diff --git a/sycl/include/CL/sycl/detail/spirv.hpp b/sycl/include/CL/sycl/detail/spirv.hpp
index 6294bf89d928f..92d5f7499c603 100644
--- a/sycl/include/CL/sycl/detail/spirv.hpp
+++ b/sycl/include/CL/sycl/detail/spirv.hpp
@@ -55,7 +55,7 @@ template <typename T, typename Functor>
 void GenericCall(const Functor &ApplyToBytes) {
   if (sizeof(T) >= sizeof(ShuffleChunkT)) {
 #pragma unroll
-    for (size_t Offset = 0; Offset < sizeof(T);
+    for (size_t Offset = 0; Offset + sizeof(ShuffleChunkT) <= sizeof(T);
          Offset += sizeof(ShuffleChunkT)) {
       ApplyToBytes(Offset, sizeof(ShuffleChunkT));
     }
@@ -160,9 +160,9 @@ EnableIfGenericBroadcast<T, IdT> GroupBroadcast(T x, IdT local_id) {
   char *ResultBytes = reinterpret_cast<char *>(&Result);
   auto BroadcastBytes = [=](size_t Offset, size_t Size) {
     uint64_t BroadcastX, BroadcastResult;
-    detail::memcpy(&BroadcastX, XBytes + Offset, Size);
+    std::memcpy(&BroadcastX, XBytes + Offset, Size);
     BroadcastResult = GroupBroadcast<Group>(BroadcastX, local_id);
-    detail::memcpy(ResultBytes + Offset, &BroadcastResult, Size);
+    std::memcpy(ResultBytes + Offset, &BroadcastResult, Size);
   };
   GenericCall<T>(BroadcastBytes);
   return Result;
@@ -213,9 +213,9 @@ EnableIfGenericBroadcast<T> GroupBroadcast(T x, id<Dimensions> local_id) {
   char *ResultBytes = reinterpret_cast<char *>(&Result);
   auto BroadcastBytes = [=](size_t Offset, size_t Size) {
     uint64_t BroadcastX, BroadcastResult;
-    detail::memcpy(&BroadcastX, XBytes + Offset, Size);
+    std::memcpy(&BroadcastX, XBytes + Offset, Size);
     BroadcastResult = GroupBroadcast<Group>(BroadcastX, local_id);
-    detail::memcpy(ResultBytes + Offset, &BroadcastResult, Size);
+    std::memcpy(ResultBytes + Offset, &BroadcastResult, Size);
   };
   GenericCall<T>(BroadcastBytes);
   return Result;
@@ -697,9 +697,9 @@ EnableIfGenericShuffle<T> SubgroupShuffle(T x, id<1> local_id) {
   char *ResultBytes = reinterpret_cast<char *>(&Result);
   auto ShuffleBytes = [=](size_t Offset, size_t Size) {
     ShuffleChunkT ShuffleX, ShuffleResult;
-    detail::memcpy(&ShuffleX, XBytes + Offset, Size);
+    std::memcpy(&ShuffleX, XBytes + Offset, Size);
     ShuffleResult = SubgroupShuffle(ShuffleX, local_id);
-    detail::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
+    std::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
   };
   GenericCall<T>(ShuffleBytes);
   return Result;
@@ -712,9 +712,9 @@ EnableIfGenericShuffle<T> SubgroupShuffleXor(T x, id<1> local_id) {
   char *ResultBytes = reinterpret_cast<char *>(&Result);
   auto ShuffleBytes = [=](size_t Offset, size_t Size) {
     ShuffleChunkT ShuffleX, ShuffleResult;
-    detail::memcpy(&ShuffleX, XBytes + Offset, Size);
+    std::memcpy(&ShuffleX, XBytes + Offset, Size);
     ShuffleResult = SubgroupShuffleXor(ShuffleX, local_id);
-    detail::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
+    std::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
   };
   GenericCall<T>(ShuffleBytes);
   return Result;
@@ -727,9 +727,9 @@ EnableIfGenericShuffle<T> SubgroupShuffleDown(T x, id<1> local_id) {
   char *ResultBytes = reinterpret_cast<char *>(&Result);
   auto ShuffleBytes = [=](size_t Offset, size_t Size) {
     ShuffleChunkT ShuffleX, ShuffleResult;
-    detail::memcpy(&ShuffleX, XBytes + Offset, Size);
+    std::memcpy(&ShuffleX, XBytes + Offset, Size);
     ShuffleResult = SubgroupShuffleDown(ShuffleX, local_id);
-    detail::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
+    std::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
   };
   GenericCall<T>(ShuffleBytes);
   return Result;
@@ -742,9 +742,9 @@ EnableIfGenericShuffle<T> SubgroupShuffleUp(T x, id<1> local_id) {
   char *ResultBytes = reinterpret_cast<char *>(&Result);
   auto ShuffleBytes = [=](size_t Offset, size_t Size) {
     ShuffleChunkT ShuffleX, ShuffleResult;
-    detail::memcpy(&ShuffleX, XBytes + Offset, Size);
+    std::memcpy(&ShuffleX, XBytes + Offset, Size);
     ShuffleResult = SubgroupShuffleUp(ShuffleX, local_id);
-    detail::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
+    std::memcpy(ResultBytes + Offset, &ShuffleResult, Size);
   };
   GenericCall<T>(ShuffleBytes);
   return Result;