Updating XLA code to account for the device lib files location change in ROCm 3.9

The location of the ROCm device lib files is changing in ROCm 3.9

Current (ROCm 3.8 and before) location is $ROCM_PATH/lib

```
root@ixt-rack-04:/opt/rocm-3.8.0# find . -name *.bc
./lib/oclc_isa_version_701.amdgcn.bc
./lib/ocml.amdgcn.bc
./lib/oclc_daz_opt_on.amdgcn.bc
./lib/oclc_isa_version_700.amdgcn.bc
./lib/oclc_isa_version_810.amdgcn.bc
./lib/oclc_unsafe_math_off.amdgcn.bc
./lib/oclc_wavefrontsize64_off.amdgcn.bc
./lib/oclc_isa_version_803.amdgcn.bc
./lib/oclc_isa_version_1011.amdgcn.bc
./lib/oclc_isa_version_1012.amdgcn.bc
./lib/opencl.amdgcn.bc
./lib/oclc_unsafe_math_on.amdgcn.bc
./lib/oclc_isa_version_1010.amdgcn.bc
./lib/oclc_finite_only_off.amdgcn.bc
./lib/oclc_correctly_rounded_sqrt_on.amdgcn.bc
./lib/oclc_daz_opt_off.amdgcn.bc
./lib/oclc_isa_version_802.amdgcn.bc
./lib/ockl.amdgcn.bc
./lib/oclc_isa_version_906.amdgcn.bc
./lib/oclc_isa_version_1030.amdgcn.bc
./lib/oclc_correctly_rounded_sqrt_off.amdgcn.bc
./lib/hip.amdgcn.bc
./lib/oclc_isa_version_908.amdgcn.bc
./lib/oclc_isa_version_900.amdgcn.bc
./lib/oclc_isa_version_702.amdgcn.bc
./lib/oclc_wavefrontsize64_on.amdgcn.bc
./lib/hc.amdgcn.bc
./lib/oclc_isa_version_902.amdgcn.bc
./lib/oclc_isa_version_801.amdgcn.bc
./lib/oclc_finite_only_on.amdgcn.bc
./lib/oclc_isa_version_904.amdgcn.bc
```

New (ROCm 3.9 and above) location is $ROCM_PATH/amdgcn/bitcode
```
root@ixt-hq-99:/opt/rocm-3.9.0-3703# find -name *.bc
./amdgcn/bitcode/oclc_isa_version_700.bc
./amdgcn/bitcode/ocml.bc
./amdgcn/bitcode/oclc_isa_version_1030.bc
./amdgcn/bitcode/oclc_isa_version_1010.bc
./amdgcn/bitcode/oclc_isa_version_904.bc
./amdgcn/bitcode/hip.bc
./amdgcn/bitcode/hc.bc
./amdgcn/bitcode/oclc_daz_opt_off.bc
./amdgcn/bitcode/oclc_wavefrontsize64_off.bc
./amdgcn/bitcode/oclc_wavefrontsize64_on.bc
./amdgcn/bitcode/oclc_isa_version_900.bc
./amdgcn/bitcode/oclc_isa_version_1012.bc
./amdgcn/bitcode/oclc_isa_version_702.bc
./amdgcn/bitcode/oclc_daz_opt_on.bc
./amdgcn/bitcode/oclc_unsafe_math_off.bc
./amdgcn/bitcode/ockl.bc
./amdgcn/bitcode/oclc_isa_version_803.bc
./amdgcn/bitcode/oclc_isa_version_908.bc
./amdgcn/bitcode/oclc_isa_version_802.bc
./amdgcn/bitcode/oclc_correctly_rounded_sqrt_off.bc
./amdgcn/bitcode/oclc_finite_only_on.bc
./amdgcn/bitcode/oclc_isa_version_701.bc
./amdgcn/bitcode/oclc_unsafe_math_on.bc
./amdgcn/bitcode/oclc_isa_version_902.bc
./amdgcn/bitcode/oclc_finite_only_off.bc
./amdgcn/bitcode/opencl.bc
./amdgcn/bitcode/oclc_isa_version_906.bc
./amdgcn/bitcode/oclc_isa_version_810.bc
./amdgcn/bitcode/oclc_isa_version_801.bc
./amdgcn/bitcode/oclc_correctly_rounded_sqrt_on.bc
./amdgcn/bitcode/oclc_isa_version_1011.bc
```

