ffmpeg | branch: master | Niklas Haas <g...@haasn.dev> | Sat Jun 21 17:23:16 2025 +0200| [daef348574898f3126ae2b1b70e9a09f773bcca6] | committer: Niklas Haas
avfilter/x86/f_ebur128: implement AVX peak calculation Stereo only, for simplicity. Slightly faster than the C code. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=daef348574898f3126ae2b1b70e9a09f773bcca6 --- libavfilter/f_ebur128.c | 8 +++----- libavfilter/f_ebur128.h | 2 +- libavfilter/x86/f_ebur128.asm | 20 ++++++++++++++++++++ libavfilter/x86/f_ebur128_init.c | 9 +++++++-- 4 files changed, 31 insertions(+), 8 deletions(-) diff --git a/libavfilter/f_ebur128.c b/libavfilter/f_ebur128.c index 514be7a8e3..80cbe20330 100644 --- a/libavfilter/f_ebur128.c +++ b/libavfilter/f_ebur128.c @@ -502,6 +502,9 @@ static int config_audio_output(AVFilterLink *outlink) return AVERROR(ENOMEM); } +#if ARCH_X86 + ff_ebur128_init_x86(&ebur128->dsp, nb_channels); +#endif return 0; } @@ -581,11 +584,6 @@ static av_cold int init(AVFilterContext *ctx) ebur128->dsp.filter_channels = ff_ebur128_filter_channels_c; ebur128->dsp.find_peak = ff_ebur128_find_peak_c; - -#if ARCH_X86 - ff_ebur128_init_x86(&ebur128->dsp); -#endif - return 0; } diff --git a/libavfilter/f_ebur128.h b/libavfilter/f_ebur128.h index ad24a3c129..19564f562b 100644 --- a/libavfilter/f_ebur128.h +++ b/libavfilter/f_ebur128.h @@ -55,7 +55,7 @@ static_assert(offsetof(EBUR128DSPContext, pre) == 0, "struct l static_assert(offsetof(EBUR128DSPContext, rlb) == 5 * sizeof(double), "struct layout mismatch"); static_assert(offsetof(EBUR128DSPContext, y) == 10 * sizeof(double), "struct layout mismatch"); -void ff_ebur128_init_x86(EBUR128DSPContext *dsp); +void ff_ebur128_init_x86(EBUR128DSPContext *dsp, int nb_channels); void ff_ebur128_filter_channels_c(const EBUR128DSPContext *, const double *, double *, double *, double *, double *, int); diff --git a/libavfilter/x86/f_ebur128.asm b/libavfilter/x86/f_ebur128.asm index f8515d811d..c7a3e1846d 100644 --- a/libavfilter/x86/f_ebur128.asm +++ b/libavfilter/x86/f_ebur128.asm @@ -37,6 +37,10 @@ struc DSP .z resq 1 endstruc +SECTION_RODATA + +abs_mask: dq 0x7FFFFFFFFFFFFFFF + SECTION .text %macro MOVNQ 3 ; num, dst, src @@ -140,4 +144,20 @@ cglobal ebur128_filter_channels, 7, 9, 14, dsp, samples, cache400, cache3000, su jnz .loop RET +cglobal ebur128_find_peak_2ch, 4, 5, 3, ch_peaks, channels, samples, nb_samples + vpbroadcastq m2, [abs_mask] + movupd m0, [ch_peaksq] +.loop: + movupd m1, [samplesq] + add samplesq, mmsize + pand m1, m2 + maxpd m0, m1 + dec nb_samplesd + jg .loop + movupd [ch_peaksq], m0 + shufpd m1, m0, m0, 1 + maxpd m0, m1 + movq rax, m0 + RET + %endif ; ARCH_X86_64 diff --git a/libavfilter/x86/f_ebur128_init.c b/libavfilter/x86/f_ebur128_init.c index 8f38aee967..4a7109746d 100644 --- a/libavfilter/x86/f_ebur128_init.c +++ b/libavfilter/x86/f_ebur128_init.c @@ -26,10 +26,15 @@ void ff_ebur128_filter_channels_avx(const EBUR128DSPContext *, const double *, double *, double *, double *, double *, int); -av_cold void ff_ebur128_init_x86(EBUR128DSPContext *dsp) +double ff_ebur128_find_peak_2ch_avx(double *, int, const double *, int); + +av_cold void ff_ebur128_init_x86(EBUR128DSPContext *dsp, int nb_channels) { int cpu_flags = av_get_cpu_flags(); - if (ARCH_X86_64 && EXTERNAL_AVX(cpu_flags)) + if (ARCH_X86_64 && EXTERNAL_AVX(cpu_flags)) { dsp->filter_channels = ff_ebur128_filter_channels_avx; + if (nb_channels == 2) + dsp->find_peak = ff_ebur128_find_peak_2ch_avx; + } } _______________________________________________ ffmpeg-cvslog mailing list ffmpeg-cvslog@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-cvslog To unsubscribe, visit link above, or email ffmpeg-cvslog-requ...@ffmpeg.org with subject "unsubscribe".
