forked from tensorflow/tensorflow
-
Notifications
You must be signed in to change notification settings - Fork 94
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Develop upstream sync 230731 #2170
Merged
Merged
Changes from 1 commit
Commits
Show all changes
453 commits
Select commit
Hold shift + click to select a range
63abf15
Added attributed to IFRT CallOp that specifies if an atom program is …
tensorflower-gardener a034b3d
Add python and numpy headers to the local_config_python folder in the…
tensorflower-gardener daa9a34
Integrate LLVM at llvm/llvm-project@365d6eb1f7d8
tensorflower-gardener f178576
Update sample_stable_delegate for promotion of experimental/accelerat…
fergushenderson 0ef963c
Increase the memory limit for the dtensor GPU test.
qlzh727 0321ee1
Remove trigraph
mihaimaruseac cc5aa34
Remove trigraph
mihaimaruseac 7ba36c1
Remove unnecessary 'const' from pass-by-value function parameters.
fergushenderson d9ec8c5
Internal visibility change only.
ddunl cf4afb6
special allocations' aggregated metrics need to consider memory color.
trisolaran 7f8be6e
deprecate instruction name, it is changed over 1 years ago.
trisolaran 0cc2c30
Add macros for working with TF_Status in C++ code
LionOfJewdah 3874ea2
Correct the device assignment for tf._XlaCompile
tensorflower-gardener 69541bb
#tf-data-service Graduate `"data_transfer"` experiment.
mpcallanan 2d04d5c
Merge pull request #60026 from milpuz01:node_rewrite_to_mkl_heuristics
tensorflower-gardener c2a9dd3
Potential improvement opportunity to eliminate extra transpose op.
tensorflower-gardener f7e9b91
Merge pull request #61176 from tensorflow:pjpratik-patch-6
tensorflower-gardener f9e2045
Remove one use of inlining in XlaCallModule shape refinement.
junwhanahn fe33928
Merge consecutive Pad operators
tensorflower-gardener 1c26b1c
update internal files for release
tensorflower-gardener 0d22c4b
[XLA:Python] Use int64_t instead of ssize_t (part 2) to fix another M…
skye 6021733
Added a workaround for broadcast.
54eeb2a
[PJRT] Add PjRtDevice::PoisonExecution.
cezheng 333bb69
[IFRT] Update ShardingParam to also support scalars.
tensorflower-gardener 60574d4
Merge pull request #61384 from elfringham:fewer_jobs
tensorflower-gardener b03ea08
[xla:gpu] Fix the BFS algorithm in dataflow analysis
anlunx d5422e3
Add uniform quantized `stablehlo.convolution` -> `tfl.conv_2d` conver…
dansuh17 7d50115
[XLA] Update PatternMatchUnmergeSharding to avoid illegal function ca…
tensorflower-gardener 9b2016c
[XLA:GPU] Enable using region analysis in the CopyInsertion pass.
bchetioui e5e0228
Add a tflite model containing tensors that store string data.
f07a290
Update GraphDef version to 1570.
tensorflower-gardener dab618e
compat: Update forward compatibility horizon to 2023-07-27
tensorflower-gardener d0d3444
Merge pull request #56813 from Intel-tensorflow:kanvi/support_3d
tensorflower-gardener 2401754
Parse compiler flags passed as string through env_option_overrides.
tensorflower-gardener 703993d
Make oneDNN ACL default on Neoverse V1 cores
cfRod 5a9e625
[XLA:GPU][NFC] Refactor the propagation of dimension orders in the Tr…
e397708
[XLA:GPU] Fix for "double free in RegisterXlaFfiStreamProvider" when …
tdanyluk 3382f0b
Fix a bug where the XNNPACK delegate plugin wasn't honouring the 'flags'
fergushenderson 3273c45
[Linaro:ARM_CI] Stop using python venv for building and testing
elfringham 4d47879
BatchMatMul implicitly transpose the RHS by changing the order from R…
alankelly 8023633
[FTLite] Fix pthreadpool CMake integration
daniel-lang cdebcea
In Subgraph::AddTensors, return an error if `tensors_to_add` paramete…
fergushenderson 66f5417
[XLA:GPU] Simplify operand index computations (NFC).
olegshyshkov 3e3e3ca
Integrate LLVM at llvm/llvm-project@4706251a3186
tensorflower-gardener 4ca12e7
This is an internal change.
tensorflower-gardener f5946e3
Remove a redundant `alwayslink` declaration for the targets
tensorflower-gardener 8d71992
[XLA:GPU][NFC] Cleanup Triton GEMM rewriter.
