diff options
100 files changed, 3045 insertions, 3747 deletions
diff --git a/.gitignore b/.gitignore index dfeae9971..aa95d571c 100644 --- a/.gitignore +++ b/.gitignore @@ -1,15 +1,19 @@ +*.S *.a *.asm.s *.d -*.gcno *.gcda +*.gcno *.o *~ -/*.ivf -/*.ivf.md5 +.cproject +.project +.settings /*-*.mk /*.asm /*.doxy +/*.ivf +/*.ivf.md5 /.bins /.deps /.docs @@ -18,16 +22,24 @@ /Makefile /config.log /config.mk -/decode_to_md5 -/decode_to_md5.dox -/decode_with_drops -/decode_with_drops.dox /docs/ /doxyfile -/error_resilient -/error_resilient.dox -/force_keyframe -/force_keyframe.dox +/examples/*.dox +/examples/decode_to_md5 +/examples/decode_with_drops +/examples/decode_with_partial_drops +/examples/error_resilient +/examples/example_xma +/examples/force_keyframe +/examples/postproc +/examples/simple_decoder +/examples/simple_encoder +/examples/twopass_encoder +/examples/vp8_multi_resolution_encoder +/examples/vp8_set_maps +/examples/vp8cx_set_ref +/examples/vp9_spatial_scalable_encoder +/examples/vpx_temporal_scalable_patterns /ivfdec /ivfdec.dox /ivfenc @@ -35,31 +47,17 @@ /libvpx.so* /libvpx.ver /obj_int_extract -/postproc -/postproc.dox /samples.dox -/simple_decoder -/simple_decoder.dox -/simple_encoder -/simple_encoder.dox /test_libvpx -/twopass_encoder -/twopass_encoder.dox /vp8_api1_migration.dox -/vp8_scalable_patterns -/vp8_scalable_patterns.dox -/vp8_set_maps -/vp8_set_maps.dox -/vp8cx_set_ref -/vp8cx_set_ref.dox +/vp[89x]_rtcd.h /vpx.pc /vpx_config.c /vpx_config.h -/vpx_rtcd.h +/vpx_scale_rtcd.h /vpx_version.h /vpxdec +/vpxdec.dox /vpxenc +/vpxenc.dox TAGS -.cproject -.project -.settings diff --git a/build/arm-msvs/obj_int_extract.bat b/build/arm-msvs/obj_int_extract.bat index 7fd16a303..267ed61d5 100644 --- a/build/arm-msvs/obj_int_extract.bat +++ b/build/arm-msvs/obj_int_extract.bat @@ -7,8 +7,12 @@ REM in the file PATENTS. All contributing project authors may REM be found in the AUTHORS file in the root of the source tree. echo on +REM Arguments: +REM %1 - Relative path to the directory containing the vp8 and vpx_scale +REM source directories. +REM %2 - Path to obj_int_extract.exe. cl /I "./" /I "%1" /nologo /c /DWINAPI_FAMILY=WINAPI_FAMILY_PHONE_APP "%1/vp8/encoder/vp8_asm_enc_offsets.c" -obj_int_extract.exe rvds "vp8_asm_enc_offsets.obj" > "vp8_asm_enc_offsets.asm" +%2\obj_int_extract.exe rvds "vp8_asm_enc_offsets.obj" > "vp8_asm_enc_offsets.asm" cl /I "./" /I "%1" /nologo /c /DWINAPI_FAMILY=WINAPI_FAMILY_PHONE_APP "%1/vpx_scale/vpx_scale_asm_offsets.c" -obj_int_extract.exe rvds "vpx_scale_asm_offsets.obj" > "vpx_scale_asm_offsets.asm" +%2\obj_int_extract.exe rvds "vpx_scale_asm_offsets.obj" > "vpx_scale_asm_offsets.asm" diff --git a/build/make/gen_msvs_proj.sh b/build/make/gen_msvs_proj.sh index 7df0334c6..5936370a7 100755 --- a/build/make/gen_msvs_proj.sh +++ b/build/make/gen_msvs_proj.sh @@ -371,7 +371,7 @@ generate_vcproj() { vpx) tag Tool \ Name="VCPreBuildEventTool" \ - CommandLine="call obj_int_extract.bat $src_path_bare" \ + CommandLine="call obj_int_extract.bat $src_path_bare $plat_no_ws\\\$(ConfigurationName)" \ tag Tool \ Name="VCCLCompilerTool" \ @@ -412,7 +412,6 @@ generate_vcproj() { obj_int_extract) tag Tool \ Name="VCLinkerTool" \ - OutputFile="${name}.exe" \ GenerateDebugInformation="true" \ ;; *) @@ -479,7 +478,7 @@ generate_vcproj() { vpx) tag Tool \ Name="VCPreBuildEventTool" \ - CommandLine="call obj_int_extract.bat $src_path_bare" \ + CommandLine="call obj_int_extract.bat $src_path_bare $plat_no_ws\\\$(ConfigurationName)" \ tag Tool \ Name="VCCLCompilerTool" \ @@ -522,7 +521,6 @@ generate_vcproj() { obj_int_extract) tag Tool \ Name="VCLinkerTool" \ - OutputFile="${name}.exe" \ GenerateDebugInformation="true" \ ;; *) diff --git a/build/make/gen_msvs_vcxproj.sh b/build/make/gen_msvs_vcxproj.sh index a6315b9ee..ce5f93f2d 100755 --- a/build/make/gen_msvs_vcxproj.sh +++ b/build/make/gen_msvs_vcxproj.sh @@ -430,6 +430,14 @@ generate_vcxproj() { Condition="'\$(Configuration)|\$(Platform)'=='$config|$plat'" tag_content OutDir "\$(SolutionDir)$plat_no_ws\\\$(Configuration)\\" tag_content IntDir "$plat_no_ws\\\$(Configuration)\\${name}\\" + if [ "$proj_kind" == "lib" ]; then + if [ "$config" == "Debug" ]; then + config_suffix=d + else + config_suffix="" + fi + tag_content TargetName "${name}${lib_sfx}${config_suffix}" + fi close_tag PropertyGroup done done @@ -438,9 +446,13 @@ generate_vcxproj() { for config in Debug Release; do open_tag ItemDefinitionGroup \ Condition="'\$(Configuration)|\$(Platform)'=='$config|$plat'" - if [ "$name" = "vpx" ]; then + if [ "$name" == "vpx" ]; then + hostplat=$plat + if [ "$hostplat" == "ARM" ]; then + hostplat=Win32 + fi open_tag PreBuildEvent - tag_content Command "call obj_int_extract.bat $src_path_bare" + tag_content Command "call obj_int_extract.bat $src_path_bare $hostplat\\\$(Configuration)" close_tag PreBuildEvent fi open_tag ClCompile @@ -448,7 +460,6 @@ generate_vcxproj() { opt=Disabled runtime=$debug_runtime curlibs=$debug_libs - confsuffix=d case "$name" in obj_int_extract) debug=DEBUG @@ -461,7 +472,6 @@ generate_vcxproj() { opt=MaxSpeed runtime=$release_runtime curlibs=$libs - confsuffix="" tag_content FavorSizeOrSpeed Speed debug=NDEBUG fi @@ -483,9 +493,7 @@ generate_vcxproj() { case "$proj_kind" in exe) open_tag Link - if [ "$name" = "obj_int_extract" ]; then - tag_content OutputFile "${name}.exe" - else + if [ "$name" != "obj_int_extract" ]; then tag_content AdditionalDependencies "$curlibs" tag_content AdditionalLibraryDirectories "$libdirs;%(AdditionalLibraryDirectories)" fi @@ -499,9 +507,6 @@ generate_vcxproj() { close_tag Link ;; lib) - open_tag Lib - tag_content OutputFile "\$(OutDir)${name}${lib_sfx}${confsuffix}.lib" - close_tag Lib ;; esac close_tag ItemDefinitionGroup diff --git a/build/x86-msvs/obj_int_extract.bat b/build/x86-msvs/obj_int_extract.bat index 4e9b0ecd0..44d095dc9 100644 --- a/build/x86-msvs/obj_int_extract.bat +++ b/build/x86-msvs/obj_int_extract.bat @@ -7,6 +7,9 @@ REM in the file PATENTS. All contributing project authors may REM be found in the AUTHORS file in the root of the source tree. echo on +REM Arguments: +REM %1 - Relative path to the directory containing the vp8 source directory. +REM %2 - Path to obj_int_extract.exe. cl /I "./" /I "%1" /nologo /c "%1/vp8/encoder/vp8_asm_enc_offsets.c" -obj_int_extract.exe rvds "vp8_asm_enc_offsets.obj" > "vp8_asm_enc_offsets.asm" +%2\obj_int_extract.exe rvds "vp8_asm_enc_offsets.obj" > "vp8_asm_enc_offsets.asm" @@ -30,7 +30,9 @@ TXT_DOX = $(call enabled,TXT_DOX) EXAMPLE_PATH += $(SRC_PATH_BARE) #for CHANGELOG, README, etc +EXAMPLE_PATH += $(SRC_PATH_BARE)/examples +doxyfile: $(if $(findstring examples, $(ALL_TARGETS)),examples.doxy) doxyfile: libs.doxy_template libs.doxy @echo " [CREATE] $@" @cat $^ > $@ diff --git a/examples.mk b/examples.mk index e4abcf702..5fcf6a350 100644 --- a/examples.mk +++ b/examples.mk @@ -26,13 +26,13 @@ vpxdec.SRCS += ivfdec.c ivfdec.h vpxdec.SRCS += tools_common.c tools_common.h vpxdec.SRCS += webmdec.c webmdec.h vpxdec.SRCS += y4menc.c y4menc.h -vpxdec.SRCS += nestegg/halloc/halloc.h -vpxdec.SRCS += nestegg/halloc/src/align.h -vpxdec.SRCS += nestegg/halloc/src/halloc.c -vpxdec.SRCS += nestegg/halloc/src/hlist.h -vpxdec.SRCS += nestegg/halloc/src/macros.h -vpxdec.SRCS += nestegg/include/nestegg/nestegg.h -vpxdec.SRCS += nestegg/src/nestegg.c +vpxdec.SRCS += third_party/nestegg/halloc/halloc.h +vpxdec.SRCS += third_party/nestegg/halloc/src/align.h +vpxdec.SRCS += third_party/nestegg/halloc/src/halloc.c +vpxdec.SRCS += third_party/nestegg/halloc/src/hlist.h +vpxdec.SRCS += third_party/nestegg/halloc/src/macros.h +vpxdec.SRCS += third_party/nestegg/include/nestegg/nestegg.h +vpxdec.SRCS += third_party/nestegg/src/nestegg.c vpxdec.SRCS += $(LIBYUV_SRCS) vpxdec.GUID = BA5FE66F-38DD-E034-F542-B1578C5FB950 vpxdec.DESCRIPTION = Full featured decoder @@ -54,7 +54,7 @@ vpxenc.SRCS += third_party/libmkv/EbmlWriter.h vpxenc.SRCS += $(LIBYUV_SRCS) vpxenc.GUID = 548DEC74-7A15-4B2B-AFC3-AA102E7C25C1 vpxenc.DESCRIPTION = Full featured encoder -UTILS-$(CONFIG_VP9_ENCODER) += vp9_spatial_scalable_encoder.c +EXAMPLES-$(CONFIG_VP9_ENCODER) += vp9_spatial_scalable_encoder.c vp9_spatial_scalable_encoder.SRCS += args.c args.h vp9_spatial_scalable_encoder.SRCS += ivfenc.c ivfenc.h vp9_spatial_scalable_encoder.SRCS += tools_common.c tools_common.h @@ -285,3 +285,36 @@ INSTALL-BINS-$(CONFIG_MSVS) += $(foreach p,$(VS_PLATFORMS),\ $(addprefix bin/$(p)/,$(ALL_EXAMPLES_BASENAME:.c=.exe))) $(foreach proj,$(call enabled,PROJECTS),\ $(eval $(call vcproj_template,$(proj)))) + +# +# Documentation Rules +# +%.dox: %.c + @echo " [DOXY] $@" + @echo "/*!\page example_$(@F:.dox=) $(@F:.dox=)" > $@ + @echo " \includelineno $(<F)" >> $@ + @echo "*/" >> $@ + +samples.dox: examples.mk + @echo " [DOXY] $@" + @echo "/*!\page samples Sample Code" > $@ + @echo " This SDK includes a number of sample applications."\ + "Each sample documents a feature of the SDK in both prose"\ + "and the associated C code."\ + "The following samples are included: ">>$@ + @$(foreach ex,$(sort $(notdir $(EXAMPLES:.c=))),\ + echo " - \subpage example_$(ex) $($(ex).DESCRIPTION)" >> $@;) + @echo >> $@ + @echo " In addition, the SDK contains a number of utilities."\ + "Since these utilities are built upon the concepts described"\ + "in the sample code listed above, they are not documented in"\ + "pieces like the samples are. Their source is included here"\ + "for reference. The following utilities are included:" >> $@ + @$(foreach ex,$(sort $(UTILS:.c=)),\ + echo " - \subpage example_$(ex) $($(ex).DESCRIPTION)" >> $@;) + @echo "*/" >> $@ + +CLEAN-OBJS += examples.doxy samples.dox $(ALL_EXAMPLES:.c=.dox) +DOCS-yes += examples.doxy samples.dox +examples.doxy: samples.dox $(ALL_EXAMPLES:.c=.dox) + @echo "INPUT += $^" > $@ diff --git a/examples/decode_to_md5.c b/examples/decode_to_md5.c index aabac60ee..28d1ad569 100644 --- a/examples/decode_to_md5.c +++ b/examples/decode_to_md5.c @@ -115,7 +115,7 @@ int main(int argc, char **argv) { size_t frame_size = 0; const unsigned char *frame = vpx_video_reader_get_frame(reader, &frame_size); - if (vpx_codec_decode(&codec, frame, frame_size, NULL, 0)) + if (vpx_codec_decode(&codec, frame, (unsigned int)frame_size, NULL, 0)) die_codec(&codec, "Failed to decode frame"); while ((img = vpx_codec_get_frame(&codec, &iter)) != NULL) { diff --git a/examples/decode_with_drops.c b/examples/decode_with_drops.c index c6f7d43b9..af1aa636b 100644 --- a/examples/decode_with_drops.c +++ b/examples/decode_with_drops.c @@ -120,7 +120,7 @@ int main(int argc, char **argv) { int skip; const unsigned char *frame = vpx_video_reader_get_frame(reader, &frame_size); - if (vpx_codec_decode(&codec, frame, frame_size, NULL, 0)) + if (vpx_codec_decode(&codec, frame, (unsigned int)frame_size, NULL, 0)) die_codec(&codec, "Failed to decode frame."); ++frame_cnt; diff --git a/examples/error_resilient.c b/examples/error_resilient.c index ef0a6c38d..19235c817 100644 --- a/examples/error_resilient.c +++ b/examples/error_resilient.c @@ -118,7 +118,7 @@ static void write_ivf_frame_header(FILE *outfile, return; pts = pkt->data.frame.pts; - mem_put_le32(header, pkt->data.frame.sz); + mem_put_le32(header, (unsigned int)pkt->data.frame.sz); mem_put_le32(header+4, pts&0xFFFFFFFF); mem_put_le32(header+8, pts >> 32); diff --git a/examples/force_keyframe.c b/examples/force_keyframe.c index f03b3d0d8..6531e4713 100644 --- a/examples/force_keyframe.c +++ b/examples/force_keyframe.c @@ -119,7 +119,7 @@ static void write_ivf_frame_header(FILE *outfile, return; pts = pkt->data.frame.pts; - mem_put_le32(header, pkt->data.frame.sz); + mem_put_le32(header, (unsigned int)pkt->data.frame.sz); mem_put_le32(header+4, pts&0xFFFFFFFF); mem_put_le32(header+8, pts >> 32); diff --git a/examples/postproc.c b/examples/postproc.c index 2912fe6de..be08e920b 100644 --- a/examples/postproc.c +++ b/examples/postproc.c @@ -118,7 +118,7 @@ int main(int argc, char **argv) { }; // Decode the frame with 15ms deadline - if (vpx_codec_decode(&codec, frame, frame_size, NULL, 15000)) + if (vpx_codec_decode(&codec, frame, (unsigned int)frame_size, NULL, 15000)) die_codec(&codec, "Failed to decode frame"); while ((img = vpx_codec_get_frame(&codec, &iter)) != NULL) { diff --git a/examples/simple_decoder.c b/examples/simple_decoder.c index b0ca77d1d..8c150510a 100644 --- a/examples/simple_decoder.c +++ b/examples/simple_decoder.c @@ -134,7 +134,7 @@ int main(int argc, char **argv) { size_t frame_size = 0; const unsigned char *frame = vpx_video_reader_get_frame(reader, &frame_size); - if (vpx_codec_decode(&codec, frame, frame_size, NULL, 0)) + if (vpx_codec_decode(&codec, frame, (unsigned int)frame_size, NULL, 0)) die_codec(&codec, "Failed to decode frame."); while ((img = vpx_codec_get_frame(&codec, &iter)) != NULL) { diff --git a/examples/twopass_encoder.c b/examples/twopass_encoder.c index f16db663d..8bca18c7e 100644 --- a/examples/twopass_encoder.c +++ b/examples/twopass_encoder.c @@ -69,9 +69,9 @@ void usage_exit() { static void get_frame_stats(vpx_codec_ctx_t *ctx, const vpx_image_t *img, vpx_codec_pts_t pts, - uint64_t duration, + unsigned int duration, vpx_enc_frame_flags_t flags, - uint64_t deadline, + unsigned int deadline, vpx_fixed_buf_t *stats) { vpx_codec_iter_t iter = NULL; const vpx_codec_cx_pkt_t *pkt = NULL; @@ -94,9 +94,9 @@ static void get_frame_stats(vpx_codec_ctx_t *ctx, static void encode_frame(vpx_codec_ctx_t *ctx, const vpx_image_t *img, vpx_codec_pts_t pts, - uint64_t duration, + unsigned int duration, vpx_enc_frame_flags_t flags, - uint64_t deadline, + unsigned int deadline, VpxVideoWriter *writer) { vpx_codec_iter_t iter = NULL; const vpx_codec_cx_pkt_t *pkt = NULL; diff --git a/examples/vp8_set_maps.c b/examples/vp8_set_maps.c index ec9bc65f5..f3cc9a768 100644 --- a/examples/vp8_set_maps.c +++ b/examples/vp8_set_maps.c @@ -62,7 +62,7 @@ void usage_exit() { static void set_roi_map(const vpx_codec_enc_cfg_t *cfg, vpx_codec_ctx_t *codec) { - int i; + unsigned int i; vpx_roi_map_t roi = {0}; roi.rows = cfg->g_h / 16; @@ -95,7 +95,7 @@ static void set_roi_map(const vpx_codec_enc_cfg_t *cfg, static void set_active_map(const vpx_codec_enc_cfg_t *cfg, vpx_codec_ctx_t *codec) { - int i; + unsigned int i; vpx_active_map_t map = {0}; map.rows = cfg->g_h / 16; diff --git a/examples/vp8cx_set_ref.c b/examples/vp8cx_set_ref.c index 5a675788a..f87dd35c3 100644 --- a/examples/vp8cx_set_ref.c +++ b/examples/vp8cx_set_ref.c @@ -139,7 +139,7 @@ static void write_ivf_frame_header(FILE *outfile, return; pts = pkt->data.frame.pts; - mem_put_le32(header, pkt->data.frame.sz); + mem_put_le32(header, (unsigned int)pkt->data.frame.sz); mem_put_le32(header+4, pts&0xFFFFFFFF); mem_put_le32(header+8, pts >> 32); diff --git a/vp9_spatial_scalable_encoder.c b/examples/vp9_spatial_scalable_encoder.c index 9f526b0a4..9f526b0a4 100644 --- a/vp9_spatial_scalable_encoder.c +++ b/examples/vp9_spatial_scalable_encoder.c diff --git a/examples/vpx_temporal_scalable_patterns.c b/examples/vpx_temporal_scalable_patterns.c index 28bb7ff4d..29a266d29 100644 --- a/examples/vpx_temporal_scalable_patterns.c +++ b/examples/vpx_temporal_scalable_patterns.c @@ -41,23 +41,23 @@ struct RateControlMetrics { // Number of encoded non-key frames per layer. int layer_enc_frames[VPX_TS_MAX_LAYERS]; // Framerate per layer layer (cumulative). - float layer_framerate[VPX_TS_MAX_LAYERS]; + double layer_framerate[VPX_TS_MAX_LAYERS]; // Target average frame size per layer (per-frame-bandwidth per layer). - float layer_pfb[VPX_TS_MAX_LAYERS]; + double layer_pfb[VPX_TS_MAX_LAYERS]; // Actual average frame size per layer. - float layer_avg_frame_size[VPX_TS_MAX_LAYERS]; + double layer_avg_frame_size[VPX_TS_MAX_LAYERS]; // Average rate mismatch per layer (|target - actual| / target). - float layer_avg_rate_mismatch[VPX_TS_MAX_LAYERS]; + double layer_avg_rate_mismatch[VPX_TS_MAX_LAYERS]; // Actual encoding bitrate per layer (cumulative). - float layer_encoding_bitrate[VPX_TS_MAX_LAYERS]; + double layer_encoding_bitrate[VPX_TS_MAX_LAYERS]; }; static void set_rate_control_metrics(struct RateControlMetrics *rc, vpx_codec_enc_cfg_t *cfg) { - int i = 0; + unsigned int i = 0; // Set the layer (cumulative) framerate and the target layer (non-cumulative) // per-frame-bandwidth, for the rate control encoding stats below. - float framerate = cfg->g_timebase.den / cfg->g_timebase.num; + const double framerate = cfg->g_timebase.den / cfg->g_timebase.num; rc->layer_framerate[0] = framerate / cfg->ts_rate_decimator[0]; rc->layer_pfb[0] = 1000.0 * cfg->ts_target_bitrate[0] / rc->layer_framerate[0]; @@ -80,8 +80,8 @@ static void set_rate_control_metrics(struct RateControlMetrics *rc, static void printout_rate_control_summary(struct RateControlMetrics *rc, vpx_codec_enc_cfg_t *cfg, int frame_cnt) { - int i = 0; - int check_num_frames = 0; + unsigned int i = 0; + int tot_num_frames = 0; printf("Total number of processed frames: %d\n\n", frame_cnt -1); printf("Rate control layer stats for %d layer(s):\n\n", cfg->ts_number_layers); @@ -89,8 +89,9 @@ static void printout_rate_control_summary(struct RateControlMetrics *rc, const int num_dropped = (i > 0) ? (rc->layer_input_frames[i] - rc->layer_enc_frames[i]) : (rc->layer_input_frames[i] - rc->layer_enc_frames[i] - 1); + tot_num_frames += rc->layer_input_frames[i]; rc->layer_encoding_bitrate[i] = 0.001 * rc->layer_framerate[i] * - rc->layer_encoding_bitrate[i] / rc->layer_tot_enc_frames[i]; + rc->layer_encoding_bitrate[i] / tot_num_frames; rc->layer_avg_frame_size[i] = rc->layer_avg_frame_size[i] / rc->layer_enc_frames[i]; rc->layer_avg_rate_mismatch[i] = 100.0 * rc->layer_avg_rate_mismatch[i] / @@ -105,10 +106,9 @@ static void printout_rate_control_summary(struct RateControlMetrics *rc, "and perc dropped frames: %d %d %f \n", rc->layer_input_frames[i], rc->layer_enc_frames[i], 100.0 * num_dropped / rc->layer_input_frames[i]); - check_num_frames += rc->layer_input_frames[i]; printf("\n"); } - if ((frame_cnt - 1) != check_num_frames) + if ((frame_cnt - 1) != tot_num_frames) die("Error: Number of input frames not equal to output! \n"); } @@ -432,7 +432,7 @@ int main(int argc, char **argv) { int frame_avail; int got_data; int flags = 0; - int i; + unsigned int i; int pts = 0; // PTS starts at 0. int frame_duration = 1; // 1 timebase tick per frame. int layering_mode = 0; @@ -492,7 +492,7 @@ int main(int argc, char **argv) { cfg.g_timebase.num = strtol(argv[6], NULL, 0); cfg.g_timebase.den = strtol(argv[7], NULL, 0); - for (i = 10; i < 10 + mode_to_num_layers[layering_mode]; ++i) { + for (i = 10; (int)i < 10 + mode_to_num_layers[layering_mode]; ++i) { cfg.ts_target_bitrate[i - 10] = strtol(argv[i], NULL, 0); } @@ -516,9 +516,6 @@ int main(int argc, char **argv) { // Disable automatic keyframe placement. cfg.kf_min_dist = cfg.kf_max_dist = 3000; - // Default setting for bitrate: used in special case of 1 layer (case 0). - cfg.rc_target_bitrate = cfg.ts_target_bitrate[0]; - set_temporal_layer_pattern(layering_mode, &cfg, layer_flags, @@ -526,6 +523,10 @@ int main(int argc, char **argv) { set_rate_control_metrics(&rc, &cfg); + // Target bandwidth for the whole stream. + // Set to ts_target_bitrate for highest layer (total bitrate). + cfg.rc_target_bitrate = cfg.ts_target_bitrate[cfg.ts_number_layers - 1]; + // Open input file. if (!(infile = fopen(argv[1], "rb"))) { die("Failed to open %s for reading", argv[1]); diff --git a/nestegg/.gitignore b/nestegg/.gitignore deleted file mode 100644 index b2ba99c3f..000000000 --- a/nestegg/.gitignore +++ /dev/null @@ -1,40 +0,0 @@ -*.lo -*.o -*.swp -*~ -.deps -.dirstamp -.libs -Makefile -Makefile.in -_stdint.h -aclocal.m4 -autom4te.cache -compile -config.guess -config.h -config.h.in -config.log -config.status -config.sub -configure -depcomp -docs/Doxyfile -docs/doxygen-build.stamp -docs/html -install-sh -libtool -ltmain.sh -m4/libtool.m4 -m4/ltoptions.m4 -m4/ltsugar.m4 -m4/ltversion.m4 -m4/lt~obsolete.m4 -missing -nestegg-uninstalled.pc -nestegg.pc -src/.dirstamp -src/libnestegg.la -stamp-h1 -test/test -include/nestegg/nestegg-stdint.h diff --git a/nestegg/AUTHORS b/nestegg/AUTHORS deleted file mode 100644 index 8204f40f4..000000000 --- a/nestegg/AUTHORS +++ /dev/null @@ -1 +0,0 @@ -Matthew Gregan <kinetik@flim.org> diff --git a/nestegg/Makefile.am b/nestegg/Makefile.am deleted file mode 100644 index 500699160..000000000 --- a/nestegg/Makefile.am +++ /dev/null @@ -1,51 +0,0 @@ -AUTOMAKE_OPTIONS = foreign 1.11 no-dist-gzip dist-bzip2 subdir-objects -ACLOCAL_AMFLAGS = -I m4 - -INCLUDES = -I$(top_srcdir)/include -I. -I$(top_srcdir)/halloc -AM_CFLAGS = -ansi -pedantic -Wall -Wextra -Wno-long-long -O0 -g - -SUBDIRS = docs - -EXTRA_DIST = \ - AUTHORS README LICENSE \ - nestegg-uninstalled.pc.in \ - m4/as-ac-expand.m4 \ - m4/pkg.m4 \ - m4/ax_create_stdint_h.m4 \ - halloc/src/halloc.c \ - halloc/halloc.h \ - halloc/src/align.h \ - halloc/src/hlist.h \ - halloc/src/macros.h - -pkgconfigdir = $(libdir)/pkgconfig -pkgconfig_DATA = nestegg.pc - -nesteggincludedir = $(includedir)/nestegg -nestegginclude_HEADERS = include/nestegg/nestegg.h include/nestegg/nestegg-stdint.h - -lib_LTLIBRARIES = src/libnestegg.la - -src_libnestegg_la_SOURCES = \ - src/nestegg.c \ - halloc/src/halloc.c \ - halloc/halloc.h \ - halloc/src/align.h \ - halloc/src/hlist.h \ - halloc/src/macros.h - -check_PROGRAMS = test/test - -test_test_SOURCES = test/test.c -test_test_LDADD = src/libnestegg.la - -DISTCLEANFILES = include/nestegg/nestegg-stdint.h - -dist-hook: - find $(distdir) -type d -name '.git' | xargs rm -rf - -debug: - $(MAKE) all CFLAGS="@DEBUG@" - -profile: - $(MAKE) all CFLAGS="@PROFILE@" diff --git a/nestegg/configure.ac b/nestegg/configure.ac deleted file mode 100644 index 70f6e0d59..000000000 --- a/nestegg/configure.ac +++ /dev/null @@ -1,124 +0,0 @@ -dnl ------------------------------------------------ -dnl Initialization and Versioning -dnl ------------------------------------------------ - -AC_INIT(libnestegg,[0.1git]) - -AC_CANONICAL_HOST -AC_CANONICAL_TARGET - -AC_CONFIG_MACRO_DIR([m4]) - -AM_CONFIG_HEADER([config.h]) -AC_CONFIG_SRCDIR([src/nestegg.c]) -AM_INIT_AUTOMAKE - -m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES([yes])]) - -dnl Library versioning -dnl CURRENT, REVISION, AGE -dnl - library source changed -> increment REVISION -dnl - interfaces added/removed/changed -> increment CURRENT, REVISION = 0 -dnl - interfaces added -> increment AGE -dnl - interfaces removed -> AGE = 0 - -NESTEGG_CURRENT=0 -NESTEGG_REVISION=0 -NESTEGG_AGE=1 -AC_SUBST(NESTEGG_CURRENT) -AC_SUBST(NESTEGG_REVISION) -AC_SUBST(NESTEGG_AGE) - - -dnl -------------------------------------------------- -dnl Check for programs -dnl -------------------------------------------------- - -dnl save $CFLAGS since AC_PROG_CC likes to insert "-g -O2" -dnl if $CFLAGS is blank -cflags_save="$CFLAGS" -AC_PROG_CC -AC_PROG_CPP -CFLAGS="$cflags_save" - -AM_PROG_CC_C_O -AC_LIBTOOL_WIN32_DLL -AM_PROG_LIBTOOL - -dnl Check for doxygen -AC_ARG_ENABLE([doc], - AS_HELP_STRING([--enable-doc], [Build API documentation]), - [ac_enable_doc=$enableval], [ac_enable_doc=auto]) - -if test "x$ac_enable_doc" != "xno"; then - AC_CHECK_PROG(HAVE_DOXYGEN, doxygen, true, false) - - if test "x$HAVE_DOXYGEN" = "xfalse" -a "x$ac_enable_doc" = "xyes"; then - AC_MSG_ERROR([*** API documentation explicitly requested but Doxygen not found]) - fi -else - HAVE_DOXYGEN=false -fi -AM_CONDITIONAL(HAVE_DOXYGEN,$HAVE_DOXYGEN) -if test $HAVE_DOXYGEN = "false"; then - AC_MSG_WARN([*** doxygen not found, API documentation will not be built]) -fi - -# Generate portable stdint.h replacement -AX_CREATE_STDINT_H(include/nestegg/nestegg-stdint.h) - -# Test whenever ld supports -version-script -AC_PROG_LD -AC_PROG_LD_GNU -AC_MSG_CHECKING([how to control symbol export]) - -dnl -------------------------------------------------- -dnl Do substitutions -dnl -------------------------------------------------- - -AC_SUBST(DEBUG) -AC_SUBST(PROFILE) - -AC_OUTPUT([ - Makefile - docs/Makefile - docs/Doxyfile - nestegg.pc - nestegg-uninstalled.pc -]) - -AS_AC_EXPAND(LIBDIR, ${libdir}) -AS_AC_EXPAND(INCLUDEDIR, ${includedir}) -AS_AC_EXPAND(BINDIR, ${bindir}) -AS_AC_EXPAND(DOCDIR, ${docdir}) - -if test $HAVE_DOXYGEN = "false"; then - doc_build="no" -else - doc_build="yes" -fi - -AC_MSG_RESULT([ ------------------------------------------------------------------------- - $PACKAGE $VERSION: Automatic configuration OK. - - General configuration: - - API Documentation: .......... ${doc_build} - - Installation paths: - - libnestegg: .................. ${LIBDIR} - C header files: .............. ${INCLUDEDIR}/nestegg - Documentation: ............... ${DOCDIR} - - Building: - - Type 'make' to compile $PACKAGE. - - Type 'make install' to install $PACKAGE. - - Example programs will be built but not installed. ------------------------------------------------------------------------- -]) - diff --git a/nestegg/docs/Doxyfile.in b/nestegg/docs/Doxyfile.in deleted file mode 100644 index e0e9249aa..000000000 --- a/nestegg/docs/Doxyfile.in +++ /dev/null @@ -1,1551 +0,0 @@ -# Doxyfile 1.6.2 - -# This file describes the settings to be used by the documentation system -# doxygen (www.doxygen.org) for a project -# -# All text after a hash (#) is considered a comment and will be ignored -# The format is: -# TAG = value [value, ...] -# For lists items can also be appended using: -# TAG += value [value, ...] -# Values that contain spaces should be placed between quotes (" ") - -#--------------------------------------------------------------------------- -# Project related configuration options -#--------------------------------------------------------------------------- - -# This tag specifies the encoding used for all characters in the config file -# that follow. The default is UTF-8 which is also the encoding used for all -# text before the first occurrence of this tag. Doxygen uses libiconv (or the -# iconv built into libc) for the transcoding. See -# http://www.gnu.org/software/libiconv for the list of possible encodings. - -DOXYFILE_ENCODING = UTF-8 - -# The PROJECT_NAME tag is a single word (or a sequence of words surrounded -# by quotes) that should identify the project. - -PROJECT_NAME = @PACKAGE@ - -# The PROJECT_NUMBER tag can be used to enter a project or revision number. -# This could be handy for archiving the generated documentation or -# if some version control system is used. - -PROJECT_NUMBER = @VERSION@ - -# The OUTPUT_DIRECTORY tag is used to specify the (relative or absolute) -# base path where the generated documentation will be put. -# If a relative path is entered, it will be relative to the location -# where doxygen was started. If left blank the current directory will be used. - -OUTPUT_DIRECTORY = . - -# If the CREATE_SUBDIRS tag is set to YES, then doxygen will create -# 4096 sub-directories (in 2 levels) under the output directory of each output -# format and will distribute the generated files over these directories. -# Enabling this option can be useful when feeding doxygen a huge amount of -# source files, where putting all generated files in the same directory would -# otherwise cause performance problems for the file system. - -CREATE_SUBDIRS = NO - -# The OUTPUT_LANGUAGE tag is used to specify the language in which all -# documentation generated by doxygen is written. Doxygen will use this -# information to generate all constant output in the proper language. -# The default language is English, other supported languages are: -# Afrikaans, Arabic, Brazilian, Catalan, Chinese, Chinese-Traditional, -# Croatian, Czech, Danish, Dutch, Esperanto, Farsi, Finnish, French, German, -# Greek, Hungarian, Italian, Japanese, Japanese-en (Japanese with English -# messages), Korean, Korean-en, Lithuanian, Norwegian, Macedonian, Persian, -# Polish, Portuguese, Romanian, Russian, Serbian, Serbian-Cyrilic, Slovak, -# Slovene, Spanish, Swedish, Ukrainian, and Vietnamese. - -OUTPUT_LANGUAGE = English - -# If the BRIEF_MEMBER_DESC tag is set to YES (the default) Doxygen will -# include brief member descriptions after the members that are listed in -# the file and class documentation (similar to JavaDoc). -# Set to NO to disable this. - -BRIEF_MEMBER_DESC = YES - -# If the REPEAT_BRIEF tag is set to YES (the default) Doxygen will prepend -# the brief description of a member or function before the detailed description. -# Note: if both HIDE_UNDOC_MEMBERS and BRIEF_MEMBER_DESC are set to NO, the -# brief descriptions will be completely suppressed. - -REPEAT_BRIEF = YES - -# This tag implements a quasi-intelligent brief description abbreviator -# that is used to form the text in various listings. Each string -# in this list, if found as the leading text of the brief description, will be -# stripped from the text and the result after processing the whole list, is -# used as the annotated text. Otherwise, the brief description is used as-is. -# If left blank, the following values are used ("$name" is automatically -# replaced with the name of the entity): "The $name class" "The $name widget" -# "The $name file" "is" "provides" "specifies" "contains" -# "represents" "a" "an" "the" - -ABBREVIATE_BRIEF = - -# If the ALWAYS_DETAILED_SEC and REPEAT_BRIEF tags are both set to YES then -# Doxygen will generate a detailed section even if there is only a brief -# description. - -ALWAYS_DETAILED_SEC = NO - -# If the INLINE_INHERITED_MEMB tag is set to YES, doxygen will show all -# inherited members of a class in the documentation of that class as if those -# members were ordinary class members. Constructors, destructors and assignment -# operators of the base classes will not be shown. - -INLINE_INHERITED_MEMB = NO - -# If the FULL_PATH_NAMES tag is set to YES then Doxygen will prepend the full -# path before files name in the file list and in the header files. If set -# to NO the shortest path that makes the file name unique will be used. - -FULL_PATH_NAMES = YES - -# If the FULL_PATH_NAMES tag is set to YES then the STRIP_FROM_PATH tag -# can be used to strip a user-defined part of the path. Stripping is -# only done if one of the specified strings matches the left-hand part of -# the path. The tag can be used to show relative paths in the file list. -# If left blank the directory from which doxygen is run is used as the -# path to strip. - -STRIP_FROM_PATH = - -# The STRIP_FROM_INC_PATH tag can be used to strip a user-defined part of -# the path mentioned in the documentation of a class, which tells -# the reader which header file to include in order to use a class. -# If left blank only the name of the header file containing the class -# definition is used. Otherwise one should specify the include paths that -# are normally passed to the compiler using the -I flag. - -STRIP_FROM_INC_PATH = - -# If the SHORT_NAMES tag is set to YES, doxygen will generate much shorter -# (but less readable) file names. This can be useful is your file systems -# doesn't support long names like on DOS, Mac, or CD-ROM. - -SHORT_NAMES = NO - -# If the JAVADOC_AUTOBRIEF tag is set to YES then Doxygen -# will interpret the first line (until the first dot) of a JavaDoc-style -# comment as the brief description. If set to NO, the JavaDoc -# comments will behave just like regular Qt-style comments -# (thus requiring an explicit @brief command for a brief description.) - -JAVADOC_AUTOBRIEF = YES - -# If the QT_AUTOBRIEF tag is set to YES then Doxygen will -# interpret the first line (until the first dot) of a Qt-style -# comment as the brief description. If set to NO, the comments -# will behave just like regular Qt-style comments (thus requiring -# an explicit \brief command for a brief description.) - -QT_AUTOBRIEF = NO - -# The MULTILINE_CPP_IS_BRIEF tag can be set to YES to make Doxygen -# treat a multi-line C++ special comment block (i.e. a block of //! or /// -# comments) as a brief description. This used to be the default behaviour. -# The new default is to treat a multi-line C++ comment block as a detailed -# description. Set this tag to YES if you prefer the old behaviour instead. - -MULTILINE_CPP_IS_BRIEF = NO - -# If the INHERIT_DOCS tag is set to YES (the default) then an undocumented -# member inherits the documentation from any documented member that it -# re-implements. - -INHERIT_DOCS = YES - -# If the SEPARATE_MEMBER_PAGES tag is set to YES, then doxygen will produce -# a new page for each member. If set to NO, the documentation of a member will -# be part of the file/class/namespace that contains it. - -SEPARATE_MEMBER_PAGES = NO - -# The TAB_SIZE tag can be used to set the number of spaces in a tab. -# Doxygen uses this value to replace tabs by spaces in code fragments. - -TAB_SIZE = 8 - -# This tag can be used to specify a number of aliases that acts -# as commands in the documentation. An alias has the form "name=value". -# For example adding "sideeffect=\par Side Effects:\n" will allow you to -# put the command \sideeffect (or @sideeffect) in the documentation, which -# will result in a user-defined paragraph with heading "Side Effects:". -# You can put \n's in the value part of an alias to insert newlines. - -ALIASES = - -# Set the OPTIMIZE_OUTPUT_FOR_C tag to YES if your project consists of C -# sources only. Doxygen will then generate output that is more tailored for C. -# For instance, some of the names that are used will be different. The list -# of all members will be omitted, etc. - -OPTIMIZE_OUTPUT_FOR_C = YES - -# Set the OPTIMIZE_OUTPUT_JAVA tag to YES if your project consists of Java -# sources only. Doxygen will then generate output that is more tailored for -# Java. For instance, namespaces will be presented as packages, qualified -# scopes will look different, etc. - -OPTIMIZE_OUTPUT_JAVA = NO - -# Set the OPTIMIZE_FOR_FORTRAN tag to YES if your project consists of Fortran -# sources only. Doxygen will then generate output that is more tailored for -# Fortran. - -OPTIMIZE_FOR_FORTRAN = NO - -# Set the OPTIMIZE_OUTPUT_VHDL tag to YES if your project consists of VHDL -# sources. Doxygen will then generate output that is tailored for -# VHDL. - -OPTIMIZE_OUTPUT_VHDL = NO - -# Doxygen selects the parser to use depending on the extension of the files it parses. -# With this tag you can assign which parser to use for a given extension. -# Doxygen has a built-in mapping, but you can override or extend it using this tag. -# The format is ext=language, where ext is a file extension, and language is one of -# the parsers supported by doxygen: IDL, Java, Javascript, C#, C, C++, D, PHP, -# Objective-C, Python, Fortran, VHDL, C, C++. For instance to make doxygen treat -# .inc files as Fortran files (default is PHP), and .f files as C (default is Fortran), -# use: inc=Fortran f=C. Note that for custom extensions you also need to set FILE_PATTERNS otherwise the files are not read by doxygen. - -EXTENSION_MAPPING = - -# If you use STL classes (i.e. std::string, std::vector, etc.) but do not want -# to include (a tag file for) the STL sources as input, then you should -# set this tag to YES in order to let doxygen match functions declarations and -# definitions whose arguments contain STL classes (e.g. func(std::string); v.s. -# func(std::string) {}). This also make the inheritance and collaboration -# diagrams that involve STL classes more complete and accurate. - -BUILTIN_STL_SUPPORT = NO - -# If you use Microsoft's C++/CLI language, you should set this option to YES to -# enable parsing support. - -CPP_CLI_SUPPORT = NO - -# Set the SIP_SUPPORT tag to YES if your project consists of sip sources only. -# Doxygen will parse them like normal C++ but will assume all classes use public -# instead of private inheritance when no explicit protection keyword is present. - -SIP_SUPPORT = NO - -# For Microsoft's IDL there are propget and propput attributes to indicate getter -# and setter methods for a property. Setting this option to YES (the default) -# will make doxygen to replace the get and set methods by a property in the -# documentation. This will only work if the methods are indeed getting or -# setting a simple type. If this is not the case, or you want to show the -# methods anyway, you should set this option to NO. - -IDL_PROPERTY_SUPPORT = YES - -# If member grouping is used in the documentation and the DISTRIBUTE_GROUP_DOC -# tag is set to YES, then doxygen will reuse the documentation of the first -# member in the group (if any) for the other members of the group. By default -# all members of a group must be documented explicitly. - -DISTRIBUTE_GROUP_DOC = NO - -# Set the SUBGROUPING tag to YES (the default) to allow class member groups of -# the same type (for instance a group of public functions) to be put as a -# subgroup of that type (e.g. under the Public Functions section). Set it to -# NO to prevent subgrouping. Alternatively, this can be done per class using -# the \nosubgrouping command. - -SUBGROUPING = YES - -# When TYPEDEF_HIDES_STRUCT is enabled, a typedef of a struct, union, or enum -# is documented as struct, union, or enum with the name of the typedef. So -# typedef struct TypeS {} TypeT, will appear in the documentation as a struct -# with name TypeT. When disabled the typedef will appear as a member of a file, -# namespace, or class. And the struct will be named TypeS. This can typically -# be useful for C code in case the coding convention dictates that all compound -# types are typedef'ed and only the typedef is referenced, never the tag name. - -TYPEDEF_HIDES_STRUCT = NO - -# The SYMBOL_CACHE_SIZE determines the size of the internal cache use to -# determine which symbols to keep in memory and which to flush to disk. -# When the cache is full, less often used symbols will be written to disk. -# For small to medium size projects (<1000 input files) the default value is -# probably good enough. For larger projects a too small cache size can cause -# doxygen to be busy swapping symbols to and from disk most of the time -# causing a significant performance penality. -# If the system has enough physical memory increasing the cache will improve the -# performance by keeping more symbols in memory. Note that the value works on -# a logarithmic scale so increasing the size by one will rougly double the -# memory usage. The cache size is given by this formula: -# 2^(16+SYMBOL_CACHE_SIZE). The valid range is 0..9, the default is 0, -# corresponding to a cache size of 2^16 = 65536 symbols - -SYMBOL_CACHE_SIZE = 0 - -#--------------------------------------------------------------------------- -# Build related configuration options -#--------------------------------------------------------------------------- - -# If the EXTRACT_ALL tag is set to YES doxygen will assume all entities in -# documentation are documented, even if no documentation was available. -# Private class members and static file members will be hidden unless -# the EXTRACT_PRIVATE and EXTRACT_STATIC tags are set to YES - -EXTRACT_ALL = NO - -# If the EXTRACT_PRIVATE tag is set to YES all private members of a class -# will be included in the documentation. - -EXTRACT_PRIVATE = NO - -# If the EXTRACT_STATIC tag is set to YES all static members of a file -# will be included in the documentation. - -EXTRACT_STATIC = NO - -# If the EXTRACT_LOCAL_CLASSES tag is set to YES classes (and structs) -# defined locally in source files will be included in the documentation. -# If set to NO only classes defined in header files are included. - -EXTRACT_LOCAL_CLASSES = YES - -# This flag is only useful for Objective-C code. When set to YES local -# methods, which are defined in the implementation section but not in -# the interface are included in the documentation. -# If set to NO (the default) only methods in the interface are included. - -EXTRACT_LOCAL_METHODS = NO - -# If this flag is set to YES, the members of anonymous namespaces will be -# extracted and appear in the documentation as a namespace called -# 'anonymous_namespace{file}', where file will be replaced with the base -# name of the file that contains the anonymous namespace. By default -# anonymous namespace are hidden. - -EXTRACT_ANON_NSPACES = NO - -# If the HIDE_UNDOC_MEMBERS tag is set to YES, Doxygen will hide all -# undocumented members of documented classes, files or namespaces. -# If set to NO (the default) these members will be included in the -# various overviews, but no documentation section is generated. -# This option has no effect if EXTRACT_ALL is enabled. - -HIDE_UNDOC_MEMBERS = NO - -# If the HIDE_UNDOC_CLASSES tag is set to YES, Doxygen will hide all -# undocumented classes that are normally visible in the class hierarchy. -# If set to NO (the default) these classes will be included in the various -# overviews. This option has no effect if EXTRACT_ALL is enabled. - -HIDE_UNDOC_CLASSES = NO - -# If the HIDE_FRIEND_COMPOUNDS tag is set to YES, Doxygen will hide all -# friend (class|struct|union) declarations. -# If set to NO (the default) these declarations will be included in the -# documentation. - -HIDE_FRIEND_COMPOUNDS = NO - -# If the HIDE_IN_BODY_DOCS tag is set to YES, Doxygen will hide any -# documentation blocks found inside the body of a function. -# If set to NO (the default) these blocks will be appended to the -# function's detailed documentation block. - -HIDE_IN_BODY_DOCS = NO - -# The INTERNAL_DOCS tag determines if documentation -# that is typed after a \internal command is included. If the tag is set -# to NO (the default) then the documentation will be excluded. -# Set it to YES to include the internal documentation. - -INTERNAL_DOCS = NO - -# If the CASE_SENSE_NAMES tag is set to NO then Doxygen will only generate -# file names in lower-case letters. If set to YES upper-case letters are also -# allowed. This is useful if you have classes or files whose names only differ -# in case and if your file system supports case sensitive file names. Windows -# and Mac users are advised to set this option to NO. - -CASE_SENSE_NAMES = NO - -# If the HIDE_SCOPE_NAMES tag is set to NO (the default) then Doxygen -# will show members with their full class and namespace scopes in the -# documentation. If set to YES the scope will be hidden. - -HIDE_SCOPE_NAMES = NO - -# If the SHOW_INCLUDE_FILES tag is set to YES (the default) then Doxygen -# will put a list of the files that are included by a file in the documentation -# of that file. - -SHOW_INCLUDE_FILES = YES - -# If the FORCE_LOCAL_INCLUDES tag is set to YES then Doxygen -# will list include files with double quotes in the documentation -# rather than with sharp brackets. - -FORCE_LOCAL_INCLUDES = NO - -# If the INLINE_INFO tag is set to YES (the default) then a tag [inline] -# is inserted in the documentation for inline members. - -INLINE_INFO = YES - -# If the SORT_MEMBER_DOCS tag is set to YES (the default) then doxygen -# will sort the (detailed) documentation of file and class members -# alphabetically by member name. If set to NO the members will appear in -# declaration order. - -SORT_MEMBER_DOCS = NO - -# If the SORT_BRIEF_DOCS tag is set to YES then doxygen will sort the -# brief documentation of file, namespace and class members alphabetically -# by member name. If set to NO (the default) the members will appear in -# declaration order. - -SORT_BRIEF_DOCS = NO - -# If the SORT_MEMBERS_CTORS_1ST tag is set to YES then doxygen will sort the (brief and detailed) documentation of class members so