summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* lavc/hevc_ps: compactify ShortTermRPShevcAnton Khirnov5 days
| | | | | | Do not use larger fields than needed, use size-1 bitfields for flags. Reduces sizeof(HEVCSPS) by 1280 bytes.
* lavc/hevc_ps: reduce the size of ShortTermRPS.usedAnton Khirnov5 days
| | | | | | | It is currently an array of 32 uint8_t, each storing a single flag. A single uint32_t is sufficient. Reduces sizeof(HEVCSPS) by 1792 bytes.
* lavc/hevc_ps: do not store delta_poc_s[01] in ShortTermRPSAnton Khirnov5 days
| | | | | | | | | They are only used in vulkan_hevc and are not actually needed, as they can be computed from delta_poc. Reduces sizeof(HEVCSPS) by 16kB. Also, fix a typo (s0->s1) in the code being touched.
* lavc/hevc_ps: fix variable signedness in ff_hevc_decode_short_term_rps()Anton Khirnov5 days
| | | | | | It is actually supposed to go negative in the loop over num_negative pics, but underflow does not break anything as the result is then assigned to a signed int.
* lavc/hevc_ps/HEVCSPS: change flags into uint8_tAnton Khirnov5 days
| | | | | | | Reduces sizeof(HEVCSPS) by 64 bytes. Also improve flag names: drop redundant suffixes and prefixes, and consistently use disabled/enabled.
* lavc/hevc_ps: reduce the size of used_by_curr_pic_lt_sps_flagAnton Khirnov5 days
| | | | | It is currently an array of 32 uint8_t, each storing a single flag. A single uint32_t is sufficient.
* lavc/hevcdec: drop a useless execute() call with 1 jobAnton Khirnov5 days
|
* lavc/hevcdec: allocate local_ctx as array of structs rather than pointersAnton Khirnov5 days
| | | | It is more efficient and easier to manage.
* lavc/hevcdec: track local context count separately from WPP thread countAnton Khirnov11 days
| | | | The latter can be lowered while decoding, which would lead to memleaks.
* lavc/hevcdec: rename HEVCContext.HEVClcList to local_ctxAnton Khirnov11 days
| | | | It is more consistent with our naming conventions.
* avformat/mov: avoid seeking back to 0 on HEVC open GOP filesllyyr12 days
| | | | | | | | | | | | | | | | | | | ab77b878f1 attempted to fix the issue of broken packets being sent to the decoder by implementing logic that kept attempting to PTS-step backwards until it reached a valid point, however applying this heuristic meant that in files that had no valid points (such as HEVC videos shot on iPhones), we'd seek back to sample 0 on every seek attempt. This meant that files that were previously seekable, albeit with some skipped frames, were not seekable at all now. Relax this heuristic a bit by giving up on seeking to a valid point if we've tried a different sample and we still don't have a valid point to seek to. This may some frames to be skipped on seeking but it's better than not being able to seek at all in such files. Fixes: ab77b878f1 ("avformat/mov: fix seeking with HEVC open GOP files") Fixes: #10585 Signed-off-by: Philip Langdale <philipl@overt.org>
* lavc/vp9dsp: R-V V mc avgsunyuechi12 days
| | | | | | | | | | | | | | | | C908: vp9_avg4_8bpp_c: 1.2 vp9_avg4_8bpp_rvv_i64: 1.0 vp9_avg8_8bpp_c: 3.7 vp9_avg8_8bpp_rvv_i64: 1.5 vp9_avg16_8bpp_c: 14.7 vp9_avg16_8bpp_rvv_i64: 3.5 vp9_avg32_8bpp_c: 57.7 vp9_avg32_8bpp_rvv_i64: 10.0 vp9_avg64_8bpp_c: 229.0 vp9_avg64_8bpp_rvv_i64: 31.7 Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
* Revert "lavc/sbrdsp: R-V V neg_odd_64"Rémi Denis-Courmont12 days
| | | | | | | | | | | | | | | | | | | | | | | While this function can easily be written with vectors, it just fails to get any performance improvement. For reference, this is a simpler loop-free implementation that does get better performance than the current one depending on hardware, but still more or less the same metrics as the C code: func ff_sbr_neg_odd_64_rvv, zve64x li a1, 32 addi a0, a0, 7 li t0, 8 vsetvli zero, a1, e8, m2, ta, ma li t1, 0x80 vlse8.v v8, (a0), t0 vxor.vx v8, v8, t1 vsse8.v v8, (a0), t0 ret endfunc This reverts commit d06fd18f8f4c6a81ef94cbb600620d83ad51269d.
