]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
ggml: fix cuda kernel launch configuration for k_compute_batched_ptrs to support...
authorleejet <redacted>
Sun, 26 Oct 2025 18:13:31 +0000 (02:13 +0800)
committerGeorgi Gerganov <redacted>
Sat, 1 Nov 2025 07:41:35 +0000 (09:41 +0200)
commitebc499315cfac3b56ccebe3fdd9f006989815d28
treec8342dcb75149b624059e7beaedcdb30485f9849
parentd8ba8a34a68048f5b667017e776d168f54daee26
ggml: fix cuda kernel launch configuration for k_compute_batched_ptrs to support large batch (llama/16744)

* fix k_compute_batched_ptrs

* add backend ops test

* Update ggml/src/ggml-cuda/ggml-cuda.cu

Co-authored-by: Johannes Gäßler <redacted>
* reduce the batch size

---------

Co-authored-by: Johannes Gäßler <redacted>
src/ggml-cuda/ggml-cuda.cu
tests/test-backend-ops.cpp