June 2022 - ffmpeg-devel - lists.ffmpeg.org

[PATCH 0/2] lavc/aarch64: Provide neon implementations
by Hubert Mazur 13 Jul '22

13 Jul '22

Provide neon implementations for motion estimation functions. Hubert Mazur (2): lavc/aarch64: Assign callback with function lavc/aarch64: Add pix_abs16_x2 neon implementation libavcodec/aarch64/me_cmp_init_aarch64.c | 5 + libavcodec/aarch64/me_cmp_neon.S | 134 +++++++++++++++++++++++ 2 files changed, 139 insertions(+) -- 2.34.1

3 9

[PATCH] avfilter/xstack: Add support for fixed size grid
by Vignesh Venkatasubramanian 13 Jul '22

13 Jul '22

Add a short hand parameter for making a fixed size grid. The existing xstack layout parameter syntax gets tedious if all one wants is a matrix like grid of the input streams. Add a grid option to the xstack filter that simplifies this use case by simply specifying the number of rows and columns instead of specific x/y co-ordinate for each stream. Also updating the filter documentation to explain the new option. Signed-off-by: Vignesh Venkatasubramanian <vigneshv(a)google.com> --- doc/filters.texi | 19 +++++++++-- libavfilter/vf_stack.c | 73 ++++++++++++++++++++++++++++++++++++------ 2 files changed, 81 insertions(+), 11 deletions(-) diff --git a/doc/filters.texi b/doc/filters.texi index e525e87b3c..9d800a0fd6 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -24381,8 +24381,23 @@ the output video frame will be filled. Similarly, videos can overlap each other if their position doesn't leave enough space for the full frame of adjoining videos. -For 2 inputs, a default layout of @code{0_0|w0_0} is set. In all other cases, -a layout must be set by the user. +For 2 inputs, a default layout of @code{0_0|w0_0} is set. In all other cases, a +layout or a grid must be set by the user. Either grid or layout option can be +specified at a time. Specifying both will result in an error. + +@item grid +Specify a fixed size grid of inputs. +This option is used to create a fixed size grid of the input streams. The +option is of the form <rows>x<columns> (e.g. 2x4). There must be <rows> * +<columns> input streams and they will be arranged as a grid with <rows> rows and +<columns> columns. When using this option, all the input streams must have the +same width and height. + +Either grid or layout option can be specified at a time. Specifying both will +result in an error. + +If grid is set, then inputs option is ignored and is implicitly set to +<rows>*<columns>. @item shortest If set to 1, force the output to terminate when the shortest input diff --git a/libavfilter/vf_stack.c b/libavfilter/vf_stack.c index aa32a1bf5e..b38a193355 100644 --- a/libavfilter/vf_stack.c +++ b/libavfilter/vf_stack.c @@ -48,6 +48,9 @@ typedef struct StackContext { int is_vertical; int is_horizontal; int nb_planes; + char *grid; + int nb_grid_rows; + int nb_grid_columns; uint8_t fillcolor[4]; char *fillcolor_str; int fillcolor_enable; @@ -85,14 +88,6 @@ static av_cold int init(AVFilterContext *ctx) if (!strcmp(ctx->filter->name, "hstack")) s->is_horizontal = 1; - s->frames = av_calloc(s->nb_inputs, sizeof(*s->frames)); - if (!s->frames) - return AVERROR(ENOMEM); - - s->items = av_calloc(s->nb_inputs, sizeof(*s->items)); - if (!s->items) - return AVERROR(ENOMEM); - if (!strcmp(ctx->filter->name, "xstack")) { if (strcmp(s->fillcolor_str, "none") && av_parse_color(s->fillcolor, s->fillcolor_str, -1, ctx) >= 0) { @@ -100,7 +95,21 @@ static av_cold int init(AVFilterContext *ctx) } else { s->fillcolor_enable = 0; } - if (!s->layout) { + if (s->grid && s->layout) { + av_log(ctx, AV_LOG_ERROR, "Both layout and grid were specified. Only one is allowed.