hdkv/yuzu
1
Fork 0
Code Issues Pull Requests Releases 1 Activity

early-access version 1680

Browse Source
This commit is contained in:
pineappleEA
2021-05-13 11:45:27 +02:00
parent 1434d96e7d
commit 66ed389c6f
311 changed files with 6452 additions and 2597 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
externals/ffmpeg/libswscale/aarch64/swscale_unscaled.c vendored
View File

@@ -42,15 +42,14 @@ static int ifmt##_to_##ofmt##_neon_wrapper(SwsContext *c, const uint8_t *src[],
uint8_t *dst[], int dstStride[]) { \
const int16_t yuv2rgb_table[] = { YUV_TO_RGB_TABLE }; \
\
ff_##ifmt##_to_##ofmt##_neon(c->srcW, srcSliceH, \
dst[0] + srcSliceY * dstStride[0], dstStride[0], \
src[0], srcStride[0], \
src[1], srcStride[1], \
src[2], srcStride[2], \
yuv2rgb_table, \
c->yuv2rgb_y_offset >> 6, \
c->yuv2rgb_y_coeff); \
return 0; \
return ff_##ifmt##_to_##ofmt##_neon(c->srcW, srcSliceH, \
dst[0] + srcSliceY * dstStride[0], dstStride[0], \
src[0], srcStride[0], \
src[1], srcStride[1], \
src[2], srcStride[2], \
yuv2rgb_table, \
c->yuv2rgb_y_offset >> 6, \
c->yuv2rgb_y_coeff); \
} \
#define DECLARE_FF_YUVX_TO_ALL_RGBX_FUNCS(yuvx) \
@@ -76,14 +75,12 @@ static int ifmt##_to_##ofmt##_neon_wrapper(SwsContext *c, const uint8_t *src[],
uint8_t *dst[], int dstStride[]) { \
const int16_t yuv2rgb_table[] = { YUV_TO_RGB_TABLE }; \
\
ff_##ifmt##_to_##ofmt##_neon(c->srcW, srcSliceH, \
dst[0] + srcSliceY * dstStride[0], dstStride[0], \
src[0], srcStride[0], src[1], srcStride[1], \
yuv2rgb_table, \
c->yuv2rgb_y_offset >> 6, \
c->yuv2rgb_y_coeff); \
\
return 0; \
return ff_##ifmt##_to_##ofmt##_neon(c->srcW, srcSliceH, \
dst[0] + srcSliceY * dstStride[0], dstStride[0], \
src[0], srcStride[0], src[1], srcStride[1], \
yuv2rgb_table, \
c->yuv2rgb_y_offset >> 6, \
c->yuv2rgb_y_coeff); \
} \
#define DECLARE_FF_NVX_TO_ALL_RGBX_FUNCS(nvx) \

2
externals/ffmpeg/libswscale/aarch64/yuv2rgb_neon.S vendored
View File

@@ -142,6 +142,7 @@
.macro declare_func ifmt ofmt
function ff_\ifmt\()_to_\ofmt\()_neon, export=1
load_args_\ifmt
mov w9, w1
1:
mov w8, w0 // w8 = width
2:
@@ -193,6 +194,7 @@ function ff_\ifmt\()_to_\ofmt\()_neon, export=1
increment_\ifmt
subs w1, w1, #1 // height -= 1
b.gt 1b
mov w0, w9
ret
endfunc
.endm

13
externals/ffmpeg/libswscale/input.c vendored
View File

@@ -244,7 +244,8 @@ rgb48funcs(bgr, BE, AV_PIX_FMT_BGR48BE)
#define input_pixel(i) ((origin == AV_PIX_FMT_RGBA || \
origin == AV_PIX_FMT_BGRA || \
origin == AV_PIX_FMT_ARGB || \
origin == AV_PIX_FMT_ABGR) \
origin == AV_PIX_FMT_ABGR || \
origin == AV_PIX_FMT_X2RGB10) \
? AV_RN32A(&src[(i) * 4]) \
: (isBE(origin) ? AV_RB16(&src[(i) * 2]) \
: AV_RL16(&src[(i) * 2])))
@@ -391,6 +392,7 @@ rgb16_32_wrapper(AV_PIX_FMT_BGR444BE, bgr12be, 0, 0, 0, 0, 0x000F, 0x00F0,
rgb16_32_wrapper(AV_PIX_FMT_RGB565BE, rgb16be, 0, 0, 0, 0, 0xF800, 0x07E0, 0x001F, 0, 5, 11, RGB2YUV_SHIFT + 8)
rgb16_32_wrapper(AV_PIX_FMT_RGB555BE, rgb15be, 0, 0, 0, 0, 0x7C00, 0x03E0, 0x001F, 0, 5, 10, RGB2YUV_SHIFT + 7)
rgb16_32_wrapper(AV_PIX_FMT_RGB444BE, rgb12be, 0, 0, 0, 0, 0x0F00, 0x00F0, 0x000F, 0, 4, 8, RGB2YUV_SHIFT + 4)
rgb16_32_wrapper(AV_PIX_FMT_X2RGB10LE, rgb30le, 16, 6, 0, 0, 0x3FF00000, 0xFFC00, 0x3FF, 0, 0, 4, RGB2YUV_SHIFT + 6)
static void gbr24pToUV_half_c(uint8_t *_dstU, uint8_t *_dstV,
const uint8_t *gsrc, const uint8_t *bsrc, const uint8_t *rsrc,
@@ -1341,6 +1343,9 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
case AV_PIX_FMT_RGB444BE:
c->chrToYV12 = rgb12beToUV_half_c;
break;
case AV_PIX_FMT_X2RGB10LE:
c->chrToYV12 = rgb30leToUV_half_c;
break;
}
} else {
switch (srcFormat) {
@@ -1422,6 +1427,9 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
case AV_PIX_FMT_RGB444BE:
c->chrToYV12 = rgb12beToUV_c;
break;
case AV_PIX_FMT_X2RGB10LE:
c->chrToYV12 = rgb30leToUV_c;
break;
}
}
@@ -1701,6 +1709,9 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
case AV_PIX_FMT_Y210LE:
c->lumToYV12 = y210le_Y_c;
break;
case AV_PIX_FMT_X2RGB10LE:
c->lumToYV12 =rgb30leToY_c;
break;
}
if (c->needAlpha) {
if (is16BPS(srcFormat) || isNBPS(srcFormat)) {

33
externals/ffmpeg/libswscale/output.c vendored
View File

@@ -180,7 +180,8 @@ yuv2planeX_16_c_template(const int16_t *filter, int filterSize,
}
}
static void yuv2p016cX_c(SwsContext *c, const int16_t *chrFilter, int chrFilterSize,
static void yuv2p016cX_c(enum AVPixelFormat dstFormat, const uint8_t *chrDither,
const int16_t *chrFilter, int chrFilterSize,
const int16_t **chrUSrc, const int16_t **chrVSrc,
uint8_t *dest8, int chrDstW)
{
@@ -188,7 +189,7 @@ static void yuv2p016cX_c(SwsContext *c, const int16_t *chrFilter, int chrFilterS
const int32_t **uSrc = (const int32_t **)chrUSrc;
const int32_t **vSrc = (const int32_t **)chrVSrc;
int shift = 15;
int big_endian = c->dstFormat == AV_PIX_FMT_P016BE;
int big_endian = dstFormat == AV_PIX_FMT_P016BE;
int i, j;
for (i = 0; i < chrDstW; i++) {
@@ -402,12 +403,11 @@ static void yuv2plane1_8_c(const int16_t *src, uint8_t *dest, int dstW,
}
}
static void yuv2nv12cX_c(SwsContext *c, const int16_t *chrFilter, int chrFilterSize,
const int16_t **chrUSrc, const int16_t **chrVSrc,
uint8_t *dest, int chrDstW)
static void yuv2nv12cX_c(enum AVPixelFormat dstFormat, const uint8_t *chrDither,
const int16_t *chrFilter, int chrFilterSize,
const int16_t **chrUSrc, const int16_t **chrVSrc,
uint8_t *dest, int chrDstW)
{
enum AVPixelFormat dstFormat = c->dstFormat;
const uint8_t *chrDither = c->chrDither8;
int i;
if (dstFormat == AV_PIX_FMT_NV12 ||
@@ -477,13 +477,14 @@ static void yuv2p010lX_c(const int16_t *filter, int filterSize,
}
}
static void yuv2p010cX_c(SwsContext *c, const int16_t *chrFilter, int chrFilterSize,
static void yuv2p010cX_c(enum AVPixelFormat dstFormat, const uint8_t *chrDither,
const int16_t *chrFilter, int chrFilterSize,
const int16_t **chrUSrc, const int16_t **chrVSrc,
uint8_t *dest8, int chrDstW)
{
uint16_t *dest = (uint16_t*)dest8;
int shift = 17;
int big_endian = c->dstFormat == AV_PIX_FMT_P010BE;
int big_endian = dstFormat == AV_PIX_FMT_P010BE;
int i, j;
for (i = 0; i < chrDstW; i++) {
@@ -1602,6 +1603,13 @@ yuv2rgb_write(uint8_t *_dest, int i, int Y1, int Y2,
dest[i * 2 + 0] = r[Y1 + dr1] + g[Y1 + dg1] + b[Y1 + db1];
dest[i * 2 + 1] = r[Y2 + dr2] + g[Y2 + dg2] + b[Y2 + db2];
} else if (target == AV_PIX_FMT_X2RGB10) {
uint32_t *dest = (uint32_t *) _dest;
const uint32_t *r = (const uint32_t *) _r;
const uint32_t *g = (const uint32_t *) _g;
const uint32_t *b = (const uint32_t *) _b;
dest[i * 2 + 0] = r[Y1] + g[Y1] + b[Y1];
dest[i * 2 + 1] = r[Y2] + g[Y2] + b[Y2];
} else /* 8/4 bits */ {
uint8_t *dest = (uint8_t *) _dest;
const uint8_t *r = (const uint8_t *) _r;
@@ -1839,6 +1847,7 @@ YUV2RGBWRAPPER(yuv2rgb,, 12, AV_PIX_FMT_RGB444, 0)
YUV2RGBWRAPPER(yuv2rgb,, 8, AV_PIX_FMT_RGB8, 0)
YUV2RGBWRAPPER(yuv2rgb,, 4, AV_PIX_FMT_RGB4, 0)
YUV2RGBWRAPPER(yuv2rgb,, 4b, AV_PIX_FMT_RGB4_BYTE, 0)
YUV2RGBWRAPPER(yuv2, rgb, x2rgb10, AV_PIX_FMT_X2RGB10, 0)
static av_always_inline void yuv2rgb_write_full(SwsContext *c,
uint8_t *dest, int i, int Y, int A, int U, int V,
@@ -2974,6 +2983,12 @@ av_cold void ff_sws_init_output_funcs(SwsContext *c,
*yuv2packed2 = yuv2rgb4b_2_c;
*yuv2packedX = yuv2rgb4b_X_c;
break;
case AV_PIX_FMT_X2RGB10LE:
case AV_PIX_FMT_X2RGB10BE:
*yuv2packed1 = yuv2x2rgb10_1_c;
*yuv2packed2 = yuv2x2rgb10_2_c;
*yuv2packedX = yuv2x2rgb10_X_c;
break;
}
}
switch (dstFormat) {

