summaryrefslogtreecommitdiffstats
path: root/video/decode/vd_lavc.c
diff options
context:
space:
mode:
authorwm4 <wm4@nowhere>2012-11-05 14:25:04 +0100
committerwm4 <wm4@nowhere>2013-01-13 20:04:10 +0100
commitc54fc507da8edcc2c5d3bc3f50b0881d1c1406d7 (patch)
tree530d112301256e1c3ea50d7bb416b7ba2109130b /video/decode/vd_lavc.c
parent1c412169aca2f0ad38380b0c89f2485e6a256766 (diff)
downloadmpv-c54fc507da8edcc2c5d3bc3f50b0881d1c1406d7.tar.bz2
mpv-c54fc507da8edcc2c5d3bc3f50b0881d1c1406d7.tar.xz
video/filter: change filter API, use refcounting, remove filter DR
Change the entire filter API to use reference counted images instead of vf_get_image(). Remove filter "direct rendering". This was useful for vf_expand and (in rare cases) vf_sub: DR allowed these filters to pass a cropped image to the filters before them. Then, on filtering, the image was "uncropped", so that black bars could be added around the image without copying. This means that in some cases, vf_expand will be slower (-vf gradfun,expand for example). Note that another form of DR used for in-place filters has been replaced by simpler logic. Instead of trying to do DR, filters can check if the image is writeable (with mp_image_is_writeable()), and do true in-place if that's the case. This affects filters like vf_gradfun and vf_sub. Everything has to support strides now. If something doesn't, making a copy of the image data is required.
Diffstat (limited to 'video/decode/vd_lavc.c')
-rw-r--r--video/decode/vd_lavc.c59
1 files changed, 19 insertions, 40 deletions
diff --git a/video/decode/vd_lavc.c b/video/decode/vd_lavc.c
index 3ee8cc7932..7fd7437a3b 100644
--- a/video/decode/vd_lavc.c
+++ b/video/decode/vd_lavc.c
@@ -41,6 +41,7 @@
#include "vd.h"
#include "video/img_format.h"
+#include "video/mp_image_pool.h"
#include "video/filter/vf.h"
#include "demux/stheader.h"
#include "demux/demux_packet.h"
@@ -244,6 +245,7 @@ static int init(sh_video_t *sh)
ctx = sh->context = talloc_zero(NULL, vd_ffmpeg_ctx);
ctx->rawvideo_fmt = PIX_FMT_NONE;
+ ctx->non_dr1_pool = talloc_steal(ctx, mp_image_pool_new(16));
if (sh->codec->dll) {
lavc_codec = avcodec_find_decoder_by_name(sh->codec->dll);
@@ -478,7 +480,6 @@ static void uninit_avctx(sh_video_t *sh)
av_freep(&avctx);
avcodec_free_frame(&ctx->pic);
- mp_image_unrefp(&ctx->last_mpi);
mp_buffer_pool_free(&ctx->dr1_buffer_pool);
}
@@ -518,7 +519,7 @@ static int init_vo(sh_video_t *sh)
width != sh->disp_w || height != sh->disp_h ||
avctx->pix_fmt != ctx->pix_fmt || !ctx->vo_initialized)
{
- mp_image_unrefp(&ctx->last_mpi);
+ mp_image_pool_clear(ctx->non_dr1_pool);
ctx->vo_initialized = 0;
mp_msg(MSGT_DECVIDEO, MSGL_V, "[ffmpeg] aspect_ratio: %f\n", aspect);
@@ -579,22 +580,6 @@ static void draw_slice_hwdec(struct AVCodecContext *s,
vf->control(vf, VFCTRL_HWDEC_DECODER_RENDER, state_ptr);
}
-static struct mp_image *get_image_hwdec(vd_ffmpeg_ctx *ctx)
-{
- for (int n = 0; n < MAX_NUM_MPI; n++) {
- struct mp_image *cur = &ctx->hwdec_mpi[n];
- if (cur->usage_count == 0) {
- *cur = (struct mp_image) {
- .number = n,
- .imgfmt = ctx->best_csp,
- .usage_count = 1,
- };
- return cur;
- }
- }
- return NULL;
-}
-
static int get_buffer_hwdec(AVCodecContext *avctx, AVFrame *pic)
{
sh_video_t *sh = avctx->opaque;
@@ -620,12 +605,12 @@ static int get_buffer_hwdec(AVCodecContext *avctx, AVFrame *pic)
assert(IMGFMT_IS_HWACCEL(ctx->best_csp));
- struct mp_image *mpi = get_image_hwdec(ctx);
- if (!mpi)
- return -1;
+ struct mp_image *mpi = NULL;
struct vf_instance *vf = sh->vfilter;
- vf->control(vf, VFCTRL_HWDEC_GET_SURFACE, mpi);
+ vf->control(vf, VFCTRL_HWDEC_ALLOC_SURFACE, &mpi);
+ if (!mpi)
+ return -1;
for (int i = 0; i < 4; i++)
pic->data[i] = mpi->planes[i];
@@ -647,9 +632,8 @@ static void release_buffer_hwdec(AVCodecContext *avctx, AVFrame *pic)
assert(pic->type == FF_BUFFER_TYPE_USER);
assert(mpi);
- assert(mpi->usage_count > 0);
- mpi->usage_count--;
+ talloc_free(mpi);
for (int i = 0; i < 4; i++)
pic->data[i] = NULL;
@@ -726,13 +710,14 @@ static int decode(struct sh_video *sh, struct demux_packet *packet, void *data,
return -1;
struct mp_image *mpi = NULL;
- if (ctx->do_hw_dr1 && pic->opaque)
+ if (ctx->do_hw_dr1 && pic->opaque) {
mpi = pic->opaque; // reordered frame
+ assert(mpi);
+ mpi = mp_image_new_ref(mpi);
+ }
if (!mpi) {
struct mp_image new = {0};
- new.type = MP_IMGTYPE_EXPORT;
- new.flags = MP_IMGFLAG_PRESERVE;
mp_image_set_size(&new, avctx->width, avctx->height);
mp_image_setfmt(&new, ctx->best_csp);
for (int i = 0; i < 4; i++) {
@@ -741,23 +726,17 @@ static int decode(struct sh_video *sh, struct demux_packet *packet, void *data,
}
if (ctx->do_dr1 && pic->opaque) {
struct FrameBuffer *fb = pic->opaque;
- mp_image_unrefp(&ctx->last_mpi);
- mp_buffer_ref(fb); // reference for last_mpi
- ctx->last_mpi = mp_image_new_external_ref(&new, fb, fb_ref,
- fb_unref, fb_is_unique);
+ mp_buffer_ref(fb); // initial reference for mpi
+ mpi = mp_image_new_external_ref(&new, fb, fb_ref, fb_unref,
+ fb_is_unique);
} else {
- if (!ctx->last_mpi)
- ctx->last_mpi = mp_image_alloc(ctx->best_csp, new.w, new.h);
- mp_image_make_writeable(&ctx->last_mpi);
- assert(ctx->last_mpi->w == new.w && ctx->last_mpi->h == new.h);
- assert(ctx->last_mpi->imgfmt == new.imgfmt);
- mp_image_copy(ctx->last_mpi, &new);
+ mpi = mp_image_pool_get(ctx->non_dr1_pool, new.imgfmt,
+ new.w, new.h);
+ mp_image_copy(mpi, &new);
}
- mpi = ctx->last_mpi;
}
- if (!mpi->planes[0])
- return 0; // ?
+ assert(mpi->planes[0]);
assert(mpi->imgfmt == pixfmt2imgfmt(avctx->pix_fmt));