ggml : remove ggml_task_type and GGML_PERF#8017
Merged
Conversation
485df9a to
486d061
Compare
ggerganov
approved these changes
Jun 20, 2024
| if (!inplace) { | ||
| if (ith == 0) { | ||
| // memcpy needs to be synchronized across threads to avoid race conditions. | ||
| // => do it in INIT phase |
Member
There was a problem hiding this comment.
The comment about INIT phase is no longer relevant
Member
Author
|
The vulkan build is failing because it has some old code that still has references to |
1a119aa to
61b96a5
Compare
27 tasks
Nexesenex
added a commit
to Nexesenex/croco.cpp
that referenced
this pull request
Jun 26, 2024
This reverts commit 95f57bb.
MagnusS0
pushed a commit
to MagnusS0/llama.cpp-normistral-tokenizer
that referenced
this pull request
Jul 1, 2024
* ggml : remove ggml_task_type and GGML_PERF * check abort_callback on main thread only * vulkan : remove usage of ggml_compute_params * remove LLAMA_PERF
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Removes the phases in
ggml_task_typein favor of a single phase that operations can split into any number of phases by using barriers.Since the barriers require all threads to pass the barrier, operations are always called with the maximum number of threads, and they are responsible of skipping the excess threads if the implementation cannot use them.
Additionally, parallelizes the conversion of
src1tovec_dot_typeinmul_matandmul_mat_id. Since the threads are available regardless, there is no reason to not use them.Removes the
GGML_PERFoption and its related fields inggml_tensorand other structs.