[Issue]: Building CK fails due to -fuse-ld=path deprecation on Fedora 42

[Aethor]

Problem Description

Trying to build CK results in the following error on Fedora 42:

clang++: error: '-fuse-ld=' taking a path is deprecated; use '--ld-path=' instead [-Werror,-Wfuse-ld-path]

The problem is the same as #2097, where the user is also on Fedora 42. However, I can confirm that this error does not appear when building with Docker, so it might be specific to the Fedora environment.

I am using Rocm 6.3.1

Operating System

Fedora 42

CPU

AMD Ryzen 7 7800X3D

GPU

Other

Other

AMD Radeon RX 9070 XT

ROCm Version

ROCm 6.3.1

ROCm Component

No response

Steps to Reproduce

run the following on Fedora 42 in the "build" folder:

cmake                                                                                             \
-D CMAKE_PREFIX_PATH=$(hipconfig -R)                                                                    \
-D CMAKE_CXX_COMPILER=$(hipconfig -R)/bin/hipcc                                                         \
-D CMAKE_BUILD_TYPE=Release                                                                       \
-D GPU_TARGETS=gfx1100                                                                    \
..

(Optional for Linux users) Output of /opt/rocm/bin/rocminfo --support

ROCk module is loaded

HSA System Attributes

Runtime Version: 1.1
Runtime Ext Version: 1.6
System Timestamp Freq.: 1000.000000MHz
Sig. Max Wait Duration: 18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count)
Machine Model: LARGE
System Endianness: LITTLE
Mwaitx: DISABLED
DMAbuf Support: YES

==========
HSA Agents

---

Agent 1

---

Name: AMD Ryzen 7 7800X3D 8-Core Processor
Uuid: CPU-XX
Marketing Name: AMD Ryzen 7 7800X3D 8-Core Processor
Vendor Name: CPU
Feature: None specified
Profile: FULL_PROFILE
Float Round Mode: NEAR
Max Queue Number: 0(0x0)
Queue Min Size: 0(0x0)
Queue Max Size: 0(0x0)
Queue Type: MULTI
Node: 0
Device Type: CPU
Cache Info:
L1: 32768(0x8000) KB
Chip ID: 0(0x0)
ASIC Revision: 0(0x0)
Cacheline Size: 64(0x40)
Max Clock Freq. (MHz): 5053
BDFID: 0
Internal Node ID: 0
Compute Unit: 16
SIMDs per CU: 0
Shader Engines: 0
Shader Arrs. per Eng.: 0
WatchPts on Addr. Ranges:1
Memory Properties:
Features: None
Pool Info:
Pool 1
Segment: GLOBAL; FLAGS: FINE GRAINED
Size: 32392232(0x1ee4428) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Recommended Granule:4KB
Alloc Alignment: 4KB
Accessible by all: TRUE
Pool 2
Segment: GLOBAL; FLAGS: EXTENDED FINE GRAINED
Size: 32392232(0x1ee4428) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Recommended Granule:4KB
Alloc Alignment: 4KB
Accessible by all: TRUE
Pool 3
Segment: GLOBAL; FLAGS: KERNARG, FINE GRAINED
Size: 32392232(0x1ee4428) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Recommended Granule:4KB
Alloc Alignment: 4KB
Accessible by all: TRUE
Pool 4
Segment: GLOBAL; FLAGS: COARSE GRAINED
Size: 32392232(0x1ee4428) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Recommended Granule:4KB
Alloc Alignment: 4KB
Accessible by all: TRUE
ISA Info:

