October 2019 - ffmpeg-cvslog - lists.ffmpeg.org

avutil/common: Fix underflow for ROUNDED_DIV with unsigned integer
by Mengye Lv 06 Oct '19

06 Oct '19

ffmpeg | branch: master | Mengye Lv <mengyelv(a)tencent.com> | Wed Oct 2 10:30:45 2019 +0800| [9f353e376b161eb16a78fd7ad727c9513ac60d13] | committer: Jun Zhao avutil/common: Fix underflow for ROUNDED_DIV with unsigned integer When used ROUNDED_DIV(a,b), if a is unsigned integer zero, it's will lead to an underflow issue(it called unsigned integer wrapping). Fixes #8062 Reviewed-by: Michael Niedermayer <michael(a)niedermayer.cc> Signed-off-by: Mengye Lv <mengyelv(a)tencent.com> Signed-off-by: Jun Zhao <barryjzhao(a)tencent.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=9f353e376b161eb16… --- libavutil/common.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavutil/common.h b/libavutil/common.h index af35397eb9..f09f0b486b 100644 --- a/libavutil/common.h +++ b/libavutil/common.h @@ -53,7 +53,7 @@ //rounded division & shift #define RSHIFT(a,b) ((a) > 0 ? ((a) + ((1<<(b))>>1))>>(b) : ((a) + ((1<<(b))>>1)-1)>>(b)) /* assume b>0 */ -#define ROUNDED_DIV(a,b) (((a)>0 ? (a) + ((b)>>1) : (a) - ((b)>>1))/(b)) +#define ROUNDED_DIV(a,b) (((a)>=0 ? (a) + ((b)>>1) : (a) - ((b)>>1))/(b)) /* Fast a/(1<<b) rounded toward +inf. Assume a>=0 and b>=0 */ #define AV_CEIL_RSHIFT(a,b) (!av_builtin_constant_p(b) ? -((-(a)) >> (b)) \ : ((a) + (1<<(b)) - 1) >> (b))

1 0

avformat/matroskadec: don't rescale mastering display values
by James Almer 06 Oct '19

06 Oct '19

ffmpeg | branch: master | James Almer <jamrial(a)gmail.com> | Thu Oct 3 20:52:20 2019 -0300| [3b4e9a31ea263830d6dfbc402c02e3c8d017094f] | committer: James Almer avformat/matroskadec: don't rescale mastering display values Simplifies code. Signed-off-by: James Almer <jamrial(a)gmail.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=3b4e9a31ea263830d… --- libavformat/matroskadec.c | 33 ++++++++++----------------------- 1 file changed, 10 insertions(+), 23 deletions(-) diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c index a5f120b54d..c224c3447a 100644 --- a/libavformat/matroskadec.c +++ b/libavformat/matroskadec.c @@ -2116,9 +2116,6 @@ static int mkv_parse_video_color(AVStream *st, const MatroskaTrack *track) { } if (has_mastering_primaries || has_mastering_luminance) { - // Use similar rationals as other standards. - const int chroma_den = 50000; - const int luma_den = 10000; AVMasteringDisplayMetadata *metadata = (AVMasteringDisplayMetadata*) av_stream_new_side_data( st, AV_PKT_DATA_MASTERING_DISPLAY_METADATA, @@ -2128,29 +2125,19 @@ static int mkv_parse_video_color(AVStream *st, const MatroskaTrack *track) { } memset(metadata, 0, sizeof(AVMasteringDisplayMetadata)); if (has_mastering_primaries) { - metadata->display_primaries[0][0] = av_make_q( - round(mastering_meta->r_x * chroma_den), chroma_den); - metadata->display_primaries[0][1] = av_make_q( - round(mastering_meta->r_y * chroma_den), chroma_den); - metadata->display_primaries[1][0] = av_make_q( - round(mastering_meta->g_x * chroma_den), chroma_den); - metadata->display_primaries[1][1] = av_make_q( - round(mastering_meta->g_y * chroma_den), chroma_den); - metadata->display_primaries[2][0] = av_make_q( - round(mastering_meta->b_x * chroma_den), chroma_den); - metadata->display_primaries[2][1] = av_make_q( - round(mastering_meta->b_y * chroma_den), chroma_den); - metadata->white_point[0] = av_make_q( - round(mastering_meta->white_x * chroma_den), chroma_den); - metadata->white_point[1] = av_make_q( - round(mastering_meta->white_y * chroma_den), chroma_den); + metadata->display_primaries[0][0] = av_d2q(mastering_meta->r_x, INT_MAX); + metadata->display_primaries[0][1] = av_d2q(mastering_meta->r_y, INT_MAX); + metadata->display_primaries[1][0] = av_d2q(mastering_meta->g_x, INT_MAX); + metadata->display_primaries[1][1] = av_d2q(mastering_meta->g_y, INT_MAX); + metadata->display_primaries[2][0] = av_d2q(mastering_meta->b_x, INT_MAX); + metadata->display_primaries[2][1] = av_d2q(mastering_meta->b_y, INT_MAX); + metadata->white_point[0] = av_d2q(mastering_meta->white_x, INT_MAX); + metadata->white_point[1] = av_d2q(mastering_meta->white_y, INT_MAX); metadata->has_primaries = 1; } if (has_mastering_luminance) { - metadata->max_luminance = av_make_q( - round(mastering_meta->max_luminance * luma_den), luma_den); - metadata->min_luminance = av_make_q( - round(mastering_meta->min_luminance * luma_den), luma_den); + metadata->max_luminance = av_d2q(mastering_meta->max_luminance, INT_MAX); + metadata->min_luminance = av_d2q(mastering_meta->min_luminance, INT_MAX); metadata->has_luminance = 1; } }

1 0

avformat/mpeg: better fix for MLP versus PCM-DVD misdetection
by Paul B Mahol 05 Oct '19

05 Oct '19

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Tue Oct 1 11:34:09 2019 +0200| [7da57875b57012478aed546818b2d52985d13793] | committer: Paul B Mahol avformat/mpeg: better fix for MLP versus PCM-DVD misdetection > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=7da57875b57012478… --- libavformat/mpeg.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/libavformat/mpeg.c b/libavformat/mpeg.c index 3205f209e6..6f132aae05 100644 --- a/libavformat/mpeg.c +++ b/libavformat/mpeg.c @@ -490,6 +490,7 @@ static int mpegps_read_packet(AVFormatContext *s, AVStream *st; int len, startcode, i, es_type, ret; int lpcm_header_len = -1; //Init to suppress warning + int pcm_dvd = 0; int request_probe= 0; enum AVCodecID codec_id = AV_CODEC_ID_NONE; enum AVMediaType type; @@ -513,6 +514,12 @@ redo: /* MLP/TrueHD audio has a 4-byte header */ avio_r8(s->pb); len--; + } else if (startcode >= 0xa0 && startcode <= 0xaf) { + ret = ffio_ensure_seekback(s->pb, 3); + if (ret < 0) + return ret; + pcm_dvd = (avio_rb24(s->pb) & 0xFF) == 0x80; + avio_skip(s->pb, -3); } } } @@ -591,7 +598,7 @@ redo: codec_id = AV_CODEC_ID_DTS; } else if (startcode >= 0xa0 && startcode <= 0xaf) { type = AVMEDIA_TYPE_AUDIO; - if (lpcm_header_len >= 6 && startcode == 0xa1) { + if (!pcm_dvd) { codec_id = AV_CODEC_ID_MLP; } else { codec_id = AV_CODEC_ID_PCM_DVD;

1 0

[ffmpeg-web] branch master updated. a5b4e10 index: typo on url
by ffmpeg-git＠ffmpeg.org 05 Oct '19

05 Oct '19

The branch, master has been updated via a5b4e10bcb0341b80cdb7758fdfc7de5cd3120f0 (commit) from ea79bbe1ab7edb2ff4a3c0c6d87fb435e4f34013 (commit) - Log ----------------------------------------------------------------- commit a5b4e10bcb0341b80cdb7758fdfc7de5cd3120f0 Author: compn <compn(a)ffmpeg.org> AuthorDate: Sat Oct 5 00:30:43 2019 -0700 Commit: compn <compn(a)ffmpeg.org> CommitDate: Sat Oct 5 00:30:43 2019 -0700 index: typo on url diff --git a/src/index b/src/index index a72f01a..59e79d5 100644 --- a/src/index +++ b/src/index @@ -57,7 +57,7 @@ work better on, please report them to us on our <a href="http://trac.ffmpeg.org">trac</a>. </p> <p> - <a href="="http://ffmpeg.org/~compn/output20p8.mp4">See for yourself</a>. + <a href="http://ffmpeg.org/~compn/output20p8.mp4">See for yourself</a>. Example was made with -vf photosensitivity=20:0.8 </p> <p> ----------------------------------------------------------------------- Summary of changes: src/index | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) hooks/post-receive --

1 0

[ffmpeg-web] branch master updated. ea79bbe index: remove video tag stretches the page too much
by ffmpeg-git＠ffmpeg.org 05 Oct '19

05 Oct '19

The branch, master has been updated via ea79bbe1ab7edb2ff4a3c0c6d87fb435e4f34013 (commit) from a696545f331107bb83208789686b6d1b642af938 (commit) - Log ----------------------------------------------------------------- commit ea79bbe1ab7edb2ff4a3c0c6d87fb435e4f34013 Author: compn <compn(a)ffmpeg.org> AuthorDate: Sat Oct 5 00:27:54 2019 -0700 Commit: compn <compn(a)ffmpeg.org> CommitDate: Sat Oct 5 00:27:54 2019 -0700 index: remove video tag stretches the page too much diff --git a/src/index b/src/index index a29802a..a72f01a 100644 --- a/src/index +++ b/src/index @@ -57,12 +57,8 @@ work better on, please report them to us on our <a href="http://trac.ffmpeg.org">trac</a>. </p> <p> - See for yourself with your eyes. -vf photosensitivity=20:0.8 - </p> - <p> - <video> - <source src="http://ffmpeg.org/~compn/output20p8.mp4" type="video/mp4"> - </video> + <a href="="http://ffmpeg.org/~compn/output20p8.mp4">See for yourself</a>. + Example was made with -vf photosensitivity=20:0.8 </p> <p> We are not professionals. Please use this in your medical studies to ----------------------------------------------------------------------- Summary of changes: src/index | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) hooks/post-receive --

1 0

[ffmpeg-web] branch master updated. a696545 index: news entry for photosensitivity filter
by ffmpeg-git＠ffmpeg.org 05 Oct '19