6
externals/ffmpeg/libswscale/swscale_internal.h vendored
View File

@@ -119,7 +119,8 @@ typedef void (*yuv2planarX_fn)(const int16_t *filter, int filterSize,
* Write one line of horizontally scaled chroma to interleaved output
* with multi-point vertical scaling between input pixels.
*
* @param c SWS scaling context
* @param dstFormat destination pixel format
* @param chrDither ordered dither array of type uint8_t and size 8
* @param chrFilter vertical chroma scaling coefficients, 12 bits [0,4096]
* @param chrUSrc scaled chroma (U) source data, 15 bits for 8-10-bit
* output, 19 bits for 16-bit output (in int32_t)
@@ -130,7 +131,8 @@ typedef void (*yuv2planarX_fn)(const int16_t *filter, int filterSize,
* output, this is in uint16_t
* @param dstW width of chroma planes
*/
typedef void (*yuv2interleavedX_fn)(struct SwsContext *c,
typedef void (*yuv2interleavedX_fn)(enum AVPixelFormat dstFormat,
const uint8_t *chrDither,
const int16_t *chrFilter,
int chrFilterSize,
const int16_t **chrUSrc,

4
externals/ffmpeg/libswscale/tests/swscale.c vendored
View File

@@ -422,7 +422,9 @@ bad_option:
for (y = 0; y < H; y++)
for (x = 0; x < W * 4; x++)
rgb_data[ x + y * 4 * W] = av_lfg_get(&rand);
sws_scale(sws, rgb_src, rgb_stride, 0, H / 12, (uint8_t * const *) src, stride);
res = sws_scale(sws, rgb_src, rgb_stride, 0, H / 12, (uint8_t * const *) src, stride);
if (res < 0 || res != (H / 12))
goto error;
sws_freeContext(sws);
av_free(rgb_data);

