1
0
mirror of https://github.com/mpv-player/mpv synced 2024-12-26 00:42:57 +00:00
mpv/video/vaapi.h

127 lines
4.4 KiB
C
Raw Normal View History

2013-08-12 21:59:47 +00:00
#ifndef MPV_VAAPI_H
#define MPV_VAAPI_H
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
#include <stdbool.h>
#include <inttypes.h>
#include <va/va.h>
#include <va/va_x11.h>
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
/* Compatibility glue with VA-API >= 0.31 */
#if defined VA_CHECK_VERSION
#if VA_CHECK_VERSION(0,31,0)
#define vaPutImage2 vaPutImage
#define vaAssociateSubpicture2 vaAssociateSubpicture
#endif
#endif
/* Compatibility glue with VA-API >= 0.34 */
#if VA_CHECK_VERSION(0,34,0)
#include <va/va_compat.h>
#endif
/* Compatibility glue with upstream libva */
#ifndef VA_SDS_VERSION
#define VA_SDS_VERSION 0
#endif
/* Compatibility glue with VA-API >= 0.30 */
#ifndef VA_INVALID_ID
#define VA_INVALID_ID 0xffffffff
#endif
#ifndef VA_FOURCC
#define VA_FOURCC(ch0, ch1, ch2, ch3) \
((uint32_t)(uint8_t)(ch0) | \
((uint32_t)(uint8_t)(ch1) << 8) | \
((uint32_t)(uint8_t)(ch2) << 16) | \
((uint32_t)(uint8_t)(ch3) << 24 ))
#endif
#if defined VA_SRC_BT601 && defined VA_SRC_BT709
# define USE_VAAPI_COLORSPACE 1
#else
# define USE_VAAPI_COLORSPACE 0
#endif
/* Compatibility glue with VA-API >= 0.31.1 */
#ifndef VA_SRC_SMPTE_240
#define VA_SRC_SMPTE_240 0x00000040
#endif
#if defined VA_FILTER_SCALING_MASK
# define USE_VAAPI_SCALING 1
#else
# define USE_VAAPI_SCALING 0
#endif
#ifndef VA_FOURCC_YV12
#define VA_FOURCC_YV12 0x32315659
#endif
#ifndef VA_FOURCC_IYUV
#define VA_FOURCC_IYUV 0x56555949
#endif
#ifndef VA_FOURCC_I420
#define VA_FOURCC_I420 VA_FOURCC('I', '4', '2', '0')
#endif
#ifndef VA_FOURCC_RGBX
#define VA_FOURCC_RGBX 0x58424752
#endif
#ifndef VA_FOURCC_BGRX
#define VA_FOURCC_BGRX 0x58524742
#endif
#define VA_STR_FOURCC(fcc) \
(const char[]){(fcc), (fcc) >> 8u, (fcc) >> 16u, (fcc) >> 24u, 0}
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
#include "mp_image.h"
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
struct mp_image_pool;
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
struct mp_vaapi_ctx {
VADisplay display;
struct va_image_formats *image_formats;
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
};
struct va_surface_pool;
struct va_image_formats;
struct va_surface {
VASurfaceID id; // VA_INVALID_ID if unallocated
int w, h, rt_format; // parameters of allocated image (0/0/-1 unallocated)
struct va_surface_priv *p;
};
bool check_va_status(VAStatus status, const char *msg);
int va_get_colorspace_flag(enum mp_csp csp);
struct mp_vaapi_ctx *va_initialize(VADisplay *display);
void va_destroy(struct mp_vaapi_ctx *ctx);
enum mp_imgfmt va_fourcc_to_imgfmt(uint32_t fourcc);
uint32_t va_fourcc_from_imgfmt(int imgfmt);
VAImageFormat * va_image_format_from_imgfmt(const struct va_image_formats *formats, int imgfmt);
bool va_image_map(VADisplay display, VAImage *image, struct mp_image *mpi);
bool va_image_unmap(VADisplay display, VAImage *image);
struct va_surface_pool * va_surface_pool_alloc(VADisplay display, int rt_format);
void va_surface_pool_release(struct va_surface_pool *pool);
void va_surface_pool_releasep(struct va_surface_pool **pool);
void va_surface_pool_clear(struct va_surface_pool *pool);
bool va_surface_pool_reserve(struct va_surface_pool *pool, int count, int w, int h);
int va_surface_pool_rt_format(const struct va_surface_pool *pool);
struct va_surface * va_surface_pool_get(struct va_surface_pool *pool, int w, int h);
struct va_surface * va_surface_pool_get_by_imgfmt(struct va_surface_pool *pool, const struct va_image_formats *formats, int imgfmt, int w, int h);
struct mp_image * va_surface_pool_get_wrapped(struct va_surface_pool *pool, const struct va_image_formats *formats, int imgfmt, int w, int h);
void va_surface_release(struct va_surface *surface);
void va_surface_releasep(struct va_surface **surface);
struct va_surface * va_surface_in_mp_image(struct mp_image *mpi);
struct mp_image * va_surface_wrap(struct va_surface *surface); // takes ownership
VASurfaceID va_surface_id(const struct va_surface *surface);
VASurfaceID va_surface_id_in_mp_image(const struct mp_image *mpi);
bool va_surface_upload(struct va_surface *surface, struct mp_image *mpi);
struct mp_image * va_surface_download(struct va_surface *surface,
const struct va_image_formats *formats,
struct mp_image_pool *pool);
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
#endif