---

Agent 2

---

Name: gfx1201
Uuid: GPU-8c7c68d970d211b8
Marketing Name: AMD Radeon RX 9070 XT
Vendor Name: AMD
Feature: KERNEL_DISPATCH
Profile: BASE_PROFILE
Float Round Mode: NEAR
Max Queue Number: 128(0x80)
Queue Min Size: 64(0x40)
Queue Max Size: 131072(0x20000)
Queue Type: MULTI
Node: 1
Device Type: GPU
Cache Info:
L1: 32(0x20) KB
L2: 8192(0x2000) KB
L3: 65536(0x10000) KB
Chip ID: 30032(0x7550)
ASIC Revision: 1(0x1)
Cacheline Size: 256(0x100)
Max Clock Freq. (MHz): 2460
BDFID: 768
Internal Node ID: 1
Compute Unit: 64
SIMDs per CU: 2
Shader Engines: 4
Shader Arrs. per Eng.: 2
WatchPts on Addr. Ranges:4
Coherent Host Access: FALSE
Memory Properties:
Features: KERNEL_DISPATCH
Fast F16 Operation: TRUE
Wavefront Size: 32(0x20)
Workgroup Max Size: 1024(0x400)
Workgroup Max Size per Dimension:
x 1024(0x400)
y 1024(0x400)
z 1024(0x400)
Max Waves Per CU: 32(0x20)
Max Work-item Per CU: 1024(0x400)
Grid Max Size: 4294967295(0xffffffff)
Grid Max Size per Dimension:
x 4294967295(0xffffffff)
y 4294967295(0xffffffff)
z 4294967295(0xffffffff)
Max fbarriers/Workgrp: 32
Packet Processor uCode:: 58
SDMA engine uCode:: 380
IOMMU Support:: None
Pool Info:
Pool 1
Segment: GLOBAL; FLAGS: COARSE GRAINED
Size: 16695296(0xfec000) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Recommended Granule:2048KB
Alloc Alignment: 4KB
Accessible by all: FALSE
Pool 2
Segment: GROUP
Size: 64(0x40) KB
Allocatable: FALSE
Alloc Granule: 0KB
Alloc Recommended Granule:0KB
Alloc Alignment: 0KB
Accessible by all: FALSE
ISA Info:
ISA 1
Name: amdgcn-amd-amdhsa--gfx1201
Machine Models: HSA_MACHINE_MODEL_LARGE
Profiles: HSA_PROFILE_BASE
Default Rounding Mode: NEAR
Default Rounding Mode: NEAR
Fast f16: TRUE
Workgroup Max Size: 1024(0x400)
Workgroup Max Size per Dimension:
x 1024(0x400)
y 1024(0x400)
z 1024(0x400)
Grid Max Size: 4294967295(0xffffffff)
Grid Max Size per Dimension:
x 4294967295(0xffffffff)
y 4294967295(0xffffffff)
z 4294967295(0xffffffff)
FBarrier Max Size: 32
*** Done ***

Additional Information

No response

[ppanchad-amd]

Hi @Aethor. Internal ticket has been created to investigate this issue. Thanks!

[lucbruni-amd]

Hi @Aethor, could you provide your ROCm installation steps?

Also, your clang version and full build log (which should include the clang version) will also be helpful. Thank you!

[Aethor]

Hi @lucbruni-amd, I installed rocm using the package on Fedora (sudo dnf install rocm rocm-devel as mentioned in the documentation).

By the clang version, do you mean the one reported by hipcc? In that case, $(hipconfig -R)/bin/hipcc --version gives me:

HIP version: 6.3.42133-0
clang version 18.0.0git
Target: x86_64-redhat-linux-gnu
Thread model: posix
InstalledDir: /usr/lib64/rocm/llvm/bin

Here are the build logs. This is the results of running cmake -D CMAKE_PREFIX_PATH=$(hipconfig -R) -D CMAKE_CXX_COMPILER=$(hipconfig -R)/bin/hipcc -D CMAKE_BUILD_TYPE=Release -D GPU_TARGETS=gfx1100 ..:

