2017-08-04 17:09:46 +00:00
|
|
|
#pragma once
|
2015-01-28 18:40:46 +00:00
|
|
|
|
2017-08-04 17:09:46 +00:00
|
|
|
#include <stdbool.h>
|
2017-07-29 18:14:48 +00:00
|
|
|
#include <math.h>
|
|
|
|
|
2017-09-10 22:27:27 +00:00
|
|
|
#include "video/out/vo.h"
|
2017-07-29 18:14:48 +00:00
|
|
|
#include "ra.h"
|
2015-01-28 18:40:46 +00:00
|
|
|
|
2015-03-13 20:14:18 +00:00
|
|
|
// A 3x2 matrix, with the translation part separate.
|
|
|
|
struct gl_transform {
|
2016-03-28 14:16:09 +00:00
|
|
|
// row-major, e.g. in mathematical notation:
|
|
|
|
// | m[0][0] m[0][1] |
|
|
|
|
// | m[1][0] m[1][1] |
|
2015-03-13 20:14:18 +00:00
|
|
|
float m[2][2];
|
|
|
|
float t[2];
|
|
|
|
};
|
|
|
|
|
vo_opengl: refactor pass_read_video and texture binding
This is a pretty major rewrite of the internal texture binding
mechanic, which makes it more flexible.
In general, the difference between the old and current approaches is
that now, all texture description is held in a struct img_tex and only
explicitly bound with pass_bind. (Once bound, a texture unit is assumed
to be set in stone and no longer tied to the img_tex)
This approach makes the code inside pass_read_video significantly more
flexible and cuts down on the number of weird special cases and
spaghetti logic.
It also has some improvements, e.g. cutting down greatly on the number
of unnecessary conversion passes inside pass_read_video (which was
previously mostly done to cope with the fact that the alternative would
have resulted in a combinatorial explosion of code complexity).
Some other notable changes (and potential improvements):
- texture expansion is now *always* handled in pass_read_video, and the
colormatrix never does this anymore. (Which means the code could
probably be removed from the colormatrix generation logic, modulo some
other VOs)
- struct fbo_tex now stores both its "physical" and "logical"
(configured) size, which cuts down on the amount of width/height
baggage on some function calls
- vo_opengl can now technically support textures with different bit
depths (e.g. 10 bit luma, 8 bit chroma) - but the APIs it queries
inside img_format.c doesn't export this (nor does ffmpeg support it,
really) so the status quo of using the same tex_mul for all planes is
kept.
- dumb_mode is now only needed because of the indirect_fbo being in the
main rendering pipeline. If we reintroduce p->use_indirect and thread
a transform through the entire program this could be skipped where
unnecessary, allowing for the removal of dumb_mode. But I'm not sure
how to do this in a clean way. (Which is part of why it got introduced
to begin with)
- It would be trivial to resurrect source-shader now (it would just be
one extra 'if' inside pass_read_video).
2016-03-05 10:29:19 +00:00
|
|
|
static const struct gl_transform identity_trans = {
|
|
|
|
.m = {{1.0, 0.0}, {0.0, 1.0}},
|
|
|
|
.t = {0.0, 0.0},
|
|
|
|
};
|
|
|
|
|
2015-03-13 20:14:18 +00:00
|
|
|
void gl_transform_ortho(struct gl_transform *t, float x0, float x1,
|
|
|
|
float y0, float y1);
|
2015-01-29 13:58:26 +00:00
|
|
|
|
vo_opengl: refactor shader generation (part 2)
This adds stuff related to gamma, linear light, sigmoid, BT.2020-CL,
etc, as well as color management. Also adds a new gamma function (gamma22).
This adds new parameters to configure the CMS settings, in particular
letting us target simple colorspaces without requiring usage of a 3DLUT.
This adds smoothmotion. Mostly working, but it's still sensitive to
timing issues. It's based on an actual queue now, but the queue size
is kept small to avoid larger amounts of latency.
Also makes “upscale before blending” the default strategy.
This is justified because the "render after blending" thing doesn't seme
to work consistently any way (introduces stutter due to the way vsync
timing works, or something), so this behavior is a bit closer to master
and makes pausing/unpausing less weird/jumpy.
This adds the remaining scalers, including bicubic_fast, sharpen3,
sharpen5, polar filters and antiringing. Apparently, sharpen3/5 also
consult scale-param1, which was undocumented in master.
