Fix time complexity of string replacement #9163
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
This change fixes a bug where replacing text in a very long string could cause llama.cpp to hang indefinitely. This is because the algorithm used was quadratic, due to memmove() when s.replace() is called in a loop. It seems most search results and LLM responses actually provide the O(n**2) algorithm, which is a great tragedy. Using a builder string fixes things
compilade
approved these changes
Aug 26, 2024
There was a problem hiding this comment.
Can confirm this helps when tokenizing a file with a million spaces with a sentencepiece tokenizer.
$ python3 -c 'print(" " * (2**20))' > spaces-1M.txt
$ ./build/bin/llama-tokenize -m models/ggml-vocab-llama-spm.gguf -f spaces-1M.txt > /dev/nullSeems to hang without this change, and perf says most time (99%+) is spent in memmove.
With this change, it finishes in at most a few seconds, and the perf report is much more balanced.
I've also tried a few test cases with the new replace_all, and it seems to behave correctly.
arthw
pushed a commit
to arthw/llama.cpp
that referenced
this pull request
Nov 15, 2024
This change fixes a bug where replacing text in a very long string could cause llama.cpp to hang indefinitely. This is because the algorithm used was quadratic, due to memmove() when s.replace() is called in a loop. It seems most search results and LLM responses actually provide the O(n**2) algorithm, which is a great tragedy. Using a builder string fixes things
arthw
pushed a commit
to arthw/llama.cpp
that referenced
this pull request
Nov 18, 2024
This change fixes a bug where replacing text in a very long string could cause llama.cpp to hang indefinitely. This is because the algorithm used was quadratic, due to memmove() when s.replace() is called in a loop. It seems most search results and LLM responses actually provide the O(n**2) algorithm, which is a great tragedy. Using a builder string fixes things
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.
This change fixes a bug where replacing text in a very long string could cause llama.cpp to hang indefinitely. This is because the algorithm used was quadratic, due to memmove() when s.replace() is called in a loop. It seems most search results and LLM responses actually provide the O(n**2) algorithm, which is a great tragedy. Using a builder string fixes things