-- The CXX compiler identification is Clang 18.0.0
-- The HIP compiler identification is Clang 18.0.0
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/hipcc - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Detecting HIP compiler ABI info
-- Detecting HIP compiler ABI info - done
-- Check for working HIP compiler: /usr/lib64/rocm/llvm/bin/clang++ - skipped
-- Detecting HIP compile features
-- Detecting HIP compile features - done
-- Found Python3: /usr/bin/python3.13 (found suitable version "3.13.7", minimum required is "3.8") found components: Interpreter
-- Found Git: /usr/bin/git (found version "2.51.0")
-- GPU_TARGETS= gfx1100
-- GPU_ARCHS= 
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
-- Found Threads: TRUE
-- Performing Test HIP_CLANG_SUPPORTS_PARALLEL_JOBS
-- Performing Test HIP_CLANG_SUPPORTS_PARALLEL_JOBS - Success
-- hip_version_flat=600342133
-- checking which targets are supported
-- Performing Test COMPILER_HAS_TARGET_ID_gfx1100
-- Performing Test COMPILER_HAS_TARGET_ID_gfx1100 - Success
-- Building CK for the following targets: gfx1100
-- Enabling WMMA instances
-- Performing Test HAS_NO_OFFLOAD_UNIFORM_BLOCK
-- Performing Test HAS_NO_OFFLOAD_UNIFORM_BLOCK - Success
-- Adding the fno-offload-uniform-block compiler flag
-- Performing Test HAS_LSR_DROP_SOLUTION
-- Performing Test HAS_LSR_DROP_SOLUTION - Success
-- Adding the lsr-drop-solution=1 compiler flag
-- Performing Test HAS_ENABLE_POST_MISCHED
-- Performing Test HAS_ENABLE_POST_MISCHED - Success
-- Adding the enable-post-misched=0 compiler flag
-- Performing Test check-coerce
-- Performing Test check-coerce - Success
-- Adding the amdgpu-coerce-illegal-types=1
-- Adding -amdgpu-early-inline-all=true and -amdgpu-function-calls=false
-- CMAKE_CXX_COMPILER: /usr/bin/hipcc
-- CMAKE_HIP_COMPILER: /usr/bin/hipcc
-- OpenMP_CXX_LIB_NAMES: libomp;libgomp;libiomp5
-- OpenMP_gomp_LIBRARY: 
-- OpenMP_pthread_LIBRARY: 
-- OpenMP_CXX_FLAGS: -fopenmp=libomp -Wno-unused-command-line-argument
-- Build with HIP 
-- Clang tidy found: 20.1.8
-- Clang tidy checks: *,-abseil-*,-android-cloexec-fopen,-cert-msc30-c,-bugprone-exception-escape,-bugprone-macro-parentheses,-cert-env33-c,-cert-msc32-c,-cert-msc50-cpp,-cert-msc51-cpp,-cert-dcl37-c,-cert-dcl51-cpp,-clang-analyzer-alpha.core.CastToStruct,-clang-analyzer-optin.performance.Padding,-clang-diagnostic-deprecated-declarations,-clang-diagnostic-extern-c-compat,-clang-diagnostic-unused-command-line-argument,-cppcoreguidelines-avoid-c-arrays,-cppcoreguidelines-avoid-magic-numbers,-cppcoreguidelines-explicit-virtual-functions,-cppcoreguidelines-init-variables,-cppcoreguidelines-macro-usage,-cppcoreguidelines-non-private-member-variables-in-classes,-cppcoreguidelines-pro-bounds-array-to-pointer-decay,-cppcoreguidelines-pro-bounds-constant-array-index,-cppcoreguidelines-pro-bounds-pointer-arithmetic,-cppcoreguidelines-pro-type-member-init,-cppcoreguidelines-pro-type-reinterpret-cast,-cppcoreguidelines-pro-type-union-access,-cppcoreguidelines-pro-type-vararg,-cppcoreguidelines-special-member-functions,-fuchsia-*,-google-explicit-constructor,-google-readability-braces-around-statements,-google-readability-todo,-google-runtime-int,-google-runtime-references,-hicpp-vararg,-hicpp-braces-around-statements,-hicpp-explicit-conversions,-hicpp-named-parameter,-hicpp-no-array-decay,-hicpp-avoid-c-arrays,-hicpp-signed-bitwise,-hicpp-special