<div dir="ltr"><div>This is proof that discrete Intel Arc use the same old hardware acceleration block for encoding</div><div><br></div><div>Best Regards,</div><div>Andrey</div><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div dir="ltr"><br></div></div></div></div></div></div></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">вт, 10 сент. 2024 г. в 18:35, Terje J. Hanssen via Cin <<a href="mailto:cin@lists.cinelerra-gg.org">cin@lists.cinelerra-gg.org</a>>:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><u></u>
<div>
<br>
<br>
<div>Den 31.08.2024 00:58, skrev Terje J.
Hanssen:<br>
</div>
<blockquote type="cite">
I am stuck how to configure QSV access to an enabled iGPU onboard
an Intel Core i7-6700K (Skylake). This is a legacy, rebuilt
workstation with additional, discrete Nvidia GeForce GTX 960 GPU
on a MSI-Z170-A Pro mobo. Any idea and code modification below?<br>
<br>
Details as follows:<br>
<br>
<font face="monospace">inxi -CGMSz<br>
System:<br>
Kernel: 6.10.5-1-default arch: x86_64 bits: 64<br>
Desktop: GNOME v: 46.3.1 Distro: openSUSE Tumbleweed-Slowroll
20240803<br>
Machine:<br>
Type: Desktop Mobo: MSI model: Z170-A PRO (MS-7971) v: 1.0<br>
serial: <superuser required> UEFI: American Megatrends
v: 1.K0<br>
date: 07/10/2018<br>
CPU:<br>
Info: quad core model: Intel Core i7-6700K bits: 64 type: MT
MCP cache:<br>
L2: 1024 KiB<br>
Speed (MHz): avg: 800 min/max: 800/4200 cores: 1: 800 2: 800
3: 800 4: 800<br>
5: 800 6: 800 7: 800 8: 800<br>
Graphics:<br>
Device-1: Intel HD Graphics 530 driver: i915 v: kernel<br>
Device-2: NVIDIA GM206 [GeForce GTX 960] driver: nvidia v:
550.100<br>
Display: wayland server: X.org v: 1.21.1.12 with: Xwayland v:
24.1.1<br>
compositor: gnome-shell driver: X: loaded: modesetting
unloaded: fbdev,vesa<br>
gpu: nvidia,nvidia-nvswitch resolution: 1920x1080~60Hz<br>
API: OpenGL v: 4.6.0 vendor: nvidia v: 550.100 renderer:
NVIDIA GeForce<br>
GTX 960/PCIe/SSE2<br>
API: Vulkan v: 1.3.290 drivers: N/A surfaces: xcb,xlib,wayland<br>
API: EGL Message: EGL data requires eglinfo. Check
--recommends.<br>
<br>
<br>
lspci -k | grep -EA3 'VGA|3D|Display'<br>
00:02.0 Display controller: Intel Corporation HD Graphics 530
(rev 06)<br>
DeviceName: Onboard IGD<br>
Subsystem: Micro-Star International Co., Ltd. [MSI] Device
7971<br>
Kernel driver in use: i915<br>
--<br>
01:00.0 VGA compatible controller: NVIDIA Corporation GM206
[GeForce GTX 960] (rev a1)<br>
Subsystem: Micro-Star International Co., Ltd. [MSI] Device
3201<br>
Kernel driver in use: nvidia<br>
Kernel modules: nouveau, nvidia_drm, nvidia<br>
<br>
=====================================================================<br>
<br>
</font><font face="Courier New, Courier, monospace">Trying to test
QSV via FFmpeg benchmark similar like this Vulkan code<br>
ffmpeg -init_hw_device "vulkan=vk:0" -hwaccel vulkan
-hwaccel_output_format vulkan -i INPUT -f null - -benchmark<br>
<a href="https://trac.ffmpeg.org/wiki/HWAccelIntro#Vulkan" target="_blank">https://trac.ffmpeg.org/wiki/HWAccelIntro#Vulkan</a><br>
<br>
cause an output error which reports: "device type qsv needed".<br>
Intel Skylake with HD 530 graphics gen 9, should be capable to
decode and encode MPEG-2/H.264/H.265 according to<br>
<a href="https://trac.ffmpeg.org/wiki/Hardware/QuickSync#HardwareSupport" target="_blank">https://trac.ffmpeg.org/wiki/Hardware/QuickSync#HardwareSupport</a><br>
</font><font face="Courier New, Courier, monospace"><br>
ffmpeg -hide_banner -hwaccel qsv -hwaccel_output_format qsv -i
h264_8bit_yuv420p.mp4 -f null - -benchmark<br>
</font><font face="monospace">[AVHWDeviceContext @ 0x557526873500]
Error creating a MFX session: -9.<br>
Device creation failed: -1313558101.<br>
[vist#0:0/h264 @ 0x55752685c6c0] [dec:h264_qsv @ 0x55752685b300]
No device available for decoder: device type qsv needed for
codec h264_qsv.<br>
[vist#0:0/h264 @ 0x55752685c6c0] [dec:h264_qsv @ 0x55752685b300]
Hardware device setup failed for decoder: Unknown error occurred<br>
[vost#0:0/wrapped_avframe @ 0x55752685a740] Error initializing a
simple filtergraph<br>
Error opening output file -.<br>
Error opening output files: Unknown error occurred<br>
bench: maxrss=67452KiB<br>
<br>
<br>
</font>I have tried to define and add a qsv device according to
the last section here<br>
<a href="https://trac.ffmpeg.org/wiki/Hardware/QuickSync#Transcode" target="_blank">https://trac.ffmpeg.org/wiki/Hardware/QuickSync#Transcode</a><br>
<p> <code>-qsv_device</code> is an qsv customized option can be
used to specify a hardware device and avoid the default device
initialization failure when multiple devices usable (eg: an
Intel integrated GPU and an AMD/Nvidia discrete graphics card).
One example on Linux (more details please see <a href="https://trac.ffmpeg.org/ticket/7649" target="_blank"><span></span>https://trac.ffmpeg.org/ticket/7649</a>)
<br>
</p>
<p><font face="Courier New, Courier, monospace">ffmpeg -hwaccel
qsv -qsv_device /dev/dri/renderD128 -c:v h264_qsv -i input.mp4
-c:v h264_qsv output.mp4</font></p>
and also had a look at this answer<br>
<a href="https://stackoverflow.com/questions/44747381/i-cant-use-hw-encoder-of-qsv-by-ffmpeg#47145215" target="_blank">https://stackoverflow.com/questions/44747381/i-cant-use-hw-encoder-of-qsv-by-ffmpeg#47145215</a>
<br>
</blockquote>
<br>
..... snip<br>
<br>
<br>
In the meantime I have also queried this QSV issue on the
ffmpeg-user mailing list, where I got some hints that helped me to
solve it.<br>
So to not leave this as an open issue here, I repost my reply to
that list:<br>
<br>
Yes, now I have verified that "libmfx" runtime was not installed by
default on the Skylake workstation (possibly because I enabled its
iGPU in BIOS afterwards).<br>
My ffmpeg had "libvpl" enabled, and "libmfx-gen1_2" was installed by
default. When<br>
<br>
When I add-installed "libmfx1", QSV via FFmpeg worked as it should:<br>
<br>
<font face="Courier New, Courier, monospace">ffmpeg -hide_banner
-hwaccel qsv -hwaccel_output_format qsv -i h264_8bit_yuv420p.mp4
-f null -</font><br>
<br>
1) On Skylake i7-6700K / HD-530 iGPU:<br>
<font face="Courier New, Courier, monospace">frame= 1780 fps=882
q=-0.0 Lsize=N/A time=00:01:11.20 bitrate=N/A speed=35.3x </font><br>
<br>
2) On KabyLake i7-8550U / UHD-620 iGPU:<br>
<font face="Courier New, Courier, monospace">frame= 1780 fps=795
q=-0.0 Lsize=N/A time=00:01:11.20 bitrate=N/A speed=31.8x </font><br>
<br>
3) On AlderLake i7-12700KF + DG2 A750 GPU:<br>
<font face="Courier New, Courier, monospace">frame= 1780 fps=833
q=-0.0 Lsize=N/A time=00:01:11.20 bitrate=N/A speed=33.3x </font><br>
<br>
What surprise me here, is that the current most powerful workstation
3), is not faster here than the legacy 1) and 2)<br>
<br>
<br>
References:<br>
<a href="https://github.com/intel/libvpl?tab=readme-ov-file#architecture" target="_blank">https://github.com/intel/libvpl?tab=readme-ov-file#architecture</a><br>
<a href="https://github.com/Intel-Media-SDK/MediaSDK?tab=readme-ov-file#id10" target="_blank">https://github.com/Intel-Media-SDK/MediaSDK?tab=readme-ov-file#id10</a><br>
<a href="https://wiki.archlinux.org/title/FFmpeg#Intel_QuickSync_(QSV)" target="_blank">https://wiki.archlinux.org/title/FFmpeg#Intel_QuickSync_(QSV)</a><br>
<br>
And as orientation for possibly other QSV interested in reading more
of this thread at ffmpeg-user, here is the url:<br>
<a href="https://ffmpeg.org/pipermail/ffmpeg-user/2024-September/058677.html" target="_blank">https://ffmpeg.org/pipermail/ffmpeg-user/2024-September/058677.html</a><br>
<br>
</div>
-- <br>
Cin mailing list<br>
<a href="mailto:Cin@lists.cinelerra-gg.org" target="_blank">Cin@lists.cinelerra-gg.org</a><br>
<a href="https://lists.cinelerra-gg.org/mailman/listinfo/cin" rel="noreferrer" target="_blank">https://lists.cinelerra-gg.org/mailman/listinfo/cin</a><br>
</blockquote></div>