aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/compiler/xla/service/hlo_computation.h
Commit message (Collapse)AuthorAge
* Support removing side effecting instructions with ↵Gravatar A. Unique TensorFlower2018-10-10
| | | | | | | | | | RemoveInstructionAndUnusedOperands If the caller explicitly asks to remove a side effceting instruction (e.g. all-reduce) then we should respect it instead of silently ignoring the request. PiperOrigin-RevId: 216505133
* [XLA] Migrate from gtl::FlatSet to absl::flat_hash_setGravatar Benjamin Kramer2018-10-01
| | | | PiperOrigin-RevId: 215324035
* [XLA] Migrate from gtl::FlatMap to absl::flat_hash_mapGravatar Benjamin Kramer2018-10-01
| | | | PiperOrigin-RevId: 215272497
* [XLA] Make HloComputation::instruction_count() constant-time.Gravatar Michael Kuperstein2018-09-25
| | | | | | | * Use a FlatMap for instruction_iterators_, and actually remove elements from it (which is cheap for a FlatMap). * Use the size of the map (which is O(1)) rather than the size of the list (which is O(n)) for instruction_count(). PiperOrigin-RevId: 214459259
* [XLA] Allow adding extra instructions in HloComputation::CloneWithReplacementsGravatar Michael Kuperstein2018-09-17
| | | | PiperOrigin-RevId: 213316504
* Conditionally allow changing a non-fusion computation root_instruction shape.Gravatar A. Unique TensorFlower2018-09-16
| | | | PiperOrigin-RevId: 213191899
* Add HloSchedule to HloModule.Gravatar Mark Heffernan2018-09-06
| | | | | | | | Add HloSchedule as a field on HloModule. This will enable scheduling to be a normal HLO pass and enable some passes such as copy insertion to more easily use tighter instruction live ranges based on the schedule. This change required adding HloSchedule to the "hlo" library because of circular dependencies. Nothing except for tests actually sets the schedule at the moment, but follow up cls will add a scheduling pass which will do so. PiperOrigin-RevId: 211815293
* Change headers to directly include absl::Span, and clean up the buildGravatar Tim Shen2018-08-30
| | | | | | dependencies as well. PiperOrigin-RevId: 211038094
* [XLA] Rename all (Mutable)ArraySlice to absl::Span.Gravatar Tim Shen2018-08-30
| | | | PiperOrigin-RevId: 210998142
* Fix perfromance of HloComputation::ComputeChannelDependenciesGravatar A. Unique TensorFlower2018-08-28
| | | | | | | | | | | | Previously it used an std::map containing std::vector's what added a large overhead to HloComputation::MakeInstructionPostOrder when a model contained a large number of channels. The new implementation replaced it with a FlatMap and an InlineVector what eliminates a large number of allocations and improves perfromance by a lot. PiperOrigin-RevId: 210531816
* [XLA] Use absl string types and functions instead of the TF versions.Gravatar Justin Lebar2018-08-23
| | | | | | | Unfortunately this has to be one big patch, because e.g. absl::StrCat doesn't accept a TF StringPiece, but as soon as we switch to absl::string_view, we have to switch away from all of the TF functions. PiperOrigin-RevId: 209957896
* Handle communicating instructions in HloComputation::ComputeReachabilityGravatar A. Unique TensorFlower2018-08-21
| | | | | | | | | | | | Send&recv instructions and cross-replica-sum instructions are imposing extra dependencies via the channel id or all reduce id. This CL teaches the reachability calculation logic in hlo computation to correctly account for these "invisible" dependencies. The main purpose is to stop multi output fusion from generating dependency cyclies via communicating instructions. PiperOrigin-RevId: 209593997
* Co-allocate call operands with callee parametersGravatar Sanjoy Das2018-07-16
| | | | | | | | | | | | | Right now we give these parameters an additional allocation and then don't use it. This is interesting for the CPU backend because we use call instructions to represent fork-join parallelism (i.e. a specially annotated kCall instruction tells the CPU IR emitter to shard the called computation across CPU threads). Moreover, I need this for a principled fix to b/111116907. PiperOrigin-RevId: 204820965