that constructors and destructors are listed first. If set to NO (the default) the constructors will appear in the respective orders defined by SORT_MEMBER_DOCS and SORT_BRIEF_DOCS. This tag will be ignored for brief docs if SORT_BRIEF_DOCS is set to NO and ignored for detailed docs if SORT_MEMBER_DOCS is set to NO. - -SORT_MEMBERS_CTORS_1ST = NO - -# If the SORT_GROUP_NAMES tag is set to YES then doxygen will sort the -# hierarchy of group names into alphabetical order. If set to NO (the default) -# the group names will appear in their defined order. - -SORT_GROUP_NAMES = NO - -# If the SORT_BY_SCOPE_NAME tag is set to YES, the class list will be -# sorted by fully-qualified names, including namespaces. If set to -# NO (the default), the class list will be sorted only by class name, -# not including the namespace part. -# Note: This option is not very useful if HIDE_SCOPE_NAMES is set to YES. -# Note: This option applies only to the class list, not to the -# alphabetical list. - -SORT_BY_SCOPE_NAME = NO - -# The GENERATE_TODOLIST tag can be used to enable (YES) or -# disable (NO) the todo list. This list is created by putting \todo -# commands in the documentation. - -GENERATE_TODOLIST = YES - -# The GENERATE_TESTLIST tag can be used to enable (YES) or -# disable (NO) the test list. This list is created by putting \test -# commands in the documentation. - -GENERATE_TESTLIST = YES - -# The GENERATE_BUGLIST tag can be used to enable (YES) or -# disable (NO) the bug list. This list is created by putting \bug -# commands in the documentation. - -GENERATE_BUGLIST = YES - -# The GENERATE_DEPRECATEDLIST tag can be used to enable (YES) or -# disable (NO) the deprecated list. This list is created by putting -# \deprecated commands in the documentation. - -GENERATE_DEPRECATEDLIST= YES - -# The ENABLED_SECTIONS tag can be used to enable conditional -# documentation sections, marked by \if sectionname ... \endif. - -ENABLED_SECTIONS = - -# The MAX_INITIALIZER_LINES tag determines the maximum number of lines -# the initial value of a variable or define consists of for it to appear in -# the documentation. If the initializer consists of more lines than specified -# here it will be hidden. Use a value of 0 to hide initializers completely. -# The appearance of the initializer of individual variables and defines in the -# documentation can be controlled using \showinitializer or \hideinitializer -# command in the documentation regardless of this setting. - -MAX_INITIALIZER_LINES = 30 - -# Set the SHOW_USED_FILES tag to NO to disable the list of files generated -# at the bottom of the documentation of classes and structs. If set to YES the -# list will mention the files that were used to generate the documentation. - -SHOW_USED_FILES = YES - -# If the sources in your project are distributed over multiple directories -# then setting the SHOW_DIRECTORIES tag to YES will show the directory hierarchy -# in the documentation. The default is NO. - -SHOW_DIRECTORIES = NO - -# Set the SHOW_FILES tag to NO to disable the generation of the Files page. -# This will remove the Files entry from the Quick Index and from the -# Folder Tree View (if specified). The default is YES. - -SHOW_FILES = YES - -# Set the SHOW_NAMESPACES tag to NO to disable the generation of the -# Namespaces page. -# This will remove the Namespaces entry from the Quick Index -# and from the Folder Tree View (if specified). The default is YES. - -SHOW_NAMESPACES = YES - -# The FILE_VERSION_FILTER tag can be used to specify a program or script that -# doxygen should invoke to get the current version for each file (typically from -# the version control system). Doxygen will invoke the program by executing (via -# popen()) the command <command> <input-file>, where <command> is the value of -# the FILE_VERSION_FILTER tag, and <input-file> is the name of an input file -# provided by doxygen. Whatever the program writes to standard output -# is used as the file version. See the manual for examples. - -FILE_VERSION_FILTER = - -# The LAYOUT_FILE tag can be used to specify a layout file which will be parsed by -# doxygen. The layout file controls the global structure of the generated output files -# in an output format independent way. The create the layout file that represents -# doxygen's defaults, run doxygen with the -l option. You can optionally specify a -# file name after the option, if omitted DoxygenLayout.xml will be used as the name -# of the layout file. - -LAYOUT_FILE = - -#--------------------------------------------------------------------------- -# configuration options related to warning and progress messages -#--------------------------------------------------------------------------- - -# The QUIET tag can be used to turn on/off the messages that are generated -# by doxygen. Possible values are YES and NO. If left blank NO is used. - -QUIET = YES - -# The WARNINGS tag can be used to turn on/off the warning messages that are -# generated by doxygen. Possible values are YES and NO. If left blank -# NO is used. - -WARNINGS = YES - -# If WARN_IF_UNDOCUMENTED is set to YES, then doxygen will generate warnings -# for undocumented members. If EXTRACT_ALL is set to YES then this flag will -# automatically be disabled. - -WARN_IF_UNDOCUMENTED = YES - -# If WARN_IF_DOC_ERROR is set to YES, doxygen will generate warnings for -# potential errors in the documentation, such as not documenting some -# parameters in a documented function, or documenting parameters that -# don't exist or using markup commands wrongly. - -WARN_IF_DOC_ERROR = YES - -# This WARN_NO_PARAMDOC option can be abled to get warnings for -# functions that are documented, but have no documentation for their parameters -# or return value. If set to NO (the default) doxygen will only warn about -# wrong or incomplete parameter documentation, but not about the absence of -# documentation. - -WARN_NO_PARAMDOC = YES - -# The WARN_FORMAT tag determines the format of the warning messages that -# doxygen can produce. The string should contain the $file, $line, and $text -# tags, which will be replaced by the file and line number from which the -# warning originated and the warning text. Optionally the format may contain -# $version, which will be replaced by the version of the file (if it could -# be obtained via FILE_VERSION_FILTER) - -WARN_FORMAT = "$file:$line: $text" - -# The WARN_LOGFILE tag can be used to specify a file to which warning -# and error messages should be written. If left blank the output is written -# to stderr. - -WARN_LOGFILE = - -#--------------------------------------------------------------------------- -# configuration options related to the input files -#--------------------------------------------------------------------------- - -# The INPUT tag can be used to specify the files and/or directories that contain -# documented source files. You may enter file names like "myfile.cpp" or -# directories like "/usr/src/myproject". Separate the files or directories -# with spaces. - -INPUT = @top_srcdir@/include/nestegg - -# This tag can be used to specify the character encoding of the source files -# that doxygen parses. Internally doxygen uses the UTF-8 encoding, which is -# also the default input encoding. Doxygen uses libiconv (or the iconv built -# into libc) for the transcoding. See http://www.gnu.org/software/libiconv for -# the list of possible encodings. - -INPUT_ENCODING = UTF-8 - -# If the value of the INPUT tag contains directories, you can use the -# FILE_PATTERNS tag to specify one or more wildcard pattern (like *.cpp -# and *.h) to filter out the source-files in the directories. If left -# blank the following patterns are tested: -# *.c *.cc *.cxx *.cpp *.c++ *.java *.ii *.ixx *.ipp *.i++ *.inl *.h *.hh *.hxx -# *.hpp *.h++ *.idl *.odl *.cs *.php *.php3 *.inc *.m *.mm *.py *.f90 - -FILE_PATTERNS = - -# The RECURSIVE tag can be used to turn specify whether or not subdirectories -# should be searched for input files as well. Possible values are YES and NO. -# If left blank NO is used. - -RECURSIVE = NO - -# The EXCLUDE tag can be used to specify files and/or directories that should -# excluded from the INPUT source files. This way you can easily exclude a -# subdirectory from a directory tree whose root is specified with the INPUT tag. - -EXCLUDE = - -# The EXCLUDE_SYMLINKS tag can be used select whether or not files or -# directories that are symbolic links (a Unix filesystem feature) are excluded -# from the input. - -EXCLUDE_SYMLINKS = NO - -# If the value of the INPUT tag contains directories, you can use the -# EXCLUDE_PATTERNS tag to specify one or more wildcard patterns to exclude -# certain files from those directories. Note that the wildcards are matched -# against the file with absolute path, so to exclude all test directories -# for example use the pattern */test/* - -EXCLUDE_PATTERNS = - -# The EXCLUDE_SYMBOLS tag can be used to specify one or more symbol names -# (namespaces, classes, functions, etc.) that should be excluded from the -# output. The symbol name can be a fully qualified name, a word, or if the -# wildcard * is used, a substring. Examples: ANamespace, AClass, -# AClass::ANamespace, ANamespace::*Test - -EXCLUDE_SYMBOLS = - -# The EXAMPLE_PATH tag can be used to specify one or more files or -# directories that contain example code fragments that are included (see -# the \include command). - -EXAMPLE_PATH = - -# If the value of the EXAMPLE_PATH tag contains directories, you can use the -# EXAMPLE_PATTERNS tag to specify one or more wildcard pattern (like *.cpp -# and *.h) to filter out the source-files in the directories. If left -# blank all files are included. - -EXAMPLE_PATTERNS = - -# If the EXAMPLE_RECURSIVE tag is set to YES then subdirectories will be -# searched for input files to be used with the \include or \dontinclude -# commands irrespective of the value of the RECURSIVE tag. -# Possible values are YES and NO. If left blank NO is used. - -EXAMPLE_RECURSIVE = NO - -# The IMAGE_PATH tag can be used to specify one or more files or -# directories that contain image that are included in the documentation (see -# the \image command). - -IMAGE_PATH = - -# The INPUT_FILTER tag can be used to specify a program that doxygen should -# invoke to filter for each input file. Doxygen will invoke the filter program -# by executing (via popen()) the command <filter> <input-file>, where <filter> -# is the value of the INPUT_FILTER tag, and <input-file> is the name of an -# input file. Doxygen will then use the output that the filter program writes -# to standard output. -# If FILTER_PATTERNS is specified, this tag will be -# ignored. - -INPUT_FILTER = - -# The FILTER_PATTERNS tag can be used to specify filters on a per file pattern -# basis. -# Doxygen will compare the file name with each pattern and apply the -# filter if there is a match. -# The filters are a list of the form: -# pattern=filter (like *.cpp=my_cpp_filter). See INPUT_FILTER for further -# info on how filters are used. If FILTER_PATTERNS is empty, INPUT_FILTER -# is applied to all files. - -FILTER_PATTERNS = - -# If the FILTER_SOURCE_FILES tag is set to YES, the input filter (if set using -# INPUT_FILTER) will be used to filter the input files when producing source -# files to browse (i.e. when SOURCE_BROWSER is set to YES). - -FILTER_SOURCE_FILES = NO - -#--------------------------------------------------------------------------- -# configuration options related to source browsing -#--------------------------------------------------------------------------- - -# If the SOURCE_BROWSER tag is set to YES then a list of source files will -# be generated. Documented entities will be cross-referenced with these sources. -# Note: To get rid of all source code in the generated output, make sure also -# VERBATIM_HEADERS is set to NO. - -SOURCE_BROWSER = NO - -# Setting the INLINE_SOURCES tag to YES will include the body -# of functions and classes directly in the documentation. - -INLINE_SOURCES = NO - -# Setting the STRIP_CODE_COMMENTS tag to YES (the default) will instruct -# doxygen to hide any special comment blocks from generated source code -# fragments. Normal C and C++ comments will always remain visible. - -STRIP_CODE_COMMENTS = YES - -# If the REFERENCED_BY_RELATION tag is set to YES -# then for each documented function all documented -# functions referencing it will be listed. - -REFERENCED_BY_RELATION = NO - -# If the REFERENCES_RELATION tag is set to YES -# then for each documented function all documented entities -# called/used by that function will be listed. - -REFERENCES_RELATION = NO - -# If the REFERENCES_LINK_SOURCE tag is set to YES (the default) -# and SOURCE_BROWSER tag is set to YES, then the hyperlinks from -# functions in REFERENCES_RELATION and REFERENCED_BY_RELATION lists will -# link to the source code. -# Otherwise they will link to the documentation. - -REFERENCES_LINK_SOURCE = YES - -# If the USE_HTAGS tag is set to YES then the references to source code -# will point to the HTML generated by the htags(1) tool instead of doxygen -# built-in source browser. The htags tool is part of GNU's global source -# tagging system (see http://www.gnu.org/software/global/global.html). You -# will need version 4.8.6 or higher. - -USE_HTAGS = NO - -# If the VERBATIM_HEADERS tag is set to YES (the default) then Doxygen -# will generate a verbatim copy of the header file for each class for -# which an include is specified. Set to NO to disable this. - -VERBATIM_HEADERS = YES - -#--------------------------------------------------------------------------- -# configuration options related to the alphabetical class index -#--------------------------------------------------------------------------- - -# If the ALPHABETICAL_INDEX tag is set to YES, an alphabetical index -# of all compounds will be generated. Enable this if the project -# contains a lot of classes, structs, unions or interfaces. - -ALPHABETICAL_INDEX = NO - -# If the alphabetical index is enabled (see ALPHABETICAL_INDEX) then -# the COLS_IN_ALPHA_INDEX tag can be used to specify the number of columns -# in which this list will be split (can be a number in the range [1..20]) - -COLS_IN_ALPHA_INDEX = 5 - -# In case all classes in a project start with a common prefix, all -# classes will be put under the same header in the alphabetical index. -# The IGNORE_PREFIX tag can be used to specify one or more prefixes that -# should be ignored while generating the index headers. - -IGNORE_PREFIX = - -#--------------------------------------------------------------------------- -# configuration options related to the HTML output -#--------------------------------------------------------------------------- - -# If the GENERATE_HTML tag is set to YES (the default) Doxygen will -# generate HTML output. - -GENERATE_HTML = YES - -# The HTML_OUTPUT tag is used to specify where the HTML docs will be put. -# If a relative path is entered the value of OUTPUT_DIRECTORY will be -# put in front of it. If left blank `html' will be used as the default path. - -HTML_OUTPUT = html - -# The HTML_FILE_EXTENSION tag can be used to specify the file extension for -# each generated HTML page (for example: .htm,.php,.asp). If it is left blank -# doxygen will generate files with .html extension. - -HTML_FILE_EXTENSION = .html - -# The HTML_HEADER tag can be used to specify a personal HTML header for -# each generated HTML page. If it is left blank doxygen will generate a -# standard header. - -HTML_HEADER = - -# The HTML_FOOTER tag can be used to specify a personal HTML footer for -# each generated HTML page. If it is left blank doxygen will generate a -# standard footer. - -HTML_FOOTER = - -# The HTML_STYLESHEET tag can be used to specify a user-defined cascading -# style sheet that is used by each HTML page. It can be used to -# fine-tune the look of the HTML output. If the tag is left blank doxygen -# will generate a default style sheet. Note that doxygen will try to copy -# the style sheet file to the HTML output directory, so don't put your own -# stylesheet in the HTML output directory as well, or it will be erased! - -HTML_STYLESHEET = - -# If the HTML_TIMESTAMP tag is set to YES then the footer of each generated HTML -# page will contain the date and time when the page was generated. Setting -# this to NO can help when comparing the output of multiple runs. - -HTML_TIMESTAMP = NO - -# If the HTML_ALIGN_MEMBERS tag is set to YES, the members of classes, -# files or namespaces will be aligned in HTML using tables. If set to -# NO a bullet list will be used. - -HTML_ALIGN_MEMBERS = YES - -# If the HTML_DYNAMIC_SECTIONS tag is set to YES then the generated HTML -# documentation will contain sections that can be hidden and shown after the -# page has loaded. For this to work a browser that supports -# JavaScript and DHTML is required (for instance Mozilla 1.0+, Firefox -# Netscape 6.0+, Internet explorer 5.0+, Konqueror, or Safari). - -HTML_DYNAMIC_SECTIONS = NO - -# If the GENERATE_DOCSET tag is set to YES, additional index files -# will be generated that can be used as input for Apple's Xcode 3 -# integrated development environment, introduced with OSX 10.5 (Leopard). -# To create a documentation set, doxygen will generate a Makefile in the -# HTML output directory. Running make will produce the docset in that -# directory and running "make install" will install the docset in -# ~/Library/Developer/Shared/Documentation/DocSets so that Xcode will find -# it at startup. -# See http://developer.apple.com/tools/creatingdocsetswithdoxygen.html for more information. - -GENERATE_DOCSET = NO - -# When GENERATE_DOCSET tag is set to YES, this tag determines the name of the -# feed. A documentation feed provides an umbrella under which multiple -# documentation sets from a single provider (such as a company or product suite) -# can be grouped. - -DOCSET_FEEDNAME = "Doxygen generated docs" - -# When GENERATE_DOCSET tag is set to YES, this tag specifies a string that -# should uniquely identify the documentation set bundle. This should be a -# reverse domain-name style string, e.g. com.mycompany.MyDocSet. Doxygen -# will append .docset to the name. - -DOCSET_BUNDLE_ID = org.doxygen.Project - -# If the GENERATE_HTMLHELP tag is set to YES, additional index files -# will be generated that can be used as input for tools like the -# Microsoft HTML help workshop to generate a compiled HTML help file (.chm) -# of the generated HTML documentation. - -GENERATE_HTMLHELP = NO - -# If the GENERATE_HTMLHELP tag is set to YES, the CHM_FILE tag can -# be used to specify the file name of the resulting .chm file. You -# can add a path in front of the file if the result should not be -# written to the html output directory. - -CHM_FILE = - -# If the GENERATE_HTMLHELP tag is set to YES, the HHC_LOCATION tag can -# be used to specify the location (absolute path including file name) of -# the HTML help compiler (hhc.exe). If non-empty doxygen will try to run -# the HTML help compiler on the generated index.hhp. - -HHC_LOCATION = - -# If the GENERATE_HTMLHELP tag is set to YES, the GENERATE_CHI flag -# controls if a separate .chi index file is generated (YES) or that -# it should be included in the master .chm file (NO). - -GENERATE_CHI = NO - -# If the GENERATE_HTMLHELP tag is set to YES, the CHM_INDEX_ENCODING -# is used to encode HtmlHelp index (hhk), content (hhc) and project file -# content. - -CHM_INDEX_ENCODING = - -# If the GENERATE_HTMLHELP tag is set to YES, the BINARY_TOC flag -# controls whether a binary table of contents is generated (YES) or a -# normal table of contents (NO) in the .chm file. - -BINARY_TOC = NO - -# The TOC_EXPAND flag can be set to YES to add extra items for group members -# to the contents of the HTML help documentation and to the tree view. - -TOC_EXPAND = NO - -# If the GENERATE_QHP tag is set to YES and both QHP_NAMESPACE and QHP_VIRTUAL_FOLDER -# are set, an additional index file will be generated that can be used as input for -# Qt's qhelpgenerator to generate a Qt Compressed Help (.qch) of the generated -# HTML documentation. - -GENERATE_QHP = NO - -# If the QHG_LOCATION tag is specified, the QCH_FILE tag can -# be used to specify the file name of the resulting .qch file. -# The path specified is relative to the HTML output folder. - -QCH_FILE = - -# The QHP_NAMESPACE tag specifies the namespace to use when generating -# Qt Help Project output. For more information please see -# http://doc.trolltech.com/qthelpproject.html#namespace - -QHP_NAMESPACE = org.doxygen.Project - -# The QHP_VIRTUAL_FOLDER tag specifies the namespace to use when generating -# Qt Help Project output. For more information please see -# http://doc.trolltech.com/qthelpproject.html#virtual-folders - -QHP_VIRTUAL_FOLDER = doc - -# If QHP_CUST_FILTER_NAME is set, it specifies the name of a custom filter to add. -# For more information please see -# http://doc.trolltech.com/qthelpproject.html#custom-filters - -QHP_CUST_FILTER_NAME = - -# The QHP_CUST_FILT_ATTRS tag specifies the list of the attributes of the custom filter to add.For more information please see -# <a href="http://doc.trolltech.com/qthelpproject.html#custom-filters">Qt Help Project / Custom Filters</a>. - -QHP_CUST_FILTER_ATTRS = - -# The QHP_SECT_FILTER_ATTRS tag specifies the list of the attributes this project's -# filter section matches. -# <a href="http://doc.trolltech.com/qthelpproject.html#filter-attributes">Qt Help Project / Filter Attributes</a>. - -QHP_SECT_FILTER_ATTRS = - -# If the GENERATE_QHP tag is set to YES, the QHG_LOCATION tag can -# be used to specify the location of Qt's qhelpgenerator. -# If non-empty doxygen will try to run qhelpgenerator on the generated -# .qhp file. - -QHG_LOCATION = - -# If the GENERATE_ECLIPSEHELP tag is set to YES, additional index files -# will be generated, which together with the HTML files, form an Eclipse help -# plugin. To install this plugin and make it available under the help contents -# menu in Eclipse, the contents of the directory containing the HTML and XML -# files needs to be copied into the plugins directory of eclipse. The name of -# the directory within the plugins directory should be the same as -# the ECLIPSE_DOC_ID value. After copying Eclipse needs to be restarted before the help appears. - -GENERATE_ECLIPSEHELP = NO - -# A unique identifier for the eclipse help plugin. When installing the plugin -# the directory name containing the HTML and XML files should also have -# this name. - -ECLIPSE_DOC_ID = org.doxygen.Project - -# The DISABLE_INDEX tag can be used to turn on/off the condensed index at -# top of each HTML page. The value NO (the default) enables the index and -# the value YES disables it. - -DISABLE_INDEX = NO - -# This tag can be used to set the number of enum values (range [1..20]) -# that doxygen will group on one line in the generated HTML documentation. - -ENUM_VALUES_PER_LINE = 4 - -# The GENERATE_TREEVIEW tag is used to specify whether a tree-like index -# structure should be generated to display hierarchical information. -# If the tag value is set to YES, a side panel will be generated -# containing a tree-like index structure (just like the one that -# is generated for HTML Help). For this to work a browser that supports -# JavaScript, DHTML, CSS and frames is required (i.e. any modern browser). -# Windows users are probably better off using the HTML help feature. - -GENERATE_TREEVIEW = NO - -# By enabling USE_INLINE_TREES, doxygen will generate the Groups, Directories, -# and Class Hierarchy pages using a tree view instead of an ordered list. - -USE_INLINE_TREES = NO - -# If the treeview is enabled (see GENERATE_TREEVIEW) then this tag can be -# used to set the initial width (in pixels) of the frame in which the tree -# is shown. - -TREEVIEW_WIDTH = 250 - -# Use this tag to change the font size of Latex formulas included -# as images in the HTML documentation. The default is 10. Note that -# when you change the font size after a successful doxygen run you need -# to manually remove any form_*.png images from the HTML output directory -# to force them to be regenerated. - -FORMULA_FONTSIZE = 10 - -# When the SEARCHENGINE tag is enabled doxygen will generate a search box for the HTML output. The underlying search engine uses javascript -# and DHTML and should work on any modern browser. Note that when using HTML help (GENERATE_HTMLHELP), Qt help (GENERATE_QHP), or docsets (GENERATE_DOCSET) there is already a search function so this one should -# typically be disabled. For large projects the javascript based search engine -# can be slow, then enabling SERVER_BASED_SEARCH may provide a better solution. - -SEARCHENGINE = YES - -# When the SERVER_BASED_SEARCH tag is enabled the search engine will be implemented using a PHP enabled web server instead of at the web client using Javascript. Doxygen will generate the search PHP script and index -# file to put on the web server. The advantage of the server based approach is that it scales better to large projects and allows full text search. The disadvances is that it is more difficult to setup -# and does not have live searching capabilities. - -SERVER_BASED_SEARCH = NO - -#--------------------------------------------------------------------------- -# configuration options related to the LaTeX output -#--------------------------------------------------------------------------- - -# If the GENERATE_LATEX tag is set to YES (the default) Doxygen will -# generate Latex output. - -GENERATE_LATEX = NO - -# The LATEX_OUTPUT tag is used to specify where the LaTeX docs will be put. -# If a relative path is entered the value of OUTPUT_DIRECTORY will be -# put in front of it. If left blank `latex' will be used as the default path. - -LATEX_OUTPUT = latex - -# The LATEX_CMD_NAME tag can be used to specify the LaTeX command name to be -# invoked. If left blank `latex' will be used as the default command name. -# Note that when enabling USE_PDFLATEX this option is only used for -# generating bitmaps for formulas in the HTML output, but not in the -# Makefile that is written to the output directory. - -LATEX_CMD_NAME = latex - -# The MAKEINDEX_CMD_NAME tag can be used to specify the command name to -# generate index for LaTeX. If left blank `makeindex' will be used as the -# default command name. - -MAKEINDEX_CMD_NAME = makeindex - -# If the COMPACT_LATEX tag is set to YES Doxygen generates more compact -# LaTeX documents. This may be useful for small projects and may help to -# save some trees in general. - -COMPACT_LATEX = NO - -# The PAPER_TYPE tag can be used to set the paper type that is used -# by the printer. Possible values are: a4, a4wide, letter, legal and -# executive. If left blank a4wide will be used. - -PAPER_TYPE = a4wide - -# The EXTRA_PACKAGES tag can be to specify one or more names of LaTeX -# packages that should be included in the LaTeX output. - -EXTRA_PACKAGES = - -# The LATEX_HEADER tag can be used to specify a personal LaTeX header for -# the generated latex document. The header should contain everything until -# the first chapter. If it is left blank doxygen will generate a -# standard header. Notice: only use this tag if you know what you are doing! - -LATEX_HEADER = - -# If the PDF_HYPERLINKS tag is set to YES, the LaTeX that is generated -# is prepared for conversion to pdf (using ps2pdf). The pdf file will -# contain links (just like the HTML output) instead of page references -# This makes the output suitable for online browsing using a pdf viewer. - -PDF_HYPERLINKS = YES - -# If the USE_PDFLATEX tag is set to YES, pdflatex will be used instead of -# plain latex in the generated Makefile. Set this option to YES to get a -# higher quality PDF documentation. - -USE_PDFLATEX = YES - -# If the LATEX_BATCHMODE tag is set to YES, doxygen will add the \\batchmode. -# command to the generated LaTeX files. This will instruct LaTeX to keep -# running if errors occur, instead of asking the user for help. -# This option is also used when generating formulas in HTML. - -LATEX_BATCHMODE = NO - -# If LATEX_HIDE_INDICES is set to YES then doxygen will not -# include the index chapters (such as File Index, Compound Index, etc.) -# in the output. - -LATEX_HIDE_INDICES = NO - -# If LATEX_SOURCE_CODE is set to YES then doxygen will include source code with syntax highlighting in the LaTeX output. Note that which sources are shown also depends on other settings such as SOURCE_BROWSER. - -LATEX_SOURCE_CODE = NO - -#--------------------------------------------------------------------------- -# configuration options related to the RTF output -#--------------------------------------------------------------------------- - -# If the GENERATE_RTF tag is set to YES Doxygen will generate RTF output -# The RTF output is optimized for Word 97 and may not look very pretty with -# other RTF readers or editors. - -GENERATE_RTF = NO - -# The RTF_OUTPUT tag is used to specify where the RTF docs will be put. -# If a relative path is entered the value of OUTPUT_DIRECTORY will be -# put in front of it. If left blank `rtf' will be used as the default path. - -RTF_OUTPUT = rtf - -# If the COMPACT_RTF tag is set to YES Doxygen generates more compact -# RTF documents. This may be useful for small projects and may help to -# save some trees in general. - -COMPACT_RTF = NO - -# If the RTF_HYPERLINKS tag is set to YES, the RTF that is generated -# will contain hyperlink fields. The RTF file will -# contain links (just like the HTML output) instead of page references. -# This makes the output suitable for online browsing using WORD or other -# programs which support those fields. -# Note: wordpad (write) and others do not support links. - -RTF_HYPERLINKS = NO - -# Load stylesheet definitions from file. Syntax is similar to doxygen's -# config file, i.e. a series of assignments. You only have to provide -# replacements, missing definitions are set to their default value. - -RTF_STYLESHEET_FILE = - -# Set optional variables used in the generation of an rtf document. -# Syntax is similar to doxygen's config file. - -RTF_EXTENSIONS_FILE = - -#--------------------------------------------------------------------------- -# configuration options related to the man page output -#--------------------------------------------------------------------------- - -# If the GENERATE_MAN tag is set to YES (the default) Doxygen will -# generate man pages - -GENERATE_MAN = NO - -# The MAN_OUTPUT tag is used to specify where the man pages will be put. -# If a relative path is entered the value of OUTPUT_DIRECTORY will be -# put in front of it. If left blank `man' will be used as the default path. - -MAN_OUTPUT = man - -# The MAN_EXTENSION tag determines the extension that is added to -# the generated man pages (default is the subroutine's section .3) - -MAN_EXTENSION = .3 - -# If the MAN_LINKS tag is set to YES and Doxygen generates man output, -# then it will generate one additional man file for each entity -# documented in the real man page(s). These additional files -# only source the real man page, but without them the man command -# would be unable to find the correct page. The default is NO. - -MAN_LINKS = NO - -#--------------------------------------------------------------------------- -# configuration options related to the XML output -#--------------------------------------------------------------------------- - -# If the GENERATE_XML tag is set to YES Doxygen will -# generate an XML file that captures the structure of -# the code including all documentation. - -GENERATE_XML = NO - -# The XML_OUTPUT tag is used to specify where the XML pages will be put. -# If a relative path is entered the value of OUTPUT_DIRECTORY will be -# put in front of it. If left blank `xml' will be used as the default path. - -XML_OUTPUT = xml - -# The XML_SCHEMA tag can be used to specify an XML schema, -# which can be used by a validating XML parser to check the -# syntax of the XML files. - -XML_SCHEMA = - -# The XML_DTD tag can be used to specify an XML DTD, -# which can be used by a validating XML parser to check the -# syntax of the XML files. - -XML_DTD = - -# If the XML_PROGRAMLISTING tag is set to YES Doxygen will -# dump the program listings (including syntax highlighting -# and cross-referencing information) to the XML output. Note that -# enabling this will significantly increase the size of the XML output. - -XML_PROGRAMLISTING = YES - -#--------------------------------------------------------------------------- -# configuration options for the AutoGen Definitions output -#--------------------------------------------------------------------------- - -# If the GENERATE_AUTOGEN_DEF tag is set to YES Doxygen will -# generate an AutoGen Definitions (see autogen.sf.net) file -# that captures the structure of the code including all -# documentation. Note that this feature is still experimental -# and incomplete at the moment. - -GENERATE_AUTOGEN_DEF = NO - -#--------------------------------------------------------------------------- -# configuration options related to the Perl module output -#--------------------------------------------------------------------------- - -# If the GENERATE_PERLMOD tag is set to YES Doxygen will -# generate a Perl module file that captures the structure of -# the code including all documentation. Note that this -# feature is still experimental and incomplete at the -# moment. - -GENERATE_PERLMOD = NO - -# If the PERLMOD_LATEX tag is set to YES Doxygen will generate -# the necessary Makefile rules, Perl scripts and LaTeX code to be able -# to generate PDF and DVI output from the Perl module output. - -PERLMOD_LATEX = NO - -# If the PERLMOD_PRETTY tag is set to YES the Perl module output will be -# nicely formatted so it can be parsed by a human reader. -# This is useful -# if you want to understand what is going on. -# On the other hand, if this -# tag is set to NO the size of the Perl module output will be much smaller -# and Perl will parse it just the same. - -PERLMOD_PRETTY = YES - -# The names of the make variables in the generated doxyrules.make file -# are prefixed with the string contained in PERLMOD_MAKEVAR_PREFIX. -# This is useful so different doxyrules.make files included by the same -# Makefile don't overwrite each other's variables. - -PERLMOD_MAKEVAR_PREFIX = - -#--------------------------------------------------------------------------- -# Configuration options related to the preprocessor -#--------------------------------------------------------------------------- - -# If the ENABLE_PREPROCESSING tag is set to YES (the default) Doxygen will -# evaluate all C-preprocessor directives found in the sources and include -# files. - -ENABLE_PREPROCESSING = YES - -# If the MACRO_EXPANSION tag is set to YES Doxygen will expand all macro -# names in the source code. If set to NO (the default) only conditional -# compilation will be performed. Macro expansion can be done in a controlled -# way by setting EXPAND_ONLY_PREDEF to YES. - -MACRO_EXPANSION = NO - -# If the EXPAND_ONLY_PREDEF and MACRO_EXPANSION tags are both set to YES -# then the macro expansion is limited to the macros specified with the -# PREDEFINED and EXPAND_AS_DEFINED tags. - -EXPAND_ONLY_PREDEF = NO - -# If the SEARCH_INCLUDES tag is set to YES (the default) the includes files -# in the INCLUDE_PATH (see below) will be search if a #include is found. - -SEARCH_INCLUDES = YES - -# The INCLUDE_PATH tag can be used to specify one or more directories that -# contain include files that are not input files but should be processed by -# the preprocessor. - -INCLUDE_PATH = - -# You can use the INCLUDE_FILE_PATTERNS tag to specify one or more wildcard -# patterns (like *.h and *.hpp) to filter out the header-files in the -# directories. If left blank, the patterns specified with FILE_PATTERNS will -# be used. - -INCLUDE_FILE_PATTERNS = - -# The PREDEFINED tag can be used to specify one or more macro names that -# are defined before the preprocessor is started (similar to the -D option of -# gcc). The argument of the tag is a list of macros of the form: name -# or name=definition (no spaces). If the definition and the = are -# omitted =1 is assumed. To prevent a macro definition from being -# undefined via #undef or recursively expanded use the := operator -# instead of the = operator. - -PREDEFINED = - -# If the MACRO_EXPANSION and EXPAND_ONLY_PREDEF tags are set to YES then -# this tag can be used to specify a list of macro names that should be expanded. -# The macro definition that is found in the sources will be used. -# Use the PREDEFINED tag if you want to use a different macro definition. - -EXPAND_AS_DEFINED = - -# If the SKIP_FUNCTION_MACROS tag is set to YES (the default) then -# doxygen's preprocessor will remove all function-like macros that are alone -# on a line, have an all uppercase name, and do not end with a semicolon. Such -# function macros are typically used for boiler-plate code, and will confuse -# the parser if not removed. - -SKIP_FUNCTION_MACROS = YES - -#--------------------------------------------------------------------------- -# Configuration::additions related to external references -#--------------------------------------------------------------------------- - -# The TAGFILES option can be used to specify one or more tagfiles. -# Optionally an initial location of the external documentation -# can be added for each tagfile. The format of a tag file without -# this location is as follows: -# -# TAGFILES = file1 file2 ... -# Adding location for the tag files is done as follows: -# -# TAGFILES = file1=loc1 "file2 = loc2" ... -# where "loc1" and "loc2" can be relative or absolute paths or -# URLs. If a location is present for each tag, the installdox tool -# does not have to be run to correct the links. -# Note that each tag file must have a unique name -# (where the name does NOT include the path) -# If a tag file is not located in the directory in which doxygen -# is run, you must also specify the path to the tagfile here. - -TAGFILES = - -# When a file name is specified after GENERATE_TAGFILE, doxygen will create -# a tag file that is based on the input files it reads. - -GENERATE_TAGFILE = - -# If the ALLEXTERNALS tag is set to YES all external classes will be listed -# in the class index. If set to NO only the inherited external classes -# will be listed. - -ALLEXTERNALS = NO - -# If the EXTERNAL_GROUPS tag is set to YES all external groups will be listed -# in the modules index. If set to NO, only the current project's groups will -# be listed. - -EXTERNAL_GROUPS = YES - -# The PERL_PATH should be the absolute path and name of the perl script -# interpreter (i.e. the result of `which perl'). - -PERL_PATH = /usr/bin/perl - -#--------------------------------------------------------------------------- -# Configuration options related to the dot tool -#--------------------------------------------------------------------------- - -# If the CLASS_DIAGRAMS tag is set to YES (the default) Doxygen will -# generate a inheritance diagram (in HTML, RTF and LaTeX) for classes with base -# or super classes. Setting the tag to NO turns the diagrams off. Note that -# this option is superseded by the HAVE_DOT option below. This is only a -# fallback. It is recommended to install and use dot, since it yields more -# powerful graphs. - -CLASS_DIAGRAMS = YES - -# You can define message sequence charts within doxygen comments using the \msc -# command. Doxygen will then run the mscgen tool (see -# http://www.mcternan.me.uk/mscgen/) to produce the chart and insert it in the -# documentation. The MSCGEN_PATH tag allows you to specify the directory where -# the mscgen tool resides. If left empty the tool is assumed to be found in the -# default search path. - -MSCGEN_PATH = - -# If set to YES, the inheritance and collaboration graphs will hide -# inheritance and usage relations if the target is undocumented -# or is not a class. - -HIDE_UNDOC_RELATIONS = YES - -# If you set the HAVE_DOT tag to YES then doxygen will assume the dot tool is -# available from the path. This tool is part of Graphviz, a graph visualization -# toolkit from AT&T and Lucent Bell Labs. The other options in this section -# have no effect if this option is set to NO (the default) - -HAVE_DOT = NO - -# By default doxygen will write a font called FreeSans.ttf to the output -# directory and reference it in all dot files that doxygen generates. This -# font does not include all possible unicode characters however, so when you need -# these (or just want a differently looking font) you can specify the font name -# using DOT_FONTNAME. You need need to make sure dot is able to find the font, -# which can be done by putting it in a standard location or by setting the -# DOTFONTPATH environment variable or by setting DOT_FONTPATH to the directory -# containing the font. - -DOT_FONTNAME = FreeSans - -# The DOT_FONTSIZE tag can be used to set the size of the font of dot graphs. -# The default size is 10pt. - -DOT_FONTSIZE = 10 - -# By default doxygen will tell dot to use the output directory to look for the -# FreeSans.ttf font (which doxygen will put there itself). If you specify a -# different font using DOT_FONTNAME you can set the path where dot -# can find it using this tag. - -DOT_FONTPATH = - -# If the CLASS_GRAPH and HAVE_DOT tags are set to YES then doxygen -# will generate a graph for each documented class showing the direct and -# indirect inheritance relations. Setting this tag to YES will force the -# the CLASS_DIAGRAMS tag to NO. - -CLASS_GRAPH = YES - -# If the COLLABORATION_GRAPH and HAVE_DOT tags are set to YES then doxygen -# will generate a graph for each documented class showing the direct and -# indirect implementation dependencies (inheritance, containment, and -# class references variables) of the class with other documented classes. - -COLLABORATION_GRAPH = YES - -# If the GROUP_GRAPHS and HAVE_DOT tags are set to YES then doxygen -# will generate a graph for groups, showing the direct groups dependencies - -GROUP_GRAPHS = YES - -# If the UML_LOOK tag is set to YES doxygen will generate inheritance and -# collaboration diagrams in a style similar to the OMG's Unified Modeling -# Language. - -UML_LOOK = NO - -# If set to YES, the inheritance and collaboration graphs will show the -# relations between templates and their instances. - -TEMPLATE_RELATIONS = NO - -# If the ENABLE_PREPROCESSING, SEARCH_INCLUDES, INCLUDE_GRAPH, and HAVE_DOT -# tags are set to YES then doxygen will generate a graph for each documented -# file showing the direct and indirect include dependencies of the file with -# other documented files. - -INCLUDE_GRAPH = YES - -# If the ENABLE_PREPROCESSING, SEARCH_INCLUDES, INCLUDED_BY_GRAPH, and -# HAVE_DOT tags are set to YES then doxygen will generate a graph for each -# documented header file showing the documented files that directly or -# indirectly include this file. - -INCLUDED_BY_GRAPH = YES - -# If the CALL_GRAPH and HAVE_DOT options are set to YES then -# doxygen will generate a call dependency graph for every global function -# or class method. Note that enabling this option will significantly increase -# the time of a run. So in most cases it will be better to enable call graphs -# for selected functions only using the \callgraph command. - -CALL_GRAPH = NO - -# If the CALLER_GRAPH and HAVE_DOT tags are set to YES then -# doxygen will generate a caller dependency graph for every global function -# or class method. Note that enabling this option will significantly increase -# the time of a run. So in most cases it will be better to enable caller -# graphs for selected functions only using the \callergraph command. - -CALLER_GRAPH = NO - -# If the GRAPHICAL_HIERARCHY and HAVE_DOT tags are set to YES then doxygen -# will graphical hierarchy of all classes instead of a textual one. - -GRAPHICAL_HIERARCHY = YES - -# If the DIRECTORY_GRAPH, SHOW_DIRECTORIES and HAVE_DOT tags are set to YES -# then doxygen will show the dependencies a directory has on other directories -# in a graphical way. The dependency relations are determined by the #include -# relations between the files in the directories. - -DIRECTORY_GRAPH = YES - -# The DOT_IMAGE_FORMAT tag can be used to set the image format of the images -# generated by dot. Possible values are png, jpg, or gif -# If left blank png will be used. - -DOT_IMAGE_FORMAT = png - -# The tag DOT_PATH can be used to specify the path where the dot tool can be -# found. If left blank, it is assumed the dot tool can be found in the path. - -DOT_PATH = - -# The DOTFILE_DIRS tag can be used to specify one or more directories that -# contain dot files that are included in the documentation (see the -# \dotfile command). - -DOTFILE_DIRS = - -# The DOT_GRAPH_MAX_NODES tag can be used to set the maximum number of -# nodes that will be shown in the graph. If the number of nodes in a graph -# becomes larger than this value, doxygen will truncate the graph, which is -# visualized by representing a node as a red box. Note that doxygen if the -# number of direct children of the root node in a graph is already larger than -# DOT_GRAPH_MAX_NODES then the graph will not be shown at all. Also note -# that the size of a graph can be further restricted by MAX_DOT_GRAPH_DEPTH. - -DOT_GRAPH_MAX_NODES = 50 - -# The MAX_DOT_GRAPH_DEPTH tag can be used to set the maximum depth of the -# graphs generated by dot. A depth value of 3 means that only nodes reachable -# from the root by following a path via at most 3 edges will be shown. Nodes -# that lay further from the root node will be omitted. Note that setting this -# option to 1 or 2 may greatly reduce the computation time needed for large -# code bases. Also note that the size of a graph can be further restricted by -# DOT_GRAPH_MAX_NODES. Using a depth of 0 means no depth restriction. - -MAX_DOT_GRAPH_DEPTH = 0 - -# Set the DOT_TRANSPARENT tag to YES to generate images with a transparent -# background. This is disabled by default, because dot on Windows does not -# seem to support this out of the box. Warning: Depending on the platform used, -# enabling this option may lead to badly anti-aliased labels on the edges of -# a graph (i.e. they become hard to read). - -DOT_TRANSPARENT = NO - -# Set the DOT_MULTI_TARGETS tag to YES allow dot to generate multiple output -# files in one run (i.e. multiple -o and -T options on the command line). This -# makes dot run faster, but since only newer versions of dot (>1.8.10) -# support this, this feature is disabled by default. - -DOT_MULTI_TARGETS = NO - -# If the GENERATE_LEGEND tag is set to YES (the default) Doxygen will -# generate a legend page explaining the meaning of the various boxes and -# arrows in the dot generated graphs. - -GENERATE_LEGEND = YES - -# If the DOT_CLEANUP tag is set to YES (the default) Doxygen will -# remove the intermediate dot files that are used to generate -# the various graphs. - -DOT_CLEANUP = YES diff --git a/nestegg/docs/Makefile.am b/nestegg/docs/Makefile.am deleted file mode 100644 index 42cf8eec4..000000000 --- a/nestegg/docs/Makefile.am +++ /dev/null @@ -1,38 +0,0 @@ -doc_DATA = doxygen-build.stamp - -EXTRA_DIST = Doxyfile.in - -if HAVE_DOXYGEN -doxygen-build.stamp: Doxyfile - doxygen - touch doxygen-build.stamp -else -doxygen-build.stamp: - echo "*** Warning: Doxygen not found; documentation will not be built." - touch doxygen-build.stamp -endif - -dist_docdir = $(distdir)/libnestegg - -dist-hook: - if test -d html; then \ - mkdir $(dist_docdir); \ - echo -n "copying built documenation..."; \ - cp -rp html $(dist_docdir)/html; \ - echo "OK"; \ - fi - - -install-data-local: doxygen-build.stamp - $(mkinstalldirs) $(DESTDIR)$(docdir) - if test -d html; then \ - cp -rp html $(DESTDIR)$(docdir)/html; \ - fi - -uninstall-local: - rm -rf $(DESTDIR)$(docdir) - -clean-local: - if test -d html; then rm -rf html; fi - if test -f doxygen-build.stamp; then rm -f doxygen-build.stamp; fi - diff --git a/nestegg/m4/as-ac-expand.m4 b/nestegg/m4/as-ac-expand.m4 deleted file mode 100644 index d6c9e3306..000000000 --- a/nestegg/m4/as-ac-expand.m4 +++ /dev/null @@ -1,43 +0,0 @@ -dnl as-ac-expand.m4 0.2.0 -dnl autostars m4 macro for expanding directories using configure's prefix -dnl thomas@apestaart.org - -dnl AS_AC_EXPAND(VAR, CONFIGURE_VAR) -dnl example -dnl AS_AC_EXPAND(SYSCONFDIR, $sysconfdir) -dnl will set SYSCONFDIR to /usr/local/etc if prefix=/usr/local - -AC_DEFUN([AS_AC_EXPAND], -[ - EXP_VAR=[$1] - FROM_VAR=[$2] - - dnl first expand prefix and exec_prefix if necessary - prefix_save=$prefix - exec_prefix_save=$exec_prefix - - dnl if no prefix given, then use /usr/local, the default prefix - if test "x$prefix" = "xNONE"; then - prefix="$ac_default_prefix" - fi - dnl if no exec_prefix given, then use prefix - if test "x$exec_prefix" = "xNONE"; then - exec_prefix=$prefix - fi - - full_var="$FROM_VAR" - dnl loop until it doesn't change anymore - while true; do - new_full_var="`eval echo $full_var`" - if test "x$new_full_var" = "x$full_var"; then break; fi - full_var=$new_full_var - done - - dnl clean up - full_var=$new_full_var - AC_SUBST([$1], "$full_var") - - dnl restore prefix and exec_prefix - prefix=$prefix_save - exec_prefix=$exec_prefix_save -]) diff --git a/nestegg/m4/ax_create_stdint_h.m4 b/nestegg/m4/ax_create_stdint_h.m4 deleted file mode 100644 index 228105b11..000000000 --- a/nestegg/m4/ax_create_stdint_h.m4 +++ /dev/null @@ -1,695 +0,0 @@ -dnl @synopsis AX_CREATE_STDINT_H [( HEADER-TO-GENERATE [, HEDERS-TO-CHECK])] -dnl -dnl the "ISO C9X: 7.18 Integer types <stdint.h>" section requires the -dnl existence of an include file <stdint.h> that defines a set of -dnl typedefs, especially uint8_t,int32_t,uintptr_t. Many older -dnl installations will not provide this file, but some will have the -dnl very same definitions in <inttypes.h>. In other enviroments we can -dnl use the inet-types in <sys/types.h> which would define the typedefs -dnl int8_t and u_int8_t respectivly. -dnl -dnl This macros will create a local "_stdint.h" or the headerfile given -dnl as an argument. In many cases that file will just "#include -dnl <stdint.h>" or "#include <inttypes.h>", while in other environments -dnl it will provide the set of basic 'stdint's definitions/typedefs: -dnl -dnl int8_t,uint8_t,int16_t,uint16_t,int32_t,uint32_t,intptr_t,uintptr_t -dnl int_least32_t.. int_fast32_t.. intmax_t -dnl -dnl which may or may not rely on the definitions of other files, or -dnl using the AC_CHECK_SIZEOF macro to determine the actual sizeof each -dnl type. -dnl -dnl if your header files require the stdint-types you will want to -dnl create an installable file mylib-int.h that all your other -dnl installable header may include. So if you have a library package -dnl named "mylib", just use -dnl -dnl AX_CREATE_STDINT_H(mylib-int.h) -dnl -dnl in configure.ac and go to install that very header file in -dnl Makefile.am along with the other headers (mylib.h) - and the -dnl mylib-specific headers can simply use "#include <mylib-int.h>" to -dnl obtain the stdint-types. -dnl -dnl Remember, if the system already had a valid <stdint.h>, the -dnl generated file will include it directly. No need for fuzzy -dnl HAVE_STDINT_H things... (oops, GCC 4.2.x has deliberatly disabled -dnl its stdint.h for non-c99 compilation and the c99-mode is not the -dnl default. Therefore this macro will not use the compiler's stdint.h -dnl - please complain to the GCC developers). -dnl -dnl @category C -dnl @author Guido U. Draheim <guidod@gmx.de> -dnl @version 2006-10-13 -dnl @license GPLWithACException - -AC_DEFUN([AX_CHECK_DATA_MODEL],[ - AC_CHECK_SIZEOF(char) - AC_CHECK_SIZEOF(short) - AC_CHECK_SIZEOF(int) - AC_CHECK_SIZEOF(long) - AC_CHECK_SIZEOF(void*) - ac_cv_char_data_model="" - ac_cv_char_data_model="$ac_cv_char_data_model$ac_cv_sizeof_char" - ac_cv_char_data_model="$ac_cv_char_data_model$ac_cv_sizeof_short" - ac_cv_char_data_model="$ac_cv_char_data_model$ac_cv_sizeof_int" - ac_cv_long_data_model="" - ac_cv_long_data_model="$ac_cv_long_data_model$ac_cv_sizeof_int" - ac_cv_long_data_model="$ac_cv_long_data_model$ac_cv_sizeof_long" - ac_cv_long_data_model="$ac_cv_long_data_model$ac_cv_sizeof_voidp" - AC_MSG_CHECKING([data model]) - case "$ac_cv_char_data_model/$ac_cv_long_data_model" in - 122/242) ac_cv_data_model="IP16" ; n="standard 16bit machine" ;; - 122/244) ac_cv_data_model="LP32" ; n="standard 32bit machine" ;; - 122/*) ac_cv_data_model="i16" ; n="unusual int16 model" ;; - 124/444) ac_cv_data_model="ILP32" ; n="standard 32bit unixish" ;; - 124/488) ac_cv_data_model="LP64" ; n="standard 64bit unixish" ;; - 124/448) ac_cv_data_model="LLP64" ; n="unusual 64bit unixish" ;; - 124/*) ac_cv_data_model="i32" ; n="unusual int32 model" ;; - 128/888) ac_cv_data_model="ILP64" ; n="unusual 64bit numeric" ;; - 128/*) ac_cv_data_model="i64" ; n="unusual int64 model" ;; - 222/*2) ac_cv_data_model="DSP16" ; n="strict 16bit dsptype" ;; - 333/*3) ac_cv_data_model="DSP24" ; n="strict 24bit dsptype" ;; - 444/*4) ac_cv_data_model="DSP32" ; n="strict 32bit dsptype" ;; - 666/*6) ac_cv_data_model="DSP48" ; n="strict 48bit dsptype" ;; - 888/*8) ac_cv_data_model="DSP64" ; n="strict 64bit dsptype" ;; - 222/*|333/*|444/*|666/*|888/*) : - ac_cv_data_model="iDSP" ; n="unusual dsptype" ;; - *) ac_cv_data_model="none" ; n="very unusual model" ;; - esac - AC_MSG_RESULT([$ac_cv_data_model ($ac_cv_long_data_model, $n)]) -]) - -dnl AX_CHECK_HEADER_STDINT_X([HEADERLIST][,ACTION-IF]) -AC_DEFUN([AX_CHECK_HEADER_STDINT_X],[ -AC_CACHE_CHECK([for stdint uintptr_t], [ac_cv_header_stdint_x],[ - ac_cv_header_stdint_x="" # the 1997 typedefs (inttypes.h) - AC_MSG_RESULT([(..)]) - for i in m4_ifval([$1],[$1],[stdint.h inttypes.h sys/inttypes.h sys/types.h]) - do - unset ac_cv_type_uintptr_t - unset ac_cv_type_uint64_t - AC_CHECK_TYPE(uintptr_t,[ac_cv_header_stdint_x=$i],continue,[#include <$i>]) - AC_CHECK_TYPE(uint64_t,[and64="/uint64_t"],[and64=""],[#include<$i>]) - m4_ifvaln([$1],[$1]) break - done - AC_MSG_CHECKING([for stdint uintptr_t]) - ]) -]) - -AC_DEFUN([AX_CHECK_HEADER_STDINT_O],[ -AC_CACHE_CHECK([for stdint uint32_t], [ac_cv_header_stdint_o],[ - ac_cv_header_stdint_o="" # the 1995 typedefs (sys/inttypes.h) - AC_MSG_RESULT([(..)]) - for i in m4_ifval([$1],[$1],[inttypes.h sys/inttypes.h sys/types.h stdint.h]) - do - unset ac_cv_type_uint32_t - unset ac_cv_type_uint64_t - AC_CHECK_TYPE(uint32_t,[ac_cv_header_stdint_o=$i],continue,[#include <$i>]) - AC_CHECK_TYPE(uint64_t,[and64="/uint64_t"],[and64=""],[#include<$i>]) - m4_ifvaln([$1],[$1]) break - break; - done - AC_MSG_CHECKING([for stdint uint32_t]) - ]) -]) - -AC_DEFUN([AX_CHECK_HEADER_STDINT_U],[ -AC_CACHE_CHECK([for stdint u_int32_t], [ac_cv_header_stdint_u],[ - ac_cv_header_stdint_u="" # the BSD typedefs (sys/types.h) - AC_MSG_RESULT([(..)]) - for i in m4_ifval([$1],[$1],[sys/types.h inttypes.h sys/inttypes.h]) ; do - unset ac_cv_type_u_int32_t - unset ac_cv_type_u_int64_t - AC_CHECK_TYPE(u_int32_t,[ac_cv_header_stdint_u=$i],continue,[#include <$i>]) - AC_CHECK_TYPE(u_int64_t,[and64="/u_int64_t"],[and64=""],[#include<$i>]) - m4_ifvaln([$1],[$1]) break - break; - done - AC_MSG_CHECKING([for stdint u_int32_t]) - ]) -]) - -AC_DEFUN([AX_CREATE_STDINT_H], -[# ------ AX CREATE STDINT H ------------------------------------- -AC_MSG_CHECKING([for stdint types]) -ac_stdint_h=`echo ifelse($1, , _stdint.h, $1)` -# try to shortcircuit - if the default include path of the compiler -# can find a "stdint.h" header then we assume that all compilers can. -AC_CACHE_VAL([ac_cv_header_stdint_t],[ -old_CXXFLAGS="$CXXFLAGS" ; CXXFLAGS="" -old_CPPFLAGS="$CPPFLAGS" ; CPPFLAGS="" -old_CFLAGS="$CFLAGS" ; CFLAGS="" -AC_TRY_COMPILE([#include <stdint.h>],[int_least32_t v = 0;], -[ac_cv_stdint_result="(assuming C99 compatible system)" - ac_cv_header_stdint_t="stdint.h"; ], -[ac_cv_header_stdint_t=""]) -if test "$GCC" = "yes" && test ".$ac_cv_header_stdint_t" = "."; then -CFLAGS="-std=c99" -AC_TRY_COMPILE([#include <stdint.h>],[int_least32_t v = 0;], -[AC_MSG_WARN(your GCC compiler has a defunct stdint.h for its default-mode)]) -fi -CXXFLAGS="$old_CXXFLAGS" -CPPFLAGS="$old_CPPFLAGS" -CFLAGS="$old_CFLAGS" ]) - -v="... $ac_cv_header_stdint_h" -if test "$ac_stdint_h" = "stdint.h" ; then - AC_MSG_RESULT([(are you sure you want them in ./stdint.h?)]) -elif test "$ac_stdint_h" = "inttypes.h" ; then - AC_MSG_RESULT([(are you sure you want them in ./inttypes.h?)]) -elif test "_$ac_cv_header_stdint_t" = "_" ; then - AC_MSG_RESULT([(putting them into $ac_stdint_h)$v]) -else - ac_cv_header_stdint="$ac_cv_header_stdint_t" - AC_MSG_RESULT([$ac_cv_header_stdint (shortcircuit)]) -fi - -if test "_$ac_cv_header_stdint_t" = "_" ; then # can not shortcircuit.. - -dnl .....intro message done, now do a few system checks..... -dnl btw, all old CHECK_TYPE macros do automatically "DEFINE" a type, -dnl therefore we use the autoconf implementation detail CHECK_TYPE_NEW -dnl instead that is triggered with 3 or more arguments (see types.m4) - -inttype_headers=`echo $2 | sed -e 's/,/ /g'` - -ac_cv_stdint_result="(no helpful system typedefs seen)" -AX_CHECK_HEADER_STDINT_X(dnl - stdint.h inttypes.h sys/inttypes.h $inttype_headers, - ac_cv_stdint_result="(seen uintptr_t$and64 in $i)") - -if test "_$ac_cv_header_stdint_x" = "_" ; then -AX_CHECK_HEADER_STDINT_O(dnl, - inttypes.h sys/inttypes.h stdint.h $inttype_headers, - ac_cv_stdint_result="(seen uint32_t$and64 in $i)") -fi - -if test "_$ac_cv_header_stdint_x" = "_" ; then -if test "_$ac_cv_header_stdint_o" = "_" ; then -AX_CHECK_HEADER_STDINT_U(dnl, - sys/types.h inttypes.h sys/inttypes.h $inttype_headers, - ac_cv_stdint_result="(seen u_int32_t$and64 in $i)") -fi fi - -dnl if there was no good C99 header file, do some typedef checks... -if test "_$ac_cv_header_stdint_x" = "_" ; then - AC_MSG_CHECKING([for stdint datatype model]) - AC_MSG_RESULT([(..)]) - AX_CHECK_DATA_MODEL -fi - -if test "_$ac_cv_header_stdint_x" != "_" ; then - ac_cv_header_stdint="$ac_cv_header_stdint_x" -elif test "_$ac_cv_header_stdint_o" != "_" ; then - ac_cv_header_stdint="$ac_cv_header_stdint_o" -elif test "_$ac_cv_header_stdint_u" != "_" ; then - ac_cv_header_stdint="$ac_cv_header_stdint_u" -else - ac_cv_header_stdint="stddef.h" -fi - -AC_MSG_CHECKING([for extra inttypes in chosen header]) -AC_MSG_RESULT([($ac_cv_header_stdint)]) -dnl see if int_least and int_fast types are present in _this_ header. -unset ac_cv_type_int_least32_t -unset ac_cv_type_int_fast32_t -AC_CHECK_TYPE(int_least32_t,,,[#include <$ac_cv_header_stdint>]) -AC_CHECK_TYPE(int_fast32_t,,,[#include<$ac_cv_header_stdint>]) -AC_CHECK_TYPE(intmax_t,,,[#include <$ac_cv_header_stdint>]) - -fi # shortcircut to system "stdint.h" -# ------------------ PREPARE VARIABLES ------------------------------ -if test "$GCC" = "yes" ; then -ac_cv_stdint_message="using gnu compiler "`$CC --version | head -1` -else -ac_cv_stdint_message="using $CC" -fi - -AC_MSG_RESULT([make use of $ac_cv_header_stdint in $ac_stdint_h dnl -$ac_cv_stdint_result]) - -dnl ----------------------------------------------------------------- -# ----------------- DONE inttypes.h checks START header ------------- -AC_CONFIG_COMMANDS([$ac_stdint_h],[ -AC_MSG_NOTICE(creating $ac_stdint_h : $_ac_stdint_h) -ac_stdint=$tmp/_stdint.h - -echo "#ifndef" $_ac_stdint_h >$ac_stdint -echo "#define" $_ac_stdint_h "1" >>$ac_stdint -echo "#ifndef" _GENERATED_STDINT_H >>$ac_stdint -echo "#define" _GENERATED_STDINT_H '"'$PACKAGE $VERSION'"' >>$ac_stdint -echo "/* generated $ac_cv_stdint_message */" >>$ac_stdint -if test "_$ac_cv_header_stdint_t" != "_" ; then -echo "#define _STDINT_HAVE_STDINT_H" "1" >>$ac_stdint -echo "#include <stdint.h>" >>$ac_stdint -echo "#endif" >>$ac_stdint -echo "#endif" >>$ac_stdint -else - -cat >>$ac_stdint <<STDINT_EOF - -/* ................... shortcircuit part ........................... */ - -#if defined HAVE_STDINT_H || defined _STDINT_HAVE_STDINT_H -#include <stdint.h> -#else -#include <stddef.h> - -/* .................... configured part ............................ */ - -STDINT_EOF - -echo "/* whether we have a C99 compatible stdint header file */" >>$ac_stdint -if test "_$ac_cv_header_stdint_x" != "_" ; then - ac_header="$ac_cv_header_stdint_x" - echo "#define _STDINT_HEADER_INTPTR" '"'"$ac_header"'"' >>$ac_stdint -else - echo "/* #undef _STDINT_HEADER_INTPTR */" >>$ac_stdint -fi - -echo "/* whether we have a C96 compatible inttypes header file */" >>$ac_stdint -if test "_$ac_cv_header_stdint_o" != "_" ; then - ac_header="$ac_cv_header_stdint_o" - echo "#define _STDINT_HEADER_UINT32" '"'"$ac_header"'"' >>$ac_stdint -else - echo "/* #undef _STDINT_HEADER_UINT32 */" >>$ac_stdint -fi - -echo "/* whether we have a BSD compatible inet types header */" >>$ac_stdint -if test "_$ac_cv_header_stdint_u" != "_" ; then - ac_header="$ac_cv_header_stdint_u" - echo "#define _STDINT_HEADER_U_INT32" '"'"$ac_header"'"' >>$ac_stdint -else - echo "/* #undef _STDINT_HEADER_U_INT32 */" >>$ac_stdint -fi - -echo "" >>$ac_stdint - -if test "_$ac_header" != "_" ; then if test "$ac_header" != "stddef.h" ; then - echo "#include <$ac_header>" >>$ac_stdint - echo "" >>$ac_stdint -fi fi - -echo "/* which 64bit typedef has been found */" >>$ac_stdint -if test "$ac_cv_type_uint64_t" = "yes" ; then -echo "#define _STDINT_HAVE_UINT64_T" "1" >>$ac_stdint -else -echo "/* #undef _STDINT_HAVE_UINT64_T */" >>$ac_stdint -fi -if test "$ac_cv_type_u_int64_t" = "yes" ; then -echo "#define _STDINT_HAVE_U_INT64_T" "1" >>$ac_stdint -else -echo "/* #undef _STDINT_HAVE_U_INT64_T */" >>$ac_stdint -fi -echo "" >>$ac_stdint - -echo "/* which type model has been detected */" >>$ac_stdint -if test "_$ac_cv_char_data_model" != "_" ; then -echo "#define _STDINT_CHAR_MODEL" "$ac_cv_char_data_model" >>$ac_stdint -echo "#define _STDINT_LONG_MODEL" "$ac_cv_long_data_model" >>$ac_stdint -else -echo "/* #undef _STDINT_CHAR_MODEL // skipped */" >>$ac_stdint -echo "/* #undef _STDINT_LONG_MODEL // skipped */" >>$ac_stdint -fi -echo "" >>$ac_stdint - -echo "/* whether int_least types were detected */" >>$ac_stdint -if test "$ac_cv_type_int_least32_t" = "yes"; then -echo "#define _STDINT_HAVE_INT_LEAST32_T" "1" >>$ac_stdint -else -echo "/* #undef _STDINT_HAVE_INT_LEAST32_T */" >>$ac_stdint -fi -echo "/* whether int_fast types were detected */" >>$ac_stdint -if test "$ac_cv_type_int_fast32_t" = "yes"; then -echo "#define _STDINT_HAVE_INT_FAST32_T" "1" >>$ac_stdint -else -echo "/* #undef _STDINT_HAVE_INT_FAST32_T */" >>$ac_stdint -fi -echo "/* whether intmax_t type was detected */" >>$ac_stdint -if test "$ac_cv_type_intmax_t" = "yes"; then -echo "#define _STDINT_HAVE_INTMAX_T" "1" >>$ac_stdint -else -echo "/* #undef _STDINT_HAVE_INTMAX_T */" >>$ac_stdint -fi -echo "" >>$ac_stdint - - cat >>$ac_stdint <<STDINT_EOF -/* .................... detections part ............................ */ - -/* whether we need to define bitspecific types from compiler base types */ -#ifndef _STDINT_HEADER_INTPTR -#ifndef _STDINT_HEADER_UINT32 -#ifndef _STDINT_HEADER_U_INT32 -#define _STDINT_NEED_INT_MODEL_T -#else -#define _STDINT_HAVE_U_INT_TYPES -#endif -#endif -#endif - -#ifdef _STDINT_HAVE_U_INT_TYPES -#undef _STDINT_NEED_INT_MODEL_T -#endif - -#ifdef _STDINT_CHAR_MODEL -#if _STDINT_CHAR_MODEL+0 == 122 || _STDINT_CHAR_MODEL+0 == 124 -#ifndef _STDINT_BYTE_MODEL -#define _STDINT_BYTE_MODEL 12 -#endif -#endif -#endif - -#ifndef _STDINT_HAVE_INT_LEAST32_T -#define _STDINT_NEED_INT_LEAST_T -#endif - -#ifndef _STDINT_HAVE_INT_FAST32_T -#define _STDINT_NEED_INT_FAST_T -#endif - -#ifndef _STDINT_HEADER_INTPTR -#define _STDINT_NEED_INTPTR_T -#ifndef _STDINT_HAVE_INTMAX_T -#define _STDINT_NEED_INTMAX_T -#endif -#endif - - -/* .................... definition part ............................ */ - -/* some system headers have good uint64_t */ -#ifndef _HAVE_UINT64_T -#if defined _STDINT_HAVE_UINT64_T || defined HAVE_UINT64_T -#define _HAVE_UINT64_T -#elif defined _STDINT_HAVE_U_INT64_T || defined HAVE_U_INT64_T -#define _HAVE_UINT64_T -typedef u_int64_t uint64_t; -#endif -#endif - -#ifndef _HAVE_UINT64_T -/* .. here are some common heuristics using compiler runtime specifics */ -#if defined __STDC_VERSION__ && defined __STDC_VERSION__ >= 199901L -#define _HAVE_UINT64_T -#define _HAVE_LONGLONG_UINT64_T -typedef long long int64_t; -typedef unsigned long long uint64_t; - -#elif !defined __STRICT_ANSI__ -#if defined _MSC_VER || defined __WATCOMC__ || defined __BORLANDC__ -#define _HAVE_UINT64_T -typedef __int64 int64_t; -typedef unsigned __int64 uint64_t; - -#elif defined __GNUC__ || defined __MWERKS__ || defined __ELF__ -/* note: all ELF-systems seem to have loff-support which needs 64-bit */ -#if !defined _NO_LONGLONG -#define _HAVE_UINT64_T -#define _HAVE_LONGLONG_UINT64_T -typedef long long int64_t; -typedef unsigned long long uint64_t; -#endif - -#elif defined __alpha || (defined __mips && defined _ABIN32) -#if !defined _NO_LONGLONG -typedef long int64_t; -typedef unsigned long uint64_t; -#endif - /* compiler/cpu type to define int64_t */ -#endif -#endif -#endif - -#if defined _STDINT_HAVE_U_INT_TYPES -/* int8_t int16_t int32_t defined by inet code, redeclare the u_intXX types */ -typedef u_int8_t uint8_t; -typedef u_int16_t uint16_t; -typedef u_int32_t uint32_t; - -/* glibc compatibility */ -#ifndef __int8_t_defined -#define __int8_t_defined -#endif -#endif - -#ifdef _STDINT_NEED_INT_MODEL_T -/* we must guess all the basic types. Apart from byte-adressable system, */ -/* there a few 32-bit-only dsp-systems that we guard with BYTE_MODEL 8-} */ -/* (btw, those nibble-addressable systems are way off, or so we assume) */ - -dnl /* have a look at "64bit and data size neutrality" at */ -dnl /* http://unix.org/version2/whatsnew/login_64bit.html */ -dnl /* (the shorthand "ILP" types always have a "P" part) */ - -#if defined _STDINT_BYTE_MODEL -#if _STDINT_LONG_MODEL+0 == 242 -/* 2:4:2 = IP16 = a normal 16-bit system */ -typedef unsigned char uint8_t; -typedef unsigned short uint16_t; -typedef unsigned long uint32_t; -#ifndef __int8_t_defined -#define __int8_t_defined -typedef char int8_t; -typedef short int16_t; -typedef long int32_t; -#endif -#elif _STDINT_LONG_MODEL+0 == 244 || _STDINT_LONG_MODEL == 444 -/* 2:4:4 = LP32 = a 32-bit system derived from a 16-bit */ -/* 4:4:4 = ILP32 = a normal 32-bit system */ -typedef unsigned char uint8_t; -typedef unsigned short uint16_t; -typedef unsigned int uint32_t; -#ifndef __int8_t_defined -#define __int8_t_defined -typedef char int8_t; -typedef short int16_t; -typedef int int32_t; -#endif -#elif _STDINT_LONG_MODEL+0 == 484 || _STDINT_LONG_MODEL+0 == 488 -/* 4:8:4 = IP32 = a 32-bit system prepared for 64-bit */ -/* 4:8:8 = LP64 = a normal 64-bit system */ -typedef unsigned char uint8_t; -typedef unsigned short uint16_t; -typedef unsigned int uint32_t; -#ifndef __int8_t_defined -#define __int8_t_defined -typedef char int8_t; -typedef short int16_t; -typedef int int32_t; -#endif -/* this system has a "long" of 64bit */ -#ifndef _HAVE_UINT64_T -#define _HAVE_UINT64_T -typedef unsigned long uint64_t; -typedef long int64_t; -#endif -#elif _STDINT_LONG_MODEL+0 == 448 -/* LLP64 a 64-bit system derived from a 32-bit system */ -typedef unsigned char uint8_t; -typedef unsigned short uint16_t; -typedef unsigned int uint32_t; -#ifndef __int8_t_defined -#define __int8_t_defined -typedef char int8_t; -typedef short int16_t; -typedef int int32_t; -#endif -/* assuming the system has a "long long" */ -#ifndef _HAVE_UINT64_T -#define _HAVE_UINT64_T -#define _HAVE_LONGLONG_UINT64_T -typedef unsigned long long uint64_t; -typedef long long int64_t; -#endif -#else -#define _STDINT_NO_INT32_T -#endif -#else -#define _STDINT_NO_INT8_T -#define _STDINT_NO_INT32_T -#endif -#endif - -/* - * quote from SunOS-5.8 sys/inttypes.h: - * Use at your own risk. As of February 1996, the committee is squarely - * behind the fixed sized types; the "least" and "fast" types are still being - * discussed. The probability that the "fast" types may be removed before - * the standard is finalized is high enough that they are not currently - * implemented. - */ - -#if defined _STDINT_NEED_INT_LEAST_T -typedef int8_t int_least8_t; -typedef int16_t int_least16_t; -typedef int32_t int_least32_t; -#ifdef _HAVE_UINT64_T -typedef int64_t int_least64_t; -#endif - -typedef uint8_t uint_least8_t; -typedef uint16_t uint_least16_t; -typedef uint32_t uint_least32_t; -#ifdef _HAVE_UINT64_T -typedef uint64_t uint_least64_t; -#endif - /* least types */ -#endif - -#if defined _STDINT_NEED_INT_FAST_T -typedef int8_t int_fast8_t; -typedef int int_fast16_t; -typedef int32_t int_fast32_t; -#ifdef _HAVE_UINT64_T -typedef int64_t int_fast64_t; -#endif - -typedef uint8_t uint_fast8_t; -typedef unsigned uint_fast16_t; -typedef uint32_t uint_fast32_t; -#ifdef _HAVE_UINT64_T -typedef uint64_t uint_fast64_t; -#endif - /* fast types */ -#endif - -#ifdef _STDINT_NEED_INTMAX_T -#ifdef _HAVE_UINT64_T -typedef int64_t intmax_t; -typedef uint64_t uintmax_t; -#else -typedef long intmax_t; -typedef unsigned long uintmax_t; -#endif -#endif - -#ifdef _STDINT_NEED_INTPTR_T -#ifndef __intptr_t_defined -#define __intptr_t_defined -/* we encourage using "long" to store pointer values, never use "int" ! */ -#if _STDINT_LONG_MODEL+0 == 242 || _STDINT_LONG_MODEL+0 == 484 -typedef unsigned int uintptr_t; -typedef int intptr_t; -#elif _STDINT_LONG_MODEL+0 == 244 || _STDINT_LONG_MODEL+0 == 444 -typedef unsigned long uintptr_t; -typedef long intptr_t; -#elif _STDINT_LONG_MODEL+0 == 448 && defined _HAVE_UINT64_T -typedef uint64_t uintptr_t; -typedef int64_t intptr_t; -#else /* matches typical system types ILP32 and LP64 - but not IP16 or LLP64 */ -typedef unsigned long uintptr_t; -typedef long intptr_t; -#endif -#endif -#endif - -/* The ISO C99 standard specifies that in C++ implementations these - should only be defined if explicitly requested. */ -#if !defined __cplusplus || defined __STDC_CONSTANT_MACROS -#ifndef UINT32_C - -/* Signed. */ -# define INT8_C(c) c -# define INT16_C(c) c -# define INT32_C(c) c -# ifdef _HAVE_LONGLONG_UINT64_T -# define INT64_C(c) c ## L -# else -# define INT64_C(c) c ## LL -# endif - -/* Unsigned. */ -# define UINT8_C(c) c ## U -# define UINT16_C(c) c ## U -# define UINT32_C(c) c ## U -# ifdef _HAVE_LONGLONG_UINT64_T -# define UINT64_C(c) c ## UL -# else -# define UINT64_C(c) c ## ULL -# endif - -/* Maximal type. */ -# ifdef _HAVE_LONGLONG_UINT64_T -# define INTMAX_C(c) c ## L -# define UINTMAX_C(c) c ## UL -# else -# define INTMAX_C(c) c ## LL -# define UINTMAX_C(c) c ## ULL -# endif - - /* literalnumbers */ -#endif -#endif - -/* These limits are merily those of a two complement byte-oriented system */ - -/* Minimum of signed integral types. */ -# define INT8_MIN (-128) -# define INT16_MIN (-32767-1) -# define INT32_MIN (-2147483647-1) -# define INT64_MIN (-__INT64_C(9223372036854775807)-1) -/* Maximum of signed integral types. */ -# define INT8_MAX (127) -# define INT16_MAX (32767) -# define INT32_MAX (2147483647) -# define INT64_MAX (__INT64_C(9223372036854775807)) - -/* Maximum of unsigned integral types. */ -# define UINT8_MAX (255) -# define UINT16_MAX (65535) -# define UINT32_MAX (4294967295U) -# define UINT64_MAX (__UINT64_C(18446744073709551615)) - -/* Minimum of signed integral types having a minimum size. */ -# define INT_LEAST8_MIN INT8_MIN -# define INT_LEAST16_MIN INT16_MIN -# define INT_LEAST32_MIN INT32_MIN -# define INT_LEAST64_MIN INT64_MIN -/* Maximum of signed integral types having a minimum size. */ -# define INT_LEAST8_MAX INT8_MAX -# define INT_LEAST16_MAX INT16_MAX -# define INT_LEAST32_MAX INT32_MAX -# define INT_LEAST64_MAX INT64_MAX - -/* Maximum of unsigned integral types having a minimum size. */ -# define UINT_LEAST8_MAX UINT8_MAX -# define UINT_LEAST16_MAX UINT16_MAX -# define UINT_LEAST32_MAX UINT32_MAX -# define UINT_LEAST64_MAX UINT64_MAX - - /* shortcircuit*/ -#endif - /* once */ -#endif -#endif -STDINT_EOF -fi - if cmp -s $ac_stdint_h $ac_stdint 2>/dev/null; then - AC_MSG_NOTICE([$ac_stdint_h is unchanged]) - else - ac_dir=`AS_DIRNAME(["$ac_stdint_h"])` - AS_MKDIR_P(["$ac_dir"]) - rm -f $ac_stdint_h - mv $ac_stdint $ac_stdint_h - fi -],[# variables for create stdint.h replacement -PACKAGE="$PACKAGE" -VERSION="$VERSION" -ac_stdint_h="$ac_stdint_h" -_ac_stdint_h=AS_TR_CPP(_$PACKAGE-$ac_stdint_h) -ac_cv_stdint_message="$ac_cv_stdint_message" -ac_cv_header_stdint_t="$ac_cv_header_stdint_t" -ac_cv_header_stdint_x="$ac_cv_header_stdint_x" -ac_cv_header_stdint_o="$ac_cv_header_stdint_o" -ac_cv_header_stdint_u="$ac_cv_header_stdint_u" -ac_cv_type_uint64_t="$ac_cv_type_uint64_t" -ac_cv_type_u_int64_t="$ac_cv_type_u_int64_t" -ac_cv_char_data_model="$ac_cv_char_data_model" -ac_cv_long_data_model="$ac_cv_long_data_model" -ac_cv_type_int_least32_t="$ac_cv_type_int_least32_t" -ac_cv_type_int_fast32_t="$ac_cv_type_int_fast32_t" -ac_cv_type_intmax_t="$ac_cv_type_intmax_t" -]) -]) diff --git a/nestegg/m4/pkg.m4 b/nestegg/m4/pkg.m4 deleted file mode 100644 index 996e29454..000000000 --- a/nestegg/m4/pkg.m4 +++ /dev/null @@ -1,157 +0,0 @@ -# pkg.m4 - Macros to locate and utilise pkg-config. -*- Autoconf -*- -# -# Copyright © 2004 Scott James Remnant <scott@netsplit.com>. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2 of the License, or -# (at your option) any later version. -# -# This program is distributed in the hope that it will be useful, but -# WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU -# General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; if not, write to the Free Software -# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -# -# As a special exception to the GNU General Public License, if you -# distribute this file as part of a program that contains a -# configuration script generated by Autoconf, you may include it under -# the same distribution terms that you use for the rest of that program. - -# PKG_PROG_PKG_CONFIG([MIN-VERSION]) -# ---------------------------------- -AC_DEFUN([PKG_PROG_PKG_CONFIG], -[m4_pattern_forbid([^_?PKG_[A-Z_]+$]) -m4_pattern_allow([^PKG_CONFIG(_PATH)?$]) -AC_ARG_VAR([PKG_CONFIG], [path to pkg-config utility])dnl -if test "x$ac_cv_env_PKG_CONFIG_set" != "xset"; then - AC_PATH_TOOL([PKG_CONFIG], [pkg-config]) -fi -if test -n "$PKG_CONFIG"; then - _pkg_min_version=m4_default([$1], [0.9.0]) - AC_MSG_CHECKING([pkg-config is at least version $_pkg_min_version]) - if $PKG_CONFIG --atleast-pkgconfig-version $_pkg_min_version; then - AC_MSG_RESULT([yes]) - else - AC_MSG_RESULT([no]) - PKG_CONFIG="" - fi - -fi[]dnl -])# PKG_PROG_PKG_CONFIG - -# PKG_CHECK_EXISTS(MODULES, [ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND]) -# -# Check to see whether a particular set of modules exists. Similar -# to PKG_CHECK_MODULES(), but does not set variables or print errors. -# -# -# Similar to PKG_CHECK_MODULES, make sure that the first instance of -# this or PKG_CHECK_MODULES is called, or make sure to call -# PKG_CHECK_EXISTS manually -# -------------------------------------------------------------- -AC_DEFUN([PKG_CHECK_EXISTS], -[AC_REQUIRE([PKG_PROG_PKG_CONFIG])dnl -if test -n "$PKG_CONFIG" && \ - AC_RUN_LOG([$PKG_CONFIG --exists --print-errors "$1"]); then - m4_ifval([$2], [$2], [:]) -m4_ifvaln([$3], [else - $3])dnl -fi]) - - -# _PKG_CONFIG([VARIABLE], [COMMAND], [MODULES]) -# --------------------------------------------- -m4_define([_PKG_CONFIG], -[if test -n "$PKG_CONFIG"; then - if test -n "$$1"; then - pkg_cv_[]$1="$$1" - else - PKG_CHECK_EXISTS([$3], - [pkg_cv_[]$1=`$PKG_CONFIG --[]$2 "$3" 2>/dev/null`], - [pkg_failed=yes]) - fi -else - pkg_failed=untried -fi[]dnl -])# _PKG_CONFIG - -# _PKG_SHORT_ERRORS_SUPPORTED -# ----------------------------- -AC_DEFUN([_PKG_SHORT_ERRORS_SUPPORTED], -[AC_REQUIRE([PKG_PROG_PKG_CONFIG]) -if $PKG_CONFIG --atleast-pkgconfig-version 0.20; then - _pkg_short_errors_supported=yes -else - _pkg_short_errors_supported=no -fi[]dnl -])# _PKG_SHORT_ERRORS_SUPPORTED - - -# PKG_CHECK_MODULES(VARIABLE-PREFIX, MODULES, [ACTION-IF-FOUND], -# [ACTION-IF-NOT-FOUND]) -# -# -# Note that if there is a possibility the first call to -# PKG_CHECK_MODULES might not happen, you should be sure to include an -# explicit call to PKG_PROG_PKG_CONFIG in your configure.ac -# -# -# -------------------------------------------------------------- -AC_DEFUN([PKG_CHECK_MODULES], -[AC_REQUIRE([PKG_PROG_PKG_CONFIG])dnl -AC_ARG_VAR([$1][_CFLAGS], [C compiler flags for $1, overriding pkg-config])dnl -AC_ARG_VAR([$1][_LIBS], [linker flags for $1, overriding pkg-config])dnl - -pkg_failed=no -AC_MSG_CHECKING([for $1]) - -_PKG_CONFIG([$1][_CFLAGS], [cflags], [$2]) -_PKG_CONFIG([$1][_LIBS], [libs], [$2]) - -m4_define([_PKG_TEXT], [Alternatively, you may set the environment variables $1[]_CFLAGS -and $1[]_LIBS to avoid the need to call pkg-config. -See the pkg-config man page for more details.]) - -if test $pkg_failed = yes; then - _PKG_SHORT_ERRORS_SUPPORTED - if test $_pkg_short_errors_supported = yes; then - $1[]_PKG_ERRORS=`$PKG_CONFIG --short-errors --errors-to-stdout --print-errors "$2"` - else - $1[]_PKG_ERRORS=`$PKG_CONFIG --errors-to-stdout --print-errors "$2"` - fi - # Put the nasty error message in config.log where it belongs - echo "$$1[]_PKG_ERRORS" >&AS_MESSAGE_LOG_FD - - ifelse([$4], , [AC_MSG_ERROR(dnl -[Package requirements ($2) were not met: - -$$1_PKG_ERRORS - -Consider adjusting the PKG_CONFIG_PATH environment variable if you -installed software in a non-standard prefix. - -_PKG_TEXT -])], - [AC_MSG_RESULT([no]) - $4]) -elif test $pkg_failed = untried; then - ifelse([$4], , [AC_MSG_FAILURE(dnl -[The pkg-config script could not be found or is too old. Make sure it -is in your PATH or set the PKG_CONFIG environment variable to the full -path to pkg-config. - -_PKG_TEXT - -To get pkg-config, see <http://pkg-config.freedesktop.org/>.])], - [$4]) -else - $1[]_CFLAGS=$pkg_cv_[]$1[]_CFLAGS - $1[]_LIBS=$pkg_cv_[]$1[]_LIBS - AC_MSG_RESULT([yes]) - ifelse([$3], , :, [$3]) -fi[]dnl -])# PKG_CHECK_MODULES diff --git a/nestegg/nestegg-uninstalled.pc.in b/nestegg/nestegg-uninstalled.pc.in deleted file mode 100644 index 19bb680ac..000000000 --- a/nestegg/nestegg-uninstalled.pc.in +++ /dev/null @@ -1,13 +0,0 @@ -# nestegg uninstalled pkg-config file - -prefix=@prefix@ -exec_prefix=@exec_prefix@ -libdir=@libdir@ -includedir=@includedir@ - -Name: nestegg -Description: WebM/Matroska demuxer -Version: @VERSION@ -Conflicts: -Libs: -L${libdir} -lnestegg -Cflags: -I${includedir} diff --git a/nestegg/nestegg.pc.in b/nestegg/nestegg.pc.in deleted file mode 100644 index 32c09d79d..000000000 --- a/nestegg/nestegg.pc.in +++ /dev/null @@ -1,13 +0,0 @@ -# nestegg installed pkg-config file - -prefix=@prefix@ -exec_prefix=@exec_prefix@ -libdir=@libdir@ -includedir=@includedir@ - -Name: nestegg -Description: WebM/Matroska demuxer -Version: @VERSION@ -Conflicts: -Libs: -L${libdir} -lnestegg -Cflags: -I${includedir} diff --git a/test/datarate_test.cc b/test/datarate_test.cc index 5ae76d761..39c9a5a67 100644 --- a/test/datarate_test.cc +++ b/test/datarate_test.cc @@ -381,7 +381,6 @@ TEST_P(DatarateTestVP9, BasicRateTargeting) { } } -#if CONFIG_NON420 // Check basic rate targeting, TEST_P(DatarateTestVP9, BasicRateTargeting444) { ::libvpx_test::Y4mVideoSource video("rush_hour_444.y4m", 0, 140); @@ -410,7 +409,6 @@ TEST_P(DatarateTestVP9, BasicRateTargeting444) { << cfg_.rc_target_bitrate << " "<< effective_datarate_; } } -#endif // Check that (1) the first dropped frame gets earlier and earlier // as the drop frame threshold is increased, and (2) that the total number of diff --git a/test/ivf_video_source.h b/test/ivf_video_source.h index 3fbafbd70..824a39d7e 100644 --- a/test/ivf_video_source.h +++ b/test/ivf_video_source.h @@ -94,14 +94,14 @@ class IVFVideoSource : public CompressedVideoSource { virtual const uint8_t *cxdata() const { return end_of_file_ ? NULL : compressed_frame_buf_; } - virtual const unsigned int frame_size() const { return frame_sz_; } - virtual const unsigned int frame_number() const { return frame_; } + virtual size_t frame_size() const { return frame_sz_; } + virtual unsigned int frame_number() const { return frame_; } protected: std::string file_name_; FILE *input_file_; uint8_t *compressed_frame_buf_; - unsigned int frame_sz_; + size_t frame_sz_; unsigned int frame_; bool end_of_file_; }; diff --git a/test/test-data.sha1 b/test/test-data.sha1 index a28773133..6f718ef6c 100644 --- a/test/test-data.sha1 +++ b/test/test-data.sha1 @@ -576,3 +576,15 @@ d48c5db1b0f8e60521a7c749696b8067886033a3 vp90-2-09-aq2.webm 54638c38009198c38c8f3b25c182b709b6c1fd2e vp90-2-09-lf_deltas.webm.md5 510d95f3beb3b51c572611fdaeeece12277dac30 vp90-2-10-show-existing-frame.webm 14d631096f4bfa2d71f7f739aec1448fb3c33bad vp90-2-10-show-existing-frame.webm.md5 +b4318e75f73a6a08992c7326de2fb589c2a794c7 vp90-2-11-size-351x287.webm +b3c48382cf7d0454e83a02497c229d27720f9e20 vp90-2-11-size-351x287.webm.md5 +8e0096475ea2535bac71d3e2fc09e0c451c444df vp90-2-11-size-351x288.webm +19e003804ec1dfc5464813b32339a15d5ba7b42f vp90-2-11-size-351x288.webm.md5 +40cd1d6a188d7a88b21ebac1e573d3f270ab261e vp90-2-11-size-352x287.webm +68f515abe3858fc1eded46c8e6b2f727d43b5331 vp90-2-11-size-352x287.webm.md5 +9a510769ff23db410880ec3029d433e87d17f7fc vp90-2-12-droppable_1.ivf +952eaac6eefa6f62179ed1db3e922fd42fecc624 vp90-2-12-droppable_1.ivf.md5 +9a510769ff23db410880ec3029d433e87d17f7fc vp90-2-12-droppable_2.ivf +92a756469fa438220524e7fa6ac1d38c89514d17 vp90-2-12-droppable_2.ivf.md5 +c21e97e4ba486520118d78b01a5cb6e6dc33e190 vp90-2-12-droppable_3.ivf +601abc9e4176c70f82ac0381365e9b151fdd24cd vp90-2-12-droppable_3.ivf.md5 diff --git a/test/test.mk b/test/test.mk index b598d6b32..8488008d4 100644 --- a/test/test.mk +++ b/test/test.mk @@ -18,7 +18,7 @@ LIBVPX_TEST_SRCS-yes += test_vectors.cc LIBVPX_TEST_SRCS-$(CONFIG_VP8_ENCODER) += altref_test.cc LIBVPX_TEST_SRCS-$(CONFIG_VP8_ENCODER) += config_test.cc LIBVPX_TEST_SRCS-$(CONFIG_VP8_ENCODER) += cq_test.cc -LIBVPX_TEST_SRCS-$(CONFIG_VP8_ENCODER) += datarate_test.cc +LIBVPX_TEST_SRCS-$(CONFIG_ENCODERS) += datarate_test.cc LIBVPX_TEST_SRCS-yes += encode_test_driver.cc LIBVPX_TEST_SRCS-yes += encode_test_driver.h @@ -39,12 +39,12 @@ LIBVPX_TEST_SRCS-$(CONFIG_DECODERS) += ivf_video_source.h LIBVPX_TEST_SRCS-$(CONFIG_VP9_DECODER) += external_frame_buffer_test.cc ## WebM Parsing -NESTEGG_SRCS += ../nestegg/halloc/halloc.h -NESTEGG_SRCS += ../nestegg/halloc/src/align.h -NESTEGG_SRCS += ../nestegg/halloc/src/halloc.c -NESTEGG_SRCS += ../nestegg/halloc/src/hlist.h -NESTEGG_SRCS += ../nestegg/include/nestegg/nestegg.h -NESTEGG_SRCS += ../nestegg/src/nestegg.c +NESTEGG_SRCS += ../third_party/nestegg/halloc/halloc.h +NESTEGG_SRCS += ../third_party/nestegg/halloc/src/align.h +NESTEGG_SRCS += ../third_party/nestegg/halloc/src/halloc.c +NESTEGG_SRCS += ../third_party/nestegg/halloc/src/hlist.h +NESTEGG_SRCS += ../third_party/nestegg/include/nestegg/nestegg.h +NESTEGG_SRCS += ../third_party/nestegg/src/nestegg.c LIBVPX_TEST_SRCS-$(CONFIG_DECODERS) += $(NESTEGG_SRCS) LIBVPX_TEST_SRCS-$(CONFIG_DECODERS) += webm_video_source.h @@ -679,6 +679,18 @@ LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp91-2-04-yv444.webm LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp91-2-04-yv444.webm.md5 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-10-show-existing-frame.webm LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-10-show-existing-frame.webm.md5 +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-11-size-351x287.webm +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-11-size-351x287.webm.md5 +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-11-size-351x288.webm +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-11-size-351x288.webm.md5 +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-11-size-352x287.webm +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-11-size-352x287.webm.md5 +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-12-droppable_1.ivf +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-12-droppable_1.ivf.md5 +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-12-droppable_2.ivf +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-12-droppable_2.ivf.md5 +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-12-droppable_3.ivf +LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-12-droppable_3.ivf.md5 ifeq ($(CONFIG_DECODE_PERF_TESTS),yes) # BBB VP9 streams diff --git a/test/test_vectors.cc b/test/test_vectors.cc index 3227f5253..aba8a3c17 100644 --- a/test/test_vectors.cc +++ b/test/test_vectors.cc @@ -150,20 +150,18 @@ const char *kVP9TestVectors[kNumVp9TestVectors] = { "vp90-2-03-size-226x202.webm", "vp90-2-03-size-226x208.webm", "vp90-2-03-size-226x210.webm", "vp90-2-03-size-226x224.webm", "vp90-2-03-size-226x226.webm", "vp90-2-03-deltaq.webm", - "vp90-2-05-resize.ivf", "vp90-2-06-bilinear.webm", - "vp90-2-07-frame_parallel.webm", - "vp90-2-08-tile_1x2_frame_parallel.webm", "vp90-2-08-tile_1x2.webm", - "vp90-2-08-tile_1x4_frame_parallel.webm", "vp90-2-08-tile_1x4.webm", - "vp90-2-08-tile_1x8_frame_parallel.webm", "vp90-2-08-tile_1x8.webm", - "vp90-2-08-tile-4x4.webm", "vp90-2-08-tile-4x1.webm", - "vp90-2-09-subpixel-00.ivf", - "vp90-2-02-size-lf-1920x1080.webm", - "vp90-2-09-aq2.webm", - "vp90-2-09-lf_deltas.webm", - "vp90-2-10-show-existing-frame.webm", -#if CONFIG_NON420 + "vp90-2-05-resize.ivf", "vp90-2-06-bilinear.webm", + "vp90-2-07-frame_parallel.webm", "vp90-2-08-tile_1x2_frame_parallel.webm", + "vp90-2-08-tile_1x2.webm", "vp90-2-08-tile_1x4_frame_parallel.webm", + "vp90-2-08-tile_1x4.webm", "vp90-2-08-tile_1x8_frame_parallel.webm", + "vp90-2-08-tile_1x8.webm", "vp90-2-08-tile-4x4.webm", + "vp90-2-08-tile-4x1.webm", "vp90-2-09-subpixel-00.ivf", + "vp90-2-02-size-lf-1920x1080.webm", "vp90-2-09-aq2.webm", + "vp90-2-09-lf_deltas.webm", "vp90-2-10-show-existing-frame.webm", + "vp90-2-11-size-351x287.webm", "vp90-2-11-size-351x288.webm", + "vp90-2-11-size-352x287.webm", "vp90-2-12-droppable_1.ivf", + "vp90-2-12-droppable_2.ivf", "vp90-2-12-droppable_3.ivf", "vp91-2-04-yv444.webm" -#endif }; #endif // CONFIG_VP9_DECODER diff --git a/test/test_vectors.h b/test/test_vectors.h index eb592de64..d5ecc96d6 100644 --- a/test/test_vectors.h +++ b/test/test_vectors.h @@ -21,11 +21,7 @@ extern const char *kVP8TestVectors[kNumVp8TestVectors]; #endif #if CONFIG_VP9_DECODER -#if CONFIG_NON420 -const int kNumVp9TestVectors = 217; -#else -const int kNumVp9TestVectors = 216; -#endif +const int kNumVp9TestVectors = 223; extern const char *kVP9TestVectors[kNumVp9TestVectors]; #endif // CONFIG_VP9_DECODER diff --git a/test/video_source.h b/test/video_source.h index 3d01d39b2..6d1855ae3 100644 --- a/test/video_source.h +++ b/test/video_source.h @@ -184,9 +184,9 @@ class CompressedVideoSource { virtual