142
externals/ffmpeg/libswscale/utils.c vendored
View File

@@ -271,6 +271,7 @@ static const FormatEntry format_entries[] = {
[AV_PIX_FMT_NV24] = { 1, 1 },
[AV_PIX_FMT_NV42] = { 1, 1 },
[AV_PIX_FMT_Y210LE] = { 1, 0 },
[AV_PIX_FMT_X2RGB10LE] = { 1, 1 },
};
int sws_isSupportedInput(enum AVPixelFormat pix_fmt)
@@ -352,13 +353,14 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
emms_c(); // FIXME should not be required but IS (even for non-MMX versions)
// NOTE: the +3 is for the MMX(+1) / SSE(+3) scaler which reads over the end
FF_ALLOC_ARRAY_OR_GOTO(NULL, *filterPos, (dstW + 3), sizeof(**filterPos), fail);
if (!FF_ALLOC_TYPED_ARRAY(*filterPos, dstW + 3))
goto nomem;
if (FFABS(xInc - 0x10000) < 10 && srcPos == dstPos) { // unscaled
int i;
filterSize = 1;
FF_ALLOCZ_ARRAY_OR_GOTO(NULL, filter,
dstW, sizeof(*filter) * filterSize, fail);
if (!FF_ALLOCZ_TYPED_ARRAY(filter, dstW * filterSize))
goto nomem;
for (i = 0; i < dstW; i++) {
filter[i * filterSize] = fone;
@@ -368,8 +370,8 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
int i;
int64_t xDstInSrc;
filterSize = 1;
FF_ALLOC_ARRAY_OR_GOTO(NULL, filter,
dstW, sizeof(*filter) * filterSize, fail);
if (!FF_ALLOC_TYPED_ARRAY(filter, dstW * filterSize))
goto nomem;
xDstInSrc = ((dstPos*(int64_t)xInc)>>8) - ((srcPos*0x8000LL)>>7);
for (i = 0; i < dstW; i++) {
@@ -384,8 +386,8 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
int i;
int64_t xDstInSrc;
filterSize = 2;
FF_ALLOC_ARRAY_OR_GOTO(NULL, filter,
dstW, sizeof(*filter) * filterSize, fail);
if (!FF_ALLOC_TYPED_ARRAY(filter, dstW * filterSize))
goto nomem;
xDstInSrc = ((dstPos*(int64_t)xInc)>>8) - ((srcPos*0x8000LL)>>7);
for (i = 0; i < dstW; i++) {
@@ -425,9 +427,8 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
filterSize = FFMIN(filterSize, srcW - 2);
filterSize = FFMAX(filterSize, 1);
FF_ALLOC_ARRAY_OR_GOTO(NULL, filter,
dstW, sizeof(*filter) * filterSize, fail);
if (!FF_ALLOC_TYPED_ARRAY(filter, dstW * filterSize))
goto nomem;
xDstInSrc = ((dstPos*(int64_t)xInc)>>7) - ((srcPos*0x10000LL)>>7);
for (i = 0; i < dstW; i++) {
int xx = (xDstInSrc - (filterSize - 2) * (1LL<<16)) / (1 << 17);
@@ -525,8 +526,8 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
if (dstFilter)
filter2Size += dstFilter->length - 1;
av_assert0(filter2Size > 0);
FF_ALLOCZ_ARRAY_OR_GOTO(NULL, filter2, dstW, filter2Size * sizeof(*filter2), fail);
if (!FF_ALLOCZ_TYPED_ARRAY(filter2, dstW * filter2Size))
goto nomem;
for (i = 0; i < dstW; i++) {
int j, k;
@@ -611,7 +612,7 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
av_assert0(filterSize > 0);
filter = av_malloc_array(dstW, filterSize * sizeof(*filter));
if (!filter)
goto fail;
goto nomem;
if (filterSize >= MAX_FILTER_SIZE * 16 /
((flags & SWS_ACCURATE_RND) ? APCK_SIZE : 16)) {
ret = RETCODE_USE_CASCADE;
@@ -684,8 +685,8 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
// Note the +1 is for the MMX scaler which reads over the end
/* align at 16 for AltiVec (needed by hScale_altivec_real) */
FF_ALLOCZ_ARRAY_OR_GOTO(NULL, *outFilter,
(dstW + 3), *outFilterSize * sizeof(int16_t), fail);
if (!FF_ALLOCZ_TYPED_ARRAY(*outFilter, *outFilterSize * (dstW + 3)))
goto nomem;
/* normalize & store in outFilter */
for (i = 0; i < dstW; i++) {
@@ -721,10 +722,13 @@ static av_cold int initFilter(int16_t **outFilter, int32_t **filterPos,
}
ret = 0;
goto done;
nomem:
ret = AVERROR(ENOMEM);
fail:
if(ret < 0)
av_log(NULL, ret == RETCODE_USE_CASCADE ? AV_LOG_DEBUG : AV_LOG_ERROR, "sws: initFilter failed\n");