This also implements cropping and chroma transformation, plus
rotation/flipping. These are inherently part of the same logic, although
it's a bit rough around the edges in some case, mainly due to the fallback
code paths (for bilinear scaling without indirection).
2015-03-12 21:18:16 +00:00
|
|
|
// This treats m as an affine transformation, in other words m[2][n] gets
|
|
|
|
// added to the output.
|
2015-03-13 20:14:18 +00:00
|
|
|
static inline void gl_transform_vec(struct gl_transform t, float *x, float *y)
|
vo_opengl: refactor shader generation (part 1)
The basic idea is to use dynamically generated shaders instead of a
single monolithic file + a ton of ifdefs. Instead of having to setup
every aspect of it separately (like compiling shaders, setting uniforms,
perfoming the actual rendering steps, the GLSL parts), we generate the
GLSL on the fly, and perform the rendering at the same time. The GLSL
is regenerated every frame, but the actual compiled OpenGL-level shaders
are cached, which makes it fast again. Almost all logic can be in a
single place.
The new code is significantly more flexible, which allows us to improve
the code clarity, performance and add more features easily.
This commit is incomplete. It drops almost all previous code, and
readds only the most important things (some of them actually buggy).
The next commit will complete it - it's separate to preserve authorship
information.
2015-03-12 20:57:54 +00:00
|
|
|
{
|
|
|
|
float vx = *x, vy = *y;
|
2016-03-28 14:16:09 +00:00
|
|
|
*x = vx * t.m[0][0] + vy * t.m[0][1] + t.t[0];
|
|
|
|
*y = vx * t.m[1][0] + vy * t.m[1][1] + t.t[1];
|
vo_opengl: refactor shader generation (part 1)
The basic idea is to use dynamically generated shaders instead of a
single monolithic file + a ton of ifdefs. Instead of having to setup
every aspect of it separately (like compiling shaders, setting uniforms,
perfoming the actual rendering steps, the GLSL parts), we generate the
GLSL on the fly, and perform the rendering at the same time. The GLSL
is regenerated every frame, but the actual compiled OpenGL-level shaders
are cached, which makes it fast again. Almost all logic can be in a
single place.
The new code is significantly more flexible, which allows us to improve
the code clarity, performance and add more features easily.
This commit is incomplete. It drops almost all previous code, and
readds only the most important things (some of them actually buggy).
The next commit will complete it - it's separate to preserve authorship
information.
2015-03-12 20:57:54 +00:00
|
|
|
}
|
|
|
|
|
vo_opengl: refactor shader generation (part 2)
This adds stuff related to gamma, linear light, sigmoid, BT.2020-CL,
etc, as well as color management. Also adds a new gamma function (gamma22).
This adds new parameters to configure the CMS settings, in particular
letting us target simple colorspaces without requiring usage of a 3DLUT.
This adds smoothmotion. Mostly working, but it's still sensitive to
timing issues. It's based on an actual queue now, but the queue size
is kept small to avoid larger amounts of latency.
Also makes “upscale before blending” the default strategy.
This is justified because the "render after blending" thing doesn't seme
to work consistently any way (introduces stutter due to the way vsync
timing works, or something), so this behavior is a bit closer to master
and makes pausing/unpausing less weird/jumpy.
This adds the remaining scalers, including bicubic_fast, sharpen3,
sharpen5, polar filters and antiringing. Apparently, sharpen3/5 also
consult scale-param1, which was undocumented in master.
This also implements cropping and chroma transformation, plus
rotation/flipping. These are inherently part of the same logic, although
it's a bit rough around the edges in some case, mainly due to the fallback
code paths (for bilinear scaling without indirection).
2015-03-12 21:18:16 +00:00
|
|
|
struct mp_rect_f {
|
|
|
|
float x0, y0, x1, y1;
|
|
|
|
};
|
|
|
|
|
2016-04-16 16:14:32 +00:00
|
|
|
// Semantic equality (fuzzy comparison)
|
|
|
|
static inline bool mp_rect_f_seq(struct mp_rect_f a, struct mp_rect_f b)
|
|
|
|
{
|
|
|
|
return fabs(a.x0 - b.x0) < 1e-6 && fabs(a.x1 - b.x1) < 1e-6 &&
|
|
|
|
fabs(a.y0 - b.y0) < 1e-6 && fabs(a.y1 - b.y1) < 1e-6;
|
|
|
|
}
|
|
|
|
|
2015-03-13 20:14:18 +00:00
|
|
|
static inline void gl_transform_rect(struct gl_transform t, struct mp_rect_f *r)
|
vo_opengl: refactor shader generation (part 2)
This adds stuff related to gamma, linear light, sigmoid, BT.2020-CL,
etc, as well as color management. Also adds a new gamma function (gamma22).