const uint8_t *cxdata() const = 0; - virtual const unsigned int frame_size() const = 0; + virtual size_t frame_size() const = 0; - virtual const unsigned int frame_number() const = 0; + virtual unsigned int frame_number() const = 0; }; } // namespace libvpx_test diff --git a/test/vp9_lossless_test.cc b/test/vp9_lossless_test.cc index 2282687dc..ad7ba44a8 100644 --- a/test/vp9_lossless_test.cc +++ b/test/vp9_lossless_test.cc @@ -73,7 +73,6 @@ TEST_P(LossLessTest, TestLossLessEncoding) { EXPECT_GE(psnr_lossless, kMaxPsnr); } -#if CONFIG_NON420 TEST_P(LossLessTest, TestLossLessEncoding444) { libvpx_test::Y4mVideoSource video("rush_hour_444.y4m", 0, 10); @@ -90,7 +89,6 @@ TEST_P(LossLessTest, TestLossLessEncoding444) { const double psnr_lossless = GetMinPsnr(); EXPECT_GE(psnr_lossless, kMaxPsnr); } -#endif VP9_INSTANTIATE_TEST_CASE(LossLessTest, ALL_TEST_MODES); } // namespace diff --git a/test/webm_video_source.h b/test/webm_video_source.h index 53b0ba2b4..9d88ae360 100644 --- a/test/webm_video_source.h +++ b/test/webm_video_source.h @@ -14,7 +14,7 @@ #include <cstdlib> #include <new> #include <string> -#include "nestegg/include/nestegg/nestegg.h" +#include "third_party/nestegg/include/nestegg/nestegg.h" #include "test/video_source.h" namespace libvpx_test { @@ -108,7 +108,7 @@ class WebMVideoSource : public CompressedVideoSource { nestegg_io io = {nestegg_read_cb, nestegg_seek_cb, nestegg_tell_cb, input_file_}; - ASSERT_FALSE(nestegg_init(&nestegg_ctx_, io, NULL)) + ASSERT_FALSE(nestegg_init(&nestegg_ctx_, io, NULL, -1)) << "nestegg_init failed"; unsigned int n; @@ -169,8 +169,8 @@ class WebMVideoSource : public CompressedVideoSource { virtual const uint8_t *cxdata() const { return end_of_file_ ? NULL : buf_; } - virtual const unsigned int frame_size() const { return buf_sz_; } - virtual const unsigned int frame_number() const { return frame_; } + virtual size_t frame_size() const { return buf_sz_; } + virtual unsigned int frame_number() const { return frame_; } protected: std::string file_name_; diff --git a/third_party/nestegg/0001-include-paths.diff b/third_party/nestegg/0001-include-paths.diff new file mode 100644 index 000000000..a704ebdcd --- /dev/null +++ b/third_party/nestegg/0001-include-paths.diff @@ -0,0 +1,41 @@ +diff --git a/nestegg/halloc/src/halloc.c b/nestegg/halloc/src/halloc.c +index 5758fc0..837b3ff 100644 +--- a/nestegg/halloc/src/halloc.c ++++ b/nestegg/halloc/src/halloc.c +@@ -15,7 +15,7 @@ + #include <stdlib.h> /* realloc */ + #include <string.h> /* memset & co */ + +-#include "halloc.h" ++#include "third_party/nestegg/halloc/halloc.h" + #include "align.h" + #include "hlist.h" + +diff --git a/nestegg/include/nestegg/nestegg.h b/nestegg/include/nestegg/nestegg.h +index ff13728..c18d1d3 100644 +--- a/nestegg/include/nestegg/nestegg.h ++++ b/nestegg/include/nestegg/nestegg.h +@@ -7,7 +7,7 @@ + #if !defined(NESTEGG_671cac2a_365d_ed69_d7a3_4491d3538d79) + #define NESTEGG_671cac2a_365d_ed69_d7a3_4491d3538d79 + +-#include <nestegg/nestegg-stdint.h> ++#include "vpx/vpx_integer.h" + + #if defined(__cplusplus) + extern "C" { +diff --git a/nestegg/src/nestegg.c b/nestegg/src/nestegg.c +index daf1eed..4fb10e7 100644 +--- a/nestegg/src/nestegg.c ++++ b/nestegg/src/nestegg.c +@@ -8,8 +8,8 @@ + #include <stdlib.h> + #include <string.h> + +-#include "halloc.h" +-#include "nestegg/nestegg.h" ++#include "third_party/nestegg/halloc/halloc.h" ++#include "third_party/nestegg/include/nestegg/nestegg.h" + + /* EBML Elements */ + #define ID_EBML 0x1a45dfa3 diff --git a/third_party/nestegg/0002-ne_read_simple-uninitialized_variable.diff b/third_party/nestegg/0002-ne_read_simple-uninitialized_variable.diff new file mode 100644 index 000000000..c3bc9e575 --- /dev/null +++ b/third_party/nestegg/0002-ne_read_simple-uninitialized_variable.diff @@ -0,0 +1,21 @@ +diff --git a/nestegg/src/nestegg.c b/nestegg/src/nestegg.c +index 4fb10e7..b6bc460 100644 +--- a/nestegg/src/nestegg.c ++++ b/nestegg/src/nestegg.c +@@ -934,7 +934,7 @@ static int + ne_read_simple(nestegg * ctx, struct ebml_element_desc * desc, size_t length) + { + struct ebml_type * storage; +- int r; ++ int r = 0; + + storage = (struct ebml_type *) (ctx->ancestor->data + desc->offset); + +@@ -968,7 +968,6 @@ ne_read_simple(nestegg * ctx, struct ebml_element_desc * desc, size_t length) + case TYPE_MASTER: + case TYPE_UNKNOWN: + assert(0); +- r = 0; + break; + } + diff --git a/third_party/nestegg/AUTHORS b/third_party/nestegg/AUTHORS new file mode 100644 index 000000000..7d2c61265 --- /dev/null +++ b/third_party/nestegg/AUTHORS @@ -0,0 +1,3 @@ +Matthew Gregan <kinetik@flim.org> +Steve Workman <sjhworkman@gmail.com> +Paul Adenot <paul@paul.cx> diff --git a/nestegg/INSTALL b/third_party/nestegg/INSTALL index 401df4184..401df4184 100644 --- a/nestegg/INSTALL +++ b/third_party/nestegg/INSTALL diff --git a/nestegg/LICENSE b/third_party/nestegg/LICENSE index a67984a61..a67984a61 100644 --- a/nestegg/LICENSE +++ b/third_party/nestegg/LICENSE diff --git a/nestegg/README b/third_party/nestegg/README index 47c8237d2..47c8237d2 100644 --- a/nestegg/README +++ b/third_party/nestegg/README diff --git a/third_party/nestegg/README.webm b/third_party/nestegg/README.webm new file mode 100644 index 000000000..c93116882 --- /dev/null +++ b/third_party/nestegg/README.webm @@ -0,0 +1,16 @@ +URL: https://github.com/kinetiknz/nestegg.git +Version: f46223191d8116a36bf299b5b9793fcb798417b1 +License: ISC-style +License File: LICENSE + +Description: +The source under the halloc/ directory is licensed under a BSD license. See +halloc/halloc.h for details. + +Local Modifications: +- delete unnecessary docs and build files +- nestegg/0001-include-paths.diff + include path modifications for the libvpx build system +- 0002-ne_read_simple-uninitialized_variable.diff + fixes: +nestegg.c|975 col 6| warning: ‘r’ may be used uninitialized in this function [-Wuninitialized] diff --git a/nestegg/TODO b/third_party/nestegg/TODO index bf0cb04c4..bf0cb04c4 100644 --- a/nestegg/TODO +++ b/third_party/nestegg/TODO diff --git a/nestegg/halloc/README b/third_party/nestegg/halloc/README index 380fba2b8..380fba2b8 100644 --- a/nestegg/halloc/README +++ b/third_party/nestegg/halloc/README diff --git a/nestegg/halloc/halloc.h b/third_party/nestegg/halloc/halloc.h index 10af4e8d8..10af4e8d8 100644 --- a/nestegg/halloc/halloc.h +++ b/third_party/nestegg/halloc/halloc.h diff --git a/nestegg/halloc/src/align.h b/third_party/nestegg/halloc/src/align.h index 4c6e1831f..4c6e1831f 100644 --- a/nestegg/halloc/src/align.h +++ b/third_party/nestegg/halloc/src/align.h diff --git a/nestegg/halloc/src/halloc.c b/third_party/nestegg/halloc/src/halloc.c index 38fd6c11a..8860d736a 100644 --- a/nestegg/halloc/src/halloc.c +++ b/third_party/nestegg/halloc/src/halloc.c @@ -15,7 +15,7 @@ #include <stdlib.h> /* realloc */ #include <string.h> /* memset & co */ -#include "../halloc.h" +#include "third_party/nestegg/halloc/halloc.h" #include "align.h" #include "hlist.h" diff --git a/nestegg/halloc/src/hlist.h b/third_party/nestegg/halloc/src/hlist.h index 2791f78c7..2791f78c7 100644 --- a/nestegg/halloc/src/hlist.h +++ b/third_party/nestegg/halloc/src/hlist.h diff --git a/nestegg/halloc/src/macros.h b/third_party/nestegg/halloc/src/macros.h index 1f84bc277..1f84bc277 100644 --- a/nestegg/halloc/src/macros.h +++ b/third_party/nestegg/halloc/src/macros.h diff --git a/nestegg/include/nestegg/nestegg.h b/third_party/nestegg/include/nestegg/nestegg.h index 6510694ba..c18d1d3bf 100644 --- a/nestegg/include/nestegg/nestegg.h +++ b/third_party/nestegg/include/nestegg/nestegg.h @@ -4,12 +4,12 @@ * This program is made available under an ISC-style license. See the * accompanying file LICENSE for details. */ -#ifndef NESTEGG_671cac2a_365d_ed69_d7a3_4491d3538d79 -#define NESTEGG_671cac2a_365d_ed69_d7a3_4491d3538d79 +#if !defined(NESTEGG_671cac2a_365d_ed69_d7a3_4491d3538d79) +#define NESTEGG_671cac2a_365d_ed69_d7a3_4491d3538d79 #include "vpx/vpx_integer.h" -#ifdef __cplusplus +#if defined(__cplusplus) extern "C" { #endif @@ -17,10 +17,10 @@ extern "C" { @section intro Introduction - This is the documentation fot the <tt>libnestegg</tt> C API. + This is the documentation for the <tt>libnestegg</tt> C API. <tt>libnestegg</tt> is a demultiplexing library for <a - href="http://www.matroska.org/">Matroska</a> and <a - href="http://www.webmproject.org/">WebMedia</a> media files. + href="http://www.webmproject.org/code/specs/container/">WebM</a> + media files. @section example Example code @@ -68,6 +68,13 @@ extern "C" { #define NESTEGG_CODEC_VP8 0 /**< Track uses Google On2 VP8 codec. */ #define NESTEGG_CODEC_VORBIS 1 /**< Track uses Xiph Vorbis codec. */ #define NESTEGG_CODEC_VP9 2 /**< Track uses Google On2 VP9 codec. */ +#define NESTEGG_CODEC_OPUS 3 /**< Track uses Xiph Opus codec. */ + +#define NESTEGG_VIDEO_MONO 0 /**< Track is mono video. */ +#define NESTEGG_VIDEO_STEREO_LEFT_RIGHT 1 /**< Track is side-by-side stereo video. Left first. */ +#define NESTEGG_VIDEO_STEREO_BOTTOM_TOP 2 /**< Track is top-bottom stereo video. Right first. */ +#define NESTEGG_VIDEO_STEREO_TOP_BOTTOM 3 /**< Track is top-bottom stereo video. Left first. */ +#define NESTEGG_VIDEO_STEREO_RIGHT_LEFT 11 /**< Track is side-by-side stereo video. Right first. */ #define NESTEGG_SEEK_SET 0 /**< Seek offset relative to beginning of stream. */ #define NESTEGG_SEEK_CUR 1 /**< Seek offset relative to current position in stream. */ @@ -114,6 +121,10 @@ typedef struct { /** Parameters specific to a video track. */ typedef struct { + unsigned int stereo_mode; /**< Video mode. One of #NESTEGG_VIDEO_MONO, + #NESTEGG_VIDEO_STEREO_LEFT_RIGHT, + #NESTEGG_VIDEO_STEREO_BOTTOM_TOP, or + #NESTEGG_VIDEO_STEREO_TOP_BOTTOM. */ unsigned int width; /**< Width of the video frame in pixels. */ unsigned int height; /**< Height of the video frame in pixels. */ unsigned int display_width; /**< Display width of the video frame in pixels. */ @@ -129,6 +140,8 @@ typedef struct { double rate; /**< Sampling rate in Hz. */ unsigned int channels; /**< Number of audio channels. */ unsigned int depth; /**< Bits per sample. */ + uint64_t codec_delay; /**< Nanoseconds that must be discarded from the start. */ + uint64_t seek_preroll;/**< Nanoseconds that must be discarded after a seek. */ } nestegg_audio_params; /** Logging callback function pointer. */ @@ -140,9 +153,10 @@ typedef void (* nestegg_log)(nestegg * context, unsigned int severity, char cons @param context Storage for the new nestegg context. @see nestegg_destroy @param io User supplied IO context. @param callback Optional logging callback function pointer. May be NULL. + @param max_offset Optional maximum offset to be read. Set -1 to ignore. @retval 0 Success. @retval -1 Error. */ -int nestegg_init(nestegg ** context, nestegg_io io, nestegg_log callback); +int nestegg_init(nestegg ** context, nestegg_io io, nestegg_log callback, int64_t max_offset); /** Destroy a nestegg context and free associated memory. @param context #nestegg context to be freed. @see nestegg_init */ @@ -171,6 +185,29 @@ int nestegg_tstamp_scale(nestegg * context, uint64_t * scale); @retval -1 Error. */ int nestegg_track_count(nestegg * context, unsigned int * tracks); +/** Query the start and end offset for a particular cluster. + @param context Stream context initialized by #nestegg_init. + @param cluster_num Zero-based cluster number; order they appear in cues. + @param max_offset Optional maximum offset to be read. Set -1 to ignore. + @param start_pos Starting offset of the cluster. -1 means non-existant. + @param end_pos Starting offset of the cluster. -1 means non-existant or + final cluster. + @param tstamp Starting timestamp of the cluster. + @retval 0 Success. + @retval -1 Error. */ +int nestegg_get_cue_point(nestegg * context, unsigned int cluster_num, + int64_t max_offset, int64_t * start_pos, + int64_t * end_pos, uint64_t * tstamp); + +/** Seek to @a offset. Stream will seek directly to offset. + Should be used to seek to the start of a resync point, i.e. cluster; the + parser will not be able to understand other offsets. + @param context Stream context initialized by #nestegg_init. + @param offset Absolute offset in bytes. + @retval 0 Success. + @retval -1 Error. */ +int nestegg_offset_seek(nestegg * context, uint64_t offset); + /** Seek @a track to @a tstamp. Stream seek will terminate at the earliest key point in the stream at or before @a tstamp. Other tracks in the stream will output packets with unspecified but nearby timestamps. @@ -286,7 +323,30 @@ int nestegg_packet_count(nestegg_packet * packet, unsigned int * count); int nestegg_packet_data(nestegg_packet * packet, unsigned int item, unsigned char ** data, size_t * length); -#ifdef __cplusplus +/** Returns discard_padding for given packet + @param packet Packet initialized by #nestegg_read_packet. + @param discard_padding pointer to store discard padding in. + @retval 0 Success. + @retval -1 Error. */ +int nestegg_packet_discard_padding(nestegg_packet * packet, + int64_t * discard_padding); + +/** Query the presence of cues. + @param context Stream context initialized by #nestegg_init. + @retval 0 The media has no cues. + @retval 1 The media has cues. */ +int nestegg_has_cues(nestegg * context); + +/** + * Try to determine if the buffer looks like the beginning of a WebM file. + * + * @param buffer A buffer containing the beginning of a media file. + * @param length The size of the buffer. + * @retval 0 The file is not a WebM file. + * @retval 1 The file is a WebM file. */ +int nestegg_sniff(unsigned char const * buffer, size_t length); + +#if defined(__cplusplus) } #endif diff --git a/nestegg/src/nestegg.c b/third_party/nestegg/src/nestegg.c index ae87e8f8f..30e0e2bb8 100644 --- a/nestegg/src/nestegg.c +++ b/third_party/nestegg/src/nestegg.c @@ -8,8 +8,8 @@ #include <stdlib.h> #include <string.h> -#include "nestegg/halloc/halloc.h" -#include "nestegg/include/nestegg/nestegg.h" +#include "third_party/nestegg/halloc/halloc.h" +#include "third_party/nestegg/include/nestegg/nestegg.h" /* EBML Elements */ #define ID_EBML 0x1a45dfa3 @@ -25,7 +25,7 @@ #define ID_VOID 0xec #define ID_CRC32 0xbf -/* WebMedia Elements */ +/* WebM Elements */ #define ID_SEGMENT 0x18538067 /* Seek Head Elements */ @@ -49,6 +49,7 @@ #define ID_BLOCK 0xa1 #define ID_BLOCK_DURATION 0x9b #define ID_REFERENCE_BLOCK 0xfb +#define ID_DISCARD_PADDING 0x75a2 /* Tracks Elements */ #define ID_TRACKS 0x1654ae6b @@ -63,9 +64,12 @@ #define ID_LANGUAGE 0x22b59c #define ID_CODEC_ID 0x86 #define ID_CODEC_PRIVATE 0x63a2 +#define ID_CODEC_DELAY 0x56aa +#define ID_SEEK_PREROLL 0x56bb /* Video Elements */ #define ID_VIDEO 0xe0 +#define ID_STEREO_MODE 0x53b8 #define ID_PIXEL_WIDTH 0xb0 #define ID_PIXEL_HEIGHT 0xba #define ID_PIXEL_CROP_BOTTOM 0x54aa @@ -129,6 +133,7 @@ enum ebml_type_enum { #define TRACK_ID_VP8 "V_VP8" #define TRACK_ID_VP9 "V_VP9" #define TRACK_ID_VORBIS "A_VORBIS" +#define TRACK_ID_OPUS "A_OPUS" enum vint_mask { MASK_NONE, @@ -192,6 +197,7 @@ struct info { struct block_group { struct ebml_type duration; struct ebml_type reference_block; + struct ebml_type discard_padding; }; struct cluster { @@ -200,6 +206,7 @@ struct cluster { }; struct video { + struct ebml_type stereo_mode; struct ebml_type pixel_width; struct ebml_type pixel_height; struct ebml_type pixel_crop_bottom; @@ -227,6 +234,8 @@ struct track_entry { struct ebml_type language; struct ebml_type codec_id; struct ebml_type codec_private; + struct ebml_type codec_delay; + struct ebml_type seek_preroll; struct video video; struct audio audio; }; @@ -274,6 +283,7 @@ struct saved_state { struct list_node * ancestor; uint64_t last_id; uint64_t last_size; + int last_valid; }; struct frame { @@ -289,6 +299,7 @@ struct nestegg { struct pool_ctx * alloc_pool; uint64_t last_id; uint64_t last_size; + int last_valid; struct list_node * ancestor; struct ebml ebml; struct segment segment; @@ -300,6 +311,7 @@ struct nestegg_packet { uint64_t track; uint64_t timecode; struct frame * frame; + int64_t discard_padding; }; /* Element Descriptor */ @@ -341,7 +353,7 @@ static struct ebml_element_desc ne_ebml_elements[] = { E_LAST }; -/* WebMedia Element Lists */ +/* WebM Element Lists */ static struct ebml_element_desc ne_seek_elements[] = { E_FIELD(ID_SEEK_ID, TYPE_BINARY, struct seek, id), E_FIELD(ID_SEEK_POSITION, TYPE_UINT, struct seek, position), @@ -363,6 +375,7 @@ static struct ebml_element_desc ne_block_group_elements[] = { E_SUSPEND(ID_BLOCK, TYPE_BINARY), E_FIELD(ID_BLOCK_DURATION, TYPE_UINT, struct block_group, duration), E_FIELD(ID_REFERENCE_BLOCK, TYPE_INT, struct block_group, reference_block), + E_FIELD(ID_DISCARD_PADDING, TYPE_INT, struct block_group, discard_padding), E_LAST }; @@ -374,6 +387,7 @@ static struct ebml_element_desc ne_cluster_elements[] = { }; static struct ebml_element_desc ne_video_elements[] = { + E_FIELD(ID_STEREO_MODE, TYPE_UINT, struct video, stereo_mode), E_FIELD(ID_PIXEL_WIDTH, TYPE_UINT, struct video, pixel_width), E_FIELD(ID_PIXEL_HEIGHT, TYPE_UINT, struct video, pixel_height), E_FIELD(ID_PIXEL_CROP_BOTTOM, TYPE_UINT, struct video, pixel_crop_bottom), @@ -403,6 +417,8 @@ static struct ebml_element_desc ne_track_entry_elements[] = { E_FIELD(ID_LANGUAGE, TYPE_STRING, struct track_entry, language), E_FIELD(ID_CODEC_ID, TYPE_STRING, struct track_entry, codec_id), E_FIELD(ID_CODEC_PRIVATE, TYPE_BINARY, struct track_entry, codec_private), + E_FIELD(ID_CODEC_DELAY, TYPE_UINT, struct track_entry, codec_delay), + E_FIELD(ID_SEEK_PREROLL, TYPE_UINT, struct track_entry, seek_preroll), E_SINGLE_MASTER(ID_VIDEO, TYPE_MASTER, struct track_entry, video), E_SINGLE_MASTER(ID_AUDIO, TYPE_MASTER, struct track_entry, audio), E_LAST @@ -660,7 +676,7 @@ ne_read_float(nestegg_io * io, double * val, uint64_t length) } value; int r; - /* length == 10 not implemented */ + /* Length == 10 not implemented. */ if (length != 4 && length != 8) return -1; r = ne_read_uint(io, &value.u, length); @@ -808,6 +824,7 @@ ne_ctx_save(nestegg * ctx, struct saved_state * s) s->ancestor = ctx->ancestor; s->last_id = ctx->last_id; s->last_size = ctx->last_size; + s->last_valid = ctx->last_valid; return 0; } @@ -822,6 +839,7 @@ ne_ctx_restore(nestegg * ctx, struct saved_state * s) ctx->ancestor = s->ancestor; ctx->last_id = s->last_id; ctx->last_size = s->last_size; + ctx->last_valid = s->last_valid; return 0; } @@ -830,7 +848,7 @@ ne_peek_element(nestegg * ctx, uint64_t * id, uint64_t * size) { int r; - if (ctx->last_id && ctx->last_size) { + if (ctx->last_valid) { if (id) *id = ctx->last_id; if (size) @@ -851,6 +869,8 @@ ne_peek_element(nestegg * ctx, uint64_t * id, uint64_t * size) if (size) *size = ctx->last_size; + ctx->last_valid = 1; + return 1; } @@ -863,8 +883,7 @@ ne_read_element(nestegg * ctx, uint64_t * id, uint64_t * size) if (r != 1) return r; - ctx->last_id = 0; - ctx->last_size = 0; + ctx->last_valid = 0; return 1; } @@ -915,7 +934,7 @@ static int ne_read_simple(nestegg * ctx, struct ebml_element_desc * desc, size_t length) { struct ebml_type * storage; - int r; + int r = 0; storage = (struct ebml_type *) (ctx->ancestor->data + desc->offset); @@ -930,8 +949,6 @@ ne_read_simple(nestegg * ctx, struct ebml_element_desc * desc, size_t length) ctx->log(ctx, NESTEGG_LOG_DEBUG, "element %llx (%s) -> %p (%u)", desc->id, desc->name, storage, desc->offset); - r = -1; - switch (desc->type) { case TYPE_UINT: r = ne_read_uint(ctx->io, &storage->v.u, length); @@ -961,29 +978,26 @@ ne_read_simple(nestegg * ctx, struct ebml_element_desc * desc, size_t length) } static int -ne_parse(nestegg * ctx, struct ebml_element_desc * top_level) +ne_parse(nestegg * ctx, struct ebml_element_desc * top_level, int64_t max_offset) { int r; int64_t * data_offset; - uint64_t id, size; + uint64_t id, size, peeked_id; struct ebml_element_desc * element; - /* loop until we need to return: - - hit suspend point - - parse complete - - error occurred */ - - /* loop over elements at current level reading them if sublevel found, - push ctx onto stack and continue if sublevel ended, pop ctx off stack - and continue */ - if (!ctx->ancestor) return -1; for (;;) { + if (max_offset > 0 && ne_io_tell(ctx->io) >= max_offset) { + /* Reached end of offset allowed for parsing - return gracefully */ + r = 1; + break; + } r = ne_peek_element(ctx, &id, &size); if (r != 1) break; + peeked_id = id; element = ne_find_element(id, ctx->ancestor->node); if (element) { @@ -997,6 +1011,7 @@ ne_parse(nestegg * ctx, struct ebml_element_desc * top_level) r = ne_read_element(ctx, &id, &size); if (r != 1) break; + assert(id == peeked_id); if (element->flags & DESC_FLAG_OFFSET) { data_offset = (int64_t *) (ctx->ancestor->data + element->data_offset); @@ -1106,7 +1121,7 @@ ne_read_xiph_lacing(nestegg_io * io, size_t block, size_t * read, uint64_t n, ui if (*read + sum > block) return -1; - /* last frame is the remainder of the block */ + /* Last frame is the remainder of the block. */ sizes[i] = block - *read - sum; return 1; } @@ -1143,7 +1158,7 @@ ne_read_ebml_lacing(nestegg_io * io, size_t block, size_t * read, uint64_t n, ui if (*read + sum > block) return -1; - /* last frame is the remainder of the block */ + /* Last frame is the remainder of the block. */ sizes[i] = block - *read - sum; return 1; } @@ -1159,6 +1174,37 @@ ne_get_timecode_scale(nestegg * ctx) return scale; } +static int +ne_map_track_number_to_index(nestegg * ctx, + unsigned int track_number, + unsigned int * track_index) +{ + struct ebml_list_node * node; + struct track_entry * t_entry; + uint64_t t_number = 0; + + if (!track_index) + return -1; + *track_index = 0; + + if (track_number == 0) + return -1; + + node = ctx->segment.tracks.track_entry.head; + while (node) { + assert(node->id == ID_TRACK_ENTRY); + t_entry = node->data; + if (ne_get_uint(t_entry->number, &t_number) != 0) + return -1; + if (t_number == track_number) + return 0; + *track_index += 1; + node = node->next; + } + + return -1; +} + static struct track_entry * ne_find_track_entry(nestegg * ctx, unsigned int track) { @@ -1187,8 +1233,8 @@ ne_read_block(nestegg * ctx, uint64_t block_id, uint64_t block_size, nestegg_pac struct frame * f, * last; struct track_entry * entry; double track_scale; - uint64_t track, length, frame_sizes[256], cluster_tc, flags, frames, tc_scale, total; - unsigned int i, lacing; + uint64_t track_number, length, frame_sizes[256], cluster_tc, flags, frames, tc_scale, total; + unsigned int i, lacing, track; size_t consumed = 0; *data = NULL; @@ -1196,11 +1242,11 @@ ne_read_block(nestegg * ctx, uint64_t block_id, uint64_t block_size, nestegg_pac if (block_size > LIMIT_BLOCK) return -1; - r = ne_read_vint(ctx->io, &track, &length); + r = ne_read_vint(ctx->io, &track_number, &length); if (r != 1) return r; - if (track == 0 || track > ctx->track_count) + if (track_number == 0) return -1; consumed += length; @@ -1219,8 +1265,8 @@ ne_read_block(nestegg * ctx, uint64_t block_id, uint64_t block_size, nestegg_pac frames = 0; - /* flags are different between block and simpleblock, but lacing is - encoded the same way */ + /* Flags are different between Block and SimpleBlock, but lacing is + encoded the same way. */ lacing = (flags & BLOCK_FLAGS_LACING) >> 1; switch (lacing) { @@ -1266,14 +1312,17 @@ ne_read_block(nestegg * ctx, uint64_t block_id, uint64_t block_size, nestegg_pac break; } - /* sanity check unlaced frame sizes against total block size. */ + /* Sanity check unlaced frame sizes against total block size. */ total = consumed; for (i = 0; i < frames; ++i) total += frame_sizes[i]; if (total > block_size) return -1; - entry = ne_find_track_entry(ctx, (unsigned int)(track - 1)); + if (ne_map_track_number_to_index(ctx, track_number, &track) != 0) + return -1; + + entry = ne_find_track_entry(ctx, track); if (!entry) return -1; @@ -1291,8 +1340,8 @@ ne_read_block(nestegg * ctx, uint64_t block_id, uint64_t block_size, nestegg_pac return -1; pkt = ne_alloc(sizeof(*pkt)); - pkt->track = track - 1; - pkt->timecode = (uint64_t)(abs_timecode * tc_scale * track_scale); + pkt->track = track; + pkt->timecode = abs_timecode * tc_scale * track_scale; ctx->log(ctx, NESTEGG_LOG_DEBUG, "%sblock t %lld pts %f f %llx frames: %llu", block_id == ID_BLOCK ? "" : "simple", pkt->track, pkt->timecode / 1e9, flags, frames); @@ -1326,6 +1375,35 @@ ne_read_block(nestegg * ctx, uint64_t block_id, uint64_t block_size, nestegg_pac return 1; } +static int +ne_read_discard_padding(nestegg * ctx, nestegg_packet * pkt) +{ + int r; + uint64_t id, size; + struct ebml_element_desc * element; + struct ebml_type * storage; + + r = ne_peek_element(ctx, &id, &size); + if (r != 1) + return r; + + if (id != ID_DISCARD_PADDING) + return 1; + + element = ne_find_element(id, ctx->ancestor->node); + if (!element) + return 1; + + r = ne_read_simple(ctx, element, size); + if (r != 1) + return r; + storage = (struct ebml_type *) (ctx->ancestor->data + element->offset); + pkt->discard_padding = storage->v.i; + + return 1; +} + + static uint64_t ne_buf_read_id(unsigned char const * p, size_t length) { @@ -1369,8 +1447,33 @@ ne_find_seek_for_id(struct ebml_list_node * seek_head, uint64_t id) return NULL; } +static struct cue_track_positions * +ne_find_cue_position_for_track(nestegg * ctx, struct ebml_list_node * node, unsigned int track) +{ + struct cue_track_positions * pos = NULL; + uint64_t track_number; + unsigned int t; + + while (node) { + assert(node->id == ID_CUE_TRACK_POSITIONS); + pos = node->data; + if (ne_get_uint(pos->track, &track_number) != 0) + return NULL; + + if (ne_map_track_number_to_index(ctx, track_number, &t) != 0) + return NULL; + + if (t == track) + return pos; + + node = node->next; + } + + return NULL; +} + static struct cue_point * -ne_find_cue_point_for_tstamp(struct ebml_list_node * cue_point, uint64_t scale, uint64_t tstamp) +ne_find_cue_point_for_tstamp(nestegg * ctx, struct ebml_list_node * cue_point, unsigned int track, uint64_t scale, uint64_t tstamp) { uint64_t time; struct cue_point * c, * prev = NULL; @@ -1385,7 +1488,9 @@ ne_find_cue_point_for_tstamp(struct ebml_list_node * cue_point, uint64_t scale, if (ne_get_uint(c->time, &time) == 0 && time * scale > tstamp) break; - prev = cue_point->data; + if (ne_find_cue_position_for_track(ctx, c->cue_track_positions.head, track) != NULL) + prev = c; + cue_point = cue_point->next; } @@ -1395,7 +1500,6 @@ ne_find_cue_point_for_tstamp(struct ebml_list_node * cue_point, uint64_t scale, static int ne_is_suspend_element(uint64_t id) { - /* this could search the tree of elements for DESC_FLAG_SUSPEND */ if (id == ID_SIMPLE_BLOCK || id == ID_BLOCK) return 1; return 0; @@ -1408,14 +1512,180 @@ ne_null_log_callback(nestegg * ctx, unsigned int severity, char const * fmt, ... return; } +static int +ne_init_cue_points(nestegg * ctx, int64_t max_offset) +{ + int r; + struct ebml_list_node * node = ctx->segment.cues.cue_point.head; + struct seek * found; + uint64_t seek_pos, id; + struct saved_state state; + + /* If there are no cues loaded, check for cues element in the seek head + and load it. */ + if (!node) { + found = ne_find_seek_for_id(ctx->segment.seek_head.head, ID_CUES); + if (!found) + return -1; + + if (ne_get_uint(found->position, &seek_pos) != 0) + return -1; + + /* Save old parser state. */ + r = ne_ctx_save(ctx, &state); + if (r != 0) + return -1; + + /* Seek and set up parser state for segment-level element (Cues). */ + r = ne_io_seek(ctx->io, ctx->segment_offset + seek_pos, NESTEGG_SEEK_SET); + if (r != 0) + return -1; + ctx->last_valid = 0; + + r = ne_read_element(ctx, &id, NULL); + if (r != 1) + return -1; + + if (id != ID_CUES) + return -1; + + ctx->ancestor = NULL; + ne_ctx_push(ctx, ne_top_level_elements, ctx); + ne_ctx_push(ctx, ne_segment_elements, &ctx->segment); + ne_ctx_push(ctx, ne_cues_elements, &ctx->segment.cues); + /* parser will run until end of cues element. */ + ctx->log(ctx, NESTEGG_LOG_DEBUG, "seek: parsing cue elements"); + r = ne_parse(ctx, ne_cues_elements, max_offset); + while (ctx->ancestor) + ne_ctx_pop(ctx); + + /* Reset parser state to original state and seek back to old position. */ + if (ne_ctx_restore(ctx, &state) != 0) + return -1; + + if (r < 0) + return -1; + + node = ctx->segment.cues.cue_point.head; + if (!node) + return -1; + } + + return 0; +} + +/* Three functions that implement the nestegg_io interface, operating on a + * sniff_buffer. */ +struct sniff_buffer { + unsigned char const * buffer; + size_t length; + int64_t offset; +}; + +static int +ne_buffer_read(void * buffer, size_t length, void * user_data) +{ + struct sniff_buffer * sb = user_data; + + int rv = 1; + size_t available = sb->length - sb->offset; + + if (available < length) + return 0; + + memcpy(buffer, sb->buffer + sb->offset, length); + sb->offset += length; + + return rv; +} + +static int +ne_buffer_seek(int64_t offset, int whence, void * user_data) +{ + struct sniff_buffer * sb = user_data; + int64_t o = sb->offset; + + switch(whence) { + case NESTEGG_SEEK_SET: + o = offset; + break; + case NESTEGG_SEEK_CUR: + o += offset; + break; + case NESTEGG_SEEK_END: + o = sb->length + offset; + break; + } + + if (o < 0 || o > (int64_t) sb->length) + return -1; + + sb->offset = o; + return 0; +} + +static int64_t +ne_buffer_tell(void * user_data) +{ + struct sniff_buffer * sb = user_data; + return sb->offset; +} + +static int +ne_match_webm(nestegg_io io, int64_t max_offset) +{ + int r; + uint64_t id; + char * doctype; + nestegg * ctx; + + if (!(io.read && io.seek && io.tell)) + return -1; + + ctx = ne_alloc(sizeof(*ctx)); + + ctx->io = ne_alloc(sizeof(*ctx->io)); + *ctx->io = io; + ctx->alloc_pool = ne_pool_init(); + ctx->log = ne_null_log_callback; + + r = ne_peek_element(ctx, &id, NULL); + if (r != 1) { + nestegg_destroy(ctx); + return 0; + } + + if (id != ID_EBML) { + nestegg_destroy(ctx); + return 0; + } + + ne_ctx_push(ctx, ne_top_level_elements, ctx); + + /* we don't check the return value of ne_parse, that might fail because + * max_offset is not on a valid element end point. We only want to check + * the EBML ID and that the doctype is "webm". */ + ne_parse(ctx, NULL, max_offset); + + if (ne_get_string(ctx->ebml.doctype, &doctype) != 0 || + strcmp(doctype, "webm") != 0) { + nestegg_destroy(ctx); + return 0; + } + + nestegg_destroy(ctx); + + return 1; +} + int -nestegg_init(nestegg ** context, nestegg_io io, nestegg_log callback) +nestegg_init(nestegg ** context, nestegg_io io, nestegg_log callback, int64_t max_offset) { int r; uint64_t id, version, docversion; struct ebml_list_node * track; char * doctype; - nestegg * ctx = NULL; + nestegg * ctx; if (!(io.read && io.seek && io.tell)) return -1; @@ -1445,7 +1715,7 @@ nestegg_init(nestegg ** context, nestegg_io io, nestegg_log callback) ne_ctx_push(ctx, ne_top_level_elements, ctx); - r = ne_parse(ctx, NULL); + r = ne_parse(ctx, NULL, max_offset); if (r != 1) { nestegg_destroy(ctx); @@ -1531,98 +1801,130 @@ nestegg_track_count(nestegg * ctx, unsigned int * tracks) } int -nestegg_track_seek(nestegg * ctx, unsigned int track, uint64_t tstamp) +nestegg_get_cue_point(nestegg * ctx, unsigned int cluster_num, int64_t max_offset, + int64_t * start_pos, int64_t * end_pos, uint64_t * tstamp) { - int r; + int range_obtained = 0; + unsigned int cluster_count = 0; struct cue_point * cue_point; struct cue_track_positions * pos; - struct saved_state state; - struct seek * found; - uint64_t seek_pos, tc_scale, t, id; - struct ebml_list_node * node = ctx->segment.cues.cue_point.head; + uint64_t seek_pos, track_number, tc_scale, time; + struct ebml_list_node * cues_node = ctx->segment.cues.cue_point.head; + struct ebml_list_node * cue_pos_node = NULL; + unsigned int track = 0, track_count = 0, track_index; - /* If there are no cues loaded, check for cues element in the seek head - and load it. */ - if (!node) { - found = ne_find_seek_for_id(ctx->segment.seek_head.head, ID_CUES); - if (!found) - return -1; + if (!start_pos || !end_pos || !tstamp) + return -1; - if (ne_get_uint(found->position, &seek_pos) != 0) - return -1; + /* Initialise return values */ + *start_pos = -1; + *end_pos = -1; + *tstamp = 0; - /* Save old parser state. */ - r = ne_ctx_save(ctx, &state); - if (r != 0) + if (!cues_node) { + ne_init_cue_points(ctx, max_offset); + cues_node = ctx->segment.cues.cue_point.head; + /* Verify cues have been added to context. */ + if (!cues_node) return -1; + } - /* Seek and set up parser state for segment-level element (Cues). */ - r = ne_io_seek(ctx->io, ctx->segment_offset + seek_pos, NESTEGG_SEEK_SET); - if (r != 0) - return -1; - ctx->last_id = 0; - ctx->last_size = 0; + nestegg_track_count(ctx, &track_count); - r = ne_read_element(ctx, &id, NULL); - if (r != 1) - return -1; + tc_scale = ne_get_timecode_scale(ctx); - if (id != ID_CUES) - return -1; + while (cues_node && !range_obtained) { + assert(cues_node->id == ID_CUE_POINT); + cue_point = cues_node->data; + cue_pos_node = cue_point->cue_track_positions.head; + while (cue_pos_node) { + assert(cue_pos_node->id == ID_CUE_TRACK_POSITIONS); + pos = cue_pos_node->data; + for (track = 0; track < track_count; track++) { + if (ne_get_uint(pos->track, &track_number) != 0) + return -1; + + if (ne_map_track_number_to_index(ctx, track_number, &track_index) != 0) + return -1; + + if (track_index == track) { + if (ne_get_uint(pos->cluster_position, &seek_pos) != 0) + return -1; + if (cluster_count == cluster_num) { + *start_pos = ctx->segment_offset+seek_pos; + if (ne_get_uint(cue_point->time, &time) != 0) + return -1; + *tstamp = time * tc_scale; + } else if (cluster_count == cluster_num+1) { + *end_pos = (ctx->segment_offset+seek_pos)-1; + range_obtained = 1; + break; + } + cluster_count++; + } + } + cue_pos_node = cue_pos_node->next; + } + cues_node = cues_node->next; + } - ctx->ancestor = NULL; - ne_ctx_push(ctx, ne_top_level_elements, ctx); - ne_ctx_push(ctx, ne_segment_elements, &ctx->segment); - ne_ctx_push(ctx, ne_cues_elements, &ctx->segment.cues); - /* parser will run until end of cues element. */ - ctx->log(ctx, NESTEGG_LOG_DEBUG, "seek: parsing cue elements"); - r = ne_parse(ctx, ne_cues_elements); - while (ctx->ancestor) - ne_ctx_pop(ctx); + return 0; +} - /* Reset parser state to original state and seek back to old position. */ - if (ne_ctx_restore(ctx, &state) != 0) - return -1; +int +nestegg_offset_seek(nestegg * ctx, uint64_t offset) +{ + int r; - if (r < 0) + /* Seek and set up parser state for segment-level element (Cluster). */ + r = ne_io_seek(ctx->io, offset, NESTEGG_SEEK_SET); + if (r != 0) + return -1; + ctx->last_valid = 0; + + while (ctx->ancestor) + ne_ctx_pop(ctx); + + ne_ctx_push(ctx, ne_top_level_elements, ctx); + ne_ctx_push(ctx, ne_segment_elements, &ctx->segment); + + return 0; +} + +int +nestegg_track_seek(nestegg * ctx, unsigned int track, uint64_t tstamp) +{ + int r; + struct cue_point * cue_point; + struct cue_track_positions * pos; + uint64_t seek_pos, tc_scale; + + /* If there are no cues loaded, check for cues element in the seek head + and load it. */ + if (!ctx->segment.cues.cue_point.head) { + r = ne_init_cue_points(ctx, -1); + if (r != 0) return -1; } tc_scale = ne_get_timecode_scale(ctx); - cue_point = ne_find_cue_point_for_tstamp(ctx->segment.cues.cue_point.head, tc_scale, tstamp); + cue_point = ne_find_cue_point_for_tstamp(ctx, ctx->segment.cues.cue_point.head, + track, tc_scale, tstamp); if (!cue_point) return -1; - node = cue_point->cue_track_positions.head; - - seek_pos = 0; - - while (node) { - assert(node->id == ID_CUE_TRACK_POSITIONS); - pos = node->data; - if (ne_get_uint(pos->track, &t) == 0 && t - 1 == track) { - if (ne_get_uint(pos->cluster_position, &seek_pos) != 0) - return -1; - break; - } - node = node->next; - } - - /* Seek and set up parser state for segment-level element (Cluster). */ - r = ne_io_seek(ctx->io, ctx->segment_offset + seek_pos, NESTEGG_SEEK_SET); - if (r != 0) + pos = ne_find_cue_position_for_track(ctx, cue_point->cue_track_positions.head, track); + if (pos == NULL) return -1; - ctx->last_id = 0; - ctx->last_size = 0; - while (ctx->ancestor) - ne_ctx_pop(ctx); + if (ne_get_uint(pos->cluster_position, &seek_pos) != 0) + return -1; - ne_ctx_push(ctx, ne_top_level_elements, ctx); - ne_ctx_push(ctx, ne_segment_elements, &ctx->segment); + /* Seek and set up parser state for segment-level element (Cluster). */ + r = nestegg_offset_seek(ctx, ctx->segment_offset + seek_pos); ctx->log(ctx, NESTEGG_LOG_DEBUG, "seek: parsing cluster elements"); - r = ne_parse(ctx, NULL); + r = ne_parse(ctx, NULL, -1); if (r != 1) return -1; @@ -1676,6 +1978,9 @@ nestegg_track_codec_id(nestegg * ctx, unsigned int track) if (strcmp(codec_id, TRACK_ID_VORBIS) == 0) return NESTEGG_CODEC_VORBIS; + if (strcmp(codec_id, TRACK_ID_OPUS) == 0) + return NESTEGG_CODEC_OPUS; + return -1; } @@ -1728,34 +2033,40 @@ nestegg_track_codec_data(nestegg * ctx, unsigned int track, unsigned int item, if (!entry) return -1; - if (nestegg_track_codec_id(ctx, track) != NESTEGG_CODEC_VORBIS) + if (nestegg_track_codec_id(ctx, track) != NESTEGG_CODEC_VORBIS + && nestegg_track_codec_id(ctx, track) != NESTEGG_CODEC_OPUS) return -1; if (ne_get_binary(entry->codec_private, &codec_private) != 0) return -1; - p = codec_private.data; - count = *p++ + 1; + if (nestegg_track_codec_id(ctx, track) == NESTEGG_CODEC_VORBIS) { + p = codec_private.data; + count = *p++ + 1; - if (count > 3) - return -1; + if (count > 3) + return -1; - i = 0; - total = 0; - while (--count) { - sizes[i] = ne_xiph_lace_value(&p); - total += sizes[i]; - i += 1; - } - sizes[i] = codec_private.length - total - (p - codec_private.data); + i = 0; + total = 0; + while (--count) { + sizes[i] = ne_xiph_lace_value(&p); + total += sizes[i]; + i += 1; + } + sizes[i] = codec_private.length - total - (p - codec_private.data); - for (i = 0; i < item; ++i) { - if (sizes[i] > LIMIT_FRAME) - return -1; - p += sizes[i]; + for (i = 0; i < item; ++i) { + if (sizes[i] > LIMIT_FRAME) + return -1; + p += sizes[i]; + } + *data = p; + *length = sizes[item]; + } else { + *data = codec_private.data; + *length = codec_private.length; } - *data = p; - *length = sizes[item]; return 0; } @@ -1776,37 +2087,43 @@ nestegg_track_video_params(nestegg * ctx, unsigned int track, if (nestegg_track_type(ctx, track) != NESTEGG_TRACK_VIDEO) return -1; + value = 0; + ne_get_uint(entry->video.stereo_mode, &value); + if (value <= NESTEGG_VIDEO_STEREO_TOP_BOTTOM || + value == NESTEGG_VIDEO_STEREO_RIGHT_LEFT) + params->stereo_mode = value; + if (ne_get_uint(entry->video.pixel_width, &value) != 0) return -1; - params->width = (unsigned int)value; + params->width = value; if (ne_get_uint(entry->video.pixel_height, &value) != 0) return -1; - params->height = (unsigned int)value; + params->height = value; value = 0; ne_get_uint(entry->video.pixel_crop_bottom, &value); - params->crop_bottom = (unsigned int)value; + params->crop_bottom = value; value = 0; ne_get_uint(entry->video.pixel_crop_top, &value); - params->crop_top = (unsigned int)value; + params->crop_top = value; value = 0; ne_get_uint(entry->video.pixel_crop_left, &value); - params->crop_left = (unsigned int)value; + params->crop_left = value; value = 0; ne_get_uint(entry->video.pixel_crop_right, &value); - params->crop_right = (unsigned int)value; + params->crop_right = value; value = params->width; ne_get_uint(entry->video.display_width, &value); - params->display_width = (unsigned int)value; + params->display_width = value; value = params->height; ne_get_uint(entry->video.display_height, &value); - params->display_height = (unsigned int)value; + params->display_height = value; return 0; } @@ -1832,11 +2149,19 @@ nestegg_track_audio_params(nestegg * ctx, unsigned int track, value = 1; ne_get_uint(entry->audio.channels, &value); - params->channels = (unsigned int)value; + params->channels = value; value = 16; ne_get_uint(entry->audio.bit_depth, &value); - params->depth = (unsigned int)value; + params->depth = value; + + value = 0; + ne_get_uint(entry->codec_delay, &value); + params->codec_delay = value; + + value = 0; + ne_get_uint(entry->seek_preroll, &value); + params->seek_preroll = value; return 0; } @@ -1854,19 +2179,26 @@ nestegg_read_packet(nestegg * ctx, nestegg_packet ** pkt) if (r != 1) return r; - /* any suspend fields must be handled here */ + /* Any DESC_FLAG_SUSPEND fields must be handled here. */ if (ne_is_suspend_element(id)) { r = ne_read_element(ctx, &id, &size); if (r != 1) return r; - /* the only suspend fields are blocks and simple blocks, which we + /* The only DESC_FLAG_SUSPEND fields are Blocks and SimpleBlocks, which we handle directly. */ r = ne_read_block(ctx, id, size, pkt); + if (r != 1) + return r; + + r = ne_read_discard_padding(ctx, *pkt); + if (r != 1) + return r; + return r; } - r = ne_parse(ctx, NULL); + r = ne_parse(ctx, NULL, -1); if (r != 1) return r; } @@ -1892,7 +2224,7 @@ nestegg_free_packet(nestegg_packet * pkt) int nestegg_packet_track(nestegg_packet * pkt, unsigned int * track) { - *track = (unsigned int)pkt->track; + *track = pkt->track; return 0; } @@ -1904,6 +2236,13 @@ nestegg_packet_tstamp(nestegg_packet * pkt, uint64_t * tstamp) } int +nestegg_packet_discard_padding(nestegg_packet * pkt, int64_t * discard_padding) +{ + *discard_padding = pkt->discard_padding; + return 0; +} + +int nestegg_packet_count(nestegg_packet * pkt, unsigned int * count) { struct frame * f = pkt->frame; @@ -1940,3 +2279,28 @@ nestegg_packet_data(nestegg_packet * pkt, unsigned int item, return -1; } + +int +nestegg_has_cues(nestegg * ctx) +{ + return ctx->segment.cues.cue_point.head || + ne_find_seek_for_id(ctx->segment.seek_head.head, ID_CUES); +} + +int +nestegg_sniff(unsigned char const * buffer, size_t length) +{ + nestegg_io io; + struct sniff_buffer user_data; + + user_data.buffer = buffer; + user_data.length = length; + user_data.offset = 0; + + io.read = ne_buffer_read; + io.seek = ne_buffer_seek; + io.tell = ne_buffer_tell; + io.userdata = &user_data; + return ne_match_webm(io, length); +} + diff --git a/nestegg/test/test.c b/third_party/nestegg/test/test.c index 210b640c7..cc0753de7 100644 --- a/nestegg/test/test.c +++ b/third_party/nestegg/test/test.c @@ -15,9 +15,10 @@ #define SEEK_TEST static int -stdio_read(void * p, size_t length, void * fp) +stdio_read(void * p, size_t length, void * file) { size_t r; + FILE * fp = file; r = fread(p, length, 1, fp); if (r == 0 && feof(fp)) @@ -26,8 +27,9 @@ stdio_read(void * p, size_t length, void * fp) } static int -stdio_seek(int64_t offset, int whence, void * fp) +stdio_seek(int64_t offset, int whence, void * file) { + FILE * fp = file; return fseek(fp, offset, whence); } @@ -43,7 +45,7 @@ log_callback(nestegg * ctx, unsigned int severity, char const * fmt, ...) va_list ap; char const * sev = NULL; -#ifndef DEBUG +#if !defined(DEBUG) if (severity < NESTEGG_LOG_WARNING) return; #endif @@ -102,32 +104,32 @@ main(int argc, char * argv[]) io.userdata = fp; ctx = NULL; - r = nestegg_init(&ctx, io, log_callback); + r = nestegg_init(&ctx, io, log_callback, -1); if (r != 0) return EXIT_FAILURE; nestegg_track_count(ctx, &tracks); nestegg_duration(ctx, &duration); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "media has %u tracks and duration %fs\n", tracks, duration / 1e9); #endif for (i = 0; i < tracks; ++i) { type = nestegg_track_type(ctx, i); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "track %u: type: %d codec: %d", i, type, nestegg_track_codec_id(ctx, i)); #endif nestegg_track_codec_data_count(ctx, i, &data_items); for (j = 0; j < data_items; ++j) { nestegg_track_codec_data(ctx, i, j, &codec_data, &length); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, " (%p, %u)", codec_data, (unsigned int) length); #endif } if (type == NESTEGG_TRACK_VIDEO) { nestegg_track_video_params(ctx, i, &vparams); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, " video: %ux%u (d: %ux%u %ux%ux%ux%u)", vparams.width, vparams.height, vparams.display_width, vparams.display_height, @@ -135,23 +137,23 @@ main(int argc, char * argv[]) #endif } else if (type == NESTEGG_TRACK_AUDIO) { nestegg_track_audio_params(ctx, i, &aparams); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, " audio: %.2fhz %u bit %u channels", aparams.rate, aparams.depth, aparams.channels); #endif } -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "\n"); #endif } -#ifdef SEEK_TEST -#ifdef DEBUG +#if defined(SEEK_TEST) +#if defined(DEBUG) fprintf(stderr, "seek to middle\n"); #endif r = nestegg_track_seek(ctx, 0, duration / 2); if (r == 0) { -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "middle "); #endif r = nestegg_read_packet(ctx, &pkt); @@ -159,23 +161,23 @@ main(int argc, char * argv[]) nestegg_packet_track(pkt, &track); nestegg_packet_count(pkt, &cnt); nestegg_packet_tstamp(pkt, &tstamp); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "* t %u pts %f frames %u\n", track, tstamp / 1e9, cnt); #endif nestegg_free_packet(pkt); } else { -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "middle seek failed\n"); #endif } } -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "seek to ~end\n"); #endif r = nestegg_track_seek(ctx, 0, duration - (duration / 10)); if (r == 0) { -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "end "); #endif r = nestegg_read_packet(ctx, &pkt); @@ -183,23 +185,23 @@ main(int argc, char * argv[]) nestegg_packet_track(pkt, &track); nestegg_packet_count(pkt, &cnt); nestegg_packet_tstamp(pkt, &tstamp); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "* t %u pts %f frames %u\n", track, tstamp / 1e9, cnt); #endif nestegg_free_packet(pkt); } else { -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "end seek failed\n"); #endif } } -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "seek to ~start\n"); #endif r = nestegg_track_seek(ctx, 0, duration / 10); if (r == 0) { -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "start "); #endif r = nestegg_read_packet(ctx, &pkt); @@ -207,12 +209,12 @@ main(int argc, char * argv[]) nestegg_packet_track(pkt, &track); nestegg_packet_count(pkt, &cnt); nestegg_packet_tstamp(pkt, &tstamp); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "* t %u pts %f frames %u\n", track, tstamp / 1e9, cnt); #endif nestegg_free_packet(pkt); } else { -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "start seek failed\n"); #endif } @@ -224,17 +226,17 @@ main(int argc, char * argv[]) nestegg_packet_count(pkt, &pkt_cnt); nestegg_packet_tstamp(pkt, &pkt_tstamp); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "t %u pts %f frames %u: ", pkt_track, pkt_tstamp / 1e9, pkt_cnt); #endif for (i = 0; i < pkt_cnt; ++i) { nestegg_packet_data(pkt, i, &ptr, &size); -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "%u ", (unsigned int) size); #endif } -#ifdef DEBUG +#if defined(DEBUG) fprintf(stderr, "\n"); #endif diff --git a/vp9/common/arm/neon/vp9_reconintra_neon.asm b/vp9/common/arm/neon/vp9_reconintra_neon.asm index 3aa15d2d0..dc9856fa8 100644 --- a/vp9/common/arm/neon/vp9_reconintra_neon.asm +++ b/vp9/common/arm/neon/vp9_reconintra_neon.asm @@ -315,8 +315,8 @@ loop_h vdup.u16 q2, r2 vadd.s16 q1, q1, q3 vadd.s16 q2, q2, q3 - vqshrun.s16 d0, q1, #0 - vqshrun.s16 d1, q2, #0 + vqmovun.s16 d0, q1 + vqmovun.s16 d1, q2 vst1.32 {d0[0]}, [r0], r1 vst1.32 {d1[0]}, [r0], r1 @@ -327,8 +327,8 @@ loop_h vdup.u16 q2, r2 vadd.s16 q1, q1, q3 vadd.s16 q2, q2, q3 - vqshrun.s16 d0, q1, #0 - vqshrun.s16 d1, q2, #0 + vqmovun.s16 d0, q1 + vqmovun.s16 d1, q2 vst1.32 {d0[0]}, [r0], r1 vst1.32 {d1[0]}, [r0], r1 bx lr @@ -372,10 +372,10 @@ loop_h vadd.s16 q8, q3, q8 vadd.s16 q9, q3, q9 - vqshrun.s16 d0, q0, #0 - vqshrun.s16 d1, q1, #0 - vqshrun.s16 d2, q8, #0 - vqshrun.s16 d3, q9, #0 + vqmovun.s16 d0, q0 + vqmovun.s16 d1, q1 + vqmovun.s16 d2, q8 + vqmovun.s16 d3, q9 vst1.64 {d0}, [r0], r1 vst1.64 {d1}, [r0], r1 @@ -394,10 +394,10 @@ loop_h vadd.s16 q8, q3, q8 vadd.s16 q9, q3, q9 - vqshrun.s16 d0, q0, #0 - vqshrun.s16 d1, q1, #0 - vqshrun.s16 d2, q8, #0 - vqshrun.s16 d3, q9, #0 + vqmovun.s16 d0, q0 + vqmovun.s16 d1, q1 + vqmovun.s16 d2, q8 + vqmovun.s16 d3, q9 vst1.64 {d0}, [r0], r1 vst1.64 {d1}, [r0], r1 @@ -445,10 +445,10 @@ loop_16x16_neon vadd.s16 q0, q0, q3 vadd.s16 q11, q8, q2 vadd.s16 q8, q8, q3 - vqshrun.s16 d2, q1, #0 - vqshrun.s16 d3, q0, #0 - vqshrun.s16 d22, q11, #0 - vqshrun.s16 d23, q8, #0 + vqmovun.s16 d2, q1 + vqmovun.s16 d3, q0 + vqmovun.s16 d22, q11 + vqmovun.s16 d23, q8 vdup.16 q0, d20[2] ; proload next 2 rows data vdup.16 q8, d20[3] vst1.64 {d2,d3}, [r0], r1 @@ -459,10 +459,10 @@ loop_16x16_neon vadd.s16 q0, q0, q3 vadd.s16 q11, q8, q2 vadd.s16 q8, q8, q3 - vqshrun.s16 d2, q1, #0 - vqshrun.s16 d3, q0, #0 - vqshrun.s16 d22, q11, #0 - vqshrun.s16 d23, q8, #0 + vqmovun.s16 d2, q1 + vqmovun.s16 d3, q0 + vqmovun.s16 d22, q11 + vqmovun.s16 d23, q8 vdup.16 q0, d21[0] ; proload next 2 rows data vdup.16 q8, d21[1] vst1.64 {d2,d3}, [r0], r1 @@ -472,10 +472,10 @@ loop_16x16_neon vadd.s16 q0, q0, q3 vadd.s16 q11, q8, q2 vadd.s16 q8, q8, q3 - vqshrun.s16 d2, q1, #0 - vqshrun.s16 d3, q0, #0 - vqshrun.s16 d22, q11, #0 - vqshrun.s16 d23, q8, #0 + vqmovun.s16 d2, q1 + vqmovun.s16 d3, q0 + vqmovun.s16 d22, q11 + vqmovun.s16 d23, q8 vdup.16 q0, d21[2] ; proload next 2 rows data vdup.16 q8, d21[3] vst1.64 {d2,d3}, [r0], r1 @@ -486,10 +486,10 @@ loop_16x16_neon vadd.s16 q0, q0, q3 vadd.s16 q11, q8, q2 vadd.s16 q8, q8, q3 - vqshrun.s16 d2, q1, #0 - vqshrun.s16 d3, q0, #0 - vqshrun.s16 d22, q11, #0 - vqshrun.s16 d23, q8, #0 + vqmovun.s16 d2, q1 + vqmovun.s16 d3, q0 + vqmovun.s16 d22, q11 + vqmovun.s16 d23, q8 vld1.8 {d18}, [r3]! ; preload 8 left into r12 vmovl.u8 q10, d18 vst1.64 {d2,d3}, [r0], r1 @@ -542,19 +542,19 @@ loop_32x32_neon vadd.s16 q13, q0, q9 vadd.s16 q14, q0, q10 vadd.s16 q15, q0, q11 - vqshrun.s16 d0, q12, #0 - vqshrun.s16 d1, q13, #0 + vqmovun.s16 d0, q12 + vqmovun.s16 d1, q13 vadd.s16 q12, q2, q8 vadd.s16 q13, q2, q9 - vqshrun.s16 d2, q14, #0 - vqshrun.s16 d3, q15, #0 + vqmovun.s16 d2, q14 + vqmovun.s16 d3, q15 vadd.s16 q14, q2, q10 vadd.s16 q15, q2, q11 vst1.64 {d0-d3}, [r0], r1 - vqshrun.s16 d24, q12, #0 - vqshrun.s16 d25, q13, #0 - vqshrun.s16 d26, q14, #0 - vqshrun.s16 d27, q15, #0 + vqmovun.s16 d24, q12 + vqmovun.s16 d25, q13 + vqmovun.s16 d26, q14 + vqmovun.s16 d27, q15 vdup.16 q1, d6[2] vdup.16 q2, d6[3] vst1.64 {d24-d27}, [r0], r1 @@ -564,19 +564,19 @@ loop_32x32_neon vadd.s16 q13, q1, q9 vadd.s16 q14, q1, q10 vadd.s16 q15, q1, q11 - vqshrun.s16 d0, q12, #0 - vqshrun.s16 d1, q13, #0 + vqmovun.s16 d0, q12 + vqmovun.s16 d1, q13 vadd.s16 q12, q2, q8 vadd.s16 q13, q2, q9 - vqshrun.s16 d2, q14, #0 - vqshrun.s16 d3, q15, #0 + vqmovun.s16 d2, q14 + vqmovun.s16 d3, q15 vadd.s16 q14, q2, q10 vadd.s16 q15, q2, q11 vst1.64 {d0-d3}, [r0], r1 - vqshrun.s16 d24, q12, #0 - vqshrun.s16 d25, q13, #0 - vqshrun.s16 d26, q14, #0 - vqshrun.s16 d27, q15, #0 + vqmovun.s16 d24, q12 + vqmovun.s16 d25, q13 + vqmovun.s16 d26, q14 + vqmovun.s16 d27, q15 vdup.16 q0, d7[0] vdup.16 q2, d7[1] vst1.64 {d24-d27}, [r0], r1 @@ -586,19 +586,19 @@ loop_32x32_neon vadd.s16 q13, q0, q9 vadd.s16 q14, q0, q10 vadd.s16 q15, q0, q11 - vqshrun.s16 d0, q12, #0 - vqshrun.s16 d1, q13, #0 + vqmovun.s16 d0, q12 + vqmovun.s16 d1, q13 vadd.s16 q12, q2, q8 vadd.s16 q13, q2, q9 - vqshrun.s16 d2, q14, #0 - vqshrun.s16 d3, q15, #0 + vqmovun.s16 d2, q14 + vqmovun.s16 d3, q15 vadd.s16 q14, q2, q10 vadd.s16 q15, q2, q11 vst1.64 {d0-d3}, [r0], r1 - vqshrun.s16 d24, q12, #0 - vqshrun.s16 d25, q13, #0 - vqshrun.s16 d26, q14, #0 - vqshrun.s16 d27, q15, #0 + vqmovun.s16 d24, q12 + vqmovun.s16 d25, q13 + vqmovun.s16 d26, q14 + vqmovun.s16 d27, q15 vdup.16 q0, d7[2] vdup.16 q2, d7[3] vst1.64 {d24-d27}, [r0], r1 @@ -608,20 +608,20 @@ loop_32x32_neon vadd.s16 q13, q0, q9 vadd.s16 q14, q0, q10 vadd.s16 q15, q0, q11 - vqshrun.s16 d0, q12, #0 - vqshrun.s16 d1, q13, #0 + vqmovun.s16 d0, q12 + vqmovun.s16 d1, q13 vadd.s16 q12, q2, q8 vadd.s16 q13, q2, q9 - vqshrun.s16 d2, q14, #0 - vqshrun.s16 d3, q15, #0 + vqmovun.s16 d2, q14 + vqmovun.s16 d3, q15 vadd.s16 q14, q2, q10 vadd.s16 q15, q2, q11 vst1.64 {d0-d3}, [r0], r1 - vqshrun.s16 d24, q12, #0 - vqshrun.s16 d25, q13, #0 + vqmovun.s16 d24, q12 + vqmovun.s16 d25, q13 vld1.8 {d0}, [r3]! ; preload 8 left pixels - vqshrun.s16 d26, q14, #0 - vqshrun.s16 d27, q15, #0 + vqmovun.s16 d26, q14 + vqmovun.s16 d27, q15 vmovl.u8 q3, d0 vst1.64 {d24-d27}, [r0], r1 diff --git a/vp9/common/vp9_entropy.c b/vp9/common/vp9_entropy.c index 13e954efe..bc12f9aa2 100644 --- a/vp9/common/vp9_entropy.c +++ b/vp9/common/vp9_entropy.c @@ -16,7 +16,7 @@ #include "vpx/vpx_integer.h" -DECLARE_ALIGNED(16, const uint8_t, vp9_coefband_trans_8x8plus[1024]) = { +const uint8_t vp9_coefband_trans_8x8plus[1024] = { 0, 1, 1, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 5, // beyond MAXBAND_INDEX+1 all values are filled as 5 @@ -85,11 +85,11 @@ DECLARE_ALIGNED(16, const uint8_t, vp9_coefband_trans_8x8plus[1024]) = { 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, }; -DECLARE_ALIGNED(16, const uint8_t, vp9_coefband_trans_4x4[16]) = { +const uint8_t vp9_coefband_trans_4x4[16] = { 0, 1, 1, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 5, 5, 5, }; -DECLARE_ALIGNED(16, const uint8_t, vp9_pt_energy_class[ENTROPY_TOKENS]) = { +const uint8_t vp9_pt_energy_class[ENTROPY_TOKENS] = { 0, 1, 2, 3, 3, 4, 4, 5, 5, 5, 5, 5 }; diff --git a/vp9/common/vp9_entropy.h b/vp9/common/vp9_entropy.h index d6b380fd5..aab8b5388 100644 --- a/vp9/common/vp9_entropy.h +++ b/vp9/common/vp9_entropy.h @@ -42,7 +42,7 @@ extern "C" { #define ENTROPY_NODES 11 -extern DECLARE_ALIGNED(16, const uint8_t, vp9_pt_energy_class[ENTROPY_TOKENS]); +DECLARE_ALIGNED(16, extern const uint8_t, vp9_pt_energy_class[ENTROPY_TOKENS]); #define EOB_MODEL_TOKEN 3 extern const vp9_tree_index vp9_coefmodel_tree[]; @@ -116,8 +116,8 @@ static INLINE void reset_skip_context(MACROBLOCKD *xd, BLOCK_SIZE bsize) { // This macro is currently unused but may be used by certain implementations #define MAXBAND_INDEX 21 -extern DECLARE_ALIGNED(16, const uint8_t, vp9_coefband_trans_8x8plus[1024]); -extern DECLARE_ALIGNED(16, const uint8_t, vp9_coefband_trans_4x4[16]); +DECLARE_ALIGNED(16, extern const uint8_t, vp9_coefband_trans_8x8plus[1024]); +DECLARE_ALIGNED(16, extern const uint8_t, vp9_coefband_trans_4x4[16]); static INLINE const uint8_t *get_band_translate(TX_SIZE tx_size) { return tx_size == TX_4X4 ? vp9_coefband_trans_4x4 diff --git a/vp9/common/vp9_filter.c b/vp9/common/vp9_filter.c index 546f603b6..7474a88bc 100644 --- a/vp9/common/vp9_filter.c +++ b/vp9/common/vp9_filter.c @@ -10,12 +10,9 @@ #include <assert.h> -#include "vpx_ports/mem.h" - #include "vp9/common/vp9_filter.h" -DECLARE_ALIGNED(256, const InterpKernel, - vp9_bilinear_filters[SUBPEL_SHIFTS]) = { +const InterpKernel vp9_bilinear_filters[SUBPEL_SHIFTS] = { { 0, 0, 0, 128, 0, 0, 0, 0 }, { 0, 0, 0, 120, 8, 0, 0, 0 }, { 0, 0, 0, 112, 16, 0, 0, 0 }, @@ -35,8 +32,7 @@ DECLARE_ALIGNED(256, const InterpKernel, }; // Lagrangian interpolation filter -DECLARE_ALIGNED(256, const InterpKernel, - vp9_sub_pel_filters_8[SUBPEL_SHIFTS]) = { +const InterpKernel vp9_sub_pel_filters_8[SUBPEL_SHIFTS] = { { 0, 0, 0, 128, 0, 0, 0, 0}, { 0, 1, -5, 126, 8, -3, 1, 0}, { -1, 3, -10, 122, 18, -6, 2, 0}, @@ -56,8 +52,7 @@ DECLARE_ALIGNED(256, const InterpKernel, }; // DCT based filter -DECLARE_ALIGNED(256, const InterpKernel, - vp9_sub_pel_filters_8s[SUBPEL_SHIFTS]) = { +const InterpKernel vp9_sub_pel_filters_8s[SUBPEL_SHIFTS] = { {0, 0, 0, 128, 0, 0, 0, 0}, {-1, 3, -7, 127, 8, -3, 1, 0}, {-2, 5, -13, 125, 17, -6, 3, -1}, @@ -77,8 +72,7 @@ DECLARE_ALIGNED(256, const InterpKernel, }; // freqmultiplier = 0.5 -DECLARE_ALIGNED(256, const InterpKernel, - vp9_sub_pel_filters_8lp[SUBPEL_SHIFTS]) = { +const InterpKernel vp9_sub_pel_filters_8lp[SUBPEL_SHIFTS] = { { 0, 0, 0, 128, 0, 0, 0, 0}, {-3, -1, 32, 64, 38, 1, -3, 0}, {-2, -2, 29, 63, 41, 2, -3, 0}, diff --git a/vp9/common/vp9_filter.h b/vp9/common/vp9_filter.h index 15610d781..29d3867c9 100644 --- a/vp9/common/vp9_filter.h +++ b/vp9/common/vp9_filter.h @@ -13,6 +13,8 @@ #include "./vpx_config.h" #include "vpx/vpx_integer.h" +#include "vpx_ports/mem.h" + #ifdef __cplusplus extern "C" { @@ -37,10 +39,14 @@ typedef int16_t InterpKernel[SUBPEL_TAPS]; const InterpKernel *vp9_get_interp_kernel(INTERP_FILTER filter); -extern const InterpKernel vp9_bilinear_filters[SUBPEL_SHIFTS]; -extern const InterpKernel vp9_sub_pel_filters_8[SUBPEL_SHIFTS]; -extern const InterpKernel vp9_sub_pel_filters_8s[SUBPEL_SHIFTS]; -extern const InterpKernel vp9_sub_pel_filters_8lp[SUBPEL_SHIFTS]; +DECLARE_ALIGNED(256, extern const InterpKernel, + vp9_bilinear_filters[SUBPEL_SHIFTS]); +DECLARE_ALIGNED(256, extern const InterpKernel, + vp9_sub_pel_filters_8[SUBPEL_SHIFTS]); +DECLARE_ALIGNED(256, extern const InterpKernel, + vp9_sub_pel_filters_8s[SUBPEL_SHIFTS]); +DECLARE_ALIGNED(256, extern const InterpKernel, + vp9_sub_pel_filters_8lp[SUBPEL_SHIFTS]); // The VP9_BILINEAR_FILTERS_2TAP macro returns a pointer to the bilinear // filter kernel as a 2 tap filter. diff --git a/vp9/common/vp9_loopfilter.c b/vp9/common/vp9_loopfilter.c index 04f8934c4..868a66ae4 100644 --- a/vp9/common/vp9_loopfilter.c +++ b/vp9/common/vp9_loopfilter.c @@ -262,9 +262,9 @@ void vp9_loop_filter_frame_init(VP9_COMMON *cm, int default_filt_lvl) { int lvl_seg = default_filt_lvl; if (vp9_segfeature_active(seg, seg_id, SEG_LVL_ALT_LF)) { const int data = vp9_get_segdata(seg, seg_id, SEG_LVL_ALT_LF); - lvl_seg = seg->abs_delta == SEGMENT_ABSDATA - ? data - : clamp(default_filt_lvl + data, 0, MAX_LOOP_FILTER); + lvl_seg = clamp(seg->abs_delta == SEGMENT_ABSDATA ? + data : default_filt_lvl + data, + 0, MAX_LOOP_FILTER); } if (!lf->mode_ref_delta_enabled) { @@ -868,7 +868,6 @@ void vp9_setup_mask(VP9_COMMON *const cm, const int mi_row, const int mi_col, assert(!(lfm->int_4x4_uv & lfm->above_uv[TX_16X16])); } -#if CONFIG_NON420 static uint8_t build_lfi(const loop_filter_info_n *lfi_n, const MB_MODE_INFO *mbmi) { const int seg = mbmi->segment_id; @@ -1046,7 +1045,6 @@ static void filter_block_plane_non420(VP9_COMMON *cm, dst->buf += 8 * dst->stride; } } -#endif void vp9_filter_block_plane(VP9_COMMON *const cm, struct macroblockd_plane *const plane, @@ -1206,10 +1204,8 @@ void vp9_loop_filter_rows(const YV12_BUFFER_CONFIG *frame_buffer, const int num_planes = y_only ? 1 : MAX_MB_PLANE; int mi_row, mi_col; LOOP_FILTER_MASK lfm; -#if CONFIG_NON420 int use_420 = y_only || (xd->plane[1].subsampling_y == 1 && xd->plane[1].subsampling_x == 1); -#endif for (mi_row = start; mi_row < stop; mi_row += MI_BLOCK_SIZE) { MODE_INFO **mi_8x8 = cm->mi_grid_visible + mi_row * cm->mode_info_stride; @@ -1220,22 +1216,16 @@ void vp9_loop_filter_rows(const YV12_BUFFER_CONFIG *frame_buffer, setup_dst_planes(xd, frame_buffer, mi_row, mi_col); // TODO(JBB): Make setup_mask work for non 420. -#if CONFIG_NON420 if (use_420) -#endif vp9_setup_mask(cm, mi_row, mi_col, mi_8x8 + mi_col, cm->mode_info_stride, &lfm); for (plane = 0; plane < num_planes; ++plane) { -#if CONFIG_NON420 if (use_420) -#endif vp9_filter_block_plane(cm, &xd->plane[plane], mi_row, &lfm); -#if CONFIG_NON420 else filter_block_plane_non420(cm, &xd->plane[plane], mi_8x8 + mi_col, mi_row, mi_col); -#endif } } } diff --git a/vp9/common/vp9_mvref_common.c b/vp9/common/vp9_mvref_common.c index 3b0d1e982..e5f3fed45 100644 --- a/vp9/common/vp9_mvref_common.c +++ b/vp9/common/vp9_mvref_common.c @@ -186,13 +186,12 @@ static INLINE int is_inside(const TileInfo *const tile, // This function searches the neighbourhood of a given MB/SB // to try and find candidate reference vectors. -void vp9_find_mv_refs_idx(const VP9_COMMON *cm, const MACROBLOCKD *xd, - const TileInfo *const tile, - MODE_INFO *mi, const MODE_INFO *prev_mi, - MV_REFERENCE_FRAME ref_frame, - int_mv *mv_ref_list, - int block_idx, - int mi_row, int mi_col) { +static void find_mv_refs_idx(const VP9_COMMON *cm, const MACROBLOCKD *xd, + const TileInfo *const tile, + MODE_INFO *mi, const MODE_INFO *prev_mi, + MV_REFERENCE_FRAME ref_frame, + int_mv *mv_ref_list, + int block_idx, int mi_row, int mi_col) { const int *ref_sign_bias = cm->ref_frame_sign_bias; int i, refmv_count = 0; const POSITION *const mv_ref_search = mv_ref_blocks[mi->mbmi.sb_type]; @@ -291,6 +290,16 @@ void vp9_find_mv_refs_idx(const VP9_COMMON *cm, const MACROBLOCKD *xd, clamp_mv_ref(&mv_ref_list[i].as_mv, xd); } +void vp9_find_mv_refs(const VP9_COMMON *cm, const MACROBLOCKD *xd, + const TileInfo *const tile, + MODE_INFO *mi, const MODE_INFO *prev_mi, + MV_REFERENCE_FRAME ref_frame, + int_mv *mv_ref_list, + int mi_row, int mi_col) { + find_mv_refs_idx(cm, xd, tile, mi, prev_mi, ref_frame, mv_ref_list, -1, + mi_row, mi_col); +} + static void lower_mv_precision(MV *mv, int allow_hp) { const int use_hp = allow_hp && vp9_use_mv_hp(mv); if (!use_hp) { @@ -325,8 +334,8 @@ void vp9_append_sub8x8_mvs_for_idx(VP9_COMMON *cm, MACROBLOCKD *xd, assert(MAX_MV_REF_CANDIDATES == 2); - vp9_find_mv_refs_idx(cm, xd, tile, mi, xd->last_mi, mi->mbmi.ref_frame[ref], - mv_list, block, mi_row, mi_col); + find_mv_refs_idx(cm, xd, tile, mi, xd->last_mi, mi->mbmi.ref_frame[ref], + mv_list, block, mi_row, mi_col); near->as_int = 0; switch (block) { diff --git a/vp9/common/vp9_mvref_common.h b/vp9/common/vp9_mvref_common.h index f99952f3c..04cb000ef 100644 --- a/vp9/common/vp9_mvref_common.h +++ b/vp9/common/vp9_mvref_common.h @@ -17,29 +17,24 @@ extern "C" { #endif +#define LEFT_TOP_MARGIN ((VP9_ENC_BORDER_IN_PIXELS - VP9_INTERP_EXTEND) << 3) +#define RIGHT_BOTTOM_MARGIN ((VP9_ENC_BORDER_IN_PIXELS -\ + VP9_INTERP_EXTEND) << 3) -void vp9_find_mv_refs_idx(const VP9_COMMON *cm, const MACROBLOCKD *xd, - const TileInfo *const tile, - MODE_INFO *mi, const MODE_INFO *prev_mi, - MV_REFERENCE_FRAME ref_frame, - int_mv *mv_ref_list, - int block_idx, - int mi_row, int mi_col); - -static INLINE void vp9_find_mv_refs(const VP9_COMMON *cm, const MACROBLOCKD *xd, - const TileInfo *const tile, - MODE_INFO *mi, const MODE_INFO *prev_mi, - MV_REFERENCE_FRAME ref_frame, - int_mv *mv_ref_list, - int mi_row, int mi_col) { - vp9_find_mv_refs_idx(cm, xd, tile, mi, prev_mi, ref_frame, - mv_ref_list, -1, mi_row, mi_col); +// TODO(jingning): this mv clamping function should be block size dependent. +static INLINE void clamp_mv2(MV *mv, const MACROBLOCKD *xd) { + clamp_mv(mv, xd->mb_to_left_edge - LEFT_TOP_MARGIN, + xd->mb_to_right_edge + RIGHT_BOTTOM_MARGIN, + xd->mb_to_top_edge - LEFT_TOP_MARGIN, + xd->mb_to_bottom_edge + RIGHT_BOTTOM_MARGIN); } -#define LEFT_TOP_MARGIN ((VP9_ENC_BORDER_IN_PIXELS \ - - VP9_INTERP_EXTEND) << 3) -#define RIGHT_BOTTOM_MARGIN ((VP9_ENC_BORDER_IN_PIXELS \ - - VP9_INTERP_EXTEND) << 3) +void vp9_find_mv_refs(const VP9_COMMON *cm, const MACROBLOCKD *xd, + const TileInfo *const tile, + MODE_INFO *mi, const MODE_INFO *prev_mi, + MV_REFERENCE_FRAME ref_frame, + int_mv *mv_ref_list, + int mi_row, int mi_col); // check a list of motion vectors by sad score using a number rows of pixels // above and a number cols of pixels in the left to select the one with best @@ -47,14 +42,6 @@ static INLINE void vp9_find_mv_refs(const VP9_COMMON *cm, const MACROBLOCKD *xd, void vp9_find_best_ref_mvs(MACROBLOCKD *xd, int allow_hp, int_mv *mvlist, int_mv *nearest, int_mv *near); -// TODO(jingning): this mv clamping function should be block size dependent. -static INLINE void clamp_mv2(MV *mv, const MACROBLOCKD *xd) { - clamp_mv(mv, xd->mb_to_left_edge - LEFT_TOP_MARGIN, - xd->mb_to_right_edge + RIGHT_BOTTOM_MARGIN, - xd->mb_to_top_edge - LEFT_TOP_MARGIN, - xd->mb_to_bottom_edge + RIGHT_BOTTOM_MARGIN); -} - void vp9_append_sub8x8_mvs_for_idx(VP9_COMMON *cm, MACROBLOCKD *xd, const TileInfo *const tile, int block, int ref, int mi_row, int mi_col, diff --git a/vp9/common/vp9_onyx.h b/vp9/common/vp9_onyx.h index ac39a98fd..ab27ca523 100644 --- a/vp9/common/vp9_onyx.h +++ b/vp9/common/vp9_onyx.h @@ -112,7 +112,6 @@ extern "C" { int auto_key; // autodetect cut scenes and set the keyframes int key_freq; // maximum distance to key frame. - int allow_lag; // allow lagged compression (if 0 lagin frames is ignored) int lag_in_frames; // how many frames lag before we start encoding // ---------------------------------------------------------------- diff --git a/vp9/common/vp9_pred_common.h b/vp9/common/vp9_pred_common.h index 7f9e56392..6c7a0d383 100644 --- a/vp9/common/vp9_pred_common.h +++ b/vp9/common/vp9_pred_common.h @@ -39,7 +39,7 @@ static INLINE int vp9_get_pred_context_seg_id(const MACROBLOCKD *xd) { return above_sip + left_sip; } -static INLINE vp9_prob vp9_get_pred_prob_seg_id(struct segmentation *seg, +static INLINE vp9_prob vp9_get_pred_prob_seg_id(const struct segmentation *seg, const MACROBLOCKD *xd) { return seg->pred_probs[vp9_get_pred_context_seg_id(xd)]; } diff --git a/vp9/common/vp9_prob.c b/vp9/common/vp9_prob.c index f9bc06ecf..a1befc63e 100644 --- a/vp9/common/vp9_prob.c +++ b/vp9/common/vp9_prob.c @@ -10,7 +10,7 @@ #include "vp9/common/vp9_prob.h" -DECLARE_ALIGNED(16, const uint8_t, vp9_norm[256]) = { +const uint8_t vp9_norm[256] = { 0, 7, 6, 6, 5, 5, 5, 5, 4, 4, 4, 4, 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, diff --git a/vp9/common/vp9_quant_common.c b/vp9/common/vp9_quant_common.c index 9fef8b1ef..def12554d 100644 --- a/vp9/common/vp9_quant_common.c +++ b/vp9/common/vp9_quant_common.c @@ -134,9 +134,9 @@ int vp9_get_qindex(const struct segmentation *seg, int segment_id, int base_qindex) { if (vp9_segfeature_active(seg, segment_id, SEG_LVL_ALT_Q)) { const int data = vp9_get_segdata(seg, segment_id, SEG_LVL_ALT_Q); - return seg->abs_delta == SEGMENT_ABSDATA ? - data : // Abs value - clamp(base_qindex + data, 0, MAXQ); // Delta value + const int seg_qindex = seg->abs_delta == SEGMENT_ABSDATA ? + data : base_qindex + data; + return clamp(seg_qindex, 0, MAXQ); } else { return base_qindex; } diff --git a/vp9/common/vp9_rtcd_defs.sh b/vp9/common/vp9_rtcd_defs.sh index 63171033c..a18ae9bab 100644 --- a/vp9/common/vp9_rtcd_defs.sh +++ b/vp9/common/vp9_rtcd_defs.sh @@ -386,7 +386,7 @@ prototype unsigned int vp9_variance4x4 "const uint8_t *src_ptr, int source_strid specialize vp9_variance4x4 mmx $sse2_x86inc prototype unsigned int vp9_sub_pixel_variance64x64 "const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse" -specialize vp9_sub_pixel_variance64x64 $sse2_x86inc $ssse3_x86inc +specialize vp9_sub_pixel_variance64x64 $sse2_x86inc $ssse3_x86inc avx2 prototype unsigned int vp9_sub_pixel_avg_variance64x64 "const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred" specialize vp9_sub_pixel_avg_variance64x64 $sse2_x86inc $ssse3_x86inc @@ -416,7 +416,7 @@ prototype unsigned int vp9_sub_pixel_avg_variance16x32 "const uint8_t *src_ptr, specialize vp9_sub_pixel_avg_variance16x32 $sse2_x86inc $ssse3_x86inc prototype unsigned int vp9_sub_pixel_variance32x32 "const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse" -specialize vp9_sub_pixel_variance32x32 $sse2_x86inc $ssse3_x86inc +specialize vp9_sub_pixel_variance32x32 $sse2_x86inc $ssse3_x86inc avx2 prototype unsigned int vp9_sub_pixel_avg_variance32x32 "const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred" specialize vp9_sub_pixel_avg_variance32x32 $sse2_x86inc $ssse3_x86inc diff --git a/vp9/common/x86/vp9_asm_stubs.c b/vp9/common/x86/vp9_asm_stubs.c index a2cf910a4..1b4904c39 100644 --- a/vp9/common/x86/vp9_asm_stubs.c +++ b/vp9/common/x86/vp9_asm_stubs.c @@ -142,20 +142,29 @@ void vp9_convolve8_##avg##opt(const uint8_t *src, ptrdiff_t src_stride, \ #if HAVE_AVX2 filter8_1dfunction vp9_filter_block1d16_v8_avx2; filter8_1dfunction vp9_filter_block1d16_h8_avx2; +filter8_1dfunction vp9_filter_block1d4_v8_ssse3; +#if (ARCH_X86_64) +filter8_1dfunction vp9_filter_block1d8_v8_intrin_ssse3; +filter8_1dfunction vp9_filter_block1d8_h8_intrin_ssse3; +filter8_1dfunction vp9_filter_block1d4_h8_intrin_ssse3; +#define vp9_filter_block1d8_v8_avx2 vp9_filter_block1d8_v8_intrin_ssse3 +#define vp9_filter_block1d8_h8_avx2 vp9_filter_block1d8_h8_intrin_ssse3 +#define vp9_filter_block1d4_h8_avx2 vp9_filter_block1d4_h8_intrin_ssse3 +#else filter8_1dfunction vp9_filter_block1d8_v8_ssse3; filter8_1dfunction vp9_filter_block1d8_h8_ssse3; -filter8_1dfunction vp9_filter_block1d4_v8_ssse3; filter8_1dfunction vp9_filter_block1d4_h8_ssse3; +#define vp9_filter_block1d8_v8_avx2 vp9_filter_block1d8_v8_ssse3 +#define vp9_filter_block1d8_h8_avx2 vp9_filter_block1d8_h8_ssse3 +#define vp9_filter_block1d4_h8_avx2 vp9_filter_block1d4_h8_ssse3 +#endif filter8_1dfunction vp9_filter_block1d16_v2_ssse3; filter8_1dfunction vp9_filter_block1d16_h2_ssse3; filter8_1dfunction vp9_filter_block1d8_v2_ssse3; filter8_1dfunction vp9_filter_block1d8_h2_ssse3; filter8_1dfunction vp9_filter_block1d4_v2_ssse3; filter8_1dfunction vp9_filter_block1d4_h2_ssse3; -#define vp9_filter_block1d8_v8_avx2 vp9_filter_block1d8_v8_ssse3 -#define vp9_filter_block1d8_h8_avx2 vp9_filter_block1d8_h8_ssse3 #define vp9_filter_block1d4_v8_avx2 vp9_filter_block1d4_v8_ssse3 -#define vp9_filter_block1d4_h8_avx2 vp9_filter_block1d4_h8_ssse3 #define vp9_filter_block1d16_v2_avx2 vp9_filter_block1d16_v2_ssse3 #define vp9_filter_block1d16_h2_avx2 vp9_filter_block1d16_h2_ssse3 #define vp9_filter_block1d8_v2_avx2 vp9_filter_block1d8_v2_ssse3 @@ -183,12 +192,26 @@ FUN_CONV_1D(vert, y_step_q4, filter_y, v, src - src_stride * 3, , avx2); FUN_CONV_2D(, avx2); #endif #if HAVE_SSSE3 +#if (ARCH_X86_64) +filter8_1dfunction vp9_filter_block1d16_v8_intrin_ssse3; +filter8_1dfunction vp9_filter_block1d16_h8_intrin_ssse3; +filter8_1dfunction vp9_filter_block1d8_v8_intrin_ssse3; +filter8_1dfunction vp9_filter_block1d8_h8_intrin_ssse3; +filter8_1dfunction vp9_filter_block1d4_v8_ssse3; +filter8_1dfunction vp9_filter_block1d4_h8_intrin_ssse3; +#define vp9_filter_block1d16_v8_ssse3 vp9_filter_block1d16_v8_intrin_ssse3 +#define vp9_filter_block1d16_h8_ssse3 vp9_filter_block1d16_h8_intrin_ssse3 +#define vp9_filter_block1d8_v8_ssse3 vp9_filter_block1d8_v8_intrin_ssse3 +#define vp9_filter_block1d8_h8_ssse3 vp9_filter_block1d8_h8_intrin_ssse3 +#define vp9_filter_block1d4_h8_ssse3 vp9_filter_block1d4_h8_intrin_ssse3 +#else filter8_1dfunction vp9_filter_block1d16_v8_ssse3; filter8_1dfunction vp9_filter_block1d16_h8_ssse3; filter8_1dfunction vp9_filter_block1d8_v8_ssse3; filter8_1dfunction vp9_filter_block1d8_h8_ssse3; filter8_1dfunction vp9_filter_block1d4_v8_ssse3; filter8_1dfunction vp9_filter_block1d4_h8_ssse3; +#endif filter8_1dfunction vp9_filter_block1d16_v8_avg_ssse3; filter8_1dfunction vp9_filter_block1d16_h8_avg_ssse3; filter8_1dfunction vp9_filter_block1d8_v8_avg_ssse3; diff --git a/vp9/common/x86/vp9_subpixel_8t_intrin_avx2.c b/vp9/common/x86/vp9_subpixel_8t_intrin_avx2.c index 0ffb1bce3..efa960c66 100644 --- a/vp9/common/x86/vp9_subpixel_8t_intrin_avx2.c +++ b/vp9/common/x86/vp9_subpixel_8t_intrin_avx2.c @@ -12,22 +12,25 @@ #include "vpx_ports/mem.h" // filters for 16_h8 and 16_v8 -DECLARE_ALIGNED(32, const unsigned char, filt1_global_avx2[32])= { +DECLARE_ALIGNED(32, static const uint8_t, filt1_global_avx2[32]) = { 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, - 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8}; + 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8 +}; -DECLARE_ALIGNED(32, const unsigned char, filt2_global_avx2[32])= { +DECLARE_ALIGNED(32, static const uint8_t, filt2_global_avx2[32]) = { 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, - 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10}; + 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10 +}; -DECLARE_ALIGNED(32, const unsigned char, filt3_global_avx2[32])= { +DECLARE_ALIGNED(32, static const uint8_t, filt3_global_avx2[32]) = { 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, - 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12}; + 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12 +}; -DECLARE_ALIGNED(32, const unsigned char, filt4_global_avx2[32])= { +DECLARE_ALIGNED(32, static const uint8_t, filt4_global_avx2[32]) = { 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13, 14, - 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13, 14}; - + 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13, 14 +}; void vp9_filter_block1d16_h8_avx2(unsigned char *src_ptr, unsigned int src_pixels_per_line, diff --git a/vp9/common/x86/vp9_subpixel_8t_intrin_ssse3.c b/vp9/common/x86/vp9_subpixel_8t_intrin_ssse3.c new file mode 100644 index 000000000..cf28d8d2b --- /dev/null +++ b/vp9/common/x86/vp9_subpixel_8t_intrin_ssse3.c @@ -0,0 +1,490 @@ +/* + * Copyright (c) 2010 The WebM project authors. All Rights Reserved. + * + * Use of this source code is governed by a BSD-style license + * that can be found in the LICENSE file in the root of the source + * tree. An additional intellectual property rights grant can be found + * in the file PATENTS. All contributing project authors may + * be found in the AUTHORS file in the root of the source tree. + */ + +#include <tmmintrin.h> +#include "vpx_ports/mem.h" +#include "vpx_ports/emmintrin_compat.h" + +// filters only for the 4_h8 convolution +DECLARE_ALIGNED(16, static const uint8_t, filt1_4_h8[16]) = { + 0, 1, 1, 2, 2, 3, 3, 4, 2, 3, 3, 4, 4, 5, 5, 6 +}; + +DECLARE_ALIGNED(16, static const uint8_t, filt2_4_h8[16]) = { + 4, 5, 5, 6, 6, 7, 7, 8, 6, 7, 7, 8, 8, 9, 9, 10 +}; + +// filters for 8_h8 and 16_h8 +DECLARE_ALIGNED(16, static const uint8_t, filt1_global[16]) = { + 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8 +}; + +DECLARE_ALIGNED(16, static const uint8_t, filt2_global[16]) = { + 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10 +}; + +DECLARE_ALIGNED(16, static const uint8_t, filt3_global[16]) = { + 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12 +}; + +DECLARE_ALIGNED(16, static const uint8_t, filt4_global[16]) = { + 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13, 14 +}; + +void vp9_filter_block1d4_h8_intrin_ssse3(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + unsigned int output_pitch, + unsigned int output_height, + int16_t *filter) { + __m128i firstFilters, secondFilters, thirdFilters, forthFilters; + __m128i srcRegFilt1, srcRegFilt2, srcRegFilt3, srcRegFilt4; + __m128i addFilterReg64, filtersReg, srcReg, minReg; + unsigned int i; + + // create a register with 0,64,0,64,0,64,0,64,0,64,0,64,0,64,0,64 + addFilterReg64 =_mm_set1_epi32((int)0x0400040u); + filtersReg = _mm_loadu_si128((__m128i *)filter); + // converting the 16 bit (short) to 8 bit (byte) and have the same data + // in both lanes of 128 bit register. + filtersReg =_mm_packs_epi16(filtersReg, filtersReg); + + // duplicate only the first 16 bits in the filter into the first lane + firstFilters = _mm_shufflelo_epi16(filtersReg, 0); + // duplicate only the third 16 bit in the filter into the first lane + secondFilters = _mm_shufflelo_epi16(filtersReg, 0xAAu); + // duplicate only the seconds 16 bits in the filter into the second lane + firstFilters = _mm_shufflehi_epi16(firstFilters, 0x55u); + // duplicate only the forth 16 bits in the filter into the second lane + secondFilters = _mm_shufflehi_epi16(secondFilters, 0xFFu); + + // loading the local filters + thirdFilters =_mm_load_si128((__m128i const *)filt1_4_h8); + forthFilters = _mm_load_si128((__m128i const *)filt2_4_h8); + + for (i = 0; i < output_height; i++) { + srcReg = _mm_loadu_si128((__m128i *)(src_ptr-3)); + + // filter the source buffer + srcRegFilt1= _mm_shuffle_epi8(srcReg, thirdFilters); + srcRegFilt2= _mm_shuffle_epi8(srcReg, forthFilters); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt1 = _mm_maddubs_epi16(srcRegFilt1, firstFilters); + srcRegFilt2 = _mm_maddubs_epi16(srcRegFilt2, secondFilters); + + // extract the higher half of the lane + srcRegFilt3 = _mm_srli_si128(srcRegFilt1, 8); + srcRegFilt4 = _mm_srli_si128(srcRegFilt2, 8); + + minReg = _mm_min_epi16(srcRegFilt3, srcRegFilt2); + + // add and saturate all the results together + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, srcRegFilt4); + srcRegFilt3 = _mm_max_epi16(srcRegFilt3, srcRegFilt2); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, minReg); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, srcRegFilt3); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, addFilterReg64); + + // shift by 7 bit each 16 bits + srcRegFilt1 = _mm_srai_epi16(srcRegFilt1, 7); + + // shrink to 8 bit each 16 bits + srcRegFilt1 = _mm_packus_epi16(srcRegFilt1, srcRegFilt1); + src_ptr+=src_pixels_per_line; + + // save only 4 bytes + *((int*)&output_ptr[0])= _mm_cvtsi128_si32(srcRegFilt1); + + output_ptr+=output_pitch; + } +} + +void vp9_filter_block1d8_h8_intrin_ssse3(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + unsigned int output_pitch, + unsigned int output_height, + int16_t *filter) { + __m128i firstFilters, secondFilters, thirdFilters, forthFilters, srcReg; + __m128i filt1Reg, filt2Reg, filt3Reg, filt4Reg; + __m128i srcRegFilt1, srcRegFilt2, srcRegFilt3, srcRegFilt4; + __m128i addFilterReg64, filtersReg, minReg; + unsigned int i; + + // create a register with 0,64,0,64,0,64,0,64,0,64,0,64,0,64,0,64 + addFilterReg64 = _mm_set1_epi32((int)0x0400040u); + filtersReg = _mm_loadu_si128((__m128i *)filter); + // converting the 16 bit (short) to 8 bit (byte) and have the same data + // in both lanes of 128 bit register. + filtersReg =_mm_packs_epi16(filtersReg, filtersReg); + + // duplicate only the first 16 bits (first and second byte) + // across 128 bit register + firstFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x100u)); + // duplicate only the second 16 bits (third and forth byte) + // across 128 bit register + secondFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x302u)); + // duplicate only the third 16 bits (fifth and sixth byte) + // across 128 bit register + thirdFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x504u)); + // duplicate only the forth 16 bits (seventh and eighth byte) + // across 128 bit register + forthFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x706u)); + + filt1Reg = _mm_load_si128((__m128i const *)filt1_global); + filt2Reg = _mm_load_si128((__m128i const *)filt2_global); + filt3Reg = _mm_load_si128((__m128i const *)filt3_global); + filt4Reg = _mm_load_si128((__m128i const *)filt4_global); + + for (i = 0; i < output_height; i++) { + srcReg = _mm_loadu_si128((__m128i *)(src_ptr-3)); + + // filter the source buffer + srcRegFilt1= _mm_shuffle_epi8(srcReg, filt1Reg); + srcRegFilt2= _mm_shuffle_epi8(srcReg, filt2Reg); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt1 = _mm_maddubs_epi16(srcRegFilt1, firstFilters); + srcRegFilt2 = _mm_maddubs_epi16(srcRegFilt2, secondFilters); + + // filter the source buffer + srcRegFilt3= _mm_shuffle_epi8(srcReg, filt3Reg); + srcRegFilt4= _mm_shuffle_epi8(srcReg, filt4Reg); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt3 = _mm_maddubs_epi16(srcRegFilt3, thirdFilters); + srcRegFilt4 = _mm_maddubs_epi16(srcRegFilt4, forthFilters); + + // add and saturate all the results together + minReg = _mm_min_epi16(srcRegFilt4, srcRegFilt3); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, srcRegFilt2); + + srcRegFilt4= _mm_max_epi16(srcRegFilt4, srcRegFilt3); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, minReg); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, srcRegFilt4); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, addFilterReg64); + + // shift by 7 bit each 16 bits + srcRegFilt1 = _mm_srai_epi16(srcRegFilt1, 7); + + // shrink to 8 bit each 16 bits + srcRegFilt1 = _mm_packus_epi16(srcRegFilt1, srcRegFilt1); + + src_ptr+=src_pixels_per_line; + + // save only 8 bytes + _mm_storel_epi64((__m128i*)&output_ptr[0], srcRegFilt1); + + output_ptr+=output_pitch; + } +} + +void vp9_filter_block1d16_h8_intrin_ssse3(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + unsigned int output_pitch, + unsigned int output_height, + int16_t *filter) { + __m128i addFilterReg64, filtersReg, srcReg1, srcReg2; + __m128i filt1Reg, filt2Reg, filt3Reg, filt4Reg; + __m128i firstFilters, secondFilters, thirdFilters, forthFilters; + __m128i srcRegFilt1_1, srcRegFilt2_1, srcRegFilt2, srcRegFilt3; + unsigned int i; + + // create a register with 0,64,0,64,0,64,0,64,0,64,0,64,0,64,0,64 + addFilterReg64 = _mm_set1_epi32((int)0x0400040u); + filtersReg = _mm_loadu_si128((__m128i *)filter); + // converting the 16 bit (short) to 8 bit (byte) and have the same data + // in both lanes of 128 bit register. + filtersReg =_mm_packs_epi16(filtersReg, filtersReg); + + // duplicate only the first 16 bits (first and second byte) + // across 128 bit register + firstFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x100u)); + // duplicate only the second 16 bits (third and forth byte) + // across 128 bit register + secondFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x302u)); + // duplicate only the third 16 bits (fifth and sixth byte) + // across 128 bit register + thirdFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x504u)); + // duplicate only the forth 16 bits (seventh and eighth byte) + // across 128 bit register + forthFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x706u)); + + filt1Reg = _mm_load_si128((__m128i const *)filt1_global); + filt2Reg = _mm_load_si128((__m128i const *)filt2_global); + filt3Reg = _mm_load_si128((__m128i const *)filt3_global); + filt4Reg = _mm_load_si128((__m128i const *)filt4_global); + + for (i = 0; i < output_height; i++) { + srcReg1 = _mm_loadu_si128((__m128i *)(src_ptr-3)); + + // filter the source buffer + srcRegFilt1_1= _mm_shuffle_epi8(srcReg1, filt1Reg); + srcRegFilt2= _mm_shuffle_epi8(srcReg1, filt2Reg); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt1_1 = _mm_maddubs_epi16(srcRegFilt1_1, firstFilters); + srcRegFilt2 = _mm_maddubs_epi16(srcRegFilt2, secondFilters); + + // add and saturate the results together + srcRegFilt1_1 = _mm_adds_epi16(srcRegFilt1_1, srcRegFilt2); + + // filter the source buffer + srcRegFilt3= _mm_shuffle_epi8(srcReg1, filt4Reg); + srcRegFilt2= _mm_shuffle_epi8(srcReg1, filt3Reg); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt3 = _mm_maddubs_epi16(srcRegFilt3, forthFilters); + srcRegFilt2 = _mm_maddubs_epi16(srcRegFilt2, thirdFilters); + + // add and saturate the results together + srcRegFilt1_1 = _mm_adds_epi16(srcRegFilt1_1, + _mm_min_epi16(srcRegFilt3, srcRegFilt2)); + + // reading the next 16 bytes. + // (part of it was being read by earlier read) + srcReg2 = _mm_loadu_si128((__m128i *)(src_ptr+5)); + + // add and saturate the results together + srcRegFilt1_1 = _mm_adds_epi16(srcRegFilt1_1, + _mm_max_epi16(srcRegFilt3, srcRegFilt2)); + + // filter the source buffer + srcRegFilt2_1= _mm_shuffle_epi8(srcReg2, filt1Reg); + srcRegFilt2= _mm_shuffle_epi8(srcReg2, filt2Reg); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt2_1 = _mm_maddubs_epi16(srcRegFilt2_1, firstFilters); + srcRegFilt2 = _mm_maddubs_epi16(srcRegFilt2, secondFilters); + + // add and saturate the results together + srcRegFilt2_1 = _mm_adds_epi16(srcRegFilt2_1, srcRegFilt2); + + // filter the source buffer + srcRegFilt3= _mm_shuffle_epi8(srcReg2, filt4Reg); + srcRegFilt2= _mm_shuffle_epi8(srcReg2, filt3Reg); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt3 = _mm_maddubs_epi16(srcRegFilt3, forthFilters); + srcRegFilt2 = _mm_maddubs_epi16(srcRegFilt2, thirdFilters); + + // add and saturate the results together + srcRegFilt2_1 = _mm_adds_epi16(srcRegFilt2_1, + _mm_min_epi16(srcRegFilt3, srcRegFilt2)); + srcRegFilt2_1 = _mm_adds_epi16(srcRegFilt2_1, + _mm_max_epi16(srcRegFilt3, srcRegFilt2)); + + srcRegFilt1_1 = _mm_adds_epi16(srcRegFilt1_1, addFilterReg64); + srcRegFilt2_1 = _mm_adds_epi16(srcRegFilt2_1, addFilterReg64); + + // shift by 7 bit each 16 bit + srcRegFilt1_1 = _mm_srai_epi16(srcRegFilt1_1, 7); + srcRegFilt2_1 = _mm_srai_epi16(srcRegFilt2_1, 7); + + // shrink to 8 bit each 16 bits, the first lane contain the first + // convolve result and the second lane contain the second convolve + // result + srcRegFilt1_1 = _mm_packus_epi16(srcRegFilt1_1, srcRegFilt2_1); + + src_ptr+=src_pixels_per_line; + + // save 16 bytes + _mm_store_si128((__m128i*)output_ptr, srcRegFilt1_1); + + output_ptr+=output_pitch; + } +} + +void vp9_filter_block1d8_v8_intrin_ssse3(unsigned char *src_ptr, + unsigned int src_pitch, + unsigned char *output_ptr, + unsigned int out_pitch, + unsigned int output_height, + int16_t *filter) { + __m128i addFilterReg64, filtersReg, minReg, srcRegFilt6; + __m128i firstFilters, secondFilters, thirdFilters, forthFilters; + __m128i srcRegFilt1, srcRegFilt2, srcRegFilt3, srcRegFilt4, srcRegFilt5; + unsigned int i; + + // create a register with 0,64,0,64,0,64,0,64,0,64,0,64,0,64,0,64 + addFilterReg64 = _mm_set1_epi32((int)0x0400040u); + filtersReg = _mm_loadu_si128((__m128i *)filter); + // converting the 16 bit (short) to 8 bit (byte) and have the same data + // in both lanes of 128 bit register. + filtersReg =_mm_packs_epi16(filtersReg, filtersReg); + + // duplicate only the first 16 bits in the filter + firstFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x100u)); + // duplicate only the second 16 bits in the filter + secondFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x302u)); + // duplicate only the third 16 bits in the filter + thirdFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x504u)); + // duplicate only the forth 16 bits in the filter + forthFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x706u)); + + for (i = 0; i < output_height; i++) { + // load the first 8 bytes + srcRegFilt1 = _mm_loadl_epi64((__m128i *)&src_ptr[0]); + // load the next 8 bytes in stride of src_pitch + srcRegFilt2 = _mm_loadl_epi64((__m128i *)&(src_ptr+src_pitch)[0]); + srcRegFilt3 = _mm_loadl_epi64((__m128i *)&(src_ptr+src_pitch*2)[0]); + srcRegFilt4 = _mm_loadl_epi64((__m128i *)&(src_ptr+src_pitch*3)[0]); + + // merge the result together + srcRegFilt1 = _mm_unpacklo_epi8(srcRegFilt1, srcRegFilt2); + srcRegFilt3 = _mm_unpacklo_epi8(srcRegFilt3, srcRegFilt4); + + // load the next 8 bytes in stride of src_pitch + srcRegFilt2 = _mm_loadl_epi64((__m128i *)&(src_ptr+src_pitch*4)[0]); + srcRegFilt4 = _mm_loadl_epi64((__m128i *)&(src_ptr+src_pitch*5)[0]); + srcRegFilt5 = _mm_loadl_epi64((__m128i *)&(src_ptr+src_pitch*6)[0]); + srcRegFilt6 = _mm_loadl_epi64((__m128i *)&(src_ptr+src_pitch*7)[0]); + + // merge the result together + srcRegFilt2 = _mm_unpacklo_epi8(srcRegFilt2, srcRegFilt4); + srcRegFilt5 = _mm_unpacklo_epi8(srcRegFilt5, srcRegFilt6); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt1 = _mm_maddubs_epi16(srcRegFilt1, firstFilters); + srcRegFilt3 = _mm_maddubs_epi16(srcRegFilt3, secondFilters); + srcRegFilt2 = _mm_maddubs_epi16(srcRegFilt2, thirdFilters); + srcRegFilt5 = _mm_maddubs_epi16(srcRegFilt5, forthFilters); + + // add and saturate the results together + minReg = _mm_min_epi16(srcRegFilt2, srcRegFilt3); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, srcRegFilt5); + srcRegFilt2 = _mm_max_epi16(srcRegFilt2, srcRegFilt3); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, minReg); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, srcRegFilt2); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, addFilterReg64); + + // shift