule: "wacom" (II) Microsoft Microsoft 5-Button Mouse with IntelliEye(TM): Close (II) UnloadModule: "evdev" (II) Power Button: Close (II) UnloadModule: "evdev" (II) Power Button: Close (II) UnloadModule: "evdev" (II) NOUVEAU(0): NVLeaveVT is called. (II) NOUVEAU(0): Closed GPU channel 2 No errors seem to be reported in that log
Also now had the same problem occur with the older RPM: xorg-x11-drv-nouveau-0.0.15-17.20091105gite1c2ef Also, the fullscreen app (mplayer) was not running when it occurred. Had the same messages for dmesg, /var/log/messages, /var/log/Xorg.0.log as before. (except no cursor hide/unhide messages) This time I also got the extra shutdown message (after trying to logout) dmesg: [drm] nouveau 0000:01:00.0: nouveau_channel_free: freeing fifo 2 /var/log/messages: Jul 30 12:34:08 ... kernel: [drm] nouveau 0000:01:00.0: nouveau_channel_free: freeing fifo 2 However, I have found another problem with my system that may have something to do with the nouveau bug presenting itself. My USB mouse is possibly failing and switches on and off sometimes ... easy for me to see while dragging windows around, because it sometimes lets go of a window but I haven't released the left mouse button (I only realised what was going on after my previous comments)
Created attachment 451152 [details] Provide info on the Evo channel in debugfs I was able to do some debugging on this issue today, and have a mosly reproducible test case (~90-95%) on my laptop (also using an external monitor). Here's nouveau's spiel about the card on boot of the 2.6.34.7-56.fc13.x86_64 kernel. [drm] nouveau 0000:01:00.0: Detected an NV50 generation card (0x086900a2) [drm] nouveau 0000:01:00.0: Attempting to load BIOS image from PRAMIN [drm] nouveau 0000:01:00.0: ... appears to be valid [drm] nouveau 0000:01:00.0: BIT BIOS found [drm] nouveau 0000:01:00.0: Bios version 60.86.68.00 [drm] nouveau 0000:01:00.0: TMDS table revision 2.0 not currently supported [drm] nouveau 0000:01:00.0: Found Display Configuration Block version 4.0 [drm] nouveau 0000:01:00.0: Raw DCB entry 0: 01000323 00010034 [drm] nouveau 0000:01:00.0: Raw DCB entry 1: 02011300 00000028 [drm] nouveau 0000:01:00.0: Raw DCB entry 2: 02022312 00010010 [drm] nouveau 0000:01:00.0: Raw DCB entry 3: 010333f1 00c0c070 [drm] nouveau 0000:01:00.0: Raw DCB entry 4: 0000000e 00000000 [drm] nouveau 0000:01:00.0: DCB connector table: VHER 0x40 5 14 2 [drm] nouveau 0000:01:00.0: 0: 0x00000041: type 0x41 idx 0 tag 0xff [drm] nouveau 0000:01:00.0: unknown type, using 0x40 [drm] nouveau 0000:01:00.0: 1: 0x00000100: type 0x00 idx 1 tag 0xff [drm] nouveau 0000:01:00.0: 2: 0x00001255: type 0x55 idx 2 tag 0x07 [drm] nouveau 0000:01:00.0: unknown type, using 0x31 [drm] nouveau 0000:01:00.0: 3: 0x00000310: type 0x10 idx 3 tag 0xff [drm] nouveau 0000:01:00.0: 4: 0x00000311: type 0x11 idx 4 tag 0xff [drm] nouveau 0000:01:00.0: 5: 0x00000313: type 0x13 idx 5 tag 0xff [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 0 at offset 0xC11C [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 1 at offset 0xC486 [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 2 at offset 0xD0C7 [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 3 at offset 0xD1B9 [drm] nouveau 0000:01:00.0: Parsing VBIOS init table 4 at offset 0xD3B3 [drm] nouveau 0000:01:00.0: Parsing VBIOS init table at offset 0xD418 [drm] nouveau 0000:01:00.0: 0xD418: Condition still not met after 20ms, skipping following opcodes [drm] nouveau 0000:01:00.0: 0xB1D6: parsing output script 0 [drm] nouveau 0000:01:00.0: 0xB34C: parsing output script 0 [drm] nouveau 0000:01:00.0: 0xA228: parsing output script 0 [drm] nouveau 0000:01:00.0: Detected 256MiB VRAM [drm] nouveau 0000:01:00.0: 512 MiB GART (aperture) [drm] nouveau 0000:01:00.0: DCB encoder 1 unknown [drm] nouveau 0000:01:00.0: TV-1 has no encoders, removing [drm] nouveau 0000:01:00.0: gpio tag 0xff not found [drm] nouveau 0000:01:00.0: gpio tag 0xff not found [drm] nouveau 0000:01:00.0: Allocating FIFO number 1 [drm] nouveau 0000:01:00.0: nouveau_channel_alloc: initialised FIFO 1 [drm] nouveau 0000:01:00.0: allocated 1920x1200 fb: 0x40230000, bo ffff8800790a2400 fbcon: nouveaufb (fb0) is primary device [drm] nouveau 0000:01:00.0: 0xB083: parsing clock script 0 fb0: nouveaufb frame buffer device [drm] Initialized nouveau 0.0.16 20090420 for 0000:01:00.0 on minor 0 [drm] nouveau 0000:01:00.0: 0xB199: parsing clock script 1 [drm] nouveau 0000:01:00.0: 0x1085: parsing clock script 0 [drm] nouveau 0000:01:00.0: Allocating FIFO number 2 [drm] nouveau 0000:01:00.0: nouveau_channel_alloc: initialised FIFO 2 I find that as we rewind the Evo's push buffer, we somehow trigger a display error, giving the message we've come to hate: [drm] nouveau 0000:01:00.0: EvoCh 0 Mthd 0x0000 Data 0x00000400 (0x0002 0x01) So far, this is closely correlated with the driver setting the ring to jump to offset 0 and wait for it to come out of the SKIPS (line 297 in nouveau.git's nouveau_dma.c -- commit 0100828245). I have never seen the error message without a wrap of the Evo ring, and some 30+ occurring in close proximity. I instrumented the code, and we usually wrap at about index 0x3f9 or 0x3fb, and evo->dma.cur is 0x27 when the message is printed. It's usually nv50_cursor_show() or nv50_cursor_hide() that is running at the time. The quickest way for me to reproduce this has been to boot, log into GNOME, open a few terminal windows, and watch /sys/kernel/debug/dri/0/channels/evo. I find that locking/unlocking the screen with the blank screen saver adds about ~0x120 to the current location each iteration. Once I'm close to a wrap, I move the mouse between the windows, which advances cur by ~0x20 or so until it wraps. Once we get a wrap and the error message, the GPU's get location for this channel will be forever 0. We'll continue to put commands in the ring, and once we fill it again, we'll start timing out in READ_GET() and that causes the stuttering mouse movements. I've gone about as far as I can, I think. I currently don't have the knowledge to figure out why we get the display error on most wraps but not all. It is perhaps timing related -- using nouveau_bo_rd32() on the push buffer to verify contents after we do the WRITE_PUT(NOUVEAU_DMA_SKIPS) made it 3 wraps before it died once, but then it died on the first wrap the next boot. If you guys can give me a pointer as to where to look next, I'm game. Until then, I hope this helps.
Wow, I'm impressed, many thanks for getting that. FWIW I always have at least 10 xterms open and "xlock -mode blank" is my screensaver, so I always saw the issue after a couple of days at most.
I'm adding a 'me too' to this bug under F13. Dual Screens, when manifested cursor has issues drawing on left hand screen, appears ok on right hand screen. lspci; 01:00.0 VGA compatible controller: nVidia Corporation GT216 [GeForce GT 220] (rev a2) Driver; xorg-x11-drv-nouveau-0.0.16-8.20100423git13c1043.fc13.i686 happens with 2 hours of a reboot under kernel-PAE-2.6.34.7-56.fc13.i686, can take several days with kernel-PAE-2.6.34.6-54.fc13.i686. /var/log/messages Oct 18 11:09:00 cbr-sjw-dt kernel: [drm] nouveau 0000:01:00.0: EvoCh 0 Mthd 0x0000 Data 0x00000400 (0x0002 0x01) Happy to work with people on this one, it's making this workstation nearly unusable.
I have not seen the original behaviour I reported since long, but now that I switched to F14 I guess the best course of action is to close this report and let users on a different release (this was for F12, which is EOL in about one month anyway) or different cards from mine open more specific bug reports.
Same for me. Has not happened once in a very long time (more than 6 months).