This adds new parameters to configure the CMS settings, in particular
letting us target simple colorspaces without requiring usage of a 3DLUT.
This adds smoothmotion. Mostly working, but it's still sensitive to
timing issues. It's based on an actual queue now, but the queue size
is kept small to avoid larger amounts of latency.
Also makes “upscale before blending” the default strategy.
This is justified because the "render after blending" thing doesn't seme
to work consistently any way (introduces stutter due to the way vsync
timing works, or something), so this behavior is a bit closer to master
and makes pausing/unpausing less weird/jumpy.
This adds the remaining scalers, including bicubic_fast, sharpen3,
sharpen5, polar filters and antiringing. Apparently, sharpen3/5 also
consult scale-param1, which was undocumented in master.
This also implements cropping and chroma transformation, plus
rotation/flipping. These are inherently part of the same logic, although
it's a bit rough around the edges in some case, mainly due to the fallback
code paths (for bilinear scaling without indirection).
2015-03-12 21:18:16 +00:00
|
|
|
{
|
2015-03-13 20:14:18 +00:00
|
|
|
gl_transform_vec(t, &r->x0, &r->y0);
|
|
|
|
gl_transform_vec(t, &r->x1, &r->y1);
|
vo_opengl: refactor shader generation (part 2)
This adds stuff related to gamma, linear light, sigmoid, BT.2020-CL,
etc, as well as color management. Also adds a new gamma function (gamma22).
This adds new parameters to configure the CMS settings, in particular
letting us target simple colorspaces without requiring usage of a 3DLUT.
This adds smoothmotion. Mostly working, but it's still sensitive to
timing issues. It's based on an actual queue now, but the queue size
is kept small to avoid larger amounts of latency.
Also makes “upscale before blending” the default strategy.
This is justified because the "render after blending" thing doesn't seme
to work consistently any way (introduces stutter due to the way vsync
timing works, or something), so this behavior is a bit closer to master
and makes pausing/unpausing less weird/jumpy.
This adds the remaining scalers, including bicubic_fast, sharpen3,
sharpen5, polar filters and antiringing. Apparently, sharpen3/5 also
consult scale-param1, which was undocumented in master.
This also implements cropping and chroma transformation, plus
rotation/flipping. These are inherently part of the same logic, although
it's a bit rough around the edges in some case, mainly due to the fallback
code paths (for bilinear scaling without indirection).
2015-03-12 21:18:16 +00:00
|
|
|
}
|
|
|
|
|
vo_opengl: refactor pass_read_video and texture binding
This is a pretty major rewrite of the internal texture binding
mechanic, which makes it more flexible.
In general, the difference between the old and current approaches is
that now, all texture description is held in a struct img_tex and only
explicitly bound with pass_bind. (Once bound, a texture unit is assumed
to be set in stone and no longer tied to the img_tex)
This approach makes the code inside pass_read_video significantly more
flexible and cuts down on the number of weird special cases and
spaghetti logic.
It also has some improvements, e.g. cutting down greatly on the number
of unnecessary conversion passes inside pass_read_video (which was
previously mostly done to cope with the fact that the alternative would
have resulted in a combinatorial explosion of code complexity).
Some other notable changes (and potential improvements):
- texture expansion is now *always* handled in pass_read_video, and the
colormatrix never does this anymore. (Which means the code could
probably be removed from the colormatrix generation logic, modulo some
other VOs)
- struct fbo_tex now stores both its "physical" and "logical"
(configured) size, which cuts down on the amount of width/height
baggage on some function calls
- vo_opengl can now technically support textures with different bit
depths (e.g. 10 bit luma, 8 bit chroma) - but the APIs it queries
inside img_format.c doesn't export this (nor does ffmpeg support it,
really) so the status quo of using the same tex_mul for all planes is
kept.
