Skip to content
Projects
Groups
Snippets
Help
Loading...
Sign in / Register
Toggle navigation
F
ffmpeg
Project
Project
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Packages
Packages
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
submodule
ffmpeg
Commits
277408b7
Commit
277408b7
authored
Mar 25, 2016
by
Clément Bœsch
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
sws/aarch64/yuv2rgb: save a few mul and add
27ms to 26ms with UHD 2160 input.
parent
99388eb0
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
11 additions
and
14 deletions
+11
-14
yuv2rgb_neon.S
libswscale/aarch64/yuv2rgb_neon.S
+11
-14
No files found.
libswscale/aarch64/yuv2rgb_neon.S
View file @
277408b7
...
@@ -143,20 +143,17 @@ function ff_\ifmt\()_to_\ofmt\()_neon, export=1
...
@@ -143,20 +143,17 @@ function ff_\ifmt\()_to_\ofmt\()_neon, export=1
load_chroma_\ifmt
load_chroma_\ifmt
sub v18.8H, v18.8H, v5.8H // U*(1<<3) - 128*(1<<3)
sub v18.8H, v18.8H, v5.8H // U*(1<<3) - 128*(1<<3)
sub v19.8H, v19.8H, v5.8H // V*(1<<3) - 128*(1<<3)
sub v19.8H, v19.8H, v5.8H // V*(1<<3) - 128*(1<<3)
zip1 v6.8H, v19.8H, v19.8H // V1
sqdmulh v20.8H, v19.8H, v1.H[0] // V * v2r (R)
zip2 v7.8H, v19.8H, v19.8H // V2
sqdmulh v22.8H, v18.8H, v1.H[1] // U * u2g
zip1 v4.8H, v18.8H, v18.8H // U1
sqdmulh v19.8H, v19.8H, v1.H[2] // V * v2g
zip2 v5.8H, v18.8H, v18.8H // U2
add v22.8H, v22.8H, v19.8H // U * u2g + V * v2g (G)
sqdmulh v20.8H, v6.8H, v1.H[0] // V1 * v2r (R1)
sqdmulh v24.8H, v18.8H, v1.H[3] // U * u2b (B)
sqdmulh v21.8H, v7.8H, v1.H[0] // V2 * v2r (R2)
zip2 v21.8H, v20.8H, v20.8H // R2
sqdmulh v22.8H, v4.8H, v1.H[1] // U1 * u2g
zip1 v20.8H, v20.8H, v20.8H // R1
sqdmulh v23.8H, v5.8H, v1.H[1] // U2 * u2g
zip2 v23.8H, v22.8H, v22.8H // G2
sqdmulh v6.8H, v6.8H, v1.H[2] // V1 * v2g
zip1 v22.8H, v22.8H, v22.8H // G1
sqdmulh v7.8H, v7.8H, v1.H[2] // V2 * v2g
zip2 v25.8H, v24.8H, v24.8H // B2
add v22.8H, v22.8H, v6.8H // U1 * u2g + V1 * v2g (G1)
zip1 v24.8H, v24.8H, v24.8H // B1
add v23.8H, v23.8H, v7.8H // U2 * u2g + V2 * v2g (G2)
sqdmulh v24.8H, v4.8H, v1.H[3] // U1 * u2b (B1)
sqdmulh v25.8H, v5.8H, v1.H[3] // U2 * u2b (B2)
ld1 {v2.16B}, [x4], #16 // load luma
ld1 {v2.16B}, [x4], #16 // load luma
ushll v26.8H, v2.8B, #3 // Y1*(1<<3)
ushll v26.8H, v2.8B, #3 // Y1*(1<<3)
ushll2 v27.8H, v2.16B, #3 // Y2*(1<<3)
ushll2 v27.8H, v2.16B, #3 // Y2*(1<<3)
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment