From 7f56abab92ebdd2c0638668410fe8a6b51653eec Mon Sep 17 00:00:00 2001
From: Masashi Yoshimura <yoshimura.masashi.frbs@gmail.com>
Date: Wed, 7 Oct 2026 16:56:53 +0900
Subject: [PATCH] ggml-webgpu: fix flash_attn supports_op check for overlapping
 KV (llama/28205)

---
 ggml/src/ggml-webgpu/ggml-webgpu.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ggml/src/ggml-webgpu/ggml-webgpu.cpp b/ggml/src/ggml-webgpu/ggml-webgpu.cpp
index ebc8a5e78..11a4fc46f 100644
--- a/ggml/src/ggml-webgpu/ggml-webgpu.cpp
+++ b/ggml/src/ggml-webgpu/ggml-webgpu.cpp
@@ -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;
                 }