PulseAugur
EN
LIVE 18:48:55

llama.cpp b12245 fixes iGPU lazy-mode fallback bug

By PulseAugur Editorial · [1 sources] ·

The llama.cpp project has released version b12245, which addresses an issue where the iGPU lazy-mode fallback was incorrectly applied when mmap was disabled. This fix ensures that llama_prefetch_rows functions correctly by only disabling lazy-mode when mmap is explicitly turned off, preventing inefficient page-by-page faulting of large tensor rows. The change was tagged by oscarmherrera and resolves issue #31943. AI

IMPACT Improves performance and stability for users of the llama.cpp project, particularly those utilizing integrated graphics.

RANK_REASON This is a software update for a specific tool, not a frontier release or significant industry event.

Read on llama.cpp — Releases →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

llama.cpp b12245 fixes iGPU lazy-mode fallback bug

How we ranked this

Signal score
3 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
This is a software update for a specific tool, not a frontier release or significant industry event.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Same-day
Cluster formed today. Ranking reflects the current source set at time of score.

Full methodology in our editorial standards.

COVERAGE [1]

  1. llama.cpp — Releases TIER_1 English(EN) · oscarmherrera ·

    b12245: llama : apply the iGPU lazy-mode fallback only when mmap is off (#31957)

    <p>On a device without mmap support, --lazy-mode auto resolved to off even<br /> with --load-mode mmap, so llama_prefetch_rows never ran and the rows of<br /> the large tensors faulted in one page at a time. Resolve it to off only<br /> when mmap is off; the default iGPU load pat…

RELATED ENTITIES

RELATED TOPICS