Le sunnuntaina 28. heinäkuuta 2024, 20.10.51 EEST Nathan E. Egge a écrit : > The implementation of ff_read_time() for RISC-V uses rdtime which has > precision on existing hardware too low (!) for benchmarking purposes. > Deleting this implementation falls back on clock_gettime() which was > added as the default ff_read_time() implementation in 33e4cc9.
So the results seem fairly stable, unlike the Linux kperf cycle count, so I am fine with this. Except that if we are switching to nanoseconds, the new FF_TIMER_UNITS value should be "ns", not "UNITS". -- レミ・デニ-クールモン http://www.remlab.net/ _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
