avcodec/videotoolbox: add hevc support

Signed-off-by: Aman Gupta <aman@tmm1.net>

6 files changed, 219 insertions, 2 deletions

diff --git a/configure b/configure
index c2f5fc2c98..9cf98c8cc2 100755
--- a/configure
+++ b/configure
@@ -2081,6 +2081,7 @@ TOOLCHAIN_FEATURES="
 
 TYPES_LIST="
     CONDITION_VARIABLE_Ptr
+    kCMVideoCodecType_HEVC
     socklen_t
     struct_addrinfo
     struct_group_source_req
@@ -2696,6 +2697,8 @@ hevc_vaapi_hwaccel_deps="vaapi VAPictureParameterBufferHEVC"
 hevc_vaapi_hwaccel_select="hevc_decoder"
 hevc_vdpau_hwaccel_deps="vdpau VdpPictureInfoHEVC"
 hevc_vdpau_hwaccel_select="hevc_decoder"
+hevc_videotoolbox_hwaccel_deps="videotoolbox"
+hevc_videotoolbox_hwaccel_select="hevc_decoder"
 mjpeg_cuvid_hwaccel_deps="cuda cuvid"
 mjpeg_cuvid_hwaccel_select="mjpeg_cuvid_decoder"
 mpeg_xvmc_hwaccel_deps="xvmc"
@@ -5814,8 +5817,10 @@ enabled avfoundation && {
     check_lib avfoundation CoreGraphics/CoreGraphics.h               CGGetActiveDisplayList "-framework CoreGraphics" ||
     check_lib avfoundation ApplicationServices/ApplicationServices.h CGGetActiveDisplayList "-framework ApplicationServices"; }
 
-enabled videotoolbox &&
+enabled videotoolbox && {
     check_lib coreservices CoreServices/CoreServices.h UTGetOSTypeFromString "-framework CoreServices"
+    check_func_headers CoreMedia/CMFormatDescription.h kCMVideoCodecType_HEVC "-framework CoreMedia"
+}
 
 check_struct "sys/time.h sys/resource.h" "struct rusage" ru_maxrss
 
diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c
index 7ae9e8cea1..4f34312e67 100644
--- a/libavcodec/allcodecs.c
+++ b/libavcodec/allcodecs.c
@@ -84,6 +84,7 @@ static void register_all(void)
     REGISTER_HWACCEL(HEVC_QSV,          hevc_qsv);
     REGISTER_HWACCEL(HEVC_VAAPI,        hevc_vaapi);
     REGISTER_HWACCEL(HEVC_VDPAU,        hevc_vdpau);
+    REGISTER_HWACCEL(HEVC_VIDEOTOOLBOX, hevc_videotoolbox);
     REGISTER_HWACCEL(MJPEG_CUVID,       mjpeg_cuvid);
     REGISTER_HWACCEL(MPEG1_CUVID,       mpeg1_cuvid);
     REGISTER_HWACCEL(MPEG1_XVMC,        mpeg1_xvmc);
diff --git a/libavcodec/hevc_refs.c b/libavcodec/hevc_refs.c
index 68c730edcc..ac462d350b 100644
--- a/libavcodec/hevc_refs.c
+++ b/libavcodec/hevc_refs.c
@@ -208,6 +208,9 @@ int ff_hevc_output_frame(HEVCContext *s, AVFrame *out, int flush)
         if (nb_output) {
             HEVCFrame *frame = &s->DPB[min_idx];
 
+            if (frame->frame->format == AV_PIX_FMT_VIDEOTOOLBOX && frame->frame->buf[0]->size == 1)
+                return 0;
+
             ret = av_frame_ref(out, frame->frame);
             if (frame->flags & HEVC_FRAME_FLAG_BUMPING)
                 ff_hevc_unref_frame(s, frame, HEVC_FRAME_FLAG_OUTPUT | HEVC_FRAME_FLAG_BUMPING);
diff --git a/libavcodec/hevcdec.c b/libavcodec/hevcdec.c
index 2306c51ed3..2e4add2ae3 100644
--- a/libavcodec/hevcdec.c
+++ b/libavcodec/hevcdec.c
@@ -352,7 +352,11 @@ static void export_stream_params(AVCodecContext *avctx, const HEVCParamSets *ps,
 