done:
av_free(filter);
av_free(filter2);
return ret;
@@ -1417,7 +1421,8 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
c->chrDstW = AV_CEIL_RSHIFT(dstW, c->chrDstHSubSample);
c->chrDstH = AV_CEIL_RSHIFT(dstH, c->chrDstVSubSample);
FF_ALLOCZ_OR_GOTO(c, c->formatConvBuffer, FFALIGN(srcW*2+78, 16) * 2, fail);
if (!FF_ALLOCZ_TYPED_ARRAY(c->formatConvBuffer, FFALIGN(srcW * 2 + 78, 16) * 2))
goto nomem;
c->srcBpc = desc_src->comp[0].depth;
if (c->srcBpc < 8)
@@ -1486,7 +1491,7 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
srcW, srcH, tmpFmt,
flags, NULL, NULL, c->param);
if (!c->cascaded_context[0]) {
return -1;
return AVERROR(ENOMEM);
}
c->cascaded_context[1] = sws_getContext(srcW, srcH, tmpFmt,
@@ -1494,7 +1499,7 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
flags, srcFilter, dstFilter, c->param);
if (!c->cascaded_context[1])
return -1;
return AVERROR(ENOMEM);
c2 = c->cascaded_context[1];
c2->is_internal_gamma = 1;
@@ -1507,10 +1512,10 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
// to properly create the gamma convert FilterDescriptor
// we have to re-initialize it
ff_free_filters(c2);
if (ff_init_filters(c2) < 0) {
if ((ret = ff_init_filters(c2)) < 0) {
sws_freeContext(c2);
c->cascaded_context[1] = NULL;
return -1;
return ret;
}
c->cascaded_context[2] = NULL;
@@ -1524,7 +1529,7 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
dstW, dstH, dstFormat,
flags, NULL, NULL, c->param);
if (!c->cascaded_context[2])
return -1;
return AVERROR(ENOMEM);
}
return 0;
}
@@ -1543,13 +1548,13 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
srcW, srcH, tmpFormat,
flags, srcFilter, NULL, c->param);
if (!c->cascaded_context[0])
return -1;
return AVERROR(ENOMEM);
c->cascaded_context[1] = sws_getContext(srcW, srcH, tmpFormat,
dstW, dstH, dstFormat,
flags, NULL, dstFilter, c->param);
if (!c->cascaded_context[1])
return -1;
return AVERROR(ENOMEM);
return 0;
}
}
@@ -1570,41 +1575,42 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
if (CONFIG_SWSCALE_ALPHA && isALPHA(srcFormat) && !isALPHA(dstFormat)) {
enum AVPixelFormat tmpFormat = alphaless_fmt(srcFormat);
if (tmpFormat != AV_PIX_FMT_NONE && c->alphablend != SWS_ALPHA_BLEND_NONE)
if (!unscaled ||
dstFormat != tmpFormat ||
usesHFilter || usesVFilter ||
c->srcRange != c->dstRange
) {
c->cascaded_mainindex = 1;
ret = av_image_alloc(c->cascaded_tmp, c->cascaded_tmpStride,
srcW, srcH, tmpFormat, 64);
if (ret < 0)
return ret;
if (tmpFormat != AV_PIX_FMT_NONE && c->alphablend != SWS_ALPHA_BLEND_NONE) {
if (!unscaled ||
dstFormat != tmpFormat ||
usesHFilter || usesVFilter ||
c->srcRange != c->dstRange
) {
c->cascaded_mainindex = 1;
ret = av_image_alloc(c->cascaded_tmp, c->cascaded_tmpStride,
srcW, srcH, tmpFormat, 64);
if (ret < 0)
return ret;
c->cascaded_context[0] = sws_alloc_set_opts(srcW, srcH, srcFormat,
srcW, srcH, tmpFormat,
flags, c->param);
if (!c->cascaded_context[0])
return -1;
c->cascaded_context[0]->alphablend = c->alphablend;
ret = sws_init_context(c->cascaded_context[0], NULL , NULL);
if (ret < 0)
return ret;
c->cascaded_context[0] = sws_alloc_set_opts(srcW, srcH, srcFormat,
srcW, srcH, tmpFormat,
flags, c->param);
if (!c->cascaded_context[0])
return AVERROR(EINVAL);
c->cascaded_context[0]->alphablend = c->alphablend;
ret = sws_init_context(c->cascaded_context[0], NULL , NULL);
if (ret < 0)
return ret;
c->cascaded_context[1] = sws_alloc_set_opts(srcW, srcH, tmpFormat,
dstW, dstH, dstFormat,
flags, c->param);
