December 2018 - ffmpeg-cvslog

lavc/kvazaar: fix auto thread flag in kvazaar wrapper.
by Jun Zhao 02 Dec '18

02 Dec '18

ffmpeg | branch: master | Jun Zhao <mypopydev(a)gmail.com> | Mon Nov 19 00:12:21 2018 +0800| [72b047a7a706b11f9c5f33d89434d288b7ec1384] | committer: Jun Zhao lavc/kvazaar: fix auto thread flag in kvazaar wrapper. Now the kvazaar warpper didn't setting the threads for kvazaar API, and kavzaar will auto selecte the thread number. Signed-off-by: Jun Zhao <mypopydev(a)gmail.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=72b047a7a706b11f9… --- libavcodec/libkvazaar.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavcodec/libkvazaar.c b/libavcodec/libkvazaar.c index 5bc5b4ebf1..8f50bef669 100644 --- a/libavcodec/libkvazaar.c +++ b/libavcodec/libkvazaar.c @@ -293,7 +293,7 @@ AVCodec ff_libkvazaar_encoder = { .long_name = NULL_IF_CONFIG_SMALL("libkvazaar H.265 / HEVC"), .type = AVMEDIA_TYPE_VIDEO, .id = AV_CODEC_ID_HEVC, - .capabilities = AV_CODEC_CAP_DELAY, + .capabilities = AV_CODEC_CAP_DELAY | AV_CODEC_CAP_AUTO_THREADS, .pix_fmts = pix_fmts, .priv_class = &class,

1 0

avdevice/decklink_enc: add support for setting genlock timing offset
by Marton Balint 02 Dec '18

02 Dec '18

ffmpeg | branch: master | Marton Balint <cus(a)passwd.hu> | Sun Nov 25 01:38:05 2018 +0100| [c0479010125b8ec7eafe6cd55574cee9c3aeb475] | committer: Marton Balint avdevice/decklink_enc: add support for setting genlock timing offset Signed-off-by: Marton Balint <cus(a)passwd.hu> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=c0479010125b8ec7e… --- doc/outdevs.texi | 4 ++++ libavdevice/decklink_common.cpp | 5 +++++ libavdevice/decklink_common_c.h | 1 + libavdevice/decklink_enc_c.c | 2 ++ libavdevice/version.h | 2 +- 5 files changed, 13 insertions(+), 1 deletion(-) diff --git a/doc/outdevs.texi b/doc/outdevs.texi index 7509ac695d..1fd83c28cc 100644 --- a/doc/outdevs.texi +++ b/doc/outdevs.texi @@ -155,6 +155,10 @@ Defaults to @option{0.5}. Sets the decklink device duplex mode. Must be @samp{unset}, @samp{half} or @samp{full}. Defaults to @samp{unset}. +@item timing_offset +Sets the genlock timing pixel offset on the used output. +Defaults to @samp{unset}. + @end table @subsection Examples diff --git a/libavdevice/decklink_common.cpp b/libavdevice/decklink_common.cpp index 130e70b2ca..d3cc1eb3d1 100644 --- a/libavdevice/decklink_common.cpp +++ b/libavdevice/decklink_common.cpp @@ -171,6 +171,11 @@ int ff_decklink_set_configs(AVFormatContext *avctx, if (ret < 0) return ret; } + if (direction == DIRECTION_OUT && cctx->timing_offset != INT_MIN) { + res = ctx->cfg->SetInt(bmdDeckLinkConfigReferenceInputTimingOffset, cctx->timing_offset); + if (res != S_OK) + av_log(avctx, AV_LOG_WARNING, "Setting timing offset failed.\n"); + } return 0; } diff --git a/libavdevice/decklink_common_c.h b/libavdevice/decklink_common_c.h index 8e3bbeb7df..ca85ec2504 100644 --- a/libavdevice/decklink_common_c.h +++ b/libavdevice/decklink_common_c.h @@ -57,6 +57,7 @@ struct decklink_cctx { int64_t queue_size; int copyts; int64_t timestamp_align; + int timing_offset; }; #endif /* AVDEVICE_DECKLINK_COMMON_C_H */ diff --git a/libavdevice/decklink_enc_c.c b/libavdevice/decklink_enc_c.c index 6169078159..63cbd39ecd 100644 --- a/libavdevice/decklink_enc_c.c +++ b/libavdevice/decklink_enc_c.c @@ -35,6 +35,8 @@ static const AVOption options[] = { { "unset" , NULL , 0 , AV_OPT_TYPE_CONST , { .i64 = 0 }, 0, 0, ENC, "duplex_mode"}, { "half" , NULL , 0 , AV_OPT_TYPE_CONST , { .i64 = 1 }, 0, 0, ENC, "duplex_mode"}, { "full" , NULL , 0 , AV_OPT_TYPE_CONST , { .i64 = 2 }, 0, 0, ENC, "duplex_mode"}, + { "timing_offset", "genlock timing pixel offset", OFFSET(timing_offset), AV_OPT_TYPE_INT, { .i64 = INT_MIN }, INT_MIN, INT_MAX, ENC, "timing_offset"}, + { "unset" , NULL , 0 , AV_OPT_TYPE_CONST, { .i64 = INT_MIN }, 0, 0, ENC, "timing_offset"}, { NULL }, }; diff --git a/libavdevice/version.h b/libavdevice/version.h index edf26bdc1a..e6ae2c44fc 100644 --- a/libavdevice/version.h +++ b/libavdevice/version.h @@ -29,7 +29,7 @@ #define LIBAVDEVICE_VERSION_MAJOR 58 #define LIBAVDEVICE_VERSION_MINOR 6 -#define LIBAVDEVICE_VERSION_MICRO 100 +#define LIBAVDEVICE_VERSION_MICRO 101 #define LIBAVDEVICE_VERSION_INT AV_VERSION_INT(LIBAVDEVICE_VERSION_MAJOR, \ LIBAVDEVICE_VERSION_MINOR, \

1 0

ffplay: fix -x and -y options when only one of them is used
by Marton Balint 02 Dec '18

02 Dec '18

ffmpeg | branch: master | Marton Balint <cus(a)passwd.hu> | Sat Nov 24 23:09:24 2018 +0100| [418c90faac6522c73481650d000e9f5f030187cb] | committer: Marton Balint ffplay: fix -x and -y options when only one of them is used Previously if only -y was used then nothing happened, if only -x then zero window height was set which is undefined in SDL and caused a black window. >From now on if only one dimension is set we calculate the other based on size/aspect of the shown content. Signed-off-by: Marton Balint <cus(a)passwd.hu> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=418c90faac6522c73… --- fftools/ffplay.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/fftools/ffplay.c b/fftools/ffplay.c index 37a853fbd9..6a195e5542 100644 --- a/fftools/ffplay.c +++ b/fftools/ffplay.c @@ -1322,7 +1322,11 @@ static void sigterm_handler(int sig) static void set_default_window_size(int width, int height, AVRational sar) { SDL_Rect rect; - calculate_display_rect(&rect, 0, 0, INT_MAX, height, width, height, sar); + int max_width = screen_width ? screen_width : INT_MAX; + int max_height = screen_height ? screen_height : INT_MAX; + if (max_width == INT_MAX && max_height == INT_MAX) + max_height = height; + calculate_display_rect(&rect, 0, 0, max_width, max_height, width, height, sar); default_width = rect.w; default_height = rect.h; } @@ -1331,13 +1335,8 @@ static int video_open(VideoState *is) { int w,h; - if (screen_width) { - w = screen_width; - h = screen_height; - } else { - w = default_width; - h = default_height; - } + w = screen_width ? screen_width : default_width; + h = screen_height ? screen_height : default_height; if (!window_title) window_title = input_filename;

1 0

ffplay: convert float math to int math in calculate_display_rect
by Marton Balint 02 Dec '18