* [XLA] BFloat16 propagation: add de-aliasing copies before while loop inputs.Gravatar Yuanzhong Xu2018-07-09
| | | | | | | | The while loop input and output alias each other, so as long as an input is also used by other ops that could not use BF16, the propagation pass could not change such an input/ouput to BF16 even if all uses in the while loop could use BF16. Add copies for each while loop operand. This increases the chance to propagate BF16 through the while loop; if some of these copies do not help, they will remain same-shape copies and be removed at the end. This can sometimes increase HBM usage because both BF16 and F32 copies are alive, and can sometimes reduce HBM usage. PiperOrigin-RevId: 203848348
* [XLA] Avoid fusion nodes to have duplicate operands during replacing uses.Gravatar Yuanzhong Xu2018-06-25
| | | | PiperOrigin-RevId: 202049336
* [XLA] Switch PostOrder accessors to use std::vector instead of std::list.Gravatar Benjamin Kramer2018-06-15
| | | | | | | std::list is just hilariously inefficient and the postorder list creation has been rewritten not to not depend on splicing anymore so there's no need for the list. While there remove the old unused postorder list creation code. PiperOrigin-RevId: 200743677
* Introduced kDomain HLO instruction set isolation to bound connected sets of ↵Gravatar A. Unique TensorFlower2018-05-29
| | | | | | | | instructions with similar attributes (ie, sharding). This CL simply adds the infrastructure, but leaves the wire-on to a separate CL. PiperOrigin-RevId: 198503625
* [XLA] Improve documentation on HloModule, HloComputation, and HloInstruction.Gravatar Justin Lebar2018-05-16
| | | | PiperOrigin-RevId: 196915982
* [XLA] Make hlo deserialization stable for HloModule by sorting by ids when ↵Gravatar A. Unique TensorFlower2018-05-09
| | | | | | | | | | creating from proto. Also, delete the HloModule parameter HloInstruction::CreateFromProto, it's not used anywhere. Also, in ToProto, set sharding to proto if there is sharding. PiperOrigin-RevId: 196049173
* Add infrastructure for a backend-specific configuration for each op. This is ↵Gravatar Bjarke Hammersholt Roune2018-05-05
| | | | | | intentionally not exposed in ComputationBuilder and is not intended for use or to be set at all prior to the last backend-specific part of compilation. PiperOrigin-RevId: 195493500
* [XLA] Use IDs instead of names to represent the edges of HLO graph in hlo.proto.Gravatar A. Unique TensorFlower2018-03-20
| | | | PiperOrigin-RevId: 189831057
* [XLA] Simplify the HLO proto: don't nest the fusion computation in an fusion ↵Gravatar A. Unique TensorFlower2018-03-20
| | | | | | HloInstructionProto. PiperOrigin-RevId: 189811729
* [XLA] Store the program shape in the HloModuleProto and HloComputationProto.Gravatar A. Unique TensorFlower2018-03-06
| | | | PiperOrigin-RevId: 188100425
* Automated g4 rollback of changelist 185623948Gravatar A. Unique TensorFlower2018-02-16
| | | | PiperOrigin-RevId: 186038783
* Automated g4 rollback of changelist 185598764Gravatar A. Unique TensorFlower2018-02-13
| | | | PiperOrigin-RevId: 185623948
* Wire in support for XLA kConditional instruction.Gravatar A. Unique TensorFlower2018-02-13
| | | | PiperOrigin-RevId: 185598764
* Internal changeGravatar Justin Lebar2018-02-02
| | | | PiperOrigin-RevId: 184239740
* [XLA] Express the default options to ToString using an overload, rather than ↵Gravatar Justin Lebar2017-12-14
| | | | | | | | | | | a default param. No functional change. The motivation for this is that GDB ignores default params, but resolves overloads just fine. PiperOrigin-RevId: 179125588
* [XLA] Gather the bool parameters into one thing to control the text format.Gravatar A. Unique TensorFlower2017-12-14
