[Nouveau] [PATCH] drm/nouveau: consider CLASS_DISPLAY_3D devices while detecting dsm/optimus

2013-10-09 Thread Emil Velikov
The present code assumes that optimus is present whenever two VGA (PCI_CLASS_DISPLAY_VGA) devices are present. This does not seem to be the case of newer laptops with optimus, in which case the nvidia gpu is a PCI_CLASS_DISPLAY_3D device. Rework the logic so that we count both VGA and 3D devices,

[Nouveau] [Bug 69168] [NVA8]Arbitrary crashes

2013-10-09 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=69168 --- Comment #3 from Ilia Mirkin --- The xorg log that you attached seems fine. That was produced after the hang? If so, it must be the compositor hanging. Try disabling the compositor. As for your gnome-related question, I don't know -- I stay a

[Nouveau] [Bug 67255] [NVC3] black screen after resuming from Hibernate

2013-10-09 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=67255 --- Comment #7 from Mike --- Created attachment 87327 --> https://bugs.freedesktop.org/attachment.cgi?id=87327&action=edit dmesg after resuming from hibernation with nouveu.debug=trace -- You are receiving this mail because: You are the assig

[Nouveau] [Bug 67255] [NVC3] black screen after resuming from Hibernate

2013-10-09 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=67255 --- Comment #8 from Mike --- Created attachment 87328 --> https://bugs.freedesktop.org/attachment.cgi?id=87328&action=edit dmesg after resuming from suspention with nouveu.debug=trace -- You are receiving this mail because: You are the assign

[Nouveau] [Bug 67255] [NVC3] black screen after resuming from Hibernate

2013-10-09 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=67255 --- Comment #9 from Mike --- I have tried with 3.12-rc4 and the problem persists. I have attached dmesg after resuming from suspention (which works) and resuming from hibernation (wich does not work) with nouveau.debug=trace. -- You are receivi

[Nouveau] [Bug 70326] New: [NVF0] Kernel/Driver fails to initialize acceleration on GTX Titan

2013-10-09 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=70326 Priority: medium Bug ID: 70326 Assignee: nouveau@lists.freedesktop.org Summary: [NVF0] Kernel/Driver fails to initialize acceleration on GTX Titan QA Contact: xorg-t...@lists.x

[Nouveau] [Bug 70326] [NVF0] Kernel/Driver fails to initialize acceleration on GTX Titan

2013-10-09 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=70326 --- Comment #1 from Ilia Mirkin --- You need to mmiotrace the blob and get the graph firmware. See http://nouveau.freedesktop.org/wiki/NVC0_Firmware/ (up to "Video Firmware", which you don't need to worry about). Once you get that, stick the file

[Nouveau] Panic in nouveau on resume after 5addcf0a

2013-10-09 Thread Martin Nyhus
After 5addcf0a (nouveau: add runtime PM support (v0.9)) I'm getting kernel panics immediately after resuming from suspend. Unfortunately it happens so early that I don't get any more info than the blinking keyboard lights... The nouveau tree still panics up to 2fb9d6c, which fails before I can

[Nouveau] [PATCH] drm/nouveau: Do not show an error if the TMDS table pointer is invalid

2013-10-09 Thread Albert Pool
Optimus cards with no outputs trigger this message (NVD7, for example) This commit changes the warning into an informational message. The return value of the function is not used, so I left it at -EINVAL. Signed-off-by: Albert Pool Cc: Emil Velikov --- drivers/gpu/drm/nouveau/nouveau_bios.c |

[Nouveau] Null pointer deref in nv50.c after 2fb9d6cf

2013-10-09 Thread Martin Nyhus
After 2fb9d6cf (drm/nouveau/core: convert event handler apis to split create/enable semantics) from the nouveau tree I'm getting a kernel panic on boot because pdisp is 0 at core/engine/software/nv50.c:179. The nouveau part of the call trace is: nv50_software_context_ctor+0x80/0x130 nouveau_obj

Re: [Nouveau] [PATCH] drm/nouveau: Do not show an error if the TMDS table pointer is invalid

2013-10-09 Thread Ben Skeggs
On Thu, Oct 10, 2013 at 5:40 AM, Albert Pool wrote: > Optimus cards with no outputs trigger this message (NVD7, for example) > This commit changes the warning into an informational message. > The return value of the function is not used, so I left it at -EINVAL. > > Signed-off-by: Albert Pool > C

Re: [Nouveau] Panic in nouveau on resume after 5addcf0a

2013-10-09 Thread Dave Airlie
On Thu, Oct 10, 2013 at 5:35 AM, Martin Nyhus wrote: > After 5addcf0a (nouveau: add runtime PM support (v0.9)) I'm getting kernel > panics immediately after resuming from suspend. Unfortunately it happens so > early that I don't get any more info than the blinking keyboard lights... > > The nouvea

Re: [Nouveau] [PATCH] drm/nouveau: Do not show an error if the TMDS table pointer is invalid

2013-10-09 Thread Albert Pool
Thank you for your reply. I hope I won't forget to upload the BIOS of Joey4712's NVE7 to the repo tomorrow, then you guys can take a look at it too. That is a GPU without any outputs and it triggers this message.I'm writing this on my phone, so don't have the file at hand right now. Albert --

Re: [Nouveau] Panic in nouveau on resume after 5addcf0a

2013-10-09 Thread Martin Nyhus
On 10/09/2013 11:34 PM, Dave Airlie wrote: On Thu, Oct 10, 2013 at 5:35 AM, Martin Nyhus wrote: After 5addcf0a (nouveau: add runtime PM support (v0.9)) I'm getting kernel panics immediately after resuming from suspend. Unfortunately it happens so early that I don't get any more info than the bl