5ea4397
Update TFRT dependency to use revision
tensorflower-gardener 7241dde
Merge pull request #60436 from Intel-tensorflow:amin/mkl-eager-fix
tensorflower-gardener f4330b0
Merge pull request #61391 from Intel-tensorflow:hmap_fix
tensorflower-gardener 48fba8d
Clean up dead launched flag to always enable op by op fallback.
changm 47b65c8
Add type annotations to parameter_server_strategy_v2.
tensorflower-gardener 4765c72
[xla] Do not depend on any iree targets by default
ezhulenev 55368a5
Implement XLA kernel for general Conv Op.
nicolagp 7e48fb0
[Linaro:ARM_CI] Retry flaky tests on AARCH64 as temp measure
elfringham 36747a8
Bump gemm rewrite test tolerance
wenscarl 9da78ff
Temporarily reenable flaky_test_attempts for upcoming TensorFlow 2.14…
MichaelHudgins b0f659b
Fix typo in error message.
haoyuz 007324c
Disable the failing test for ASAN for now.
qlzh727 d044bd6
Add EagerWeakTensor and GraphWeakTensor.
tensorflower-gardener 7d5d69a
Merge pull request #61385 from Intel-tensorflow:karen/curl-8.1.2
tensorflower-gardener d640178
Update image_searcher.md
pjpratik 79f5dee
Update nl_classifier.md
pjpratik 7b9c7e6
Update text_searcher.md
pjpratik a53ed37
Update bert_nl_classifier.md
pjpratik 4088118
Update bert_question_answerer.md
pjpratik c2aba63
Move legalize-tf-types pass to MOT directory
tensorflower-gardener d198343
Update audio_classifier.md
pjpratik 04075f4
Go: Update generated wrapper functions for TensorFlow ops.
tensorflower-gardener 362f2fb
Allow dynamic shaped starting indices when legalizing mhlo.gather op.
yishuangP 97a878f
Move the ComplexType handling into the type lowering pass for collect…
tensorflower-gardener 9c91df1
Update legacy reference to tensor.Tensor.
fionalang 5368a3a
Delete all API exports from the stale tensorflow/python/keras/ direct…
fionalang 78be6ff
Update C++ UnsortedSegmentMin/Max grad tests to avoid potential flaki…
tensorflower-gardener 2dd1d39
Tighten up DVariable constructor.
rainwoodman 5c1559b
Include indirect dependencies check under `DependOn()` in AllReduce c…
tensorflower-gardener a9e383e
Add mlir tests for the current FFT SPMD expander(CL 550657034).
tensorflower-gardener 94f2933
Fix paths and convert continuous envs from cache to RBE
angerson c5eb50d
Add C++ gradient for BatchMatMulV3
tensorflower-gardener 98446be
[XLA:Python] Fix build failure on Mac OS.
hawkinsp a693975
Unique stacks and frames when attaching debug information to graphs.
tensorflower-gardener d65b9f2
Guard ARM64 CPU info changes when on Apple or OpenBSD.
penpornk c73eaf4
Removing jit backend from tfrt.
tensorflower-gardener e57d550
Provide interface to enable XLA outside compilation for Cloud TPU VMs.
LionOfJewdah ce15c1b
Use VLOG instead of LOG_EVERY_N_SEC for logging the optimal number of…
armandouv e281561
[PJRT C] Add PJRT_Plugin_Initialize for plugin to provide initializat…
00cbb02
[TF:PJRT] Set device ordinal in compile option to be physical device …
dd21cba
Update TFRT dependency to use revision
tensorflower-gardener 2c8505a
Fix comments in .bazelrc
DrMarcII 141bb08
[xla:gpu] Improve the mechanism for terminating the concurrent region
anlunx 17f11ef
Move `TopologicalIterator` to a separate file for reusing.
tensorflower-gardener e9ab841
Move platform imports in python/__init__.py to modules_with_exports.py.
fionalang b694605
Add type annotations to the return values of convert_to_tensor_v1* an…
tensorflower-gardener f4d81e8
Register Iterations automatically when reporting a time
armandouv 766d993
Implementing BEF Generation to saved_model_aot_compile and minima…
IMMZEK 7983c59
Don't add input shardings for conditional statements when generating …
tensorflower-gardener 7e72326
#tf-data Turn down `"file_locality"` (v1) experiment.
mpcallanan bd2ad32
Add TF::CollectiveReduceV2Op to AllowedTf2XlaFallback so it is not ou…
kenfranko dd5286e
Merge pull request #61417 from tensorflow:DrMarcII-patch-1
tensorflower-gardener a08c4bc
When sharding propagation cannot infer shardings for the operand of a…
tensorflower-gardener 4d7ea08
Add type annotations to the return values of convert_to_tensor_v1* an…
5050599
Add support for mesh shapes with >3 dimensions, but will <3 dimension…
tensorflower-gardener f433fe3
#tf-data Ramp up `"file_locality_v2"` experiment to 50%.
