1
0
mirror of https://github.com/mpv-player/mpv synced 2025-01-20 06:11:10 +00:00
Commit Graph

13 Commits

Author SHA1 Message Date
wm4
dcd167ca37 dxva2: fix handling of cropped video
Basically, we need to make sure to allocate enough data for the pretty
dumb copy_nv12 function. (It could be avoided by making the function
less dumb, but this fix is simpler.)
2015-07-06 21:25:07 +02:00
wm4
561416597e client API, dxva2: add a workaround for OpenGL fullscreen issues
This is basically a hack for drivers which prevent the mpv DXVA2 decoder
glue from working if OpenGL is in fullscreen mode.

Since it doesn't add any "hard" new API to the client API, some of the
code would be required for a true zero-copy hw decoding pipeline, and
sine it isn't too much code after all, this is probably acceptable.
2015-07-03 16:38:12 +02:00
wm4
b85321d057 vo_direct3d, dxva2: use the same D3D device
Since we still read-back (and don't have hard plans on changing this),
this doesn't have much of an advantage.
2015-07-03 16:04:42 +02:00
wm4
b6dc11810b dxva2: move device creation code
Preparation for the following commit.
2015-07-03 15:42:52 +02:00
wm4
a2235b0280 dxva2: support HEVC
This is pretty much copy&pasted from Libav commit
a7e0380497306d9723dec8440a4c52e8bf0263cf.

Note that if FFmpeg was not compiled with HEVC DXVA2 support or your
video drivers do not support HEVC, the player will not fallback and
just fail decoding any video. This is because libavcodec appears not
to return an error in this case. The situation is made worse by the
fact that MSYS2 is on an ancient MinGW-w64 release, which does not
have the required headers for HEVC DXVA2 support.
2015-05-14 17:44:48 +02:00
Avi Halachmi (:avih)
e2326bda82 dxva2: fix broken build with gcc 5.1
gpu_mempcy should to be called from code which targets SSE

Signed-off-by: wm4 <wm4@nowhere>
2015-04-24 17:40:39 +02:00
wm4
f61b8b312d win32: request UTF-16 API variants, Vista+ APIs, and COM C macros
Put the Vista+ (_WIN32_WINNT) and the COM C (COBJMACROS) defines into
the build system, instead of defining them over and over in the code.
2015-01-07 21:42:44 +01:00
James Ross-Gowan
a095958263 dxva2: copy mp_image attributes to sw_img 2015-01-07 23:00:22 +11:00
James Ross-Gowan
a95ce07273 dxva2: use gpu_memcpy with unmatched strides
Like memcpy_pic, this checks if the strides match first.
2014-10-27 12:37:25 +01:00
James Ross-Gowan
ac35e3b308 dxva2: fix copying surfaces with different stride 2014-10-27 10:56:00 +01:00
wm4
f22acd94af dxva2: fix crash on initialization failure
If dxva2_init() fails, dxva2_uninit() will be called twice.
2014-10-26 15:42:47 +01:00
James Ross-Gowan
52b52800ce dxva2: use optimized memcpy
At least on my machine, reading back the frame with system memcpy is
slower than just using software rendering. Use the optimized gpu_memcpy
from LAV to speed things up.
2014-10-26 02:34:15 +02:00
wm4
423a7de676 video: initial dxva2 support
Shamelessly stolen from ffmpeg. It probably doesn't work - you can debug
it yourself.
2014-10-25 19:25:22 +02:00