05 Oct '19

The branch, master has been updated via a696545f331107bb83208789686b6d1b642af938 (commit) from 0d1e4ae63c9bbcebacdff37383325083f4798f9f (commit) - Log ----------------------------------------------------------------- commit a696545f331107bb83208789686b6d1b642af938 Author: compn <compn(a)ffmpeg.org> AuthorDate: Sat Oct 5 00:20:02 2019 -0700 Commit: compn <compn(a)ffmpeg.org> CommitDate: Sat Oct 5 00:20:02 2019 -0700 index: news entry for photosensitivity filter diff --git a/src/index b/src/index index 8b92a72..a29802a 100644 --- a/src/index +++ b/src/index @@ -37,6 +37,42 @@ News </h1> + <h3 id="photosensitivity">October 5th, 2019, Bright Lights</h3> + <p> + FFmpeg has added a realtime bright flash removal filter to libavfilter. + </p> + <p> + Note that this filter is not FDA approved, nor are we medical professionals. + Nor has this filter been tested with anyone who has photosensitive epilepsy. + FFmpeg and its photosensitivity filter are not making any medical claims. + </p> + <p> + That said, this is a new video filter that may help photosensitive people + watch tv, play video games or even be used with a VR headset to block + out epiletic triggers such as filtered sunlight when they are outside. + Or you could use it against those annoying white flashes on your tv screen. + The filter fails on some input, such as the + <a href="https://www.youtube.com/watch?v=8L_9hXnUzRk">Incredibles 2 Screen Slaver</a> + scene. It is not perfect. If you have other clips that you want this filter to + work better on, please report them to us on our <a href="http://trac.ffmpeg.org">trac</a>. + </p> + <p> + See for yourself with your eyes. -vf photosensitivity=20:0.8 + </p> + <p> + <video> + <source src="http://ffmpeg.org/~compn/output20p8.mp4" type="video/mp4"> + </video> + </p> + <p> + We are not professionals. Please use this in your medical studies to + advance epilepsy research. If you decide to use this in a medical + setting, or make a hardware hdmi input output realtime tv filter, + or find another use for this, <a href="mailto:compn@ffmpeg.org">please let me know</a>. + This filter was a feature request of mine + <a href="https://trac.ffmpeg.org/ticket/2104">since 2013</a>. + </p> + <h3 id="pr4.2">August 5th, 2019, FFmpeg 4.2 "Ada"</h3> <p> <a href="download.html#release_4.2">FFmpeg 4.2 "Ada"</a>, a new ----------------------------------------------------------------------- Summary of changes: src/index | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) hooks/post-receive --

1 0

FATE/dnn: fix stack buffer overflow
by Zhao Zhili 04 Oct '19

04 Oct '19

ffmpeg | branch: master | Zhao Zhili <zhilizhao(a)tencent.com> | Tue Oct 1 14:36:49 2019 +0800| [7c145b64417fbe674f90785d2e5a5afea1a1b821] | committer: Pedro Arthur FATE/dnn: fix stack buffer overflow Signed-off-by: Pedro Arthur <bygrandao(a)gmail.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=7c145b64417fbe674… --- tests/dnn/dnn-layer-pad-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/dnn/dnn-layer-pad-test.c b/tests/dnn/dnn-layer-pad-test.c index 1fb2be1590..ea8c824d1e 100644 --- a/tests/dnn/dnn-layer-pad-test.c +++ b/tests/dnn/dnn-layer-pad-test.c @@ -203,7 +203,7 @@ static int test_with_mode_constant(void) params.paddings[3][1] = 2; operands[0].data = input; - operands[0].dims[0] = 3; + operands[0].dims[0] = 1; operands[0].dims[1] = 2; operands[0].dims[2] = 2; operands[0].dims[3] = 3;

1 0

swscale: Fix AltiVec/VSX build with recent GCC
by Daniel Kolesa 04 Oct '19

04 Oct '19

ffmpeg | branch: master | Daniel Kolesa <daniel(a)octaforge.org> | Wed Aug 7 19:39:07 2019 +0200| [e6625ca41f85ed10b3108d37bd897f0ae063250e] | committer: Lauri Kasanen swscale: Fix AltiVec/VSX build with recent GCC The argument to vec_splat_u16 must be a literal. By making the function always inline and marking the arguments const, gcc can turn those into literals, and avoid build errors like: swscale_vsx.c:165:53: error: argument 1 must be a 5-bit signed literal Fixes #7861. Signed-off-by: Daniel Kolesa <daniel(a)octaforge.org> Signed-off-by: Lauri Kasanen <cand(a)gmx.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=e6625ca41f85ed10b… --- libswscale/ppc/swscale_vsx.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/libswscale/ppc/swscale_vsx.c b/libswscale/ppc/swscale_vsx.c index e6a35d3f78..af8b0e1fa3 100644 --- a/libswscale/ppc/swscale_vsx.c +++ b/libswscale/ppc/swscale_vsx.c @@ -154,8 +154,10 @@ static void yuv2plane1_nbps_u(const int16_t *src, uint16_t *dest, int dstW, } } -static void yuv2plane1_nbps_vsx(const int16_t *src, uint16_t *dest, int dstW, - int big_endian, int output_bits) +static av_always_inline void yuv2plane1_nbps_vsx(const int16_t *src, + uint16_t *dest, int dstW, + const int big_endian, + const int output_bits) { const int dst_u = -(uintptr_t)dest & 7; const int shift = 15 - output_bits; @@ -273,8 +275,10 @@ static void yuv2plane1_16_u(const int32_t *src, uint16_t *dest, int dstW, } } -static void yuv2plane1_16_vsx(const int32_t *src, uint16_t *dest, int dstW, - int big_endian, int output_bits) +static av_always_inline void yuv2plane1_16_vsx(const int32_t *src, + uint16_t *dest, int dstW, + const int big_endian, + int output_bits) { const int dst_u = -(uintptr_t)dest & 7; const int shift = 3;

1 0

swscale: Replace illegal vector keyword usage in altivec code
by Daniel Kolesa 04 Oct '19

