2012-03-30 23:13:38 +00:00
|
|
|
/*
|
2013-10-23 17:06:14 +00:00
|
|
|
* Based on vo_gl.c by Reimar Doeffinger.
|
|
|
|
*
|
2015-04-13 07:36:54 +00:00
|
|
|
* This file is part of mpv.
|
|
|
|
*
|
2016-01-19 17:36:34 +00:00
|
|
|
* mpv is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU Lesser General Public
|
|
|
|
* License as published by the Free Software Foundation; either
|
|
|
|
* version 2.1 of the License, or (at your option) any later version.
|
2012-03-30 23:13:38 +00:00
|
|
|
*
|
2015-04-13 07:36:54 +00:00
|
|
|
* mpv is distributed in the hope that it will be useful,
|
2012-03-30 23:13:38 +00:00
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
2016-01-19 17:36:34 +00:00
|
|
|
* GNU Lesser General Public License for more details.
|
2012-03-30 23:13:38 +00:00
|
|
|
*
|
2016-01-19 17:36:34 +00:00
|
|
|
* You should have received a copy of the GNU Lesser General Public
|
|
|
|
* License along with mpv. If not, see <http://www.gnu.org/licenses/>.
|
2012-03-30 23:13:38 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <math.h>
|
|
|
|
#include <stdbool.h>
|
|
|
|
#include <assert.h>
|
|
|
|
|
2012-08-15 20:23:02 +00:00
|
|
|
#include <libavutil/common.h>
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2016-01-11 18:03:40 +00:00
|
|
|
#include "mpv_talloc.h"
|
2013-12-17 01:39:45 +00:00
|
|
|
#include "common/common.h"
|
2014-08-29 10:09:04 +00:00
|
|
|
#include "misc/bstr.h"
|
2013-12-17 01:39:45 +00:00
|
|
|
#include "common/msg.h"
|
2015-11-09 10:57:11 +00:00
|
|
|
#include "common/global.h"
|
2013-12-17 01:02:25 +00:00
|
|
|
#include "options/m_config.h"
|
2012-11-09 00:06:43 +00:00
|
|
|
#include "vo.h"
|
|
|
|
#include "video/mp_image.h"
|
2013-11-24 11:58:06 +00:00
|
|
|
#include "sub/osd.h"
|
2012-03-30 23:13:38 +00:00
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
#include "gpu/context.h"
|
|
|
|
#include "gpu/hwdec.h"
|
|
|
|
#include "gpu/video.h"
|
2016-09-02 13:59:40 +00:00
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv {
|
2015-01-26 01:11:14 +00:00
|
|
|
struct mp_log *log;
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct ra_ctx *ctx;
|
2016-09-02 13:59:40 +00:00
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
char *context_name;
|
|
|
|
char *context_type;
|
2013-03-01 20:19:20 +00:00
|
|
|
struct gl_video *renderer;
|
2013-11-03 23:00:18 +00:00
|
|
|
|
2016-06-05 10:21:47 +00:00
|
|
|
int events;
|
2013-03-01 20:19:20 +00:00
|
|
|
};
|
2017-10-17 07:09:00 +00:00
|
|
|
static void resize(struct vo *vo)
|
2012-03-30 23:13:38 +00:00
|
|
|
{
|
2017-10-17 07:09:00 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
vo_gpu: vulkan: initial implementation
This time based on ra/vo_gpu. 2017 is the year of the vulkan desktop!
Current problems / limitations / improvement opportunities:
1. The swapchain/flipping code violates the vulkan spec, by assuming
that the presentation queue will be bounded (in cases where rendering
is significantly faster than vsync). But apparently, there's simply
no better way to do this right now, to the point where even the
stupid cube.c examples from LunarG etc. do it wrong.
(cf. https://github.com/KhronosGroup/Vulkan-Docs/issues/370)
2. The memory allocator could be improved. (This is a universal
constant)
3. Could explore using push descriptors instead of descriptor sets,
especially since we expect to switch descriptors semi-often for some
passes (like interpolation). Probably won't make a difference, but
the synchronization overhead might be a factor. Who knows.
4. Parallelism across frames / async transfer is not well-defined, we
either need to use a better semaphore / command buffer strategy or a
resource pooling layer to safely handle cross-frame parallelism.
(That said, I gave resource pooling a try and was not happy with the
result at all - so I'm still exploring the semaphore strategy)
5. We aggressively use pipeline barriers where events would offer a much
more fine-grained synchronization mechanism. As a result of this, we
might be suffering from GPU bubbles due to too-short dependencies on
objects. (That said, I'm also exploring the use of semaphores as a an
ordering tactic which would allow cross-frame time slicing in theory)
Some minor changes to the vo_gpu and infrastructure, but nothing
consequential.
