Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
jerrrrry
infinicore
Commits
9a0f2505
Commit
9a0f2505
authored
Mar 05, 2026
by
wooway777
Browse files
issue/1050 - fix paged caching and paged prefill on metax
parent
a9503148
Changes
3
Expand all
Show whitespace changes
Inline
Side-by-side
Showing
3 changed files
with
73 additions
and
73 deletions
+73
-73
src/infiniop/ops/paged_attention_prefill/metax/paged_attention_prefill_metax.maca
...ttention_prefill/metax/paged_attention_prefill_metax.maca
+60
-58
src/infiniop/ops/paged_attention_prefill/nvidia/paged_attention_prefill_nvidia.cu
...ttention_prefill/nvidia/paged_attention_prefill_nvidia.cu
+12
-14
src/infiniop/ops/paged_caching/metax/paged_caching_metax.maca
...infiniop/ops/paged_caching/metax/paged_caching_metax.maca
+1
-1
No files found.
src/infiniop/ops/paged_attention_prefill/metax/paged_attention_prefill_metax.maca
View file @
9a0f2505
This diff is collapsed.
Click to expand it.
src/infiniop/ops/paged_attention_prefill/nvidia/paged_attention_prefill_nvidia.cu
View file @
9a0f2505
This diff is collapsed.
Click to expand it.
src/infiniop/ops/paged_caching/metax/paged_caching_metax.maca
View file @
9a0f2505
...
@@ -12,7 +12,7 @@ INFINIOP_METAX_KERNEL pagedCaching(
...
@@ -12,7 +12,7 @@ INFINIOP_METAX_KERNEL pagedCaching(
const ptrdiff_t k_src_stride, const ptrdiff_t v_src_stride,
const ptrdiff_t k_src_stride, const ptrdiff_t v_src_stride,
const ptrdiff_t k_cache_block_stride, const ptrdiff_t v_cache_block_stride,
const ptrdiff_t k_cache_block_stride, const ptrdiff_t v_cache_block_stride,
const ptrdiff_t k_cache_head_stride, const ptrdiff_t v_cache_head_stride,
const ptrdiff_t k_cache_head_stride, const ptrdiff_t v_cache_head_stride,
const ptrdiff_t k_cache_slot_stride, const ptrdiff_t v_cache_slot_strid) {
const ptrdiff_t k_cache_slot_stride, const ptrdiff_t v_cache_slot_strid
e
) {
op::paged_caching::cuda::pagedCachingKernel<Tdata, NUM_THREADS>(
op::paged_caching::cuda::pagedCachingKernel<Tdata, NUM_THREADS>(
k_cache, v_cache, k, v, slot_mapping, head_size,
k_cache, v_cache, k, v, slot_mapping, head_size,
block_size, k_src_stride, v_src_stride,
block_size, k_src_stride, v_src_stride,
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment