1 files changed, 5 insertions, 5 deletions
diff --git a/ggml-cuda/dequantize.cuh b/ggml-cuda/dequantize.cuh
index b5440063..bd3c2d9d 100644
--- a/ggml-cuda/dequantize.cuh
+++ b/ggml-cuda/dequantize.cuh
@@ -1,6 +1,6 @@
 #include "common.cuh"
 
-static __device__ __forceinline__ void dequantize_q4_0(const void * vx, const int ib, const int iqs, dfloat2 & v){
+static __device__ __forceinline__ void dequantize_q4_0(const void * vx, const int64_t ib, const int iqs, dfloat2 & v){
     const block_q4_0 * x = (const block_q4_0 *) vx;
 
     const dfloat d = x[ib].d;
@@ -19,7 +19,7 @@ static __device__ __forceinline__ void dequantize_q4_0(const void * vx, const in
 #endif // GGML_CUDA_F16
 }
 
-static __device__ __forceinline__ void dequantize_q4_1(const void * vx, const int ib, const int iqs, dfloat2 & v){
+static __device__ __forceinline__ void dequantize_q4_1(const void * vx, const int64_t ib, const int iqs, dfloat2 & v){
     const block_q4_1 * x = (const block_q4_1 *) vx;
 
     const dfloat d = __low2half(x[ib].dm);
@@ -39,7 +39,7 @@ static __device__ __forceinline__ void dequantize_q4_1(const void * vx, const in
 #endif // GGML_CUDA_F16
 }
 
-static __device__ __forceinline__ void dequantize_q5_0(const void * vx, const int ib, const int iqs, dfloat2 & v){
+static __device__ __forceinline__ void dequantize_q5_0(const void * vx, const int64_t ib, const int iqs, dfloat2 & v){
     const block_q5_0 * x = (const block_q5_0 *) vx;
 
     const dfloat d = x[ib].d;
@@ -62,7 +62,7 @@ static __device__ __forceinline__ void dequantize_q5_0(const void * vx, const in
 #endif // GGML_CUDA_F16
 }
 
-static __device__ __forceinline__ void dequantize_q5_1(const void * vx, const int ib, const int iqs, dfloat2 & v){
+static __device__ __forceinline__ void dequantize_q5_1(const void * vx, const int64_t ib, const int iqs, dfloat2 & v){
     const block_q5_1 * x = (const block_q5_1 *) vx;
 
     const dfloat d = __low2half(x[ib].dm);
@@ -86,7 +86,7 @@ static __device__ __forceinline__ void dequantize_q5_1(const void * vx, const in
 #endif // GGML_CUDA_F16
 }
 
-static __device__ __forceinline__ void dequantize_q8_0(const void * vx, const int ib, const int iqs, dfloat2 & v){
+static __device__ __forceinline__ void dequantize_q8_0(const void * vx, const int64_t ib, const int iqs, dfloat2 & v){
     const block_q8_0 * x = (const block_q8_0 *) vx;
 
     const dfloat d = x[ib].d;