Skip to content

LLM inference in C/C++

C++ ggml

webgpu : reorder includes since V that appears in common_decls.tmpl may be defined as K in flash_attn_decls.tmpl if KV_OVERLAP (#27545)

fairydreamingc1c766da59ba
diff --git a/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn.wgsl b/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn.wgsl
index d5bf2af8d2..a7dee65128 100644
--- a/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn.wgsl
+++ b/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn.wgsl
@@ -5,10 +5,9 @@ enable subgroups;
enable chromium_experimental_subgroup_matrix;
#define BYTE_HELPERS
-#include "common_decls.tmpl"
-
#define FLASH_ATTN_SCALAR_KV
#include "flash_attn_decls.tmpl"
+#include "common_decls.tmpl"
// Default values
// The actual values are defined in shader-lib.
diff --git a/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_tile.wgsl b/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_tile.wgsl
index 8cd18b9218..7edca84fc0 100644
--- a/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_tile.wgsl
+++ b/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_tile.wgsl
@@ -2,8 +2,8 @@ enable f16;
enable subgroups;
#define BYTE_HELPERS
-#include "common_decls.tmpl"
#include "flash_attn_decls.tmpl"
+#include "common_decls.tmpl"
// Default values
// The actual values are defined in shader-lib.
diff --git a/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_vec_split.wgsl b/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_vec_split.wgsl
index 42f3b10890..ae941245cd 100644
--- a/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_vec_split.wgsl
+++ b/ggml/src/ggml-webgpu/wgsl-shaders/flash_attn_vec_split.wgsl
@@ -3,9 +3,9 @@ enable f16;
enable subgroups;
#define BYTE_HELPERS
-#include "common_decls.tmpl"
#define FLASH_ATTN_VEC_SPLIT
#include "flash_attn_decls.tmpl"
+#include "common_decls.tmpl"
// Default values
// The actual values are defined in shader-lib.