STT-tensorflow/tensorflow/compiler/xla
Deven Desai 868395d7d0 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
2020-12-04 02:26:58 +00:00
..
client [XLA] Switch implementation of erf to use the same rational polynomial approximation as Eigen. 2020-10-15 11:29:25 -07:00
experimental/xla_sharding [XLA:SPMD] Add partial sharding API to SPMD and bridge support 2020-08-12 17:37:57 -07:00
g3doc Merge pull request #42747 from yongwww:patch-2 2020-09-02 01:13:09 -07:00
pjrt Refactor PJRT. 2020-10-20 12:08:59 -07:00
python Refactor PJRT. 2020-10-20 12:08:59 -07:00
python_api Python 3 fixes for ConvertLiteralToNumpyArray. 2020-06-26 16:49:42 -07:00
rpc Internal change 2020-10-02 20:33:42 -07:00
service Updating XLA code to account for the device lib files location change in ROCm 3.9 2020-12-04 02:26:58 +00:00
tests Small update to prng_test for more consistent results. 2020-10-16 20:39:03 -07:00
tools Remove dependencies on aliases in tensorflow/core/BUILD 2020-10-11 21:03:09 -07:00
.clang-format
array2d_test.cc
array2d.h
array3d_test.cc
array3d.h
array4d_test.cc
array4d.h Add a default constructor to Array4D 2020-02-07 15:23:01 -08:00
array_test.cc
array.cc Add specialization of method Array::FillRandom() for complex64 type. 2019-12-13 16:56:06 -08:00
array.h Fix C++14 build. 2020-08-08 13:12:15 -07:00
bit_cast_test.cc
bit_cast.h Move bfloat16 header to tensorflow/core/platform. 2020-08-14 13:44:57 -07:00
BUILD Remove dependencies on aliases in tensorflow/core/BUILD 2020-10-11 21:03:09 -07:00
comparison_util.cc Adding total-order comparison support in proto and HloInstruction. 2020-08-10 09:40:29 -07:00
comparison_util.h Adding total-order comparison support in proto and HloInstruction. 2020-08-10 09:40:29 -07:00
cpu_function_runtime.cc
cpu_function_runtime.h [XLA:CPU] Reconsider alignment heuristics 2020-04-24 02:51:55 -07:00
debug_options_flags.cc Replace instances of "blacklist" with "denylist" where possible. See Google Developer guidelines at https://developers.google.com/style/word-list#blacklist for more information. 2020-07-20 16:05:25 -07:00
debug_options_flags.h minor spelling tweaks 2019-12-09 18:21:12 +09:00
debug_options_parsers_test.cc [ROCm] Fix for compile error in //tensorflow/compiler/xla:debug_options_parsers_test 2020-01-19 02:38:56 +00:00
debug_options_parsers.h
device_util.h
error_spec.h
executable_run_options.cc Add run_id to TraceMes. 2020-03-13 10:42:34 -07:00
executable_run_options.h Enable XRT cache to be shared among multiple GPU devices. 2020-05-07 09:34:58 -07:00
execution_options_util.cc
execution_options_util.h minor spelling tweaks 2019-12-09 18:21:12 +09:00
index_util_test.cc
index_util.cc xla directory resolutions 2020-07-26 22:14:33 +00:00
index_util.h
iterator_util_test.cc
iterator_util.h
layout_test.cc [XLA] Remove unsupported sparse layout 2020-01-02 18:04:52 -08:00
layout_util_test.cc [XLA] Remove unsupported sparse layout 2020-01-02 18:04:52 -08:00
layout_util.cc Merge pull request #41752 from tg-at-google:wsign-compare-semi-final-xla 2020-07-29 19:10:26 -07:00
layout_util.h [XLA] Remove unsupported sparse layout 2020-01-02 18:04:52 -08:00
layout.cc This CL optimizes C++11 range-based for loops where the variable is copied in each iteration but it would suffice to obtain it by const reference. This is only applied to loop variables of types that are expensive to copy which means they are not trivially copyable or have a non-trivial copy constructor or destructor. 2020-04-06 17:55:23 -07:00
layout.h [XLA] Remove unsupported sparse layout 2020-01-02 18:04:52 -08:00
literal_comparison.cc Merge pull request #40962 from Intel-tensorflow:yang/eigen-bf16 2020-07-14 17:21:38 -07:00
literal_comparison.h
literal_test.cc [Resubmit] Dynamic literal support 2020-07-17 18:19:30 -07:00
literal_util.cc xla directory resolutions 2020-07-26 22:14:33 +00:00
literal_util.h Introduce a few new literal conversion functions for various floating point types 2020-01-27 07:47:30 -08:00