\n"); + return AVERROR(EINVAL); + } + if (s->grid) { + if (sscanf(s->grid, "%dx%d", &s->nb_grid_rows, &s->nb_grid_columns) != 2) { + av_log(ctx, AV_LOG_ERROR, "grid string is not of the form rowsxcolumns.\n"); + return AVERROR(EINVAL); + } + s->nb_inputs = s->nb_grid_rows * s->nb_grid_columns; + } else { + s->nb_grid_rows = 0; + s->nb_grid_columns = 0; + } + if (!s->layout && !s->grid) { if (s->nb_inputs == 2) { s->layout = av_strdup("0_0|w0_0"); if (!s->layout) @@ -112,6 +121,15 @@ static av_cold int init(AVFilterContext *ctx) } } + s->frames = av_calloc(s->nb_inputs, sizeof(*s->frames)); + if (!s->frames) + return AVERROR(ENOMEM); + + s->items = av_calloc(s->nb_inputs, sizeof(*s->items)); + if (!s->items) + return AVERROR(ENOMEM); + + for (i = 0; i < s->nb_inputs; i++) { AVFilterPad pad = { 0 }; @@ -244,6 +262,42 @@ static int config_output(AVFilterLink *outlink) width += ctx->inputs[i]->w; } } + } else if (s->nb_grid_rows > 0 && s->nb_grid_columns > 0) { + int inw = 0, inh = 0; + int k = 0; + for (i = 0; i < s->nb_grid_rows; i++, inh += height) { + inw = 0; + for (int j = 0; j < s->nb_grid_columns; j++, k++, inw += width) { + AVFilterLink *inlink = ctx->inputs[k]; + StackItem *item = &s->items[k]; + + if (ctx->inputs[k]->w != width) { + av_log(ctx, AV_LOG_ERROR, "Input %d width %d does not match input %d width %d.\n", k, ctx->inputs[k]->w, 0, width); + return AVERROR(EINVAL); + } + + if (ctx->inputs[k]->h != height) { + av_log(ctx, AV_LOG_ERROR, "Input %d height %d does not match input %d height %d.\n", k, ctx->inputs[k]->h, 0, height); + return AVERROR(EINVAL); + } + + if ((ret = av_image_fill_linesizes(item->linesize, inlink->format, inlink->w)) < 0) { + return ret; + } + + item->height[1] = item->height[2] = AV_CEIL_RSHIFT(inlink->h, s->desc->log2_chroma_h); + item->height[0] = item->height[3] = inlink->h; + + if ((ret = av_image_fill_linesizes(item->x, inlink->format, inw)) < 0) { + return ret; + } + + item->y[1] = item->y[2] = AV_CEIL_RSHIFT(inh, s->desc->log2_chroma_h); + item->y[0] = item->y[3] = inh; + } + } + width *= s->nb_grid_columns; + height *= s->nb_grid_rows; } else { char *arg, *p = s->layout, *saveptr = NULL; char *arg2, *p2, *saveptr2 = NULL; @@ -436,6 +490,7 @@ const AVFilter ff_vf_vstack = { static const AVOption xstack_options[] = { { "inputs", "set number of inputs", OFFSET(nb_inputs), AV_OPT_TYPE_INT, {.i64=2}, 2, INT_MAX, .flags = FLAGS }, { "layout", "set custom layout", OFFSET(layout), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 0, .flags = FLAGS }, + { "grid", "set fixed size grid layout", OFFSET(grid), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 0, .flags = FLAGS }, { "shortest", "force termination when the shortest input terminates", OFFSET(shortest), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, .flags = FLAGS }, { "fill", "set the color for unused pixels", OFFSET(fillcolor_str), AV_OPT_TYPE_STRING, {.str = "none"}, .flags = FLAGS }, { NULL }, -- 2.37.0.rc0.161.g10f37bed90-goog

2 7

[PATCH v2 3/5] libswscale: Avx2 hscale can process inputs of any size.
by Alan Kelly 13 Jul '22

