Shilei Tian 3bc125490a

[AMDGPU][Verifier] Check address space of alloca instruction (#135820) ...

This PR updates the `Verifier` to enforce that `alloca` instructions on
AMDGPU must be in AS5. This prevents hitting a misleading backend error
like "unable to select FrameIndex," which makes it look like a backend
bug when it's actually an IR-level issue.

2025-04-26 00:54:00 -04:00

..

add_attributes_amdgcn.ll

…

add_attributes.ll

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

always_inline_device.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

attributor_module_slice_reproducer.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

attributor_pointer_offset_crash.ll

…

attributor_recursion_crash.ll

…

barrier_removal.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

bug66687.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

custom_state_machines_pre_lto.ll

[AMDGPU][Verifier] Check address space of alloca instruction (#135820)

2025-04-26 00:54:00 -04:00

custom_state_machines_remarks.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

custom_state_machines.ll

[AMDGPU][Verifier] Check address space of alloca instruction (#135820)

2025-04-26 00:54:00 -04:00

dead_lib_func.ll

…

dead_use.ll

…

deduplication_remarks.ll

…

deduplication_soundness.ll

…

deduplication_target.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

deduplication.ll

…

get_hardware_num_threads_in_block_fold_optnone.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

get_hardware_num_threads_in_block_fold.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

global_alias.ll

…

global_constructor.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

globalization_remarks.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

gpu_kernel_detection_remarks.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

gpu_state_machine_function_ptr_replacement.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

heap-to-shared-missing-declarations.ll

…

hide_mem_transfer_latency.ll

…

icv_remarks.ll

…

icv_tracking_out_of_scope.ll

…

icv_tracking.ll

[clang][OpenMP] New OpenMP 6.0 assumption clause, 'no_openmp_constructs' (#125933)

2025-02-06 12:41:10 -08:00

indirect_call_kernel_info_crash.ll

…

internals_cgscc.ll

…

is_spmd_exec_mode_fold.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

keep_rpc_client.ll

…

nested_parallelism.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

no-insert-nvvm-annotations.ll

…

parallel_deletion_cg_update.ll

…

parallel_deletion_remarks.ll

…

parallel_deletion.ll

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

parallel_level_fold.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

parallel_region_merging.ll

[mlir][OpenMP] fix crash outlining infinite loop (#129872)

2025-03-07 11:02:52 +00:00

reduced_pointer_info_assertion.ll

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

remove_globalization.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

remove_noinline_attributes.ll

…

remove_rpc_client.ll

…

replace_globalization.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

rtf_type_checking.ll

…

single_threaded_execution.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

spmdization_assumes.ll

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

spmdization_constant_prop.ll

[AMDGPU][Verifier] Check address space of alloca instruction (#135820)

2025-04-26 00:54:00 -04:00

spmdization_guarding_two_reaching_kernels.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

spmdization_guarding.ll

[IR] Convert from nocapture to captures(none) (#123181)

2025-01-29 16:56:47 +01:00

spmdization_indirect.ll

[AMDGPU][Verifier] Check address space of alloca instruction (#135820)

2025-04-26 00:54:00 -04:00

spmdization_kernel_env_dep.ll

…

spmdization_no_guarding_two_reaching_kernels.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

spmdization_remarks.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

spmdization.ll

[AMDGPU][Verifier] Check address space of alloca instruction (#135820)

2025-04-26 00:54:00 -04:00

value-simplify-openmp-opt.ll

[OpenMP] Replace nvvm.annotation usage with kernel calling conventions (#122320)

2025-01-24 16:56:10 -08:00

values_in_offload_arrays.alloca.ll

…

values_in_offload_arrays.ll

…