by 7 bit each 16 bit + srcRegFilt1 = _mm_srai_epi16(srcRegFilt1, 7); + + // shrink to 8 bit each 16 bits + srcRegFilt1 = _mm_packus_epi16(srcRegFilt1, srcRegFilt1); + + src_ptr+=src_pitch; + + // save only 8 bytes convolve result + _mm_storel_epi64((__m128i*)&output_ptr[0], srcRegFilt1); + + output_ptr+=out_pitch; + } +} + +void vp9_filter_block1d16_v8_intrin_ssse3(unsigned char *src_ptr, + unsigned int src_pitch, + unsigned char *output_ptr, + unsigned int out_pitch, + unsigned int output_height, + int16_t *filter) { + __m128i addFilterReg64, filtersReg, srcRegFilt1, srcRegFilt2, srcRegFilt3; + __m128i firstFilters, secondFilters, thirdFilters, forthFilters; + __m128i srcRegFilt4, srcRegFilt5, srcRegFilt6, srcRegFilt7, srcRegFilt8; + unsigned int i; + + // create a register with 0,64,0,64,0,64,0,64,0,64,0,64,0,64,0,64 + addFilterReg64 = _mm_set1_epi32((int)0x0400040u); + filtersReg = _mm_loadu_si128((__m128i *)filter); + // converting the 16 bit (short) to 8 bit (byte) and have the same data + // in both lanes of 128 bit register. + filtersReg =_mm_packs_epi16(filtersReg, filtersReg); + + // duplicate only the first 16 bits in the filter + firstFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x100u)); + // duplicate only the second 16 bits in the filter + secondFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x302u)); + // duplicate only the third 16 bits in the filter + thirdFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x504u)); + // duplicate only the forth 16 bits in the filter + forthFilters = _mm_shuffle_epi8(filtersReg, _mm_set1_epi16(0x706u)); + + for (i = 0; i < output_height; i++) { + // load the first 16 bytes + srcRegFilt1 = _mm_loadu_si128((__m128i *)(src_ptr)); + // load the next 16 bytes in stride of src_pitch + srcRegFilt2 = _mm_loadu_si128((__m128i *)(src_ptr+src_pitch)); + srcRegFilt3 = _mm_loadu_si128((__m128i *)(src_ptr+src_pitch*6)); + srcRegFilt4 = _mm_loadu_si128((__m128i *)(src_ptr+src_pitch*7)); + + // merge the result together + srcRegFilt5 = _mm_unpacklo_epi8(srcRegFilt1, srcRegFilt2); + srcRegFilt6 = _mm_unpacklo_epi8(srcRegFilt3, srcRegFilt4); + srcRegFilt1 = _mm_unpackhi_epi8(srcRegFilt1, srcRegFilt2); + srcRegFilt3 = _mm_unpackhi_epi8(srcRegFilt3, srcRegFilt4); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt5 = _mm_maddubs_epi16(srcRegFilt5, firstFilters); + srcRegFilt6 = _mm_maddubs_epi16(srcRegFilt6, forthFilters); + srcRegFilt1 = _mm_maddubs_epi16(srcRegFilt1, firstFilters); + srcRegFilt3 = _mm_maddubs_epi16(srcRegFilt3, forthFilters); + + // add and saturate the results together + srcRegFilt5 = _mm_adds_epi16(srcRegFilt5, srcRegFilt6); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, srcRegFilt3); + + // load the next 16 bytes in stride of two/three src_pitch + srcRegFilt2 = _mm_loadu_si128((__m128i *)(src_ptr+src_pitch*2)); + srcRegFilt3 = _mm_loadu_si128((__m128i *)(src_ptr+src_pitch*3)); + + // merge the result together + srcRegFilt4 = _mm_unpacklo_epi8(srcRegFilt2, srcRegFilt3); + srcRegFilt6 = _mm_unpackhi_epi8(srcRegFilt2, srcRegFilt3); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt4 = _mm_maddubs_epi16(srcRegFilt4, secondFilters); + srcRegFilt6 = _mm_maddubs_epi16(srcRegFilt6, secondFilters); + + // load the next 16 bytes in stride of four/five src_pitch + srcRegFilt2 = _mm_loadu_si128((__m128i *)(src_ptr+src_pitch*4)); + srcRegFilt3 = _mm_loadu_si128((__m128i *)(src_ptr+src_pitch*5)); + + // merge the result together + srcRegFilt7 = _mm_unpacklo_epi8(srcRegFilt2, srcRegFilt3); + srcRegFilt8 = _mm_unpackhi_epi8(srcRegFilt2, srcRegFilt3); + + // multiply 2 adjacent elements with the filter and add the result + srcRegFilt7 = _mm_maddubs_epi16(srcRegFilt7, thirdFilters); + srcRegFilt8 = _mm_maddubs_epi16(srcRegFilt8, thirdFilters); + + // add and saturate the results together + srcRegFilt5 = _mm_adds_epi16(srcRegFilt5, + _mm_min_epi16(srcRegFilt4, srcRegFilt7)); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, + _mm_min_epi16(srcRegFilt6, srcRegFilt8)); + + // add and saturate the results together + srcRegFilt5 = _mm_adds_epi16(srcRegFilt5, + _mm_max_epi16(srcRegFilt4, srcRegFilt7)); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, + _mm_max_epi16(srcRegFilt6, srcRegFilt8)); + srcRegFilt5 = _mm_adds_epi16(srcRegFilt5, addFilterReg64); + srcRegFilt1 = _mm_adds_epi16(srcRegFilt1, addFilterReg64); + + // shift by 7 bit each 16 bit + srcRegFilt5 = _mm_srai_epi16(srcRegFilt5, 7); + srcRegFilt1 = _mm_srai_epi16(srcRegFilt1, 7); + + // shrink to 8 bit each 16 bits, the first lane contain the first + // convolve result and the second lane contain the second convolve + // result + srcRegFilt1 = _mm_packus_epi16(srcRegFilt5, srcRegFilt1); + + src_ptr+=src_pitch; + + // save 16 bytes convolve result + _mm_store_si128((__m128i*)output_ptr, srcRegFilt1); + + output_ptr+=out_pitch; + } +} diff --git a/vp9/decoder/vp9_decodeframe.c b/vp9/decoder/vp9_decodeframe.c index 791d0f2e7..4f3aca1d2 100644 --- a/vp9/decoder/vp9_decodeframe.c +++ b/vp9/decoder/vp9_decodeframe.c @@ -1120,6 +1120,12 @@ static size_t read_uncompressed_header(VP9D_COMP *pbi, if (cm->show_existing_frame) { // Show an existing frame directly. const int frame_to_show = cm->ref_frame_map[vp9_rb_read_literal(rb, 3)]; + + if (cm->frame_bufs[frame_to_show].ref_count < 1) + vpx_internal_error(&cm->error, VPX_CODEC_UNSUP_BITSTREAM, + "Buffer %d does not contain a decoded frame", + frame_to_show); + ref_cnt_fb(cm->frame_bufs, &cm->new_fb_idx, frame_to_show); pbi->refresh_frame_flags = 0; cm->lf.filter_level = 0; diff --git a/vp9/decoder/vp9_decodemv.c b/vp9/decoder/vp9_decodemv.c index 856c8b533..4de8db37c 100644 --- a/vp9/decoder/vp9_decodemv.c +++ b/vp9/decoder/vp9_decodemv.c @@ -356,6 +356,11 @@ static void read_intra_block_mode_info(VP9_COMMON *const cm, MODE_INFO *mi, mbmi->uv_mode = read_intra_mode_uv(cm, r, mbmi->mode); } +static INLINE int is_mv_valid(const MV *mv) { + return mv->row > MV_LOW && mv->row < MV_UPP && + mv->col > MV_LOW && mv->col < MV_UPP; +} + static INLINE int assign_mv(VP9_COMMON *cm, MB_PREDICTION_MODE mode, int_mv mv[2], int_mv ref_mv[2], int_mv nearest_mv[2], int_mv near_mv[2], @@ -367,14 +372,10 @@ static INLINE int assign_mv(VP9_COMMON *cm, MB_PREDICTION_MODE mode, case NEWMV: { nmv_context_counts *const mv_counts = cm->frame_parallel_decoding_mode ? NULL : &cm->counts.mv; - read_mv(r, &mv[0].as_mv, &ref_mv[0].as_mv, - &cm->fc.nmvc, mv_counts, allow_hp); - if (is_compound) - read_mv(r, &mv[1].as_mv, &ref_mv[1].as_mv, - &cm->fc.nmvc, mv_counts, allow_hp); for (i = 0; i < 1 + is_compound; ++i) { - ret = ret && mv[i].as_mv.row < MV_UPP && mv[i].as_mv.row > MV_LOW; - ret = ret && mv[i].as_mv.col < MV_UPP && mv[i].as_mv.col > MV_LOW; + read_mv(r, &mv[i].as_mv, &ref_mv[i].as_mv, &cm->fc.nmvc, mv_counts, + allow_hp); + ret = ret && is_mv_valid(&mv[i].as_mv); } break; } diff --git a/vp9/encoder/vp9_bitstream.c b/vp9/encoder/vp9_bitstream.c index 2a1836f40..31ec069d0 100644 --- a/vp9/encoder/vp9_bitstream.c +++ b/vp9/encoder/vp9_bitstream.c @@ -34,7 +34,6 @@ #include "vp9/encoder/vp9_write_bit_buffer.h" #ifdef ENTROPY_STATS -vp9_coeff_stats tree_update_hist[TX_SIZES][PLANE_TYPES]; extern unsigned int active_section; #endif @@ -247,15 +246,15 @@ static void pack_inter_mode_mvs(VP9_COMP *cpi, MODE_INFO *m, vp9_writer *bc) { const nmv_context *nmvc = &cm->fc.nmvc; MACROBLOCK *const x = &cpi->mb; MACROBLOCKD *const xd = &x->e_mbd; - struct segmentation *seg = &cm->seg; + const struct segmentation *const seg = &cm->seg; MB_MODE_INFO *const mi = &m->mbmi; - const MV_REFERENCE_FRAME rf = mi->ref_frame[0]; - const MV_REFERENCE_FRAME sec_rf = mi->ref_frame[1]; + const MV_REFERENCE_FRAME ref0 = mi->ref_frame[0]; + const MV_REFERENCE_FRAME ref1 = mi->ref_frame[1]; const MB_PREDICTION_MODE mode = mi->mode; const int segment_id = mi->segment_id; - int skip; const BLOCK_SIZE bsize = mi->sb_type; const int allow_hp = cm->allow_high_precision_mv; + int skip; #ifdef ENTROPY_STATS active_section = 9; @@ -276,15 +275,15 @@ static void pack_inter_mode_mvs(VP9_COMP *cpi, MODE_INFO *m, vp9_writer *bc) { skip = write_skip(cpi, segment_id, m, bc); if (!vp9_segfeature_active(seg, segment_id, SEG_LVL_REF_FRAME)) - vp9_write(bc, rf != INTRA_FRAME, vp9_get_intra_inter_prob(cm, xd)); + vp9_write(bc, ref0 != INTRA_FRAME, vp9_get_intra_inter_prob(cm, xd)); if (bsize >= BLOCK_8X8 && cm->tx_mode == TX_MODE_SELECT && - !(rf != INTRA_FRAME && + !(ref0 != INTRA_FRAME && (skip || vp9_segfeature_active(seg, segment_id, SEG_LVL_SKIP)))) { write_selected_tx_size(cpi, m, mi->tx_size, bsize, bc); } - if (rf == INTRA_FRAME) { + if (ref0 == INTRA_FRAME) { #ifdef ENTROPY_STATS active_section = 6; #endif @@ -306,7 +305,7 @@ static void pack_inter_mode_mvs(VP9_COMP *cpi, MODE_INFO *m, vp9_writer *bc) { } else { vp9_prob *mv_ref_p; encode_ref_frame(cpi, bc); - mv_ref_p = cpi->common.fc.inter_mode_probs[mi->mode_context[rf]]; + mv_ref_p = cm->fc.inter_mode_probs[mi->mode_context[ref0]]; #ifdef ENTROPY_STATS active_section = 3; @@ -316,7 +315,7 @@ static void pack_inter_mode_mvs(VP9_COMP *cpi, MODE_INFO *m, vp9_writer *bc) { if (!vp9_segfeature_active(seg, segment_id, SEG_LVL_SKIP)) { if (bsize >= BLOCK_8X8) { write_inter_mode(bc, mode, mv_ref_p); - ++cm->counts.inter_mode[mi->mode_context[rf]][INTER_OFFSET(mode)]; + ++cm->counts.inter_mode[mi->mode_context[ref0]][INTER_OFFSET(mode)]; } } @@ -336,21 +335,19 @@ static void pack_inter_mode_mvs(VP9_COMP *cpi, MODE_INFO *m, vp9_writer *bc) { for (idy = 0; idy < 2; idy += num_4x4_blocks_high) { for (idx = 0; idx < 2; idx += num_4x4_blocks_wide) { const int j = idy * 2 + idx; - const MB_PREDICTION_MODE blockmode = m->bmi[j].as_mode; - write_inter_mode(bc, blockmode, mv_ref_p); - ++cm->counts.inter_mode[mi->mode_context[rf]] - [INTER_OFFSET(blockmode)]; - - if (blockmode == NEWMV) { + const MB_PREDICTION_MODE b_mode = m->bmi[j].as_mode; + write_inter_mode(bc, b_mode, mv_ref_p); + ++cm->counts.inter_mode[mi->mode_context[ref0]][INTER_OFFSET(b_mode)]; + if (b_mode == NEWMV) { #ifdef ENTROPY_STATS active_section = 11; #endif vp9_encode_mv(cpi, bc, &m->bmi[j].as_mv[0].as_mv, - &mi->ref_mvs[rf][0].as_mv, nmvc, allow_hp); + &mi->ref_mvs[ref0][0].as_mv, nmvc, allow_hp); if (has_second_ref(mi)) vp9_encode_mv(cpi, bc, &m->bmi[j].as_mv[1].as_mv, - &mi->ref_mvs[sec_rf][0].as_mv, nmvc, allow_hp); + &mi->ref_mvs[ref1][0].as_mv, nmvc, allow_hp); } } } @@ -359,11 +356,11 @@ static void pack_inter_mode_mvs(VP9_COMP *cpi, MODE_INFO *m, vp9_writer *bc) { active_section = 5; #endif vp9_encode_mv(cpi, bc, &mi->mv[0].as_mv, - &mi->ref_mvs[rf][0].as_mv, nmvc, allow_hp); + &mi->ref_mvs[ref0][0].as_mv, nmvc, allow_hp); if (has_second_ref(mi)) vp9_encode_mv(cpi, bc, &mi->mv[1].as_mv, - &mi->ref_mvs[sec_rf][0].as_mv, nmvc, allow_hp); + &mi->ref_mvs[ref1][0].as_mv, nmvc, allow_hp); } } } @@ -628,10 +625,6 @@ static void update_coef_probs_common(vp9_writer* const bc, VP9_COMP *cpi, if (s > 0 && newp != *oldp) u = 1; vp9_write(bc, u, upd); -#ifdef ENTROPY_STATS - if (!cpi->dummy_packing) - ++tree_update_hist[tx_size][i][j][k][l][t][u]; -#endif if (u) { /* send/use new probability */ vp9_write_prob_diff_update(bc, newp, *oldp); @@ -683,10 +676,6 @@ static void update_coef_probs_common(vp9_writer* const bc, VP9_COMP *cpi, updates += u; if (u == 0 && updates == 0) { noupdates_before_first++; -#ifdef ENTROPY_STATS - if (!cpi->dummy_packing) - ++tree_update_hist[tx_size][i][j][k][l][t][u]; -#endif continue; } if (u == 1 && updates == 1) { @@ -697,10 +686,6 @@ static void update_coef_probs_common(vp9_writer* const bc, VP9_COMP *cpi, vp9_write(bc, 0, upd); } vp9_write(bc, u, upd); -#ifdef ENTROPY_STATS - if (!cpi->dummy_packing) - ++tree_update_hist[tx_size][i][j][k][l][t][u]; -#endif if (u) { /* send/use new probability */ vp9_write_prob_diff_update(bc, newp, *oldp); @@ -1272,7 +1257,7 @@ void vp9_pack_bitstream(VP9_COMP *cpi, uint8_t *dest, size_t *size) { active_section = 7; #endif - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); first_part_size = write_compressed_header(cpi, data); data += first_part_size; diff --git a/vp9/encoder/vp9_block.h b/vp9/encoder/vp9_block.h index 7cbdfce52..85f6c97af 100644 --- a/vp9/encoder/vp9_block.h +++ b/vp9/encoder/vp9_block.h @@ -49,7 +49,6 @@ typedef struct { int_mv ref_mvs[MAX_REF_FRAMES][MAX_MV_REF_CANDIDATES]; int rate; int distortion; - int64_t intra_error; int best_mode_index; int rddiv; int rdmult; @@ -63,9 +62,6 @@ typedef struct { // search loop int_mv pred_mv[MAX_REF_FRAMES]; INTERP_FILTER pred_interp_filter; - - // Bit flag for each mode whether it has high error in comparison to others. - unsigned int modes_with_high_error; } PICK_MODE_CONTEXT; struct macroblock_plane { diff --git a/vp9/encoder/vp9_encodeframe.c b/vp9/encoder/vp9_encodeframe.c index 802346632..f529c9336 100644 --- a/vp9/encoder/vp9_encodeframe.c +++ b/vp9/encoder/vp9_encodeframe.c @@ -398,7 +398,6 @@ static void update_state(VP9_COMP *cpi, PICK_MODE_CONTEXT *ctx, MB_MODE_INFO *const mbmi = &xd->mi_8x8[0]->mbmi; MODE_INFO *mi_addr = xd->mi_8x8[0]; - const int mb_mode_index = ctx->best_mode_index; const int mis = cm->mode_info_stride; const int mi_width = num_8x8_blocks_wide_lookup[bsize]; const int mi_height = num_8x8_blocks_high_lookup[bsize]; @@ -470,8 +469,8 @@ static void update_state(VP9_COMP *cpi, PICK_MODE_CONTEXT *ctx, cpi->rd_tx_select_diff[i] += ctx->tx_rd_diff[i]; } - if (frame_is_intra_only(cm)) { #if CONFIG_INTERNAL_STATS + if (frame_is_intra_only(cm)) { static const int kf_mode_index[] = { THR_DC /*DC_PRED*/, THR_V_PRED /*V_PRED*/, @@ -484,12 +483,13 @@ static void update_state(VP9_COMP *cpi, PICK_MODE_CONTEXT *ctx, THR_D63_PRED /*D63_PRED*/, THR_TM /*TM_PRED*/, }; - cpi->mode_chosen_counts[kf_mode_index[mbmi->mode]]++; -#endif + ++cpi->mode_chosen_counts[kf_mode_index[mbmi->mode]]; } else { // Note how often each mode chosen as best - cpi->mode_chosen_counts[mb_mode_index]++; - + ++cpi->mode_chosen_counts[ctx->best_mode_index]; + } +#endif + if (!frame_is_intra_only(cm)) { if (is_inter_block(mbmi)) { if (mbmi->sb_type < BLOCK_8X8 || mbmi->mode == NEWMV) { int_mv best_mv[2]; @@ -627,7 +627,7 @@ static void rd_pick_sb_modes(VP9_COMP *cpi, const TileInfo *const tile, int orig_rdmult = x->rdmult; double rdmult_ratio; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); rdmult_ratio = 1.0; // avoid uninitialized warnings // Use the lower precision, but faster, 32x32 fdct for mode selection. @@ -683,8 +683,8 @@ static void rd_pick_sb_modes(VP9_COMP *cpi, const TileInfo *const tile, activity_masking(cpi, x); if (cpi->oxcf.aq_mode == VARIANCE_AQ) { - vp9_clear_system_state(); // __asm emms; - x->rdmult = round(x->rdmult * rdmult_ratio); + vp9_clear_system_state(); + x->rdmult = (int)round(x->rdmult * rdmult_ratio); } else if (cpi->oxcf.aq_mode == COMPLEXITY_AQ) { const int mi_offset = mi_row * cm->mi_cols + mi_col; unsigned char complexity = cpi->complexity_map[mi_offset]; @@ -713,8 +713,8 @@ static void rd_pick_sb_modes(VP9_COMP *cpi, const TileInfo *const tile, if (cpi->oxcf.aq_mode == VARIANCE_AQ) { x->rdmult = orig_rdmult; if (*totalrate != INT_MAX) { - vp9_clear_system_state(); // __asm emms; - *totalrate = round(*totalrate * rdmult_ratio); + vp9_clear_system_state(); + *totalrate = (int)round(*totalrate * rdmult_ratio); } } else if (cpi->oxcf.aq_mode == COMPLEXITY_AQ) { @@ -965,9 +965,9 @@ static BLOCK_SIZE find_partition_size(BLOCK_SIZE bsize, // may not be allowed in which case this code attempts to choose the largest // allowable partition. static void set_partitioning(VP9_COMP *cpi, const TileInfo *const tile, - MODE_INFO **mi_8x8, int mi_row, int mi_col) { + MODE_INFO **mi_8x8, int mi_row, int mi_col, + BLOCK_SIZE bsize) { VP9_COMMON *const cm = &cpi->common; - BLOCK_SIZE bsize = cpi->sf.always_this_block_size; const int mis = cm->mode_info_stride; int row8x8_remaining = tile->mi_row_end - mi_row; int col8x8_remaining = tile->mi_col_end - mi_col; @@ -1047,32 +1047,12 @@ static void update_state_rt(VP9_COMP *cpi, PICK_MODE_CONTEXT *ctx, VP9_COMMON *const cm = &cpi->common; MACROBLOCK *const x = &cpi->mb; MACROBLOCKD *const xd = &x->e_mbd; - struct macroblock_plane *const p = x->plane; - struct macroblockd_plane *const pd = xd->plane; MB_MODE_INFO *const mbmi = &xd->mi_8x8[0]->mbmi; - const int mb_mode_index = ctx->best_mode_index; - int max_plane; - - max_plane = is_inter_block(mbmi) ? MAX_MB_PLANE : 1; - for (i = 0; i < max_plane; ++i) { - p[i].coeff = ctx->coeff_pbuf[i][1]; - p[i].qcoeff = ctx->qcoeff_pbuf[i][1]; - pd[i].dqcoeff = ctx->dqcoeff_pbuf[i][1]; - p[i].eobs = ctx->eobs_pbuf[i][1]; - } - - for (i = max_plane; i < MAX_MB_PLANE; ++i) { - p[i].coeff = ctx->coeff_pbuf[i][2]; - p[i].qcoeff = ctx->qcoeff_pbuf[i][2]; - pd[i].dqcoeff = ctx->dqcoeff_pbuf[i][2]; - p[i].eobs = ctx->eobs_pbuf[i][2]; - } - x->skip = ctx->skip; - if (frame_is_intra_only(cm)) { #if CONFIG_INTERNAL_STATS + if (frame_is_intra_only(cm)) { static const int kf_mode_index[] = { THR_DC /*DC_PRED*/, THR_V_PRED /*V_PRED*/, @@ -1086,10 +1066,12 @@ static void update_state_rt(VP9_COMP *cpi, PICK_MODE_CONTEXT *ctx, THR_TM /*TM_PRED*/, }; ++cpi->mode_chosen_counts[kf_mode_index[mbmi->mode]]; -#endif } else { // Note how often each mode chosen as best - cpi->mode_chosen_counts[mb_mode_index]++; + ++cpi->mode_chosen_counts[ctx->best_mode_index]; + } +#endif + if (!frame_is_intra_only(cm)) { if (is_inter_block(mbmi)) { if (mbmi->sb_type < BLOCK_8X8 || mbmi->mode == NEWMV) { int_mv best_mv[2]; @@ -1128,8 +1110,8 @@ static void encode_b_rt(VP9_COMP *cpi, const TileInfo *const tile, } static void encode_sb_rt(VP9_COMP *cpi, const TileInfo *const tile, - TOKENEXTRA **tp, int mi_row, int mi_col, - int output_enabled, BLOCK_SIZE bsize) { + TOKENEXTRA **tp, int mi_row, int mi_col, + int output_enabled, BLOCK_SIZE bsize) { VP9_COMMON *const cm = &cpi->common; MACROBLOCK *const x = &cpi->mb; const int bsl = b_width_log2(bsize), hbs = (1 << bsl) / 4; @@ -1147,7 +1129,6 @@ static void encode_sb_rt(VP9_COMP *cpi, const TileInfo *const tile, ctx = partition_plane_context(cpi->above_seg_context, cpi->left_seg_context, mi_row, mi_col, bsize); subsize = mi_8x8[0]->mbmi.sb_type; - } else { ctx = 0; subsize = BLOCK_4X4; @@ -1230,13 +1211,11 @@ static void rd_use_partition(VP9_COMP *cpi, ENTROPY_CONTEXT l[16 * MAX_MB_PLANE], a[16 * MAX_MB_PLANE]; PARTITION_CONTEXT sl[8], sa[8]; int last_part_rate = INT_MAX; - int64_t last_part_dist = INT_MAX; - int split_rate = INT_MAX; - int64_t split_dist = INT_MAX; + int64_t last_part_dist = INT64_MAX; int none_rate = INT_MAX; - int64_t none_dist = INT_MAX; + int64_t none_dist = INT64_MAX; int chosen_rate = INT_MAX; - int64_t chosen_dist = INT_MAX; + int64_t chosen_dist = INT64_MAX; BLOCK_SIZE sub_subsize = BLOCK_4X4; int splits_below = 0; BLOCK_SIZE bs_type = mi_8x8[0]->mbmi.sb_type; @@ -1319,9 +1298,9 @@ static void rd_use_partition(VP9_COMP *cpi, *get_sb_index(x, subsize) = 1; rd_pick_sb_modes(cpi, tile, mi_row + (ms >> 1), mi_col, &rt, &dt, subsize, get_block_context(x, subsize), INT64_MAX); - if (rt == INT_MAX || dt == INT_MAX) { + if (rt == INT_MAX || dt == INT64_MAX) { last_part_rate = INT_MAX; - last_part_dist = INT_MAX; + last_part_dist = INT64_MAX; break; } @@ -1343,9 +1322,9 @@ static void rd_use_partition(VP9_COMP *cpi, *get_sb_index(x, subsize) = 1; rd_pick_sb_modes(cpi, tile, mi_row, mi_col + (ms >> 1), &rt, &dt, subsize, get_block_context(x, subsize), INT64_MAX); - if (rt == INT_MAX || dt == INT_MAX) { + if (rt == INT_MAX || dt == INT64_MAX) { last_part_rate = INT_MAX; - last_part_dist = INT_MAX; + last_part_dist = INT64_MAX; break; } last_part_rate += rt; @@ -1371,9 +1350,9 @@ static void rd_use_partition(VP9_COMP *cpi, rd_use_partition(cpi, tile, mi_8x8 + jj * bss * mis + ii * bss, tp, mi_row + y_idx, mi_col + x_idx, subsize, &rt, &dt, i != 3); - if (rt == INT_MAX || dt == INT_MAX) { + if (rt == INT_MAX || dt == INT64_MAX) { last_part_rate = INT_MAX; - last_part_dist = INT_MAX; + last_part_dist = INT64_MAX; break; } last_part_rate += rt; @@ -1394,8 +1373,8 @@ static void rd_use_partition(VP9_COMP *cpi, && (mi_row + ms < cm->mi_rows || mi_row + (ms >> 1) == cm->mi_rows) && (mi_col + ms < cm->mi_cols || mi_col + (ms >> 1) == cm->mi_cols)) { BLOCK_SIZE split_subsize = get_subsize(bsize, PARTITION_SPLIT); - split_rate = 0; - split_dist = 0; + chosen_rate = 0; + chosen_dist = 0; restore_context(cpi, mi_row, mi_col, a, l, sa, sl, bsize); // Split partition. @@ -1422,31 +1401,29 @@ static void rd_use_partition(VP9_COMP *cpi, restore_context(cpi, mi_row, mi_col, a, l, sa, sl, bsize); - if (rt == INT_MAX || dt == INT_MAX) { - split_rate = INT_MAX; - split_dist = INT_MAX; + if (rt == INT_MAX || dt == INT64_MAX) { + chosen_rate = INT_MAX; + chosen_dist = INT64_MAX; break; } + chosen_rate += rt; + chosen_dist += dt; + if (i != 3) encode_sb(cpi, tile, tp, mi_row + y_idx, mi_col + x_idx, 0, split_subsize); - split_rate += rt; - split_dist += dt; pl = partition_plane_context(cpi->above_seg_context, cpi->left_seg_context, mi_row + y_idx, mi_col + x_idx, split_subsize); - split_rate += x->partition_cost[pl][PARTITION_NONE]; + chosen_rate += x->partition_cost[pl][PARTITION_NONE]; } pl = partition_plane_context(cpi->above_seg_context, cpi->left_seg_context, mi_row, mi_col, bsize); - if (split_rate < INT_MAX) { - split_rate += x->partition_cost[pl][PARTITION_SPLIT]; - - chosen_rate = split_rate; - chosen_dist = split_dist; + if (chosen_rate < INT_MAX) { + chosen_rate += x->partition_cost[pl][PARTITION_SPLIT]; } } @@ -1473,7 +1450,7 @@ static void rd_use_partition(VP9_COMP *cpi, // We must have chosen a partitioning and encoding or we'll fail later on. // No other opportunities for success. if ( bsize == BLOCK_64X64) - assert(chosen_rate < INT_MAX && chosen_dist < INT_MAX); + assert(chosen_rate < INT_MAX && chosen_dist < INT64_MAX); if (do_recon) { int output_enabled = (bsize == BLOCK_64X64); @@ -1931,7 +1908,7 @@ static void rd_pick_partition(VP9_COMP *cpi, const TileInfo *const tile, if (bsize == BLOCK_64X64) { assert(tp_orig < *tp); assert(best_rate < INT_MAX); - assert(best_dist < INT_MAX); + assert(best_dist < INT64_MAX); } else { assert(tp_orig == *tp); } @@ -1954,14 +1931,17 @@ static void encode_sb_row(VP9_COMP *cpi, const TileInfo *const tile, BLOCK_SIZE i; MACROBLOCK *x = &cpi->mb; - for (i = BLOCK_4X4; i < BLOCK_8X8; ++i) { - const int num_4x4_w = num_4x4_blocks_wide_lookup[i]; - const int num_4x4_h = num_4x4_blocks_high_lookup[i]; - const int num_4x4_blk = MAX(4, num_4x4_w * num_4x4_h); - for (x->sb_index = 0; x->sb_index < 4; ++x->sb_index) - for (x->mb_index = 0; x->mb_index < 4; ++x->mb_index) - for (x->b_index = 0; x->b_index < 16 / num_4x4_blk; ++x->b_index) - get_block_context(x, i)->pred_interp_filter = SWITCHABLE; + + if (cpi->sf.adaptive_pred_interp_filter) { + for (i = BLOCK_4X4; i < BLOCK_8X8; ++i) { + const int num_4x4_w = num_4x4_blocks_wide_lookup[i]; + const int num_4x4_h = num_4x4_blocks_high_lookup[i]; + const int num_4x4_blk = MAX(4, num_4x4_w * num_4x4_h); + for (x->sb_index = 0; x->sb_index < 4; ++x->sb_index) + for (x->mb_index = 0; x->mb_index < 4; ++x->mb_index) + for (x->b_index = 0; x->b_index < 16 / num_4x4_blk; ++x->b_index) + get_block_context(x, i)->pred_interp_filter = SWITCHABLE; + } } vp9_zero(cpi->mb.pred_mv); @@ -1975,7 +1955,8 @@ static void encode_sb_row(VP9_COMP *cpi, const TileInfo *const tile, cpi->mb.source_variance = UINT_MAX; if (cpi->sf.use_one_partition_size_always) { set_offsets(cpi, tile, mi_row, mi_col, BLOCK_64X64); - set_partitioning(cpi, tile, mi_8x8, mi_row, mi_col); + set_partitioning(cpi, tile, mi_8x8, mi_row, mi_col, + cpi->sf.always_this_block_size); rd_use_partition(cpi, tile, mi_8x8, tp, mi_row, mi_col, BLOCK_64X64, &dummy_rate, &dummy_dist, 1); } else { @@ -2272,68 +2253,56 @@ static INLINE int get_block_col(int b32i, int b16i, int b8i) { } static void rtc_use_partition(VP9_COMP *cpi, - const TileInfo *const tile, - MODE_INFO **mi_8x8, - TOKENEXTRA **tp, int mi_row, int mi_col, - BLOCK_SIZE bsize, int *rate, int64_t *dist, - int do_recon) { + const TileInfo *const tile, + MODE_INFO **mi_8x8, + TOKENEXTRA **tp, int mi_row, int mi_col, + BLOCK_SIZE bsize, int *rate, int64_t *dist, + int do_recon) { VP9_COMMON *const cm = &cpi->common; MACROBLOCK *const x = &cpi->mb; MACROBLOCKD *const xd = &cpi->mb.e_mbd; - const int mis = cm->mode_info_stride; - int mi_width = num_8x8_blocks_wide_lookup[cpi->sf.always_this_block_size]; - int mi_height = num_8x8_blocks_high_lookup[cpi->sf.always_this_block_size]; + int mis = cm->mode_info_stride; + int br, bc; int i, j; int chosen_rate = INT_MAX; - int64_t chosen_dist = INT_MAX; + int64_t chosen_dist = INT64_MAX; MB_PREDICTION_MODE mode = DC_PRED; - int row8x8_remaining = tile->mi_row_end - mi_row; - int col8x8_remaining = tile->mi_col_end - mi_col; - int b32i; - for (b32i = 0; b32i < 4; b32i++) { - int b16i; - for (b16i = 0; b16i < 4; b16i++) { - int b8i; - int block_row = get_block_row(b32i, b16i, 0); - int block_col = get_block_col(b32i, b16i, 0); - int index = block_row * mis + block_col; - int rate; - int64_t dist; - - // Find a partition size that fits - bsize = find_partition_size(cpi->sf.always_this_block_size, - (row8x8_remaining - block_row), - (col8x8_remaining - block_col), - &mi_height, &mi_width); - mi_8x8[index] = mi_8x8[0] + index; - - set_mi_row_col(xd, tile, mi_row + block_row, mi_height, - mi_col + block_col, mi_width, cm->mi_rows, cm->mi_cols); - - xd->mi_8x8 = mi_8x8 + index; - - if (cm->frame_type != KEY_FRAME) { - set_offsets(cpi, tile, mi_row + block_row, mi_col + block_col, bsize); - - vp9_pick_inter_mode(cpi, x, tile, - mi_row + block_row, mi_col + block_col, - &rate, &dist, bsize); - } else { - set_mode_info(&mi_8x8[index]->mbmi, bsize, mode, - mi_row + block_row, mi_col + block_col); - } + int rows = MIN(MI_BLOCK_SIZE, tile->mi_row_end - mi_row); + int cols = MIN(MI_BLOCK_SIZE, tile->mi_col_end - mi_col); + + int mi_8x8_width = num_8x8_blocks_wide_lookup[bsize]; + int mi_8x8_hight = num_8x8_blocks_high_lookup[bsize]; + + int brate; + int64_t bdist; + *rate = 0; + *dist = 0; + + // find prediction mode for each 8x8 block + for (br = 0; br < rows; br += mi_8x8_hight) { + for (bc = 0; bc < cols; bc += mi_8x8_width) { + int row = mi_row + br; + int col = mi_col + bc; + int bh = 0, bw = 0; + BLOCK_SIZE bs = find_partition_size(bsize, rows - br, cols - bc, + &bh, &bw); + + set_offsets(cpi, tile, row, col, bs); + + if (cm->frame_type != KEY_FRAME) + vp9_pick_inter_mode(cpi, x, tile, row, col, &brate, &bdist, bs); + else + set_mode_info(&xd->mi_8x8[0]->mbmi, bs, mode, row, col); - for (j = 0; j < mi_height; j++) - for (i = 0; i < mi_width; i++) - if ((xd->mb_to_right_edge >> (3 + MI_SIZE_LOG2)) + mi_width > i - && (xd->mb_to_bottom_edge >> (3 + MI_SIZE_LOG2)) + mi_height > j) { - mi_8x8[index+ i + j * mis] = mi_8x8[index]; - } + *rate += brate; + *dist += bdist; - for (b8i = 0; b8i < 4; b8i++) { - } + for (j = 0; j < bh; ++j) + for (i = 0; i < bw; ++i) + xd->mi_8x8[j * mis + i] = xd->mi_8x8[0]; } } + encode_sb_rt(cpi, tile, tp, mi_row, mi_col, 1, BLOCK_64X64); *rate = chosen_rate; @@ -2359,9 +2328,9 @@ static void encode_rtc_sb_row(VP9_COMP *cpi, const TileInfo *const tile, MODE_INFO **mi_8x8 = cm->mi_grid_visible + idx_str; cpi->mb.source_variance = UINT_MAX; - set_partitioning(cpi, tile, mi_8x8, mi_row, mi_col); - rtc_use_partition(cpi, tile, mi_8x8, tp, mi_row, mi_col, BLOCK_64X64, - &dummy_rate, &dummy_dist, 1); + rtc_use_partition(cpi, tile, mi_8x8, tp, mi_row, mi_col, + cpi->sf.always_this_block_size, + &dummy_rate, &dummy_dist, 1); } } // end RTC play code @@ -2417,6 +2386,22 @@ static void encode_frame_internal(VP9_COMP *cpi) { set_prev_mi(cm); + if (cpi->sf.use_pick_mode) { + // Initialize internal buffer pointers for rtc coding, where non-RD + // mode decision is used and hence no buffer pointer swap needed. + int i; + struct macroblock_plane *const p = x->plane; + struct macroblockd_plane *const pd = xd->plane; + PICK_MODE_CONTEXT *ctx = &cpi->mb.sb64_context; + + for (i = 0; i < MAX_MB_PLANE; ++i) { + p[i].coeff = ctx->coeff_pbuf[i][0]; + p[i].qcoeff = ctx->qcoeff_pbuf[i][0]; + pd[i].dqcoeff = ctx->dqcoeff_pbuf[i][0]; + p[i].eobs = ctx->eobs_pbuf[i][0]; + } + } + { struct vpx_usec_timer emr_timer; vpx_usec_timer_start(&emr_timer); @@ -2436,7 +2421,7 @@ static void encode_frame_internal(VP9_COMP *cpi) { // For each row of SBs in the frame vp9_tile_init(&tile, cm, tile_row, tile_col); for (mi_row = tile.mi_row_start; - mi_row < tile.mi_row_end; mi_row += 8) { + mi_row < tile.mi_row_end; mi_row += MI_BLOCK_SIZE) { if (cpi->sf.use_pick_mode) encode_rtc_sb_row(cpi, &tile, mi_row, &tp); else @@ -2700,6 +2685,7 @@ static void encode_superblock(VP9_COMP *cpi, TOKENEXTRA **t, int output_enabled, const int mis = cm->mode_info_stride; const int mi_width = num_8x8_blocks_wide_lookup[bsize]; const int mi_height = num_8x8_blocks_high_lookup[bsize]; + x->skip_recode = !x->select_txfm_size && mbmi->sb_type >= BLOCK_8X8 && (cpi->oxcf.aq_mode != COMPLEXITY_AQ) && !cpi->sf.use_pick_mode; @@ -2739,6 +2725,7 @@ static void encode_superblock(VP9_COMP *cpi, TOKENEXTRA **t, int output_enabled, vp9_encode_intra_block_plane(x, MAX(bsize, BLOCK_8X8), plane); if (output_enabled) sum_intra_stats(&cm->counts, mi); + vp9_tokenize_sb(cpi, t, !output_enabled, MAX(bsize, BLOCK_8X8)); } else { int ref; const int is_compound = has_second_ref(mbmi); @@ -2748,19 +2735,17 @@ static void encode_superblock(VP9_COMP *cpi, TOKENEXTRA **t, int output_enabled, setup_pre_planes(xd, ref, cfg, mi_row, mi_col, &xd->block_refs[ref]->sf); } vp9_build_inter_predictors_sb(xd, mi_row, mi_col, MAX(bsize, BLOCK_8X8)); - } - if (!is_inter_block(mbmi)) { - vp9_tokenize_sb(cpi, t, !output_enabled, MAX(bsize, BLOCK_8X8)); - } else if (!x->skip) { - mbmi->skip = 1; - vp9_encode_sb(x, MAX(bsize, BLOCK_8X8)); - vp9_tokenize_sb(cpi, t, !output_enabled, MAX(bsize, BLOCK_8X8)); - } else { - mbmi->skip = 1; - if (output_enabled) - cm->counts.skip[vp9_get_skip_context(xd)][1]++; - reset_skip_context(xd, MAX(bsize, BLOCK_8X8)); + if (!x->skip) { + mbmi->skip = 1; + vp9_encode_sb(x, MAX(bsize, BLOCK_8X8)); + vp9_tokenize_sb(cpi, t, !output_enabled, MAX(bsize, BLOCK_8X8)); + } else { + mbmi->skip = 1; + if (output_enabled) + cm->counts.skip[vp9_get_skip_context(xd)][1]++; + reset_skip_context(xd, MAX(bsize, BLOCK_8X8)); + } } if (output_enabled) { diff --git a/vp9/encoder/vp9_encodemb.c b/vp9/encoder/vp9_encodemb.c index 6f976f584..19421aaac 100644 --- a/vp9/encoder/vp9_encodemb.c +++ b/vp9/encoder/vp9_encodemb.c @@ -345,6 +345,15 @@ static void optimize_init_b(int plane, BLOCK_SIZE bsize, pd->above_context, pd->left_context, num_4x4_w, num_4x4_h); } + +static INLINE void fdct32x32(int rd_transform, + const int16_t *src, int16_t *dst, int src_stride) { + if (rd_transform) + vp9_fdct32x32_rd(src, dst, src_stride); + else + vp9_fdct32x32(src, dst, src_stride); +} + void vp9_xform_quant(MACROBLOCK *x, int plane, int block, BLOCK_SIZE plane_bsize, TX_SIZE tx_size) { MACROBLOCKD *const xd = &x->e_mbd; @@ -364,10 +373,7 @@ void vp9_xform_quant(MACROBLOCK *x, int plane, int block, switch (tx_size) { case TX_32X32: scan_order = &vp9_default_scan_orders[TX_32X32]; - if (x->use_lp32x32fdct) - vp9_fdct32x32_rd(src_diff, coeff, diff_stride); - else - vp9_fdct32x32(src_diff, coeff, diff_stride); + fdct32x32(x->use_lp32x32fdct, src_diff, coeff, diff_stride); vp9_quantize_b_32x32(coeff, 1024, x->skip_block, p->zbin, p->round, p->quant, p->quant_shift, qcoeff, dqcoeff, pd->dequant, p->zbin_extra, eob, scan_order->scan, @@ -533,10 +539,12 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, uint8_t *src, *dst; int16_t *src_diff; uint16_t *eob = &p->eobs[block]; + const int src_stride = p->src.stride; + const int dst_stride = pd->dst.stride; int i, j; txfrm_block_to_raster_xy(plane_bsize, tx_size, block, &i, &j); - dst = &pd->dst.buf[4 * (j * pd->dst.stride + i)]; - src = &p->src.buf[4 * (j * p->src.stride + i)]; + dst = &pd->dst.buf[4 * (j * dst_stride + i)]; + src = &p->src.buf[4 * (j * src_stride + i)]; src_diff = &p->src_diff[4 * (j * diff_stride + i)]; // if (x->optimize) @@ -548,22 +556,19 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, mode = plane == 0 ? mbmi->mode : mbmi->uv_mode; vp9_predict_intra_block(xd, block >> 6, bwl, TX_32X32, mode, x->skip_encode ? src : dst, - x->skip_encode ? p->src.stride : pd->dst.stride, - dst, pd->dst.stride, i, j, plane); + x->skip_encode ? src_stride : dst_stride, + dst, dst_stride, i, j, plane); if (!x->skip_recode) { vp9_subtract_block(32, 32, src_diff, diff_stride, - src, p->src.stride, dst, pd->dst.stride); - if (x->use_lp32x32fdct) - vp9_fdct32x32_rd(src_diff, coeff, diff_stride); - else - vp9_fdct32x32(src_diff, coeff, diff_stride); + src, src_stride, dst, dst_stride); + fdct32x32(x->use_lp32x32fdct, src_diff, coeff, diff_stride); vp9_quantize_b_32x32(coeff, 1024, x->skip_block, p->zbin, p->round, p->quant, p->quant_shift, qcoeff, dqcoeff, pd->dequant, p->zbin_extra, eob, scan_order->scan, scan_order->iscan); } if (!x->skip_encode && *eob) - vp9_idct32x32_add(dqcoeff, dst, pd->dst.stride, *eob); + vp9_idct32x32_add(dqcoeff, dst, dst_stride, *eob); break; case TX_16X16: tx_type = get_tx_type_16x16(pd->plane_type, xd); @@ -571,11 +576,11 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, mode = plane == 0 ? mbmi->mode : mbmi->uv_mode; vp9_predict_intra_block(xd, block >> 4, bwl, TX_16X16, mode, x->skip_encode ? src : dst, - x->skip_encode ? p->src.stride : pd->dst.stride, - dst, pd->dst.stride, i, j, plane); + x->skip_encode ? src_stride : dst_stride, + dst, dst_stride, i, j, plane); if (!x->skip_recode) { vp9_subtract_block(16, 16, src_diff, diff_stride, - src, p->src.stride, dst, pd->dst.stride); + src, src_stride, dst, dst_stride); vp9_fht16x16(src_diff, coeff, diff_stride, tx_type); vp9_quantize_b(coeff, 256, x->skip_block, p->zbin, p->round, p->quant, p->quant_shift, qcoeff, dqcoeff, @@ -583,7 +588,7 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, scan_order->iscan); } if (!x->skip_encode && *eob) - vp9_iht16x16_add(tx_type, dqcoeff, dst, pd->dst.stride, *eob); + vp9_iht16x16_add(tx_type, dqcoeff, dst, dst_stride, *eob); break; case TX_8X8: tx_type = get_tx_type_8x8(pd->plane_type, xd); @@ -591,11 +596,11 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, mode = plane == 0 ? mbmi->mode : mbmi->uv_mode; vp9_predict_intra_block(xd, block >> 2, bwl, TX_8X8, mode, x->skip_encode ? src : dst, - x->skip_encode ? p->src.stride : pd->dst.stride, - dst, pd->dst.stride, i, j, plane); + x->skip_encode ? src_stride : dst_stride, + dst, dst_stride, i, j, plane); if (!x->skip_recode) { vp9_subtract_block(8, 8, src_diff, diff_stride, - src, p->src.stride, dst, pd->dst.stride); + src, src_stride, dst, dst_stride); vp9_fht8x8(src_diff, coeff, diff_stride, tx_type); vp9_quantize_b(coeff, 64, x->skip_block, p->zbin, p->round, p->quant, p->quant_shift, qcoeff, dqcoeff, @@ -603,7 +608,7 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, scan_order->iscan); } if (!x->skip_encode && *eob) - vp9_iht8x8_add(tx_type, dqcoeff, dst, pd->dst.stride, *eob); + vp9_iht8x8_add(tx_type, dqcoeff, dst, dst_stride, *eob); break; case TX_4X4: tx_type = get_tx_type_4x4(pd->plane_type, xd, block); @@ -615,12 +620,12 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, vp9_predict_intra_block(xd, block, bwl, TX_4X4, mode, x->skip_encode ? src : dst, - x->skip_encode ? p->src.stride : pd->dst.stride, - dst, pd->dst.stride, i, j, plane); + x->skip_encode ? src_stride : dst_stride, + dst, dst_stride, i, j, plane); if (!x->skip_recode) { vp9_subtract_block(4, 4, src_diff, diff_stride, - src, p->src.stride, dst, pd->dst.stride); + src, src_stride, dst, dst_stride); if (tx_type != DCT_DCT) vp9_fht4x4(src_diff, coeff, diff_stride, tx_type); else @@ -636,9 +641,9 @@ static void encode_block_intra(int plane, int block, BLOCK_SIZE plane_bsize, // this is like vp9_short_idct4x4 but has a special case around eob<=1 // which is significant (not just an optimization) for the lossless // case. - xd->itxm_add(dqcoeff, dst, pd->dst.stride, *eob); + xd->itxm_add(dqcoeff, dst, dst_stride, *eob); else - vp9_iht4x4_16_add(dqcoeff, dst, pd->dst.stride, tx_type); + vp9_iht4x4_16_add(dqcoeff, dst, dst_stride, tx_type); } break; default: diff --git a/vp9/encoder/vp9_firstpass.c b/vp9/encoder/vp9_firstpass.c index 556197cd4..ddb901dd1 100644 --- a/vp9/encoder/vp9_firstpass.c +++ b/vp9/encoder/vp9_firstpass.c @@ -65,7 +65,7 @@ static int select_cq_level(int qindex) { double target_q = (vp9_convert_qindex_to_q(qindex) * 0.5847) + 1.0; - for (i = 0; i < QINDEX_RANGE; i++) { + for (i = 0; i < QINDEX_RANGE; ++i) { if (target_q <= vp9_convert_qindex_to_q(i)) { ret_val = i; break; @@ -399,7 +399,7 @@ static void first_pass_motion_search(VP9_COMP *cpi, MACROBLOCK *x, // Refine the motion search range according to the frame dimension // for first pass test. while ((quart_frm << sr) < MAX_FULL_PEL_VAL) - sr++; + ++sr; step_param += sr; further_steps -= sr; @@ -427,10 +427,10 @@ static void first_pass_motion_search(VP9_COMP *cpi, MACROBLOCK *x, num00 = 0; while (n < further_steps) { - n++; + ++n; if (num00) { - num00--; + --num00; } else { tmp_err = cpi->diamond_search_sad(x, &ref_mv_full, &tmp_mv, step_param + n, x->sadperbit16, @@ -495,7 +495,7 @@ void vp9_first_pass(VP9_COMP *cpi) { struct twopass_rc *const twopass = &cpi->twopass; const MV zero_mv = {0, 0}; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); vp9_setup_src_planes(x, cpi->Source, 0, 0); setup_pre_planes(xd, 0, lst_yv12, 0, 0, NULL); @@ -522,7 +522,7 @@ void vp9_first_pass(VP9_COMP *cpi) { // Tiling is ignored in the first pass. vp9_tile_init(&tile, cm, 0, 0); - for (mb_row = 0; mb_row < cm->mb_rows; mb_row++) { + for (mb_row = 0; mb_row < cm->mb_rows; ++mb_row) { int_mv best_ref_mv; best_ref_mv.as_int = 0; @@ -538,13 +538,13 @@ void vp9_first_pass(VP9_COMP *cpi) { x->mv_row_max = ((cm->mb_rows - 1 - mb_row) * 16) + BORDER_MV_PIXELS_B16; - for (mb_col = 0; mb_col < cm->mb_cols; mb_col++) { + for (mb_col = 0; mb_col < cm->mb_cols; ++mb_col) { int this_error; const int use_dc_pred = (mb_col || mb_row) && (!mb_col || !mb_row); double error_weight = 1.0; const BLOCK_SIZE bsize = get_bsize(cm, mb_row, mb_col); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); xd->plane[0].dst.buf = new_yv12->y_buffer + recon_yoffset; xd->plane[1].dst.buf = new_yv12->u_buffer + recon_uvoffset; @@ -565,7 +565,7 @@ void vp9_first_pass(VP9_COMP *cpi) { // Do intra 16x16 prediction. this_error = vp9_encode_intra(x, use_dc_pred); if (cpi->oxcf.aq_mode == VARIANCE_AQ) { - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); this_error = (int)(this_error * error_weight); } @@ -601,7 +601,7 @@ void vp9_first_pass(VP9_COMP *cpi) { first_pass_motion_search(cpi, x, &best_ref_mv.as_mv, &mv.as_mv, &motion_error); if (cpi->oxcf.aq_mode == VARIANCE_AQ) { - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); motion_error = (int)(motion_error * error_weight); } @@ -612,7 +612,7 @@ void vp9_first_pass(VP9_COMP *cpi) { first_pass_motion_search(cpi, x, &zero_mv, &tmp_mv.as_mv, &tmp_err); if (cpi->oxcf.aq_mode == VARIANCE_AQ) { - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); tmp_err = (int)(tmp_err * error_weight); } @@ -633,12 +633,12 @@ void vp9_first_pass(VP9_COMP *cpi) { first_pass_motion_search(cpi, x, &zero_mv, &tmp_mv.as_mv, &gf_motion_error); if (cpi->oxcf.aq_mode == VARIANCE_AQ) { - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); gf_motion_error = (int)(gf_motion_error * error_weight); } if (gf_motion_error < motion_error && gf_motion_error < this_error) - second_ref_count++; + ++second_ref_count; // Reset to last frame as reference buffer. xd->plane[0].pre[0].buf = lst_yv12->y_buffer + recon_yoffset; @@ -665,7 +665,7 @@ void vp9_first_pass(VP9_COMP *cpi) { // cropped clips with black bars at the sides or top and bottom. if (((this_error - intrapenalty) * 9 <= motion_error * 10) && this_error < 2 * intrapenalty) - neutral_count++; + ++neutral_count; mv.as_mv.row *= 8; mv.as_mv.col *= 8; @@ -682,42 +682,42 @@ void vp9_first_pass(VP9_COMP *cpi) { sum_mvc_abs += abs(mv.as_mv.col); sum_mvrs += mv.as_mv.row * mv.as_mv.row; sum_mvcs += mv.as_mv.col * mv.as_mv.col; - intercount++; + ++intercount; best_ref_mv.as_int = mv.as_int; if (mv.as_int) { - mvcount++; + ++mvcount; // Non-zero vector, was it different from the last non zero vector? if (mv.as_int != lastmv_as_int) - new_mv_count++; + ++new_mv_count; lastmv_as_int = mv.as_int; // Does the row vector point inwards or outwards? if (mb_row < cm->mb_rows / 2) { if (mv.as_mv.row > 0) - sum_in_vectors--; + --sum_in_vectors; else if (mv.as_mv.row < 0) - sum_in_vectors++; + ++sum_in_vectors; } else if (mb_row > cm->mb_rows / 2) { if (mv.as_mv.row > 0) - sum_in_vectors++; + ++sum_in_vectors; else if (mv.as_mv.row < 0) - sum_in_vectors--; + --sum_in_vectors; } // Does the col vector point inwards or outwards? if (mb_col < cm->mb_cols / 2) { if (mv.as_mv.col > 0) - sum_in_vectors--; + --sum_in_vectors; else if (mv.as_mv.col < 0) - sum_in_vectors++; + ++sum_in_vectors; } else if (mb_col > cm->mb_cols / 2) { if (mv.as_mv.col > 0) - sum_in_vectors++; + ++sum_in_vectors; else if (mv.as_mv.col < 0) - sum_in_vectors--; + --sum_in_vectors; } } } @@ -742,10 +742,10 @@ void vp9_first_pass(VP9_COMP *cpi) { x->plane[2].src.buf += uv_mb_height * x->plane[1].src.stride - uv_mb_height * cm->mb_cols; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); } - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); { FIRSTPASS_STATS fps; @@ -802,7 +802,7 @@ void vp9_first_pass(VP9_COMP *cpi) { vp8_yv12_copy_frame(lst_yv12, gld_yv12); twopass->sr_update_lag = 1; } else { - twopass->sr_update_lag++; + ++twopass->sr_update_lag; } // Swap frame pointers so last frame refers to the frame we just compressed. swap_yv12(lst_yv12, new_yv12); @@ -830,7 +830,7 @@ void vp9_first_pass(VP9_COMP *cpi) { fclose(recon_file); } - cm->current_video_frame++; + ++cm->current_video_frame; } // Estimate a cost per mb attributable to overheads such as the coding of modes @@ -910,7 +910,7 @@ int vp9_twopass_worst_quality(VP9_COMP *cpi, FIRSTPASS_STATS *fpstats, // Try and pick a max Q that will be high enough to encode the // content at the given rate. - for (q = rc->best_quality; q < rc->worst_quality; q++) { + for (q = rc->best_quality; q < rc->worst_quality; ++q) { const double err_correction_factor = calc_correction_factor(err_per_mb, ERR_DIVISOR, 0.5, 0.90, q); const int bits_per_mb_at_this_q = vp9_rc_bits_per_mb(INTER_FRAME, q, @@ -1045,7 +1045,7 @@ static int detect_transition_to_still(VP9_COMP *cpi, int frame_interval, FIRSTPASS_STATS tmp_next_frame; // Look ahead a few frames to see if static condition persists... - for (j = 0; j < still_interval; j++) { + for (j = 0; j < still_interval; ++j) { if (EOF == input_stats(&cpi->twopass, &tmp_next_frame)) break; @@ -1164,7 +1164,7 @@ static int calc_arf_boost(VP9_COMP *cpi, int offset, int flash_detected = 0; // Search forward from the proposed arf/next gf position. - for (i = 0; i < f_frames; i++) { + for (i = 0; i < f_frames; ++i) { if (read_frame_stats(twopass, &this_frame, (i + offset)) == EOF) break; @@ -1201,7 +1201,7 @@ static int calc_arf_boost(VP9_COMP *cpi, int offset, abs_mv_in_out_accumulator = 0.0; // Search backward towards last gf position. - for (i = -1; i >= -b_frames; i--) { + for (i = -1; i >= -b_frames; --i) { if (read_frame_stats(twopass, &this_frame, (i + offset)) == EOF) break; @@ -1409,7 +1409,7 @@ static void define_gf_group(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { twopass->gf_group_bits = 0; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); start_pos = twopass->stats_in; @@ -1443,7 +1443,7 @@ static void define_gf_group(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { i = 0; while (i < twopass->static_scene_max_gf_interval && i < rc->frames_to_key) { - i++; + ++i; // Accumulate error score of frames in this gf group. mod_frame_err = calculate_modified_err(cpi, this_frame); @@ -1515,7 +1515,7 @@ static void define_gf_group(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { // Don't allow a gf too near the next kf. if ((rc->frames_to_key - i) < MIN_GF_INTERVAL) { while (i < (rc->frames_to_key + !rc->next_key_frame_forced)) { - i++; + ++i; if (EOF == input_stats(twopass, this_frame)) break; @@ -1752,7 +1752,7 @@ static void define_gf_group(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { zero_stats(§ionstats); reset_fpf_position(twopass, start_pos); - for (i = 0; i < rc->baseline_gf_interval; i++) { + for (i = 0; i < rc->baseline_gf_interval; ++i) { input_stats(twopass, &next_frame); accumulate_stats(§ionstats, &next_frame); } @@ -1837,7 +1837,7 @@ static int test_candidate_kf(VP9_COMP *cpi, start_pos = cpi->twopass.stats_in; // Examine how well the key frame predicts subsequent frames. - for (i = 0; i < 16; i++) { + for (i = 0; i < 16; ++i) { double next_iiratio = (IIKFACTOR1 * local_next_frame.intra_error / DOUBLE_DIVIDE_CHECK(local_next_frame.coded_error)); @@ -1907,7 +1907,7 @@ static void find_next_key_frame(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { vp9_zero(next_frame); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); start_position = twopass->stats_in; cpi->common.frame_type = KEY_FRAME; @@ -1956,7 +1956,7 @@ static void find_next_key_frame(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { // quality since the last GF or KF. recent_loop_decay[i % 8] = loop_decay_rate; decay_accumulator = 1.0; - for (j = 0; j < 8; j++) + for (j = 0; j < 8; ++j) decay_accumulator *= recent_loop_decay[j]; // Special check for transition or high motion followed by a @@ -1966,7 +1966,7 @@ static void find_next_key_frame(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { break; // Step on to the next frame. - rc->frames_to_key++; + ++rc->frames_to_key; // If we don't have a real key frame within the next two // key_frame_frequency intervals then break out of the loop. @@ -1975,7 +1975,7 @@ static void find_next_key_frame(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { } else { ++rc->frames_to_key; } - i++; + ++i; } // If there is a max kf interval set by the user we must obey it. @@ -1997,7 +1997,7 @@ static void find_next_key_frame(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { kf_group_err = 0; // Rescan to get the correct error data for the forced kf group. - for (i = 0; i < rc->frames_to_key; i++) { + for (i = 0; i < rc->frames_to_key; ++i) { // Accumulate kf group errors. kf_group_err += calculate_modified_err(cpi, &tmp_frame); @@ -2046,7 +2046,7 @@ static void find_next_key_frame(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { boost_score = 0.0; // Scan through the kf group collating various stats. - for (i = 0; i < rc->frames_to_key; i++) { + for (i = 0; i < rc->frames_to_key; ++i) { double r; if (EOF == input_stats(twopass, &next_frame)) @@ -2089,7 +2089,7 @@ static void find_next_key_frame(VP9_COMP *cpi, FIRSTPASS_STATS *this_frame) { zero_stats(§ionstats); reset_fpf_position(twopass, start_position); - for (i = 0; i < rc->frames_to_key; i++) { + for (i = 0; i < rc->frames_to_key; ++i) { input_stats(twopass, &next_frame); accumulate_stats(§ionstats, &next_frame); } diff --git a/vp9/encoder/vp9_mbgraph.c b/vp9/encoder/vp9_mbgraph.c index 7eacda217..44c1f9078 100644 --- a/vp9/encoder/vp9_mbgraph.c +++ b/vp9/encoder/vp9_mbgraph.c @@ -419,7 +419,7 @@ void vp9_update_mbgraph_stats(VP9_COMP *cpi) { golden_ref, cpi->Source); } - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); separate_arf_mbs(cpi); } diff --git a/vp9/encoder/vp9_mcomp.c b/vp9/encoder/vp9_mcomp.c index 8b45f8b3e..f18f55042 100644 --- a/vp9/encoder/vp9_mcomp.c +++ b/vp9/encoder/vp9_mcomp.c @@ -855,6 +855,184 @@ int vp9_square_search(const MACROBLOCK *x, square_num_candidates, square_candidates); }; +// Number of candidates in first hex search +#define FIRST_HEX_CANDIDATES 6 +// Index of previous hex search's best match +#define PRE_BEST_CANDIDATE 6 +// Number of candidates in following hex search +#define NEXT_HEX_CANDIDATES 3 +// Number of candidates in refining search +#define REFINE_CANDIDATES 4 + +int vp9_fast_hex_search(const MACROBLOCK *x, + MV *ref_mv, + int search_param, + int sad_per_bit, + const vp9_variance_fn_ptr_t *vfp, + int use_mvcost, + const MV *center_mv, + MV *best_mv) { + const MACROBLOCKD* const xd = &x->e_mbd; + static const MV hex[FIRST_HEX_CANDIDATES] = { + { -1, -2}, {1, -2}, {2, 0}, {1, 2}, { -1, 2}, { -2, 0} + }; + static const MV next_chkpts[PRE_BEST_CANDIDATE][NEXT_HEX_CANDIDATES] = { + {{ -2, 0}, { -1, -2}, {1, -2}}, + {{ -1, -2}, {1, -2}, {2, 0}}, + {{1, -2}, {2, 0}, {1, 2}}, + {{2, 0}, {1, 2}, { -1, 2}}, + {{1, 2}, { -1, 2}, { -2, 0}}, + {{ -1, 2}, { -2, 0}, { -1, -2}} + }; + static const MV neighbors[REFINE_CANDIDATES] = { + {0, -1}, { -1, 0}, {1, 0}, {0, 1} + }; + int i, j; + + const uint8_t *what = x->plane[0].src.buf; + const int what_stride = x->plane[0].src.stride; + const int in_what_stride = xd->plane[0].pre[0].stride; + int br, bc; + MV this_mv; + unsigned int bestsad = 0x7fffffff; + unsigned int thissad; + const uint8_t *base_offset; + const uint8_t *this_offset; + int k = -1; + int best_site = -1; + const int max_hex_search = 512; + const int max_dia_search = 32; + + const int *mvjsadcost = x->nmvjointsadcost; + int *mvsadcost[2] = {x->nmvsadcost[0], x->nmvsadcost[1]}; + + const MV fcenter_mv = {center_mv->row >> 3, center_mv->col >> 3}; + + // Adjust ref_mv to make sure it is within MV range + clamp_mv(ref_mv, x->mv_col_min, x->mv_col_max, x->mv_row_min, x->mv_row_max); + br = ref_mv->row; + bc = ref_mv->col; + + // Check the start point + base_offset = xd->plane[0].pre[0].buf; + this_offset = base_offset + (br * in_what_stride) + bc; + this_mv.row = br; + this_mv.col = bc; + bestsad = vfp->sdf(what, what_stride, this_offset, in_what_stride, 0x7fffffff) + + mvsad_err_cost(&this_mv, &fcenter_mv, mvjsadcost, mvsadcost, + sad_per_bit); + + // Initial 6-point hex search + if (check_bounds(x, br, bc, 2)) { + for (i = 0; i < FIRST_HEX_CANDIDATES; i++) { + this_mv.row = br + hex[i].row; + this_mv.col = bc + hex[i].col; + this_offset = base_offset + (this_mv.row * in_what_stride) + this_mv.col; + thissad = vfp->sdf(what, what_stride, this_offset, in_what_stride, + bestsad); + CHECK_BETTER + } + } else { + for (i = 0; i < FIRST_HEX_CANDIDATES; i++) { + this_mv.row = br + hex[i].row; + this_mv.col = bc + hex[i].col; + if (!is_mv_in(x, &this_mv)) + continue; + this_offset = base_offset + (this_mv.row * in_what_stride) + this_mv.col; + thissad = vfp->sdf(what, what_stride, this_offset, in_what_stride, + bestsad); + CHECK_BETTER + } + } + + // Continue hex search if we find a better match in first round + if (best_site != -1) { + br += hex[best_site].row; + bc += hex[best_site].col; + k = best_site; + + // Allow search covering maximum MV range + for (j = 1; j < max_hex_search; j++) { + best_site = -1; + + if (check_bounds(x, br, bc, 2)) { + for (i = 0; i < 3; i++) { + this_mv.row = br + next_chkpts[k][i].row; + this_mv.col = bc + next_chkpts[k][i].col; + this_offset = base_offset + (this_mv.row * in_what_stride) + + this_mv.col; + thissad = vfp->sdf(what, what_stride, this_offset, in_what_stride, + bestsad); + CHECK_BETTER + } + } else { + for (i = 0; i < 3; i++) { + this_mv.row = br + next_chkpts[k][i].row; + this_mv.col = bc + next_chkpts[k][i].col; + if (!is_mv_in(x, &this_mv)) + continue; + this_offset = base_offset + (this_mv.row * in_what_stride) + + this_mv.col; + thissad = vfp->sdf(what, what_stride, this_offset, in_what_stride, + bestsad); + CHECK_BETTER + } + } + + if (best_site == -1) { + break; + } else { + br += next_chkpts[k][best_site].row; + bc += next_chkpts[k][best_site].col; + k += 5 + best_site; + if (k >= 12) k -= 12; + else if (k >= 6) k -= 6; + } + } + } + + // Check 4 1-away neighbors + for (j = 0; j < max_dia_search; j++) { + best_site = -1; + + if (check_bounds(x, br, bc, 1)) { + for (i = 0; i < REFINE_CANDIDATES; i++) { + this_mv.row = br + neighbors[i].row; + this_mv.col = bc + neighbors[i].col; + this_offset = base_offset + (this_mv.row * in_what_stride) + + this_mv.col; + thissad = vfp->sdf(what, what_stride, this_offset, in_what_stride, + bestsad); + CHECK_BETTER + } + } else { + for (i = 0; i < REFINE_CANDIDATES; i++) { + this_mv.row = br + neighbors[i].row; + this_mv.col = bc + neighbors[i].col; + if (!is_mv_in(x, &this_mv)) + continue; + this_offset = base_offset + (this_mv.row * in_what_stride) + + this_mv.col; + thissad = vfp->sdf(what, what_stride, this_offset, in_what_stride, + bestsad); + CHECK_BETTER + } + } + + if (best_site == -1) { + break; + } else { + br += neighbors[best_site].row; + bc += neighbors[best_site].col; + } + } + + best_mv->row = br; + best_mv->col = bc; + + return bestsad; +} + #undef CHECK_BETTER int vp9_full_range_search_c(const MACROBLOCK *x, MV *ref_mv, MV *best_mv, diff --git a/vp9/encoder/vp9_mcomp.h b/vp9/encoder/vp9_mcomp.h index 4414f3d49..ff4b1df75 100644 --- a/vp9/encoder/vp9_mcomp.h +++ b/vp9/encoder/vp9_mcomp.h @@ -75,6 +75,14 @@ int vp9_square_search(const MACROBLOCK *x, int use_mvcost, const MV *center_mv, MV *best_mv); +int vp9_fast_hex_search(const MACROBLOCK *x, + MV *ref_mv, + int search_param, + int sad_per_bit, + const vp9_variance_fn_ptr_t *vfp, + int use_mvcost, + const MV *center_mv, + MV *best_mv); typedef int (fractional_mv_step_fp) ( const MACROBLOCK *x, diff --git a/vp9/encoder/vp9_onyx_if.c b/vp9/encoder/vp9_onyx_if.c index 4561e7671..8f3d82570 100644 --- a/vp9/encoder/vp9_onyx_if.c +++ b/vp9/encoder/vp9_onyx_if.c @@ -154,20 +154,22 @@ void vp9_initialize_enc() { } static void dealloc_compressor_data(VP9_COMP *cpi) { + VP9_COMMON *const cm = &cpi->common; + // Delete sementation map vpx_free(cpi->segmentation_map); - cpi->segmentation_map = 0; - vpx_free(cpi->common.last_frame_seg_map); - cpi->common.last_frame_seg_map = 0; + cpi->segmentation_map = NULL; + vpx_free(cm->last_frame_seg_map); + cm->last_frame_seg_map = NULL; vpx_free(cpi->coding_context.last_frame_seg_map_copy); - cpi->coding_context.last_frame_seg_map_copy = 0; + cpi->coding_context.last_frame_seg_map_copy = NULL; vpx_free(cpi->complexity_map); cpi->complexity_map = 0; vpx_free(cpi->active_map); cpi->active_map = 0; - vp9_free_frame_buffers(&cpi->common); + vp9_free_frame_buffers(cm); vp9_free_frame_buffer(&cpi->last_frame_uf); vp9_free_frame_buffer(&cpi->scaled_source); @@ -194,19 +196,20 @@ static void dealloc_compressor_data(VP9_COMP *cpi) { // to a target value // target q value int vp9_compute_qdelta(const VP9_COMP *cpi, double qstart, double qtarget) { + const RATE_CONTROL *const rc = &cpi->rc; + int start_index = rc->worst_quality; + int target_index = rc->worst_quality; int i; - int start_index = cpi->rc.worst_quality; - int target_index = cpi->rc.worst_quality; // Convert the average q value to an index. - for (i = cpi->rc.best_quality; i < cpi->rc.worst_quality; i++) { + for (i = rc->best_quality; i < rc->worst_quality; ++i) { start_index = i; if (vp9_convert_qindex_to_q(i) >= qstart) break; } // Convert the q target to an index - for (i = cpi->rc.best_quality; i < cpi->rc.worst_quality; i++) { + for (i = rc->best_quality; i < rc->worst_quality; ++i) { target_index = i; if (vp9_convert_qindex_to_q(i) >= qtarget) break; @@ -217,28 +220,24 @@ int vp9_compute_qdelta(const VP9_COMP *cpi, double qstart, double qtarget) { // Computes a q delta (in "q index" terms) to get from a starting q value // to a value that should equate to thegiven rate ratio. + static int compute_qdelta_by_rate(VP9_COMP *cpi, int base_q_index, double rate_target_ratio) { int i; - int base_bits_per_mb; - int target_bits_per_mb; int target_index = cpi->rc.worst_quality; - // Make SURE use of floating point in this function is safe. - vp9_clear_system_state(); - // Look up the current projected bits per block for the base index - base_bits_per_mb = vp9_rc_bits_per_mb(cpi->common.frame_type, - base_q_index, 1.0); + const int base_bits_per_mb = vp9_rc_bits_per_mb(cpi->common.frame_type, + base_q_index, 1.0); // Find the target bits per mb based on the base value and given ratio. - target_bits_per_mb = (int)(rate_target_ratio * base_bits_per_mb); + const int target_bits_per_mb = (int)(rate_target_ratio * base_bits_per_mb); // Convert the q target to an index - for (i = cpi->rc.best_quality; i < cpi->rc.worst_quality; i++) { + for (i = cpi->rc.best_quality; i < cpi->rc.worst_quality; ++i) { target_index = i; - if (vp9_rc_bits_per_mb(cpi->common.frame_type, - i, 1.0) <= target_bits_per_mb ) + if (vp9_rc_bits_per_mb(cpi->common.frame_type, i, 1.0) <= + target_bits_per_mb ) break; } @@ -248,11 +247,8 @@ static int compute_qdelta_by_rate(VP9_COMP *cpi, int base_q_index, // This function sets up a set of segments with delta Q values around // the baseline frame quantizer. static void setup_in_frame_q_adj(VP9_COMP *cpi) { - VP9_COMMON *cm = &cpi->common; - struct segmentation *seg = &cm->seg; - // double q_ratio; - int segment; - int qindex_delta; + VP9_COMMON *const cm = &cpi->common; + struct segmentation *const seg = &cm->seg; // Make SURE use of floating point in this function is safe. vp9_clear_system_state(); @@ -260,6 +256,8 @@ static void setup_in_frame_q_adj(VP9_COMP *cpi) { if (cm->frame_type == KEY_FRAME || cpi->refresh_alt_ref_frame || (cpi->refresh_golden_frame && !cpi->rc.is_src_frame_alt_ref)) { + int segment; + // Clear down the segment map vpx_memset(cpi->segmentation_map, 0, cm->mi_rows * cm->mi_cols); @@ -277,16 +275,16 @@ static void setup_in_frame_q_adj(VP9_COMP *cpi) { // Use some of the segments for in frame Q adjustment for (segment = 1; segment < 2; segment++) { - qindex_delta = compute_qdelta_by_rate(cpi, cm->base_qindex, - in_frame_q_adj_ratio[segment]); + const int qindex_delta = compute_qdelta_by_rate(cpi, cm->base_qindex, + in_frame_q_adj_ratio[segment]); vp9_enable_segfeature(seg, segment, SEG_LVL_ALT_Q); vp9_set_segdata(seg, segment, SEG_LVL_ALT_Q, qindex_delta); } } } static void configure_static_seg_features(VP9_COMP *cpi) { - VP9_COMMON *cm = &cpi->common; - struct segmentation *seg = &cm->seg; + VP9_COMMON *const cm = &cpi->common; + struct segmentation *const seg = &cm->seg; int high_q = (int)(cpi->rc.avg_q > 48.0); int qi_delta; @@ -430,13 +428,13 @@ static void print_seg_map(VP9_COMP *cpi) { static void update_reference_segmentation_map(VP9_COMP *cpi) { VP9_COMMON *const cm = &cpi->common; + MODE_INFO **mi_8x8_ptr = cm->mi_grid_visible; + uint8_t *cache_ptr = cm->last_frame_seg_map; int row, col; - MODE_INFO **mi_8x8, **mi_8x8_ptr = cm->mi_grid_visible; - uint8_t *cache_ptr = cm->last_frame_seg_map, *cache; for (row = 0; row < cm->mi_rows; row++) { - mi_8x8 = mi_8x8_ptr; - cache = cache_ptr; + MODE_INFO **mi_8x8 = mi_8x8_ptr; + uint8_t *cache = cache_ptr; for (col = 0; col < cm->mi_cols; col++, mi_8x8++, cache++) cache[0] = mi_8x8[0]->mbmi.segment_id; mi_8x8_ptr += cm->mode_info_stride; @@ -663,6 +661,7 @@ static void set_good_speed_feature(VP9_COMMON *cm, sf->reference_masking = 1; sf->auto_mv_step_size = 1; + sf->disable_split_var_thresh = 32; sf->disable_filter_search_var_thresh = 100; sf->comp_inter_joint_search_thresh = BLOCK_SIZES; @@ -698,6 +697,7 @@ static void set_good_speed_feature(VP9_COMMON *cm, sf->reference_masking = 1; sf->auto_mv_step_size = 1; + sf->disable_split_var_thresh = 64; sf->disable_filter_search_var_thresh = 200; sf->comp_inter_joint_search_thresh = BLOCK_SIZES; @@ -715,7 +715,7 @@ static void set_good_speed_feature(VP9_COMMON *cm, sf->adaptive_rd_thresh = 4; sf->mode_skip_start = 6; } - if (speed == 5) { + if (speed >= 5) { sf->comp_inter_joint_search_thresh = BLOCK_SIZES; sf->use_one_partition_size_always = 1; sf->always_this_block_size = BLOCK_16X16; @@ -859,11 +859,16 @@ static void set_rt_speed_feature(VP9_COMMON *cm, } sf->frame_parameter_update = 0; sf->encode_breakout_thresh = 1000; + + sf->search_method = FAST_HEX; } if (speed >= 6) { + sf->use_one_partition_size_always = 1; + sf->always_this_block_size = BLOCK_32X32; + } + if (speed >= 7) { sf->always_this_block_size = BLOCK_16X16; sf->use_pick_mode = 1; - sf->encode_breakout_thresh = 1000; } } @@ -877,8 +882,10 @@ void vp9_set_speed_features(VP9_COMP *cpi) { if (speed < 0) speed = -speed; +#if CONFIG_INTERNAL_STATS for (i = 0; i < MAX_MODES; ++i) cpi->mode_chosen_counts[i] = 0; +#endif // best quality defaults sf->frame_parameter_update = 1; @@ -970,20 +977,24 @@ void vp9_set_speed_features(VP9_COMP *cpi) { if (cpi->encode_breakout && cpi->oxcf.mode == MODE_REALTIME && sf->encode_breakout_thresh > cpi->encode_breakout) cpi->encode_breakout = sf->encode_breakout_thresh; + + if (sf->disable_split_mask == DISABLE_ALL_SPLIT) + sf->adaptive_pred_interp_filter = 0; } static void alloc_raw_frame_buffers(VP9_COMP *cpi) { VP9_COMMON *cm = &cpi->common; + const VP9_CONFIG *oxcf = &cpi->oxcf; - cpi->lookahead = vp9_lookahead_init(cpi->oxcf.width, cpi->oxcf.height, + cpi->lookahead = vp9_lookahead_init(oxcf->width, oxcf->height, cm->subsampling_x, cm->subsampling_y, - cpi->oxcf.lag_in_frames); + oxcf->lag_in_frames); if (!cpi->lookahead) vpx_internal_error(&cm->error, VPX_CODEC_MEM_ERROR, "Failed to allocate lag buffers"); if (vp9_realloc_frame_buffer(&cpi->alt_ref_buffer, - cpi->oxcf.width, cpi->oxcf.height, + oxcf->width, oxcf->height, cm->subsampling_x, cm->subsampling_y, VP9_ENC_BORDER_IN_PIXELS, NULL, NULL, NULL)) vpx_internal_error(&cm->error, VPX_CODEC_MEM_ERROR, @@ -1110,7 +1121,7 @@ int vp9_reverse_trans(int x) { void vp9_new_framerate(VP9_COMP *cpi, double framerate) { VP9_COMMON *const cm = &cpi->common; - int64_t vbr_max_bits; + int vbr_max_bits; if (framerate < 0.1) framerate = 30; @@ -1134,10 +1145,10 @@ void vp9_new_framerate(VP9_COMP *cpi, double framerate) { // be acheived because of a user specificed max q (e.g. when the user // specifies lossless encode. // - vbr_max_bits = ((int64_t)cpi->rc.av_per_frame_bandwidth * - (int64_t)cpi->oxcf.two_pass_vbrmax_section) / 100; + vbr_max_bits = (int)(((int64_t)cpi->rc.av_per_frame_bandwidth * + cpi->oxcf.two_pass_vbrmax_section) / 100); cpi->rc.max_frame_bandwidth = - MAX(MAX((cm->MBs * MAX_MB_RATE), MAXRATE_1080P), vbr_max_bits); + MAX(MAX((cm->MBs * MAX_MB_RATE), MAXRATE_1080P), vbr_max_bits); // Set Maximum gf/arf interval cpi->rc.max_gf_interval = 16; @@ -1158,7 +1169,7 @@ void vp9_new_framerate(VP9_COMP *cpi, double framerate) { cpi->rc.max_gf_interval = cpi->twopass.static_scene_max_gf_interval; } -static int64_t rescale(int val, int64_t num, int denom) { +static int64_t rescale(int64_t val, int64_t num, int denom) { int64_t llnum = num; int64_t llden = denom; int64_t llval = val; @@ -1211,9 +1222,12 @@ static void update_layer_context_change_config(VP9_COMP *const cpi, lc->target_bandwidth = oxcf->ts_target_bitrate[temporal_layer] * 1000; bitrate_alloc = (float)lc->target_bandwidth / (float)target_bandwidth; // Update buffer-related quantities. - lc->starting_buffer_level = oxcf->starting_buffer_level * bitrate_alloc; - lc->optimal_buffer_level = oxcf->optimal_buffer_level * bitrate_alloc; - lc->maximum_buffer_size = oxcf->maximum_buffer_size * bitrate_alloc; + lc->starting_buffer_level = + (int64_t)(oxcf->starting_buffer_level * bitrate_alloc); + lc->optimal_buffer_level = + (int64_t)(oxcf->optimal_buffer_level * bitrate_alloc); + lc->maximum_buffer_size = + (int64_t)(oxcf->maximum_buffer_size * bitrate_alloc); lrc->bits_off_target = MIN(lrc->bits_off_target, lc->maximum_buffer_size); lrc->buffer_level = MIN(lrc->buffer_level, lc->maximum_buffer_size); // Update framerate-related quantities. @@ -1245,8 +1259,8 @@ static void update_layer_framerate(VP9_COMP *const cpi) { int prev_layer_target_bandwidth = oxcf->ts_target_bitrate[temporal_layer - 1] * 1000; lc->avg_frame_size = - (int)(lc->target_bandwidth - prev_layer_target_bandwidth) / - (lc->framerate - prev_layer_framerate); + (int)((lc->target_bandwidth - prev_layer_target_bandwidth) / + (lc->framerate - prev_layer_framerate)); } } @@ -1274,7 +1288,7 @@ static void save_layer_context(VP9_COMP *const cpi) { int temporal_layer = cpi->svc.temporal_layer_id; LAYER_CONTEXT *lc = &cpi->svc.layer_context[temporal_layer]; lc->rc = cpi->rc; - lc->target_bandwidth = cpi->oxcf.target_bandwidth; + lc->target_bandwidth = (int)cpi->oxcf.target_bandwidth; lc->starting_buffer_level = cpi->oxcf.starting_buffer_level; lc->optimal_buffer_level = cpi->oxcf.optimal_buffer_level; lc->maximum_buffer_size = cpi->oxcf.maximum_buffer_size; @@ -1375,6 +1389,9 @@ void vp9_change_config(VP9_PTR ptr, VP9_CONFIG *oxcf) { cpi->oxcf = *oxcf; + if (cpi->oxcf.cpu_used == -6) + cpi->oxcf.play_alternate = 0; + switch (cpi->oxcf.mode) { // Real time and one pass deprecated in test code base case MODE_GOODQUALITY: @@ -1491,18 +1508,14 @@ void vp9_change_config(VP9_PTR ptr, VP9_CONFIG *oxcf) { if (cpi->svc.number_temporal_layers > 1 && cpi->oxcf.end_usage == USAGE_STREAM_FROM_SERVER) { - update_layer_context_change_config(cpi, cpi->oxcf.target_bandwidth); + update_layer_context_change_config(cpi, (int)cpi->oxcf.target_bandwidth); } cpi->speed = abs(cpi->oxcf.cpu_used); - if (cpi->oxcf.lag_in_frames == 0) { - // Force allow_lag to 0 if lag_in_frames is 0. - cpi->oxcf.allow_lag = 0; - } else if (cpi->oxcf.lag_in_frames > MAX_LAG_BUFFERS) { - // Limit on lag buffers as these are not currently dynamically allocated. + // Limit on lag buffers as these are not currently dynamically allocated. + if (cpi->oxcf.lag_in_frames > MAX_LAG_BUFFERS) cpi->oxcf.lag_in_frames = MAX_LAG_BUFFERS; - } #if CONFIG_MULTIPLE_ARF vp9_zero(cpi->alt_ref_source); @@ -1568,6 +1581,7 @@ static void alloc_mode_context(VP9_COMMON *cm, int num_4x4_blk, int num_pix = num_4x4_blk << 4; int i, k; ctx->num_4x4_blk = num_4x4_blk; + CHECK_MEM_ERROR(cm, ctx->zcoeff_blk, vpx_calloc(num_4x4_blk, sizeof(uint8_t))); for (i = 0; i < MAX_MB_PLANE; ++i) { @@ -1611,7 +1625,6 @@ static void init_pick_mode_context(VP9_COMP *cpi) { VP9_COMMON *const cm = &cpi->common; MACROBLOCK *const x = &cpi->mb; - for (i = 0; i < BLOCK_SIZES; ++i) { const int num_4x4_w = num_4x4_blocks_wide_lookup[i]; const int num_4x4_h = num_4x4_blocks_high_lookup[i]; @@ -2029,10 +2042,12 @@ void vp9_remove_compressor(VP9_PTR *ptr) { / time_encoded; if (cpi->b_calculate_psnr) { - const double total_psnr = vp9_mse2psnr(cpi->total_samples, 255.0, - cpi->total_sq_error); - const double totalp_psnr = vp9_mse2psnr(cpi->totalp_samples, 255.0, - cpi->totalp_sq_error); + const double total_psnr = + vp9_mse2psnr((double)cpi->total_samples, 255.0, + (double)cpi->total_sq_error); + const double totalp_psnr = + vp9_mse2psnr((double)cpi->totalp_samples, 255.0, + (double)cpi->totalp_sq_error); const double total_ssim = 100 * pow(cpi->summed_quality / cpi->summed_weights, 8.0); const double totalp_ssim = 100 * pow(cpi->summedp_quality / @@ -2208,12 +2223,12 @@ static void calc_psnr(const YV12_BUFFER_CONFIG *a, const YV12_BUFFER_CONFIG *b, const int w = widths[i]; const int h = heights[i]; const uint32_t samples = w * h; - const double sse = calc_plane_error(a_planes[i], a_strides[i], - b_planes[i], b_strides[i], - w, h); + const uint64_t sse = calc_plane_error(a_planes[i], a_strides[i], + b_planes[i], b_strides[i], + w, h); psnr->sse[1 + i] = sse; psnr->samples[1 + i] = samples; - psnr->psnr[1 + i] = vp9_mse2psnr(samples, 255.0, sse); + psnr->psnr[1 + i] = vp9_mse2psnr(samples, 255.0, (double) sse); total_sse += sse; total_samples += samples; @@ -2221,7 +2236,7 @@ static void calc_psnr(const YV12_BUFFER_CONFIG *a, const YV12_BUFFER_CONFIG *b, psnr->sse[0] = total_sse; psnr->samples[0] = total_samples; - psnr->psnr[0] = vp9_mse2psnr(total_samples, 255.0, total_sse); + psnr->psnr[0] = vp9_mse2psnr((double)total_samples, 255.0, (double)total_sse); } static void generate_psnr_packet(VP9_COMP *cpi) { @@ -2734,7 +2749,7 @@ static void output_frame_level_debug_stats(VP9_COMP *cpi) { FILE *const f = fopen("tmp.stt", cm->current_video_frame ? "a" : "w"); int recon_err; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); recon_err = vp9_calc_ss_err(cpi->Source, get_frame_new_buffer(cm)); @@ -2778,8 +2793,6 @@ static void output_frame_level_debug_stats(VP9_COMP *cpi) { for (i = 0; i < MAX_MODES; ++i) fprintf(fmodes, "%5d ", cpi->mode_chosen_counts[i]); - for (i = 0; i < MAX_REFS; ++i) - fprintf(fmodes, "%5d ", cpi->sub8x8_mode_chosen_counts[i]); fprintf(fmodes, "\n"); @@ -2793,7 +2806,7 @@ static void encode_without_recode_loop(VP9_COMP *cpi, uint8_t *dest, int q) { VP9_COMMON *const cm = &cpi->common; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); vp9_set_quantizer(cpi, q); // Set up entropy context depending on frame type. The decoder mandates @@ -2804,7 +2817,7 @@ static void encode_without_recode_loop(VP9_COMP *cpi, if (cm->frame_type == KEY_FRAME) { vp9_setup_key_frame(cpi); } else { - if (!cm->intra_only && !cm->error_resilient_mode) { + if (!cm->intra_only && !cm->error_resilient_mode && !cpi->use_svc) { cpi->common.frame_context_idx = cpi->refresh_alt_ref_frame; } vp9_setup_inter_frame(cpi); @@ -2822,7 +2835,7 @@ static void encode_without_recode_loop(VP9_COMP *cpi, // Update the skip mb flag probabilities based on the distribution // seen in the last encoder iteration. // update_base_skip_probs(cpi); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); } static void encode_with_recode_loop(VP9_COMP *cpi, @@ -2832,6 +2845,7 @@ static void encode_with_recode_loop(VP9_COMP *cpi, int bottom_index, int top_index) { VP9_COMMON *const cm = &cpi->common; + RATE_CONTROL *const rc = &cpi->rc; int loop_count = 0; int loop = 0; int overshoot_seen = 0; @@ -2841,12 +2855,12 @@ static void encode_with_recode_loop(VP9_COMP *cpi, int frame_under_shoot_limit; // Decide frame size bounds - vp9_rc_compute_frame_size_bounds(cpi, cpi->rc.this_frame_target, + vp9_rc_compute_frame_size_bounds(cpi, rc->this_frame_target, &frame_under_shoot_limit, &frame_over_shoot_limit); do { - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); vp9_set_quantizer(cpi, q); @@ -2859,7 +2873,7 @@ static void encode_with_recode_loop(VP9_COMP *cpi, if (cm->frame_type == KEY_FRAME) { vp9_setup_key_frame(cpi); } else { - if (!cm->intra_only && !cm->error_resilient_mode) { + if (!cm->intra_only && !cm->error_resilient_mode && !cpi->use_svc) { cpi->common.frame_context_idx = cpi->refresh_alt_ref_frame; } vp9_setup_inter_frame(cpi); @@ -2881,7 +2895,7 @@ static void encode_with_recode_loop(VP9_COMP *cpi, // seen in the last encoder iteration. // update_base_skip_probs(cpi); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); // Dummy pack of the bitstream using up to date stats to get an // accurate estimate of output frame size to determine if we need @@ -2892,7 +2906,7 @@ static void encode_with_recode_loop(VP9_COMP *cpi, if (!cpi->sf.use_pick_mode) vp9_pack_bitstream(cpi, dest, size); - cpi->rc.projected_frame_size = (*size) << 3; + rc->projected_frame_size = (int)(*size) << 3; vp9_restore_coding_context(cpi); if (frame_over_shoot_limit == 0) @@ -2903,8 +2917,8 @@ static void encode_with_recode_loop(VP9_COMP *cpi, loop = 0; } else { if ((cm->frame_type == KEY_FRAME) && - cpi->rc.this_key_frame_forced && - (cpi->rc.projected_frame_size < cpi->rc.max_frame_bandwidth)) { + rc->this_key_frame_forced && + (rc->projected_frame_size < rc->max_frame_bandwidth)) { int last_q = q; int kf_err = vp9_calc_ss_err(cpi->Source, get_frame_new_buffer(cm)); @@ -2917,9 +2931,9 @@ static void encode_with_recode_loop(VP9_COMP *cpi, // The key frame is not good enough or we can afford // to make it better without undue risk of popping. if ((kf_err > high_err_target && - cpi->rc.projected_frame_size <= frame_over_shoot_limit) || + rc->projected_frame_size <= frame_over_shoot_limit) || (kf_err > low_err_target && - cpi->rc.projected_frame_size <= frame_under_shoot_limit)) { + rc->projected_frame_size <= frame_under_shoot_limit)) { // Lower q_high q_high = q > q_low ? q - 1 : q_low; @@ -2927,7 +2941,7 @@ static void encode_with_recode_loop(VP9_COMP *cpi, q = (q * high_err_target) / kf_err; q = MIN(q, (q_high + q_low) >> 1); } else if (kf_err < low_err_target && - cpi->rc.projected_frame_size >= frame_under_shoot_limit) { + rc->projected_frame_size >= frame_under_shoot_limit) { // The key frame is much better than the previous frame // Raise q_low q_low = q < q_high ? q + 1 : q_high; @@ -2953,10 +2967,10 @@ static void encode_with_recode_loop(VP9_COMP *cpi, // Update correction factor & compute new Q to try... // Frame is too large - if (cpi->rc.projected_frame_size > cpi->rc.this_frame_target) { + if (rc->projected_frame_size > rc->this_frame_target) { // Special case if the projected size is > the max allowed. - if (cpi->rc.projected_frame_size >= cpi->rc.max_frame_bandwidth) - q_high = cpi->rc.worst_quality; + if (rc->projected_frame_size >= rc->max_frame_bandwidth) + q_high = rc->worst_quality; // Raise Qlow as to at least the current value q_low = q < q_high ? q + 1 : q_high; @@ -2970,12 +2984,12 @@ static void encode_with_recode_loop(VP9_COMP *cpi, // Update rate_correction_factor unless vp9_rc_update_rate_correction_factors(cpi, 0); - q = vp9_rc_regulate_q(cpi, cpi->rc.this_frame_target, + q = vp9_rc_regulate_q(cpi, rc->this_frame_target, bottom_index, MAX(q_high, top_index)); while (q < q_low && retries < 10) { vp9_rc_update_rate_correction_factors(cpi, 0); - q = vp9_rc_regulate_q(cpi, cpi->rc.this_frame_target, + q = vp9_rc_regulate_q(cpi, rc->this_frame_target, bottom_index, MAX(q_high, top_index)); retries++; } @@ -2991,7 +3005,7 @@ static void encode_with_recode_loop(VP9_COMP *cpi, q = (q_high + q_low) / 2; } else { vp9_rc_update_rate_correction_factors(cpi, 0); - q = vp9_rc_regulate_q(cpi, cpi->rc.this_frame_target, + q = vp9_rc_regulate_q(cpi, rc->this_frame_target, bottom_index, top_index); // Special case reset for qlow for constrained quality. // This should only trigger where there is very substantial @@ -3004,7 +3018,7 @@ static void encode_with_recode_loop(VP9_COMP *cpi, while (q > q_high && retries < 10) { vp9_rc_update_rate_correction_factors(cpi, 0); - q = vp9_rc_regulate_q(cpi, cpi->rc.this_frame_target, + q = vp9_rc_regulate_q(cpi, rc->this_frame_target, bottom_index, top_index); retries++; } @@ -3023,8 +3037,8 @@ static void encode_with_recode_loop(VP9_COMP *cpi, } // Special case for overlay frame. - if (cpi->rc.is_src_frame_alt_ref && - (cpi->rc.projected_frame_size < cpi->rc.max_frame_bandwidth)) + if (rc->is_src_frame_alt_ref && + rc->projected_frame_size < rc->max_frame_bandwidth) loop = 0; if (loop) { @@ -3092,8 +3106,8 @@ static void encode_frame_to_data_rate(VP9_COMP *cpi, int top_index; int bottom_index; - SPEED_FEATURES *const sf = &cpi->sf; - unsigned int max_mv_def = MIN(cm->width, cm->height); + const SPEED_FEATURES *const sf = &cpi->sf; + const unsigned int max_mv_def = MIN(cm->width, cm->height); struct segmentation *const seg = &cm->seg; set_ext_overrides(cpi); @@ -3232,7 +3246,7 @@ static void encode_frame_to_data_rate(VP9_COMP *cpi, if (!frame_is_intra_only(cm)) { cm->interp_filter = DEFAULT_INTERP_FILTER; /* TODO: Decide this more intelligently */ - set_high_precision_mv(cpi, (q < HIGH_PRECISION_MV_QTHRESH)); + set_high_precision_mv(cpi, q < HIGH_PRECISION_MV_QTHRESH); } if (cpi->sf.recode_loop == DISALLOW_RECODE) { @@ -3425,6 +3439,7 @@ static void Pass2Encode(VP9_COMP *cpi, size_t *size, static void check_initial_width(VP9_COMP *cpi, int subsampling_x, int subsampling_y) { VP9_COMMON *const cm = &cpi->common; + if (!cpi->initial_width) { cm->subsampling_x = subsampling_x; cm->subsampling_y = subsampling_y; @@ -3438,12 +3453,12 @@ static void check_initial_width(VP9_COMP *cpi, int subsampling_x, int vp9_receive_raw_frame(VP9_PTR ptr, unsigned int frame_flags, YV12_BUFFER_CONFIG *sd, int64_t time_stamp, int64_t end_time) { - VP9_COMP *cpi = (VP9_COMP *) ptr; - VP9_COMMON *cm = &cpi->common; - struct vpx_usec_timer timer; - int res = 0; - const int subsampling_x = sd->uv_width < sd->y_width; - const int subsampling_y = sd->uv_height < sd->y_height; + VP9_COMP *cpi = (VP9_COMP *)ptr; + VP9_COMMON *cm = &cpi->common; + struct vpx_usec_timer timer; + int res = 0; + const int subsampling_x = sd->uv_width < sd->y_width; + const int subsampling_y = sd->uv_height < sd->y_height; check_initial_width(cpi, subsampling_x, subsampling_y); vpx_usec_timer_start(&timer); @@ -3675,7 +3690,7 @@ int vp9_get_compressed_data(VP9_PTR ptr, unsigned int *frame_flags, *size = 0; // Clear down mmx registers - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); /* find a free buffer for the new frame, releasing the reference previously * held. @@ -3762,7 +3777,7 @@ int vp9_get_compressed_data(VP9_PTR ptr, unsigned int *frame_flags, #if CONFIG_INTERNAL_STATS if (cpi->pass != 1) { - cpi->bytes += *size; + cpi->bytes += (int)(*size); if (cm->show_frame) { cpi->count++; @@ -3837,22 +3852,23 @@ int vp9_get_compressed_data(VP9_PTR ptr, unsigned int *frame_flags, int vp9_get_preview_raw_frame(VP9_PTR comp, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t *flags) { - VP9_COMP *cpi = (VP9_COMP *) comp; + VP9_COMP *cpi = (VP9_COMP *)comp; + VP9_COMMON *cm = &cpi->common; - if (!cpi->common.show_frame) { + if (!cm->show_frame) { return -1; } else { int ret; #if CONFIG_VP9_POSTPROC - ret = vp9_post_proc_frame(&cpi->common, dest, flags); + ret = vp9_post_proc_frame(cm, dest, flags); #else - if (cpi->common.frame_to_show) { - *dest = *cpi->common.frame_to_show; - dest->y_width = cpi->common.width; - dest->y_height = cpi->common.height; - dest->uv_width = cpi->common.width >> cpi->common.subsampling_x; - dest->uv_height = cpi->common.height >> cpi->common.subsampling_y; + if (cm->frame_to_show) { + *dest = *cm->frame_to_show; + dest->y_width = cm->width; + dest->y_height = cm->height; + dest->uv_width = cm->width >> cm->subsampling_x; + dest->uv_height = cm->height >> cm->subsampling_y; ret = 0; } else { ret = -1; diff --git a/vp9/encoder/vp9_onyx_int.h b/vp9/encoder/vp9_onyx_int.h index 7c0135eb5..7bcceedb8 100644 --- a/vp9/encoder/vp9_onyx_int.h +++ b/vp9/encoder/vp9_onyx_int.h @@ -136,7 +136,8 @@ typedef enum { NSTEP = 1, HEX = 2, BIGDIA = 3, - SQUARE = 4 + SQUARE = 4, + FAST_HEX = 5 } SEARCH_METHODS; typedef enum { @@ -486,9 +487,6 @@ typedef struct VP9_COMP { // Ambient reconstruction err target for force key frames int ambient_err; - unsigned int mode_chosen_counts[MAX_MODES]; - unsigned int sub8x8_mode_chosen_counts[MAX_REFS]; - int rd_threshes[MAX_SEGMENTS][BLOCK_SIZES][MAX_MODES]; int rd_thresh_freq_fact[BLOCK_SIZES][MAX_MODES]; int rd_thresh_sub8x8[MAX_SEGMENTS][BLOCK_SIZES][MAX_REFS]; @@ -586,6 +584,8 @@ typedef struct VP9_COMP { int fixed_divide[512]; #if CONFIG_INTERNAL_STATS + unsigned int mode_chosen_counts[MAX_MODES]; + int count; double total_y; double total_u; diff --git a/vp9/encoder/vp9_pickmode.c b/vp9/encoder/vp9_pickmode.c index 822185a8a..383d92751 100644 --- a/vp9/encoder/vp9_pickmode.c +++ b/vp9/encoder/vp9_pickmode.c @@ -127,9 +127,50 @@ static int full_pixel_motion_search(VP9_COMP *cpi, MACROBLOCK *x, // calculate the bit cost on motion vector *rate_mv = vp9_mv_bit_cost(&tmp_mv->as_mv, &ref_mv.as_mv, x->nmvjointcost, x->mvcost, MV_COST_WEIGHT); + return bestsme; +} +static void sub_pixel_motion_search(VP9_COMP *cpi, MACROBLOCK *x, + const TileInfo *const tile, + BLOCK_SIZE bsize, int mi_row, int mi_col, + int_mv *tmp_mv) { + MACROBLOCKD *xd = &x->e_mbd; + MB_MODE_INFO *mbmi = &xd->mi_8x8[0]->mbmi; + struct buf_2d backup_yv12[MAX_MB_PLANE] = {{0}}; + int ref = mbmi->ref_frame[0]; + int_mv ref_mv = mbmi->ref_mvs[ref][0]; + int dis; - return bestsme; + const YV12_BUFFER_CONFIG *scaled_ref_frame = vp9_get_scaled_ref_frame(cpi, + ref); + if (scaled_ref_frame) { + int i; + // Swap out the reference frame for a version that's been scaled to + // match the resolution of the current frame, allowing the existing + // motion search code to be used without additional modifications. + for (i = 0; i < MAX_MB_PLANE; i++) + backup_yv12[i] = xd->plane[i].pre[0]; + + setup_pre_planes(xd, 0, scaled_ref_frame, mi_row, mi_col, NULL); + } + + tmp_mv->as_mv.col >>= 3; + tmp_mv->as_mv.row >>= 3; + + cpi->find_fractional_mv_step(x, &tmp_mv->as_mv, &ref_mv.as_mv, + cpi->common.allow_high_precision_mv, + x->errorperbit, + &cpi->fn_ptr[bsize], + cpi->sf.subpel_force_stop, + cpi->sf.subpel_iters_per_step, + x->nmvjointcost, x->mvcost, + &dis, &x->pred_sse[ref]); + + if (scaled_ref_frame) { + int i; + for (i = 0; i < MAX_MB_PLANE; i++) + xd->plane[i].pre[0] = backup_yv12[i]; + } } // TODO(jingning) placeholder for inter-frame non-RD mode decision. @@ -151,7 +192,10 @@ int64_t vp9_pick_inter_mode(VP9_COMP *cpi, MACROBLOCK *x, VP9_ALT_FLAG }; int64_t best_rd = INT64_MAX; int64_t this_rd; - int cost[4]= { 0, 100, 150, 205 }; + static const int cost[4]= { 0, 50, 75, 100 }; + + const int64_t inter_mode_thresh = 300; + const int64_t intra_mode_cost = 50; x->skip_encode = cpi->sf.skip_encode_frame && x->q_index < QIDX_SKIP_THRESH; @@ -161,6 +205,7 @@ int64_t vp9_pick_inter_mode(VP9_COMP *cpi, MACROBLOCK *x, // initialize mode decisions *returnrate = INT_MAX; + *returndistortion = INT64_MAX; vpx_memset(mbmi, 0, sizeof(MB_MODE_INFO)); mbmi->sb_type = bsize; mbmi->ref_frame[0] = NONE; @@ -200,9 +245,6 @@ int64_t vp9_pick_inter_mode(VP9_COMP *cpi, MACROBLOCK *x, int64_t dist; if (this_mode == NEWMV) { - if (this_rd < 300) - continue; - x->mode_sad[ref_frame][INTER_OFFSET(NEWMV)] = full_pixel_motion_search(cpi, x, tile, bsize, mi_row, mi_col, &frame_mv[NEWMV][ref_frame], &rate_mv); @@ -225,10 +267,39 @@ int64_t vp9_pick_inter_mode(VP9_COMP *cpi, MACROBLOCK *x, } } - // TODO(jingning) sub-pixel motion search, if NEWMV is chosen - - // TODO(jingning) intra prediction search, if the best SAD is above a certain + // Perform intra prediction search, if the best SAD is above a certain // threshold. + if (best_rd > inter_mode_thresh) { + struct macroblock_plane *const p = &x->plane[0]; + struct macroblockd_plane *const pd = &xd->plane[0]; + for (this_mode = DC_PRED; this_mode <= DC_PRED; ++this_mode) { + vp9_predict_intra_block(xd, 0, b_width_log2(bsize), + mbmi->tx_size, this_mode, + &p->src.buf[0], p->src.stride, + &pd->dst.buf[0], pd->dst.stride, 0, 0, 0); + + this_rd = cpi->fn_ptr[bsize].sdf(p->src.buf, + p->src.stride, + pd->dst.buf, + pd->dst.stride, INT_MAX); + + if (this_rd + intra_mode_cost < best_rd) { + best_rd = this_rd; + mbmi->mode = this_mode; + mbmi->ref_frame[0] = INTRA_FRAME; + mbmi->uv_mode = this_mode; + } + } + } + + // Perform sub-pixel motion search, if NEWMV is chosen + if (mbmi->mode == NEWMV) { + ref_frame = mbmi->ref_frame[0]; + sub_pixel_motion_search(cpi, x, tile, bsize, mi_row, mi_col, + &frame_mv[NEWMV][ref_frame]); + mbmi->mv[0].as_int = frame_mv[NEWMV][ref_frame].as_int; + xd->mi_8x8[0]->bmi[0].as_mv[0].as_int = mbmi->mv[0].as_int; + } return INT64_MAX; } diff --git a/vp9/encoder/vp9_ratectrl.c b/vp9/encoder/vp9_ratectrl.c index a5560631f..f78ebfe18 100644 --- a/vp9/encoder/vp9_ratectrl.c +++ b/vp9/encoder/vp9_ratectrl.c @@ -354,7 +354,7 @@ void vp9_rc_update_rate_correction_factors(VP9_COMP *cpi, int damp_var) { int projected_size_based_on_q = 0; // Clear down mmx registers to allow floating point in what follows - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); // Work out how big we would have expected the frame to be at this Q given // the current correction factor. diff --git a/vp9/encoder/vp9_rdopt.c b/vp9/encoder/vp9_rdopt.c index a3e879b82..376224e5a 100644 --- a/vp9/encoder/vp9_rdopt.c +++ b/vp9/encoder/vp9_rdopt.c @@ -274,7 +274,7 @@ void vp9_initialize_rd_consts(VP9_COMP *cpi) { MACROBLOCK *x = &cpi->mb; int qindex, i; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); // Further tests required to see if optimum is different // for key frames, golden frames and arf frames. @@ -295,13 +295,15 @@ void vp9_initialize_rd_consts(VP9_COMP *cpi) { set_block_thresholds(cpi); - fill_token_costs(x->token_costs, cm->fc.coef_probs); - if (!cpi->sf.use_pick_mode) { + fill_token_costs(x->token_costs, cm->fc.coef_probs); + for (i = 0; i < PARTITION_CONTEXTS; i++) vp9_cost_tokens(x->partition_cost[i], get_partition_probs(cm, i), vp9_partition_tree); + } + if (!cpi->sf.use_pick_mode || (cm->current_video_frame & 0x07) == 1) { fill_mode_costs(cpi); if (!frame_is_intra_only(cm)) { @@ -431,7 +433,9 @@ static void model_rd_for_sb(VP9_COMP *cpi, BLOCK_SIZE bsize, // Note our transform coeffs are 8 times an orthogonal transform. // Hence quantizer step is also 8 times. To get effective quantizer // we need to divide by 8 before sending to modeling function. - int i, rate_sum = 0, dist_sum = 0; + int i; + int64_t rate_sum = 0; + int64_t dist_sum = 0; int ref = xd->mi_8x8[0]->mbmi.ref_frame[0]; unsigned int sse; @@ -448,13 +452,13 @@ static void model_rd_for_sb(VP9_COMP *cpi, BLOCK_SIZE bsize, // Fast approximate the modelling function. if (cpi->speed > 4) { - int rate; + int64_t rate; int64_t dist; int64_t square_error = sse; int quantizer = (pd->dequant[1] >> 3); - if ( quantizer < 120) - rate = (square_error * (280-quantizer) )>> 8; + if (quantizer < 120) + rate = (square_error * (280 - quantizer)) >> 8; else rate = 0; dist = (square_error * quantizer) >> 8; @@ -466,12 +470,12 @@ static void model_rd_for_sb(VP9_COMP *cpi, BLOCK_SIZE bsize, model_rd_from_var_lapndz(sse, 1 << num_pels_log2_lookup[bs], pd->dequant[1] >> 3, &rate, &dist); rate_sum += rate; - dist_sum += (int)dist; + dist_sum += dist; } } - *out_rate_sum = rate_sum; - *out_dist_sum = (int64_t)dist_sum << 4; + *out_rate_sum = (int)rate_sum; + *out_dist_sum = dist_sum << 4; } static void model_rd_for_sb_y_tx(VP9_COMP *cpi, BLOCK_SIZE bsize, @@ -2446,7 +2450,11 @@ static void single_motion_search(VP9_COMP *cpi, MACROBLOCK *x, // Further step/diamond searches as necessary further_steps = (cpi->sf.max_step_search_steps - 1) - step_param; - if (cpi->sf.search_method == HEX) { + if (cpi->sf.search_method == FAST_HEX) { + bestsme = vp9_fast_hex_search(x, &mvp_full, step_param, sadpb, + &cpi->fn_ptr[bsize], 1, + &ref_mv.as_mv, &tmp_mv->as_mv); + } else if (cpi->sf.search_method == HEX) { bestsme = vp9_hex_search(x, &mvp_full, step_param, sadpb, 1, &cpi->fn_ptr[bsize], 1, &ref_mv.as_mv, &tmp_mv->as_mv); @@ -2708,6 +2716,8 @@ static int64_t handle_inter_mode(VP9_COMP *cpi, MACROBLOCK *x, int_mv tmp_mv; single_motion_search(cpi, x, tile, bsize, mi_row, mi_col, &tmp_mv, &rate_mv); + if (tmp_mv.as_int == INVALID_MV) + return INT64_MAX; *rate2 += rate_mv; frame_mv[refs[0]].as_int = xd->mi_8x8[0]->bmi[0].as_mv[0].as_int = tmp_mv.as_int; @@ -3119,10 +3129,10 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, BLOCK_SIZE bsize, PICK_MODE_CONTEXT *ctx, int64_t best_rd_so_far) { - VP9_COMMON *cm = &cpi->common; - MACROBLOCKD *xd = &x->e_mbd; - MB_MODE_INFO *mbmi = &xd->mi_8x8[0]->mbmi; - const struct segmentation *seg = &cm->seg; + VP9_COMMON *const cm = &cpi->common; + MACROBLOCKD *const xd = &x->e_mbd; + MB_MODE_INFO *const mbmi = &xd->mi_8x8[0]->mbmi; + const struct segmentation *const seg = &cm->seg; const BLOCK_SIZE block_size = get_plane_block_size(bsize, &xd->plane[0]); MB_PREDICTION_MODE this_mode; MV_REFERENCE_FRAME ref_frame, second_ref_frame; @@ -3158,14 +3168,14 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, const int bws = num_8x8_blocks_wide_lookup[bsize] / 2; const int bhs = num_8x8_blocks_high_lookup[bsize] / 2; int best_skip2 = 0; - int ref_frame_mask = 0; int mode_skip_mask = 0; + const int mode_skip_start = cpi->sf.mode_skip_start + 1; + const int *const rd_threshes = cpi->rd_threshes[segment_id][bsize]; + const int *const rd_thresh_freq_fact = cpi->rd_thresh_freq_fact[bsize]; + const int mode_search_skip_flags = cpi->sf.mode_search_skip_flags; x->skip_encode = cpi->sf.skip_encode_frame && x->q_index < QIDX_SKIP_THRESH; - // Everywhere the flag is set the error is much higher than its neighbors. - ctx->modes_with_high_error = 0; - estimate_ref_frame_costs(cpi, segment_id, ref_costs_single, ref_costs_comp, &comp_mode_p); @@ -3193,13 +3203,26 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, frame_mv[ZEROMV][ref_frame].as_int = 0; } - for (ref_frame = LAST_FRAME; - ref_frame <= ALTREF_FRAME && cpi->sf.reference_masking; ++ref_frame) { - int i; - for (i = LAST_FRAME; i <= ALTREF_FRAME; ++i) { - if ((x->pred_mv_sad[ref_frame] >> 2) > x->pred_mv_sad[i]) { - ref_frame_mask |= (1 << ref_frame); - break; + for (ref_frame = LAST_FRAME; ref_frame <= ALTREF_FRAME; ++ref_frame) { + // All modes from vp9_mode_order that use this frame as any ref + static const int ref_frame_mask_all[] = { + 0x123291, 0x25c444, 0x39b722 + }; + // Fixed mv modes (NEARESTMV, NEARMV, ZEROMV) from vp9_mode_order that use + // this frame as their primary ref + static const int ref_frame_mask_fixedmv[] = { + 0x121281, 0x24c404, 0x080102 + }; + if (!