github llama.cpp · just now · inference

llama.cpp b11302 fixes ThreadSanitizer data race in sparse indexer

from English

The llama.cpp project released build b11302, which addresses a critical concurrency issue identified by ThreadSanitizer in the CI pipeline. The update resolves a data race within the sparse attention mechanism where multiple CPU threads were incorrectly writing to the same memory elements.

Dead indexer slots are now assigned unique scatter rows (n_kv + slot) instead of sharing a single sentinel row.
Live slots address disjoint cells, ensuring that scatter indices for a token remain unique across all selection paths.
The fix eliminates overlaps between invisible pools selected by top_k and the tail cells of tokens.

This change ensures thread-safe operation of the sparse indexer mask construction, preventing undefined behavior caused by concurrent writes to shared memory locations.

Importance 1/3 Trust 1/3 llama.cpp Inference efficiency Local inference

Read original

github llama.cpp · just now Live

llama.cpp b11301 fixes integer overflow in ggml/gguf

The llama.cpp project has released build b11301, which includes a fix for an integer overflow issue within the ggml and gguf components. This update addresses validation logic to prevent potential errors when handling zero-element tensors.

github llama.cpp · 1h ago · 1 view Live

llama.cpp b11298 adds dflash support to convert and feature extraction

The llama.cpp project has released build b11298, which introduces support for the dflash format. This update enables both conversion of models to this format and subsequent feature extraction capabilities.

github llama.cpp · 8h ago · 2 views

llama.cpp b11284 fixes OpenVINO GET_ROWS on quantized weight views

The llama.cpp project released build b11284, which includes a fix for the OpenVINO backend to correctly handle GET_ROWS operations on quantized weight views.

github llama.cpp · 11h ago · 2 views

llama.cpp adds GLM-5.3-Flash support with k-pool and MTP fixes

The llama.cpp project has added initial support for the GLM-5.3-Flash (GLM5-Next) model architecture, including migration to hybrid-index memory and early Multi-Token Prediction (MTP) capabilities.

github llama.cpp · 14h ago · 2 views

llama.cpp b11272 optimizes Hexagon concat op

The llama.cpp project released build b11272, which includes optimizations for the Hexagon backend. The primary change focuses on optimizing the concatenation operation to improve performance on Qualcomm Snapdragon hardware.