13 Jul '22

The main loop processes blocks of 16 pixels. The tail processes blocks of size 4. --- libswscale/x86/scale_avx2.asm | 48 +++++++++++++++++++++++++++++++++-- 1 file changed, 46 insertions(+), 2 deletions(-) diff --git a/libswscale/x86/scale_avx2.asm b/libswscale/x86/scale_avx2.asm index 20acdbd633..dc42abb100 100644 --- a/libswscale/x86/scale_avx2.asm +++ b/libswscale/x86/scale_avx2.asm @@ -53,6 +53,9 @@ cglobal hscale8to15_%1, 7, 9, 16, pos0, dst, w, srcmem, filter, fltpos, fltsize, mova m14, [four] shr fltsized, 2 %endif + cmp wq, 16 + jl .tail_loop + mov countq, 0x10 .loop: movu m1, [fltposq] movu m2, [fltposq+32] @@ -97,11 +100,52 @@ cglobal hscale8to15_%1, 7, 9, 16, pos0, dst, w, srcmem, filter, fltpos, fltsize, vpsrad m6, 7 vpackssdw m5, m5, m6 vpermd m5, m15, m5 - vmovdqu [dstq + countq * 2], m5 + vmovdqu [dstq], m5 + add dstq, 0x20 add fltposq, 0x40 add countq, 0x10 cmp countq, wq - jl .loop + jle .loop + + sub countq, 0x10 + cmp countq, wq + jge .end + +.tail_loop: + movu xm1, [fltposq] +%ifidn %1, X4 + pxor xm9, xm9 + pxor xm10, xm10 + xor innerq, innerq +.tail_innerloop: +%endif + vpcmpeqd xm13, xm13 + vpgatherdd xm3,[srcmemq + xm1], xm13 + vpunpcklbw xm5, xm3, xm0 + vpunpckhbw xm6, xm3, xm0 + vpmaddwd xm5, xm5, [filterq] + vpmaddwd xm6, xm6, [filterq + 16] + add filterq, 0x20 +%ifidn %1, X4 + paddd xm9, xm5 + paddd xm10, xm6 + paddd xm1, xm14 + add innerq, 1 + cmp innerq, fltsizeq + jl .tail_innerloop + vphaddd xm5, xm9, xm10 +%else + vphaddd xm5, xm5, xm6 +%endif + vpsrad xm5, 7 + vpackssdw xm5, xm5, xm5 + vmovq [dstq], xm5 + add dstq, 0x8 + add fltposq, 0x10 + add countq, 0x4 + cmp countq, wq + jl .tail_loop +.end: REP_RET %endmacro -- 2.35.1.265.g69c8d7142f-goog

3 8

[PATCH] avcodec/h264dec: Skip late SEI
by Michael Niedermayer 12 Jul '22

12 Jul '22

Fixes: Race condition Fixes: clusterfuzz-testcase-minimized-mediasource_MP2T_AVC_pipeline_integration_fuzzer-6282675434094592 Found-by: google ClusterFuzz Tested-by: Dan Sanders <sandersd(a)google.com> Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavcodec/h264dec.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/libavcodec/h264dec.c b/libavcodec/h264dec.c index d80bc6b17f..2a5b53ea56 100644 --- a/libavcodec/h264dec.c +++ b/libavcodec/h264dec.c @@ -672,6 +672,10 @@ static int decode_nal_units(H264Context *h, const uint8_t *buf, int buf_size) avpriv_request_sample(avctx, "data partitioning"); break; case H264_NAL_SEI: + if (h->setup_finished) { + avpriv_request_sample(avctx, "Late SEI"); + break; + } ret = ff_h264_sei_decode(&h->sei, &nal->gb, &h->ps, avctx); h->has_recovery_point = h->has_recovery_point || h->sei.recovery_point.recovery_frame_cnt != -1; if (avctx->debug & FF_DEBUG_GREEN_MD) -- 2.17.1

1 1

[PATCH] avcodec/sbrdsp_fixed: Fix integer overflows in sbr_qmf_deint_neg_c()
by Michael Niedermayer 12 Jul '22

12 Jul '22

Fixes: signed integer overflow: 2147483645 + 16 cannot be represented in type 'int' Fixes: 46993/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AAC_FIXED_fuzzer-4759025234870272 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavcodec/sbrdsp_fixed.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavcodec/sbrdsp_fixed.c b/libavcodec/sbrdsp_fixed.c index 43fcc90ae5..0d34a2a710 100644 --- a/libavcodec/sbrdsp_fixed.c +++ b/libavcodec/sbrdsp_fixed.c @@ -114,8 +114,8 @@ static void sbr_qmf_deint_neg_c(int *v, const int *src) { int i; for (i = 0; i < 32; i++) { - v[ i] = ( src[63 - 2*i ] + 0x10) >> 5; - v[63 - i] = (-src[63 - 2*i - 1] + 0x10) >> 5; + v[ i] = (int)(0x10U + src[63 - 2*i ]) >> 5; + v[63 - i] = (int)(0x10U - src[63 - 2*i - 1]) >> 5; } } -- 2.17.1

