mozempk/ggml-org-llama-cpp-mirror
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-10-14 08:50:37 +02:00
Code Issues Packages Projects Releases Wiki Activity

ggml-webgpu: fix flash_attn supports_op check for overlapping KV (#28205)

Browse Source
This commit is contained in:
Masashi Yoshimura
2026-10-07 09:56:53 +02:00
committed by GitHub
parent fa3c2fab36
commit 36a73916ee
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-webgpu/ggml-webgpu.cpp
+1 -1
View File
@@ -4544,7 +4544,7 @@ static bool ggml_backend_webgpu_device_supports_op(ggml_backend_dev_t dev, const
break;
}
if (ggml_webgpu_tensor_binding_overlap(ctx->webgpu_global_ctx, src1, src2) &&
src1->type != src2->type && !ggml_is_quantized(src1->type) && !ggml_is_quantized(src2->type)) {
src1->type != src2->type && !(ggml_is_quantized(src1->type) && ggml_is_quantized(src2->type))) {
supports_op = false;
break;
}