02 Dec '18

ffmpeg | branch: master | Marton Balint <cus(a)passwd.hu> | Sat Nov 24 23:58:04 2018 +0100| [9a39c5449f5aef1c02bd46538bbbad4590064602] | committer: Marton Balint ffplay: convert float math to int math in calculate_display_rect Signed-off-by: Marton Balint <cus(a)passwd.hu> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=9a39c5449f5aef1c0… --- fftools/ffplay.c | 22 +++++++++------------- 1 file changed, 9 insertions(+), 13 deletions(-) diff --git a/fftools/ffplay.c b/fftools/ffplay.c index ab1f9faccf..37a853fbd9 100644 --- a/fftools/ffplay.c +++ b/fftools/ffplay.c @@ -861,31 +861,27 @@ static void calculate_display_rect(SDL_Rect *rect, int scr_xleft, int scr_ytop, int scr_width, int scr_height, int pic_width, int pic_height, AVRational pic_sar) { - float aspect_ratio; - int width, height, x, y; + AVRational aspect_ratio = pic_sar; + int64_t width, height, x, y; - if (pic_sar.num == 0) - aspect_ratio = 0; - else - aspect_ratio = av_q2d(pic_sar); + if (av_cmp_q(aspect_ratio, av_make_q(0, 1)) <= 0) + aspect_ratio = av_make_q(1, 1); - if (aspect_ratio <= 0.0) - aspect_ratio = 1.0; - aspect_ratio *= (float)pic_width / (float)pic_height; + aspect_ratio = av_mul_q(aspect_ratio, av_make_q(pic_width, pic_height)); /* XXX: we suppose the screen has a 1.0 pixel ratio */ height = scr_height; - width = lrint(height * aspect_ratio) & ~1; + width = av_rescale(height, aspect_ratio.num, aspect_ratio.den) & ~1; if (width > scr_width) { width = scr_width; - height = lrint(width / aspect_ratio) & ~1; + height = av_rescale(width, aspect_ratio.den, aspect_ratio.num) & ~1; } x = (scr_width - width) / 2; y = (scr_height - height) / 2; rect->x = scr_xleft + x; rect->y = scr_ytop + y; - rect->w = FFMAX(width, 1); - rect->h = FFMAX(height, 1); + rect->w = FFMAX((int)width, 1); + rect->h = FFMAX((int)height, 1); } static void get_sdl_pix_fmt_and_blendmode(int format, Uint32 *sdl_pix_fmt, SDL_BlendMode *sdl_blendmode)

1 0

avformat/mpegenc: extend muxing PCM-DVD to other depths
by Paul B Mahol 01 Dec '18

01 Dec '18

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Tue Nov 27 12:09:03 2018 +0100| [2a08faba88027ec4679910bab9ff23925d8f6142] | committer: Paul B Mahol avformat/mpegenc: extend muxing PCM-DVD to other depths Fixes #6783. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=2a08faba88027ec46… --- libavformat/mpegenc.c | 27 +++++++++++++++++++++------ libavformat/version.h | 2 +- 2 files changed, 22 insertions(+), 7 deletions(-) diff --git a/libavformat/mpegenc.c b/libavformat/mpegenc.c index 4c6fa67fb8..1389288b7f 100644 --- a/libavformat/mpegenc.c +++ b/libavformat/mpegenc.c @@ -364,12 +364,7 @@ static av_cold int mpeg_mux_init(AVFormatContext *ctx) stream->id = ac3_id++; } else if (st->codecpar->codec_id == AV_CODEC_ID_DTS) { stream->id = dts_id++; - } else if (st->codecpar->codec_id == AV_CODEC_ID_PCM_S16BE || - st->codecpar->codec_id == AV_CODEC_ID_PCM_DVD) { - if (st->codecpar->bits_per_coded_sample != 16) { - av_log(ctx, AV_LOG_ERROR, "Only 16 bit LPCM streams can be muxed.\n"); - goto fail; - } + } else if (st->codecpar->codec_id == AV_CODEC_ID_PCM_S16BE) { stream->id = lpcm_id++; for (j = 0; j < 4; j++) { if (lpcm_freq_tab[j] == st->codecpar->sample_rate) @@ -392,6 +387,26 @@ static av_cold int mpeg_mux_init(AVFormatContext *ctx) stream->lpcm_header[1] = (st->codecpar->channels - 1) | (j << 4); stream->lpcm_header[2] = 0x80; stream->lpcm_align = st->codecpar->channels * 2; + } else if (st->codecpar->codec_id == AV_CODEC_ID_PCM_DVD) { + int freq; + + switch (st->codecpar->sample_rate) { + case 48000: freq = 0; break; + case 96000: freq = 1; break; + case 44100: freq = 2; break; + case 32000: freq = 3; break; + default: + av_log(ctx, AV_LOG_ERROR, "Unsupported sample rate.\n"); + return AVERROR(EINVAL); + } + + stream->lpcm_header[0] = 0x0c; + stream->lpcm_header[1] = (freq << 4) | + (((st->codecpar->bits_per_coded_sample - 16) / 4) << 6) | + st->codecpar->channels - 1; + stream->lpcm_header[2] = 0x80; + stream->id = lpcm_id++; + stream->lpcm_align = st->codecpar->channels * st->codecpar->bits_per_coded_sample / 8; } else { stream->id = mpa_id++; } diff --git a/libavformat/version.h b/libavformat/version.h index 824b9e7cd6..30f2a84ddb 100644 --- a/libavformat/version.h +++ b/libavformat/version.h @@ -33,7 +33,7 @@ // Also please add any ticket numbers that you believe might be affected here #define LIBAVFORMAT_VERSION_MAJOR 58 #define LIBAVFORMAT_VERSION_MINOR 23 -#define LIBAVFORMAT_VERSION_MICRO 101 +#define LIBAVFORMAT_VERSION_MICRO 102 #define LIBAVFORMAT_VERSION_INT AV_VERSION_INT(LIBAVFORMAT_VERSION_MAJOR, \ LIBAVFORMAT_VERSION_MINOR, \