| | | | PiperOrigin-RevId: 179079727
* Changed to allow removing side-effect instructions from an HLO computationGravatar HyoukJoong Lee2017-11-28
| | | | | | and moved the condition to the hlo_dce pass. PiperOrigin-RevId: 177215395
* When constructing fusion computations from a proto, do not uniquify the ↵Gravatar Mark Heffernan2017-11-16
| | | | | | names. The names are already unique and uniquifying them again will mutate them resulting in inconsistent names between the proto and the constructed HLO. PiperOrigin-RevId: 176035108
* Hlo parser: support fusion.Gravatar A. Unique TensorFlower2017-11-13
| | | | | | | | | Also, - Add a HloInstruction::CreateFusion interface that creates a fusion instruction with given fusion computation. Add a HloComputation::SetFusionInstruction interface to help do that. - Change how we print fusion kind. Before this change we print fusion kind together with the opcode, e.g., fusion:kLoop, which is not easy to parse. Now we append fusion kind as an attribute. - Print fusion computation the same way as other computations, instead of nested in an instruction. PiperOrigin-RevId: 175621768
* Rollback copy insertion change because it results in a DCHECK with an ↵Gravatar Mark Heffernan2017-11-03
| | | | | | | | | | | internal model. END_PUBLIC BEGIN_PUBLIC Automated g4 rollback of changelist 174423881 PiperOrigin-RevId: 174505237
* [TF:XLA] Improve support for const HLO visitors.Gravatar A. Unique TensorFlower2017-11-03
| | | | | | Add missing const overloads of Accept methods. PiperOrigin-RevId: 174500495
* Rewrite CopyInsertion to use module-scoped HloAliasAnalysis. The net effect ↵Gravatar Mark Heffernan2017-11-02
| | | | | | | | | | | | | | | | | | | | (number of copies inserted) is roughly similar to the existing implementation, but the new implementation is much more general. The new implementation can handle entry argument buffer reuse with minimal modification, for example. Some unnecessary copies are still added due to deficiencies in buffer assignment (b/62548313), but these can be removed when buffer assignment also uses HloAliasAnalysis. Also address a few issues uncovered with this cl: (1) For inplace dynamic slice in llvm backends, truncate do not wrap the slice. This matches the behavior of the non-inplace variant. (2) Disable SelectBetweenPredTuples test on GPU. The test introduces top-level buffer ambiguity which is not tolerated by the gpu backend. (3) When deserializing HLO form a proto, do not uniquify instruction names in fused computations. (4) In dataflow analysis, don't deallocate deleted HloValues during propagation. (5) In dataflow analysis, fix issue with live_out_of_computation property. PiperOrigin-RevId: 174423881
* [XLA] Add dead tuple elem removal to WhileLoopSimplifier.Gravatar Justin Lebar2017-11-02
| | | | | | | | | | | | | Specifically, if a while loop has tuple element that - is not used by the while condition, and - is not used by the while body, except to pass it along to the next iteration of the loop, then we can reshape the while loop's computations to eliminate this tuple element. PiperOrigin-RevId: 174413683
* [TF:XLA] Add a const HLO visitor.Gravatar A. Unique TensorFlower2017-11-02
| | | | | | Use it in the HLO cost analysis pass. PiperOrigin-RevId: 174411043
* Fixed HloComputation/HloInstruction clone to allow deep clone, and avoid the ↵Gravatar A. Unique TensorFlower2017-11-01
| | | | | | cloned instruction and computations to still have live link to their parent original modules and computations. PiperOrigin-RevId: 174271432
* [XLA] Allow full dumps of constant values via boolean parameter.Gravatar Chris Leary2017-11-01
| | | | PiperOrigin-RevId: 174257660
* Add support for PRED and S64 types to MakeFakeLiteral.Gravatar Mark Heffernan2017-10-25
| | | | | | Don't uniquify names when creating and HLO module from a proto. This preserves instruction names across serialization/deserialization. PiperOrigin-RevId: 173498734
