Skip to content

Revert "cuda: reset cuda context after reading memory size (#23935)" - #24715

Merged
0cc4m merged 1 commit into
masterfrom
0cc4m/cuda-get-memory-device-reset-revert
Jun 17, 2026
Merged

Revert "cuda: reset cuda context after reading memory size (#23935)"#24715
0cc4m merged 1 commit into
masterfrom
0cc4m/cuda-get-memory-device-reset-revert

Conversation

@0cc4m

@0cc4m 0cc4m commented Jun 17, 2026

Copy link
Copy Markdown
Contributor

This reverts commit 0f7fada.

Overview

Sadly this change caused a bunch of unforeseen issues (#24608, #24682), so let's revert it for now and think of a different solution.

Requirements

@0cc4m
0cc4m requested a review from a team as a code owner June 17, 2026 05:17
@github-actions github-actions Bot added ggml changes relating to the ggml tensor library for machine learning CUDA Related to the CUDA backend labels Jun 17, 2026

@ORippler ORippler left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The general recommendation for CUDA is "Have 1 CUDA context per device for the lifetime of an application". It is quite monolithic in that regard unfortunately.

@0cc4m
0cc4m merged commit ea21e03 into master Jun 17, 2026
33 of 35 checks passed
@0cc4m
0cc4m deleted the 0cc4m/cuda-get-memory-device-reset-revert branch June 17, 2026 09:01
papamoose pushed a commit to papamoose/llama.cpp that referenced this pull request Jun 27, 2026
adrianhoehne pushed a commit to adrianhoehne/llama.cpp that referenced this pull request Jul 5, 2026
min23asdw pushed a commit to min23asdw/llama-cpp-turboquant that referenced this pull request Jul 27, 2026
TheTom added a commit to TheTom/llama-cpp-turboquant that referenced this pull request Jul 30, 2026
…del-reload

cuda: backport upstream revert of "reset cuda context after reading memory size" (ggml-org#24715)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants