Dequant improvements rebase #8255
Merged
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
JohannesGaessler
added
the
SYCL
OuadiElfarouki
approved these changes
Jul 2, 2024
Contributor
OuadiElfarouki
left a comment
OuadiElfarouki
left a comment
There was a problem hiding this comment.
Improvement observed on Nvidia A4000 & RTX 4070 as well (7B & 13B - Q4_K_*).
Thanks!
Contributor
|
Ping @airMeng to check for regressions on Intel side |
airMeng
approved these changes
Jul 2, 2024
Contributor
There was a problem hiding this comment.
It is weird that I don't see any performance improvements on Arc A770, no regression either.
ping our performance expert @luoyu-intel
Contributor
|
The code looks good. It can prevent cache misses. So you may not see the performance improvement if there are no cache misses in your case. |
arthw
pushed a commit
to arthw/llama.cpp
that referenced
this pull request
Jul 3, 2024
* Single load for half2 * Store scales in local mem * Vec load quantized values
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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Outdated suggestions cannot be applied.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 PR provides improvements to the
dequantize_block_q4_Kkernel. It focuses on improving the global memory accesses.Three main changes are implemented:
All results below collected on A100 GPU
No meaningful change in Intel GPU results have been observed.