-member-functions,-hicpp-uppercase-literal-suffix,-hicpp-use-auto,-hicpp-use-equals-default,-hicpp-use-override,-llvm-header-guard,-llvm-include-order,-llvmlibc-restrict-system-libc-headers,-llvmlibc-callee-namespace,-llvmlibc-implementation-in-namespace,-llvm-else-after-return,-llvm-qualified-auto,-misc-misplaced-const,-misc-non-private-member-variables-in-classes,-misc-no-recursion,-modernize-avoid-bind,-modernize-avoid-c-arrays,-modernize-pass-by-value,-modernize-use-auto,-modernize-use-default-member-init,-modernize-use-equals-default,-modernize-use-trailing-return-type,-modernize-use-transparent-functors,-performance-unnecessary-value-param,-readability-braces-around-statements,-readability-else-after-return,-readability-function-cognitive-complexity,-readability-isolate-declaration,-readability-magic-numbers,-readability-named-parameter,-readability-uppercase-literal-suffix,-readability-convert-member-functions-to-static,-readability-qualified-auto,-readability-redundant-string-init,-bugprone-narrowing-conversions,-cppcoreguidelines-narrowing-conversions,-altera-struct-pack-align,-cppcoreguidelines-prefer-member-initializer
-- CMAKE_CXX_FLAGS: 
-- Generating sharded instantiations for target: device_grouped_conv2d_fwd_xdl_ngchw_gkcyx_ngkhw_bf16_instances
-- Generating sharded instantiations for target: device_grouped_conv2d_fwd_xdl_ngchw_gkcyx_ngkhw_f16_instances
-- Generating sharded instantiations for target: device_grouped_conv2d_fwd_xdl_ngchw_gkcyx_ngkhw_bf16_comp_instances
-- Generating sharded instantiations for target: device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_inter_instances
-- Generating sharded instantiations for target: device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_intra_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_bf16_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_f16_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_bf16_mem_inter_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_f16_mem_inter_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_f32_mem_inter_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_bf16_mem_intra_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_f16_mem_intra_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_f32_mem_intra_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ndhwgc_gkzyxc_ndhwgk_bf16_comp_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ndhwgc_gkzyxc_ndhwgk_f16_comp_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_bf16_comp_instances
-- Generating sharded instantiations for target: device_grouped_conv3d_fwd_xdl_ngcdhw_gkczyx_ngkdhw_f16_comp_instances
-- Found Python3: /usr/bin/python3.13 (found version "3.13.7") found components: Interpreter Development Development.Module Development.Embed
-- Performing Test HAS_MISCHED_BOTTOMUP
-- Performing Test HAS_MISCHED_BOTTOMUP - Success
-- Performing Test HAS_MISCHED_PRERA_DIRECTION
-- Performing Test HAS_MISCHED_PRERA_DIRECTION - Failed
-- Performing Test HAS_MAX_OCCUPANCY_EXPERIMENTAL
-- Performing Test HAS_MAX_OCCUPANCY_EXPERIMENTAL - Failed
CMake Warning at tile_engine/ops/gemm/CMakeLists.txt:18 (message):
  Skipping Tile Engine GEMM compilation: No supported GPU targets (gfx90a,
  gfx942, gfx950) found in SUPPORTED_GPU_TARGETS: gfx1100
Call Stack (most recent call first):
  tile_engine/ops/gemm/CMakeLists.txt:167 (build_gemm_for_datatype)