if (!c->cascaded_context[1])
return -1;
c->cascaded_context[1] = sws_alloc_set_opts(srcW, srcH, tmpFormat,
dstW, dstH, dstFormat,
flags, c->param);
if (!c->cascaded_context[1])
return AVERROR(EINVAL);
c->cascaded_context[1]->srcRange = c->srcRange;
c->cascaded_context[1]->dstRange = c->dstRange;
ret = sws_init_context(c->cascaded_context[1], srcFilter , dstFilter);
if (ret < 0)
return ret;
c->cascaded_context[1]->srcRange = c->srcRange;
c->cascaded_context[1]->dstRange = c->dstRange;
ret = sws_init_context(c->cascaded_context[1], srcFilter , dstFilter);
if (ret < 0)
return ret;
return 0;
return 0;
}
}
}
@@ -1657,10 +1663,11 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
return AVERROR(ENOMEM);
}
FF_ALLOCZ_OR_GOTO(c, c->hLumFilter, (dstW / 8 + 8) * sizeof(int16_t), fail);
FF_ALLOCZ_OR_GOTO(c, c->hChrFilter, (c->chrDstW / 4 + 8) * sizeof(int16_t), fail);
FF_ALLOCZ_OR_GOTO(c, c->hLumFilterPos, (dstW / 2 / 8 + 8) * sizeof(int32_t), fail);
FF_ALLOCZ_OR_GOTO(c, c->hChrFilterPos, (c->chrDstW / 2 / 4 + 8) * sizeof(int32_t), fail);
if (!FF_ALLOCZ_TYPED_ARRAY(c->hLumFilter, dstW / 8 + 8) ||
!FF_ALLOCZ_TYPED_ARRAY(c->hChrFilter, c->chrDstW / 4 + 8) ||
!FF_ALLOCZ_TYPED_ARRAY(c->hLumFilterPos, dstW / 2 / 8 + 8) ||
!FF_ALLOCZ_TYPED_ARRAY(c->hChrFilterPos, c->chrDstW / 2 / 4 + 8))
goto nomem;
ff_init_hscaler_mmxext( dstW, c->lumXInc, c->lumMmxextFilterCode,
c->hLumFilter, (uint32_t*)c->hLumFilterPos, 8);
@@ -1671,6 +1678,7 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
if ( mprotect(c->lumMmxextFilterCode, c->lumMmxextFilterCodeSize, PROT_EXEC | PROT_READ) == -1
|| mprotect(c->chrMmxextFilterCode, c->chrMmxextFilterCodeSize, PROT_EXEC | PROT_READ) == -1) {
av_log(c, AV_LOG_ERROR, "mprotect failed, cannot use fast bilinear scaler\n");
ret = AVERROR(EINVAL);
goto fail;
}
#endif
@@ -1728,8 +1736,9 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
goto fail;
#if HAVE_ALTIVEC
FF_ALLOC_OR_GOTO(c, c->vYCoeffsBank, sizeof(vector signed short) * c->vLumFilterSize * c->dstH, fail);
FF_ALLOC_OR_GOTO(c, c->vCCoeffsBank, sizeof(vector signed short) * c->vChrFilterSize * c->chrDstH, fail);
if (!FF_ALLOC_TYPED_ARRAY(c->vYCoeffsBank, c->vLumFilterSize * c->dstH) ||
!FF_ALLOC_TYPED_ARRAY(c->vCCoeffsBank, c->vChrFilterSize * c->chrDstH))
goto nomem;
for (i = 0; i < c->vLumFilterSize * c->dstH; i++) {
int j;
@@ -1748,7 +1757,8 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
}
for (i = 0; i < 4; i++)
FF_ALLOCZ_OR_GOTO(c, c->dither_error[i], (c->dstW+2) * sizeof(int), fail);
if (!FF_ALLOCZ_TYPED_ARRAY(c->dither_error[i], c->dstW + 2))
goto nomem;
c->needAlpha = (CONFIG_SWSCALE_ALPHA && isALPHA(c->srcFormat) && isALPHA(c->dstFormat)) ? 1 : 0;
@@ -1838,6 +1848,8 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
c->swscale = ff_getSwsFunc(c);
return ff_init_filters(c);
nomem:
ret = AVERROR(ENOMEM);
fail: // FIXME replace things by appropriate error codes
if (ret == RETCODE_USE_CASCADE) {
int tmpW = sqrt(srcW * (int64_t)dstW);
@@ -1859,16 +1871,16 @@ fail: // FIXME replace things by appropriate error codes
tmpW, tmpH, tmpFormat,
flags, srcFilter, NULL, c->param);
if (!c->cascaded_context[0])
return -1;
return AVERROR(ENOMEM);
c->cascaded_context[1] = sws_getContext(tmpW, tmpH, tmpFormat,
dstW, dstH, dstFormat,
flags, NULL, dstFilter, c->param);
if (!c->cascaded_context[1])
return -1;
return AVERROR(ENOMEM);
return 0;
}
return -1;
return ret;
}
SwsContext *sws_alloc_set_opts(int srcW, int srcH, enum AVPixelFormat srcFormat,

