diff --git a/libc/cmake/modules/LLVMLibCObjectRules.cmake b/libc/cmake/modules/LLVMLibCObjectRules.cmake
index 2c3cf116fc860..709acd9ad614f 100644
--- a/libc/cmake/modules/LLVMLibCObjectRules.cmake
+++ b/libc/cmake/modules/LLVMLibCObjectRules.cmake
@@ -278,7 +278,9 @@ function(_build_gpu_objects fq_target_name internal_target_name)
     target_compile_options(${internal_target_name} BEFORE PRIVATE
                            ${common_compile_options} --target=${LIBC_GPU_TARGET_TRIPLE})
     if(LIBC_GPU_TARGET_ARCHITECTURE_IS_AMDGPU)
-      target_compile_options(${internal_target_name} PRIVATE -mcpu=${LIBC_GPU_TARGET_ARCHITECTURE} -flto)
+      target_compile_options(${internal_target_name} PRIVATE
+                             "SHELL:-Xclang -mcode-object-version=none"
+                             -mcpu=${LIBC_GPU_TARGET_ARCHITECTURE} -flto)
     elseif(LIBC_GPU_TARGET_ARCHITECTURE_IS_NVPTX)
       get_nvptx_compile_options(nvptx_options ${LIBC_GPU_TARGET_ARCHITECTURE})
       target_compile_options(${internal_target_name} PRIVATE ${nvptx_options})
diff --git a/libc/cmake/modules/LLVMLibCTestRules.cmake b/libc/cmake/modules/LLVMLibCTestRules.cmake
index 4d17dcaf06fe9..e1286b4ab9631 100644
--- a/libc/cmake/modules/LLVMLibCTestRules.cmake
+++ b/libc/cmake/modules/LLVMLibCTestRules.cmake
@@ -528,7 +528,8 @@ function(add_integration_test test_name)
   if(LIBC_GPU_TARGET_ARCHITECTURE_IS_AMDGPU)
     target_compile_options(${fq_build_target_name} PRIVATE
                            -nogpulib -mcpu=${LIBC_GPU_TARGET_ARCHITECTURE}
-                           -flto --target=${LIBC_GPU_TARGET_TRIPLE})
+                           -flto --target=${LIBC_GPU_TARGET_TRIPLE}
+                           -mcode-object-version=${LIBC_GPU_CODE_OBJECT_VERSION})
   elseif(LIBC_GPU_TARGET_ARCHITECTURE_IS_NVPTX)
     get_nvptx_compile_options(nvptx_options ${LIBC_GPU_TARGET_ARCHITECTURE})
     target_compile_options(${fq_build_target_name} PRIVATE
@@ -578,7 +579,9 @@ set(LIBC_HERMETIC_TEST_COMPILE_OPTIONS ${LIBC_COMPILE_OPTIONS_DEFAULT}
 # The GPU build requires overriding the default CMake triple and architecture.
 if(LIBC_GPU_TARGET_ARCHITECTURE_IS_AMDGPU)
   list(APPEND LIBC_HERMETIC_TEST_COMPILE_OPTIONS
-       -nogpulib -mcpu=${LIBC_GPU_TARGET_ARCHITECTURE} -flto --target=${LIBC_GPU_TARGET_TRIPLE})
+       -nogpulib -mcpu=${LIBC_GPU_TARGET_ARCHITECTURE} -flto
+       --target=${LIBC_GPU_TARGET_TRIPLE}
+       -mcode-object-version=${LIBC_GPU_CODE_OBJECT_VERSION})
 elseif(LIBC_GPU_TARGET_ARCHITECTURE_IS_NVPTX)
   get_nvptx_compile_options(nvptx_options ${LIBC_GPU_TARGET_ARCHITECTURE})
   list(APPEND LIBC_HERMETIC_TEST_COMPILE_OPTIONS
diff --git a/libc/cmake/modules/prepare_libc_gpu_build.cmake b/libc/cmake/modules/prepare_libc_gpu_build.cmake
index 063c3b3ca6506..0b6067f69775c 100644
--- a/libc/cmake/modules/prepare_libc_gpu_build.cmake
+++ b/libc/cmake/modules/prepare_libc_gpu_build.cmake
@@ -115,3 +115,13 @@ if(LIBC_GPU_TARGET_ARCHITECTURE_IS_NVPTX)
     get_filename_component(LIBC_CUDA_ROOT "${CUDAToolkit_BIN_DIR}" DIRECTORY ABSOLUTE)
   endif()
 endif()
+
+if(LIBC_GPU_TARGET_ARCHITECTURE_IS_AMDGPU)
+  # The AMDGPU environment uses different code objects to encode the ABI for
+  # kernel calls and intrinsic functions. We want to specify this manually to
+  # conform to whatever the test suite was built to handle.
+  # FIXME: The test suite currently hangs when compiled targeting version five.
+  # This occurrs during traversal of the callback array in the startup code. We
+  # deliberately use version four until this can be addressed.
+  set(LIBC_GPU_CODE_OBJECT_VERSION 4)
+endif()
diff --git a/libc/startup/gpu/amdgpu/CMakeLists.txt b/libc/startup/gpu/amdgpu/CMakeLists.txt
index e7c0aeafafd9a..723ed9bdb290c 100644
--- a/libc/startup/gpu/amdgpu/CMakeLists.txt
+++ b/libc/startup/gpu/amdgpu/CMakeLists.txt
@@ -13,6 +13,7 @@ add_startup_object(
     -nogpulib # Do not include any GPU vendor libraries.
     -mcpu=${LIBC_GPU_TARGET_ARCHITECTURE}
     -emit-llvm # AMDGPU's intermediate object file format is bitcode.
+    -mcode-object-version=${LIBC_GPU_CODE_OBJECT_VERSION} # Manually set the ABI.
     --target=${LIBC_GPU_TARGET_TRIPLE}
   NO_GPU_BUNDLE # Compile this file directly without special GPU handling.
 )
@@ -26,4 +27,5 @@ target_link_libraries(
   "--target=${LIBC_GPU_TARGET_TRIPLE}"
   "-flto"
   "-Wl,-mllvm,-amdgpu-lower-global-ctor-dtor=0"
+  "-Wl,-mllvm,-amdhsa-code-object-version=${LIBC_GPU_CODE_OBJECT_VERSION}"
 )
diff --git a/libc/test/IntegrationTest/CMakeLists.txt b/libc/test/IntegrationTest/CMakeLists.txt
index 0c4c38db12f32..3d32b6cb53be9 100644
--- a/libc/test/IntegrationTest/CMakeLists.txt
+++ b/libc/test/IntegrationTest/CMakeLists.txt
@@ -3,6 +3,7 @@ if(LIBC_GPU_TARGET_ARCHITECTURE_IS_AMDGPU)
     -mcpu=${LIBC_GPU_TARGET_ARCHITECTURE}
     -emit-llvm # AMDGPU's intermediate object file format is bitcode.
     --target=${LIBC_GPU_TARGET_TRIPLE}
+    -mcode-object-version=${LIBC_GPU_CODE_OBJECT_VERSION} # Manually set the ABI.
   )
 elseif(LIBC_GPU_TARGET_ARCHITECTURE_IS_NVPTX)
   set(TEST_COMPILE_FLAGS