 static enum AVPixelFormat get_format(HEVCContext *s, const HEVCSPS *sps)
 {
-    #define HWACCEL_MAX (CONFIG_HEVC_DXVA2_HWACCEL + CONFIG_HEVC_D3D11VA_HWACCEL * 2 + CONFIG_HEVC_VAAPI_HWACCEL + CONFIG_HEVC_VDPAU_HWACCEL)
+#define HWACCEL_MAX (CONFIG_HEVC_DXVA2_HWACCEL + \
+                     CONFIG_HEVC_D3D11VA_HWACCEL * 2 + \
+                     CONFIG_HEVC_VAAPI_HWACCEL + \
+                     CONFIG_HEVC_VIDEOTOOLBOX_HWACCEL + \
+                     CONFIG_HEVC_VDPAU_HWACCEL)
     enum AVPixelFormat pix_fmts[HWACCEL_MAX + 2], *fmt = pix_fmts;
 
     switch (sps->pix_fmt) {
@@ -371,6 +375,9 @@ static enum AVPixelFormat get_format(HEVCContext *s, const HEVCSPS *sps)
 #if CONFIG_HEVC_VDPAU_HWACCEL
         *fmt++ = AV_PIX_FMT_VDPAU;
 #endif
+#if CONFIG_HEVC_VIDEOTOOLBOX_HWACCEL
+        *fmt++ = AV_PIX_FMT_VIDEOTOOLBOX;
+#endif
         break;
     case AV_PIX_FMT_YUV420P10:
 #if CONFIG_HEVC_DXVA2_HWACCEL
@@ -383,6 +390,9 @@ static enum AVPixelFormat get_format(HEVCContext *s, const HEVCSPS *sps)
 #if CONFIG_HEVC_VAAPI_HWACCEL
         *fmt++ = AV_PIX_FMT_VAAPI;
 #endif
+#if CONFIG_HEVC_VIDEOTOOLBOX_HWACCEL
+        *fmt++ = AV_PIX_FMT_VIDEOTOOLBOX;
+#endif
         break;
     }
 
diff --git a/libavcodec/vda_vt_internal.h b/libavcodec/vda_vt_internal.h
index e55a813899..326a60a695 100644
--- a/libavcodec/vda_vt_internal.h
+++ b/libavcodec/vda_vt_internal.h
@@ -59,4 +59,5 @@ int ff_videotoolbox_h264_decode_slice(AVCodecContext *avctx,
                                       const uint8_t *buffer,
                                       uint32_t size);
 CFDataRef ff_videotoolbox_avcc_extradata_create(AVCodecContext *avctx);
+CFDataRef ff_videotoolbox_hvcc_extradata_create(AVCodecContext *avctx);
 #endif /* AVCODEC_VDA_VT_INTERNAL_H */
diff --git a/libavcodec/videotoolbox.c b/libavcodec/videotoolbox.c
index dd13e2581b..ec8b6d885c 100644
--- a/libavcodec/videotoolbox.c
+++ b/libavcodec/videotoolbox.c
@@ -32,6 +32,7 @@
 #include "libavutil/hwcontext.h"
 #include "bytestream.h"
 #include "h264dec.h"
+#include "hevcdec.h"
 #include "mpegvideo.h"
 #include <TargetConditionals.h>
 
@@ -39,6 +40,10 @@
 #  define kVTVideoDecoderSpecification_RequireHardwareAcceleratedVideoDecoder CFSTR("RequireHardwareAcceleratedVideoDecoder")
 #endif
 
+#if !HAVE_KCMVIDEOCODECTYPE_HEVC
+enum { kCMVideoCodecType_HEVC = 'hvc1' };
+#endif
+
 #define VIDEOTOOLBOX_ESDS_EXTRADATA_PADDING  12
 
 static void videotoolbox_buffer_release(void *opaque, uint8_t *data)
@@ -115,6 +120,164 @@ CFDataRef ff_videotoolbox_avcc_extradata_create(AVCodecContext *avctx)
     return data;
 }
 
