similar to: OpenGL context switching with Noveau

For details see my presentation on day 3 of XDC2016. TL;DW: A context switch on average takes ~25 microseconds, but depending on the display resolution and the load on the card times up to 130 microseconds have been observed. The average does not appear to differ much between cards as the (growing) size of the context is in balance with the increased DRAM bandwidth. Measured worst cases

Context switching = very slow on NVIDIA. Don't do it if you can avoid it. Each context is like a megabyte of data, if not more. Each time it has to get saved off and restored. On Wed, Feb 8, 2017 at 1:38 PM, Sampsa Riikonen <sampsa.riikonen at iki.fi> wrote: > Dear Devs, > > (I hope this question is not that much OT for this list..) > > My question is about fast OpenGL

Dear Devs, We have achieved a desktop of up to six monitors, with openGL running succesfully on the desktop, with the following setup/features: * Ubuntu 16+ * Xrandr * Noveau driver * Two gtx750 graphic cards Each (identical) graphic card has 2xHDMI + 2xDVI connectors, which we connect to the monitor array. So far it works with six monitors, but we'd like to achieve eight. However,

On 05/07/2017 11:12 PM, Ilia Mirkin wrote: > On Sun, May 7, 2017 at 7:17 AM, Sampsa Riikonen <sampsa.riikonen at iki.fi> wrote: >> Dear Devs, >> >> We have achieved a desktop of up to six monitors, with openGL running >> succesfully on the desktop, with the following setup/features: >> >> * Ubuntu 16+ >> * Xrandr >> * Noveau driver >>

Hi, Thanks for your advice..! I have a few follow-up questions (tagged below Q1, Q2 and Q3). Any help highly/extremely appreciated. Regarding to "reverse prime", etc. I have read the following page: https://nouveau.freedesktop.org/wiki/Optimus/ So, if we want a single "macro" xscreen that spans two cards, for example: Card 0, connected to monitor 0 Card 1, connected to

On Sun, May 7, 2017 at 7:17 AM, Sampsa Riikonen <sampsa.riikonen at iki.fi> wrote: > Dear Devs, > > We have achieved a desktop of up to six monitors, with openGL running > succesfully on the desktop, with the following setup/features: > > * Ubuntu 16+ > * Xrandr > * Noveau driver > * Two gtx750 graphic cards > > Each (identical) graphic card has 2xHDMI +

On 8 May 2017 12:10:13 a.m. GMT+02:00, Ben Skeggs <skeggsb at gmail.com> wrote: >On 05/07/2017 11:12 PM, Ilia Mirkin wrote: >> On Sun, May 7, 2017 at 7:17 AM, Sampsa Riikonen ><sampsa.riikonen at iki.fi> wrote: >>> Dear Devs, >>> >>> We have achieved a desktop of up to six monitors, with openGL >running >>> succesfully on the desktop,

Hi I am relatively new to ruby on rails. I want to use this javascript bundle: http://prototype-window.xilinus.com/index.html How would I call those functions in an rjs file? example: win = new Window(''window_id'', {title: "Sample", top:100, left:350}) win.getContent().innerHTML = "<h1>Hello world !!</h1>"; win.setDestroyOnClose();

Hello, Is it possible to produce machine code directly to memory buffer so that it could be called immediately? What I would like to do is something like this: Compiler: Scripting language (Antlr) -> LLVM IR -> (Optimization) -> Raw Machine code -> (Transport media) -> Execution The idea is that I could compile my scripts directly to asm without any extra steps and in

It seems common to use RJS for quickly adding and highlighting a list that is having content added to it like this: http://rafb.net/paste/results/Z3m32g48.html ... where "my_list" would be an UL and the partial would insert a LI. Given that example, is there an easy way to highlight only the newly inserted LI instead of the whole UL? The only way I can see to do it is to give

Hi All, I'm experiencing odd behaviour with various OpenGL apps I'm (trying to) use. I have an IBM Thinkpad X60s with the Intel GMA chipset, the 945GM to be precise. I have glxinfo/glxgears working normally, disabled composite in xorg.conf and everything seems fine. However, there are a few applications that I use that show rather strange behaviour that I think is related to the same

On tor, 2012-10-18 at 18:11 +0800, Jun Koi wrote: > On Thu, Oct 18, 2012 at 5:42 PM, Duncan Sands <baldrick at free.fr> wrote: > > Hi Jun, > > > > > >>> don't use cmake? > >>> > >> > >> you mean i should not use cmake? what is wrong here? > > > > > > LLVM cmake support is missing all kinds of features

On Thu, 17 Jul 2008 15:49:42 +0200 Micha? Wi?niewski <brylozketrzyn at gmail.com> wrote: > Hi > > I'm writing to you cause I want to help with noveau. I have GF 5200 card and I > can dump any data you need from it's registers, of course, avalaible to dump > with nvclock. Just write what test you want me to perform (if possible, with > explanations; I'll set

Hey Attila, I CC'd the nouveau mailing list. Although if you want help, you should probably provide more details (what version of mesa, etc.). St?phane On Sat, Feb 8, 2014 at 11:23 PM, Attila T?th <tothsoft at gmail.com> wrote: > Hi, > sorry, i have not better email adress of noveau team. Please send email - > mail feedback, that by moving on time in the movie, vdpau

Adding back the nouveau list... On Sun, Feb 9, 2014 at 8:28 AM, Attila T?th <tothsoft at gmail.com> wrote: > Linux mint 16 (64bit) > > Repo: Ubuntu X-SWAT <ubuntu-x at lists.ubuntu.com><ubuntu-x at lists.ubuntu.com> > Mesa: 10.2.0~git20140205.44338cd8-0ubuntu0sarvatt~saucy > libdrm: 2.4.52+git20140121.46d451c9-0ubuntu0sarvatt~saucy > libg3dvl-mesa:

On Sat, 2020-06-13 at 20:29 -0500, Aditya Pakki wrote: > nv50_disp_atomic_commit() calls calls pm_runtime_get_sync and in turn > increments the reference count. In case of failure, decrement the > ref count before returning the error. > > Signed-off-by: Aditya Pakki <pakki001 at umn.edu> > --- > drivers/gpu/drm/nouveau/dispnv50/disp.c | 4 +++- > 1 file changed, 3

nv50_disp_atomic_commit() calls calls pm_runtime_get_sync and in turn increments the reference count. In case of failure, decrement the ref count before returning the error. Signed-off-by: Aditya Pakki <pakki001 at umn.edu> --- drivers/gpu/drm/nouveau/dispnv50/disp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c

OK, I give up and come grovelling, "Fork" was suggested at 18:23, it's now 22:20 and I have been through 4 different methods, all block with a 2 second delay before returning to dialplan. Here are just some of the examples I have tried, as as per the suggestions, I am closing all possible outputs in the forked process. https://docs.python.org/3.5/library/multiprocessing.html

Thanks for the pointer. But I don't like this patch. If you find a bug, make a bug report or just fix it if you know the fix already. Or write something in IRC. Or write on the Mailing list as a general question or something else But I really don't agree on doing it this way. You would have needed like the same amount of time to actual fix the problem. Anyway, for adding a printk:

On Tue, Jul 11, 2017 at 2:08 PM, Mike Galbraith <efault at gmx.de> wrote: > On Tue, 2017-07-11 at 13:51 -0400, Ilia Mirkin wrote: >> Some details that may be useful in analysis of the bug: >> >> 1. lspci -nn -d 10de: > > 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GM204 [GeForce GTX 980] [10de:13c0] (rev a1) > 01:00.1 Audio device [0403]: NVIDIA