NOTE: For safety, all use of asynchronous commands / multiple command
pools is currently disabled completely. There are some left-over relics
of this in the code (e.g. the distinction between dev_poll and
pool_poll), but that is kept in place mostly because this will be
re-extended in the future (vulkan rev 2).
The queue count is also currently capped to 1, because of the lack of
cross-frame semaphores means we need the implicit synchronization from
the same-queue semantics to guarantee a correct result.
2016-09-14 18:54:18 +00:00
|
|
|
struct ra_swapchain *sw = p->ctx->swapchain;
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2013-07-31 19:44:21 +00:00
|
|
|
MP_VERBOSE(vo, "Resize: %dx%d\n", vo->dwidth, vo->dheight);
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2013-03-01 20:19:20 +00:00
|
|
|
struct mp_rect src, dst;
|
|
|
|
struct mp_osd_res osd;
|
|
|
|
vo_get_src_dst_rects(vo, &src, &dst, &osd);
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2017-08-07 17:14:18 +00:00
|
|
|
gl_video_resize(p->renderer, &src, &dst, &osd);
|
2012-03-30 23:13:38 +00:00
|
|
|
|
vo_gpu: vulkan: initial implementation
This time based on ra/vo_gpu. 2017 is the year of the vulkan desktop!
Current problems / limitations / improvement opportunities:
1. The swapchain/flipping code violates the vulkan spec, by assuming
that the presentation queue will be bounded (in cases where rendering
is significantly faster than vsync). But apparently, there's simply
no better way to do this right now, to the point where even the
stupid cube.c examples from LunarG etc. do it wrong.
(cf. https://github.com/KhronosGroup/Vulkan-Docs/issues/370)
2. The memory allocator could be improved. (This is a universal
constant)
3. Could explore using push descriptors instead of descriptor sets,
especially since we expect to switch descriptors semi-often for some
passes (like interpolation). Probably won't make a difference, but
the synchronization overhead might be a factor. Who knows.
4. Parallelism across frames / async transfer is not well-defined, we
either need to use a better semaphore / command buffer strategy or a
resource pooling layer to safely handle cross-frame parallelism.
(That said, I gave resource pooling a try and was not happy with the
result at all - so I'm still exploring the semaphore strategy)
5. We aggressively use pipeline barriers where events would offer a much
more fine-grained synchronization mechanism. As a result of this, we
might be suffering from GPU bubbles due to too-short dependencies on
objects. (That said, I'm also exploring the use of semaphores as a an
ordering tactic which would allow cross-frame time slicing in theory)
Some minor changes to the vo_gpu and infrastructure, but nothing
consequential.
NOTE: For safety, all use of asynchronous commands / multiple command
pools is currently disabled completely. There are some left-over relics
of this in the code (e.g. the distinction between dev_poll and
pool_poll), but that is kept in place mostly because this will be
re-extended in the future (vulkan rev 2).
The queue count is also currently capped to 1, because of the lack of
cross-frame semaphores means we need the implicit synchronization from
the same-queue semantics to guarantee a correct result.
2016-09-14 18:54:18 +00:00
|
|
|
int fb_depth = sw->fns->color_depth ? sw->fns->color_depth(sw) : 0;
|
|
|
|
if (fb_depth)
|
|
|
|
MP_VERBOSE(p, "Reported display depth: %d\n", fb_depth);
|
|
|
|
gl_video_set_fb_depth(p->renderer, fb_depth);
|
|
|
|
|
2012-03-30 23:13:38 +00:00
|
|
|
vo->want_redraw = true;
|
|
|
|
}
|
|
|
|
|
2015-10-30 19:26:43 +00:00
|
|
|
static void draw_frame(struct vo *vo, struct vo_frame *frame)
|
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
|
|
|
struct ra_swapchain *sw = p->ctx->swapchain;
|
2017-03-20 12:37:47 +00:00
|
|
|
|
2017-09-20 08:45:33 +00:00
|
|
|
struct ra_fbo fbo;
|
|
|
|
if (!sw->fns->start_frame(sw, &fbo))
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
return;
|
2015-10-30 19:26:43 +00:00
|
|
|
|
2018-02-08 00:55:31 +00:00
|
|
|
gl_video_render_frame(p->renderer, frame, fbo, RENDER_FRAME_DEF);
