]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
vulkan: fix coopmat2 flash attention for non-contiguous inputs (llama/11281)
authorJeff Bolz <redacted>
Sat, 18 Jan 2025 08:26:50 +0000 (02:26 -0600)
committerGeorgi Gerganov <redacted>
Mon, 3 Feb 2025 20:00:57 +0000 (22:00 +0200)
commit668306ff2ba2c5f46c4851b370bd15962be3fa91
treecf23c9ba662b81c81ddc1fbcdc49a824a66a6845
parentfdc21fc87b9ad152d76bede79a98313c4ceb4ac2
vulkan: fix coopmat2 flash attention for non-contiguous inputs (llama/11281)

Add code similar to mul_mm_cm2 to force alignment of strides, to avoid
a performance regression.

Add noncontiguous FA tests in test-backend-ops.

Fixes #11268.
ggml/src/ggml-vulkan/ggml-vulkan.cpp
ggml/src/ggml-vulkan/vulkan-shaders/flash_attn_cm2.comp