Also not the change in the filename(s)

This commit updates the XLA code, that has the device lib path + filename(s) hardcoded, to account for the change in location / filename

tensorflow/compiler/xla/service/gpu/llvm_gpu_backend/gpu_backend_lib.cc

@@ -67,6 +67,10 @@ limitations under the License.
 #include "tensorflow/core/profiler/lib/traceme.h"
 #include "tensorflow/core/util/env_var.h"
 
+#if TENSORFLOW_USE_ROCM
+#include "rocm/rocm_config.h"
+#endif
+
 namespace xla {
 namespace gpu {
 namespace {
@@ -560,11 +564,18 @@ namespace {
 static std::vector<string> GetROCDLPaths(int amdgpu_version,
                                          const string& rocdl_dir_path) {
   // AMDGPU version-neutral bitcodes.
+#if TF_ROCM_VERSION >= 30900
+  static std::vector<string>* rocdl_filenames = new std::vector<string>(
+      {"hc.bc", "opencl.bc", "ocml.bc", "ockl.bc", "oclc_finite_only_off.bc",
+       "oclc_daz_opt_off.bc", "oclc_correctly_rounded_sqrt_on.bc",
+       "oclc_unsafe_math_off.bc", "oclc_wavefrontsize64_on.bc"});
+#else
   static std::vector<string>* rocdl_filenames = new std::vector<string>(
       {"hc.amdgcn.bc", "opencl.amdgcn.bc", "ocml.amdgcn.bc", "ockl.amdgcn.bc",
        "oclc_finite_only_off.amdgcn.bc", "oclc_daz_opt_off.amdgcn.bc",
        "oclc_correctly_rounded_sqrt_on.amdgcn.bc",
        "oclc_unsafe_math_off.amdgcn.bc", "oclc_wavefrontsize64_on.amdgcn.bc"});
+#endif
 
   // Construct full path to ROCDL bitcode libraries.
   std::vector<string> result;
@@ -575,7 +586,11 @@ static std::vector<string> GetROCDLPaths(int amdgpu_version,
   // Add AMDGPU version-specific bitcodes.
   result.push_back(tensorflow::io::JoinPath(
       rocdl_dir_path,
+#if TF_ROCM_VERSION >= 30900
+      absl::StrCat("oclc_isa_version_", amdgpu_version, ".bc")));
+#else
       absl::StrCat("oclc_isa_version_", amdgpu_version, ".amdgcn.bc")));
+#endif
   return result;
 }
 

tensorflow/core/platform/default/rocm_rocdl_path.cc

@@ -36,7 +36,11 @@ string RocmRoot() {
 }
 
 string RocdlRoot() {
+#if TF_ROCM_VERSION >= 30900
+  return tensorflow::io::JoinPath(tensorflow::RocmRoot(), "amdgcn/bitcode");
+#else
   return tensorflow::io::JoinPath(tensorflow::RocmRoot(), "lib");
+#endif
 }
 
 }  // namespace tensorflow

tensorflow/core/platform/rocm_rocdl_path_test.cc

@@ -15,6 +15,7 @@ limitations under the License.
 
 #include "tensorflow/core/platform/rocm_rocdl_path.h"
 
+#include "rocm/rocm_config.h"
 #include "tensorflow/core/lib/core/status_test_util.h"
 #include "tensorflow/core/platform/env.h"
 #include "tensorflow/core/platform/path.h"
@@ -27,7 +28,11 @@ TEST(RocmRocdlPathTest, ROCDLPath) {
   VLOG(2) << "ROCm-Device-Libs root = " << RocdlRoot();
   std::vector<string> rocdl_files;
   TF_EXPECT_OK(Env::Default()->GetMatchingPaths(
+#if TF_ROCM_VERSION >= 30900
+      io::JoinPath(RocdlRoot(), "*.bc"), &rocdl_files));
+#else
       io::JoinPath(RocdlRoot(), "*.amdgcn.bc"), &rocdl_files));
+#endif
   EXPECT_LT(0, rocdl_files.size());
 }
 #endif