[rocm-libraries] ROCm/rocm-libraries#7111 (commit 651947f)

[CK] Fix latest batch of staging compiler warnings

## Motivation

Suppress the new batch of clang lifetimebound and invalidation warnings
with the latest staging compiler.

## Technical Details

<!-- Explain the changes along with any relevant GitHub links. -->

## Test Plan

<!-- Explain any relevant testing done to verify this PR. -->

## Test Result

<!-- Briefly summarize test outcomes. -->

## Submission Checklist

- [ ] Look over the contributing guidelines at
https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

include/ck/library/utility/gpu_verification.hpp

@@ -18,6 +18,8 @@
 #include "ck/host_utility/hip_check_error.hpp"
 #include "ck/library/utility/check_err.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace profiler {
 
@@ -517,3 +519,4 @@ float gpu_reduce_max(Iterator device_buffer, std::size_t size, hipStream_t strea
 
 } // namespace profiler
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_batched_contraction_multiple_d_wmma_cshuffle_v3.hpp

@@ -19,6 +19,8 @@
 #include "ck/host_utility/kernel_launch.hpp"
 #include "ck/utility/scheduler_enum.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 
 template <typename DeviceOp,
@@ -974,3 +976,4 @@ struct DeviceBatchedContractionMultipleD_Wmma_CShuffle_V3
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_batched_contraction_multiple_d_xdl_cshuffle.hpp

@@ -18,6 +18,8 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 
 template <typename GridwiseGemm,
@@ -1097,3 +1099,4 @@ struct DeviceBatchedContractionMultipleD_Xdl_CShuffle
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp

@@ -18,6 +18,8 @@
 #include "ck/host_utility/kernel_launch.hpp"
 #include "ck/host_utility/io.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -770,3 +772,4 @@ struct DeviceBatchedGemmMultiD_Xdl : public DeviceBatchedGemmMultiD<ALayout,
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_layernorm_wmma_cshuffle_v3.hpp

@@ -18,6 +18,8 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 
 template <typename GridwiseGemm,
@@ -919,3 +921,4 @@ struct DeviceGemmMultipleDLayernorm_Wmma_CShuffleV3
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_multiple_r_xdl_cshuffle.hpp

@@ -17,6 +17,8 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 
 template <typename GridwiseGemm,
@@ -712,3 +714,4 @@ struct DeviceGemmMultipleDMultipleR_Xdl_CShuffle
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_wmma_cshuffle_v3.hpp

@@ -18,6 +18,8 @@
 #include "ck/host_utility/flush_cache.hpp"
 #include "ck/tensor_operation/gpu/device/impl/device_gemm_wmma_cshuffle_v3_common.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -408,3 +410,4 @@ struct DeviceGemmMultipleD_Wmma_CShuffleV3
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp

@@ -19,6 +19,8 @@
 #include "ck/tensor_operation/gpu/device/matrix_padder.hpp"
 #include "ck/tensor_operation/gpu/grid/gridwise_gemm_multiple_d_xdl_cshuffle.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 
 template <typename GridwiseGemm,
@@ -941,3 +943,4 @@ struct DeviceGemmMultipleD_Xdl_CShuffle : public DeviceGemmMultipleD<ALayout,
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.hpp

@@ -34,6 +34,8 @@
 #include "ck_tile/builder/reflect/instance_traits_device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.hpp"
 #endif
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -2381,3 +2383,4 @@ struct DeviceGroupedConvFwdMultipleABD_Wmma_CShuffle_V3
 #ifdef CK_EXPERIMENTAL_BUILDER
 #include "ck_tile/builder/reflect/reflect_device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.inc"
 #endif
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp

@@ -33,6 +33,8 @@
 #include "ck_tile/builder/reflect/instance_traits_device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
 #endif
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -2233,3 +2235,4 @@ struct DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp

@@ -32,6 +32,8 @@
 #include "ck_tile/builder/reflect/instance_traits_device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp"
 #endif
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -2208,3 +2210,4 @@ struct DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_multiple_r_xdl_cshuffle.hpp

@@ -24,6 +24,8 @@
 #include "ck/host_utility/io.hpp"
 #include "ck/library/utility/numeric.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -1141,3 +1143,4 @@ struct DeviceGroupedConvFwdMultipleDMultipleR_Xdl_CShuffle
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_wmma_cshuffle.hpp

@@ -29,6 +29,8 @@
 #include "ck_tile/builder/reflect/instance_traits_device_grouped_conv_fwd_multiple_d_wmma_cshuffle.hpp"
 #endif
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -1031,3 +1033,4 @@ struct DeviceGroupedConvFwdMultipleD_Wmma_CShuffle
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_multiple_reduce_multiblock.hpp

@@ -17,6 +17,8 @@
 
 #include "ck/host_utility/kernel_launch.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -593,3 +595,4 @@ struct DeviceMultipleReduceMultiBlock : public DeviceMultipleReduce<Rank,
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_multiple_reduce_threadwise.hpp

@@ -16,6 +16,8 @@
 
 #include "ck/host_utility/kernel_launch.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -420,3 +422,4 @@ struct DeviceMultipleReduceThreadWise : public DeviceMultipleReduce<Rank,
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_permute_impl.hpp

@@ -17,6 +17,8 @@
 
 #include "ck/host_utility/kernel_launch.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 namespace tensor_operation {
 namespace device {
@@ -280,3 +282,4 @@ struct DevicePermuteImpl : DevicePermute<NumDim, InDataType, OutDataType, Elemen
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/device/impl/device_splitk_contraction_multiple_d_xdl_cshuffle.hpp

@@ -18,6 +18,8 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 
 template <typename GridwiseGemm,
@@ -1203,3 +1205,4 @@ struct DeviceSplitKContractionMultipleD_Xdl_CShuffle
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v3_multi_abd.hpp

@@ -14,6 +14,8 @@
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 #include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_common.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 
 // Currently we do not have a elegant way to put single lds buffer & double lds buffer pipe in same
@@ -1597,3 +1599,4 @@ struct GridwiseGemm_xdl_cshuffle_v3
 };
 
 } // namespace ck
+#pragma clang diagnostic pop

include/ck/utility/type_convert.hpp

@@ -13,6 +13,8 @@
 #include "ck/utility/amd_inline_asm.hpp"
 #include "ck/utility/type.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck {
 // Define the common macro for MI300 models
 #if defined(__gfx942__) || defined(__gfx950__)
@@ -2483,3 +2485,4 @@ inline __host__ __device__ void array_convert(Array<Y, NumElems>& y, const Array
 }
 
 } // namespace ck
+#pragma clang diagnostic pop

include/ck_tile/core/numeric/type_convert.hpp

@@ -13,6 +13,8 @@
 #include "ck_tile/core/numeric/int8.hpp"
 #include "ck_tile/core/numeric/mxfp_convert.hpp"
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck_tile {
 
 #if CK_TILE_USE_CUSTOM_DATA_TYPE
@@ -147,3 +149,4 @@ CK_TILE_SCALED_TYPE_CONVERT(fp16_t, fp16, pk_fp4_t, pk_fp4)
 #endif
 
 } // namespace ck_tile
+#pragma clang diagnostic pop

include/ck_tile/ref/naive_attention.hpp

@@ -9,6 +9,8 @@
 #include <thread>
 #include <string>
 
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wlifetime-safety-intra-tu-suggestions"
 namespace ck_tile {
 
 enum class naive_attention_layout_enum
@@ -824,3 +826,4 @@ CK_TILE_HOST float naive_attention_fwd(naive_attention_fwd_traits t,
 #undef CK_TILE_DISPATCH_NAIVE_ATTEN_FWD_INTERNAL_
 
 } // namespace ck_tile
+#pragma clang diagnostic pop