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
if (!sw->fns->submit_frame(sw, frame)) {
|
|
|
|
MP_ERR(vo, "Failed presenting frame!\n");
|
|
|
|
return;
|
|
|
|
}
|
2015-10-30 19:26:43 +00:00
|
|
|
}
|
|
|
|
|
2012-03-30 23:13:38 +00:00
|
|
|
static void flip_page(struct vo *vo)
|
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
|
|
|
struct ra_swapchain *sw = p->ctx->swapchain;
|
|
|
|
sw->fns->swap_buffers(sw);
|
2012-03-30 23:13:38 +00:00
|
|
|
}
|
|
|
|
|
2018-08-31 18:08:08 +00:00
|
|
|
static void get_vsync(struct vo *vo, struct vo_vsync_info *info)
|
2018-08-31 14:33:15 +00:00
|
|
|
{
|
|
|
|
struct gpu_priv *p = vo->priv;
|
|
|
|
struct ra_swapchain *sw = p->ctx->swapchain;
|
2018-08-31 18:08:08 +00:00
|
|
|
if (sw->fns->get_vsync)
|
|
|
|
sw->fns->get_vsync(sw, info);
|
2018-08-31 14:33:15 +00:00
|
|
|
}
|
|
|
|
|
2015-01-21 21:08:24 +00:00
|
|
|
static int query_format(struct vo *vo, int format)
|
2012-03-30 23:13:38 +00:00
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
2013-11-03 23:00:18 +00:00
|
|
|
if (!gl_video_check_format(p->renderer, format))
|
2012-03-30 23:13:38 +00:00
|
|
|
return 0;
|
2015-01-21 21:08:24 +00:00
|
|
|
return 1;
|
2012-03-30 23:13:38 +00:00
|
|
|
}
|
|
|
|
|
2015-10-03 16:20:16 +00:00
|
|
|
static int reconfig(struct vo *vo, struct mp_image_params *params)
|
2012-03-30 23:13:38 +00:00
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
2012-03-30 23:13:38 +00:00
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
if (!p->ctx->fns->reconfig(p->ctx))
|
2012-03-30 23:13:38 +00:00
|
|
|
return -1;
|
|
|
|
|
2017-10-17 07:09:00 +00:00
|
|
|
resize(vo);
|
2013-06-07 23:35:44 +00:00
|
|
|
gl_video_config(p->renderer, params);
|
2012-10-02 23:54:13 +00:00
|
|
|
|
2012-03-30 23:13:38 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
vo_gpu: hwdec: load hwdec interops on-demand by default
Historically, we have treated hwdec interop loading as a completely
separate step from loading the hwdecs themselves. Some hwdecs need an
interop, and some don't, and users generally configure the exact
hwdec they want, so interops that aren't relevant for that hwdec
shouldn't be loaded to save time and avoid warning/error spam.
The basic approach here is to recognise that interops are tied to
hwdecs by imgfmt. The hwdec outputs some format, and an interop is
needed to get that format to the vo without read back.
So, when we try to load an hwdec, instead of just blindly loading all
interops as we do today, let's pass the imgfmt in and only load
interops that work for that format. If more than one interop is
available for the format, the existing logic (whatever it is) will
continue to be used to pick one.
We also have one callsite in filters where we seem to pre-emptively
load all the interops. It's probably possible to trace down a specific
format but for now I'm just letting it keep loading all of them; it's
no worse than before.
You may notice there is no documentation update - and that's because
the current docs say that when the interop mode is `auto`, the interop
is loaded on demand. So reality now reflects the docs. How nice.
2022-02-06 22:18:59 +00:00
|
|
|
static void request_hwdec_api(struct vo *vo, void *data)
|
2013-11-03 23:00:18 +00:00
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
2022-03-20 19:31:38 +00:00
|
|
|
gl_video_load_hwdecs_for_img_fmt(p->renderer, vo->hwdec_devs, data);
|
2013-11-03 23:00:18 +00:00
|
|
|
}
|
|
|
|
|
2022-03-20 19:31:38 +00:00
|
|
|
static void call_request_hwdec_api(void *ctx,
|
|
|
|
struct hwdec_imgfmt_request *params)
|
2013-11-03 23:00:18 +00:00
|
|
|
{
|
video: move display and timing to a separate thread
The VO is run inside its own thread. It also does most of video timing.
The playloop hands the image data and a realtime timestamp to the VO,
and the VO does the rest.
In particular, this allows the playloop to do other things, instead of
blocking for video redraw. But if anything accesses the VO during video
timing, it will block.
This also fixes vo_sdl.c event handling; but that is only a side-effect,
since reimplementing the broken way would require more effort.
Also drop --softsleep. In theory, this option helps if the kernel's
sleeping mechanism is too inaccurate for video timing. In practice, I
haven't ever encountered a situation where it helps, and it just burns
CPU cycles. On the other hand it's probably actively harmful, because
it prevents the libavcodec decoder threads from doing real work.
Side note:
Originally, I intended that multiple frames can be queued to the VO. But
this is not done, due to problems with OSD and other certain features.
OSD in particular is simply designed in a way that it can be neither
timed nor copied, so you do have to render it into the video frame
before you can draw the next frame. (Subtitles have no such restriction.
sd_lavc was even updated to fix this.) It seems the right solution to
queuing multiple VO frames is rendering on VO-backed framebuffers, like
vo_vdpau.c does. This requires VO driver support, and is out of scope
of this commit.
As consequence, the VO has a queue size of 1. The existing video queue
is just needed to compute frame duration, and will be moved out in the
next commit.
2014-08-12 21:02:08 +00:00
|
|
|
// Roundabout way to run hwdec loading on the VO thread.
|
|
|
|
// Redirects to request_hwdec_api().
|
2022-03-20 19:31:38 +00:00
|
|
|
vo_control(ctx, VOCTRL_LOAD_HWDEC_API, params);
|
2013-11-03 23:00:18 +00:00
|
|
|
}
|
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
static void get_and_update_icc_profile(struct gpu_priv *p)
|
2014-03-24 22:30:12 +00:00
|
|
|
{
|
2016-06-03 18:35:17 +00:00
|
|
|
if (gl_video_icc_auto_enabled(p->renderer)) {
|
2015-01-26 01:11:14 +00:00
|
|
|
MP_VERBOSE(p, "Querying ICC profile...\n");
|
2015-03-07 11:32:19 +00:00
|
|
|
bstr icc = bstr0(NULL);
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
int r = p->ctx->fns->control(p->ctx, &p->events, VOCTRL_GET_ICC_PROFILE, &icc);
|
2015-01-26 01:11:14 +00:00
|
|
|
|
2015-03-07 11:32:19 +00:00
|
|
|
if (r != VO_NOTAVAIL) {
|
|
|
|
if (r == VO_FALSE) {
|
|
|
|
MP_WARN(p, "Could not retrieve an ICC profile.\n");
|
|
|
|
} else if (r == VO_NOTIMPL) {
|
|
|
|
MP_ERR(p, "icc-profile-auto not implemented on this platform.\n");
|
|
|
|
}
|
|
|
|
|
2016-06-03 18:35:17 +00:00
|
|
|
gl_video_set_icc_profile(p->renderer, icc);
|
2015-01-26 01:11:14 +00:00
|
|
|
}
|
|
|
|
}
|
2014-03-24 22:30:12 +00:00
|
|
|
}
|
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
static void get_and_update_ambient_lighting(struct gpu_priv *p)
|
2015-02-07 12:54:18 +00:00
|
|
|
{
|
|
|
|
int lux;
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
int r = p->ctx->fns->control(p->ctx, &p->events, VOCTRL_GET_AMBIENT_LUX, &lux);
|
2015-02-07 12:54:18 +00:00
|
|
|
if (r == VO_TRUE) {
|
|
|
|
gl_video_set_ambient_lux(p->renderer, lux);
|
|
|
|
}
|
2016-09-02 12:50:03 +00:00
|
|
|
if (r != VO_TRUE && gl_video_gamma_auto_enabled(p->renderer)) {
|
2015-02-07 12:54:18 +00:00
|
|
|
MP_ERR(p, "gamma_auto option provided, but querying for ambient"
|
|
|
|
" lighting is not supported on this platform\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-10-05 15:27:44 +00:00
|
|
|
static void update_ra_ctx_options(struct vo *vo)
|
|
|
|
{
|
|
|
|
struct gpu_priv *p = vo->priv;
|
|
|
|
|
|
|
|
/* Only the alpha option has any runtime toggle ability. */
|
|
|
|
struct gl_video_opts *gl_opts = mp_get_config_group(p->ctx, vo->global, &gl_video_conf);
|
|
|
|
p->ctx->opts.want_alpha = gl_opts->alpha_mode == 1;
|
|
|
|
}
|
|
|
|
|
2012-03-30 23:13:38 +00:00
|
|
|
static int control(struct vo *vo, uint32_t request, void *data)
|
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
2012-03-30 23:13:38 +00:00
|
|
|
|
|
|
|
switch (request) {
|
|
|
|
case VOCTRL_SET_PANSCAN:
|
2017-10-17 07:09:00 +00:00
|
|
|
resize(vo);
|
2012-03-30 23:13:38 +00:00
|
|
|
return VO_TRUE;
|
video: redo video equalizer option handling
I really wouldn't care much about this, but some parts of the core code
are under HAVE_GPL, so there's some need to get rid of it. Simply turn
the video equalizer from its current fine-grained handling with vf/vo
fallbacks into global options. This makes updating them much simpler.