mpcallanan 7e0eb78
[PJRT:C] Make PjRtCApiBuffer::has_dynamic_dimensions return false if …
skye 9daa492
Remove all dimensions from the mesh shape with a length of 1. This si…
tensorflower-gardener 222a63c
[Refactoring] Replace `absl::optional`, `absl::variant`, and related …
LionOfJewdah 1256c03
[XLA] Allow mixed precision types for all-gather in HLO verifier
jurahul 6ecdacc
Allow DEVICE_GPU to use PJRT for XlaCompile+Run ops.
swachhandl 79eccca
Integrate LLVM at llvm/llvm-project@2854852f4f0f
tensorflower-gardener 4741d07
[XLA] Turn
amitsabne1 b644c21
Update TFRT dependency to use revision
tensorflower-gardener 7677b1b
Target Monterey (12.0) as the minimum compatible os for arm64 wheels
nitins17 b0d2cb4
Add MLIR op definition for VirtualInfeedEnqueue op and VirtualInfeedD…
tensorflower-gardener b8cc608
Add a command line flag to provide a path to a file containing compil…
tensorflower-gardener bd083e3
Prevent redundant and excessive HLO dumps from the auto-sharding pass.
tensorflower-gardener 0f91246
Support preserving function alias in QuantizePtqDynamicRange
sngyhan 9a6fd5c
Merge pull request #61414 from elfringham:flay_retries
tensorflower-gardener 1c6adc0
Merge pull request #61412 from daniel-lang:fix-pthreadpool
tensorflower-gardener 81109af
Let the raw-ops page changes apply for dev (tf-nightly) and rc builds.
MarkDaoust ab8c59c
Internal cleanup of an incorrect patch
tensorflower-gardener 00101ba
Internal Code Change
tensorflower-gardener 7a1c8e0
[xla:gpu] Add lmhlo_gpu.gemm (cuBLAS) support
ezhulenev de1024a
Merge pull request #61413 from elfringham:no_venv
tensorflower-gardener a1e9623
Merge pull request #61415 from wenscarl:bump_gemm_rew_test_tol
tensorflower-gardener 5bb35d2
[XLA:GPU] Bubble up triton autotuner error messages.
chsigg 158426a
[xla:gpu] Add HLO tracing support
ezhulenev 97d1bc2
[XLA] NFC: simplify shape util's `ElementsIn()`, and some IWYU.
chsigg dcbb4e2
Update GraphDef version to 1571.
tensorflower-gardener e1089d6
compat: Update forward compatibility horizon to 2023-07-28
tensorflower-gardener a868ab2
[xla:gpu] Add StreamExecutor backend
ezhulenev 6d13c55
[xla:gpu] NFC: Use single flag to control experimental compiler/runti…
ezhulenev 8ba454c
[XLA:GPU] Limit the use of split-K in Triton GEMMs.
0967a0f
PR #61110: Adds matmul heuristic for oneDNN ACL builds on AArch64
cfRod f3034f9
[XLA:GPU] Enable by default more fusions in Triton GEMMs.
f374749
XlaCallModule: Increase limit of number of error message inputs for s…
gnecula d84a2cf
Remove duplicate random dependency from ROCM build.
jreiffers d672b2c
Merge pull request #60353 from cfRod:make_onednn_acl_default
tensorflower-gardener 0f3da2e
Introduce the DynamicTopKOp experiment
gnecula f1759fd
Fix typo in test name.
fergushenderson e3902ad
IF STATIC: sharing node inputs and outputs with subgraph inputs and o…
alankelly d781442
Sync 61415
wenscarl 9bc986c
Change log to only report every 10 seconds to reduce frequency of thi…
tensorflower-gardener 7636f9c
Avoid collision of channel_ids for send/recv and collectives.
kenfranko d43e620
Some CustomCalls may wish their called_computation to be sharded, too.
tensorflower-gardener 059fab8
Allow the use of different sized program keys for _XlaSendFromHost an…
kenfranko 1631dad
Integrate LLVM at llvm/llvm-project@cb924ddca514
tensorflower-gardener b11800d
Merge pull request #61400 from nluehr:cusparse-12.2-fix
tensorflower-gardener c28b940
Update TFRT dependency to use revision
tensorflower-gardener 0866c16
#tf-data-service Make data transfer server port function const and re…
mpcallanan b4edcee
#tf-data-service Rename data transfer client compatibility parameter …
mpcallanan d53dc92
memory viewer peak memory take into account fragmentation in HLO temp…
trisolaran 7334d9e
Merge pull request #61418 from tensorflow:srnitin-patch-macos_arm64-ci
tensorflower-gardener db68386
Fix type annotations and add more to parameter_server_strategy_v2.