04 Oct '19

ffmpeg | branch: master | Daniel Kolesa <daniel(a)octaforge.org> | Wed Aug 7 19:39:06 2019 +0200| [1bdb47b73484dd21ee1bf493d79ac26384eda05e] | committer: Lauri Kasanen swscale: Replace illegal vector keyword usage in altivec code While this technically compiles in current ffmpeg, this is only because ffmpeg is compiled in strict ISO C mode, which disables the builtin 'vector' keyword for AltiVec/VSX. Instead this gets replaced with a macro inside altivec.h, which defines vector to be actually __vector, which accepts random types. Normally, the vector keyword should be used only with plain scalar non-typedef types, such as unsigned int. But we have the vec_(s|u)(8|16|32) macros, which can be used in a portable manner, in util_altivec.h in libavutil. This is also consistent with other AltiVec/VSX code elsewhere in the tree. Fixes #7861. Signed-off-by: Daniel Kolesa <daniel(a)octaforge.org> Signed-off-by: Lauri Kasanen <cand(a)gmx.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=1bdb47b73484dd21e… --- libswscale/ppc/swscale_altivec.c | 36 +- libswscale/ppc/swscale_vsx.c | 882 +++++++++++++++++++-------------------- 2 files changed, 459 insertions(+), 459 deletions(-) diff --git a/libswscale/ppc/swscale_altivec.c b/libswscale/ppc/swscale_altivec.c index 6b8cc2c194..1630355f51 100644 --- a/libswscale/ppc/swscale_altivec.c +++ b/libswscale/ppc/swscale_altivec.c @@ -153,13 +153,13 @@ static void yuv2plane1_float_altivec(const int32_t *src, float *dest, int dstW) const int add = (1 << (shift - 1)); const int clip = (1 << 16) - 1; const float fmult = 1.0f / 65535.0f; - const vector uint32_t vadd = (vector uint32_t) {add, add, add, add}; - const vector uint32_t vshift = (vector uint32_t) vec_splat_u32(shift); - const vector uint32_t vlargest = (vector uint32_t) {clip, clip, clip, clip}; - const vector float vmul = (vector float) {fmult, fmult, fmult, fmult}; - const vector float vzero = (vector float) {0, 0, 0, 0}; - vector uint32_t v; - vector float vd; + const vec_u32 vadd = (vec_u32) {add, add, add, add}; + const vec_u32 vshift = (vec_u32) vec_splat_u32(shift); + const vec_u32 vlargest = (vec_u32) {clip, clip, clip, clip}; + const vec_f vmul = (vec_f) {fmult, fmult, fmult, fmult}; + const vec_f vzero = (vec_f) {0, 0, 0, 0}; + vec_u32 v; + vec_f vd; int i; yuv2plane1_float_u(src, dest, dst_u, 0); @@ -186,15 +186,15 @@ static void yuv2plane1_float_bswap_altivec(const int32_t *src, uint32_t *dest, i const int add = (1 << (shift - 1)); const int clip = (1 << 16) - 1; const float fmult = 1.0f / 65535.0f; - const vector uint32_t vadd = (vector uint32_t) {add, add, add, add}; - const vector uint32_t vshift = (vector uint32_t) vec_splat_u32(shift); - const vector uint32_t vlargest = (vector uint32_t) {clip, clip, clip, clip}; - const vector float vmul = (vector float) {fmult, fmult, fmult, fmult}; - const vector float vzero = (vector float) {0, 0, 0, 0}; - const vector uint32_t vswapbig = (vector uint32_t) {16, 16, 16, 16}; - const vector uint16_t vswapsmall = vec_splat_u16(8); - vector uint32_t v; - vector float vd; + const vec_u32 vadd = (vec_u32) {add, add, add, add}; + const vec_u32 vshift = (vec_u32) vec_splat_u32(shift); + const vec_u32 vlargest = (vec_u32) {clip, clip, clip, clip}; + const vec_f vmul = (vec_f) {fmult, fmult, fmult, fmult}; + const vec_f vzero = (vec_f) {0, 0, 0, 0}; + const vec_u32 vswapbig = (vec_u32) {16, 16, 16, 16}; + const vec_u16 vswapsmall = vec_splat_u16(8); + vec_u32 v; + vec_f vd; int i; yuv2plane1_float_bswap_u(src, dest, dst_u, 0); @@ -208,8 +208,8 @@ static void yuv2plane1_float_bswap_altivec(const int32_t *src, uint32_t *dest, i vd = vec_ctf(v, 0); vd = vec_madd(vd, vmul, vzero); - vd = (vector float) vec_rl((vector uint32_t) vd, vswapbig); - vd = (vector float) vec_rl((vector uint16_t) vd, vswapsmall); + vd = (vec_f) vec_rl((vec_u32) vd, vswapbig); + vd = (vec_f) vec_rl((vec_u16) vd, vswapsmall); vec_st(vd, 0, (float *) &dest[i]); } diff --git a/libswscale/ppc/swscale_vsx.c b/libswscale/ppc/swscale_vsx.c index 75dee5ea58..e6a35d3f78 100644 --- a/libswscale/ppc/swscale_vsx.c +++ b/libswscale/ppc/swscale_vsx.c @@ -103,9 +103,9 @@ static void yuv2plane1_8_vsx(const int16_t *src, uint8_t *dest, int dstW, const int dst_u = -(uintptr_t)dest & 15; int i, j; LOCAL_ALIGNED(16, int16_t, val, [16]); - const vector uint16_t shifts = (vector uint16_t) {7, 7, 7, 7, 7, 7, 7, 7}; - vector int16_t vi, vileft, ditherleft, ditherright; - vector uint8_t vd; + const vec_u16 shifts = (vec_u16) {7, 7, 7, 7, 7, 7, 7, 7}; + vec_s16 vi, vileft, ditherleft, ditherright; + vec_u8 vd; for (j = 0; j < 16; j++) { val[j] = dither[(dst_u + offset + j) & 7]; @@ -161,11 +161,11 @@ static void yuv2plane1_nbps_vsx(const int16_t *src, uint16_t *dest, int dstW, const int shift = 15 - output_bits; const int add = (1 << (shift - 1)); const int clip = (1 << output_bits) - 1; - const vector uint16_t vadd = (vector uint16_t) {add, add, add, add, add, add, add, add}; - const vector uint16_t vswap = (vector uint16_t) vec_splat_u16(big_endian ? 8 : 0); - const vector uint16_t vshift = (vector uint16_t) vec_splat_u16(shift); - const vector uint16_t vlargest = (vector uint16_t) {clip, clip, clip, clip, clip, clip, clip, clip}; - vector uint16_t v; + const vec_u16 vadd = (vec_u16) {add, add, add, add, add, add, add, add}; + const vec_u16 vswap = (vec_u16) vec_splat_u16(big_endian ? 8 : 0); + const vec_u16 vshift = (vec_u16) vec_splat_u16(shift); + const vec_u16 vlargest = (vec_u16) {clip, clip, clip, clip, clip, clip, clip, clip}; + vec_u16 v; int i; yuv2plane1_nbps_u(src, dest, dst_u, big_endian, output_bits, 0); @@ -209,20 +209,20 @@ static void yuv2planeX_nbps_vsx(const int16_t *filter, int filterSize, const int add = (1 << (shift - 1)); const int clip = (1 << output_bits) - 1; const uint16_t swap = big_endian ? 8 : 0; - const vector uint32_t vadd = (vector uint32_t) {add, add, add, add}; - const vector uint32_t vshift = (vector uint32_t) {shift, shift, shift, shift}; - const vector uint16_t vswap = (vector uint16_t) {swap, swap, swap, swap, swap, swap, swap, swap}; - const vector uint16_t vlargest = (vector uint16_t) {clip, clip, clip, clip, clip, clip, clip, clip}; - const vector int16_t vzero = vec_splat_s16(0); - const vector uint8_t vperm = (vector uint8_t) {0, 1, 8, 9, 2, 3, 10, 11, 4, 5, 12, 13, 6, 7, 14, 15}; - vector int16_t vfilter[MAX_FILTER_SIZE], vin; - vector uint16_t v; - vector uint32_t vleft, vright, vtmp; + const vec_u32 vadd = (vec_u32) {add, add, add, add}; + const vec_u32 vshift = (vec_u32) {shift, shift, shift, shift}; + const vec_u16 vswap = (vec_u16) {swap, swap, swap, swap, swap, swap, swap, swap}; + const vec_u16 vlargest = (vec_u16) {clip, clip, clip, clip, clip, clip, clip, clip}; + const vec_s16 vzero = vec_splat_s16(0); + const vec_u8 vperm = (vec_u8) {0, 1, 8, 9, 2, 3, 10, 11, 4, 5, 12, 13, 6, 7, 14, 15}; + vec_s16 vfilter[MAX_FILTER_SIZE], vin; + vec_u16 v; + vec_u32 vleft, vright, vtmp; int i, j; for (i = 0; i < filterSize; i++) { - vfilter[i] = (vector int16_t) {filter[i], filter[i], filter[i], filter[i], - filter[i], filter[i], filter[i], filter[i]}; + vfilter[i] = (vec_s16) {filter[i], filter[i], filter[i], filter[i], + filter[i], filter[i], filter[i], filter[i]}; } yuv2planeX_nbps_u(filter, filterSize, src, dest, dst_u, big_endian, output_bits, 0); @@ -232,16 +232,16 @@ static void yuv2planeX_nbps_vsx(const int16_t *filter, int filterSize, for (j = 0; j < filterSize; j++) { vin = vec_vsx_ld(0, &src[j][i]); - vtmp = (vector uint32_t) vec_mule(vin, vfilter[j]); + vtmp = (vec_u32) vec_mule(vin, vfilter[j]); vleft = vec_add(vleft, vtmp); - vtmp = (vector uint32_t) vec_mulo(vin, vfilter[j]); + vtmp = (vec_u32) vec_mulo(vin, vfilter[j]); vright = vec_add(vright, vtmp); } vleft = vec_sra(vleft, vshift); vright = vec_sra(vright, vshift); v = vec_packsu(vleft, vright); - v = (vector uint16_t) vec_max((vector int16_t) v, vzero); + v = (vec_u16) vec_max((vec_s16) v, vzero); v = vec_min(v, vlargest); v = vec_rl(v, vswap); v = vec_perm(v, v, vperm); @@ -279,11 +279,11 @@ static void yuv2plane1_16_vsx(const int32_t *src, uint16_t *dest, int dstW, const int dst_u = -(uintptr_t)dest & 7; const int shift = 3; const int add = (1 << (shift - 1)); - const vector uint32_t vadd = (vector uint32_t) {add, add, add, add}; - const vector uint16_t vswap = (vector uint16_t) vec_splat_u16(big_endian ? 8 : 0); - const vector uint32_t vshift = (vector uint32_t) vec_splat_u32(shift); - vector uint32_t v, v2; - vector uint16_t vd; + const vec_u32 vadd = (vec_u32) {add, add, add, add}; + const vec_u16 vswap = (vec_u16) vec_splat_u16(big_endian ? 8 : 0); + const vec_u32 vshift = (vec_u32) vec_splat_u32(shift); + vec_u32 v, v2; + vec_u16 vd; int i; yuv2plane1_16_u(src, dest, dst_u, big_endian, output_bits, 0); @@ -341,18 +341,18 @@ static void yuv2planeX_16_vsx(const int16_t *filter, int filterSize, const int bias = 0x8000; const int add = (1 << (shift - 1)) - 0x40000000; const uint16_t swap = big_endian ? 