- dumb_mode is now only needed because of the indirect_fbo being in the
main rendering pipeline. If we reintroduce p->use_indirect and thread
a transform through the entire program this could be skipped where
unnecessary, allowing for the removal of dumb_mode. But I'm not sure
how to do this in a clean way. (Which is part of why it got introduced
to begin with)
- It would be trivial to resurrect source-shader now (it would just be
one extra 'if' inside pass_read_video).
2016-03-05 10:29:19 +00:00
|
|
|
static inline bool gl_transform_eq(struct gl_transform a, struct gl_transform b)
|
|
|
|
{
|
|
|
|
for (int x = 0; x < 2; x++) {
|
|
|
|
for (int y = 0; y < 2; y++) {
|
|
|
|
if (a.m[x][y] != b.m[x][y])
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return a.t[0] == b.t[0] && a.t[1] == b.t[1];
|
|
|
|
}
|
|
|
|
|
2015-10-26 22:43:48 +00:00
|
|
|
void gl_transform_trans(struct gl_transform t, struct gl_transform *x);
|
|
|
|
|
2017-08-07 14:44:15 +00:00
|
|
|
struct fbodst {
|
|
|
|
struct ra_tex *tex;
|
|
|
|
bool flip; // mirror vertically
|
|
|
|
};
|
|
|
|
|
|
|
|
void gl_transform_ortho_fbodst(struct gl_transform *t, struct fbodst fbo);
|
|
|
|
|
2017-08-19 02:33:40 +00:00
|
|
|
// A pool of buffers, which can grow as needed
|
|
|
|
struct ra_buf_pool {
|
|
|
|
struct ra_buf_params current_params;
|
|
|
|
struct ra_buf **buffers;
|
|
|
|
int num_buffers;
|
|
|
|
int index;
|
|
|
|
};
|
|
|
|
|
|
|
|
void ra_buf_pool_uninit(struct ra *ra, struct ra_buf_pool *pool);
|
|
|
|
|
|
|
|
// Note: params->initial_data is *not* supported
|
|
|
|
struct ra_buf *ra_buf_pool_get(struct ra *ra, struct ra_buf_pool *pool,
|
|
|
|
const struct ra_buf_params *params);
|
|
|
|
|
|
|
|
// Helper that wraps ra_tex_upload using texture upload buffers to ensure that
|
|
|
|
// params->buf is always set. This is intended for RA-internal usage.
|
|
|
|
bool ra_tex_upload_pbo(struct ra *ra, struct ra_buf_pool *pbo,
|
|
|
|
const struct ra_tex_upload_params *params);
|
|
|
|
|
2017-08-04 17:09:46 +00:00
|
|
|
struct fbotex {
|
|
|
|
struct ra *ra;
|
|
|
|
struct ra_tex *tex;
|
|
|
|
int lw, lh; // logical (configured) size, <= than texture size
|
2017-08-07 14:44:15 +00:00
|
|
|
struct fbodst fbo;
|
2016-07-03 14:00:51 +00:00
|
|
|
};
|
|
|
|
|
2017-08-04 17:09:46 +00:00
|
|
|
void fbotex_uninit(struct fbotex *fbo);
|
|
|
|
bool fbotex_change(struct fbotex *fbo, struct ra *ra, struct mp_log *log,
|
|
|
|
int w, int h, const struct ra_format *fmt, int flags);
|
|
|
|
#define FBOTEX_FUZZY_W 1
|
|
|
|
#define FBOTEX_FUZZY_H 2
|
|
|
|
#define FBOTEX_FUZZY (FBOTEX_FUZZY_W | FBOTEX_FUZZY_H)
|
2017-08-05 16:20:45 +00:00
|
|
|
|
|
|
|
// A wrapper around ra_timer that does result pooling, averaging etc.
|
|
|
|
struct timer_pool;
|
|
|
|
|
|
|
|
struct timer_pool *timer_pool_create(struct ra *ra);
|
|
|
|
void timer_pool_destroy(struct timer_pool *pool);
|
|
|
|
void timer_pool_start(struct timer_pool *pool);
|
|
|
|
void timer_pool_stop(struct timer_pool *pool);
|
|
|
|
struct mp_pass_perf timer_pool_measure(struct timer_pool *pool);
|
2017-08-05 17:38:43 +00:00
|
|
|
|
|
|
|
// print a multi line string with line numbers (e.g. for shader sources)
|
|
|
|
// log, lev: module and log level, as in mp_msg()
|
|
|
|
void mp_log_source(struct mp_log *log, int lev, const char *src);
|