* Make the HLO proto representation (hlo.proto) full fidelity. Hlo modules can ↵Gravatar Mark Heffernan2017-10-13
| | | | | | | | be serialized to HLO protos and deserialized without any information loss. As part of this change, a bug is fixed in NameUniquer. Previously, passing names with numeric suffixes could result in name collisions. PiperOrigin-RevId: 172161360
* [XLA] Make HloComputation::instructions() return a view of HloInstruction*s.Gravatar Justin Lebar2017-09-28
| | | | | | | | Currently it returns a view of unique_ptr<HloInstruction>s. But the fact that these are unique_ptrs is an implementation detail, and it's ugly to leak it everywhere. PiperOrigin-RevId: 170445375
* [XLA] Replace HloComputation::ReplaceUsesOfInstruction with ↵Gravatar Justin Lebar2017-09-27
| | | | | | | | | | | | | | | HloInstruction::ReplaceAllUsesWith. RAUW used to be *almost* synonymous with RUOI, except RAUW didn't update the computation's root. This was a dangerous footgun -- if you accidentally called RAUW when you wanted RUOI (which you almost always did), your code would work perfectly, except when the relevant node happened to be the root of a computation. This change simplifies our APIs so there's just one Right Way To Do It, by making RAUW update the computation. PiperOrigin-RevId: 170290230
* [XLA] Refactor parent-fusion-instruction pointer into HloComputation, not ↵Gravatar A. Unique TensorFlower2017-08-30
| | | | | | | | | | | | HloInstruction. Presently, each instruction inside a fusion computation contains a pointer to the fusion instruction that contains the computation, which is redundant since this is common across the entire computation. This leads to lots of places where this pointer must be set when adding an instruction to the fusion computation (and bugs such as b/65177535 when one is missed), as well as code to check that it's set correctly. In addition, this is simply unnecessary data bloat. Moreover, the computation itself does not contain a pointer to the fusion instruction that references it, which leads to odd circumlocutions in the HloComputation code that retrieve the fusion instruction from the computation's root instruction. Thus, this CL moves this pointer into the HloComputation class (replacing the is_fusion_computation_ bool value), and refactor the uses as necessary. PiperOrigin-RevId: 167039280
* Add option to HloComputation::DeepCopyInstruction for selectively copying onlyGravatar Mark Heffernan2017-08-25
| | | | | | | certain indices. Also, add mechanism for returning the kCopy instructions added to create the deep copy. PiperOrigin-RevId: 166521917
* Merge changes from github.Gravatar Jonathan Hseu2017-08-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | END_PUBLIC --- Commit b30ce4714 authored by James Qin<jamesqin@google.com> Committed by TensorFlower Gardener<gardener@tensorflow.org>: Revamp CudnnRNN Saveables 1. Use a lossy way to save/restore cudnn biases during checkpointing. Cudnn uses 2 biases each gate for all RNNs while tf uses one. To allow cudnn checkpoints to be compatible with both Cudnn and platform-independent impls, previously both individual bias and summed biases each gate were stored. The new way only stores the bias sum for each gate, and split it half-half when restoring from a cudnn graph. Doing this does not cause problems since RNNs do not use weight-decay to regularize. 