1 0

avcodec: add PCM-DVD encoder
by Paul B Mahol 01 Dec '18

01 Dec '18

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Mon Nov 26 19:03:58 2018 +0100| [e9967822e41213acb95e4a1b3a9bbfd240bc12c9] | committer: Paul B Mahol avcodec: add PCM-DVD encoder Fixes #6784. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=e9967822e41213acb… --- Changelog | 1 + libavcodec/Makefile | 1 + libavcodec/allcodecs.c | 1 + libavcodec/pcm-dvdenc.c | 197 ++++++++++++++++++++++++++++++++++++++++++++++++ libavcodec/version.h | 2 +- 5 files changed, 201 insertions(+), 1 deletion(-) diff --git a/Changelog b/Changelog index 8782ac41de..1f53ff46da 100644 --- a/Changelog +++ b/Changelog @@ -9,6 +9,7 @@ version <next>: - freezedetect filter - truehd_core bitstream filter - dhav demuxer +- PCM-DVD encoder version 4.1: diff --git a/libavcodec/Makefile b/libavcodec/Makefile index 2840a8069f..5feadac729 100644 --- a/libavcodec/Makefile +++ b/libavcodec/Makefile @@ -737,6 +737,7 @@ OBJS-$(CONFIG_PCM_ALAW_DECODER) += pcm.o OBJS-$(CONFIG_PCM_ALAW_ENCODER) += pcm.o OBJS-$(CONFIG_PCM_BLURAY_DECODER) += pcm-bluray.o OBJS-$(CONFIG_PCM_DVD_DECODER) += pcm-dvd.o +OBJS-$(CONFIG_PCM_DVD_ENCODER) += pcm-dvdenc.o OBJS-$(CONFIG_PCM_F16LE_DECODER) += pcm.o OBJS-$(CONFIG_PCM_F24LE_DECODER) += pcm.o OBJS-$(CONFIG_PCM_F32BE_DECODER) += pcm.o diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c index 2c17db5a70..d70646e91a 100644 --- a/libavcodec/allcodecs.c +++ b/libavcodec/allcodecs.c @@ -494,6 +494,7 @@ extern AVCodec ff_xma2_decoder; extern AVCodec ff_pcm_alaw_encoder; extern AVCodec ff_pcm_alaw_decoder; extern AVCodec ff_pcm_bluray_decoder; +extern AVCodec ff_pcm_dvd_encoder; extern AVCodec ff_pcm_dvd_decoder; extern AVCodec ff_pcm_f16le_decoder; extern AVCodec ff_pcm_f24le_decoder; diff --git a/libavcodec/pcm-dvdenc.c b/libavcodec/pcm-dvdenc.c new file mode 100644 index 0000000000..d26eaf071c --- /dev/null +++ b/libavcodec/pcm-dvdenc.c @@ -0,0 +1,197 @@ +/* + * LPCM codecs for PCM formats found in Video DVD streams + * Copyright (c) 2018 Paul B Mahol + * + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg 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 + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with FFmpeg; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "avcodec.h" +#include "bytestream.h" +#include "internal.h" + +typedef struct PCMDVDContext { + uint8_t header[3]; // Header added to every frame + int block_size; // Size of a block of samples in bytes + int samples_per_block; // Number of samples per channel per block + int groups_per_block; // Number of 20/24-bit sample groups per block + uint8_t *extra_samples; // Pointer to leftover samples from a frame + int extra_sample_count; // Number of leftover samples in the buffer +} PCMDVDContext; + +static av_cold int pcm_dvd_encode_init(AVCodecContext *avctx) +{ + PCMDVDContext *s = avctx->priv_data; + int quant, freq, frame_size; + + switch (avctx->sample_rate) { + case 48000: + freq = 0; + break; + case 96000: + freq = 1; + break; + } + + switch (avctx->sample_fmt) { + case AV_SAMPLE_FMT_S16: + avctx->bits_per_coded_sample = 16; + quant = 0; + break; + case AV_SAMPLE_FMT_S32: + avctx->bits_per_coded_sample = 24; + quant = 2; + break; + } + + avctx->bits_per_coded_sample = 16 + quant * 4; + avctx->block_align = avctx->channels * avctx->bits_per_coded_sample / 8; + avctx->bit_rate = avctx->block_align * 8LL * avctx->sample_rate; + if (avctx->bit_rate > 9800000) { + av_log(avctx, AV_LOG_ERROR, "Too big bitrate: reduce sample rate, bitdepth or channels.\n"); + return AVERROR(EINVAL); + } + + if (avctx->sample_fmt == AV_SAMPLE_FMT_S16) { + s->samples_per_block = 1; + s->block_size = avctx->channels * 2; + frame_size = 2008 / s->block_size; + } else { + switch (avctx->channels) { + case 1: + case 2: + case 4: + /* one group has all the samples needed */ + s->block_size = 4 * avctx->bits_per_coded_sample / 8; + s->samples_per_block = 4 / avctx->channels; + s->groups_per_block = 1; + break; + case 8: + /* two groups have all the samples needed */ + s->block_size = 8 * avctx->bits_per_coded_sample / 8; + s->samples_per_block = 1; + s->groups_per_block = 2; + break; + default: + /* need avctx->channels groups */ + s->block_size = 4 * avctx->channels * + avctx->bits_per_coded_sample / 8; + s->samples_per_block = 4; + s->groups_per_block = avctx->channels; + break; + } + + frame_size = FFALIGN(2008 / s->block_size, s->samples_per_block); + } + + s->header[0] = 0x0c; + s->header[1] = (quant << 6) | (freq << 4) | (avctx->channels - 1); + s->header[2] = 0x80; + + if (!avctx->frame_size) + avctx->frame_size = frame_size; + + return 0; +} + +static int pcm_dvd_encode_frame(AVCodecContext *avctx, AVPacket *avpkt, + const AVFrame *frame, int *got_packet_ptr) +{ + PCMDVDContext *s = avctx->priv_data; + int samples = frame->nb_samples * avctx->channels; + int64_t pkt_size = (frame->nb_samples / s->samples_per_block) * s->block_size + 3; + int blocks = (pkt_size - 3) / s->block_size; + const int16_t *src16; + const int32_t *src32; + PutByteContext pb; + int ret; + + if ((ret = ff_alloc_packet2(avctx, avpkt, pkt_size, 0)) < 0) + return ret; + + memcpy(avpkt->data, s->header, 3); + + src16 = (const int16_t *)frame->data[0]; + src32 = (const int32_t *)frame->data[0]; + + bytestream2_init_writer(&pb, avpkt->data + 3, avpkt->size - 3); + + switch (avctx->sample_fmt) { + case AV_SAMPLE_FMT_S16: + do { + bytestream2_put_be16(&pb, *src16++); + } while (--samples); + break; + case AV_SAMPLE_FMT_S32: + if (avctx->channels == 1) { + do { + for (int i = 2; i; i--) { + bytestream2_put_be16(&pb, src32[0] >> 16); + bytestream2_put_be16(&pb, src32[1] >> 16); + bytestream2_put_byte(&pb, (*src32++) >> 24); + bytestream2_put_byte(&pb, (*src32++) >> 24); + } + } while (--blocks); + } else { + do { + for (int i = s->groups_per_block; i; i--) { + bytestream2_put_be16(&pb, src32[0] >> 16); + bytestream2_put_be16(&pb, src32[1] >> 16); + bytestream2_put_be16(&pb, src32[2] >> 16); + bytestream2_put_be16(&pb, src32[3] >> 16); + bytestream2_put_byte(&pb, (*src32++) >> 24); + bytestream2_put_byte(&pb, (*src32++) >> 24); + bytestream2_put_byte(&pb, (*src32++) >> 24); + bytestream2_put_byte(&pb, (*src32++) >> 24); + } + } while (--blocks); + } + break; + } + + avpkt->pts = frame->pts; + avpkt->size = pkt_size; + avpkt->duration = ff_samples_to_time_base(avctx, frame->nb_samples); + *got_packet_ptr = 1; + + return 0; +} + +static av_cold int pcm_dvd_encode_close(AVCodecContext *avctx) +{ + return 0; +} + +AVCodec ff_pcm_dvd_encoder = { + .name = "pcm_dvd", + .long_name = NULL_IF_CONFIG_SMALL("PCM signed 16|20|24-bit big-endian for DVD media"), + .type = AVMEDIA_TYPE_AUDIO, + .id = AV_CODEC_ID_PCM_DVD, + .priv_data_size = sizeof(PCMDVDContext), + .init = pcm_dvd_encode_init, + .close = pcm_dvd_encode_close, + .encode2 = pcm_dvd_encode_frame, + .capabilities = AV_CODEC_CAP_SMALL_LAST_FRAME, + .supported_samplerates = (const int[]) { 48000, 96000, 0}, + .channel_layouts = (const uint64_t[]) { AV_CH_LAYOUT_MONO, + AV_CH_LAYOUT_STEREO, + AV_CH_LAYOUT_5POINT1, + AV_CH_LAYOUT_7POINT1, + 0 }, + .sample_fmts = (const enum AVSampleFormat[]){ AV_SAMPLE_FMT_S16, + AV_SAMPLE_FMT_S32, + AV_SAMPLE_FMT_NONE }, +}; diff --git a/libavcodec/version.h b/libavcodec/version.h index 5e0b188b7a..8c81ccc2d5 100644 --- a/libavcodec/version.h +++ b/libavcodec/version.h @@ -28,7 +28,7 @@ #include "libavutil/version.h" #define LIBAVCODEC_VERSION_MAJOR 58 -#define LIBAVCODEC_VERSION_MINOR 40 +#define LIBAVCODEC_VERSION_MINOR 41 #define LIBAVCODEC_VERSION_MICRO 100 #define LIBAVCODEC_VERSION_INT AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \

1 0

avcodec/hevcdec: Check for overlapping slices
by Michael Niedermayer 01 Dec '18

01 Dec '18

ffmpeg | branch: master | Michael Niedermayer <michael(a)niedermayer.cc> | Fri Nov 16 20:03:01 2018 +0100| [1a89ae1df8585134b9524f736b71cee98de7895a] | committer: Michael Niedermayer avcodec/hevcdec: Check for overlapping slices Fixes: Timeout Fixes: 10108/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_HEVC_fuzzer-6222384351674368 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=1a89ae1df8585134b… --- libavcodec/hevcdec.c | 4 ++++ libavcodec/hevcdec.h | 1 + 2 files changed, 5 insertions(+) diff --git a/libavcodec/hevcdec.c b/libavcodec/hevcdec.c index a3b5c8cb71..10bf2563c0 100644 --- a/libavcodec/hevcdec.c +++ b/libavcodec/hevcdec.c @@ -2942,6 +2942,7 @@ static int decode_nal_unit(HEVCContext *s, const H2645NAL *nal) s->max_ra = INT_MIN; } + s->overlap ++; ret = hevc_frame_start(s); if (ret < 0) return ret; @@ -3020,6 +3021,7 @@ static int decode_nal_units(HEVCContext *s, const uint8_t *buf, int length) s->ref = NULL; s->last_eos = s->eos; s->eos = 0; + s->overlap = 0; /* split the input packet into NAL units, so we know the upper bound on the * number of slices in the frame */ @@ -3054,6 +3056,8 @@ static int decode_nal_units(HEVCContext *s, const uint8_t *buf, int length) continue; ret = decode_nal_unit(s, nal); + if (ret >= 0 && s->overlap > 2) + ret = AVERROR_INVALIDDATA; if (ret < 0) { av_log(s->avctx, AV_LOG_WARNING, "Error parsing NAL unit #%d.\n", i); diff --git a/libavcodec/hevcdec.h b/libavcodec/hevcdec.h index dd951aae06..b45969b7e2 100644 --- a/libavcodec/hevcdec.h +++ b/libavcodec/hevcdec.h @@ -430,6 +430,7 @@ typedef struct HEVCContext { int max_ra; int bs_width; int bs_height; + int overlap; int is_decoded; int no_rasl_output_flag;

1 0

avcodec/truemotion2rt: Fix rounding in input size check
by Michael Niedermayer 01 Dec '18