2
externals/ffmpeg/libswscale/version.h vendored
View File

@@ -27,7 +27,7 @@
#include "libavutil/version.h"
#define LIBSWSCALE_VERSION_MAJOR 5
#define LIBSWSCALE_VERSION_MINOR 7
#define LIBSWSCALE_VERSION_MINOR 8
#define LIBSWSCALE_VERSION_MICRO 100
#define LIBSWSCALE_VERSION_INT AV_VERSION_INT(LIBSWSCALE_VERSION_MAJOR, \

2
externals/ffmpeg/libswscale/vscale.c vendored
View File

@@ -92,7 +92,7 @@ static int chr_planar_vscale(SwsContext *c, SwsFilterDescriptor *desc, int slice
uint16_t *filter = inst->filter[0] + (inst->isMMX ? 0 : chrSliceY * inst->filter_size);
if (c->yuv2nv12cX) {
inst->pfn.yuv2interleavedX(c, filter, inst->filter_size, (const int16_t**)src1, (const int16_t**)src2, dst1[0], dstW);
inst->pfn.yuv2interleavedX(c->dstFormat, c->chrDither8, filter, inst->filter_size, (const int16_t**)src1, (const int16_t**)src2, dst1[0], dstW);
} else if (inst->filter_size == 1) {
inst->pfn.yuv2planar1((const int16_t*)src1[0], dst1[0], dstW, c->chrDither8, 0);
inst->pfn.yuv2planar1((const int16_t*)src2[0], dst2[0], dstW, c->chrDither8, 3);

126
externals/ffmpeg/libswscale/x86/output.asm vendored
View File

@@ -2,6 +2,7 @@
;* x86-optimized vertical line scaling functions
;* Copyright (c) 2011 Ronald S. Bultje <rsbultje@gmail.com>
;* Kieran Kunhya <kieran@kunhya.com>
;* (c) 2020 Nelson Gomez <nelson.gomez@microsoft.com>
;*
;* This file is part of FFmpeg.
;*
@@ -22,7 +23,7 @@
%include "libavutil/x86/x86util.asm"
SECTION_RODATA
SECTION_RODATA 32
minshort: times 8 dw 0x8000
yuv2yuvX_16_start: times 4 dd 0x4000 - 0x40000000
@@ -34,9 +35,20 @@ pd_4: times 4 dd 4
pd_4min0x40000:times 4 dd 4 - (0x40000)
pw_16: times 8 dw 16
pw_32: times 8 dw 32
pd_255: times 8 dd 255
pw_512: times 8 dw 512
pw_1024: times 8 dw 1024
yuv2nv12_shuffle_mask: times 2 db 0, 4, 8, 12, \
-1, -1, -1, -1, \
-1, -1, -1, -1, \
-1, -1, -1, -1
yuv2nv21_shuffle_mask: times 2 db 4, 0, 12, 8, \
-1, -1, -1, -1, \
-1, -1, -1, -1, \
-1, -1, -1, -1
yuv2nv12_permute_mask: dd 0, 4, 1, 2, 3, 5, 6, 7
SECTION .text
;-----------------------------------------------------------------------------
@@ -423,3 +435,115 @@ yuv2plane1_fn 9, 5, 3
yuv2plane1_fn 10, 5, 3
yuv2plane1_fn 16, 5, 3
%endif
%undef movsx
;-----------------------------------------------------------------------------
; AVX2 yuv2nv12cX implementation
;
; void ff_yuv2nv12cX_avx2(enum AVPixelFormat format, const uint8_t *dither,
; const int16_t *filter, int filterSize,
; const int16_t **u, const int16_t **v,
; uint8_t *dst, int dstWidth)
;
; void ff_yuv2nv21cX_avx2(enum AVPixelFormat format, const uint8_t *dither,
; const int16_t *filter, int filterSize,
; const int16_t **u, const int16_t **v,
; uint8_t *dst, int dstWidth)
;-----------------------------------------------------------------------------
%if ARCH_X86_64
%macro yuv2nv12cX_fn 1
cglobal %1cX, 8, 11, 13, tmp1, dither, filter, filterSize, u, v, dst, dstWidth
mov tmp1q, qword [ditherq]
movq xm0, tmp1q
ror tmp1q, 24
movq xm1, tmp1q
pmovzxbd m0, xm0
pslld m0, m0, 12 ; ditherLo
pmovzxbd m1, xm1
pslld m1, m1, 12 ; ditherHi
pxor m9, m9 ; uint8_min dwords
mova m10, [pd_255] ; uint8_max dwords
mova m11, [%1_shuffle_mask] ; shuffle_mask
mova m12, [yuv2nv12_permute_mask] ; permute mask
DEFINE_ARGS tmp1, tmp2, filter, filterSize, u, v, dst, dstWidth
xor r8q, r8q
nv12_outer_%1:
mova m2, m0 ; resultLo
mova m3, m1 ; resultHi
xor r9q, r9q
nv12_inner_%1:
movsx r10d, word [filterq + (2 * r9q)]
movd xm4, r10d
vpbroadcastd m4, xm4 ; filter
mov tmp1q, [uq + (gprsize * r9q)]
mova xm7, oword [tmp1q + 2 * r8q]
mov tmp2q, [vq + (gprsize * r9q)]
mova xm8, oword [tmp2q + 2 * r8q]
punpcklwd xm5, xm7, xm8
pmovsxwd m5, xm5 ; multiplicandsLo
punpckhwd xm6, xm7, xm8
pmovsxwd m6, xm6 ; multiplicandsHi
pmulld m7, m5, m4 ; mulResultLo
pmulld m8, m6, m4 ; mulResultHi
paddd m2, m2, m7 ; resultLo += mulResultLo
paddd m3, m3, m8 ; resultHi += mulResultHi
inc r9d
cmp r9d, filterSized
jl nv12_inner_%1
; end of inner loop
psrad m2, m2, 19
psrad m3, m3, 19
; Vectorized av_clip_uint8
pmaxsd m2, m2, m9
pmaxsd m3, m3, m9
pminsd m2, m2, m10
pminsd m3, m3, m10
; At this point we have clamped uint8s arranged in this order:
; m2: u1 0 0 0 v1 0 0 0 [...]
; m3: u5 0 0 0 v5 0 0 0 [...]
;
; First, we shuffle the bytes to make the bytes semi-contiguous.
; AVX-2 doesn't have cross-lane shuffling, so we'll end up with:
; m2: u1 v1 u2 v2 0 0 0 0 0 0 0 0 u3 v3 u4 v4
; m3: u5 v5 u6 v6 0 0 0 0 0 0 0 0 u7 v7 u8 v8
pshufb m2, m2, m11
pshufb m3, m3, m11
; To fix the cross-lane shuffling issue, we'll then use cross-lane
; permutation to combine the two segments
vpermd m2, m12, m2
vpermd m3, m12, m3
; Now we have the final results in the lower 8 bytes of each register
movq [dstq], xm2
movq [dstq + 8], xm3
add r8d, 8
add dstq, 16
cmp r8d, dstWidthd
jl nv12_outer_%1
RET
%endmacro
INIT_YMM avx2
yuv2nv12cX_fn yuv2nv12
yuv2nv12cX_fn yuv2nv21
%endif ; ARCH_X86_64

28
externals/ffmpeg/libswscale/x86/swscale.c vendored
View File

@@ -380,6 +380,17 @@ INPUT_FUNCS(sse2);
INPUT_FUNCS(ssse3);
INPUT_FUNCS(avx);
#if ARCH_X86_64
#define YUV2NV_DECL(fmt, opt) \
void ff_yuv2 ## fmt ## cX_ ## opt(enum AVPixelFormat format, const uint8_t *dither, \
const int16_t *filter, int filterSize, \
const int16_t **u, const int16_t **v, \
uint8_t *dst, int dstWidth)
YUV2NV_DECL(nv12, avx2);
YUV2NV_DECL(nv21, avx2);
#endif
av_cold void ff_sws_init_swscale_x86(SwsContext *c)
{
int cpu_flags = av_get_cpu_flags();
@@ -580,4 +591,21 @@ switch(c->dstBpc){ \
break;
}
}
#if ARCH_X86_64
if (EXTERNAL_AVX2_FAST(cpu_flags)) {
switch (c->dstFormat) {
case AV_PIX_FMT_NV12:
case AV_PIX_FMT_NV24:
c->yuv2nv12cX = ff_yuv2nv12cX_avx2;
break;
case AV_PIX_FMT_NV21:
case AV_PIX_FMT_NV42:
c->yuv2nv12cX = ff_yuv2nv21cX_avx2;
break;
default:
break;
}
}
#endif
}

22
externals/ffmpeg/libswscale/yuv2rgb.c vendored
View File

@@ -965,6 +965,28 @@ av_cold int ff_yuv2rgb_c_init_tables(SwsContext *c, const int inv_table[4],
fill_table(c->table_bU, 1, cbu, y_table + yoffs);
fill_gv_table(c->table_gV, 1, cgv);
break;
case 30:
rbase = 20;
gbase = 10;
bbase = 0;
needAlpha = CONFIG_SWSCALE_ALPHA && isALPHA(c->srcFormat);
if (!needAlpha)
abase = 30;
ALLOC_YUV_TABLE(table_plane_size * 3 * 4);
y_table32 = c->yuvTable;
yb = -(384 << 16) - YUVRGB_TABLE_LUMA_HEADROOM*cy - oy;
for (i = 0; i < table_plane_size; i++) {
unsigned yval = av_clip_uint8((yb + 0x8000) >> 16);
y_table32[i]= (yval << rbase) + (needAlpha ? 0 : (255u << abase));
y_table32[i + table_plane_size] = yval << gbase;
y_table32[i + 2 * table_plane_size] = yval << bbase;
yb += cy;
}
fill_table(c->table_rV, 4, crv, y_table32 + yoffs);
fill_table(c->table_gU, 4, cgu, y_table32 + yoffs + table_plane_size);
fill_table(c->table_bU, 4, cbu, y_table32 + yoffs + 2 * table_plane_size);
fill_gv_table(c->table_gV, 4, cgv);
break;
case 32:
case 64:
base = (c->dstFormat == AV_PIX_FMT_RGB32_1 ||