* lavc/vc1dsp: R-V V vc1_unescape_bufferRémi Denis-Courmont12 days
| | | | | | | | | | | | | | | | | | | | | | | | Notes: - The loop is biased toward no unescaped bytes as that should be most common. - The input byte array is slid rather than the (8 times smaller) bit-mask, as RISC-V V does not provide a bit-mask (or bit-wise) slide instruction. - There are two comparisons with 0 per iteration, for the same reason. - In case of match, bytes are copied until the first match, and the loop is restarted after the escape byte. Vector compression (vcompress.vm) could discard all escape bytes but that is slower if escape bytes are rare. Further optimisations should be possible, e.g.: - processing 2 bytes fewer per iteration to get rid of a 2 slides, - taking a short cut if the input vector contains less than 2 zeroes. But this is a good starting point: T-Head C908: vc1dsp.vc1_unescape_buffer_c: 12749.5 vc1dsp.vc1_unescape_buffer_rvv_i32: 6009.0 SpacemiT X60: vc1dsp.vc1_unescape_buffer_c: 11038.0 vc1dsp.vc1_unescape_buffer_rvv_i32: 2061.0
* checkasm: h264dsp: Avoid out of buffer writes when benchmarkingMartin Storsjö12 days
| | | | | | | | | | The loop filters can write before the pointer given to them; the actual test invocations correctly used an offset, while the benchmark calls were lacking an offset. Therefore, when running with benchmarking, these tests could have spurious failures. Signed-off-by: Martin Storsjö <martin@martin.st>
* checkasm: print bench runs when benchmarkingLynne12 days
| | | | Helps make sense of the possible noise in the results.
* checkasm: add runs argument to adjust during benchJ. Dekker12 days
| | | | | | | | | | | | Some timers on certain device and test combinations can produce noisy results, affecting the reliability of performance measurements. One notable example of this is the Canaan K230 RISC-V development board. An option to adjust the number of samples by an exponent (--runs) has been added, allowing developers to increase the sample count for more reliable results. Signed-off-by: J. Dekker <jdek@itanimul.li>
* checkasm: vvc_alf: Limit benchmarking to a reasonable subset of functionsMartin Storsjö12 days
| | | | | | | | | | | Don't benchmark every single combination of widths and heights; only benchmark cases which are squares (like in vvc_mc.c). Contrary to vvc_mc, which increases sizes by doubling dimensions, vvc_alf tests all sizes in increments of 4. Limit benchmarking to the cases which are powers of two. This reduces the number of benchmarked cases from 3072 down to 18.
* Changelog: add DVB compatible information for VVC decoderNuo Mi12 days
| | | | see https://dvb.org/specifications/verification-validation/vvc-test-content/
* avcodec/vvcdec: support Reference Picture ResamplingNuo Mi12 days
| | | | | | | | | passed clips: RPR_A_Alibaba_4.bit RPR_B_Alibaba_3.bit RPR_C_Alibaba_3.bit RPR_D_Qualcomm_1.bit VVC_HDR_UHDTV1_OpenGOP_Max3840x2160_50fps_HLG10_res_change_with_RPR.ts
* avcodec/vvcdec: increase edge_emu_buffer for RPRNuo Mi12 days
|
* avcodec/vvcdec: refact, remove hf_idx and vf_idx from mc_xxx's param listNuo Mi12 days
|
* avcodec/vvcdec: refact out luma_prof from luma_prof_biNuo Mi12 days
|
* avcodec/vvcdec: fix dmvr, bdof, cb_prof for RPRNuo Mi12 days
|
* avcodec/vvcdec: inter, wait reference with a different resolutionNuo Mi12 days
| | | | | For RPR, the current frame may reference a frame with a different resolution. Therefore, we need to consider frame scaling when we wait for reference pixels.