CMake Warning at tile_engine/ops/gemm/CMakeLists.txt:18 (message):
  Skipping Tile Engine GEMM compilation: No supported GPU targets (gfx90a,
  gfx942, gfx950) found in SUPPORTED_GPU_TARGETS: gfx1100
Call Stack (most recent call first):
  tile_engine/ops/gemm/CMakeLists.txt:167 (build_gemm_for_datatype)


CMake Warning at tile_engine/ops/gemm_multi_d/CMakeLists.txt:19 (message):
  Skipping Tile Engine GEMM Multi D compilation: No supported GPU targets
  (gfx90a, gfx942, gfx950) found in SUPPORTED_GPU_TARGETS: gfx1100
Call Stack (most recent call first):
  tile_engine/ops/gemm_multi_d/CMakeLists.txt:171 (build_gemm_multi_d_for_datatype_layout)


-- The C compiler identification is GNU 15.2.1
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /usr/bin/cc - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Found Python3: /usr/bin/python3.13 (found version "3.13.7") found components: Interpreter
-- Adding: test/ck_tile/utility/
-- CK_PROFILER_OP_FILTER: .+
-- CK_PROFILER_INSTANCE_FILTER: .+
-- Configuring done (17.1s)
-- Generating done (1.2s)
-- Build files have been written to: /home/aethor/Dev/composable_kernel/build

Then, running make gives:

clang++: error: '-fuse-ld=' taking a path is deprecated; use '--ld-path=' instead [-Werror,-Wfuse-ld-path]
failed to execute:/usr/lib64/rocm/llvm/bin/clang++  --offload-arch=gfx1100 --offload-arch=gfx1100  -DCK_ENABLE_BF16 -DCK_ENABLE_BF8 -DCK_ENABLE_FP16 -DCK_ENABLE_FP32 -DCK_ENABLE_FP64 -DCK_ENABLE_FP8 -DCK_ENABLE_INT8 -DCK_TILE_USE_WMMA=1 -DCK_TIME_KERNEL=1 -DCK_USE_WMMA -DDL_KERNELS -DDPP_KERNELS -DUSE_PROF_API=1 -D__HIP_PLATFORM_AMD__=1 -D__HIP_PLATFORM_HCC__=1 -I/home/aethor/Dev/composable_kernel/library/include -I/home/aethor/Dev/composable_kernel/include -I/home/aethor/Dev/composable_kernel/build/include -O3 -DNDEBUG -std=c++20 -fPIC -Wall -Wextra -Wcomment -Wendif-labels -Wformat -Winit-self -Wreturn-type -Wsequence-point -Wswitch -Wtrigraphs -Wundef -Wuninitialized -Wunreachable-code -Wunused -Wno-reserved-identifier -Wno-option-ignored -Wsign-compare -Wno-extra-semi-stmt -Wno-unused-template -Wno-missing-field-initializers -Wno-error=deprecated-declarations -Wall -Wextra -Wcomment -Wendif-labels -Wformat -Winit-self -Wreturn-type -Wsequence-point -Wswitch -Wtrigraphs -Wundef -Wuninitialized -Wunreachable-code -Wunused -Wno-reserved-identifier -Wno-option-ignored -Wsign-compare -Wno-extra-semi-stmt -Wno-unused-template -Weverything -Wno-c++98-compat -Wno-c++98-compat-pedantic -Wno-conversion -Wno-double-promotion -Wno-exit-time-destructors -Wno-extra-semi -Wno-float-conversion -Wno-gnu-anonymous-struct -Wno-gnu-zero-variadic-macro-arguments -Wno-missing-prototypes -Wno-nested-anon-types -Wno-padded -Wno-return-std-move-in-c++11 -Wno-shorten-64-to-32 -Wno-sign-conversion -Wno-unknown-warning-option -Wno-unused-command-line-argument -Wno-weak-vtables -Wno-covered-switch-default -Wno-unsafe-buffer-usage -Wno-unused-lambda-capture -Wno-nvcc-compat -Wno-bit-int-extension -Wno-pass-failed -Wno-switch-default -Wno-unique-object-duplication -Wno-nrvo -fno-offload-uniform-block -mllvm --lsr-drop-solution=1 -mllvm -enable-post-misched=0 -mllvm -amdgpu-coerce-illegal-types=1 -mllvm -amdgpu-early-inline-all=true -mllvm -amdgpu-function-calls=false -Werror -Weverything -fcolor-diagnostics --offload-compress -x hip -MD -MT library/src/tensor_operation_instance/gpu/transpose/CMakeFiles/device_transpose_instance.dir/device_transpose_instances_3d.cpp.o -MF CMakeFiles/device_transpose_instance.dir/device_transpose_instances_3d.cpp.o.d -o "CMakeFiles/device_transpose_instance.dir/device_transpose_instances_3d.cpp.o" -c /home/aethor/Dev/composable_kernel/library/src/tensor_operation_instance/gpu/transpose/device_transpose_instances_3d.cpp
make[2]: *** [library/src/tensor_operation_instance/gpu/transpose/CMakeFiles/device_transpose_instance.dir/build.make:75: library/src/tensor_operation_instance/gpu/transpose/CMakeFiles/device_transpose_instance.dir/device_transpose_instances_3d.cpp.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:23016: library/src/tensor_operation_instance/gpu/transpose/CMakeFiles/device_transpose_instance.dir/all] Error 2
make: *** [Makefile:166: all] Error 2

[lucbruni-amd]

Hi @Aethor,

Thanks for sharing the additional information.

