aboutsummaryrefslogtreecommitdiffstats
path: root/libavutil/aarch64
diff options
context:
space:
mode:
authorRĂ©mi Denis-Courmont <remi@remlab.net>2022-10-05 19:12:56 +0300
committerLynne <dev@lynne.ee>2022-10-10 02:23:17 +0200
commitbfc69297c5d8d130b9cf1082dd1728b39a0e75f6 (patch)
treeda51678275cd33ed8ec44d51d4214b626baed1d8 /libavutil/aarch64
parent97d34befea598d34e92ed384acb3dced5490ae8a (diff)
downloadffmpeg-bfc69297c5d8d130b9cf1082dd1728b39a0e75f6.tar.gz
lavc/opusdsp: RISC-V V (512-bit) postfilter
This adds a variant of the postfilter for use with 512-bit vectors. Half a vector is enough to perform the scalar product. Normally a whole vector would be used anyhow. Indeed fractional multiplers are no faster than the unit multipler. But in this particular function, a full vector makes up 16 samples, which would be loaded at each iteration of the outer loop. The minimum guaranteed CELT postfilter period is only 15. Accounting for the edges, we can only safely preload up to 13 samples. The fractional multipler is thus used to cap the selected vector length to a safe value of 8 elements or 256 bits. Likewise, we have the 1024-bit variant with the quarter multipler. In theory, a 2048-bit one would be possible with the eigth multipler, but that length is not even defined in the specifications as of yet, nor is it supported by any emulator - forget actual hardware.
Diffstat (limited to 'libavutil/aarch64')
0 files changed, 0 insertions, 0 deletions