8 : 0; - const vector uint32_t vadd = (vector uint32_t) {add, add, add, add}; - const vector uint32_t vshift = (vector uint32_t) {shift, shift, shift, shift}; - const vector uint16_t vswap = (vector uint16_t) {swap, swap, swap, swap, swap, swap, swap, swap}; - const vector uint16_t vbias = (vector uint16_t) {bias, bias, bias, bias, bias, bias, bias, bias}; - vector int32_t vfilter[MAX_FILTER_SIZE]; - vector uint16_t v; - vector uint32_t vleft, vright, vtmp; - vector int32_t vin32l, vin32r; + const vec_u32 vadd = (vec_u32) {add, add, add, add}; + const vec_u32 vshift = (vec_u32) {shift, shift, shift, shift}; + const vec_u16 vswap = (vec_u16) {swap, swap, swap, swap, swap, swap, swap, swap}; + const vec_u16 vbias = (vec_u16) {bias, bias, bias, bias, bias, bias, bias, bias}; + vec_s32 vfilter[MAX_FILTER_SIZE]; + vec_u16 v; + vec_u32 vleft, vright, vtmp; + vec_s32 vin32l, vin32r; int i, j; for (i = 0; i < filterSize; i++) { - vfilter[i] = (vector int32_t) {filter[i], filter[i], filter[i], filter[i]}; + vfilter[i] = (vec_s32) {filter[i], filter[i], filter[i], filter[i]}; } yuv2planeX_16_u(filter, filterSize, src, dest, dst_u, big_endian, output_bits, 0); @@ -364,15 +364,15 @@ static void yuv2planeX_16_vsx(const int16_t *filter, int filterSize, vin32l = vec_vsx_ld(0, &src[j][i]); vin32r = vec_vsx_ld(0, &src[j][i + 4]); - vtmp = (vector uint32_t) vec_mul(vin32l, vfilter[j]); + vtmp = (vec_u32) vec_mul(vin32l, vfilter[j]); vleft = vec_add(vleft, vtmp); - vtmp = (vector uint32_t) vec_mul(vin32r, vfilter[j]); + vtmp = (vec_u32) vec_mul(vin32r, vfilter[j]); vright = vec_add(vright, vtmp); } vleft = vec_sra(vleft, vshift); vright = vec_sra(vright, vshift); - v = (vector uint16_t) vec_packs((vector int32_t) vleft, (vector int32_t) vright); + v = (vec_u16) vec_packs((vec_s32) vleft, (vec_s32) vright); v = vec_add(v, vbias); v = vec_rl(v, vswap); vec_st(v, 0, &dest[i]); @@ -478,9 +478,9 @@ yuv2NBPSX(16, LE, 0, 16, int32_t) out0 = vec_mergeh(bd, gd); \ out1 = vec_mergeh(rd, ad); \ \ - tmp8 = (vector uint8_t) vec_mergeh((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergeh((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 0, dest); \ - tmp8 = (vector uint8_t) vec_mergel((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergel((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 16, dest); \ \ dest += 32; \ @@ -489,9 +489,9 @@ yuv2NBPSX(16, LE, 0, 16, int32_t) out0 = vec_mergeh(rd, gd); \ out1 = vec_mergeh(bd, ad); \ \ - tmp8 = (vector uint8_t) vec_mergeh((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergeh((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 0, dest); \ - tmp8 = (vector uint8_t) vec_mergel((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergel((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 16, dest); \ \ dest += 32; \ @@ -500,9 +500,9 @@ yuv2NBPSX(16, LE, 0, 16, int32_t) out0 = vec_mergeh(ad, rd); \ out1 = vec_mergeh(gd, bd); \ \ - tmp8 = (vector uint8_t) vec_mergeh((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergeh((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 0, dest); \ - tmp8 = (vector uint8_t) vec_mergel((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergel((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 16, dest); \ \ dest += 32; \ @@ -511,9 +511,9 @@ yuv2NBPSX(16, LE, 0, 16, int32_t) out0 = vec_mergeh(ad, bd); \ out1 = vec_mergeh(gd, rd); \ \ - tmp8 = (vector uint8_t) vec_mergeh((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergeh((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 0, dest); \ - tmp8 = (vector uint8_t) vec_mergel((vector uint16_t) out0, (vector uint16_t) out1); \ + tmp8 = (vec_u8) vec_mergel((vec_u16) out0, (vec_u16) out1); \ vec_vsx_st(tmp8, 16, dest); \ \ dest += 32; \ @@ -528,48 +528,48 @@ yuv2rgb_full_X_vsx_template(SwsContext *c, const int16_t *lumFilter, const int16_t **alpSrc, uint8_t *dest, int dstW, int y, enum AVPixelFormat target, int hasAlpha) { - vector int16_t vv; - vector int32_t vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32; - vector int32_t R_l, R_r, G_l, G_r, B_l, B_r; - vector int32_t tmp, tmp2, tmp3, tmp4; - vector uint16_t rd16, gd16, bd16; - vector uint8_t rd, bd, gd, ad, out0, out1, tmp8; - vector int16_t vlumFilter[MAX_FILTER_SIZE], vchrFilter[MAX_FILTER_SIZE]; - const vector int32_t ystart = vec_splats(1 << 9); - const vector int32_t uvstart = vec_splats((1 << 9) - (128 << 19)); - const vector uint16_t zero16 = vec_splat_u16(0); - const vector int32_t y_offset = vec_splats(c->yuv2rgb_y_offset); - const vector int32_t y_coeff = vec_splats(c->yuv2rgb_y_coeff); - const vector int32_t y_add = vec_splats(1 << 21); - const vector int32_t v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); - const vector int32_t v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); - const vector int32_t u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); - const vector int32_t u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); - const vector int32_t rgbclip = vec_splats(1 << 30); - const vector int32_t zero32 = vec_splat_s32(0); - const vector uint32_t shift22 = vec_splats(22U); - const vector uint32_t shift10 = vec_splat_u32(10); + vec_s16 vv; + vec_s32 vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32; + vec_s32 R_l, R_r, G_l, G_r, B_l, B_r; + vec_s32 tmp, tmp2, tmp3, tmp4; + vec_u16 rd16, gd16, bd16; + vec_u8 rd, bd, gd, ad, out0, out1, tmp8; + vec_s16 vlumFilter[MAX_FILTER_SIZE], vchrFilter[MAX_FILTER_SIZE]; + const vec_s32 ystart = vec_splats(1 << 9); + const vec_s32 uvstart = vec_splats((1 << 9) - (128 << 19)); + const vec_u16 zero16 = vec_splat_u16(0); + const vec_s32 y_offset = vec_splats(c->yuv2rgb_y_offset); + const vec_s32 y_coeff = vec_splats(c->yuv2rgb_y_coeff); + const vec_s32 y_add = vec_splats(1 << 21); + const vec_s32 v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); + const vec_s32 v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); + const vec_s32 u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); + const vec_s32 u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); + const vec_s32 rgbclip = vec_splats(1 << 30); + const vec_s32 zero32 = vec_splat_s32(0); + const vec_u32 shift22 = vec_splats(22U); + const vec_u32 shift10 = vec_splat_u32(10); int i, j; // Various permutations - const vector uint8_t perm3rg0 = (vector uint8_t) {0x0, 0x10, 0, - 0x1, 0x11, 0, - 0x2, 0x12, 0, - 0x3, 0x13, 0, - 0x4, 0x14, 0, - 0x5 }; - const vector uint8_t perm3rg1 = (vector uint8_t) { 0x15, 0, - 0x6, 0x16, 0, - 0x7, 0x17, 0 }; - const vector uint8_t perm3tb0 = (vector uint8_t) {0x0, 0x1, 0x10, - 0x3, 0x4, 0x11, - 0x6, 0x7, 0x12, - 0x9, 0xa, 0x13, - 0xc, 0xd, 0x14, - 0xf }; - const vector uint8_t perm3tb1 = (vector uint8_t) { 0x0, 0x15, - 0x2, 0x3, 0x16, - 0x5, 0x6, 0x17 }; + const vec_u8 perm3rg0 = (vec_u8) {0x0, 0x10, 0, + 0x1, 0x11, 0, + 0x2, 0x12, 0, + 0x3, 0x13, 0, + 0x4, 0x14, 0, + 0x5 }; + const vec_u8 perm3rg1 = (vec_u8) { 0x15, 0, + 0x6, 0x16, 0, + 0x7, 0x17, 0 }; + const vec_u8 perm3tb0 = (vec_u8) {0x0, 0x1, 0x10, + 0x3, 0x4, 0x11, + 0x6, 0x7, 0x12, + 0x9, 0xa, 0x13, + 0xc, 0xd, 0x14, + 0xf }; + const vec_u8 perm3tb1 = (vec_u8) { 0x0, 0x15, + 0x2, 0x3, 0x16, + 0x5, 0x6, 0x17 }; ad = vec_splats((uint8_t) 255); @@ -685,52 +685,52 @@ yuv2rgb_full_2_vsx_template(SwsContext *c, const int16_t *buf[2], *abuf1 = hasAlpha ? abuf[1] : NULL; const int16_t yalpha1 = 4096 - yalpha; const int16_t uvalpha1 = 4096 - uvalpha; - vector int16_t vy, vu, vv, A = vec_splat_s16(0); - vector int32_t vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32; - vector int32_t R_l, R_r, G_l, G_r, B_l, B_r; - vector int32_t tmp, tmp2, tmp3, tmp4, tmp5, tmp6; - vector uint16_t rd16, gd16, bd16; - vector uint8_t rd, bd, gd, ad, out0, out1, tmp8; - const vector int16_t vyalpha1 = vec_splats(yalpha1); - const vector int16_t vuvalpha1 = vec_splats(uvalpha1); - const vector int16_t vyalpha = vec_splats((int16_t) yalpha); - const vector int16_t vuvalpha = vec_splats((int16_t) uvalpha); - const vector uint16_t zero16 = vec_splat_u16(0); - const vector int32_t y_offset = vec_splats(c->yuv2rgb_y_offset); - const vector int32_t y_coeff = vec_splats(c->yuv2rgb_y_coeff); - const vector int32_t y_add = vec_splats(1 << 21); - const vector int32_t v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); - const vector int32_t v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); - const vector int32_t u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); - const vector int32_t u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); - const vector int32_t rgbclip = vec_splats(1 << 30); - const vector int32_t zero32 = vec_splat_s32(0); - const vector uint32_t shift19 = vec_splats(19U); - const vector uint32_t shift22 = vec_splats(22U); - const vector uint32_t shift10 = vec_splat_u32(10); - const vector int32_t dec128 = vec_splats(128 << 19); - const vector int32_t add18 = vec_splats(1 << 18); + vec_s16 vy, vu, vv, A = vec_splat_s16(0); + vec_s32 vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32; + vec_s32 R_l, R_r, G_l, G_r, B_l, B_r; + vec_s32 tmp, tmp2, tmp3, tmp4, tmp5, tmp6; + vec_u16 rd16, gd16, bd16; + vec_u8 rd, bd, gd, ad, out0, out1, tmp8; + const vec_s16 vyalpha1 = vec_splats(yalpha1); + const vec_s16 vuvalpha1 = vec_splats(uvalpha1); + const vec_s16 vyalpha = vec_splats((int16_t) yalpha); + const vec_s16 vuvalpha = vec_splats((int16_t) uvalpha); + const vec_u16 zero16 = vec_splat_u16(0); + const vec_s32 y_offset = vec_splats(c->yuv2rgb_y_offset); + const vec_s32 y_coeff = vec_splats(c->yuv2rgb_y_coeff); + const vec_s32 y_add = vec_splats(1 << 21); + const vec_s32 v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); + const vec_s32 v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); + const vec_s32 u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); + const vec_s32 u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); + const vec_s32 rgbclip = vec_splats(1 << 30); + const vec_s32 zero32 = vec_splat_s32(0); + const vec_u32 shift19 = vec_splats(19U); + const vec_u32 shift22 = vec_splats(22U); + const vec_u32 shift10 = vec_splat_u32(10); + const vec_s32 dec128 = vec_splats(128 << 19); + const vec_s32 add18 = vec_splats(1 << 18); int i; // Various permutations - const vector uint8_t perm3rg0 = (vector uint8_t) {0x0, 0x10, 0, - 0x1, 0x11, 0, - 0x2, 0x12, 0, - 0x3, 0x13, 0, - 0x4, 0x14, 0, - 0x5 }; - const vector uint8_t perm3rg1 = (vector uint8_t) { 0x15, 0, - 0x6, 0x16, 0, - 0x7, 0x17, 0 }; - const vector uint8_t perm3tb0 = (vector uint8_t) {0x0, 0x1, 0x10, - 0x3, 0x4, 0x11, - 0x6, 0x7, 0x12, - 0x9, 0xa, 0x13, - 0xc, 0xd, 0x14, - 0xf }; - const vector uint8_t perm3tb1 = (vector uint8_t) { 0x0, 0x15, - 0x2, 0x3, 0x16, - 0x5, 0x6, 0x17 }; + const vec_u8 perm3rg0 = (vec_u8) {0x0, 0x10, 0, + 0x1, 0x11, 0, + 0x2, 0x12, 0, + 0x3, 0x13, 0, + 0x4, 0x14, 0, + 0x5 }; + const vec_u8 perm3rg1 = (vec_u8) { 0x15, 0, + 0x6, 0x16, 0, + 0x7, 0x17, 0 }; + const vec_u8 perm3tb0 = (vec_u8) {0x0, 0x1, 0x10, + 0x3, 0x4, 0x11, + 0x6, 0x7, 0x12, + 0x9, 0xa, 0x13, + 0xc, 0xd, 0x14, + 0xf }; + const vec_u8 perm3tb1 = (vec_u8) { 0x0, 0x15, + 0x2, 0x3, 0x16, + 0x5, 0x6, 0x17 }; av_assert2(yalpha <= 4096U); av_assert2(uvalpha <= 4096U); @@ -759,7 +759,7 @@ yuv2rgb_full_2_vsx_template(SwsContext *c, const int16_t *buf[2], tmp3 = vec_sra(tmp3, shift19); tmp4 = vec_sra(tmp4, shift19); A = vec_packs(tmp3, tmp4); - ad = vec_packsu(A, (vector int16_t) zero16); + ad = vec_packsu(A, (vec_s16) zero16); } else { ad = vec_splats((uint8_t) 255); } @@ -807,60 +807,60 @@ yuv2rgb_2_vsx_template(SwsContext *c, const int16_t *buf[2], *abuf1 = hasAlpha ? abuf[1] : NULL; const int16_t yalpha1 = 4096 - yalpha; const int16_t uvalpha1 = 4096 - uvalpha; - vector int16_t vy, vu, vv, A = vec_splat_s16(0); - vector int32_t vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32; - vector int32_t R_l, R_r, G_l, G_r, B_l, B_r, vud32_l, vud32_r, vvd32_l, vvd32_r; - vector int32_t tmp, tmp2, tmp3, tmp4, tmp5, tmp6; - vector uint16_t rd16, gd16, bd16; - vector uint8_t rd, bd, gd, ad, out0, out1, tmp8; - const vector int16_t vyalpha1 = vec_splats(yalpha1); - const vector int16_t vuvalpha1 = vec_splats(uvalpha1); - const vector int16_t vyalpha = vec_splats((int16_t) yalpha); - const vector int16_t vuvalpha = vec_splats((int16_t) uvalpha); - const vector uint16_t zero16 = vec_splat_u16(0); - const vector int32_t y_offset = vec_splats(c->yuv2rgb_y_offset); - const vector int32_t y_coeff = vec_splats(c->yuv2rgb_y_coeff); - const vector int32_t y_add = vec_splats(1 << 21); - const vector int32_t v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); - const vector int32_t v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); - const vector int32_t u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); - const vector int32_t u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); - const vector int32_t rgbclip = vec_splats(1 << 30); - const vector int32_t zero32 = vec_splat_s32(0); - const vector uint32_t shift19 = vec_splats(19U); - const vector uint32_t shift22 = vec_splats(22U); - const vector uint32_t shift10 = vec_splat_u32(10); - const vector int32_t dec128 = vec_splats(128 << 19); - const vector int32_t add18 = vec_splats(1 << 18); + vec_s16 vy, vu, vv, A = vec_splat_s16(0); + vec_s32 vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32; + vec_s32 R_l, R_r, G_l, G_r, B_l, B_r, vud32_l, vud32_r, vvd32_l, vvd32_r; + vec_s32 tmp, tmp2, tmp3, tmp4, tmp5, tmp6; + vec_u16 rd16, gd16, bd16; + vec_u8 rd, bd, gd, ad, out0, out1, tmp8; + const vec_s16 vyalpha1 = vec_splats(yalpha1); + const vec_s16 vuvalpha1 = vec_splats(uvalpha1); + const vec_s16 vyalpha = vec_splats((int16_t) yalpha); + const vec_s16 vuvalpha = vec_splats((int16_t) uvalpha); + const vec_u16 zero16 = vec_splat_u16(0); + const vec_s32 y_offset = vec_splats(c->yuv2rgb_y_offset); + const vec_s32 y_coeff = vec_splats(c->yuv2rgb_y_coeff); + const vec_s32 y_add = vec_splats(1 << 21); + const vec_s32 v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); + const vec_s32 v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); + const vec_s32 u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); + const vec_s32 u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); + const vec_s32 rgbclip = vec_splats(1 << 30); + const vec_s32 zero32 = vec_splat_s32(0); + const vec_u32 shift19 = vec_splats(19U); + const vec_u32 shift22 = vec_splats(22U); + const vec_u32 shift10 = vec_splat_u32(10); + const vec_s32 dec128 = vec_splats(128 << 19); + const vec_s32 add18 = vec_splats(1 << 18); int i; // Various permutations - const vector uint8_t doubleleft = (vector uint8_t) {0, 1, 2, 3, - 0, 1, 2, 3, - 4, 5, 6, 7, - 4, 5, 6, 7 }; - const vector uint8_t doubleright = (vector uint8_t) {8, 9, 10, 11, - 8, 9, 10, 11, - 12, 13, 14, 15, - 12, 13, 14, 15 }; - const vector uint8_t perm3rg0 = (vector uint8_t) {0x0, 0x10, 0, - 0x1, 0x11, 0, - 0x2, 0x12, 0, - 0x3, 0x13, 0, - 0x4, 0x14, 0, - 0x5 }; - const vector uint8_t perm3rg1 = (vector uint8_t) { 0x15, 0, - 0x6, 0x16, 0, - 0x7, 0x17, 0 }; - const vector uint8_t perm3tb0 = (vector uint8_t) {0x0, 0x1, 0x10, - 0x3, 0x4, 0x11, - 0x6, 0x7, 0x12, - 0x9, 0xa, 0x13, - 0xc, 0xd, 0x14, - 0xf }; - const vector uint8_t perm3tb1 = (vector uint8_t) { 0x0, 0x15, - 0x2, 0x3, 0x16, - 0x5, 0x6, 0x17 }; + const vec_u8 doubleleft = (vec_u8) {0, 1, 2, 3, + 0, 1, 2, 3, + 4, 5, 6, 7, + 4, 5, 6, 7 }; + const vec_u8 doubleright = (vec_u8) {8, 9, 10, 11, + 8, 9, 10, 11, + 12, 13, 14, 15, + 12, 13, 14, 15 }; + const vec_u8 perm3rg0 = (vec_u8) {0x0, 0x10, 0, + 0x1, 0x11, 0, + 0x2, 0x12, 0, + 0x3, 0x13, 0, + 0x4, 0x14, 0, + 0x5 }; + const vec_u8 perm3rg1 = (vec_u8) { 0x15, 0, + 0x6, 0x16, 0, + 0x7, 0x17, 0 }; + const vec_u8 perm3tb0 = (vec_u8) {0x0, 0x1, 0x10, + 0x3, 0x4, 0x11, + 0x6, 0x7, 0x12, + 0x9, 0xa, 0x13, + 0xc, 0xd, 0x14, + 0xf }; + const vec_u8 perm3tb1 = (vec_u8) { 0x0, 0x15, + 0x2, 0x3, 0x16, + 0x5, 0x6, 0x17 }; av_assert2(yalpha <= 4096U); av_assert2(uvalpha <= 4096U); @@ -889,7 +889,7 @@ yuv2rgb_2_vsx_template(SwsContext *c, const int16_t *buf[2], tmp3 = vec_sra(tmp3, shift19); tmp4 = vec_sra(tmp4, shift19); A = vec_packs(tmp3, tmp4); - ad = vec_packsu(A, (vector int16_t) zero16); + ad = vec_packsu(A, (vec_s16) zero16); } else { ad = vec_splats((uint8_t) 255); } @@ -978,51 +978,51 @@ yuv2rgb_full_1_vsx_template(SwsContext *c, const int16_t *buf0, { const int16_t *ubuf0 = ubuf[0], *vbuf0 = vbuf[0]; const int16_t *ubuf1 = ubuf[1], *vbuf1 = vbuf[1]; - vector int16_t vy, vu, vv, A = vec_splat_s16(0), tmp16; - vector int32_t vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32, tmp32_2; - vector int32_t R_l, R_r, G_l, G_r, B_l, B_r; - vector uint16_t rd16, gd16, bd16; - vector uint8_t rd, bd, gd, ad, out0, out1, tmp8; - const vector uint16_t zero16 = vec_splat_u16(0); - const vector int32_t y_offset = vec_splats(c->yuv2rgb_y_offset); - const vector int32_t y_coeff = vec_splats(c->yuv2rgb_y_coeff); - const vector int32_t y_add = vec_splats(1 << 21); - const vector int32_t v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); - const vector int32_t v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); - const vector int32_t u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); - const vector int32_t u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); - const vector int32_t rgbclip = vec_splats(1 << 30); - const vector int32_t zero32 = vec_splat_s32(0); - const vector uint32_t shift2 = vec_splat_u32(2); - const vector uint32_t shift22 = vec_splats(22U); - const vector uint16_t sub7 = vec_splats((uint16_t) (128 << 7)); - const vector uint16_t sub8 = vec_splats((uint16_t) (128 << 8)); - const vector int16_t mul4 = vec_splat_s16(4); - const vector int16_t mul8 = vec_splat_s16(8); - const vector int16_t add64 = vec_splat_s16(64); - const vector uint16_t shift7 = vec_splat_u16(7); - const vector int16_t max255 = vec_splat_s16(255); + vec_s16 vy, vu, vv, A = vec_splat_s16(0), tmp16; + vec_s32 vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32, tmp32_2; + vec_s32 R_l, R_r, G_l, G_r, B_l, B_r; + vec_u16 rd16, gd16, bd16; + vec_u8 rd, bd, gd, ad, out0, out1, tmp8; + const vec_u16 zero16 = vec_splat_u16(0); + const vec_s32 y_offset = vec_splats(c->yuv2rgb_y_offset); + const vec_s32 y_coeff = vec_splats(c->yuv2rgb_y_coeff); + const vec_s32 y_add = vec_splats(1 << 21); + const vec_s32 v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); + const vec_s32 v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); + const vec_s32 u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); + const vec_s32 u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); + const vec_s32 rgbclip = vec_splats(1 << 30); + const vec_s32 zero32 = vec_splat_s32(0); + const vec_u32 shift2 = vec_splat_u32(2); + const vec_u32 shift22 = vec_splats(22U); + const vec_u16 sub7 = vec_splats((uint16_t) (128 << 7)); + const vec_u16 sub8 = vec_splats((uint16_t) (128 << 8)); + const vec_s16 mul4 = vec_splat_s16(4); + const vec_s16 mul8 = vec_splat_s16(8); + const vec_s16 add64 = vec_splat_s16(64); + const vec_u16 shift7 = vec_splat_u16(7); + const vec_s16 max255 = vec_splat_s16(255); int i; // Various permutations - const vector uint8_t perm3rg0 = (vector uint8_t) {0x0, 0x10, 0, - 0x1, 0x11, 0, - 0x2, 0x12, 0, - 0x3, 0x13, 0, - 0x4, 0x14, 0, - 0x5 }; - const vector uint8_t perm3rg1 = (vector uint8_t) { 0x15, 0, - 0x6, 0x16, 0, - 0x7, 0x17, 0 }; - const vector uint8_t perm3tb0 = (vector uint8_t) {0x0, 0x1, 0x10, - 0x3, 0x4, 0x11, - 0x6, 0x7, 0x12, - 0x9, 0xa, 0x13, - 0xc, 0xd, 0x14, - 0xf }; - const vector uint8_t perm3tb1 = (vector uint8_t) { 0x0, 0x15, - 0x2, 0x3, 0x16, - 0x5, 0x6, 0x17 }; + const vec_u8 perm3rg0 = (vec_u8) {0x0, 0x10, 0, + 0x1, 0x11, 0, + 0x2, 0x12, 0, + 0x3, 0x13, 0, + 0x4, 0x14, 0, + 0x5 }; + const vec_u8 perm3rg1 = (vec_u8) { 0x15, 0, + 0x6, 0x16, 0, + 0x7, 0x17, 0 }; + const vec_u8 perm3tb0 = (vec_u8) {0x0, 0x1, 0x10, + 0x3, 0x4, 0x11, + 0x6, 0x7, 0x12, + 0x9, 0xa, 0x13, + 0xc, 0xd, 0x14, + 0xf }; + const vec_u8 perm3tb1 = (vec_u8) { 0x0, 0x15, + 0x2, 0x3, 0x16, + 0x5, 0x6, 0x17 }; for (i = 0; i < dstW; i += 8) { // The x86 asm also overwrites padding bytes. vy = vec_ld(0, &buf0[i]); @@ -1034,8 +1034,8 @@ yuv2rgb_full_1_vsx_template(SwsContext *c, const int16_t *buf0, vu = vec_ld(0, &ubuf0[i]); vv = vec_ld(0, &vbuf0[i]); if (uvalpha < 2048) { - vu = (vector int16_t) vec_sub((vector uint16_t) vu, sub7); - vv = (vector int16_t) vec_sub((vector uint16_t) vv, sub7); + vu = (vec_s16) vec_sub((vec_u16) vu, sub7); + vv = (vec_s16) vec_sub((vec_u16) vv, sub7); tmp32 = vec_mule(vu, mul4); tmp32_2 = vec_mulo(vu, mul4); @@ -1048,10 +1048,10 @@ yuv2rgb_full_1_vsx_template(SwsContext *c, const int16_t *buf0, } else { tmp16 = vec_ld(0, &ubuf1[i]); vu = vec_add(vu, tmp16); - vu = (vector int16_t) vec_sub((vector uint16_t) vu, sub8); + vu = (vec_s16) vec_sub((vec_u16) vu, sub8); tmp16 = vec_ld(0, &vbuf1[i]); vv = vec_add(vv, tmp16); - vv = (vector int16_t) vec_sub((vector uint16_t) vv, sub8); + vv = (vec_s16) vec_sub((vec_u16) vv, sub8); vu32_l = vec_mule(vu, mul8); vu32_r = vec_mulo(vu, mul8); @@ -1064,7 +1064,7 @@ yuv2rgb_full_1_vsx_template(SwsContext *c, const int16_t *buf0, A = vec_add(A, add64); A = vec_sr(A, shift7); A = vec_max(A, max255); - ad = vec_packsu(A, (vector int16_t) zero16); + ad = vec_packsu(A, (vec_s16) zero16); } else { ad = vec_splats((uint8_t) 255); } @@ -1107,60 +1107,60 @@ yuv2rgb_1_vsx_template(SwsContext *c, const int16_t *buf0, { const int16_t *ubuf0 = ubuf[0], *vbuf0 = vbuf[0]; const int16_t *ubuf1 = ubuf[1], *vbuf1 = vbuf[1]; - vector int16_t vy, vu, vv, A = vec_splat_s16(0), tmp16; - vector int32_t vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32, tmp32_2; - vector int32_t vud32_l, vud32_r, vvd32_l, vvd32_r; - vector int32_t R_l, R_r, G_l, G_r, B_l, B_r; - vector uint16_t rd16, gd16, bd16; - vector uint8_t rd, bd, gd, ad, out0, out1, tmp8; - const