Can you try appending the following CMake flags to your configure command:

-D CMAKE_CXX_FLAGS="-Wno-error=fuse-ld-path"
-D CMAKE_HIP_FLAGS="-Wno-error=fuse-ld-path"

Let me know if this helps workaround the error. Thanks!

[Aethor]

Good call @lucbruni-amd ! Now I get a profusion of warnings since earlier errors are transformed into warnings, but at least I can build further. I have another issue down the build but I don't think this is related.

Maybe this should be included in the documentation, or set as default flags depending on the version of clang?

[lucbruni-amd]

Hey @Aethor, glad to hear you could make it a bit further down in the build. Fedora is quite strict with errors/warnings.

May I ask what your latest issue is? Perhaps we can continue working through them together.

Maybe this should be included in the documentation, or set as default flags depending on the version of clang?

I will forward this suggestion to the team. There are continual packaging efforts for Fedora which includes Composable Kernel, however I need to clarify with a few folks regarding the plan and timeline. If this component can be included in the Fedora ROCm packages, we could avoid these source build issues - but I'm not certain on when that will happen. Thanks for the suggestion!

[Aethor]

Happy to know that there is effort to package ck for Fedora :)

May I ask what your latest issue is? Perhaps we can continue working through them together.

Sure! But maybe I should open a new issue about that?
Basically I get a bunch of these error messages:

In file included from /home/aethor/Dev/composable_kernel/example/29_batched_gemm_bias_e_permute/batched_gemm_bias_e_permute_wmma_fp16.cpp:9:
In file included from /home/aethor/Dev/composable_kernel/include/ck/ck.hpp:10:
In file included from /usr/include/hip/hip_runtime.h:62:
In file included from /usr/include/hip/amd_detail/amd_hip_runtime.h:114:
In file included from /usr/include/hip/hip_runtime_api.h:575:
In file included from /usr/include/hip/texture_types.h:47:
In file included from /usr/include/hip/channel_descriptor.h:32:
In file included from /usr/include/hip/amd_detail/amd_channel_descriptor.h:29:
In file included from /usr/include/hip/amd_detail/amd_hip_vector_types.h:49:
/usr/lib/gcc/x86_64-redhat-linux/15/../../../../include/c++/15/array:210:2: error: reference to __host__ function '__glibcxx_assert_fail' in __host__ __device__ function
  210 |         __glibcxx_requires_subscript(__n);
      |         ^
/usr/lib/gcc/x86_64-redhat-linux/15/../../../../include/c++/15/debug/assertions.h:39:3: note: expanded from macro '__glibcxx_requires_subscript'
   39 |   __glibcxx_assert(_N < this->size())
      |   ^
/usr/lib/gcc/x86_64-redhat-linux/15/../../../../include/c++/15/x86_64-redhat-linux/bits/c++config.h:2572:12: note: expanded from macro '__glibcxx_assert'
 2572 |       std::__glibcxx_assert_fail();                                     \
      |            ^
/home/aethor/Dev/composable_kernel/include/ck/tensor_operation/gpu/device/impl/device_batched_contraction_multiple_d_wmma_cshuffle.hpp:524:17: note: called by 'operator()<ck::integral_constant<int, 0>>'
  524 |                 ds_offset[i] = static_cast<long_index_t>(g_idx) *
      |                 ^
/home/aethor/Dev/composable_kernel/include/ck/utility/functional2.hpp:60:10: note: called by 'operator()<(lambda at /home/aethor/Dev/composable_kernel/include/ck/tensor_operation/gpu/device/impl/device_batched_contraction_multiple_d_wmma_cshuffle.hpp:523:44)>'
   60 |         (f(Number<Is>{}), ...);
      |          ^
/home/aethor/Dev/composable_kernel/include/ck/tensor_operation/gpu/device/impl/device_batched_contraction_multiple_d_wmma_cshuffle.hpp:523:13: note: called by 'GetDsPtrOffset'
  523 |             static_for<0, NumDTensor, 1>{}([&](auto i) {
      |             ^
/home/aethor/Dev/composable_kernel/include/ck/tensor_operation/gpu/grid/gridwise_gemm_multiple_d_wmma_cshuffle.hpp:165:62: note: called by 'kernel_contraction_multiple_d_wmma_cshuffle<ck::GridwiseGemmMultipleD_Wmma<_Float16, _Float16, float, _Float16, ck::Tuple<_Float16>, _Float16, ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int>, ck::Tuple<int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int>>, ck::RightPad<int, int>, ck::RightPad<int, int>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 2>, ck::integral_constant<int, 2>, ck::integral_constant<int, 4>>, false>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 4>, ck::integral_constant<int, 16>>, false>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<7>, ck::Sequence<6>>, ck::Tuple<ck::Sequence<1, 2, 3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8, 9, 10, 11>, ck::Sequence<12, 13, 14>>, ck::Sequence<8, 12, 13, 9, 10, 14, 11>, long>, ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int>, ck::Tuple<int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int>>, ck::RightPad<int, int>, ck::RightPad<int, int>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 4>>, false>, ck::PassThrough<int>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<7>, ck::Sequence<6>>, ck::Tuple<ck::Sequence<1, 2, 3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8, 9>, ck::Sequence<10>>, ck::Sequence<8, 10, 9>, long>, ck::Tuple<ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int, int>, ck::Tuple<int, int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::RightPad<int, int>, ck::RightPad<int, int>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3, 4>, ck::Sequence<5>, ck::Sequence<6>>, ck::Tuple<ck::Sequence<1, 2, 3, 4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8>>, ck::Sequence<7, 8>, long>>, ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int, int>, ck::Tuple<int, int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::RightPad<int, int>, ck::RightPad<int, int>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3, 4>, ck::Sequence<5>, ck::Sequence<6>>, ck::Tuple<ck::Sequence<1, 2, 3, 4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8>>, ck::Sequence<7, 8>, long>, ck::tensor_operation::element_wise::PassThrough, ck::tensor_operation::element_wise::PassThrough, ck::tensor_operation::element_wise::Add, ck::InMemoryDataOperationEnum::Set, 64, 64, 64, 16, 16, 4, 1, 4, 128, ck::Sequence<4, 32, 1>, ck::Sequence<1, 0, 2>, ck::Sequence<1, 0, 2>, 2, 4, 4, false, false, false, ck::Sequence<4, 32, 1>, ck::Sequence<1, 0, 2>, ck::Sequence<1, 0, 2>, 2, 4, 4, false, true, false, 1, 1, ck::Sequence<1, 64, 1, 2>, 8, 