Still happening every so often on F12 (I swapped the flaky mouse out long ago) Current nouveau is: xorg-x11-drv-nouveau-0.0.15-21.20091105gite1c2efd.fc12.x86_64 Really is annoying sometimes to have to kill everything and shut down to fix the problem (I've ALWAYS booted 'init 3' on all Linux servers because I don't think graphics drivers should be needed to boot a server ... the drivers for a while now have been forced upon us even during an 'init 3' boot ... but I've also had 'init 3' on all my desktops too ... pity it doesn't help with this problem. My other F12 desktop - dual screen LCD + TV connected - uses that 'other' driver - obviously due to TV-Out - and though I don't work on it much - just for playing DVD's or other videos - I've still never had this problem with it)
I still see the behavior I reported (see comment 35) on F-13 once in awhile. Steve Walsh (comment 44) appears to be seeing the same thing. I'll ask again: is this really the same bug? If not, perhaps bug 548545 should be revived.
This message is a reminder that Fedora 12 is nearing its end of life. Approximately 30 (thirty) days from now Fedora will stop maintaining and issuing updates for Fedora 12. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as WONTFIX if it remains open with a Fedora 'version' of '12'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version prior to Fedora 12's end of life. Bug Reporter: Thank you for reporting this issue and we are sorry that we may not be able to fix it before Fedora 12 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora please change the 'version' of this bug to the applicable version. If you are unable to change the version, please add a comment here and someone will do it for you. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete. The process we are following is described here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping
I use Slackware, this has got nothing to do with Fedora 12, if this bug gets closed I'm giving up on this and switching to the closed nvidia driver.
Please respect the maintainer and dont be rude. You should be aware of the fact that this bugzilla instance is to be used with RHEL and Fedora only not for everydistro out there so please file your report either to slackware bugzilla instance if they have one or directly upstream with the nouveau maintainers. However you are welcome to download the latest Fedora release ( 14 ) install it and try to see if you still experience this problem and if so you can make note of that on this bug. Thank you.
As the original reporter, I need to note this was for a specific version of nuoveau (the one in F12) with my specific card (with ID in the summary) and that I did not experience anymore mouse movement issues (now I'm on F14, no issues there as well) If you are running any other combo, I advice to open another, separate report
Created attachment 457771 [details] Band-aid patch that solves the problem for me Continuing the work I started in comment 42, I was also seeing the "EvoCh 0 Mthd 0x0000 Data 0x00000400 (0x0002 0x01)" messages as everyone else. I did more investigation, and found that not using the last 32 bytes of the Evo ring buffer solved the issue for me -- I no longer get the EvoCh messages, which directly lead to the "no space" messages as seen by the original report and the jerky mouse movements reported by others. This bug is current as of F13, and is likely in F14 and rawhide, as the issue is not yet fixed upstream. Prior to the patch, dma.max was getting the value 1022. &'ing with ~7 was equivalent to subtracting another 6 entries from that value. I also tried ~3, but that did not fix the issue. I was asked on IRC to try subtracting 5, 4 etc to narrow down the minimum value, and my testing confirmed that subtracting 6 (&= ~7) was minimal. I also tried changing some of the register settings above (NV50_PDISPLAY_CHANNEL_UNK2, etc) but was not successful in teasing out meaning or a fix using them. I have not tried playing with the RAMHT for Evo. I don't like that my patch involves yet more magic numbers without explanation, but it has been stable for me for the past few weeks, and solves my cursor problem.
Created attachment 457773 [details] Test program that provokes the problem This program hides/unhides the cursor for the number of cycles given on the command line, default 1. I use this to push the Evo channel's index through the rewind and expose the EvoCh messages. It's much easier than moving the cursor from window to window for a while until things wrap. With my patch, this survives thousands of iterations.
Reassigning to 13 as suggested by Ben on IRC; that will avoid the auto-close.
(In reply to comment #54) > Created attachment 457773 [details] > Test program that provokes the problem > > This program hides/unhides the cursor for the number of cycles given on the > command line, default 1. I use this to push the Evo channel's index through the > rewind and expose the EvoCh messages. It's much easier than moving the cursor > from window to window for a while until things wrap. With my patch, this > survives thousands of iterations. Can you give the list of libs needed to build the program ? Thanks.
(In reply to comment #56) > (In reply to comment #54) > > This program hides/unhides the cursor for the number of cycles given on the > > command line, default 1. I use this to push the Evo channel's index through the > > rewind and expose the EvoCh messages. It's much easier than moving the cursor > > from window to window for a while until things wrap. With my patch, this > > survives thousands of iterations. > > Can you give the list of libs needed to build the program ? > > Thanks. cc -o cycle_cursor -O3 -Wall cycle_cursor.c -lX11
David, I've been using your patch in comment 53 and it's resolved the issue for me too, thank you very much.