vector uint16_t zero16 = vec_splat_u16(0); - const vector int32_t y_offset = vec_splats(c->yuv2rgb_y_offset); - const vector int32_t y_coeff = vec_splats(c->yuv2rgb_y_coeff); - const vector int32_t y_add = vec_splats(1 << 21); - const vector int32_t v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); - const vector int32_t v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); - const vector int32_t u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); - const vector int32_t u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); - const vector int32_t rgbclip = vec_splats(1 << 30); - const vector int32_t zero32 = vec_splat_s32(0); - const vector uint32_t shift2 = vec_splat_u32(2); - const vector uint32_t shift22 = vec_splats(22U); - const vector uint16_t sub7 = vec_splats((uint16_t) (128 << 7)); - const vector uint16_t sub8 = vec_splats((uint16_t) (128 << 8)); - const vector int16_t mul4 = vec_splat_s16(4); - const vector int16_t mul8 = vec_splat_s16(8); - const vector int16_t add64 = vec_splat_s16(64); - const vector uint16_t shift7 = vec_splat_u16(7); - const vector int16_t max255 = vec_splat_s16(255); + vec_s16 vy, vu, vv, A = vec_splat_s16(0), tmp16; + vec_s32 vy32_l, vy32_r, vu32_l, vu32_r, vv32_l, vv32_r, tmp32, tmp32_2; + vec_s32 vud32_l, vud32_r, vvd32_l, vvd32_r; + vec_s32 R_l, R_r, G_l, G_r, B_l, B_r; + vec_u16 rd16, gd16, bd16; + vec_u8 rd, bd, gd, ad, out0, out1, tmp8; + const vec_u16 zero16 = vec_splat_u16(0); + const vec_s32 y_offset = vec_splats(c->yuv2rgb_y_offset); + const vec_s32 y_coeff = vec_splats(c->yuv2rgb_y_coeff); + const vec_s32 y_add = vec_splats(1 << 21); + const vec_s32 v2r_coeff = vec_splats(c->yuv2rgb_v2r_coeff); + const vec_s32 v2g_coeff = vec_splats(c->yuv2rgb_v2g_coeff); + const vec_s32 u2g_coeff = vec_splats(c->yuv2rgb_u2g_coeff); + const vec_s32 u2b_coeff = vec_splats(c->yuv2rgb_u2b_coeff); + const vec_s32 rgbclip = vec_splats(1 << 30); + const vec_s32 zero32 = vec_splat_s32(0); + const vec_u32 shift2 = vec_splat_u32(2); + const vec_u32 shift22 = vec_splats(22U); + const vec_u16 sub7 = vec_splats((uint16_t) (128 << 7)); + const vec_u16 sub8 = vec_splats((uint16_t) (128 << 8)); + const vec_s16 mul4 = vec_splat_s16(4); + const vec_s16 mul8 = vec_splat_s16(8); + const vec_s16 add64 = vec_splat_s16(64); + const vec_u16 shift7 = vec_splat_u16(7); + const vec_s16 max255 = vec_splat_s16(255); int i; // Various permutations - const vector uint8_t doubleleft = (vector uint8_t) {0, 1, 2, 3, - 0, 1, 2, 3, - 4, 5, 6, 7, - 4, 5, 6, 7 }; - const vector uint8_t doubleright = (vector uint8_t) {8, 9, 10, 11, - 8, 9, 10, 11, - 12, 13, 14, 15, - 12, 13, 14, 15 }; - const vector uint8_t perm3rg0 = (vector uint8_t) {0x0, 0x10, 0, - 0x1, 0x11, 0, - 0x2, 0x12, 0, - 0x3, 0x13, 0, - 0x4, 0x14, 0, - 0x5 }; - const vector uint8_t perm3rg1 = (vector uint8_t) { 0x15, 0, - 0x6, 0x16, 0, - 0x7, 0x17, 0 }; - const vector uint8_t perm3tb0 = (vector uint8_t) {0x0, 0x1, 0x10, - 0x3, 0x4, 0x11, - 0x6, 0x7, 0x12, - 0x9, 0xa, 0x13, - 0xc, 0xd, 0x14, - 0xf }; - const vector uint8_t perm3tb1 = (vector uint8_t) { 0x0, 0x15, - 0x2, 0x3, 0x16, - 0x5, 0x6, 0x17 }; + const vec_u8 doubleleft = (vec_u8) {0, 1, 2, 3, + 0, 1, 2, 3, + 4, 5, 6, 7, + 4, 5, 6, 7 }; + const vec_u8 doubleright = (vec_u8) {8, 9, 10, 11, + 8, 9, 10, 11, + 12, 13, 14, 15, + 12, 13, 14, 15 }; + const vec_u8 perm3rg0 = (vec_u8) {0x0, 0x10, 0, + 0x1, 0x11, 0, + 0x2, 0x12, 0, + 0x3, 0x13, 0, + 0x4, 0x14, 0, + 0x5 }; + const vec_u8 perm3rg1 = (vec_u8) { 0x15, 0, + 0x6, 0x16, 0, + 0x7, 0x17, 0 }; + const vec_u8 perm3tb0 = (vec_u8) {0x0, 0x1, 0x10, + 0x3, 0x4, 0x11, + 0x6, 0x7, 0x12, + 0x9, 0xa, 0x13, + 0xc, 0xd, 0x14, + 0xf }; + const vec_u8 perm3tb1 = (vec_u8) { 0x0, 0x15, + 0x2, 0x3, 0x16, + 0x5, 0x6, 0x17 }; for (i = 0; i < (dstW + 1) >> 1; i += 8) { // The x86 asm also overwrites padding bytes. vy = vec_ld(0, &buf0[i * 2]); @@ -1172,8 +1172,8 @@ yuv2rgb_1_vsx_template(SwsContext *c, const int16_t *buf0, vu = vec_ld(0, &ubuf0[i]); vv = vec_ld(0, &vbuf0[i]); if (uvalpha < 2048) { - vu = (vector int16_t) vec_sub((vector uint16_t) vu, sub7); - vv = (vector int16_t) vec_sub((vector uint16_t) vv, sub7); + vu = (vec_s16) vec_sub((vec_u16) vu, sub7); + vv = (vec_s16) vec_sub((vec_u16) vv, sub7); tmp32 = vec_mule(vu, mul4); tmp32_2 = vec_mulo(vu, mul4); @@ -1186,10 +1186,10 @@ yuv2rgb_1_vsx_template(SwsContext *c, const int16_t *buf0, } else { tmp16 = vec_ld(0, &ubuf1[i]); vu = vec_add(vu, tmp16); - vu = (vector int16_t) vec_sub((vector uint16_t) vu, sub8); + vu = (vec_s16) vec_sub((vec_u16) vu, sub8); tmp16 = vec_ld(0, &vbuf1[i]); vv = vec_add(vv, tmp16); - vv = (vector int16_t) vec_sub((vector uint16_t) vv, sub8); + vv = (vec_s16) vec_sub((vec_u16) vv, sub8); vu32_l = vec_mule(vu, mul8); vu32_r = vec_mulo(vu, mul8); @@ -1202,7 +1202,7 @@ yuv2rgb_1_vsx_template(SwsContext *c, const int16_t *buf0, A = vec_add(A, add64); A = vec_sr(A, shift7); A = vec_max(A, max255); - ad = vec_packsu(A, (vector int16_t) zero16); + ad = vec_packsu(A, (vec_s16) zero16); } else { ad = vec_splats((uint8_t) 255); } @@ -1358,41 +1358,41 @@ YUV2RGBWRAPPERX(yuv2, rgb_full, rgb24_full, AV_PIX_FMT_RGB24, 0) YUV2RGBWRAPPERX(yuv2, rgb_full, bgr24_full, AV_PIX_FMT_BGR24, 0) static av_always_inline void -write422(const vector int16_t vy1, const vector int16_t vy2, - const vector int16_t vu, const vector int16_t vv, +write422(const vec_s16 vy1, const vec_s16 vy2, + const vec_s16 vu, const vec_s16 vv, uint8_t *dest, const enum AVPixelFormat target) { - vector uint8_t vd1, vd2, tmp; - const vector uint8_t yuyv1 = (vector uint8_t) { - 0x0, 0x10, 0x1, 0x18, - 0x2, 0x11, 0x3, 0x19, - 0x4, 0x12, 0x5, 0x1a, - 0x6, 0x13, 0x7, 0x1b }; - const vector uint8_t yuyv2 = (vector uint8_t) { - 0x8, 0x14, 0x9, 0x1c, - 0xa, 0x15, 0xb, 0x1d, - 0xc, 0x16, 0xd, 0x1e, - 0xe, 0x17, 0xf, 0x1f }; - const vector uint8_t yvyu1 = (vector uint8_t) { - 0x0, 0x18, 0x1, 0x10, - 0x2, 0x19, 0x3, 0x11, - 0x4, 0x1a, 0x5, 0x12, - 0x6, 0x1b, 0x7, 0x13 }; - const vector uint8_t yvyu2 = (vector uint8_t) { - 0x8, 0x1c, 0x9, 0x14, - 0xa, 0x1d, 0xb, 0x15, - 0xc, 0x1e, 0xd, 0x16, - 0xe, 0x1f, 0xf, 0x17 }; - const vector uint8_t uyvy1 = (vector uint8_t) { - 0x10, 0x0, 0x18, 0x1, - 0x11, 0x2, 0x19, 0x3, - 0x12, 0x4, 0x1a, 0x5, - 0x13, 0x6, 0x1b, 0x7 }; - const vector uint8_t uyvy2 = (vector uint8_t) { - 0x14, 0x8, 0x1c, 0x9, - 0x15, 0xa, 0x1d, 0xb, - 0x16, 0xc, 0x1e, 0xd, - 0x17, 0xe, 0x1f, 0xf }; + vec_u8 vd1, vd2, tmp; + const vec_u8 yuyv1 = (vec_u8) { + 0x0, 0x10, 0x1, 0x18, + 0x2, 0x11, 0x3, 0x19, + 0x4, 0x12, 0x5, 0x1a, + 0x6, 0x13, 0x7, 0x1b }; + const vec_u8 yuyv2 = (vec_u8) { + 0x8, 0x14, 0x9, 0x1c, + 0xa, 0x15, 0xb, 0x1d, + 0xc, 0x16, 0xd, 0x1e, + 0xe, 0x17, 0xf, 0x1f }; + const vec_u8 yvyu1 = (vec_u8) { + 0x0, 0x18, 0x1, 0x10, + 0x2, 0x19, 0x3, 0x11, + 0x4, 0x1a, 0x5, 0x12, + 0x6, 0x1b, 0x7, 0x13 }; + const vec_u8 yvyu2 = (vec_u8) { + 0x8, 0x1c, 0x9, 0x14, + 0xa, 0x1d, 0xb, 0x15, + 0xc, 0x1e, 0xd, 0x16, + 0xe, 0x1f, 0xf, 0x17 }; + const vec_u8 uyvy1 = (vec_u8) { + 0x10, 0x0, 0x18, 0x1, + 0x11, 0x2, 0x19, 0x3, + 0x12, 0x4, 0x1a, 0x5, + 0x13, 0x6, 0x1b, 0x7 }; + const vec_u8 uyvy2 = (vec_u8) { + 0x14, 0x8, 0x1c, 0x9, + 0x15, 0xa, 0x1d, 0xb, + 0x16, 0xc, 0x1e, 0xd, + 0x17, 0xe, 0x1f, 0xf }; vd1 = vec_packsu(vy1, vy2); vd2 = vec_packsu(vu, vv); @@ -1428,11 +1428,11 @@ yuv2422_X_vsx_template(SwsContext *c, const int16_t *lumFilter, int y, enum AVPixelFormat target) { int i, j; - vector int16_t vy1, vy2, vu, vv; - vector int32_t vy32[4], vu32[2], vv32[2], tmp, tmp2, tmp3, tmp4; - vector int16_t vlumFilter[MAX_FILTER_SIZE], vchrFilter[MAX_FILTER_SIZE]; - const vector int32_t start = vec_splats(1 << 18); - const vector uint32_t shift19 = vec_splats(19U); + vec_s16 vy1, vy2, vu, vv; + vec_s32 vy32[4], vu32[2], vv32[2], tmp, tmp2, tmp3, tmp4; + vec_s16 vlumFilter[MAX_FILTER_SIZE], vchrFilter[MAX_FILTER_SIZE]; + const vec_s32 start = vec_splats(1 << 18); + const vec_u32 shift19 = vec_splats(19U); for (i = 0; i < lumFilterSize; i++) vlumFilter[i] = vec_splats(lumFilter[i]); @@ -1539,11 +1539,11 @@ yuv2422_2_vsx_template(SwsContext *c, const int16_t *buf[2], *vbuf0 = vbuf[0], *vbuf1 = vbuf[1]; const int16_t yalpha1 = 4096 - yalpha; const int16_t uvalpha1 = 4096 - uvalpha; - vector int16_t vy1, vy2, vu, vv; - vector int32_t tmp, tmp2, tmp3, tmp4, tmp5, tmp6; - const vector int16_t vyalpha1 = vec_splats(yalpha1); - const vector int16_t vuvalpha1 = vec_splats(uvalpha1); - const vector uint32_t shift19 = vec_splats(19U); + vec_s16 vy1, vy2, vu, vv; + vec_s32 tmp, tmp2, tmp3, tmp4, tmp5, tmp6; + const vec_s16 vyalpha1 = vec_splats(yalpha1); + const vec_s16 vuvalpha1 = vec_splats(uvalpha1); + const vec_u32 shift19 = vec_splats(19U); int i; av_assert2(yalpha <= 4096U); av_assert2(uvalpha <= 4096U); @@ -1568,11 +1568,11 @@ yuv2422_1_vsx_template(SwsContext *c, const int16_t *buf0, int uvalpha, int y, enum AVPixelFormat target) { const int16_t *ubuf0 = ubuf[0], *vbuf0 = vbuf[0]; - vector int16_t vy1, vy2, vu, vv, tmp; - const vector int16_t add64 = vec_splats((int16_t) 64); - const vector int16_t add128 = vec_splats((int16_t) 128); - const vector uint16_t shift7 = vec_splat_u16(7); - const vector uint16_t shift8 = vec_splat_u16(8); + vec_s16 vy1, vy2, vu, vv, tmp; + const vec_s16 add64 = vec_splats((int16_t) 64); + const vec_s16 add128 = vec_splats((int16_t) 128); + const vec_u16 shift7 = vec_splat_u16(7); + const vec_u16 shift8 = vec_splat_u16(8); int i; if (uvalpha < 2048) { @@ -1666,18 +1666,18 @@ static void hyscale_fast_vsx(SwsContext *c, int16_t *dst, int dstWidth, { int i; unsigned int xpos = 0, xx; - vector uint8_t vin, vin2, vperm; - vector int8_t vmul, valpha; - vector int16_t vtmp, vtmp2, vtmp3, vtmp4; - vector uint16_t vd_l, vd_r, vcoord16[2]; - vector uint32_t vcoord[4]; - const vector uint32_t vadd = (vector uint32_t) { + vec_u8 vin, vin2, vperm; + vec_s8 vmul, valpha; + vec_s16 vtmp, vtmp2, vtmp3, vtmp4; + vec_u16 vd_l, vd_r, vcoord16[2]; + vec_u32 vcoord[4]; + const vec_u32 vadd = (vec_u32) { 0, xInc * 1, xInc * 2, xInc * 3, }; - const vector uint16_t vadd16 = (vector uint16_t) { // Modulo math + const vec_u16 vadd16 = (vec_u16) { // Modulo math 0, xInc * 1, xInc * 2, @@ -1687,10 +1687,10 @@ static void hyscale_fast_vsx(SwsContext *c, int16_t *dst, int dstWidth, xInc * 