01 Dec '18

ffmpeg | branch: master | Michael Niedermayer <michael(a)niedermayer.cc> | Sat Nov 17 09:24:30 2018 +0100| [7f22a4ebc97817fd0968f5ea8295c9a59a6292e0] | committer: Michael Niedermayer avcodec/truemotion2rt: Fix rounding in input size check Fixes: Timeout Fixes: 11332/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_TRUEMOTION2RT_fuzzer-5678456612847616 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=7f22a4ebc97817fd0… --- libavcodec/truemotion2rt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavcodec/truemotion2rt.c b/libavcodec/truemotion2rt.c index 9df0b527bb..e3ab998fda 100644 --- a/libavcodec/truemotion2rt.c +++ b/libavcodec/truemotion2rt.c @@ -116,7 +116,7 @@ static int truemotion2rt_decode_frame(AVCodecContext *avctx, void *data, if (ret < 0) return ret; - if (avctx->width / s->hscale * avctx->height * s->delta_size > avpkt->size * 8LL * 4) + if ((avctx->width + s->hscale - 1)/ s->hscale * avctx->height * s->delta_size > avpkt->size * 8LL * 4) return AVERROR_INVALIDDATA; ret = init_get_bits8(gb, avpkt->data + ret, avpkt->size - ret);

1 0

avcodec/mips: [loongson] refine optimization in h264_chroma.
by Shiyou Yin 01 Dec '18

