optimize RGBA to BGRA conversion #234
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
qbnu
force-pushed
the
optimize-format-conversion
branch
from
d386db8
to
db294bf
Compare
sylikc
reviewed
Sep 23, 2023
| *i = ((*i & 0x00FF0000) >> 16) | ((*i & 0x0000FF00)) | ((*i & 0x000000FF) << 16) | ((*i & 0xFF000000)); | ||
| uint32_t* data = (uint32_t*)pixels.data(); | ||
| for (int i = 0; i * sizeof(uint32_t) < size; i++) { | ||
| ((uint32_t*)pPixelData)[i] = _rotr(_byteswap_ulong(data[i]), 8); |
There was a problem hiding this comment.
This is an amazing optimization. wow.
sylikc
reviewed
Sep 23, 2023
| o += nchannels; | ||
| *o = _rotr(_byteswap_ulong(*p), 8); | ||
| p++; | ||
| o++; |
There was a problem hiding this comment.
I've read (possibly incorrectly) that preincrement is faster than postincrement.
There was a problem hiding this comment.
I think the compiler will optimize it anyway
sylikc
added a commit
that referenced
this pull request
Sep 23, 2023
Speed up RGBA -> BGRA conversion by ~25% for JXL and ~150% for HEIF Merge PR #234 by https://github.com/qbnu
sylikc
added
the
format support
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Speed up RGBA -> BGRA conversion by ~25% for JXL and ~150% for HEIF.
Slow:
((value & 0x00FF0000) >> 16) | ((value & 0x000000FF) << 16) | (value & 0xFF00FF00);Faster:
_rotr(value & 0x00FF00FF, 16) | (value & 0xFF00FF00);(((value >> 16) | (value << 16)) & 0x00FF00FF) | (value & 0xFF00FF00);((_byteswap_ulong(value) >> 8) & 0x00FF00FF) | (value & 0xFF00FF00);Fastest:
_rotr(_byteswap_ulong(value), 8);_byteswap_ulong(_rotl(value, 8));I also got rid of the unnecessary
memcpys.For me this saves 50-100ms for large files.