This removes any possibility of applying video equalizers in filters,
which affects vf_scale, and the previously removed vf_eq. Not a big
loss, since the preferred VOs have this builtin.
Remove video equalizer handling from vo_direct3d, vo_sdl, vo_vaapi, and
vo_xv. I'm not going to waste my time on these legacy VOs.
vo.eq_opts_cache exists _only_ to send a VOCTRL_SET_EQUALIZER, which
exists _only_ to trigger a redraw. This seems silly, but for now I feel
like this is less of a pain. The rest of the equalizer using code is
self-updating.
See commit 96b906a51d5 for how some video equalizer code was GPL only.
Some command line option names and ranges can probably be traced back to
a GPL only committer, but we don't consider these copyrightable.
2017-08-22 15:01:35 +00:00
|
|
|
case VOCTRL_SET_EQUALIZER:
|
|
|
|
vo->want_redraw = true;
|
|
|
|
return VO_TRUE;
|
2018-02-07 19:18:36 +00:00
|
|
|
case VOCTRL_SCREENSHOT: {
|
|
|
|
struct vo_frame *frame = vo_get_current_vo_frame(vo);
|
2018-04-20 15:05:53 +00:00
|
|
|
if (frame)
|
2018-02-07 19:18:36 +00:00
|
|
|
gl_video_screenshot(p->renderer, frame, data);
|
|
|
|
talloc_free(frame);
|
|
|
|
return true;
|
|
|
|
}
|
video: move display and timing to a separate thread
The VO is run inside its own thread. It also does most of video timing.
The playloop hands the image data and a realtime timestamp to the VO,
and the VO does the rest.
In particular, this allows the playloop to do other things, instead of
blocking for video redraw. But if anything accesses the VO during video
timing, it will block.
This also fixes vo_sdl.c event handling; but that is only a side-effect,
since reimplementing the broken way would require more effort.
Also drop --softsleep. In theory, this option helps if the kernel's
sleeping mechanism is too inaccurate for video timing. In practice, I
haven't ever encountered a situation where it helps, and it just burns
CPU cycles. On the other hand it's probably actively harmful, because
it prevents the libavcodec decoder threads from doing real work.
Side note:
Originally, I intended that multiple frames can be queued to the VO. But
this is not done, due to problems with OSD and other certain features.
OSD in particular is simply designed in a way that it can be neither
timed nor copied, so you do have to render it into the video frame
before you can draw the next frame. (Subtitles have no such restriction.
sd_lavc was even updated to fix this.) It seems the right solution to
queuing multiple VO frames is rendering on VO-backed framebuffers, like
vo_vdpau.c does. This requires VO driver support, and is out of scope
of this commit.
As consequence, the VO has a queue size of 1. The existing video queue
is just needed to compute frame duration, and will be moved out in the
next commit.
2014-08-12 21:02:08 +00:00
|
|
|
case VOCTRL_LOAD_HWDEC_API:
|
vo_gpu: hwdec: load hwdec interops on-demand by default
Historically, we have treated hwdec interop loading as a completely
separate step from loading the hwdecs themselves. Some hwdecs need an
interop, and some don't, and users generally configure the exact
hwdec they want, so interops that aren't relevant for that hwdec
shouldn't be loaded to save time and avoid warning/error spam.
The basic approach here is to recognise that interops are tied to
hwdecs by imgfmt. The hwdec outputs some format, and an interop is
needed to get that format to the vo without read back.
So, when we try to load an hwdec, instead of just blindly loading all
interops as we do today, let's pass the imgfmt in and only load
interops that work for that format. If more than one interop is
available for the format, the existing logic (whatever it is) will
continue to be used to pick one.
We also have one callsite in filters where we seem to pre-emptively
load all the interops. It's probably possible to trace down a specific
format but for now I'm just letting it keep loading all of them; it's
no worse than before.
You may notice there is no documentation update - and that's because
the current docs say that when the interop mode is `auto`, the interop
is loaded on demand. So reality now reflects the docs. How nice.
2022-02-06 22:18:59 +00:00
|
|
|
request_hwdec_api(vo, data);
|
video: move display and timing to a separate thread
The VO is run inside its own thread. It also does most of video timing.
The playloop hands the image data and a realtime timestamp to the VO,
and the VO does the rest.
In particular, this allows the playloop to do other things, instead of
blocking for video redraw. But if anything accesses the VO during video
timing, it will block.
This also fixes vo_sdl.c event handling; but that is only a side-effect,
since reimplementing the broken way would require more effort.
Also drop --softsleep. In theory, this option helps if the kernel's
sleeping mechanism is too inaccurate for video timing. In practice, I
haven't ever encountered a situation where it helps, and it just burns
CPU cycles. On the other hand it's probably actively harmful, because
it prevents the libavcodec decoder threads from doing real work.
Side note:
Originally, I intended that multiple frames can be queued to the VO. But
this is not done, due to problems with OSD and other certain features.
OSD in particular is simply designed in a way that it can be neither
timed nor copied, so you do have to render it into the video frame
before you can draw the next frame. (Subtitles have no such restriction.
sd_lavc was even updated to fix this.) It seems the right solution to
queuing multiple VO frames is rendering on VO-backed framebuffers, like
vo_vdpau.c does. This requires VO driver support, and is out of scope
of this commit.
As consequence, the VO has a queue size of 1. The existing video queue
is just needed to compute frame duration, and will be moved out in the
next commit.
2014-08-12 21:02:08 +00:00
|
|
|
return true;
|
2016-09-02 13:59:40 +00:00
|
|
|
case VOCTRL_UPDATE_RENDER_OPTS: {
|
2020-10-05 15:27:44 +00:00
|
|
|
update_ra_ctx_options(vo);
|
2017-10-17 07:09:00 +00:00
|
|
|
gl_video_configure_queue(p->renderer, vo);
|
2017-11-01 00:58:16 +00:00
|
|
|
get_and_update_icc_profile(p);
|
2020-10-15 17:18:59 +00:00
|
|
|
if (p->ctx->fns->update_render_opts)
|
|
|
|
p->ctx->fns->update_render_opts(p->ctx);
|
2017-10-17 07:09:00 +00:00
|
|
|
vo->want_redraw = true;
|
2016-09-02 13:59:40 +00:00
|
|
|
return true;
|
2012-07-31 23:35:58 +00:00
|
|
|
}
|
2014-11-23 19:06:05 +00:00
|
|
|
case VOCTRL_RESET:
|
|
|
|
gl_video_reset(p->renderer);
|
|
|
|
return true;
|
2015-02-04 22:37:38 +00:00
|
|
|
case VOCTRL_PAUSE:
|
2017-02-21 13:31:51 +00:00
|
|
|
if (gl_video_showing_interpolated_frame(p->renderer))
|
2015-02-04 22:37:38 +00:00
|
|
|
vo->want_redraw = true;
|
2019-10-07 12:48:34 +00:00
|
|
|
return true;
|
2016-06-06 00:44:15 +00:00
|
|
|
case VOCTRL_PERFORMANCE_DATA:
|
vo_opengl: refactor vo performance subsystem
This replaces `vo-performance` by `vo-passes`, bringing with it a number
of changes and improvements:
1. mpv users can now introspect the vo_opengl passes, which is something
that has been requested multiple times.
2. performance data is now measured per-pass, which helps both
development and debugging.
3. since adding more passes is cheap, we can now report information for
more passes (e.g. the blit pass, and the osd pass). Note: we also
switch to nanosecond scale, to be able to measure these passes
better.
4. `--user-shaders` authors can now describe their own passes, helping
users both identify which user shaders are active at any given time
as well as helping shader authors identify performance issues.
5. the timing data per pass is now exported as a full list of samples,
so projects like Argon-/mpv-stats can immediately read out all of the
samples and render a graph without having to manually poll this
option constantly.
Due to gl_timer's design being complicated (directly reading performance
data would block, so we delay the actual read-back until the next _start
command), it's vital not to conflate different passes that might be
doing different things from one frame to another. To accomplish this,
the actual timers are stored as part of the gl_shader_cache's sc_entry,
which makes them unique for that exact shader.
Starting and stopping the time measurement is easy to unify with the
gl_sc architecture, because the existing API already relies on a
"generate, render, reset" flow, so we can just put timer_start and
timer_stop in sc_generate and sc_reset, respectively.