1, ck::LoopScheduler::Default, ck::PipelineVersion::v1>, _Float16, _Float16, ck::Tuple<const _Float16 *>, _Float16, ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int>, ck::Tuple<int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int>>, ck::RightPad<int, int>, ck::RightPad<int, int>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 2>, ck::integral_constant<int, 2>, ck::integral_constant<int, 4>>, false>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 4>, ck::integral_constant<int, 16>>, false>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<7>, ck::Sequence<6>>, ck::Tuple<ck::Sequence<1, 2, 3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8, 9, 10, 11>, ck::Sequence<12, 13, 14>>, ck::Sequence<8, 12, 13, 9, 10, 14, 11>, long>, ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int>, ck::Tuple<int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int>>, ck::RightPad<int, int>, ck::RightPad<int, int>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 4>>, false>, ck::PassThrough<int>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<7>, ck::Sequence<6>>, ck::Tuple<ck::Sequence<1, 2, 3>, ck::Sequence<4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8, 9>, ck::Sequence<10>>, ck::Sequence<8, 10, 9>, long>, ck::Tuple<ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int, int>, ck::Tuple<int, int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::RightPad<int, int>, ck::RightPad<int, int>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 64>>, false>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 64>>, false>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3, 4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8>>, ck::Tuple<ck::Sequence<1, 2, 3, 4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8>, ck::Sequence<9, 10>, ck::Sequence<11, 12>>, ck::Sequence<9, 10, 11, 12>, long>>, ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int, int>, ck::Tuple<int, int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::RightPad<int, int>, ck::RightPad<int, int>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 64>>, false>, ck::UnMerge<ck::Tuple<int, ck::integral_constant<int, 64>>, false>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3, 4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8>>, ck::Tuple<ck::Sequence<1, 2, 3, 4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8>, ck::Sequence<9, 10>, ck::Sequence<11, 12>>, ck::Sequence<9, 10, 11, 12>, long>, ck::tensor_operation::element_wise::PassThrough, ck::tensor_operation::element_wise::PassThrough, ck::tensor_operation::element_wise::Add, ck::tensor_operation::device::DeviceBatchedContractionMultipleD_Wmma_CShuffle<2, 2, 2, 1, _Float16, _Float16, float, _Float16, ck::Tuple<_Float16>, _Float16, ck::tensor_operation::element_wise::PassThrough, ck::tensor_operation::element_wise::PassThrough, ck::tensor_operation::element_wise::Add, ck::tensor_operation::device::GemmSpecialization::MNKPadding, ck::tensor_operation::device::TensorSpecialization::Default, ck::tensor_operation::device::TensorSpecialization::Default, ck::tensor_operation::device::TensorSpecialization::Default, 1, 128, 64, 64, 64, 4, 16, 16, 1, 4, ck::Sequence<4, 32, 1>, ck::Sequence<1, 0, 2>, ck::Sequence<1, 0, 2>, 2, 4, 4, false, ck::Sequence<4, 32, 1>, ck::Sequence<1, 0, 2>, ck::Sequence<1, 0, 2>, 2, 4, 4, false, 1, 1, ck::Sequence<1, 64, 1, 2>, 8>::ComputePtrOffsetOfStridedBatch, ck::BlockToCTileMap_M00_N0_M01Adapt<64, 64, ck::TensorDescriptor<ck::Tuple<ck::Embed<ck::Tuple<int, int, int, int>, ck::Tuple<int, int, int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::Merge_v2_magic_division<ck::Tuple<int, int>>, ck::RightPad<int, int>, ck::RightPad<int, int>>, ck::Tuple<ck::Sequence<0>, ck::Sequence<1, 2>, ck::Sequence<3, 4>, ck::Sequence<5>, ck::Sequence<6>>, ck::Tuple<ck::Sequence<1, 2, 3, 4>, ck::Sequence<5>, ck::Sequence<6>, ck::Sequence<7>, ck::Sequence<8>>, ck::Sequence<7, 8>, long>>, true>'
  165 |     const auto ds_batch_offset = compute_ptr_offset_of_batch.GetDsPtrOffset(g_idx);
      |                                                              ^
/usr/lib/gcc/x86_64-redhat-linux/15/../../../../include/c++/15/x86_64-redhat-linux/bits/c++config.h:2566:3: note: '__glibcxx_assert_fail' declared here
 2566 |   __glibcxx_assert_fail()
      |   ^

I think it looks like this issue

[lucbruni-amd]

Hi @Aethor,

I am also seeing these issues with the official fedora:42 Docker image. Allegedly a fix was pushed to the compiler and included in the latest rocm Fedora package, but I need to confirm whether that's the case. Even if it was, that doesn't explain why we're still running into these issues.

Sure! But maybe I should open a new issue about that?

That's a good idea. While both of these issues will eventually be resolved by the packaging efforts, I think it's good to separate them into two separate issues. Feel free to tag me on the new issue and I'll respond there.

Thanks for your help in finding these issues!