I experienced the same problem on Fedora 14: "no space while unhiding cursor". This issue is very annoying because the system becomes so slow that I need to restart it.
I'd like to test the patch in comment 53, but I cannot find nv50_display.c. I've installed the kernel sources (yum install kernel-devel) but the folder "/usr/src/kernels/2.6.35.6-48.fc14.x86_64/drivers/gpu/drm/nouveau" is empty.
I think that the Severity and Priority for this bug should be higher, because, when it occurs, the system becomes so slow that is almost unusable and on my laptop occurs every day. My video card is: 01:00.0 VGA compatible controller: nVidia Corporation GT216 [Quadro FX 880M] (rev a2)
I'm having the invisible cursor issue and the cycle_cursor app doesn't help with that problem. I'm not seeing ANY errors in either the .xsession_errors file or in /var/log/messages* . 00:05.0 VGA compatible controller: nVidia Corporation C51G [GeForce 6100] (rev a2)
(In reply to comment #62) > I'm having the invisible cursor issue and the cycle_cursor app doesn't help > with that problem. > > I'm not seeing ANY errors in either the .xsession_errors file or in > /var/log/messages* . > > 00:05.0 VGA compatible controller: nVidia Corporation C51G [GeForce 6100] (rev > a2) This one is actually a hardware bug, and not related in any way.
I don't know if my issue is related to this bug, but on my laptop (Compal JFL92) with a 8600M GT card, kernel 2.6.35.8-60 killed my sound. Pulseaudio no more sees the sound card (HDA Intel) and falls back to dummy audio output. No problem at all with 2.6.35.8-59. I'm not sure if this is the right bug to report it, as the kernel changelog reads: * Fri Nov 19 2010 Ben Skeggs <bskeggs> 2.6.35.8-60 - nouveau: add quirk for iMac G4 (rhbz#505161) - nouveau: add workaround for display hang on GF8+ (rhbz#537065) - nouveau: don't reject 3D object creation on NVAF (MBA3) but this seems the only one related to my hardware. dmesg doesn't report anything suspicious, but I'm available for any further testing on this issue. This on F14 x86_64.
(In reply to comment #63) > (In reply to comment #62) > > I'm having the invisible cursor issue and the cycle_cursor app doesn't help > > with that problem. > > > > I'm not seeing ANY errors in either the .xsession_errors file or in > > /var/log/messages* . > > > > 00:05.0 VGA compatible controller: nVidia Corporation C51G [GeForce 6100] (rev > > a2) > > This one is actually a hardware bug, and not related in any way. Odd, Since in Fedora 12 it worked fine (cursor showed up without any issues). It only seemed to happen after a certain point. Next time I have to reboot, I'll swap to an external card vs. the built in video).
(In reply to comment #64) > I don't know if my issue is related to this bug, but on my laptop (Compal > JFL92) with a 8600M GT card, kernel 2.6.35.8-60 killed my sound. Pulseaudio no > more sees the sound card (HDA Intel) and falls back to dummy audio output. No > problem at all with 2.6.35.8-59. I'm not sure if this is the right bug to > report it, as the kernel changelog reads: > > * Fri Nov 19 2010 Ben Skeggs <bskeggs> 2.6.35.8-60 > - nouveau: add quirk for iMac G4 (rhbz#505161) > - nouveau: add workaround for display hang on GF8+ (rhbz#537065) > - nouveau: don't reject 3D object creation on NVAF (MBA3) > > but this seems the only one related to my hardware. > dmesg doesn't report anything suspicious, but I'm available for any further > testing on this issue. > > This on F14 x86_64. What was your previous kernel? I don't see how these changes could possibly cause what you see.
(In reply to comment #65) > (In reply to comment #63) > > (In reply to comment #62) > > > I'm having the invisible cursor issue and the cycle_cursor app doesn't help > > > with that problem. > > > > > > I'm not seeing ANY errors in either the .xsession_errors file or in > > > /var/log/messages* . > > > > > > 00:05.0 VGA compatible controller: nVidia Corporation C51G [GeForce 6100] (rev > > > a2) > > > > This one is actually a hardware bug, and not related in any way. > > Odd, Since in Fedora 12 it worked fine (cursor showed up without any issues). > It only seemed to happen after a certain point. Next time I have to reboot, > I'll swap to an external card vs. the built in video). I guess it's possible something in nouveau now triggers the hw bug more reliably. All previous reports indicate that it would usually come up find, and disappear randomly at some point until a reboot. I used to have the effected hardware (the machine has since died), and even the NVIDIA binary driver could not bring it back once it went.