01 Dec '18

ffmpeg | branch: master | Shiyou Yin <yinshiyou-hf(a)loongson.cn> | Tue Nov 6 08:59:13 2018 +0800| [5982614af1f564ac05c1aa535cbd12cbbb2aa526] | committer: Michael Niedermayer avcodec/mips: [loongson] refine optimization in h264_chroma. Remove invalid operation in the case x and y all equal 0, this refine made about 2% speedup for H264 decode on loongson platform. Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=5982614af1f564ac0… --- libavcodec/mips/h264chroma_mmi.c | 970 +++++++++++++++++---------------------- 1 file changed, 426 insertions(+), 544 deletions(-) diff --git a/libavcodec/mips/h264chroma_mmi.c b/libavcodec/mips/h264chroma_mmi.c index 91b2cc470d..739dd7d4d6 100644 --- a/libavcodec/mips/h264chroma_mmi.c +++ b/libavcodec/mips/h264chroma_mmi.c @@ -30,74 +30,177 @@ void ff_put_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int h, int x, int y) { int A = 64, B, C, D, E; - double ftmp[10]; + double ftmp[12]; uint64_t tmp[1]; if (!(x || y)) { /* x=0, y=0, A=64 */ __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "mtc1 %[tmp0], %[ftmp4] \n\t" - "1: \n\t" + MMI_ULDC1(%[ftmp0], %[src], 0x00) + PTR_ADDU "%[src], %[src], %[stride] \n\t" MMI_ULDC1(%[ftmp1], %[src], 0x00) - "addi %[h], %[h], -0x04 \n\t" PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp5], %[src], 0x00) + MMI_ULDC1(%[ftmp2], %[src], 0x00) PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp6], %[src], 0x00) + MMI_ULDC1(%[ftmp3], %[src], 0x00) PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp7], %[src], 0x00) - "punpcklbh %[ftmp2], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" - "psllh %[ftmp1], %[ftmp2], %[ftmp4] \n\t" - "psllh %[ftmp2], %[ftmp3], %[ftmp4] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp4] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) + "addi %[h], %[h], -0x04 \n\t" - "punpcklbh %[ftmp2], %[ftmp5], %[ftmp0] \n\t" - "punpckhbh %[ftmp3], %[ftmp5], %[ftmp0] \n\t" - "psllh %[ftmp1], %[ftmp2], %[ftmp4] \n\t" - "psllh %[ftmp2], %[ftmp3], %[ftmp4] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp4] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_SDC1(%[ftmp0], %[dst], 0x00) PTR_ADDU "%[dst], %[dst], %[stride] \n\t" MMI_SDC1(%[ftmp1], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + MMI_SDC1(%[ftmp2], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + MMI_SDC1(%[ftmp3], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + "bnez %[h], 1b \n\t" + : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), + [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), + [dst]"+&r"(dst), [src]"+&r"(src), + [h]"+&r"(h) + : [stride]"r"((mips_reg)stride) + : "memory" + ); + } else if (x && y) { + /* x!=0, y!=0 */ + D = x * y; + B = (x << 3) - D; + C = (y << 3) - D; + A = 64 - D - B - C; - "punpcklbh %[ftmp2], %[ftmp6], %[ftmp0] \n\t" - "punpckhbh %[ftmp3], %[ftmp6], %[ftmp0] \n\t" - "psllh %[ftmp1], %[ftmp2], %[ftmp4] \n\t" - "psllh %[ftmp2], %[ftmp3], %[ftmp4] \n\t" + __asm__ volatile ( + "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" + "dli %[tmp0], 0x06 \n\t" + "pshufh %[A], %[A], %[ftmp0] \n\t" + "pshufh %[B], %[B], %[ftmp0] \n\t" + "mtc1 %[tmp0], %[ftmp9] \n\t" + "pshufh %[C], %[C], %[ftmp0] \n\t" + "pshufh %[D], %[D], %[ftmp0] \n\t" + + "1: \n\t" + MMI_ULDC1(%[ftmp1], %[src], 0x00) + MMI_ULDC1(%[ftmp2], %[src], 0x01) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULDC1(%[ftmp3], %[src], 0x00) + MMI_ULDC1(%[ftmp4], %[src], 0x01) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULDC1(%[ftmp10], %[src], 0x00) + MMI_ULDC1(%[ftmp11], %[src], 0x01) + "addi %[h], %[h], -0x02 \n\t" + + "punpcklbh %[ftmp5], %[ftmp1], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp1], %[ftmp0] \n\t" + "punpcklbh %[ftmp7], %[ftmp2], %[ftmp0] \n\t" + "punpckhbh %[ftmp8], %[ftmp2], %[ftmp0] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" + "pmullh %[ftmp7], %[ftmp7], %[B] \n\t" + "paddh %[ftmp1], %[ftmp5], %[ftmp7] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[A] \n\t" + "pmullh %[ftmp8], %[ftmp8], %[B] \n\t" + "paddh %[ftmp2], %[ftmp6], %[ftmp8] \n\t" + "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp3], %[ftmp0] \n\t" + "punpcklbh %[ftmp7], %[ftmp4], %[ftmp0] \n\t" + "punpckhbh %[ftmp8], %[ftmp4], %[ftmp0] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" + "pmullh %[ftmp7], %[ftmp7], %[D] \n\t" + "paddh %[ftmp5], %[ftmp5], %[ftmp7] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[C] \n\t" + "pmullh %[ftmp8], %[ftmp8], %[D] \n\t" + "paddh %[ftmp6], %[ftmp6], %[ftmp8] \n\t" + "paddh %[ftmp1], %[ftmp1], %[ftmp5] \n\t" + "paddh %[ftmp2], %[ftmp2], %[ftmp6] \n\t" "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp4] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" + "psrlh %[ftmp1], %[ftmp1], %[ftmp9] \n\t" + "psrlh %[ftmp2], %[ftmp2], %[ftmp9] \n\t" "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + + "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp3], %[ftmp0] \n\t" + "punpcklbh %[ftmp7], %[ftmp4], %[ftmp0] \n\t" + "punpckhbh %[ftmp8], %[ftmp4], %[ftmp0] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" + "pmullh %[ftmp7], %[ftmp7], %[B] \n\t" + "paddh %[ftmp3], %[ftmp5], %[ftmp7] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[A] \n\t" + "pmullh %[ftmp8], %[ftmp8], %[B] \n\t" + "paddh %[ftmp4], %[ftmp6], %[ftmp8] \n\t" + "punpcklbh %[ftmp5], %[ftmp10], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp10], %[ftmp0] \n\t" + "punpcklbh %[ftmp7], %[ftmp11], %[ftmp0] \n\t" + "punpckhbh %[ftmp8], %[ftmp11], %[ftmp0] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" + "pmullh %[ftmp7], %[ftmp7], %[D] \n\t" + "paddh %[ftmp5], %[ftmp5], %[ftmp7] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[C] \n\t" + "pmullh %[ftmp8], %[ftmp8], %[D] \n\t" + "paddh %[ftmp6], %[ftmp6], %[ftmp8] \n\t" + "paddh %[ftmp3], %[ftmp3], %[ftmp5] \n\t" + "paddh %[ftmp4], %[ftmp4], %[ftmp6] \n\t" + "paddh %[ftmp3], %[ftmp3], %[ff_pw_32] \n\t" + "paddh %[ftmp4], %[ftmp4], %[ff_pw_32] \n\t" + "psrlh %[ftmp3], %[ftmp3], %[ftmp9] \n\t" + "psrlh %[ftmp4], %[ftmp4], %[ftmp9] \n\t" + "packushb %[ftmp3], %[ftmp3], %[ftmp4] \n\t" + MMI_SDC1(%[ftmp1], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + MMI_SDC1(%[ftmp3], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + "bnez %[h], 1b \n\t" + : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), + [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), + [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), + [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), + [ftmp8]"=&f"(ftmp[8]), [ftmp9]"=&f"(ftmp[9]), + [ftmp10]"=&f"(ftmp[10]), [ftmp11]"=&f"(ftmp[11]), + [tmp0]"=&r"(tmp[0]), + [dst]"+&r"(dst), [src]"+&r"(src), + [h]"+&r"(h) + : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), + [A]"f"(A), [B]"f"(B), + [C]"f"(C), [D]"f"(D) + : "memory" + ); + } else if (x) { + /* x!=0, y==0 */ + E = x << 3; + A = 64 - E; + + __asm__ volatile ( + "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" + "dli %[tmp0], 0x06 \n\t" + "pshufh %[A], %[A], %[ftmp0] \n\t" + "pshufh %[E], %[E], %[ftmp0] \n\t" + "mtc1 %[tmp0], %[ftmp7] \n\t" + + "1: \n\t" + MMI_ULDC1(%[ftmp1], %[src], 0x00) + MMI_ULDC1(%[ftmp2], %[src], 0x01) + "addi %[h], %[h], -0x01 \n\t" + PTR_ADDU "%[src], %[src], %[stride] \n\t" + + "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" + "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" + "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" + "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" + "paddh %[ftmp1], %[ftmp3], %[ftmp5] \n\t" + "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" + "paddh %[ftmp2], %[ftmp4], %[ftmp6] \n\t" - "punpcklbh %[ftmp2], %[ftmp7], %[ftmp0] \n\t" - "punpckhbh %[ftmp3], %[ftmp7], %[ftmp0] \n\t" - "psllh %[ftmp1], %[ftmp2], %[ftmp4] \n\t" - "psllh %[ftmp2], %[ftmp3], %[ftmp4] \n\t" "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp4] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" + "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" + "psrlh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" MMI_SDC1(%[ftmp1], %[dst], 0x00) - - PTR_ADDU "%[src], %[src], %[stride] \n\t" PTR_ADDU "%[dst], %[dst], %[stride] \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), @@ -107,220 +210,80 @@ void ff_put_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, [tmp0]"=&r"(tmp[0]), [dst]"+&r"(dst), [src]"+&r"(src), [h]"+&r"(h) - : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32) + : [stride]"r"((mips_reg)stride), + [ff_pw_32]"f"(ff_pw_32), + [A]"f"(A), [E]"f"(E) : "memory" ); } else { - if (x && y) { - /* x!=0, y!