* avcodec/vvcdec: add RPR dspNuo Mi12 days
|
* avcodec/vvcdec: emulated_edge, use reference frame's sps and ppsNuo Mi12 days
| | | | a preparation for Reference Picture Resampling
* avcodec/vvcdec: add vvc inter filters for RPRNuo Mi12 days
|
* avcodec/vvcdec: refact, pred_get_refs return VVCRefPic instead of VVCFrameNuo Mi12 days
|
* avcodec/vvcdec: refact out VVCRefPic from RefPicListNuo Mi12 days
|
* avcodec/vvcdec: refact, unify pred_regular_{luma, chroma} to pred_regularNuo Mi12 days
|
* avcodec/vvcdec: misc, remove unused EMULATED_EDGE_{LUMA, CHROMA}, ↵Nuo Mi12 days
| | | | EMULATED_EDGE_DMVR_{LUAM, CHROMA}
* avcodec/vvcdec: refact, unify {luma, chroma}_mc_bi to mc_biNuo Mi12 days
|
* avcodec/vvcdec: refact, unify {luma, chroma}_mc_uni to mc_uniNuo Mi12 days
|
* avcodec/vvcdec: refact, unify {luma, chroma}_mc to mcNuo Mi12 days
|
* avcodec/vvcdec: misc, inter, use is_chroma instead of is_lumaNuo Mi12 days
|
* avfilter/vf_dnn_detect: Fix null pointer dereferenceZhao Zhili13 days
| | | | Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
* lavc/vaapi_decode: Reject decoding of frames with no slicesDavid Rosca13 days
| | | | Matches other hwaccels.
* avutil/hwcontext_qsv: fix GCC 14.1 warningsoltolm13 days
| | | | | Tested-by: Tong Wu <tong1.wu@intel.com> Signed-off-by: oltolm <oleg.tolmatcev@gmail.com>
* avfilter/vf_signalstats: Reindent after the previous commitAndreas Rheinhardt13 days
| | | | | | Also use loop scope for iterators while just at it. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/vf_signalstats: Deduplicate <= 8 and > 8 bit codeAndreas Rheinhardt13 days
| | | | Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/vf_signalstats: Use av_dict_set_int() where appropriateAndreas Rheinhardt13 days
| | | | Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avformat/mp3dec: change bogus error message if read_header encounters EOFMarton Balint13 days
| | | | | | | | Because of ffio_ensure_seekback() a seek error normally should only happen if the end of file is reached during checking for the junk run-in. Also use proper error code. Signed-off-by: Marton Balint <cus@passwd.hu>
* avformat/mp3dec: simplify inner frame size check in mp3_read_headerMarton Balint13 days
| | | | | | | | | | We are protecting the checked buffer with ffio_ensure_seekback(), so if the inner check fails with a seek error, that likely means the end of file was reached when checking for the next frame. This could also be the result of a wrongly guessed (larger than normal) frame size, so let's continue the loop instead of breaking out early. It will end sooner or later anyway. Signed-off-by: Marton Balint <cus@passwd.hu>
* avformat/mp3dec: only call ffio_ensure_seekback onceMarton Balint13 days
| | | | | | | | | | | Otherwise the subsequent ffio_ensure_seekback calls destroy the buffer of the earlier. The worst case ~66kB seekback is so small it is easier to request it entirely. Fixes ticket #10837, a regression since 0d17f5228f4d3854066ec1001f69c7d1714b0df9. Signed-off-by: Marton Balint <cus@passwd.hu>
* libavfilter/vf_curves: fix a memory leak on error pathLuMingYin13 days
| | | | | Signed-off-by: LuMingYin <lumingyindetect@163.com> Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
* libavformat/rtsp: fix a memory leak on error pathLuMingYin13 days
| | | | | Signed-off-by: LuMingYin <lumingyindetect@163.com> Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
* libavformat/hlsenc: fix a memory leak on error pathLuMingYin13 days
| | | | | | Signed-off-by: LuMingYin <lumingyindetect@163.com> Reviewed-by: Steven Liu <lq@chinaffmpeg.org> Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
* avcodec/cbs_h266: read vps_ptl_max_tid before using itJames Almer13 days
| | | | | Reviewed-by: Nuo Mi <nuomi2021@gmail.com> Signed-off-by: James Almer <jamrial@gmail.com>
* avfilter/vf_v360: Add assert to suppress Coverity false positivesAndreas Rheinhardt13 days
| | | | | | | Should fix many Coverity false positives, namely #1457947-#1457994 as well as #1461195-#146210. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-02 22:55:03 +0000