more optimization for ubuf_pic and upipe_set_color #869
Conversation
This file contains hidden or 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
…c_plane_set_color
…buf_pic_plane_set_color on x86 Using gcc 12 it manages to get compiled into a hot loop of: movups, add, cmp, jb. Increases the number of calls to ubuf_pic_clear from ~6200 to ~7000 per second. Other compilers are less successful.
…sics Compiles to what is desired on gcc as old as 4.8
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
This time I managed to get things right. Again "add a basic benchmark for ubuf_pic_clear" is not supposed to be committed.
It compiles to what I expect on gcc as far back as 4.8. That being a hot loop of 4 instructions: movdqu, add, cmp, conditional jump
The performance increase is much more modest than the original improvements. From ~6500 to ~7000 calls per second on an AMD Ryzen 7 3700X desktop and from ~2600 to ~2900 on an Intel Xeon E3-1245 v5 server and from ~2000 to ~2200 on an Intel Xeon CPU E3-1265L v3 server with just 1 memory channel populated.
@nto if your previous measurements were done on an x86 system would you like to look at this patch set?