+CFDataRef ff_videotoolbox_hvcc_extradata_create(AVCodecContext *avctx)
+{
+    HEVCContext *h = avctx->priv_data;
+    const HEVCVPS *vps = (const HEVCVPS *)h->ps.vps_list[0]->data;
+    const HEVCSPS *sps = (const HEVCSPS *)h->ps.sps_list[0]->data;
+    int i, num_pps = 0;
+    const HEVCPPS *pps = h->ps.pps;
+    PTLCommon ptlc = vps->ptl.general_ptl;
+    VUI vui = sps->vui;
+    uint8_t parallelismType;
+    CFDataRef data = NULL;
+    uint8_t *p;
+    int vt_extradata_size = 23 + 5 + vps->data_size + 5 + sps->data_size + 3;
+    uint8_t *vt_extradata;
+
+    for (i = 0; i < MAX_PPS_COUNT; i++) {
+        if (h->ps.pps_list[i]) {
+            const HEVCPPS *pps = (const HEVCPPS *)h->ps.pps_list[i]->data;
+            vt_extradata_size += 2 + pps->data_size;
+            num_pps++;
+        }
+    }
+
+    vt_extradata = av_malloc(vt_extradata_size);
+    if (!vt_extradata)
+        return NULL;
+    p = vt_extradata;
+
+    /* unsigned int(8) configurationVersion = 1; */
+    AV_W8(p + 0, 1);
+
+    /*
+     * unsigned int(2) general_profile_space;
+     * unsigned int(1) general_tier_flag;
+     * unsigned int(5) general_profile_idc;
+     */
+    AV_W8(p + 1, ptlc.profile_space << 6 |
+                 ptlc.tier_flag     << 5 |
+                 ptlc.profile_idc);
+
+    /* unsigned int(32) general_profile_compatibility_flags; */
+    memcpy(p + 2, ptlc.profile_compatibility_flag, 4);
+
+    /* unsigned int(48) general_constraint_indicator_flags; */
+    AV_W8(p + 6, ptlc.progressive_source_flag    << 7 |
+                 ptlc.interlaced_source_flag     << 6 |
+                 ptlc.non_packed_constraint_flag << 5 |
+                 ptlc.frame_only_constraint_flag << 4);
+    AV_W8(p + 7, 0);
+    AV_WN32(p + 8, 0);
+
+    /* unsigned int(8) general_level_idc; */
+    AV_W8(p + 12, ptlc.level_idc);
+
+    /*
+     * bit(4) reserved = ‘1111’b;
+     * unsigned int(12) min_spatial_segmentation_idc;
+     */
+    AV_W8(p + 13, 0xf0 | (vui.min_spatial_segmentation_idc >> 4));
+    AV_W8(p + 14, vui.min_spatial_segmentation_idc & 0xff);
+
+    /*
+     * bit(6) reserved = ‘111111’b;
+     * unsigned int(2) parallelismType;
+     */
+    if (!vui.min_spatial_segmentation_idc)
+        parallelismType = 0;
+    else if (pps->entropy_coding_sync_enabled_flag && pps->tiles_enabled_flag)
+        parallelismType = 0;
+    else if (pps->entropy_coding_sync_enabled_flag)
+        parallelismType = 3;
+    else if (pps->tiles_enabled_flag)
+        parallelismType = 2;
+    else
+        parallelismType = 1;
+    AV_W8(p + 15, 0xfc | parallelismType);
+
+    /*
+     * bit(6) reserved = ‘111111’b;
+     * unsigned int(2) chromaFormat;
+     */
+    AV_W8(p + 16, sps->chroma_format_idc | 0xfc);
+
+    /*
+     * bit(5) reserved = ‘11111’b;
+     * unsigned int(3) bitDepthLumaMinus8;
+     */
+    AV_W8(p + 17, (sps->bit_depth - 8) | 0xfc);
+
+    /*
+     * bit(5) reserved = ‘11111’b;
+     * unsigned int(3) bitDepthChromaMinus8;
+     */
+    AV_W8(p + 18, (sps->bit_depth_chroma - 8) | 0xfc);
+
+    /* bit(16) avgFrameRate; */
+    AV_WB16(p + 19, 0);
+
+    /*
+     * bit(2) constantFrameRate;
+     * bit(3) numTemporalLayers;
+     * bit(1) temporalIdNested;
+     * unsigned int(2) lengthSizeMinusOne;
+     */
+    AV_W8(p + 21, 0                             << 6 |
+                  sps->max_sub_layers           << 3 |
+                  sps->temporal_id_nesting_flag << 2 |
+                  3);
+
+    /* unsigned int(8) numOfArrays; */
+    AV_W8(p + 22, 3);
+
+    p += 23;
+    /* vps */
+    /*
+     * bit(1) array_completeness;
+     * unsigned int(1) reserved = 0;
+     * unsigned int(6) NAL_unit_type;
+     */
+    AV_W8(p, 1 << 7 |
+             HEVC_NAL_VPS & 0x3f);
+    /* unsigned int(16) numNalus; */
+    AV_WB16(p + 1, 1);
+    /* unsigned int(16) nalUnitLength; */
+    AV_WB16(p + 3, vps->data_size);
+    /* bit(8*nalUnitLength) nalUnit; */
+    memcpy(p + 5, vps->data, vps->data_size);
+    p += 5 + vps->data_size;
+
+    /* sps */
+    AV_W8(p, 1 << 7 |
+             HEVC_NAL_SPS & 0x3f);
+    AV_WB16(p + 1, 1);
+    AV_WB16(p + 3, sps->data_size);
+    memcpy(p + 5, sps->data, sps->data_size);
+    p += 5 + sps->data_size;
+
+    /* pps */
+    AV_W8(p, 1 << 7 |
+             HEVC_NAL_PPS & 0x3f);
+    AV_WB16(p + 1, num_pps);
+    p += 3;
+    for (i = 0; i < MAX_PPS_COUNT; i++) {
+        if (h->ps.pps_list[i]) {
+            const HEVCPPS *pps = (const HEVCPPS *)h->ps.pps_list[i]->data;
+            AV_WB16(p, pps->data_size);
+            memcpy(p + 2, pps->data, pps->data_size);
+            p += 2 + pps->data_size;
+        }
+    }
+
+    av_assert0(p - vt_extradata == vt_extradata_size);
+
+    data = CFDataCreate(kCFAllocatorDefault, vt_extradata, vt_extradata_size);
+    av_free(vt_extradata);
+    return data;
+}
+
 int ff_videotoolbox_buffer_create(VTContext *vtctx, AVFrame *frame)
 {
     av_buffer_unref(&frame->buf[0]);
@@ -445,6 +608,18 @@ static int videotoolbox_h264_end_frame(AVCodecContext *avctx)
     return videotoolbox_common_end_frame(avctx, frame);
 }
 
