From: Georgi Gerganov <redacted>
Date: Sun, 8 Sep 2024 19:01:02 +0000 (+0300)
Subject: cuda : fix FA Q src index (1 -> 0) (llama/9374)
X-Git-Tag: upstream/1.7.4~420
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=26225f1fb0d8a00189317830c64edca0b78f17b8;p=pkg%2Fggml%2Fsources%2Fwhisper.cpp

cuda : fix FA Q src index (1 -> 0) (llama/9374)
---

diff --git a/ggml/src/ggml-cuda/fattn.cu b/ggml/src/ggml-cuda/fattn.cu
index f87f33b3..f28a19d4 100644
--- a/ggml/src/ggml-cuda/fattn.cu
+++ b/ggml/src/ggml-cuda/fattn.cu
@@ -152,7 +152,7 @@ static void ggml_cuda_flash_attn_ext_wmma_f16(ggml_backend_cuda_context & ctx, g
     }                                                                       \
 
 static void ggml_cuda_flash_attn_ext_vec_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
-    ggml_tensor * Q = dst->src[1];
+    ggml_tensor * Q = dst->src[0];
     ggml_tensor * K = dst->src[1];
     ggml_tensor * V = dst->src[2];
 
@@ -227,7 +227,7 @@ static void ggml_cuda_flash_attn_ext_vec_f16(ggml_backend_cuda_context & ctx, gg
     }                                                                       \
 
 static void ggml_cuda_flash_attn_ext_vec_f32(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
-    ggml_tensor * Q = dst->src[1];
+    ggml_tensor * Q = dst->src[0];
     ggml_tensor * K = dst->src[1];
     ggml_tensor * V = dst->src[2];