1 1

[PATCH] avcodec/pixlet: consider minimum plane header in the minimal packet size
by Michael Niedermayer 12 Jul '22

12 Jul '22

Fixes: Timeout Fixes: 46956/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_PIXLET_fuzzer-5698161106092032 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavcodec/pixlet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavcodec/pixlet.c b/libavcodec/pixlet.c index 18a6587257..b16dd4aa87 100644 --- a/libavcodec/pixlet.c +++ b/libavcodec/pixlet.c @@ -611,7 +611,7 @@ static int pixlet_decode_frame(AVCodecContext *avctx, AVFrame *p, bytestream2_init(&ctx->gb, avpkt->data, avpkt->size); pktsize = bytestream2_get_be32(&ctx->gb); - if (pktsize <= 44 || pktsize - 4 > bytestream2_get_bytes_left(&ctx->gb)) { + if (pktsize <= 44 + (NB_LEVELS * 8 + 6) * 3 || pktsize - 4 > bytestream2_get_bytes_left(&ctx->gb)) { av_log(avctx, AV_LOG_ERROR, "Invalid packet size %"PRIu32"\n", pktsize); return AVERROR_INVALIDDATA; } -- 2.17.1

1 1

[PATCH 1/2] avformat/http: Remove always true non NULL check
by Michael Niedermayer 12 Jul '22

12 Jul '22

Coverity uses the logic of "Comparing options to null implies that options might be null." But ffurl_connect() checks for NULL already and uses &tmp_opts Helps: CID1374122 Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavformat/http.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/libavformat/http.c b/libavformat/http.c index c8f3f4b6a3..81c3048934 100644 --- a/libavformat/http.c +++ b/libavformat/http.c @@ -692,8 +692,7 @@ static int http_open(URLContext *h, const char *uri, int flags, if (!s->uri) return AVERROR(ENOMEM); - if (options) - av_dict_copy(&s->chained_options, *options, 0); + av_dict_copy(&s->chained_options, *options, 0); if (s->headers) { int len = strlen(s->headers); -- 2.17.1

1 2

[PATCH 1/2] avformat/matroskadec: assert non NULL buf
by Michael Niedermayer 12 Jul '22

12 Jul '22

The code is only called if size is > 0 so buf should not be NULL Helps: CID610554 Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavformat/matroskadec.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c index 4715f1b7d4..de73f97aca 100644 --- a/libavformat/matroskadec.c +++ b/libavformat/matroskadec.c @@ -3701,6 +3701,8 @@ static int matroska_parse_block(MatroskaDemuxContext *matroska, AVBufferRef *buf uint64_t num; int trust_default_duration; + av_assert1(buf); + ffio_init_context(&pb, data, size, 0, NULL, NULL, NULL, NULL); if ((n = ebml_read_num(matroska, &pb.pub, 8, &num, 1)) < 0) -- 2.17.1

2 4

[PATCH 1/3] avcodec/jpeglsdec: fix end check for xfrm
by Michael Niedermayer 12 Jul '22

12 Jul '22

Fixes: out of array access Fixes: 47871/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AMV_fuzzer-5646305956855808 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavcodec/jpeglsdec.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/libavcodec/jpeglsdec.c b/libavcodec/jpeglsdec.c index 399837cf2f..ab663dc1fc 100644 --- a/libavcodec/jpeglsdec.c +++ b/libavcodec/jpeglsdec.c @@ -484,19 +484,19 @@ int ff_jpegls_decode_picture(MJpegDecodeContext *s, int near, for (i = 0; i < s->height; i++) { switch(s->xfrm) { case 1: - for (x = off; x < w; x += 3) { + for (x = off; x + 2 < w; x += 3) { src[x ] += src[x+1] + 128; src[x+2] += src[x+1] + 128; } break; case 2: - for (x = off; x < w; x += 3) { + for (x = off; x + 2 < w; x += 3) { src[x ] += src[x+1] + 128; src[x+2] += ((src[x ] + src[x+1])>>1) + 128; } break; case 3: - for (x = off; x < w; x += 3) { + for (x = off; x + 2 < w; x += 3) { int g = src[x+0] - ((src[x+2]+src[x+1])>>2) + 64; src[x+0] = src[x+2] + g + 128; src[x+2] = src[x+1] + g + 128; @@ -504,7 +504,7 @@ int ff_jpegls_decode_picture(MJpegDecodeContext *s, int near, } break; case 4: - for (x = off; x < w; x += 3) { + for (x = off; x + 2 < w; x += 3) { int r = src[x+0] - (( 359 * (src[x+2]-128) + 490) >> 8); int g = src[x+0] - (( 88 * (src[x+1]-128) - 183 * (src[x+2]-128) + 30) >> 8); int b = src[x+0] + ((454 * (src[x+1]-128) + 574) >> 8); -- 2.17.1

