UPSTREAM PR #18266: llama: fix magic number of 999 for GPU layers#653
Closed
UPSTREAM PR #18266: llama: fix magic number of 999 for GPU layers#653
Conversation
loci-dev
had a problem deploying
to
PROD__AL_DEMO
GitHub Actions
Failure
loci-dev
force-pushed
the
main
branch
5 times, most recently
from
26a6f0f to
cf53bc9
Compare
DajanaV
deleted the
upstream-PR18266-branch_JohannesGaessler-llama-fp-ngl-magic-number
branch
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Mirrored from ggml-org/llama.cpp#18266
Fixes ggml-org/llama.cpp#18258 .
As of right now
llama_params_fitis disabled if the user sets a values other than the default values. However, because the default value for the number of GPU layers is 999 this is a value that a user could feasibly set manually without this being recognized by the code. This PR makes it so that the llama API explicitly recognizes negative values forn_gpu_layersto mean all layers should be put on the GPU, the default is changed to -1. For CLI arguments a value of -1 meansllama_params_fit, other negative values mean all layers on GPU.Also this PR fixes a minor bug where a print in
llama_params_fitwhere the sum of total memory rather than the sum of free memory is printed.