6, xInc * 7, }; - const vector uint32_t vshift16 = vec_splats((uint32_t) 16); - const vector uint16_t vshift9 = vec_splat_u16(9); - const vector uint8_t vzero = vec_splat_u8(0); - const vector uint16_t vshift = vec_splat_u16(7); + const vec_u32 vshift16 = vec_splats((uint32_t) 16); + const vec_u16 vshift9 = vec_splat_u16(9); + const vec_u8 vzero = vec_splat_u8(0); + const vec_u16 vshift = vec_splat_u16(7); for (i = 0; i < dstWidth; i += 16) { vcoord16[0] = vec_splats((uint16_t) xpos); @@ -1701,7 +1701,7 @@ static void hyscale_fast_vsx(SwsContext *c, int16_t *dst, int dstWidth, vcoord16[0] = vec_sr(vcoord16[0], vshift9); vcoord16[1] = vec_sr(vcoord16[1], vshift9); - valpha = (vector int8_t) vec_pack(vcoord16[0], vcoord16[1]); + valpha = (vec_s8) vec_pack(vcoord16[0], vcoord16[1]); xx = xpos >> 16; vin = vec_vsx_ld(0, &src[xx]); @@ -1730,22 +1730,22 @@ static void hyscale_fast_vsx(SwsContext *c, int16_t *dst, int dstWidth, vin2 = vec_vsx_ld(1, &src[xx]); vin2 = vec_perm(vin2, vin2, vperm); - vmul = (vector int8_t) vec_sub(vin2, vin); + vmul = (vec_s8) vec_sub(vin2, vin); vtmp = vec_mule(vmul, valpha); vtmp2 = vec_mulo(vmul, valpha); vtmp3 = vec_mergeh(vtmp, vtmp2); vtmp4 = vec_mergel(vtmp, vtmp2); - vd_l = (vector uint16_t) vec_mergeh(vin, vzero); - vd_r = (vector uint16_t) vec_mergel(vin, vzero); + vd_l = (vec_u16) vec_mergeh(vin, vzero); + vd_r = (vec_u16) vec_mergel(vin, vzero); vd_l = vec_sl(vd_l, vshift); vd_r = vec_sl(vd_r, vshift); - vd_l = vec_add(vd_l, (vector uint16_t) vtmp3); - vd_r = vec_add(vd_r, (vector uint16_t) vtmp4); + vd_l = vec_add(vd_l, (vec_u16) vtmp3); + vd_r = vec_add(vd_r, (vec_u16) vtmp4); - vec_st((vector int16_t) vd_l, 0, &dst[i]); - vec_st((vector int16_t) vd_r, 0, &dst[i + 8]); + vec_st((vec_s16) vd_l, 0, &dst[i]); + vec_st((vec_s16) vd_r, 0, &dst[i + 8]); xpos += xInc * 16; } @@ -1773,8 +1773,8 @@ static void hyscale_fast_vsx(SwsContext *c, int16_t *dst, int dstWidth, vd_l = vec_add(vd_l, vtmp3); \ vd_r = vec_add(vd_r, vtmp4); \ \ - vec_st((vector int16_t) vd_l, 0, &out[i]); \ - vec_st((vector int16_t) vd_r, 0, &out[i + 8]) + vec_st((vec_s16) vd_l, 0, &out[i]); \ + vec_st((vec_s16) vd_r, 0, &out[i + 8]) static void hcscale_fast_vsx(SwsContext *c, int16_t *dst1, int16_t *dst2, int dstWidth, const uint8_t *src1, @@ -1782,19 +1782,19 @@ static void hcscale_fast_vsx(SwsContext *c, int16_t *dst1, int16_t *dst2, { int i; unsigned int xpos = 0, xx; - vector uint8_t vin, vin2, vperm; - vector uint8_t valpha, valphaxor; - vector uint16_t vtmp, vtmp2, vtmp3, vtmp4; - vector uint16_t vd_l, vd_r, vcoord16[2]; - vector uint32_t vcoord[4]; - const vector uint8_t vxor = vec_splats((uint8_t) 127); - const vector uint32_t vadd = (vector uint32_t) { + vec_u8 vin, vin2, vperm; + vec_u8 valpha, valphaxor; + vec_u16 vtmp, vtmp2, vtmp3, vtmp4; + vec_u16 vd_l, vd_r, vcoord16[2]; + vec_u32 vcoord[4]; + const vec_u8 vxor = vec_splats((uint8_t) 127); + const vec_u32 vadd = (vec_u32) { 0, xInc * 1, xInc * 2, xInc * 3, }; - const vector uint16_t vadd16 = (vector uint16_t) { // Modulo math + const vec_u16 vadd16 = (vec_u16) { // Modulo math 0, xInc * 1, xInc * 2, @@ -1804,8 +1804,8 @@ static void hcscale_fast_vsx(SwsContext *c, int16_t *dst1, int16_t *dst2, xInc * 6, xInc * 7, }; - const vector uint32_t vshift16 = vec_splats((uint32_t) 16); - const vector uint16_t vshift9 = vec_splat_u16(9); + const vec_u32 vshift16 = vec_splats((uint32_t) 16); + const vec_u16 vshift9 = vec_splat_u16(9); for (i = 0; i < dstWidth; i += 16) { vcoord16[0] = vec_splats((uint16_t) xpos); @@ -1859,29 +1859,29 @@ static void hScale8To19_vsx(SwsContext *c, int16_t *_dst, int dstW, { int i, j; int32_t *dst = (int32_t *) _dst; - vector int16_t vfilter, vin; - vector uint8_t vin8; - vector int32_t vout; - const vector uint8_t vzero = vec_splat_u8(0); - const vector uint8_t vunusedtab[8] = { - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe, 0xf}, - (vector uint8_t) {0x0, 0x1, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x10, 0x10, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0x10, 0x10}, + vec_s16 vfilter, vin; + vec_u8 vin8; + vec_s32 vout; + const vec_u8 vzero = vec_splat_u8(0); + const vec_u8 vunusedtab[8] = { + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe, 0xf}, + (vec_u8) {0x0, 0x1, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0x10, 0x10}, }; - const vector uint8_t vunused = vunusedtab[filterSize % 8]; + const vec_u8 vunused = vunusedtab[filterSize % 8]; if (filterSize == 1) { for (i = 0; i < dstW; i++) { @@ -1898,14 +1898,14 @@ static void hScale8To19_vsx(SwsContext *c, int16_t *_dst, int dstW, vout = vec_splat_s32(0); for (j = 0; j < filterSize; j += 8) { vin8 = vec_vsx_ld(0, &src[srcPos + j]); - vin = (vector int16_t) vec_mergeh(vin8, vzero); + vin = (vec_s16) vec_mergeh(vin8, vzero); if (j + 8 > filterSize) // Remove the unused elements on the last round - vin = vec_perm(vin, (vector int16_t) vzero, vunused); + vin = vec_perm(vin, (vec_s16) vzero, vunused); vfilter = vec_vsx_ld(0, &filter[filterSize * i + j]); vout = vec_msums(vin, vfilter, vout); } - vout = vec_sums(vout, (vector int32_t) vzero); + vout = vec_sums(vout, (vec_s32) vzero); dst[i] = FFMIN(vout[3] >> 3, (1 << 19) - 1); } } @@ -1921,28 +1921,28 @@ static void hScale16To19_vsx(SwsContext *c, int16_t *_dst, int dstW, const uint16_t *src = (const uint16_t *) _src; int bits = desc->comp[0].depth - 1; int sh = bits - 4; - vector int16_t vfilter, vin; - vector int32_t vout, vtmp, vtmp2, vfilter32_l, vfilter32_r; - const vector uint8_t vzero = vec_splat_u8(0); - const vector uint8_t vunusedtab[8] = { - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe, 0xf}, - (vector uint8_t) {0x0, 0x1, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x10, 0x10, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0x10, 0x10}, + vec_s16 vfilter, vin; + vec_s32 vout, vtmp, vtmp2, vfilter32_l, vfilter32_r; + const vec_u8 vzero = vec_splat_u8(0); + const vec_u8 vunusedtab[8] = { + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe, 0xf}, + (vec_u8) {0x0, 0x1, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0x10, 0x10}, }; - const vector uint8_t vunused = vunusedtab[filterSize % 8]; + const vec_u8 vunused = vunusedtab[filterSize % 8]; if ((isAnyRGB(c->srcFormat) || c->srcFormat==AV_PIX_FMT_PAL8) && desc->comp[0].depth<16) { sh = 9; @@ -1966,16 +1966,16 @@ static void hScale16To19_vsx(SwsContext *c, int16_t *_dst, int dstW, const int srcPos = filterPos[i]; vout = vec_splat_s32(0); for (j = 0; j < filterSize; j += 8) { - vin = (vector int16_t) vec_vsx_ld(0, &src[srcPos + j]); + vin = (vec_s16) vec_vsx_ld(0, &src[srcPos + j]); if (j + 8 > filterSize) // Remove the unused elements on the last round - vin = vec_perm(vin, (vector int16_t) vzero, vunused); + vin = vec_perm(vin, (vec_s16) vzero, vunused); vfilter = vec_vsx_ld(0, &filter[filterSize * i + j]); vfilter32_l = vec_unpackh(vfilter); vfilter32_r = vec_unpackl(vfilter); - vtmp = (vector int32_t) vec_mergeh(vin, (vector int16_t) vzero); - vtmp2 = (vector int32_t) vec_mergel(vin, (vector int16_t) vzero); + vtmp = (vec_s32) vec_mergeh(vin, (vec_s16) vzero); + vtmp2 = (vec_s32) vec_mergel(vin, (vec_s16) vzero); vtmp = vec_mul(vtmp, vfilter32_l); vtmp2 = vec_mul(vtmp2, vfilter32_r); @@ -1983,7 +1983,7 @@ static void hScale16To19_vsx(SwsContext *c, int16_t *_dst, int dstW, vout = vec_adds(vout, vtmp); vout = vec_adds(vout, vtmp2); } - vout = vec_sums(vout, (vector int32_t) vzero); + vout = vec_sums(vout, (vec_s32) vzero); dst[i] = FFMIN(vout[3] >> sh, (1 << 19) - 1); } } @@ -1997,28 +1997,28 @@ static void hScale16To15_vsx(SwsContext *c, int16_t *dst, int dstW, int i, j; const uint16_t *src = (const uint16_t *) _src; int sh = desc->comp[0].depth - 1; - vector int16_t vfilter, vin; - vector int32_t vout, vtmp, vtmp2, vfilter32_l, vfilter32_r; - const vector uint8_t vzero = vec_splat_u8(0); - const vector uint8_t vunusedtab[8] = { - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe, 0xf}, - (vector uint8_t) {0x0, 0x1, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x10, 0x10, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x10, 0x10, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0x10, 0x10, 0x10, 0x10}, - (vector uint8_t) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, - 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0x10, 0x10}, + vec_s16 vfilter, vin; + vec_s32 vout, vtmp, vtmp2, vfilter32_l, vfilter32_r; + const vec_u8 vzero = vec_splat_u8(0); + const vec_u8 vunusedtab[8] = { + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe, 0xf}, + (vec_u8) {0x0, 0x1, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0x10, 0x10, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0x10, 0x10, 0x10, 0x10}, + (vec_u8) {0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7, + 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0x10, 0x10}, }; - const vector uint8_t vunused = vunusedtab[filterSize % 8]; + const vec_u8 vunused = vunusedtab[filterSize % 8]; if (sh<15) { sh = isAnyRGB(c->srcFormat) || c->srcFormat==AV_PIX_FMT_PAL8 ? 13 : (desc->comp[0].depth - 1); @@ -2042,16 +2042,16 @@ static void hScale16To15_vsx(SwsContext *c, int16_t *dst, int dstW, const int srcPos = filterPos[i]; vout = vec_splat_s32(0); for (j = 0; j < filterSize; j += 8) { - vin = (vector int16_t) vec_vsx_ld(0, &src[srcPos + j]); + vin = (vec_s16) vec_vsx_ld(0, &src[srcPos + j]); if (j + 8 > filterSize) // Remove the unused elements on the last round - vin = vec_perm(vin, (vector int16_t) vzero, vunused); + vin = vec_perm(vin, (vec_s16) vzero, vunused); vfilter = vec_vsx_ld(0, &filter[filterSize * i + j]); vfilter32_l = vec_unpackh(vfilter); vfilter32_r = vec_unpackl(vfilter); - vtmp = (vector int32_t) vec_mergeh(vin, (vector int16_t) vzero); - vtmp2 = (vector int32_t) vec_mergel(vin, (vector int16_t) vzero); + vtmp = (vec_s32) vec_mergeh(vin, (vec_s16) vzero); + vtmp2 = (vec_s32) vec_mergel(vin, (vec_s16) vzero); vtmp = vec_mul(vtmp, vfilter32_l); vtmp2 = vec_mul(vtmp2, vfilter32_r); @@ -2059,7 +2059,7 @@ static void hScale16To15_vsx(SwsContext *c, int16_t *dst, int dstW, vout = vec_adds(vout, vtmp); vout = vec_adds(vout, vtmp2); } - vout = vec_sums(vout, (vector int32_t) vzero); + vout = vec_sums(vout, (vec_s32) vzero); dst[i] = FFMIN(vout[3] >> sh, (1 << 15) - 1); } }