+static int videotoolbox_hevc_end_frame(AVCodecContext *avctx)
+{
+    HEVCContext *h = avctx->priv_data;
+    AVFrame *frame = h->ref->frame;
+    VTContext *vtctx = avctx->internal->hwaccel_priv_data;
+    int ret;
+
+    ret = videotoolbox_common_end_frame(avctx, frame);
+    vtctx->bitstream_size = 0;
+    return ret;
+}
+
 static int videotoolbox_mpeg_start_frame(AVCodecContext *avctx,
                                          const uint8_t *buffer,
                                          uint32_t size)
@@ -501,6 +676,11 @@ static CFDictionaryRef videotoolbox_decoder_config_create(CMVideoCodecType codec
             if (data)
                 CFDictionarySetValue(avc_info, CFSTR("avcC"), data);
             break;
+        case kCMVideoCodecType_HEVC :
+            data = ff_videotoolbox_hvcc_extradata_create(avctx);
+            if (data)
+                CFDictionarySetValue(avc_info, CFSTR("hvcC"), data);
+            break;
         default:
             break;
         }
@@ -600,6 +780,9 @@ static int videotoolbox_default_init(AVCodecContext *avctx)
     case AV_CODEC_ID_H264 :
         videotoolbox->cm_codec_type = kCMVideoCodecType_H264;
         break;
+    case AV_CODEC_ID_HEVC :
+        videotoolbox->cm_codec_type = kCMVideoCodecType_HEVC;
+        break;
     case AV_CODEC_ID_MPEG1VIDEO :
         videotoolbox->cm_codec_type = kCMVideoCodecType_MPEG1Video;
         break;
@@ -782,6 +965,20 @@ AVHWAccel ff_h263_videotoolbox_hwaccel = {
     .priv_data_size = sizeof(VTContext),
 };
 
+AVHWAccel ff_hevc_videotoolbox_hwaccel = {
+    .name           = "hevc_videotoolbox",
+    .type           = AVMEDIA_TYPE_VIDEO,
+    .id             = AV_CODEC_ID_HEVC,
+    .pix_fmt        = AV_PIX_FMT_VIDEOTOOLBOX,
+    .alloc_frame    = ff_videotoolbox_alloc_frame,
+    .start_frame    = ff_videotoolbox_h264_start_frame,
+    .decode_slice   = ff_videotoolbox_h264_decode_slice,
+    .end_frame      = videotoolbox_hevc_end_frame,
+    .init           = videotoolbox_common_init,
+    .uninit         = ff_videotoolbox_uninit,
+    .priv_data_size = sizeof(VTContext),
+};
+
 AVHWAccel ff_h264_videotoolbox_hwaccel = {
     .name           = "h264_videotoolbox",
     .type           = AVMEDIA_TYPE_VIDEO,