aboutsummaryrefslogtreecommitdiffstats
path: root/tests/ref/fate/filter-pixdesc-xyz12le
diff options
context:
space:
mode:
authorSwinney, Jonathan <jswinney@amazon.com>2022-05-26 02:02:13 +0000
committerMartin Storsjö <martin@martin.st>2022-05-28 01:09:05 +0300
commit0ea61725b1bd35f47d0ebc49597e73e6798c553d (patch)
treeb3e4fc037138a450c0538bfc74cc4665b5c9fc3e /tests/ref/fate/filter-pixdesc-xyz12le
parent92ea8e03dfc17cc580e8c6e0fb1923a2c02f68aa (diff)
downloadffmpeg-0ea61725b1bd35f47d0ebc49597e73e6798c553d.tar.gz
swscale/aarch64: add hscale specializations
This patch adds code to support specializations of the hscale function and adds a specialization for filterSize == 4. ff_hscale8to15_4_neon is a complete rewrite. Since the main bottleneck here is loading the data from src, this data is loaded a whole block ahead and stored back to the stack to be loaded again with ld4. This arranges the data for most efficient use of the vector instructions and removes the need for completion adds at the end. The number of iterations of the C per iteration of the assembly is increased from 4 to 8, but because of the prefetching, there must be a special section without prefetching when dstW < 16. This improves speed on Graviton 2 (Neoverse N1) dramatically in the case where previously fs=8 would have been required. before: hscale_8_to_15__fs_8_dstW_512_neon: 1962.8 after : hscale_8_to_15__fs_4_dstW_512_neon: 1220.9 Signed-off-by: Jonathan Swinney <jswinney@amazon.com> Signed-off-by: Martin Storsjö <martin@martin.st>
Diffstat (limited to 'tests/ref/fate/filter-pixdesc-xyz12le')
0 files changed, 0 insertions, 0 deletions