1 0

avformat/matroskadec: Fix demuxing ProRes
by Andreas Rheinhardt 04 Oct '19

04 Oct '19

ffmpeg | branch: master | Andreas Rheinhardt <andreas.rheinhardt(a)gmail.com> | Sat Sep 28 19:54:25 2019 +0200| [581419ea39de6619c3389b8d10ac2cbe212c62a0] | committer: James Almer avformat/matroskadec: Fix demuxing ProRes The structure of a ProRes frame in mov/mp4 is that of a typical atom: First a 32 bit BE size field, then a tag detailling the content. Said size field includes the eight bytes of the atom header. This header is actually redundant, as the size of the atom is already known from the containing atom. It is therefore stripped away when muxed into Matroska and so the Matroska demuxer has to recreate upon demuxing. But it did not account for the fact that the size field includes the size of the header and this can lead to problems when a decoder uses the in-band size field. Fixes ticket #8210. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt(a)gmail.com> Signed-off-by: James Almer <jamrial(a)gmail.com> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=581419ea39de6619c… --- libavformat/matroskadec.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c index 10c398856b..a5f120b54d 100644 --- a/libavformat/matroskadec.c +++ b/libavformat/matroskadec.c @@ -3280,15 +3280,16 @@ static int matroska_parse_prores(MatroskaTrack *track, uint8_t *src, int dstlen = *size; if (AV_RB32(&src[4]) != MKBETAG('i', 'c', 'p', 'f')) { - dst = av_malloc(dstlen + 8 + AV_INPUT_BUFFER_PADDING_SIZE); + dstlen += 8; + + dst = av_malloc(dstlen + AV_INPUT_BUFFER_PADDING_SIZE); if (!dst) return AVERROR(ENOMEM); AV_WB32(dst, dstlen); AV_WB32(dst + 4, MKBETAG('i', 'c', 'p', 'f')); - memcpy(dst + 8, src, dstlen); - memset(dst + 8 + dstlen, 0, AV_INPUT_BUFFER_PADDING_SIZE); - dstlen += 8; + memcpy(dst + 8, src, dstlen - 8); + memset(dst + dstlen, 0, AV_INPUT_BUFFER_PADDING_SIZE); } *pdst = dst;

1 0