2. Use inheritance instead of branching * Split RNNParamsSaveable to 1 base class and 4 subclasses. * Extract common routines and only overwrite rnn-type-specific pieces in subclasses. PiperOrigin-RevId: 166413989 --- Commit ebc421daf authored by Alan Yee<alyee@ucsd.edu> Committed by Jonathan Hseu<vomjom@vomjom.net>: Update documentation for contrib (#12424) * Update __init__.py Remove ## for standardization of api docs * Create README.md Add README to define this directory's purpose * Update __init.py Markdown styling does not show up well in api docs * Update README.md Add short mention of describing what to deprecate * Update README.md Capitalize title * Update README.md Revert README change * Delete README.md --- Commit fd295394d authored by A. Unique TensorFlower<gardener@tensorflow.org> Committed by TensorFlower Gardener<gardener@tensorflow.org>: Use latest version of nsync library, which now allows use of cmake on MacOS. PiperOrigin-RevId: 166411437 --- Commit 587d728e0 authored by A. Unique TensorFlower<gardener@tensorflow.org> Committed by TensorFlower Gardener<gardener@tensorflow.org>: [XLA] Refactor reduce-precision-insertion filters, add several more options. In particular, this adds the ability to add reduce-precision operations after fusion nodes based on the contents of those fusion nodes, and the ability to filter operations based on the "op_name" metadata. PiperOrigin-RevId: 166408392 --- Commit 3142f8ef5 authored by Ali Yahya<alive@google.com> Committed by TensorFlower Gardener<gardener@tensorflow.org>: Steps toward making ResourceVariables compatible with Eager. This change forces the value of the reuse flag in variable scopes to be tf.AUTO_REUSE when in Eager mode. This change also adds comprehensive Eager tests for ResourceVariable. PiperOrigin-RevId: 166408161 --- Commit b2ce45150 authored by Igor Ganichev<iga@google.com> Committed by TensorFlower Gardener<gardener@tensorflow.org>: Make Graph::IsValidNode public It can be reimplemented with existing public APIs, but instead of doing so, making this one public seems better. PiperOrigin-RevId: 166407897 --- Commit 0a2f40e92 authored by A. Unique TensorFlower<gardener@tensorflow.org> Committed by TensorFlower Gardener<gardener@tensorflow.org>: [XLA::CPU] Fix HLO profiling in parallel CPU backend. PiperOrigin-RevId: 166400211 --- Commit c4a58e3fd authored by Yao Zhang<yaozhang@google.com> Committed by TensorFlower Gardener<gardener@tensorflow.org>: Identify frame ids for all nodes in a graph. PiperOrigin-RevId: 166397615 --- Commit 989713f26 authored by A. Unique TensorFlower<gardener@tensorflow.org> Committed by TensorFlower Gardener<gardener@tensorflow.org>: BEGIN_PUBLIC Automated g4 rollback of changelist 166294015 PiperOrigin-RevId: 166521502
* Add option to HloComputation::DeepCopyInstruction for selectively copying onlyGravatar Mark Heffernan2017-08-25
| | | | | | | certain indices. Also, add mechanism for returning the kCopy instructions added to create the deep copy. PiperOrigin-RevId: 166521917
* Merging sibling fusion instruction using multi_output_fusionGravatar A. Unique TensorFlower2017-08-10
| | | | PiperOrigin-RevId: 164920220
* Make HloDataFlowAnalysis updatable after transforming the HLO graph.Gravatar Mark Heffernan2017-08-09
| | | | | | | | | | | | | | | | | | | | Updating is possible if operands/uses or computation roots change in the graph. Updating is not possible if instructions are deleted or if new instructions are added. Specific changes: * Add verification methods for asserting invariants and checking the analysis after updating. * Always add phi values at while instructions. Previously these were added only if the phi had different inputs. The advantage of using phi's unconditionally is that the set of values is fixed for a module. Updates due to changing operands/uses in the graph do not create new values. * Store values in a vector rather than a map. With unconditional phi values, the number of HloValues is fixed so the values can be held in a vector with stable references to elements. PiperOrigin-RevId: 164778750
* Consider the nested computations when checking if an instruction isGravatar HyoukJoong Lee2017-08-09
| | | | | | | removable from a computation. This is to prevent DCE from removing a while instruction that includes a send/recv instruction. PiperOrigin-RevId: 164722478
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-31 02:00:00 +0000