Chiebot-Mirror/FFmpeg - FFmpeg - Gitea: Git with a cup of tea

Author	SHA1	Message	Date
Rémi Denis-Courmont	f2c30fe15a	lavc/riscv: add forward-edge CFI landing pads	7 months ago
Rémi Denis-Courmont	187d4d066a	lavc/riscv: require B or zba explicitly	7 months ago
Rémi Denis-Courmont	4fe8f2cc43	riscv: allow passing addend to vtype_vli macro A constant (-1) is added to the length value, so we can have an added for free, and optimise the addition away if the addend is exactly 1.	9 months ago
Rémi Denis-Courmont	c53d42380d	lavc/lpc: optimise RVV vector type for compute_autocorr On SpacemiT X60 (with len == 4000): autocorr_10_c: 2303.7 autocorr_10_rvv_f64: 1411.5 (before) autocorr_10_rvv_f64: 842.2 (after)	9 months ago
Rémi Denis-Courmont	af20fb9c4e	lavc/lpc: fix off-by-one in R-V V compute_autocorr	9 months ago
Rémi Denis-Courmont	918b3ed2d5	lavc/lpc: R-V V compute_autocorr The loop iterates over the length of the vector, not the order. This is to avoid reloading the same data for each lag value. However this means the loop only works if the maximum order is no larger than VLENB. The loop is roughly equivalent to: for (size_t j = 0; j < lag; j++) autoc[j] = 1.; while (len > lag) { for (ptrdiff_t j = 0; j < lag; j++) autoc[j] += data[j] * data; data++; len--; } while (len > 0) { for (ptrdiff_t j = 0; j < len; j++) autoc[j] += data[j] *data; data++; len--; } Since register pressure is only at 50%, it should be possible to implement the same loop for order up to 2xVLENB. But this is left for future work. Performance numbers are all over the place from ~1.25x to ~4x speedups, but at least they are always noticeably better than nothing.	1 year ago
Rémi Denis-Courmont	272d0c164d	lavc/lpc: R-V V apply_welch_window apply_welch_window_even_c: 617.5 apply_welch_window_even_rvv_f64: 235.0 apply_welch_window_odd_c: 709.0 apply_welch_window_odd_rvv_f64: 256.5	1 year ago