=0 */ - D = x * y; - B = (x << 3) - D; - C = (y << 3) - D; - A = 64 - D - B - C; - - __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "pshufh %[B], %[B], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp9] \n\t" - "pshufh %[C], %[C], %[ftmp0] \n\t" - "pshufh %[D], %[D], %[ftmp0] \n\t" - - "1: \n\t" - MMI_ULDC1(%[ftmp1], %[src], 0x00) - MMI_ULDC1(%[ftmp2], %[src], 0x01) - PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp3], %[src], 0x00) - MMI_ULDC1(%[ftmp4], %[src], 0x01) - "addi %[h], %[h], -0x02 \n\t" - - "punpcklbh %[ftmp5], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp1], %[ftmp0] \n\t" - "punpcklbh %[ftmp7], %[ftmp2], %[ftmp0] \n\t" - "punpckhbh %[ftmp8], %[ftmp2], %[ftmp0] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" - "pmullh %[ftmp7], %[ftmp7], %[B] \n\t" - "paddh %[ftmp1], %[ftmp5], %[ftmp7] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[A] \n\t" - "pmullh %[ftmp8], %[ftmp8], %[B] \n\t" - "paddh %[ftmp2], %[ftmp6], %[ftmp8] \n\t" - - "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp3], %[ftmp0] \n\t" - "punpcklbh %[ftmp7], %[ftmp4], %[ftmp0] \n\t" - "punpckhbh %[ftmp8], %[ftmp4], %[ftmp0] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" - "pmullh %[ftmp7], %[ftmp7], %[D] \n\t" - "paddh %[ftmp3], %[ftmp5], %[ftmp7] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[C] \n\t" - "pmullh %[ftmp8], %[ftmp8], %[D] \n\t" - "paddh %[ftmp4], %[ftmp6], %[ftmp8] \n\t" - - "paddh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp9] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp9] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - - MMI_ULDC1(%[ftmp1], %[src], 0x00) - MMI_ULDC1(%[ftmp2], %[src], 0x01) - PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp3], %[src], 0x00) - MMI_ULDC1(%[ftmp4], %[src], 0x01) - - "punpcklbh %[ftmp5], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp1], %[ftmp0] \n\t" - "punpcklbh %[ftmp7], %[ftmp2], %[ftmp0] \n\t" - "punpckhbh %[ftmp8], %[ftmp2], %[ftmp0] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" - "pmullh %[ftmp7], %[ftmp7], %[B] \n\t" - "paddh %[ftmp1], %[ftmp5], %[ftmp7] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[A] \n\t" - "pmullh %[ftmp8], %[ftmp8], %[B] \n\t" - "paddh %[ftmp2], %[ftmp6], %[ftmp8] \n\t" - - "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp3], %[ftmp0] \n\t" - "punpcklbh %[ftmp7], %[ftmp4], %[ftmp0] \n\t" - "punpckhbh %[ftmp8], %[ftmp4], %[ftmp0] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" - "pmullh %[ftmp7], %[ftmp7], %[D] \n\t" - "paddh %[ftmp3], %[ftmp5], %[ftmp7] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[C] \n\t" - "pmullh %[ftmp8], %[ftmp8], %[D] \n\t" - "paddh %[ftmp4], %[ftmp6], %[ftmp8] \n\t" - - "paddh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp9] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp9] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - - "bnez %[h], 1b \n\t" - : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), - [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - [ftmp8]"=&f"(ftmp[8]), [ftmp9]"=&f"(ftmp[9]), - [tmp0]"=&r"(tmp[0]), - [dst]"+&r"(dst), [src]"+&r"(src), - [h]"+&r"(h) - : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), - [A]"f"(A), [B]"f"(B), - [C]"f"(C), [D]"f"(D) - : "memory" - ); - } else { - if (x) { - /* x!=0, y==0 */ - E = x << 3; - A = 64 - E; - - __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "pshufh %[E], %[E], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp7] \n\t" - - "1: \n\t" - MMI_ULDC1(%[ftmp1], %[src], 0x00) - MMI_ULDC1(%[ftmp2], %[src], 0x01) - "addi %[h], %[h], -0x01 \n\t" - PTR_ADDU "%[src], %[src], %[stride] \n\t" - - "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" - "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" - "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" - "paddh %[ftmp1], %[ftmp3], %[ftmp5] \n\t" - "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" - "paddh %[ftmp2], %[ftmp4], %[ftmp6] \n\t" - - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - "bnez %[h], 1b \n\t" - : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), - [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - [tmp0]"=&r"(tmp[0]), - [dst]"+&r"(dst), [src]"+&r"(src), - [h]"+&r"(h) - : [stride]"r"((mips_reg)stride), - [ff_pw_32]"f"(ff_pw_32), - [A]"f"(A), [E]"f"(E) - : "memory" - ); - } else { - /* x==0, y!=0 */ - E = y << 3; - A = 64 - E; - - __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "pshufh %[E], %[E], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp7] \n\t" - - "1: \n\t" - MMI_ULDC1(%[ftmp1], %[src], 0x00) - PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp2], %[src], 0x00) - "addi %[h], %[h], -0x01 \n\t" - - "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" - "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" - "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" - "paddh %[ftmp1], %[ftmp3], %[ftmp5] \n\t" - "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" - "paddh %[ftmp2], %[ftmp4], %[ftmp6] \n\t" - - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) - - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - "bnez %[h], 1b \n\t" - : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), - [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - [tmp0]"=&r"(tmp[0]), - [dst]"+&r"(dst), [src]"+&r"(src), - [h]"+&r"(h) - : [stride]"r"((mips_reg)stride), - [ff_pw_32]"f"(ff_pw_32), - [A]"f"(A), [E]"f"(E) - : "memory" - ); - } - } + /* x==0, y!=0 */ + E = y << 3; + A = 64 - E; + + __asm__ volatile ( + "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" + "dli %[tmp0], 0x06 \n\t" + "pshufh %[A], %[A], %[ftmp0] \n\t" + "pshufh %[E], %[E], %[ftmp0] \n\t" + "mtc1 %[tmp0], %[ftmp7] \n\t" + + "1: \n\t" + MMI_ULDC1(%[ftmp1], %[src], 0x00) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULDC1(%[ftmp2], %[src], 0x00) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULDC1(%[ftmp8], %[src], 0x00) + "addi %[h], %[h], -0x02 \n\t" + + "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" + "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" + "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" + "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" + "paddh %[ftmp3], %[ftmp3], %[ftmp5] \n\t" + "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" + "paddh %[ftmp4], %[ftmp4], %[ftmp6] \n\t" + "paddh %[ftmp3], %[ftmp3], %[ff_pw_32] \n\t" + "paddh %[ftmp4], %[ftmp4], %[ff_pw_32] \n\t" + "psrlh %[ftmp3], %[ftmp3], %[ftmp7] \n\t" + "psrlh %[ftmp4], %[ftmp4], %[ftmp7] \n\t" + "packushb %[ftmp1], %[ftmp3], %[ftmp4] \n\t" + + "punpcklbh %[ftmp3], %[ftmp2], %[ftmp0] \n\t" + "punpckhbh %[ftmp4], %[ftmp2], %[ftmp0] \n\t" + "punpcklbh %[ftmp5], %[ftmp8], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp8], %[ftmp0] \n\t" + "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" + "paddh %[ftmp3], %[ftmp3], %[ftmp5] \n\t" + "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" + "paddh %[ftmp4], %[ftmp4], %[ftmp6] \n\t" + "paddh %[ftmp3], %[ftmp3], %[ff_pw_32] \n\t" + "paddh %[ftmp4], %[ftmp4], %[ff_pw_32] \n\t" + "psrlh %[ftmp3], %[ftmp3], %[ftmp7] \n\t" + "psrlh %[ftmp4], %[ftmp4], %[ftmp7] \n\t" + "packushb %[ftmp2], %[ftmp3], %[ftmp4] \n\t" + + MMI_SDC1(%[ftmp1], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + MMI_SDC1(%[ftmp2], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + "bnez %[h], 1b \n\t" + : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), + [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), + [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), + [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), + [ftmp8]"=&f"(ftmp[8]), [tmp0]"=&r"(tmp[0]), + [dst]"+&r"(dst), [src]"+&r"(src), + [h]"+&r"(h) + : [stride]"r"((mips_reg)stride), + [ff_pw_32]"f"(ff_pw_32), + [A]"f"(A), [E]"f"(E) + : "memory" + ); } } @@ -334,231 +297,200 @@ void ff_avg_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, if(!(x || y)){ /* x=0, y=0, A=64 */ __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp4] \n\t" - "1: \n\t" - MMI_ULDC1(%[ftmp1], %[src], 0x00) + MMI_ULDC1(%[ftmp0], %[src], 0x00) PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp5], %[src], 0x00) + MMI_ULDC1(%[ftmp1], %[src], 0x00) PTR_ADDU "%[src], %[src], %[stride] \n\t" - - "punpcklbh %[ftmp2], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" - "pmullh %[ftmp1], %[ftmp2], %[A] \n\t" - "pmullh %[ftmp2], %[ftmp3], %[A] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp4] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" MMI_LDC1(%[ftmp2], %[dst], 0x00) - "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - - "punpcklbh %[ftmp2], %[ftmp5], %[ftmp0] \n\t" - "punpckhbh %[ftmp3], %[ftmp5], %[ftmp0] \n\t" - "pmullh %[ftmp1], %[ftmp2], %[A] \n\t" - "pmullh %[ftmp2], %[ftmp3], %[A] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp4] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_LDC1(%[ftmp2], %[dst], 0x00) - "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_LDC1(%[ftmp3], %[dst], 0x00) + PTR_SUBU "%[dst], %[dst], %[stride] \n\t" + "pavgb %[ftmp0], %[ftmp0], %[ftmp2] \n\t" + "pavgb %[ftmp1], %[ftmp1], %[ftmp3] \n\t" + MMI_SDC1(%[ftmp0], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" MMI_SDC1(%[ftmp1], %[dst], 0x00) PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - "addi %[h], %[h], -0x02 \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), + [dst]"+&r"(dst), [src]"+&r"(src), + [h]"+&r"(h) + : [stride]"r"((mips_reg)stride) + : "memory" + ); + } else if (x && y) { + /* x!=0, y!