tensorflower-gardener e98eadb
[XLA/debuggability] Add features to `hlo_slicer` that slice to the ne…
tensorflower-gardener 8c7e187
Fix up wheel verification test
angerson 0aed5ef
[XLA:Python] Remove code that existed to support ml_dtypes < 0.2.
hawkinsp 67b7e02
Take conditionals into account when estimate instruction execution
tensorflower-gardener 7e31270
[xla:gpu] Make StreamAssignmentPass a module pass
anlunx 83f43a3
Change sharding from binary to human readable while importing graphdef.
tensorflower-gardener 152dc61
Fix math_grad_test build break on mac
tensorflower-gardener f0fdcc1
[xla:gpu] NFC: Extract xla_gpu_api library + pass shared state by ref…
ezhulenev 1a72f09
[IFRT] Define MemoryKind and add it to Sharding
hyeontaek 27c3996
[XLA] Turn
bloops ebb292f
Fix boosted trees shape function issue.
cantonios 7d8e658
[IFRT] Rewrite PjRtArray::CopyToHostBuffer to use more specific shape…
skye 631bbed
Merge pull request #60807 from philipphack:u_fp8_conv_xla
tensorflower-gardener 43c256c
[jax] Rewrite IfrtHelpers::xla_dynamic_shape to use more specific sha…
skye c86bdac
Allow public linking against TSL_Status internal
hhb f826bad
Remove constant and use full path to export symbols.
DrMarcII 02a7b5f
Use TSL_Status directly inside Tensorflow
hhb e6fb864
Add support for assign/assign_add/assign_sub methods of ResourceVaria…
JW1992 53cee0b
Integrate LLVM at llvm/llvm-project@5d976edd3f2a
tensorflower-gardener 29e38e0
Update TFRT dependency to use revision
tensorflower-gardener 34db316
Replace legacy TpuStatus_ methods with TSL_Status
LionOfJewdah 9a96ad2
Update GraphDef version to 1572.
tensorflower-gardener 07ecbda
Internal Code Change
tensorflower-gardener 23abe15
compat: Update forward compatibility horizon to 2023-07-29
tensorflower-gardener 0440b9d
Update Eigen to commit:d4ae542ed1c6f3eaad29445100052489471e38ea
tensorflower-gardener 5379fe7
Fix deps issues that are preventing targets from building.
DrMarcII 6933bb5
Replace legacy TpuStatus_ methods with TSL_Status
hhb 27c7c96
Add an AoT compilation API to HLO runner.
jinliangwei 3cf13ce
Update GraphDef version to 1573.
tensorflower-gardener 3d69caf
compat: Update forward compatibility horizon to 2023-07-30
tensorflower-gardener 4ca0410
Add TfLiteTypeSizeOf to test_utils.
LukeBoyer 9a7a2f3
log raw memory allocation inside TfAllocatorAdapter::Allocate
tensorflower-gardener 8e4cd1d
Merge pull request #61416 from pjpratik:master
tensorflower-gardener d5c0d40
Rolls back log raw memory allocation inside TfAllocatorAdapter::Allocate
tensorflower-gardener fee6d2d
Update TFRT dependency to use revision
tensorflower-gardener 53812fc
Add dot general quantization conversion pattern to `ComposeUniformQua…
dansuh17 078045d
Internal Code Change
tensorflower-gardener d7c7408
Fix unable to find lit module for xla unit tests
elfringham 4ee5f59
compat: Update forward compatibility horizon to 2023-07-31
tensorflower-gardener 599baef
[TFLite] Fix gemmlowp CMake integration
daniel-lang 305c2c0
Update GraphDef version to 1574.
tensorflower-gardener fb51545
[mhlo, stablehlo] Prevent invalid indexing in Attribute pretty printers.
GleasonK b13d24f
PR #4603: [ROCm] fixed rocm build error
i-chaochen 7f6f500
[XLA:GPU][NFC] Shorten access to frequently used MLIR types.