(In reply to comment #66) > (In reply to comment #64) > > I don't know if my issue is related to this bug, but on my laptop (Compal > > JFL92) with a 8600M GT card, kernel 2.6.35.8-60 killed my sound. Pulseaudio no > > more sees the sound card (HDA Intel) and falls back to dummy audio output. No > > problem at all with 2.6.35.8-59. I'm not sure if this is the right bug to > > report it, as the kernel changelog reads: > > > > * Fri Nov 19 2010 Ben Skeggs <bskeggs> 2.6.35.8-60 > > - nouveau: add quirk for iMac G4 (rhbz#505161) > > - nouveau: add workaround for display hang on GF8+ (rhbz#537065) > > - nouveau: don't reject 3D object creation on NVAF (MBA3) > > > > but this seems the only one related to my hardware. > > dmesg doesn't report anything suspicious, but I'm available for any further > > testing on this issue. > > > > This on F14 x86_64. > What was your previous kernel? I don't see how these changes could possibly > cause what you see. Previous kernel was 2.6.35-59, and audio still works if I boot it. Yeah I know, it's weird. I'm looking after getting some more info about it, but I doubt I'll have time today...
(In reply to comment #68) > (In reply to comment #66) > > (In reply to comment #64) > > > I don't know if my issue is related to this bug, but on my laptop (Compal > > > JFL92) with a 8600M GT card, kernel 2.6.35.8-60 killed my sound. Pulseaudio no > > > more sees the sound card (HDA Intel) and falls back to dummy audio output. No > > > problem at all with 2.6.35.8-59. I'm not sure if this is the right bug to > > > report it, as the kernel changelog reads: > > > > > > * Fri Nov 19 2010 Ben Skeggs <bskeggs> 2.6.35.8-60 > > > - nouveau: add quirk for iMac G4 (rhbz#505161) > > > - nouveau: add workaround for display hang on GF8+ (rhbz#537065) > > > - nouveau: don't reject 3D object creation on NVAF (MBA3) > > > > > > but this seems the only one related to my hardware. > > > dmesg doesn't report anything suspicious, but I'm available for any further > > > testing on this issue. > > > > > > This on F14 x86_64. > > What was your previous kernel? I don't see how these changes could possibly > > cause what you see. > > Previous kernel was 2.6.35-59, and audio still works if I boot it. Yeah I know, > it's weird. I'm looking after getting some more info about it, but I doubt I'll > have time today... Sorry disregard my bug report, I verified it's a problem with the new dracut update that triggers when a new initramfs is built, not a kernel problem. Coincidentally the two things overlapped on my system. Sorry for wasting your time.
As I already asked, how may I find nv50_display.c to try the patch in comment 53?
(In reply to comment #70) > As I already asked, how may I find nv50_display.c to try the patch in comment > 53? You don't need to, it's in kernel-2.6.35.8-60.fc14 (http://koji.fedoraproject.org/koji/buildinfo?buildID=205566)
Okay, thanks a lot! I still have the kernel 2.6.35.6-48.fc14.x86_64. The kernel-2.6.35.8-60 is still a release candidate. May I install it simple by downloading the rpm or it's better that I wait for the official release (is there already a date?) and then updating with yum?
I've now installed the newer kernel (2.6.35.8-60). I let you known, if the problem ("no space while unhiding cursor") occurs again.
I'm using the newer kernel (2.6.35.8-60) for about one month and I haven't experienced the problem above ("no space while unhiding cursor") any more. In my opinion this problem is fixed. Thanks a lot!
This message is a reminder that Fedora 13 is nearing its end of life. Approximately 30 (thirty) days from now Fedora will stop maintaining and issuing updates for Fedora 13. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as WONTFIX if it remains open with a Fedora 'version' of '13'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version prior to Fedora 13's end of life. Bug Reporter: Thank you for reporting this issue and we are sorry that we may not be able to fix it before Fedora 13 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora please change the 'version' of this bug to the applicable version. If you are unable to change the version, please add a comment here and someone will do it for you. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete. The process we are following is described here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping
Fedora 13 changed to end-of-life (EOL) status on 2011-06-25. Fedora 13 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. Thank you for reporting this bug and we are sorry it could not be fixed.