The ugliest thing about this code is that due to the need to keep pass
information relatively stable in between frames, we need to distinguish
between "new" and "redrawn" frames, which bloats the code somewhat and
also feels hacky and vo_opengl-specific. (But then again, this entire
thing is vo_opengl-specific)
2017-06-29 15:00:06 +00:00
|
|
|
gl_video_perfdata(p->renderer, (struct voctrl_performance_data *)data);
|
2016-06-06 00:44:15 +00:00
|
|
|
return true;
|
2017-12-26 00:38:32 +00:00
|
|
|
case VOCTRL_EXTERNAL_RESIZE:
|
|
|
|
p->ctx->fns->reconfig(p->ctx);
|
|
|
|
resize(vo);
|
|
|
|
return true;
|
2012-03-30 23:13:38 +00:00
|
|
|
}
|
2013-05-15 16:17:18 +00:00
|
|
|
|
|
|
|
int events = 0;
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
int r = p->ctx->fns->control(p->ctx, &events, request, data);
|
2015-01-26 01:21:00 +00:00
|
|
|
if (events & VO_EVENT_ICC_PROFILE_CHANGED) {
|
2016-06-05 10:21:47 +00:00
|
|
|
get_and_update_icc_profile(p);
|
2015-01-08 17:27:22 +00:00
|
|
|
vo->want_redraw = true;
|
|
|
|
}
|
2015-02-07 12:54:18 +00:00
|
|
|
if (events & VO_EVENT_AMBIENT_LIGHTING_CHANGED) {
|
2016-06-05 10:21:47 +00:00
|
|
|
get_and_update_ambient_lighting(p);
|
2015-02-07 12:54:18 +00:00
|
|
|
vo->want_redraw = true;
|
|
|
|
}
|
2016-06-05 10:21:47 +00:00
|
|
|
events |= p->events;
|
|
|
|
p->events = 0;
|
2013-05-15 16:17:18 +00:00
|
|
|
if (events & VO_EVENT_RESIZE)
|
2017-10-17 07:09:00 +00:00
|
|
|
resize(vo);
|
2013-05-15 16:17:18 +00:00
|
|
|
if (events & VO_EVENT_EXPOSE)
|
|
|
|
vo->want_redraw = true;
|
2014-11-02 19:26:51 +00:00
|
|
|
vo_event(vo, events);
|
2013-05-15 16:17:18 +00:00
|
|
|
|
|
|
|
return r;
|
2012-03-30 23:13:38 +00:00
|
|
|
}
|
|
|
|
|
2016-07-20 18:42:30 +00:00
|
|
|
static void wakeup(struct vo *vo)
|
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
|
|
|
if (p->ctx && p->ctx->fns->wakeup)
|
|
|
|
p->ctx->fns->wakeup(p->ctx);
|
2016-07-20 18:42:30 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void wait_events(struct vo *vo, int64_t until_time_us)
|
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
|
|
|
if (p->ctx && p->ctx->fns->wait_events) {
|
|
|
|
p->ctx->fns->wait_events(p->ctx, until_time_us);
|
2016-07-20 18:42:30 +00:00
|
|
|
} else {
|
|
|
|
vo_wait_default(vo, until_time_us);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-07-23 07:41:51 +00:00
|
|
|
static struct mp_image *get_image(struct vo *vo, int imgfmt, int w, int h,
|
2023-01-21 14:21:49 +00:00
|
|
|
int stride_align, int flags)
|
2017-07-23 07:41:51 +00:00
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
2017-07-23 07:41:51 +00:00
|
|
|
|
2023-01-21 14:21:49 +00:00
|
|
|
return gl_video_get_image(p->renderer, imgfmt, w, h, stride_align, flags);
|
2017-07-23 07:41:51 +00:00
|
|
|
}
|
|
|
|
|
2012-03-30 23:13:38 +00:00
|
|
|
static void uninit(struct vo *vo)
|
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2014-12-03 21:37:39 +00:00
|
|
|
gl_video_uninit(p->renderer);
|
2016-05-11 15:04:30 +00:00
|
|
|
if (vo->hwdec_devs) {
|
|
|
|
hwdec_devices_set_loader(vo->hwdec_devs, NULL, NULL);
|
|
|
|
hwdec_devices_destroy(vo->hwdec_devs);
|
|
|
|
}
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
ra_ctx_destroy(&p->ctx);
|
2012-03-30 23:13:38 +00:00
|
|
|
}
|
|
|
|
|
2013-07-22 20:52:42 +00:00
|
|
|
static int preinit(struct vo *vo)
|
2012-03-30 23:13:38 +00:00
|
|
|
{
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
struct gpu_priv *p = vo->priv;
|
2015-01-26 01:11:14 +00:00
|
|
|
p->log = vo->log;
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2021-04-18 09:48:35 +00:00
|
|
|
struct ra_ctx_opts *ctx_opts = mp_get_config_group(vo, vo->global, &ra_ctx_conf);
|
|
|
|
struct gl_video_opts *gl_opts = mp_get_config_group(vo, vo->global, &gl_video_conf);
|
|
|
|
