similar to: Scripted video decoding firmware extraction

Hi Ilia, A fortnight ago, you wrote: > > The video plays, CPU load is less (my aim), but there's ‘tearing’ of > > the picture as if small rectangles that are updates are appearing in > > the wrong location, off by a little. If I step through the frames > > with mpv's ‘.’ and ‘,’ then I've found a pattern: one frame's > > picture is good, followed by

Às 14:46 de 27/07/2019, Ilia Mirkin escreveu: > On Sat, Jul 27, 2019 at 7:37 AM Ralph Corderoy <ralph at inputplus.co.uk> wrote: >> The video plays, CPU load is less (my aim), but there's ‘tearing’ of the >> picture as if small rectangles that are updates are appearing in the >> wrong location, off by a little. If I step through the frames with >> mpv's

On Sat, Jul 27, 2019 at 7:37 AM Ralph Corderoy <ralph at inputplus.co.uk> wrote: > The video plays, CPU load is less (my aim), but there's ‘tearing’ of the > picture as if small rectangles that are updates are appearing in the > wrong location, off by a little. If I step through the frames with > mpv's ‘.’ and ‘,’ then I've found a pattern: one frame's picture is

Hi, I'm having trouble with hardware decoding of video with mpv(1) having switched from Nvidia's proprietary driver to nouveau due to Arch Linux removing its nvidia-340xx package. The machine is an Acer Revo. $ lspci -nn | grep VGA 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GT218 [ION] [10de:0a64] (rev a2) I think this is codename NVA8 with the

First, make sure there's a mentor available -- I doubt I'll have time to do it, but perhaps someone else will be able to. I would recommend writing a minimal application that e.g. encodes a fixed group of frames. You should then trace this app under the blob with both mmiotrace to figure out how to configure the underlying engine (probably just firmware to upload), as well as get a mmt

Hi all, I have developed an Android VR player, which supports 4K video playback so far. I want to include the support for 8K video. The current Snapdragon 845 mobile platform supports up to 4K video playback, which lacks the support for my requirement. So I choose Android-x86 as my platform for development, which is Android OS runs on Intel x86 architecture. I've tried to play 8K(7680*3840)

Hi Martin, if you refer to my kernel version. 3.10.17 is the latest, stable version in the official gentoo repository for the amd64 architeture. See here [4] https://packages.gentoo.org/package/sys-kernel/gentoo-sources?full_cat As long as I do not miss any features, I stay with the stable version. Hence, if I understand you correctly, there are kernel version (newer than some unknown point in

On Thu, Dec 12, 2013 at 2:32 AM, Matthias Nagel <matthias.h.nagel at gmail.com> wrote: > Hello, > > I run the Gentoo Linux distribution and use a self-compiled Linux 3.10.17 > kernel. According to > > [1] http://nouveau.freedesktop.org/wiki/InstallDRM/ > [2] http://nouveau.freedesktop.org/wiki/VideoAcceleration/ > > I need the original firmware from the binary

On 12/12/2013 08:32, Matthias Nagel wrote: > Hello, > > I run the Gentoo Linux distribution and use a self-compiled Linux > 3.10.17 kernel. According to > > [1] http://nouveau.freedesktop.org/wiki/InstallDRM/ > [2] http://nouveau.freedesktop.org/wiki/VideoAcceleration/ > > I need the original firmware from the binary driver in order to > sucessully use DRM and to

Hello there, With the use of CUDA and ptx I managed to make a disassembler for at least one of the nv50 microcode/shader formats. This might come in handy for some people, hence I'm posting it here: http://www.cs.rug.nl/~wladimir/decuda/ I'm not yet sure if pixel/vertex shaders use the same instruction format as the compute shaders used by CUDA, but you'd think so, as nv50 is

Hi, just a note for everybody who stumble across this thread in the future in order to bring this thread to a proper end. If I run the kernel with nouveau.debug="PVP=debug,PBSP=debug,PPPP=debug" I see messages about the firmware files gettings loaded. If the firmware files are missing, I get some warnings/errors. So far everything works as expected. But if I invoke

Hello, I run the Gentoo Linux distribution and use a self-compiled Linux 3.10.17 kernel. According to [1] http://nouveau.freedesktop.org/wiki/InstallDRM/ [2] http://nouveau.freedesktop.org/wiki/VideoAcceleration/ I need the original firmware from the binary driver in order to sucessully use DRM and to use VDPAU video acceleration. I used the python script from [2] and I had a look at the ebuild

Hi, Good to know that! Yes, it should be faster than page faulting, although I haven't done the benchmarking yet. And yes, it is not needed to disable all but one CPU. In my current implementation, I use an ordered workqueue to send the data to the mmapped output buffer (where they will be read from from the user space) and that ensures the order of events is kept. May be less than ideal but

Hi, First of all, congratulations to the X.org team for selection into Google Summer of Code 2016. I would like to participate to Google Summer of Code 2016 by contributing to Kepler Accelerated Video Encoding for Nouveau. I request you to kindly guide me as to how to get started with this project. Regards, Saket Sinha

On Fri, 18 Oct 2013 00:11:15 +0400 Eugene Shatokhin <euspectre at gmail.com> wrote: > Hi, > > Good to know that! > > Yes, it should be faster than page faulting, although I haven't done the > benchmarking yet. And yes, it is not needed to disable all but one CPU. In > my current implementation, I use an ordered workqueue to send the data to > the mmapped output

Hi, > Ah, you are not using the ftrace framework nor relayfs? Mmiotrace used to be relayfs at one point and then converted to ftrace. Yes, I considered these when I started working on KernelStrider but finally borrowed ideas from Perf and implemented them. A mmapped ring buffer does its job well and has a higher throughput than Ftrace in my case. > Are you saying that you intercept

On Sat, 30 Nov 2013 12:54:45 -0800 Ilia Mirkin <imirkin at alum.mit.edu> wrote: > On Thu, Nov 21, 2013 at 5:22 PM, Ilia Mirkin <imirkin at alum.mit.edu> > wrote: > > On Thu, Nov 21, 2013 at 5:07 PM, Benjamin Morris > > <bmorris at nvidia.com> wrote: > >> On 11/19/2013 08:16 PM, Ilia Mirkin wrote: > >>> Hello, > >>> >

On Fri, Oct 19, 2018 at 8:53 AM 孫偉程 <dmvictor at gmail.com> wrote: > I can imagine there is a long way to go to fulfill my requirements. But I know a driver for my NVIDIA card is a must for the hardware decoder to work. Can somebody shed some lights that if I want to play an 8K HEVC encoded video on Android-x86 with NVIDIA's decoder enabled, what should I do in this project? And how

Hey, Probably mostly for Ben, but I had this idea that it might be possible to write a userspace wrapper for the binary driver that run as root could be used to do the card init for the current or even a fake GPU then produce traces to pull the firmwares out. Not sure how insane it is, vs having to mmiotrace, but I just thought I'd see if anyone had considered it. Dave.

On Mon, 14 Oct 2013 22:45:09 +0400 Eugene Shatokhin <euspectre at gmail.com> wrote: > Hi, > > There is an interesting TODO item on MmioTraceDeveloper page: > "kprobes has a generic instruction decoding facility, use that instead of > homebrewn (or KVM), and use emulation instead of page faulting" > > Actually, I have done something similar in one of my