[FFmpeg-cvslog] checkasm: add helper to report a fatal signal

ffmpeg | branch: master | Rémi Denis-Courmont | Thu Nov 16 17:45:02 2023 +0200| [286d6742218ba0235c32876b50bf593cb1986353] | committer: Rémi Denis-Courmont checkasm: add helper to report a fatal signal > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=286d6742218ba0235c32876b50bf593c

[FFmpeg-cvslog] checkasm/riscv: report an error upon SIGILL

ffmpeg | branch: master | Rémi Denis-Courmont | Wed Jul 12 22:48:53 2023 +0300| [7212466e735aa187d82f51dadbce957fe3da77f0] | committer: Rémi Denis-Courmont checkasm/riscv: report an error upon SIGILL Terminating the whole checkasm process is not very helpful. This will report if an illegal ins

[FFmpeg-cvslog] lavu/fixed_dsp: R-V V fmul_window_scaled

ffmpeg | branch: master | Rémi Denis-Courmont | Wed Oct 4 22:50:47 2023 +0300| [e33ce0d9dd8136d9d1a5eb17863c2f6d4d5f131f] | committer: Rémi Denis-Courmont lavu/fixed_dsp: R-V V fmul_window_scaled vector_fmul_window_scaled_fixed_c: 4393.7 vector_fmul_window_scaled_fixed_rvv_i64: 1642.7

[FFmpeg-cvslog] lavc/llvidencdsp: add R-V V diff_bytes

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Nov 19 21:55:58 2023 +0200| [0fa421c8f19109aa26d902d2a33d2f46e57f378b] | committer: Rémi Denis-Courmont lavc/llvidencdsp: add R-V V diff_bytes diff_bytes_c: 163.0 diff_bytes_rvv_i32: 52.7 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/

[FFmpeg-cvslog] lavc/aacpsdsp: use LMUL=2 and amortise strides

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Nov 19 17:50:49 2023 +0200| [0183c2c83091b727f3b881c8797b45c8f6915f1b] | committer: Rémi Denis-Courmont lavc/aacpsdsp: use LMUL=2 and amortise strides The input is laid out in 16 segments, of which 13 actually need to be loaded. There are no

[FFmpeg-cvslog] lavu/fixed_dsp: optimise R-V V fmul_reverse

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Nov 19 13:24:29 2023 +0200| [3a134e82994ff49b784056d2dfce0230a8256ebd] | committer: Rémi Denis-Courmont lavu/fixed_dsp: optimise R-V V fmul_reverse Gathers are (unsurprisingly) a notable exception to the rule that R-V V gets faster with large

[FFmpeg-cvslog] lavu/float_dsp: optimise R-V V fmul_reverse & fmul_window

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Nov 19 13:36:01 2023 +0200| [e49f41fb27711a19eab3f41b431ee2fc7b0e6120] | committer: Rémi Denis-Courmont lavu/float_dsp: optimise R-V V fmul_reverse & fmul_window Roll the loop to avoid slow gathers. Before: vector_fmul_reverse_c: 1561.

[FFmpeg-cvslog] lavc/g722dsp: optimise R-V V apply_qmf

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Nov 19 14:49:28 2023 +0200| [b88d4058f95de7ebf8322358d2e72cbeaffec49e] | committer: Rémi Denis-Courmont lavc/g722dsp: optimise R-V V apply_qmf This stores the constant coefficients deinterleaved, so that they can be loaded directly with NF=0.

[FFmpeg-cvslog] avfilter/asrc_afirsrc: fix by one smaller allocation of buffer

ffmpeg | branch: master | Paul B Mahol | Thu Nov 23 14:58:35 2023 +0100| [4adb93dff05dd947878c67784d98c9a4e13b57a7] | committer: Paul B Mahol avfilter/asrc_afirsrc: fix by one smaller allocation of buffer > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=4adb93dff05dd947878c67784d98c9