struct ra_ctx_opts opts = *ctx_opts;
|
2020-10-05 15:27:44 +00:00
|
|
|
opts.want_alpha = gl_opts->alpha_mode == 1;
|
2021-04-18 09:48:35 +00:00
|
|
|
p->ctx = ra_ctx_create(vo, opts);
|
|
|
|
talloc_free(ctx_opts);
|
2021-05-07 12:52:46 +00:00
|
|
|
talloc_free(gl_opts);
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
if (!p->ctx)
|
2014-12-22 11:49:20 +00:00
|
|
|
goto err_out;
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
assert(p->ctx->ra);
|
|
|
|
assert(p->ctx->swapchain);
|
2017-08-07 17:14:18 +00:00
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
p->renderer = gl_video_init(p->ctx->ra, vo->log, vo->global);
|
2015-03-23 15:32:59 +00:00
|
|
|
gl_video_set_osd_source(p->renderer, vo->osd);
|
2015-07-16 20:43:40 +00:00
|
|
|
gl_video_configure_queue(p->renderer, vo);
|
2015-01-07 17:47:27 +00:00
|
|
|
|
2016-06-05 10:21:47 +00:00
|
|
|
get_and_update_icc_profile(p);
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2016-05-09 17:42:03 +00:00
|
|
|
vo->hwdec_devs = hwdec_devices_create();
|
|
|
|
hwdec_devices_set_loader(vo->hwdec_devs, call_request_hwdec_api, vo);
|
2014-08-11 21:08:35 +00:00
|
|
|
|
2022-03-12 18:54:23 +00:00
|
|
|
gl_video_init_hwdecs(p->renderer, p->ctx, vo->hwdec_devs, false);
|
2015-07-07 12:53:58 +00:00
|
|
|
|
2013-03-01 20:19:20 +00:00
|
|
|
return 0;
|
2012-03-30 23:13:38 +00:00
|
|
|
|
2013-03-01 20:19:20 +00:00
|
|
|
err_out:
|
|
|
|
uninit(vo);
|
2012-03-30 23:13:38 +00:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
const struct vo_driver video_out_gpu = {
|
|
|
|
.description = "Shader-based GPU Renderer",
|
|
|
|
.name = "gpu",
|
|
|
|
.caps = VO_CAP_ROTATE90,
|
|
|
|
.preinit = preinit,
|
|
|
|
.query_format = query_format,
|
|
|
|
.reconfig = reconfig,
|
|
|
|
.control = control,
|
|
|
|
.get_image = get_image,
|
|
|
|
.draw_frame = draw_frame,
|
|
|
|
.flip_page = flip_page,
|
2018-08-31 18:08:08 +00:00
|
|
|
.get_vsync = get_vsync,
|
vo_opengl: refactor into vo_gpu
This is done in several steps:
1. refactor MPGLContext -> struct ra_ctx
2. move GL-specific stuff in vo_opengl into opengl/context.c
3. generalize context creation to support other APIs, and add --gpu-api
4. rename all of the --opengl- options that are no longer opengl-specific
5. move all of the stuff from opengl/* that isn't GL-specific into gpu/
(note: opengl/gl_utils.h became opengl/utils.h)
6. rename vo_opengl to vo_gpu
7. to handle window screenshots, the short-term approach was to just add
it to ra_swchain_fns. Long term (and for vulkan) this has to be moved to
ra itself (and vo_gpu altered to compensate), but this was a stop-gap
measure to prevent this commit from getting too big
8. move ra->fns->flush to ra_gl_ctx instead
9. some other minor changes that I've probably already forgotten
Note: This is one half of a major refactor, the other half of which is
provided by rossy's following commit. This commit enables support for
all linux platforms, while his version enables support for all non-linux
platforms.
Note 2: vo_opengl_cb.c also re-uses ra_gl_ctx so it benefits from the
--opengl- options like --opengl-early-flush, --opengl-finish etc. Should
be a strict superset of the old functionality.
Disclaimer: Since I have no way of compiling mpv on all platforms, some
of these ports were done blindly. Specifically, the blind ports included
context_mali_fbdev.c and context_rpi.c. Since they're both based on
egl_helpers, the port should have gone smoothly without any major
changes required. But if somebody complains about a compile error on
those platforms (assuming anybody actually uses them), you know where to
complain.
2017-09-14 06:04:55 +00:00
|
|
|
.wait_events = wait_events,
|
|
|
|
.wakeup = wakeup,
|
|
|
|
.uninit = uninit,
|
|
|
|
.priv_size = sizeof(struct gpu_priv),
|
|
|
|
};
|