(cpi->ref_frame_flags & flag_list[ref_frame])) { + // Skip modes for missing references + mode_skip_mask |= ref_frame_mask_all[ref_frame - LAST_FRAME]; + } else if (cpi->sf.reference_masking) { + for (i = LAST_FRAME; i <= ALTREF_FRAME; ++i) { + // Skip fixed mv modes for poor references + if ((x->pred_mv_sad[ref_frame] >> 2) > x->pred_mv_sad[i]) { + mode_skip_mask |= ref_frame_mask_fixedmv[ref_frame - LAST_FRAME]; + break; + } } } } @@ -3215,110 +3238,64 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, int64_t tx_cache[TX_MODES]; int i; int this_skip2 = 0; - int64_t total_sse = INT_MAX; + int64_t total_sse = INT64_MAX; int early_term = 0; - for (i = 0; i < TX_MODES; ++i) - tx_cache[i] = INT64_MAX; - - x->skip = 0; - this_mode = vp9_mode_order[mode_index].mode; - ref_frame = vp9_mode_order[mode_index].ref_frame[0]; - second_ref_frame = vp9_mode_order[mode_index].ref_frame[1]; - // Look at the reference frame of the best mode so far and set the // skip mask to look at a subset of the remaining modes. - if (mode_index > cpi->sf.mode_skip_start) { - if (mode_index == (cpi->sf.mode_skip_start + 1)) { - switch (vp9_mode_order[best_mode_index].ref_frame[0]) { - case INTRA_FRAME: - mode_skip_mask = 0; - break; - case LAST_FRAME: - mode_skip_mask = LAST_FRAME_MODE_MASK; - break; - case GOLDEN_FRAME: - mode_skip_mask = GOLDEN_FRAME_MODE_MASK; - break; - case ALTREF_FRAME: - mode_skip_mask = ALT_REF_MODE_MASK; - break; - case NONE: - case MAX_REF_FRAMES: - assert(0 && "Invalid Reference frame"); - } + if (mode_index == mode_skip_start) { + switch (vp9_mode_order[best_mode_index].ref_frame[0]) { + case INTRA_FRAME: + break; + case LAST_FRAME: + mode_skip_mask |= LAST_FRAME_MODE_MASK; + break; + case GOLDEN_FRAME: + mode_skip_mask |= GOLDEN_FRAME_MODE_MASK; + break; + case ALTREF_FRAME: + mode_skip_mask |= ALT_REF_MODE_MASK; + break; + case NONE: + case MAX_REF_FRAMES: + assert(0 && "Invalid Reference frame"); } - if (mode_skip_mask & (1 << mode_index)) - continue; } - - // Skip if the current reference frame has been masked off - if (ref_frame_mask & (1 << ref_frame) && this_mode != NEWMV) + if (mode_skip_mask & (1 << mode_index)) continue; // Test best rd so far against threshold for trying this mode. - if ((best_rd < ((int64_t)cpi->rd_threshes[segment_id][bsize][mode_index] * - cpi->rd_thresh_freq_fact[bsize][mode_index] >> 5)) || - cpi->rd_threshes[segment_id][bsize][mode_index] == INT_MAX) - continue; - - // Do not allow compound prediction if the segment level reference - // frame feature is in use as in this case there can only be one reference. - if ((second_ref_frame > INTRA_FRAME) && - vp9_segfeature_active(seg, segment_id, SEG_LVL_REF_FRAME)) - continue; + if (best_rd < ((int64_t)rd_threshes[mode_index] * + rd_thresh_freq_fact[mode_index] >> 5) || + rd_threshes[mode_index] == INT_MAX) + continue; - mbmi->ref_frame[0] = ref_frame; - mbmi->ref_frame[1] = second_ref_frame; - - if (!(ref_frame == INTRA_FRAME - || (cpi->ref_frame_flags & flag_list[ref_frame]))) { - continue; - } - if (!(second_ref_frame == NONE - || (cpi->ref_frame_flags & flag_list[second_ref_frame]))) { - continue; - } + this_mode = vp9_mode_order[mode_index].mode; + ref_frame = vp9_mode_order[mode_index].ref_frame[0]; + second_ref_frame = vp9_mode_order[mode_index].ref_frame[1]; comp_pred = second_ref_frame > INTRA_FRAME; if (comp_pred) { - if (cpi->sf.mode_search_skip_flags & FLAG_SKIP_COMP_BESTINTRA) - if (vp9_mode_order[best_mode_index].ref_frame[0] == INTRA_FRAME) - continue; - if (cpi->sf.mode_search_skip_flags & FLAG_SKIP_COMP_REFMISMATCH) - if (ref_frame != best_inter_ref_frame && - second_ref_frame != best_inter_ref_frame) - continue; - } - - set_ref_ptrs(cm, xd, ref_frame, second_ref_frame); - mbmi->uv_mode = DC_PRED; - - // Evaluate all sub-pel filters irrespective of whether we can use - // them for this frame. - mbmi->interp_filter = cm->interp_filter == SWITCHABLE ? EIGHTTAP - : cm->interp_filter; - xd->interp_kernel = vp9_get_interp_kernel(mbmi->interp_filter); - - if (comp_pred) { - if (!(cpi->ref_frame_flags & flag_list[second_ref_frame])) + // Do not allow compound prediction if the segment level reference + // frame feature is in use as in this case there can only be one + // reference. + if (vp9_segfeature_active(seg, segment_id, SEG_LVL_REF_FRAME)) continue; - - mode_excluded = mode_excluded ? mode_excluded - : cm->reference_mode == SINGLE_REFERENCE; + if ((mode_search_skip_flags & FLAG_SKIP_COMP_BESTINTRA) && + vp9_mode_order[best_mode_index].ref_frame[0] == INTRA_FRAME) + continue; + if ((mode_search_skip_flags & FLAG_SKIP_COMP_REFMISMATCH) && + ref_frame != best_inter_ref_frame && + second_ref_frame != best_inter_ref_frame) + continue; + mode_excluded = mode_excluded ? + mode_excluded : cm->reference_mode == SINGLE_REFERENCE; } else { if (ref_frame != INTRA_FRAME && second_ref_frame != INTRA_FRAME) mode_excluded = mode_excluded ? mode_excluded : cm->reference_mode == COMPOUND_REFERENCE; } - // Select prediction reference frames. - for (i = 0; i < MAX_MB_PLANE; i++) { - xd->plane[i].pre[0] = yv12_mb[ref_frame][i]; - if (comp_pred) - xd->plane[i].pre[1] = yv12_mb[second_ref_frame][i]; - } - // If the segment reference frame feature is enabled.... // then do nothing if the current ref frame is not allowed.. if (vp9_segfeature_active(seg, segment_id, SEG_LVL_REF_FRAME) && @@ -3360,38 +3337,60 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, continue; } -#ifdef MODE_TEST_HIT_STATS - // TEST/DEBUG CODE - // Keep a rcord of the number of test hits at each size - cpi->mode_test_hits[bsize]++; -#endif - - if (ref_frame == INTRA_FRAME) { - TX_SIZE uv_tx; // Disable intra modes other than DC_PRED for blocks with low variance // Threshold for intra skipping based on source variance // TODO(debargha): Specialize the threshold for super block sizes static const unsigned int skip_intra_var_thresh[BLOCK_SIZES] = { 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, }; - if ((cpi->sf.mode_search_skip_flags & FLAG_SKIP_INTRA_LOWVAR) && + if ((mode_search_skip_flags & FLAG_SKIP_INTRA_LOWVAR) && this_mode != DC_PRED && - x->source_variance < skip_intra_var_thresh[mbmi->sb_type]) + x->source_variance < skip_intra_var_thresh[bsize]) continue; // Only search the oblique modes if the best so far is // one of the neighboring directional modes - if ((cpi->sf.mode_search_skip_flags & FLAG_SKIP_INTRA_BESTINTER) && + if ((mode_search_skip_flags & FLAG_SKIP_INTRA_BESTINTER) && (this_mode >= D45_PRED && this_mode <= TM_PRED)) { if (vp9_mode_order[best_mode_index].ref_frame[0] > INTRA_FRAME) continue; } - mbmi->mode = this_mode; - if (cpi->sf.mode_search_skip_flags & FLAG_SKIP_INTRA_DIRMISMATCH) { - if (conditional_skipintra(mbmi->mode, best_intra_mode)) + if (mode_search_skip_flags & FLAG_SKIP_INTRA_DIRMISMATCH) { + if (conditional_skipintra(this_mode, best_intra_mode)) continue; } + } + + mbmi->mode = this_mode; + mbmi->uv_mode = DC_PRED; + mbmi->ref_frame[0] = ref_frame; + mbmi->ref_frame[1] = second_ref_frame; + // Evaluate all sub-pel filters irrespective of whether we can use + // them for this frame. + mbmi->interp_filter = cm->interp_filter == SWITCHABLE ? EIGHTTAP + : cm->interp_filter; + x->skip = 0; + set_ref_ptrs(cm, xd, ref_frame, second_ref_frame); + xd->interp_kernel = vp9_get_interp_kernel(mbmi->interp_filter); + + // Select prediction reference frames. + for (i = 0; i < MAX_MB_PLANE; i++) { + xd->plane[i].pre[0] = yv12_mb[ref_frame][i]; + if (comp_pred) + xd->plane[i].pre[1] = yv12_mb[second_ref_frame][i]; + } + + for (i = 0; i < TX_MODES; ++i) + tx_cache[i] = INT64_MAX; +#ifdef MODE_TEST_HIT_STATS + // TEST/DEBUG CODE + // Keep a rcord of the number of test hits at each size + cpi->mode_test_hits[bsize]++; +#endif + + if (ref_frame == INTRA_FRAME) { + TX_SIZE uv_tx; super_block_yrd(cpi, x, &rate_y, &distortion_y, &skippable, NULL, bsize, tx_cache, best_rd); @@ -3415,8 +3414,6 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, rate2 += intra_cost_penalty; distortion2 = distortion_y + distortion_uv; } else { - mbmi->mode = this_mode; - compmode_cost = vp9_cost_bit(comp_mode_p, second_ref_frame > INTRA_FRAME); this_rd = handle_inter_mode(cpi, x, tile, bsize, tx_cache, &rate2, &distortion2, &skippable, @@ -3428,14 +3425,16 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, single_newmv, &total_sse, best_rd); if (this_rd == INT64_MAX) continue; - } - if (cm->reference_mode == REFERENCE_MODE_SELECT) - rate2 += compmode_cost; + compmode_cost = vp9_cost_bit(comp_mode_p, comp_pred); + + if (cm->reference_mode == REFERENCE_MODE_SELECT) + rate2 += compmode_cost; + } // Estimate the reference frame signaling cost and add it // to the rolling cost variable. - if (second_ref_frame > INTRA_FRAME) { + if (comp_pred) { rate2 += ref_costs_comp[ref_frame]; } else { rate2 += ref_costs_single[ref_frame]; @@ -3543,7 +3542,7 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, // TODO(debargha): enhance this test with a better distortion prediction // based on qp, activity mask and history - if ((cpi->sf.mode_search_skip_flags & FLAG_EARLY_TERMINATE) && + if ((mode_search_skip_flags & FLAG_EARLY_TERMINATE) && (mode_index > MIN_EARLY_TERM_INDEX)) { const int qstep = xd->plane[0].dequant[1]; // TODO(debargha): Enhance this by specializing for each mode_index @@ -3653,17 +3652,6 @@ int64_t vp9_rd_pick_inter_mode_sb(VP9_COMP *cpi, MACROBLOCK *x, } } - // Flag all modes that have a distortion thats > 2x the best we found at - // this level. - for (mode_index = 0; mode_index < MB_MODE_COUNT; ++mode_index) { - if (mode_index == NEARESTMV || mode_index == NEARMV || mode_index == NEWMV) - continue; - - if (mode_distortions[mode_index] > 2 * *returndistortion) { - ctx->modes_with_high_error |= (1 << mode_index); - } - } - assert((cm->interp_filter == SWITCHABLE) || (cm->interp_filter == best_mbmode.interp_filter) || !is_inter_block(&best_mbmode)); @@ -4379,7 +4367,7 @@ int64_t vp9_rd_pick_inter_mode_sub8x8(VP9_COMP *cpi, MACROBLOCK *x, if (best_rd == INT64_MAX && bsize < BLOCK_8X8) { *returnrate = INT_MAX; - *returndistortion = INT_MAX; + *returndistortion = INT64_MAX; return best_rd; } diff --git a/vp9/encoder/vp9_vaq.c b/vp9/encoder/vp9_vaq.c index 896cd2ca2..600029b19 100644 --- a/vp9/encoder/vp9_vaq.c +++ b/vp9/encoder/vp9_vaq.c @@ -44,7 +44,7 @@ unsigned int vp9_vaq_segment_id(int energy) { double vp9_vaq_rdmult_ratio(int energy) { ENERGY_IN_BOUNDS(energy); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); return RDMULT_RATIO(energy); } @@ -52,7 +52,7 @@ double vp9_vaq_rdmult_ratio(int energy) { double vp9_vaq_inv_q_ratio(int energy) { ENERGY_IN_BOUNDS(energy); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); return Q_RATIO(-energy); } @@ -63,7 +63,7 @@ void vp9_vaq_init() { assert(ENERGY_SPAN <= MAX_SEGMENTS); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); base_ratio = 1.5; @@ -88,7 +88,7 @@ void vp9_vaq_frame_setup(VP9_COMP *cpi) { seg->abs_delta = SEGMENT_DELTADATA; - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); for (i = ENERGY_MIN; i <= ENERGY_MAX; i++) { int qindex_delta, segment_rdmult; @@ -141,8 +141,8 @@ int vp9_block_energy(VP9_COMP *cpi, MACROBLOCK *x, BLOCK_SIZE bs) { double energy; unsigned int var = block_variance(cpi, x, bs); - vp9_clear_system_state(); // __asm emms; + vp9_clear_system_state(); energy = 0.9 * (log(var + 1.0) - 10.0); - return clamp(round(energy), ENERGY_MIN, ENERGY_MAX); + return clamp((int)round(energy), ENERGY_MIN, ENERGY_MAX); } diff --git a/vp9/encoder/x86/vp9_subpel_variance_impl_intrin_avx2.c b/vp9/encoder/x86/vp9_subpel_variance_impl_intrin_avx2.c new file mode 100644 index 000000000..b8bfa8900 --- /dev/null +++ b/vp9/encoder/x86/vp9_subpel_variance_impl_intrin_avx2.c @@ -0,0 +1,641 @@ +/* + * Copyright (c) 2012 The WebM project authors. All Rights Reserved. + * + * Use of this source code is governed by a BSD-style license + * that can be found in the LICENSE file in the root of the source + * tree. An additional intellectual property rights grant can be found + * in the file PATENTS. All contributing project authors may + * be found in the AUTHORS file in the root of the source tree. + */ + +#include <immintrin.h> // AVX2 +#include "vpx_ports/mem.h" +#include "vp9/encoder/vp9_variance.h" + +DECLARE_ALIGNED(32, static const uint8_t, bilinear_filters_avx2[512]) = { + 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, + 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, + 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, + 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, + 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, + 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, + 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, + 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, + 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, + 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, + 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, + 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, + 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, + 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, + 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, + 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, + 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, 7, 9, + 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, + 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, 6, 10, + 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, + 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, 5, 11, + 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, + 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, 4, 12, + 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, + 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, 3, 13, + 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, + 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, 2, 14, + 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, + 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15, 1, 15 +}; + +unsigned int vp9_sub_pixel_variance32xh_avx2(const uint8_t *src, + int src_stride, + int x_offset, + int y_offset, + const uint8_t *dst, + int dst_stride, + int height, + unsigned int *sse) { + __m256i src_reg, dst_reg, exp_src_lo, exp_src_hi, exp_dst_lo, exp_dst_hi; + __m256i sse_reg, sum_reg, sse_reg_hi, res_cmp, sum_reg_lo, sum_reg_hi; + __m256i zero_reg; + int i, sum; + sum_reg = _mm256_set1_epi16(0); + sse_reg = _mm256_set1_epi16(0); + zero_reg = _mm256_set1_epi16(0); + + if (x_offset == 0) { + // x_offset = 0 and y_offset = 0 + if (y_offset == 0) { + for (i = 0; i < height ; i++) { + // load source and destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + + // expend each byte to 2 bytes + exp_src_lo = _mm256_unpacklo_epi8(src_reg, zero_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, zero_reg); + + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + src+= src_stride; + dst+= dst_stride; + } + // x_offset = 0 and y_offset = 8 + } else if (y_offset == 8) { + __m256i src_next_reg; + for (i = 0; i < height ; i++) { + // load source + next source + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) + (src + src_stride)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + // average between current and next stride source + src_reg = _mm256_avg_epu8(src_reg, src_next_reg); + + // expend each byte to 2 bytes + exp_src_lo = _mm256_unpacklo_epi8(src_reg, zero_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, zero_reg); + + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + src+= src_stride; + dst+= dst_stride; + } + // x_offset = 0 and y_offset = bilin interpolation + } else { + __m256i filter, pw8, src_next_reg; +#if (ARCH_X86_64) + int64_t y_offset64; + y_offset64 = y_offset; + y_offset64 <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + y_offset64)); +#else + y_offset <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + y_offset)); +#endif + pw8 = _mm256_set1_epi16(8); + for (i = 0; i < height ; i++) { + // load current and next source + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) + (src + src_stride)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + + // merge current and next source + exp_src_lo = _mm256_unpacklo_epi8(src_reg, src_next_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, src_next_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, filter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, filter); + + // add 8 to the source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // expand each byte to 2 byte in the destination + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + src+= src_stride; + dst+= dst_stride; + } + } + // x_offset = 8 and y_offset = 0 + } else if (x_offset == 8) { + if (y_offset == 0) { + __m256i src_next_reg; + for (i = 0; i < height ; i++) { + // load source and another source starting from the next + // following byte + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + + // average between source and the next byte following source + src_reg = _mm256_avg_epu8(src_reg, src_next_reg); + + // expand each byte to 2 bytes + exp_src_lo = _mm256_unpacklo_epi8(src_reg, zero_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, zero_reg); + + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + src+= src_stride; + dst+= dst_stride; + } + // x_offset = 8 and y_offset = 8 + } else if (y_offset == 8) { + __m256i src_next_reg, src_avg; + // load source and another source starting from the next + // following byte + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + + // average between source and the next byte following source + src_avg = _mm256_avg_epu8(src_reg, src_next_reg); + for (i = 0; i < height ; i++) { + src+= src_stride; + // load source and another source starting from the next + // following byte + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + // average between source and the next byte following source + src_reg = _mm256_avg_epu8(src_reg, src_next_reg); + + // expand each byte to 2 bytes + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // average between previous average to current average + src_avg = _mm256_avg_epu8(src_avg, src_reg); + // expand each byte to 2 bytes + exp_src_lo = _mm256_unpacklo_epi8(src_avg, zero_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_avg, zero_reg); + + // save current source average + src_avg = src_reg; + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + dst+= dst_stride; + } + // x_offset = 8 and y_offset = bilin interpolation + } else { + __m256i filter, pw8, src_next_reg, src_avg; +#if (ARCH_X86_64) + int64_t y_offset64; + y_offset64 = y_offset; + y_offset64 <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + y_offset64)); +#else + y_offset <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + y_offset)); +#endif + pw8 = _mm256_set1_epi16(8); + // load source and another source starting from the next + // following byte + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + // average between source and the next byte following source + src_avg = _mm256_avg_epu8(src_reg, src_next_reg); + for (i = 0; i < height ; i++) { + src+= src_stride; + // load source and another source starting from the next + // following byte + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + // average between source and the next byte following source + src_reg = _mm256_avg_epu8(src_reg, src_next_reg); + + // merge previous average and current average + exp_src_lo = _mm256_unpacklo_epi8(src_avg, src_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_avg, src_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, filter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, filter); + + // add 8 to the source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide the source by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // expand each byte to 2 bytes + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // save current source average + src_avg = src_reg; + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + dst+= dst_stride; + } + } + // x_offset = bilin interpolation and y_offset = 0 + } else { + if (y_offset == 0) { + __m256i filter, pw8, src_next_reg; +#if (ARCH_X86_64) + int64_t x_offset64; + x_offset64 = x_offset; + x_offset64 <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + x_offset64)); +#else + x_offset <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + x_offset)); +#endif + pw8 = _mm256_set1_epi16(8); + for (i = 0; i < height ; i++) { + // load source and another source starting from the next + // following byte + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + + // merge current and next source + exp_src_lo = _mm256_unpacklo_epi8(src_reg, src_next_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, src_next_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, filter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, filter); + + // add 8 to source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide the source by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // expand each byte to 2 bytes + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + src+= src_stride; + dst+= dst_stride; + } + // x_offset = bilin interpolation and y_offset = 8 + } else if (y_offset == 8) { + __m256i filter, pw8, src_next_reg, src_pack; +#if (ARCH_X86_64) + int64_t x_offset64; + x_offset64 = x_offset; + x_offset64 <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + x_offset64)); +#else + x_offset <<= 5; + filter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + x_offset)); +#endif + pw8 = _mm256_set1_epi16(8); + // load source and another source starting from the next + // following byte + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + + // merge current and next stride source + exp_src_lo = _mm256_unpacklo_epi8(src_reg, src_next_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, src_next_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, filter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, filter); + + // add 8 to source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide source by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // convert each 16 bit to 8 bit to each low and high lane source + src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); + for (i = 0; i < height ; i++) { + src+= src_stride; + + // load source and another source starting from the next + // following byte + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + + // merge current and next stride source + exp_src_lo = _mm256_unpacklo_epi8(src_reg, src_next_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, src_next_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, filter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, filter); + + // add 8 to source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide source by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // convert each 16 bit to 8 bit to each low and high lane source + src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); + // average between previous pack to the current + src_pack = _mm256_avg_epu8(src_pack, src_reg); + + // expand each byte to 2 bytes + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + exp_src_lo = _mm256_unpacklo_epi8(src_pack, zero_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_pack, zero_reg); + + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // calculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + // save previous pack + src_pack = src_reg; + dst+= dst_stride; + } + // x_offset = bilin interpolation and y_offset = bilin interpolation + } else { + __m256i xfilter, yfilter, pw8, src_next_reg, src_pack; +#if (ARCH_X86_64) + int64_t x_offset64, y_offset64; + x_offset64 = x_offset; + x_offset64 <<= 5; + y_offset64 = y_offset; + y_offset64 <<= 5; + xfilter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + x_offset64)); + yfilter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + y_offset64)); +#else + x_offset <<= 5; + xfilter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + x_offset)); + y_offset <<= 5; + yfilter = _mm256_load_si256( + (__m256i const *)(bilinear_filters_avx2 + y_offset)); +#endif + pw8 = _mm256_set1_epi16(8); + // load source and another source starting from the next + // following byte + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + // merge current and next stride source + exp_src_lo = _mm256_unpacklo_epi8(src_reg, src_next_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, src_next_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, xfilter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, xfilter); + + // add 8 to the source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide the source by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // convert each 16 bit to 8 bit to each low and high lane source + src_pack = _mm256_packus_epi16(exp_src_lo, exp_src_hi); + for (i = 0; i < height ; i++) { + src+= src_stride; + // load source and another source starting from the next + // following byte + destination + src_reg = _mm256_loadu_si256((__m256i const *) (src)); + src_next_reg = _mm256_loadu_si256((__m256i const *) (src + 1)); + dst_reg = _mm256_load_si256((__m256i const *) (dst)); + + // merge current and next stride source + exp_src_lo = _mm256_unpacklo_epi8(src_reg, src_next_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_reg, src_next_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, xfilter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, xfilter); + + // add 8 to source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide source by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // convert each 16 bit to 8 bit to each low and high lane source + src_reg = _mm256_packus_epi16(exp_src_lo, exp_src_hi); + + // merge previous pack to current pack source + exp_src_lo = _mm256_unpacklo_epi8(src_pack, src_reg); + exp_src_hi = _mm256_unpackhi_epi8(src_pack, src_reg); + + // filter the source + exp_src_lo = _mm256_maddubs_epi16(exp_src_lo, yfilter); + exp_src_hi = _mm256_maddubs_epi16(exp_src_hi, yfilter); + + // expand each byte to 2 bytes + exp_dst_lo = _mm256_unpacklo_epi8(dst_reg, zero_reg); + exp_dst_hi = _mm256_unpackhi_epi8(dst_reg, zero_reg); + + // add 8 to source + exp_src_lo = _mm256_add_epi16(exp_src_lo, pw8); + exp_src_hi = _mm256_add_epi16(exp_src_hi, pw8); + + // divide source by 16 + exp_src_lo = _mm256_srai_epi16(exp_src_lo, 4); + exp_src_hi = _mm256_srai_epi16(exp_src_hi, 4); + + // source - dest + exp_src_lo = _mm256_sub_epi16(exp_src_lo, exp_dst_lo); + exp_src_hi = _mm256_sub_epi16(exp_src_hi, exp_dst_hi); + + // caculate sum + sum_reg = _mm256_add_epi16(sum_reg, exp_src_lo); + exp_src_lo = _mm256_madd_epi16(exp_src_lo, exp_src_lo); + sum_reg = _mm256_add_epi16(sum_reg, exp_src_hi); + exp_src_hi = _mm256_madd_epi16(exp_src_hi, exp_src_hi); + + // calculate sse + sse_reg = _mm256_add_epi32(sse_reg, exp_src_lo); + sse_reg = _mm256_add_epi32(sse_reg, exp_src_hi); + + src_pack = src_reg; + dst+= dst_stride; + } + } + } + // sum < 0 + res_cmp = _mm256_cmpgt_epi16(zero_reg, sum_reg); + // save the next 8 bytes of each lane of sse + sse_reg_hi = _mm256_srli_si256(sse_reg, 8); + // merge the result of sum < 0 with sum to add sign to the next 16 bits + sum_reg_lo = _mm256_unpacklo_epi16(sum_reg, res_cmp); + sum_reg_hi = _mm256_unpackhi_epi16(sum_reg, res_cmp); + // add each 8 bytes from every lane of sse and sum + sse_reg = _mm256_add_epi32(sse_reg, sse_reg_hi); + sum_reg = _mm256_add_epi32(sum_reg_lo, sum_reg_hi); + + // save the next 4 bytes of each lane sse + sse_reg_hi = _mm256_srli_si256(sse_reg, 4); + // save the next 8 bytes of each lane of sum + sum_reg_hi = _mm256_srli_si256(sum_reg, 8); + + // add the first 4 bytes to the next 4 bytes sse + sse_reg = _mm256_add_epi32(sse_reg, sse_reg_hi); + // add the first 8 bytes to the next 8 bytes + sum_reg = _mm256_add_epi32(sum_reg, sum_reg_hi); + // extract the low lane and the high lane and add the results + *((int*)sse)= _mm_cvtsi128_si32(_mm256_castsi256_si128(sse_reg)) + + _mm_cvtsi128_si32(_mm256_extractf128_si256(sse_reg, 1)); + sum_reg_hi = _mm256_srli_si256(sum_reg, 4); + sum_reg = _mm256_add_epi32(sum_reg, sum_reg_hi); + sum = _mm_cvtsi128_si32(_mm256_castsi256_si128(sum_reg)) + + _mm_cvtsi128_si32(_mm256_extractf128_si256(sum_reg, 1)); + return sum; +} diff --git a/vp9/encoder/x86/vp9_variance_avx2.c b/vp9/encoder/x86/vp9_variance_avx2.c index c9b90d52d..02007a3bd 100644 --- a/vp9/encoder/x86/vp9_variance_avx2.c +++ b/vp9/encoder/x86/vp9_variance_avx2.c @@ -42,6 +42,18 @@ void vp9_get32x32var_avx2 int *Sum ); +unsigned int vp9_sub_pixel_variance32xh_avx2 +( + const uint8_t *src, + int src_stride, + int x_offset, + int y_offset, + const uint8_t *dst, + int dst_stride, + int height, + unsigned int *sse +); + static void variance_avx2(const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int recon_stride, int w, int h, unsigned int *sse, int *sum, @@ -155,3 +167,43 @@ unsigned int vp9_variance64x32_avx2(const uint8_t *src_ptr, *sse = var; return (var - (((int64_t)avg * avg) >> 11)); } + +unsigned int vp9_sub_pixel_variance64x64_avx2(const uint8_t *src, + int src_stride, + int x_offset, + int y_offset, + const uint8_t *dst, + int dst_stride, + unsigned int *sse_ptr) { + // processing 32 elements in parallel + unsigned int sse; + int se = vp9_sub_pixel_variance32xh_avx2(src, src_stride, x_offset, + y_offset, dst, dst_stride, + 64, &sse); + // processing the next 32 elements in parallel + unsigned int sse2; + int se2 = vp9_sub_pixel_variance32xh_avx2(src + 32, src_stride, + x_offset, y_offset, + dst + 32, dst_stride, + 64, &sse2); + se += se2; + sse += sse2; + *sse_ptr = sse; + return sse - (((int64_t)se * se) >> 12); +} + +unsigned int vp9_sub_pixel_variance32x32_avx2(const uint8_t *src, + int src_stride, + int x_offset, + int y_offset, + const uint8_t *dst, + int dst_stride, + unsigned int *sse_ptr) { + // processing 32 element in parallel + unsigned int sse; + int se = vp9_sub_pixel_variance32xh_avx2(src, src_stride, x_offset, + y_offset, dst, dst_stride, + 32, &sse); + *sse_ptr = sse; + return sse - (((int64_t)se * se) >> 10); +} diff --git a/vp9/vp9_common.mk b/vp9/vp9_common.mk index 48d6a7ca0..a448b3c52 100644 --- a/vp9/vp9_common.mk +++ b/vp9/vp9_common.mk @@ -80,6 +80,7 @@ VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_subpixel_bilinear_sse2.asm VP9_COMMON_SRCS-$(HAVE_SSSE3) += common/x86/vp9_subpixel_8t_ssse3.asm VP9_COMMON_SRCS-$(HAVE_SSSE3) += common/x86/vp9_subpixel_bilinear_ssse3.asm VP9_COMMON_SRCS-$(HAVE_AVX2) += common/x86/vp9_subpixel_8t_intrin_avx2.c +VP9_COMMON_SRCS-$(HAVE_SSSE3) += common/x86/vp9_subpixel_8t_intrin_ssse3.c ifeq ($(CONFIG_VP9_POSTPROC),yes) VP9_COMMON_SRCS-$(HAVE_MMX) += common/x86/vp9_postproc_mmx.asm VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_postproc_sse2.asm diff --git a/vp9/vp9_cx_iface.c b/vp9/vp9_cx_iface.c index fbdad7414..a4162e942 100644 --- a/vp9/vp9_cx_iface.c +++ b/vp9/vp9_cx_iface.c @@ -264,7 +264,7 @@ static vpx_codec_err_t validate_img(vpx_codec_alg_priv_t *ctx, static vpx_codec_err_t set_vp9e_config(VP9_CONFIG *oxcf, vpx_codec_enc_cfg_t cfg, - struct vp9_extracfg vp8_cfg) { + struct vp9_extracfg vp9_cfg) { oxcf->version = cfg.g_profile; oxcf->width = cfg.g_w; oxcf->height = cfg.g_h; @@ -289,10 +289,8 @@ static vpx_codec_err_t set_vp9e_config(VP9_CONFIG *oxcf, } if (cfg.g_pass == VPX_RC_FIRST_PASS) { - oxcf->allow_lag = 0; oxcf->lag_in_frames = 0; } else { - oxcf->allow_lag = (cfg.g_lag_in_frames) > 0; oxcf->lag_in_frames = cfg.g_lag_in_frames; } @@ -305,11 +303,11 @@ static vpx_codec_err_t set_vp9e_config(VP9_CONFIG *oxcf, oxcf->end_usage = USAGE_STREAM_FROM_SERVER; oxcf->target_bandwidth = cfg.rc_target_bitrate; - oxcf->rc_max_intra_bitrate_pct = vp8_cfg.rc_max_intra_bitrate_pct; + oxcf->rc_max_intra_bitrate_pct = vp9_cfg.rc_max_intra_bitrate_pct; oxcf->best_allowed_q = cfg.rc_min_quantizer; oxcf->worst_allowed_q = cfg.rc_max_quantizer; - oxcf->cq_level = vp8_cfg.cq_level; + oxcf->cq_level = vp9_cfg.cq_level; oxcf->fixed_q = -1; oxcf->under_shoot_pct = cfg.rc_undershoot_pct; @@ -330,30 +328,30 @@ static vpx_codec_err_t set_vp9e_config(VP9_CONFIG *oxcf, // oxcf->kf_min_dist = cfg.kf_min_dis; oxcf->key_freq = cfg.kf_max_dist; - oxcf->cpu_used = vp8_cfg.cpu_used; - oxcf->encode_breakout = vp8_cfg.static_thresh; - oxcf->play_alternate = vp8_cfg.enable_auto_alt_ref; - oxcf->noise_sensitivity = vp8_cfg.noise_sensitivity; - oxcf->sharpness = vp8_cfg.sharpness; + oxcf->cpu_used = vp9_cfg.cpu_used; + oxcf->encode_breakout = vp9_cfg.static_thresh; + oxcf->play_alternate = vp9_cfg.enable_auto_alt_ref; + oxcf->noise_sensitivity = vp9_cfg.noise_sensitivity; + oxcf->sharpness = vp9_cfg.sharpness; oxcf->two_pass_stats_in = cfg.rc_twopass_stats_in; - oxcf->output_pkt_list = vp8_cfg.pkt_list; + oxcf->output_pkt_list = vp9_cfg.pkt_list; - oxcf->arnr_max_frames = vp8_cfg.arnr_max_frames; - oxcf->arnr_strength = vp8_cfg.arnr_strength; - oxcf->arnr_type = vp8_cfg.arnr_type; + oxcf->arnr_max_frames = vp9_cfg.arnr_max_frames; + oxcf->arnr_strength = vp9_cfg.arnr_strength; + oxcf->arnr_type = vp9_cfg.arnr_type; - oxcf->tuning = vp8_cfg.tuning; + oxcf->tuning = vp9_cfg.tuning; - oxcf->tile_columns = vp8_cfg.tile_columns; - oxcf->tile_rows = vp8_cfg.tile_rows; + oxcf->tile_columns = vp9_cfg.tile_columns; + oxcf->tile_rows = vp9_cfg.tile_rows; - oxcf->lossless = vp8_cfg.lossless; + oxcf->lossless = vp9_cfg.lossless; oxcf->error_resilient_mode = cfg.g_error_resilient; - oxcf->frame_parallel_decoding_mode = vp8_cfg.frame_parallel_decoding_mode; + oxcf->frame_parallel_decoding_mode = vp9_cfg.frame_parallel_decoding_mode; - oxcf->aq_mode = vp8_cfg.aq_mode; + oxcf->aq_mode = vp9_cfg.aq_mode; oxcf->ss_number_layers = cfg.ss_number_layers; @@ -390,7 +388,6 @@ static vpx_codec_err_t set_vp9e_config(VP9_CONFIG *oxcf, printf("two_pass_vbrbias: %d\n", oxcf->two_pass_vbrbias); printf("two_pass_vbrmin_section: %d\n", oxcf->two_pass_vbrmin_section); printf("two_pass_vbrmax_section: %d\n", oxcf->two_pass_vbrmax_section); - printf("allow_lag: %d\n", oxcf->allow_lag); printf("lag_in_frames: %d\n", oxcf->lag_in_frames); printf("play_alternate: %d\n", oxcf->play_alternate); printf("Version: %d\n", oxcf->Version); diff --git a/vp9/vp9_dx_iface.c b/vp9/vp9_dx_iface.c index 76cbebf19..b85e17237 100644 --- a/vp9/vp9_dx_iface.c +++ b/vp9/vp9_dx_iface.c @@ -159,11 +159,7 @@ static vpx_codec_err_t vp9_peek_si(const uint8_t *data, unsigned int data_sz, if (frame_marker != VP9_FRAME_MARKER) return VPX_CODEC_UNSUP_BITSTREAM; -#if CONFIG_NON420 if (version > 1) return VPX_CODEC_UNSUP_BITSTREAM; -#else - if (version != 0) return VPX_CODEC_UNSUP_BITSTREAM; -#endif if (vp9_rb_read_bit(&rb)) { // show an existing frame return VPX_CODEC_OK; diff --git a/vp9/vp9cx.mk b/vp9/vp9cx.mk index c0d973b4f..27dd6f625 100644 --- a/vp9/vp9cx.mk +++ b/vp9/vp9cx.mk @@ -86,6 +86,7 @@ VP9_CX_SRCS-$(HAVE_SSE2) += encoder/x86/vp9_variance_impl_sse2.asm VP9_CX_SRCS-$(HAVE_AVX2) += encoder/x86/vp9_variance_impl_intrin_avx2.c VP9_CX_SRCS-$(HAVE_SSE2) += encoder/x86/vp9_sad4d_sse2.asm VP9_CX_SRCS-$(HAVE_SSE2) += encoder/x86/vp9_subpel_variance_impl_sse2.asm +VP9_CX_SRCS-$(HAVE_AVX2) += encoder/x86/vp9_subpel_variance_impl_intrin_avx2.c VP9_CX_SRCS-$(HAVE_SSE2) += encoder/x86/vp9_temporal_filter_apply_sse2.asm VP9_CX_SRCS-$(HAVE_SSE3) += encoder/x86/vp9_sad_sse3.asm diff --git a/vpx/src/svc_encodeframe.c b/vpx/src/svc_encodeframe.c index 4f5ba6f20..adce47637 100644 --- a/vpx/src/svc_encodeframe.c +++ b/vpx/src/svc_encodeframe.c @@ -23,11 +23,13 @@ #include "vpx/vp8cx.h" #include "vpx/vpx_encoder.h" -#if defined(__MINGW32__) && !defined(MINGW_HAS_SECURE_API) +#ifdef __MINGW32__ #define strtok_r strtok_s +#ifndef MINGW_HAS_SECURE_API // proto from /usr/x86_64-w64-mingw32/include/sec_api/string_s.h _CRTIMP char *__cdecl strtok_s(char *str, const char *delim, char **context); -#endif +#endif /* MINGW_HAS_SECURE_API */ +#endif /* __MINGW32__ */ #ifdef _MSC_VER #define strdup _strdup diff --git a/vpx/vp8cx.h b/vpx/vp8cx.h index d0ac1afc8..0b637d42f 100644 --- a/vpx/vp8cx.h +++ b/vpx/vp8cx.h @@ -195,6 +195,11 @@ enum vp8e_enc_control_id { VP9E_SET_SVC, VP9E_SET_SVC_PARAMETERS, + /*!\brief control function to set svc layer for spatial and temporal. + * \note Valid ranges: 0..#vpx_codec_enc_cfg::ss_number_layers for spatial + * layer and 0..#vpx_codec_enc_cfg::ts_number_layers for + * temporal layer. + */ VP9E_SET_SVC_LAYER_ID }; @@ -297,9 +302,16 @@ typedef struct vpx_svc_parameters { int alt_fb_idx; /**< alt reference frame frame buffer index */ } vpx_svc_parameters_t; +/*!\brief vp9 svc layer parameters + * + * This defines the spatial and temporal layer id numbers for svc encoding. + * This is used with the #VP9E_SET_SVC_LAYER_ID control to set the spatial and + * temporal layer id for the current frame. + * + */ typedef struct vpx_svc_layer_id { - int spatial_layer_id; - int temporal_layer_id; + int spatial_layer_id; /**< Spatial layer id number. */ + int temporal_layer_id; /**< Temporal layer id number. */ } vpx_svc_layer_id_t; /*!\brief VP8 encoder control function parameter type diff --git a/vpx/vpx_frame_buffer.h b/vpx/vpx_frame_buffer.h index 6803759d7..e69df4bc8 100644 --- a/vpx/vpx_frame_buffer.h +++ b/vpx/vpx_frame_buffer.h @@ -11,6 +11,10 @@ #ifndef VPX_VPX_FRAME_BUFFER_H_ #define VPX_VPX_FRAME_BUFFER_H_ +/*!\file + * \brief Describes the decoder external frame buffer interface. + */ + #ifdef __cplusplus extern "C" { #endif @@ -298,10 +298,11 @@ int file_is_raw(struct VpxInputContext *input) { return is_raw; } -void show_progress(int frame_in, int frame_out, unsigned long dx_time) { - fprintf(stderr, "%d decoded frames/%d showed frames in %lu us (%.2f fps)\r", +void show_progress(int frame_in, int frame_out, uint64_t dx_time) { + fprintf(stderr, + "%d decoded frames/%d showed frames in %"PRId64" us (%.2f fps)\r", frame_in, frame_out, dx_time, - (float)frame_out * 1000000.0 / (float)dx_time); + (double)frame_out * 1000000.0 / (double)dx_time); } struct ExternalFrameBuffer { @@ -493,7 +494,7 @@ int main_loop(int argc, const char **argv_) { int ec_enabled = 0; const VpxInterface *interface = NULL; const VpxInterface *fourcc_interface = NULL; - unsigned long dx_time = 0; + uint64_t dx_time = 0; struct arg arg; char **argv, **argi, **argj; @@ -803,7 +804,7 @@ int main_loop(int argc, const char **argv_) { } vpx_usec_timer_mark(&timer); - dx_time += (unsigned int)vpx_usec_timer_elapsed(&timer); + dx_time += vpx_usec_timer_elapsed(&timer); } } @@ -231,6 +231,10 @@ static const arg_def_t disable_warnings = static const arg_def_t disable_warning_prompt = ARG_DEF("y", "disable-warning-prompt", 0, "Display warnings, but do not prompt user to continue."); +static const arg_def_t experimental_bitstream = + ARG_DEF(NULL, "experimental-bitstream", 0, + "Allow experimental bitstream features."); + static const arg_def_t *main_args[] = { &debugmode, @@ -713,6 +717,8 @@ static void parse_global_config(struct VpxEncoderConfig *global, char **argv) { global->disable_warnings = 1; else if (arg_match(&arg, &disable_warning_prompt, argi)) global->disable_warning_prompt = 1; + else if (arg_match(&arg, &experimental_bitstream, argi)) + global->experimental_bitstream = 1; else argj++; } @@ -1002,13 +1008,20 @@ static int parse_stream_params(struct VpxEncoderConfig *global, } while (0) -static void validate_stream_config(struct stream_state *stream) { - struct stream_state *streami; +static void validate_stream_config(const struct stream_state *stream, + const struct VpxEncoderConfig *global) { + const struct stream_state *streami; if (!stream->config.cfg.g_w || !stream->config.cfg.g_h) fatal("Stream %d: Specify stream dimensions with --width (-w) " " and --height (-h)", stream->index); + if (stream->config.cfg.g_profile != 0 && !global->experimental_bitstream) { + fatal("Stream %d: profile %d is experimental and requires the --%s flag", + stream->index, stream->config.cfg.g_profile, + experimental_bitstream.long_name); + } + for (streami = stream; streami; streami = streami->next) { /* All streams require output files */ if (!streami->config.out_fn) @@ -1540,11 +1553,9 @@ int main(int argc, const char **argv_) { if (!input.filename) usage_exit(); -#if CONFIG_NON420 /* Decide if other chroma subsamplings than 4:2:0 are supported */ if (global.codec->fourcc == VP9_FOURCC) input.only_i420 = 0; -#endif for (pass = global.pass ? global.pass - 1 : 0; pass < global.passes; pass++) { int frames_in = 0, seen_frames = 0; @@ -1571,7 +1582,7 @@ int main(int argc, const char **argv_) { fatal("Specify stream dimensions with --width (-w) " " and --height (-h)"); FOREACH_STREAM(set_stream_dimensions(stream, input.width, input.height)); - FOREACH_STREAM(validate_stream_config(stream)); + FOREACH_STREAM(validate_stream_config(stream, &global)); /* Ensure that --passes and --pass are consistent. If --pass is set and * --passes=2, ensure --fpf was set. @@ -46,6 +46,7 @@ struct VpxEncoderConfig { int show_rate_hist_buckets; int disable_warnings; int disable_warning_prompt; + int experimental_bitstream; }; #ifdef __cplusplus @@ -12,7 +12,7 @@ #include <stdarg.h> -#include "nestegg/include/nestegg/nestegg.h" +#include "third_party/nestegg/include/nestegg/nestegg.h" static int nestegg_read_cb(void *buffer, size_t length, void *userdata) { FILE *f = userdata; @@ -65,7 +65,7 @@ int file_is_webm(struct WebmInputContext *webm_ctx, nestegg_video_params params; io.userdata = vpx_ctx->file; - if (nestegg_init(&webm_ctx->nestegg_ctx, io, NULL)) + if (nestegg_init(&webm_ctx->nestegg_ctx, io, NULL, -1)) goto fail; if (nestegg_track_count(webm_ctx->nestegg_ctx, &n)) |