=0 */ + D = x * y; + B = (x << 3) - D; + C = (y << 3) - D; + A = 64 - D - B - C; + __asm__ volatile ( + "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" + "dli %[tmp0], 0x06 \n\t" + "pshufh %[A], %[A], %[ftmp0] \n\t" + "pshufh %[B], %[B], %[ftmp0] \n\t" + "mtc1 %[tmp0], %[ftmp9] \n\t" + "pshufh %[C], %[C], %[ftmp0] \n\t" + "pshufh %[D], %[D], %[ftmp0] \n\t" + + "1: \n\t" + MMI_ULDC1(%[ftmp1], %[src], 0x00) + MMI_ULDC1(%[ftmp2], %[src], 0x01) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULDC1(%[ftmp3], %[src], 0x00) + MMI_ULDC1(%[ftmp4], %[src], 0x01) + "addi %[h], %[h], -0x01 \n\t" + + "punpcklbh %[ftmp5], %[ftmp1], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp1], %[ftmp0] \n\t" + "punpcklbh %[ftmp7], %[ftmp2], %[ftmp0] \n\t" + "punpckhbh %[ftmp8], %[ftmp2], %[ftmp0] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" + "pmullh %[ftmp7], %[ftmp7], %[B] \n\t" + "paddh %[ftmp1], %[ftmp5], %[ftmp7] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[A] \n\t" + "pmullh %[ftmp8], %[ftmp8], %[B] \n\t" + "paddh %[ftmp2], %[ftmp6], %[ftmp8] \n\t" + + "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp3], %[ftmp0] \n\t" + "punpcklbh %[ftmp7], %[ftmp4], %[ftmp0] \n\t" + "punpckhbh %[ftmp8], %[ftmp4], %[ftmp0] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" + "pmullh %[ftmp7], %[ftmp7], %[D] \n\t" + "paddh %[ftmp3], %[ftmp5], %[ftmp7] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[C] \n\t" + "pmullh %[ftmp8], %[ftmp8], %[D] \n\t" + "paddh %[ftmp4], %[ftmp6], %[ftmp8] \n\t" + + "paddh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" + "paddh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" + "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" + "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" + "psrlh %[ftmp1], %[ftmp1], %[ftmp9] \n\t" + "psrlh %[ftmp2], %[ftmp2], %[ftmp9] \n\t" + "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_LDC1(%[ftmp2], %[dst], 0x00) + "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_SDC1(%[ftmp1], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + "bnez %[h], 1b \n\t" + : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), + [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), + [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), + [ftmp8]"=&f"(ftmp[8]), [ftmp9]"=&f"(ftmp[9]), [tmp0]"=&r"(tmp[0]), [dst]"+&r"(dst), [src]"+&r"(src), [h]"+&r"(h) : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), - [A]"f"(A) + [A]"f"(A), [B]"f"(B), + [C]"f"(C), [D]"f"(D) + : "memory" + ); + } else if (x) { + /* x!=0, y==0 */ + E = x << 3; + A = 64 - E; + __asm__ volatile ( + "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" + "dli %[tmp0], 0x06 \n\t" + "pshufh %[A], %[A], %[ftmp0] \n\t" + "pshufh %[E], %[E], %[ftmp0] \n\t" + "mtc1 %[tmp0], %[ftmp7] \n\t" + + "1: \n\t" + MMI_ULDC1(%[ftmp1], %[src], 0x00) + MMI_ULDC1(%[ftmp2], %[src], 0x01) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + "addi %[h], %[h], -0x01 \n\t" + + "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" + "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" + "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" + "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" + "paddh %[ftmp1], %[ftmp3], %[ftmp5] \n\t" + "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" + "paddh %[ftmp2], %[ftmp4], %[ftmp6] \n\t" + + "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" + "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" + "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" + "psrlh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" + "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_LDC1(%[ftmp2], %[dst], 0x00) + "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_SDC1(%[ftmp1], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + "bnez %[h], 1b \n\t" + : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), + [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), + [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), + [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), + [tmp0]"=&r"(tmp[0]), + [dst]"+&r"(dst), [src]"+&r"(src), + [h]"+&r"(h) + : [stride]"r"((mips_reg)stride), + [ff_pw_32]"f"(ff_pw_32), + [A]"f"(A), [E]"f"(E) : "memory" ); } else { - if(x && y) { - /* x!=0, y!=0 */ - D = x * y; - B = (x << 3) - D; - C = (y << 3) - D; - A = 64 - D - B - C; - __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "pshufh %[B], %[B], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp9] \n\t" - "pshufh %[C], %[C], %[ftmp0] \n\t" - "pshufh %[D], %[D], %[ftmp0] \n\t" - - "1: \n\t" - MMI_ULDC1(%[ftmp1], %[src], 0x00) - MMI_ULDC1(%[ftmp2], %[src], 0x01) - PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp3], %[src], 0x00) - MMI_ULDC1(%[ftmp4], %[src], 0x01) - "addi %[h], %[h], -0x01 \n\t" - - "punpcklbh %[ftmp5], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp1], %[ftmp0] \n\t" - "punpcklbh %[ftmp7], %[ftmp2], %[ftmp0] \n\t" - "punpckhbh %[ftmp8], %[ftmp2], %[ftmp0] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" - "pmullh %[ftmp7], %[ftmp7], %[B] \n\t" - "paddh %[ftmp1], %[ftmp5], %[ftmp7] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[A] \n\t" - "pmullh %[ftmp8], %[ftmp8], %[B] \n\t" - "paddh %[ftmp2], %[ftmp6], %[ftmp8] \n\t" - - "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp3], %[ftmp0] \n\t" - "punpcklbh %[ftmp7], %[ftmp4], %[ftmp0] \n\t" - "punpckhbh %[ftmp8], %[ftmp4], %[ftmp0] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" - "pmullh %[ftmp7], %[ftmp7], %[D] \n\t" - "paddh %[ftmp3], %[ftmp5], %[ftmp7] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[C] \n\t" - "pmullh %[ftmp8], %[ftmp8], %[D] \n\t" - "paddh %[ftmp4], %[ftmp6], %[ftmp8] \n\t" - - "paddh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ftmp4] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp9] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp9] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_LDC1(%[ftmp2], %[dst], 0x00) - "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - "bnez %[h], 1b \n\t" - : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), - [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - [ftmp8]"=&f"(ftmp[8]), [ftmp9]"=&f"(ftmp[9]), - [tmp0]"=&r"(tmp[0]), - [dst]"+&r"(dst), [src]"+&r"(src), - [h]"+&r"(h) - : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), - [A]"f"(A), [B]"f"(B), - [C]"f"(C), [D]"f"(D) - : "memory" - ); - } else { - if(x) { - /* x!=0, y==0 */ - E = x << 3; - A = 64 - E; - __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "pshufh %[E], %[E], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp7] \n\t" - - "1: \n\t" - MMI_ULDC1(%[ftmp1], %[src], 0x00) - MMI_ULDC1(%[ftmp2], %[src], 0x01) - PTR_ADDU "%[src], %[src], %[stride] \n\t" - "addi %[h], %[h], -0x01 \n\t" - - "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" - "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" - "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" - "paddh %[ftmp1], %[ftmp3], %[ftmp5] \n\t" - "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" - "paddh %[ftmp2], %[ftmp4], %[ftmp6] \n\t" - - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_LDC1(%[ftmp2], %[dst], 0x00) - "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - "bnez %[h], 1b \n\t" - : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), - [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - [tmp0]"=&r"(tmp[0]), - [dst]"+&r"(dst), [src]"+&r"(src), - [h]"+&r"(h) - : [stride]"r"((mips_reg)stride), - [ff_pw_32]"f"(ff_pw_32), - [A]"f"(A), [E]"f"(E) - : "memory" - ); - } else { - /* x==0, y!=0 */ - E = y << 3; - A = 64 - E; - __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "pshufh %[E], %[E], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp7] \n\t" - - "1: \n\t" - MMI_ULDC1(%[ftmp1], %[src], 0x00) - PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_ULDC1(%[ftmp2], %[src], 0x00) - "addi %[h], %[h], -0x01 \n\t" - - "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" - "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" - "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" - "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" - "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" - "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" - "paddh %[ftmp1], %[ftmp3], %[ftmp5] \n\t" - "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" - "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" - "paddh %[ftmp2], %[ftmp4], %[ftmp6] \n\t" - - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" - "psrlh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_LDC1(%[ftmp2], %[dst], 0x00) - "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - MMI_SDC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - "bnez %[h], 1b \n\t" - : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), - [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - [tmp0]"=&r"(tmp[0]), - [dst]"+&r"(dst), [src]"+&r"(src), - [h]"+&r"(h) - : [stride]"r"((mips_reg)stride), - [ff_pw_32]"f"(ff_pw_32), - [A]"f"(A), [E]"f"(E) - : "memory" - ); - } - } + /* x==0, y!=0 */ + E = y << 3; + A = 64 - E; + __asm__ volatile ( + "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" + "dli %[tmp0], 0x06 \n\t" + "pshufh %[A], %[A], %[ftmp0] \n\t" + "pshufh %[E], %[E], %[ftmp0] \n\t" + "mtc1 %[tmp0], %[ftmp7] \n\t" + + "1: \n\t" + MMI_ULDC1(%[ftmp1], %[src], 0x00) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULDC1(%[ftmp2], %[src], 0x00) + "addi %[h], %[h], -0x01 \n\t" + + "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" + "punpckhbh %[ftmp4], %[ftmp1], %[ftmp0] \n\t" + "punpcklbh %[ftmp5], %[ftmp2], %[ftmp0] \n\t" + "punpckhbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" + "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" + "pmullh %[ftmp5], %[ftmp5], %[E] \n\t" + "paddh %[ftmp1], %[ftmp3], %[ftmp5] \n\t" + "pmullh %[ftmp4], %[ftmp4], %[A] \n\t" + "pmullh %[ftmp6], %[ftmp6], %[E] \n\t" + "paddh %[ftmp2], %[ftmp4], %[ftmp6] \n\t" + + "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" + "paddh %[ftmp2], %[ftmp2], %[ff_pw_32] \n\t" + "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" + "psrlh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" + "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_LDC1(%[ftmp2], %[dst], 0x00) + "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_SDC1(%[ftmp1], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + "bnez %[h], 1b \n\t" + : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), + [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), + [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), + [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), + [tmp0]"=&r"(tmp[0]), + [dst]"+&r"(dst), [src]"+&r"(src), + [h]"+&r"(h) + : [stride]"r"((mips_reg)stride), + [ff_pw_32]"f"(ff_pw_32), + [A]"f"(A), [E]"f"(E) + : "memory" + ); } } @@ -567,8 +499,8 @@ void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, { const int A = (8 - x) * (8 - y); const int B = x * (8 - y); - const int C = (8 - x) * y; - const int D = x * y; + const int C = (8 - x) * y; + const int D = x * y; const int E = B + C; double ftmp[8]; uint64_t tmp[1]; @@ -586,31 +518,29 @@ void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, "pshufh %[D], %[D], %[ftmp0] \n\t" "1: \n\t" - PTR_ADDU "%[addr0], %[src], %[stride] \n\t" MMI_ULWC1(%[ftmp1], %[src], 0x00) MMI_ULWC1(%[ftmp2], %[src], 0x01) - MMI_ULWC1(%[ftmp3], %[addr0], 0x00) - MMI_ULWC1(%[ftmp4], %[addr0], 0x01) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULWC1(%[ftmp3], %[src], 0x00) + MMI_ULWC1(%[ftmp4], %[src], 0x01) "punpcklbh %[ftmp5], %[ftmp1], %[ftmp0] \n\t" "punpcklbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" "pmullh %[ftmp6], %[ftmp6], %[B] \n\t" "paddh %[ftmp1], %[ftmp5], %[ftmp6] \n\t" - "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" "punpcklbh %[ftmp6], %[ftmp4], %[ftmp0] \n\t" "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" "pmullh %[ftmp6], %[ftmp6], %[D] \n\t" "paddh %[ftmp2], %[ftmp5], %[ftmp6] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ftmp2] \n\t" "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" + "addi %[h], %[h], -0x01 \n\t" MMI_SWC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[src], %[src], %[stride] \n\t" PTR_ADDU "%[dst], %[dst], %[stride] \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), @@ -619,7 +549,6 @@ void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), [tmp0]"=&r"(tmp[0]), RESTRICT_ASM_LOW32 - [addr0]"=&r"(addr[0]), [dst]"+&r"(dst), [src]"+&r"(src), [h]"+&r"(h) : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), @@ -629,7 +558,6 @@ void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, ); } else if (E) { const int step = C ? stride : 1; - __asm__ volatile ( "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" "dli %[tmp0], 0x06 \n\t" @@ -638,22 +566,20 @@ void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, "mtc1 %[tmp0], %[ftmp5] \n\t" "1: \n\t" - PTR_ADDU "%[addr0], %[src], %[step] \n\t" MMI_ULWC1(%[ftmp1], %[src], 0x00) + PTR_ADDU "%[addr0], %[src], %[step] \n\t" MMI_ULWC1(%[ftmp2], %[addr0], 0x00) - + PTR_ADDU "%[src], %[src], %[stride] \n\t" + "addi %[h], %[h], -0x01 \n\t" "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" "punpcklbh %[ftmp4], %[ftmp2], %[ftmp0] \n\t" "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" "pmullh %[ftmp4], %[ftmp4], %[E] \n\t" "paddh %[ftmp1], %[ftmp3], %[ftmp4] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" "psrlh %[ftmp1], %[ftmp1], %[ftmp5] \n\t" "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" - "addi %[h], %[h], -0x01 \n\t" MMI_SWC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[src], %[src], %[stride] \n\t" PTR_ADDU "%[dst], %[dst], %[stride] \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), @@ -671,42 +597,22 @@ void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, ); } else { __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp3] \n\t" - "1: \n\t" - MMI_ULWC1(%[ftmp1], %[src], 0x00) - "punpcklbh %[ftmp2], %[ftmp1], %[ftmp0] \n\t" - "pmullh %[ftmp1], %[ftmp2], %[A] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" + MMI_ULWC1(%[ftmp0], %[src], 0x00) PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_SWC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - MMI_ULWC1(%[ftmp1], %[src], 0x00) - "punpcklbh %[ftmp2], %[ftmp1], %[ftmp0] \n\t" - "pmullh %[ftmp1], %[ftmp2], %[A] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" + PTR_ADDU "%[src], %[src], %[stride] \n\t" "addi %[h], %[h], -0x02 \n\t" + MMI_SWC1(%[ftmp0], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" MMI_SWC1(%[ftmp1], %[dst], 0x00) - - PTR_ADDU "%[src], %[src], %[stride] \n\t" PTR_ADDU "%[dst], %[dst], %[stride] \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [tmp0]"=&r"(tmp[0]), - RESTRICT_ASM_LOW32 [dst]"+&r"(dst), [src]"+&r"(src), + RESTRICT_ASM_LOW32 [h]"+&r"(h) - : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), - [A]"f"(A) + : [stride]"r"((mips_reg)stride) : "memory" ); } @@ -736,33 +642,31 @@ void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, "pshufh %[D], %[D], %[ftmp0] \n\t" "1: \n\t" - PTR_ADDU "%[addr0], %[src], %[stride] \n\t" MMI_ULWC1(%[ftmp1], %[src], 0x00) MMI_ULWC1(%[ftmp2], %[src], 0x01) - MMI_ULWC1(%[ftmp3], %[addr0], 0x00) - MMI_ULWC1(%[ftmp4], %[addr0], 0x01) + PTR_ADDU "%[src], %[src], %[stride] \n\t" + MMI_ULWC1(%[ftmp3], %[src], 0x00) + MMI_ULWC1(%[ftmp4], %[src], 0x01) "punpcklbh %[ftmp5], %[ftmp1], %[ftmp0] \n\t" "punpcklbh %[ftmp6], %[ftmp2], %[ftmp0] \n\t" "pmullh %[ftmp5], %[ftmp5], %[A] \n\t" "pmullh %[ftmp6], %[ftmp6], %[B] \n\t" "paddh %[ftmp1], %[ftmp5], %[ftmp6] \n\t" - "punpcklbh %[ftmp5], %[ftmp3], %[ftmp0] \n\t" "punpcklbh %[ftmp6], %[ftmp4], %[ftmp0] \n\t" "pmullh %[ftmp5], %[ftmp5], %[C] \n\t" "pmullh %[ftmp6], %[ftmp6], %[D] \n\t" "paddh %[ftmp2], %[ftmp5], %[ftmp6] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ftmp2] \n\t" "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" "psrlh %[ftmp1], %[ftmp1], %[ftmp7] \n\t" "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" MMI_LWC1(%[ftmp2], %[dst], 0x00) "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + "addi %[h], %[h], -0x01 \n\t" MMI_SWC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[src], %[src], %[stride] \n\t" PTR_ADDU "%[dst], %[dst], %[stride] \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), @@ -771,7 +675,6 @@ void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), [tmp0]"=&r"(tmp[0]), RESTRICT_ASM_LOW32 - [addr0]"=&r"(addr[0]), [dst]"+&r"(dst), [src]"+&r"(src), [h]"+&r"(h) : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), @@ -781,32 +684,30 @@ void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, ); } else if (E) { const int step = C ? stride : 1; - __asm__ volatile ( "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" "dli %[tmp0], 0x06 \n\t" "pshufh %[A], %[A], %[ftmp0] \n\t" "pshufh %[E], %[E], %[ftmp0] \n\t" "mtc1 %[tmp0], %[ftmp5] \n\t" + "1: \n\t" - PTR_ADDU "%[addr0], %[src], %[step] \n\t" MMI_ULWC1(%[ftmp1], %[src], 0x00) + PTR_ADDU "%[addr0], %[src], %[step] \n\t" MMI_ULWC1(%[ftmp2], %[addr0], 0x00) - + PTR_ADDU "%[src], %[src], %[stride] \n\t" + "addi %[h], %[h], -0x01 \n\t" "punpcklbh %[ftmp3], %[ftmp1], %[ftmp0] \n\t" "punpcklbh %[ftmp4], %[ftmp2], %[ftmp0] \n\t" "pmullh %[ftmp3], %[ftmp3], %[A] \n\t" "pmullh %[ftmp4], %[ftmp4], %[E] \n\t" "paddh %[ftmp1], %[ftmp3], %[ftmp4] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" "psrlh %[ftmp1], %[ftmp1], %[ftmp5] \n\t" "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" MMI_LWC1(%[ftmp2], %[dst], 0x00) "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - "addi %[h], %[h], -0x01 \n\t" MMI_SWC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[src], %[src], %[stride] \n\t" PTR_ADDU "%[dst], %[dst], %[stride] \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), @@ -824,46 +725,27 @@ void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, ); } else { __asm__ volatile ( - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "dli %[tmp0], 0x06 \n\t" - "pshufh %[A], %[A], %[ftmp0] \n\t" - "mtc1 %[tmp0], %[ftmp3] \n\t" - "1: \n\t" - MMI_ULWC1(%[ftmp1], %[src], 0x00) - "punpcklbh %[ftmp2], %[ftmp1], %[ftmp0] \n\t" - "pmullh %[ftmp1], %[ftmp2], %[A] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" - MMI_LWC1(%[ftmp2], %[dst], 0x00) - "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + MMI_ULWC1(%[ftmp0], %[src], 0x00) PTR_ADDU "%[src], %[src], %[stride] \n\t" - MMI_SWC1(%[ftmp1], %[dst], 0x00) - PTR_ADDU "%[dst], %[dst], %[stride] \n\t" - MMI_ULWC1(%[ftmp1], %[src], 0x00) - "punpcklbh %[ftmp2], %[ftmp1], %[ftmp0] \n\t" - "pmullh %[ftmp1], %[ftmp2], %[A] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ff_pw_32] \n\t" - "psrlh %[ftmp1], %[ftmp1], %[ftmp3] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" - MMI_LWC1(%[ftmp2], %[dst], 0x00) - "pavgb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + PTR_ADDU "%[src], %[src], %[stride] \n\t" "addi %[h], %[h], -0x02 \n\t" + MMI_LWC1(%[ftmp2], %[dst], 0x00) + "pavgb %[ftmp0], %[ftmp0], %[ftmp2] \n\t" + MMI_SWC1(%[ftmp0], %[dst], 0x00) + PTR_ADDU "%[dst], %[dst], %[stride] \n\t" + MMI_LWC1(%[ftmp3], %[dst], 0x00) + "pavgb %[ftmp1], %[ftmp1], %[ftmp3] \n\t" MMI_SWC1(%[ftmp1], %[dst], 0x00) - - PTR_ADDU "%[src], %[src], %[stride] \n\t" PTR_ADDU "%[dst], %[dst], %[stride] \n\t" "bnez %[h], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [tmp0]"=&r"(tmp[0]), - RESTRICT_ASM_LOW32 [dst]"+&r"(dst), [src]"+&r"(src), + RESTRICT_ASM_LOW32 [h]"+&r"(h) - : [stride]"r"((mips_reg)stride),[ff_pw_32]"f"(ff_pw_32), - [A]"f"(A) + : [stride]"r"((mips_reg)stride) : "memory" ); }

1 0

configure: enable mipsfpu for loongson platform.
by Shiyou Yin 01 Dec '18

01 Dec '18

ffmpeg | branch: master | Shiyou Yin <yinshiyou-hf(a)loongson.cn> | Thu Nov 29 17:39:41 2018 +0800| [5c806d5b54a7a261569b4d3c2053a321dcf4445f] | committer: Michael Niedermayer configure: enable mipsfpu for loongson platform. mipsfpu supported by loongson 3a2000,3a3000,3a4000,2k1000, Fate tests passed. Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=5c806d5b54a7a2615… --- configure | 1 - 1 file changed, 1 deletion(-) diff --git a/configure b/configure index 3e9222ed1b..54b7e1119b 100755 --- a/configure +++ b/configure @@ -4848,7 +4848,6 @@ elif enabled mips; then enable fast_cmov enable fast_unaligned disable aligned_stack - disable mipsfpu disable mipsdsp disable mipsdspr2 # When gcc version less than 5.3.0, add -fno-expensive-optimizations flag.

1 0