Re: [FFmpeg-devel] [PATCH] lavc/lpc: R-V V apply_welch_window

2023-12-11 Thread Rémi Denis-Courmont
Le 11 décembre 2023 11:57:50 GMT+02:00, Anton Khirnov a écrit : >Quoting Rémi Denis-Courmont (2023-12-11 10:50:53) >> Le 11 décembre 2023 11:11:28 GMT+02:00, Anton Khirnov a >> écrit : >> >I think it'd look a lot less like base64 < /dev/random if you vertically >> >aligned the first operands.

Re: [FFmpeg-devel] [PATCH] lavc/lpc: R-V V apply_welch_window

2023-12-11 Thread Anton Khirnov
Quoting Rémi Denis-Courmont (2023-12-11 10:50:53) > Le 11 décembre 2023 11:11:28 GMT+02:00, Anton Khirnov a > écrit : > >I think it'd look a lot less like base64 < /dev/random if you vertically > >aligned the first operands. > > They are aligned to the 17th column. Problem is that quite a few ve

Re: [FFmpeg-devel] [PATCH] lavc/lpc: R-V V apply_welch_window

2023-12-11 Thread Rémi Denis-Courmont
Le 11 décembre 2023 11:11:28 GMT+02:00, Anton Khirnov a écrit : >Quoting Rémi Denis-Courmont (2023-12-08 18:46:51) >> +#if __riscv_xlen >= 64 >> +func ff_lpc_apply_welch_window_rvv, zve64d >> +vsetvli t0, zero, e64, m8, ta, ma >> +vid.v v0 >> +addit2, a1, -1 >> +

Re: [FFmpeg-devel] [PATCH] lavc/lpc: R-V V apply_welch_window

2023-12-11 Thread Anton Khirnov
Quoting Rémi Denis-Courmont (2023-12-08 18:46:51) > +#if __riscv_xlen >= 64 > +func ff_lpc_apply_welch_window_rvv, zve64d > +vsetvli t0, zero, e64, m8, ta, ma > +vid.v v0 > +addit2, a1, -1 > +vfcvt.f.xu.v v0, v0 > +li t3, 2 > +fcvt.d.l ft2,

[FFmpeg-devel] [PATCH] lavc/lpc: R-V V apply_welch_window

2023-12-08 Thread Rémi Denis-Courmont
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 --- libavcodec/lpc.c| 4 +- libavcodec/lpc.h| 1 + libavcodec/riscv/Makefile | 2 + libavcodec/riscv/lpc_init.c |