1 5

[PATCH] avcodec/qpeldsp: copy less for the mc0x cases
by Michael Niedermayer 12 Jul '22

12 Jul '22

Fixes: out of array access Fixes: 47936/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MPEG4_fuzzer-5745039940124672 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> --- libavcodec/qpeldsp.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/libavcodec/qpeldsp.c b/libavcodec/qpeldsp.c index 2b9146ceb1..5f937f9d9e 100644 --- a/libavcodec/qpeldsp.c +++ b/libavcodec/qpeldsp.c @@ -199,7 +199,7 @@ static void OPNAME ## qpel8_mc01_c(uint8_t *dst, const uint8_t *src, \ uint8_t full[16 * 9]; \ uint8_t half[64]; \ \ - copy_block9(full, src, 16, stride, 9); \ + copy_block8(full, src, 16, stride, 9); \ put ## RND ## mpeg4_qpel8_v_lowpass(half, full, 8, 16); \ OPNAME ## pixels8_l2_8(dst, full, half, stride, 16, 8, 8); \ } \ @@ -209,7 +209,7 @@ static void OPNAME ## qpel8_mc02_c(uint8_t *dst, const uint8_t *src, \ { \ uint8_t full[16 * 9]; \ \ - copy_block9(full, src, 16, stride, 9); \ + copy_block8(full, src, 16, stride, 9); \ OPNAME ## mpeg4_qpel8_v_lowpass(dst, full, stride, 16); \ } \ \ @@ -219,7 +219,7 @@ static void OPNAME ## qpel8_mc03_c(uint8_t *dst, const uint8_t *src, \ uint8_t full[16 * 9]; \ uint8_t half[64]; \ \ - copy_block9(full, src, 16, stride, 9); \ + copy_block8(full, src, 16, stride, 9); \ put ## RND ## mpeg4_qpel8_v_lowpass(half, full, 8, 16); \ OPNAME ## pixels8_l2_8(dst, full + 16, half, stride, 16, 8, 8); \ } \ @@ -459,7 +459,7 @@ static void OPNAME ## qpel16_mc01_c(uint8_t *dst, const uint8_t *src, \ uint8_t full[24 * 17]; \ uint8_t half[256]; \ \ - copy_block17(full, src, 24, stride, 17); \ + copy_block16(full, src, 24, stride, 17); \ put ## RND ## mpeg4_qpel16_v_lowpass(half, full, 16, 24); \ OPNAME ## pixels16_l2_8(dst, full, half, stride, 24, 16, 16); \ } \ @@ -469,7 +469,7 @@ static void OPNAME ## qpel16_mc02_c(uint8_t *dst, const uint8_t *src, \ { \ uint8_t full[24 * 17]; \ \ - copy_block17(full, src, 24, stride, 17); \ + copy_block16(full, src, 24, stride, 17); \ OPNAME ## mpeg4_qpel16_v_lowpass(dst, full, stride, 24); \ } \ \ @@ -479,7 +479,7 @@ static void OPNAME ## qpel16_mc03_c(uint8_t *dst, const uint8_t *src, \ uint8_t full[24 * 17]; \ uint8_t half[256]; \ \ - copy_block17(full, src, 24, stride, 17); \ + copy_block16(full, src, 24, stride, 17); \ put ## RND ## mpeg4_qpel16_v_lowpass(half, full, 16, 24); \ OPNAME ## pixels16_l2_8(dst, full + 24, half, stride, 24, 16, 16); \ } \ -- 2.17.1

2 3