6d2f5ac
Merge pull request #60319 from wenscarl:rank3_biasadd
tensorflower-gardener 412df1a
Integrate LLVM at llvm/llvm-project@41895843b591
d0k f1108d5
Update TFRT dependency to use revision
tensorflower-gardener a880358
[XLA:GPU] Simplify bool expression in ReductionIsRaceFree.
olegshyshkov a783e84
[XLA:GPU] Make Triton GEMM autotuner skip fusions with already assign…
af4c2ae
Move module import from python/__init__.py to python/modules_with_exp…
fionalang 5b275d6
[xla:gpu] Use profiler lock to detect active profiling sessions
ezhulenev 85db97c
[XLA] Fix 2 bugs in collective permute motion.
39af5a5
[TF:PJRT] Relax the required byte size check when the input shape is …
b63d1b6
Add release notes for bugs fixed for *FFT* ops in DTensor.
tensorflower-gardener cef42dc
Integrate LLVM at llvm/llvm-project@2dc1a27449a9
durin42 b276e31
Merge pull request #61409 from daniel-lang:fix-glemmlowp-cmake
tensorflower-gardener bbcff12
removing erroneous documentation
tensorflower-gardener 3f13278
[sharding propagation] Support propagating across call boundaries
tensorflower-gardener 80090b7
Update TFRT dependency to use revision
tensorflower-gardener 81cb103
Merge pull request #61399 from daniel-lang:fix-flatbuffer-cmake-config
tensorflower-gardener 26ab75c
Add type annotations to the return values of convert_to_tensor_v1* an…
tensorflower-gardener 4d2373a
Include the program key in the rendezvous key for host-callback in Xl…
kenfranko 8032cd2
The synchronization among the group of participants for acquiring
bixia1 4282c5f
Add shim for signature_runner.
fergushenderson 43f97d5
[xla:gpu][iree] Add support for compiled ops with multiple kernels + …
ezhulenev 2a0b819
Change `CHECK` to explicit error.
jszaday 39c23eb
[xla:runtime] Fix a bug in encoding memrefs with dynamic offset
ezhulenev ea3b03a
Merge pull request #60898 from psunn:matmul_psunn
tensorflower-gardener 3cdfbc6
Cleanup for XLA Outside Compilation in Cloud TPU VMs
LionOfJewdah c265e85
[XLA:TPU] HLO flattening fix for SPMD graphs that have outfeed.
tayo 691b916
[XLA] Make PatternMatchMergeSharding and PatternMatchUnmergeSharding …
tensorflower-gardener 08a1842
[TF:PJRT] Allow DEVICE_GPU to use PJRT for XlaCompileOnDemand op.
chuanhaozhuge 915d0fb
Add sample stable delegate code for nested control flow support.
JunyoungLim d1c9b06
Support executing multi partition GPU programs using PJRT. It mainly …
qqfish fbe8e76
Add predicate-based `wait` method to tsl::condition_variable.
cantonios 9a389b7
[XLA] Add `device_memory_size` option to ExecutableBuildOptions.
jurahul 872f84d
[xla:gpu] Add readme + style recommendation for experimental backend
ezhulenev bfc143e
Adds saved model default input support in TF.
tensorflower-gardener 82077d5
Remove reference to deprecated "long" type.
jam14j 04465a6
#tf-data Promote `"file_locality_v2"` experiment to job-level.
mpcallanan bc68c98
[xla] Make Send a control predecessor of Recv-done in the generated S…
bixia1 cd1cdc8
Fixed conflict in graph_execution_options wrapper.
IMMZEK 3c4627f
Remove setting up GCS in FindAndLoadTpuLibrary.
73a54a8
[XLA:GPU] Update intercept check of DUS and Copy in LiveRangeRegion A…
tensorflower-gardener 0eed700
Merge pull request #61428 from elfringham:fix_xla_lit
tensorflower-gardener b769fc4
Update Rendezvous API to not depend on `:tf_status`
hhb 67dbc78
Add a device count cache to CudaPlatform.
tensorflower-gardener 32803d8
Add input type inference, conversion and WeakTensor construction logi…
tensorflower-gardener 9d0fea2
Creates an optimization to fuse transpose and reshape into batch_ma…
chhe7 d1280be
When sharding propagation does not return an input sharding for opera…
tensorflower-gardener e1de22b
weekly sync 230731 before solving conflicts
weihanmines da2cefb
weekly sync 230731 after solving conflicts
weihanmines 24ddcca
remove intel's rules
weihanmines 110b0d8
turn on gpu graph and triton softmax fusion
weihanmines 0b22eac
disable float_support_test which targets for sm80
weihanmines 3131c3a
fix the tag issue
weihanmines 6dd08fb
turn off hlo_op_profiler_test for now
weihanmines File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
i-chaochen marked this conversation as resolved.
Show resolved
Hide resolved
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is added in tensorflow@56f261b we need a ticket to track this test
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sure.