literal.cc [XLA] Implement S8,S16,U16 support for Literal::GetIntegralAsS64 2020-08-06 14:48:38 -07:00
literal.h [Resubmit] Dynamic literal support 2020-07-17 18:19:30 -07:00
map_util.h
metric_table_report.cc Merge pull request #41752 from tg-at-google:wsign-compare-semi-final-xla 2020-07-29 19:10:26 -07:00
metric_table_report.h
overflow_util.h
packed_literal_reader.cc
packed_literal_reader.h
parse_flags_from_env_test.cc Handle the cases where stdout may have CRLF suffix on windows in xla parse_flags_from_env_test 2020-03-12 14:33:13 -07:00
parse_flags_from_env.cc [XLA] Use LOG(QFATAL) rather than LOG(FATAL) in XLA_FLAGS argument parsing. 2019-11-27 10:27:55 -08:00
parse_flags_from_env.h minor spelling tweaks 2019-12-09 18:21:12 +09:00
primitive_util_test.cc
primitive_util.cc Adding total-order comparison support in proto and HloInstruction. 2020-08-10 09:40:29 -07:00
primitive_util.h Adding total-order comparison support in proto and HloInstruction. 2020-08-10 09:40:29 -07:00
protobuf_util.cc
protobuf_util.h
README.md
refcounting_hash_map_test.cc [XLA:GPU] Add experimental, lightly tested support for multi-host and multi-process NCCL AllReduce. 2020-02-21 14:00:29 -08:00
refcounting_hash_map.h [XLA:GPU] Add experimental, lightly tested support for multi-host and multi-process NCCL AllReduce. 2020-02-21 14:00:29 -08:00
reference_util_test.cc
reference_util.cc
reference_util.h
service_interface.h
shape_layout.cc [XLA] Implement dynamic input and output in DynamicPadder. 2019-12-12 10:31:20 -08:00
shape_layout.h Introduce dynamic reshape op. 2020-08-14 15:48:19 -07:00
shape_test.cc minor spelling tweaks 2019-12-09 18:21:12 +09:00
shape_tree_test.cc Split ShapeTreeIterator into two classes: ShapeTreeIterator and ShapeTreeLeafIterator 2020-09-15 14:29:58 -07:00
shape_tree.h Split ShapeTreeIterator into two classes: ShapeTreeIterator and ShapeTreeLeafIterator 2020-09-15 14:29:58 -07:00
shape_util_test.cc [XLA] No-op reshapes should not return true for inserting or deleting one-sized 2020-10-15 06:43:45 -07:00
shape_util.cc [XLA] No-op reshapes should not return true for inserting or deleting one-sized 2020-10-15 06:43:45 -07:00
shape_util.h [XLA] Compute ShapedBuffer::on_host_shape from ShapedBuffer::on_device_shape. 2020-10-07 17:54:11 -07:00
shape.cc [xla::Shape::Shape] Emplace tuple shapes into tuple_shapes_. 2020-03-03 15:06:38 -08:00
shape.h [libtpu] Reimplement XLA_Shape in C API for performance. 2020-10-19 14:17:22 -07:00
status_macros_test.cc minor spelling tweaks 2019-12-09 18:21:12 +09:00
status_macros.cc
status_macros.h
status.h
statusor.h
test_helpers.h
test.h
text_literal_reader_test.cc
text_literal_reader.cc
text_literal_reader.h
text_literal_writer_test.cc Use Env::LocalTempFilename for a temp filename. 2020-02-19 17:30:20 -08:00
text_literal_writer.cc
text_literal_writer.h
types.h
union_find.h XLA Parallel reduce. 2020-08-27 15:21:33 -07:00
util_test.cc [XLA] Add NaN tests for RoundTripFpToString 2020-10-05 09:39:45 -07:00
util.cc [XLA] log a WARNING for out of range F64 constants instead of a check failure. 2020-10-08 08:47:54 -07:00
util.h [XLA] Add a utility to transform dimensions from one shape's dimensions to another. 2020-03-31 02:34:20 -07:00
window_util_test.cc
window_util.cc Support empty window and 0D convolution. For non-GPUs it's already accidentally supported; for GPUs it's not hard to support anyway. 2020-01-13 14:51:47 -08:00
window_util.h
xla_data.proto Add optimization profile type in the HLO metadata 2020-10-08 16:00:57 -07:00
xla.bzl Migrate from xla_proto_library to tf_proto_library_cc. 2019-11-15 01:00:17 -08:00
xla.proto Add a module config option to enable hlo deduplication. 2020-08-03 12:42:04 -07:00

XLA (Accelerated Linear Algebra) is a domain-specific compiler for linear algebra that optimizes TensorFlow computations. See the documentation.