llama.cpp
vulkan: allow reducing graph submission batches to avoid device timeouts
#24872

Merged

vulkan: allow reducing graph submission batches to avoid device timeouts #24872

0cc4m merged 1 commit into ggml-org:master from wbruna:ggml_vk_reduce_submit_igpu

wbruna requested a review 62 days ago

github-actions added Vulkan

github-actions added ggml

0cc4m commented on 2026-06-21

wbruna force pushed from ff52c050 to 5e37770f 62 days ago

wbruna changed the title ~~vulkan: reduce graph submission batches on iGPUs to avoid timeouts~~ vulkan: allow reducing graph submission batches to avoid device timeouts 62 days ago

wbruna requested a review from

0cc4m 62 days ago

0cc4m approved these changes on 2026-06-23

vulkan: allow reducing the graph submission batches to avoid timeouts

0f0d2af1

wbruna force pushed from 5e37770f to 0f0d2af1 59 days ago

0cc4m approved these changes on 2026-06-24

taronaeo approved these changes on 2026-06-24

0cc4m merged 51eae8cf into master 59 days ago

Reviewers

taronaeo

0cc4m

Assignees

No one assigned

Labels

Vulkan ggml

Milestone

No milestone

llama.cpp vulkan: allow reducing graph submission batches to avoid device timeouts #24872 Merged

vulkan: allow reducing graph submission batches to avoid device timeouts #24872

llama.cpp
vulkan: allow reducing graph submission batches to avoid device timeouts
#24872

Merged