Possible fix: use ne0..ne3 (dst dims) instead of ne00..ne03 in ggml_compute_forward_dup_f16 #15626
+8
−8
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
github-actions
bot
added
the
ggml
|
Yes, looks like a bug. Can you try to add a |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
I might have found a small mistake in ggml/src/ggml-cpu/ops.cpp, in the non-contiguous path of ggml_compute_forward_dup_f16 when dst->type == GGML_TYPE_F16.
The carry after each element uses ne00/ne01/ne02/ne03 (source dims). Since the indices i10..i13 point to dst, I think it should compare against ne0/ne1/ne2/ne3 (destination dims). The adjacent F32 branch seems to use ne0..ne3.
Apologies if I’m missing context. If this intentional, feel free to close.