| Commit message (Expand) | Author | Age |
* | Add kGenerateToken HLO instruction. | Mark Heffernan | 2018-06-08 |
* | Introduced kDomain HLO instruction set isolation to bound connected sets of i... | A. Unique TensorFlower | 2018-05-29 |
* | [XLA:GPU] Basic multi-output fusion for GPU. | A. Unique TensorFlower | 2018-05-24 |
* | Refactor HloInstruction::Fuse and add a method for multi-output fusion. | A. Unique TensorFlower | 2018-05-16 |
* | [XLA] Add log1p/expm1 | David Majnemer | 2018-05-09 |
* | [XLA] Always be willing to duplicate widening kConvert instructions during fu... | Justin Lebar | 2018-05-07 |
* | Simplify, test and document logic in instruction fusion that decides whether we | A. Unique TensorFlower | 2018-04-26 |
* | Initial addition of CLZ HLO | Jacques Pienaar | 2018-04-18 |
* | Automated g4 rollback of changelist 192180356 | Dimitris Vardoulakis | 2018-04-18 |
* | Add opcode for new instruction that broadcasts degenerate dimensions. | Dimitris Vardoulakis | 2018-04-09 |
* | [XLA] Minor comment fixes in instruction_fusion.cc. | Justin Lebar | 2018-03-05 |
* | [XLA] Add some plumbing, documentation, verification and shape inference for ... | Sanjoy Das | 2018-02-16 |
* | [TF:XLA] Adds HostCompute HLO - a pseudo-op to represent host-side computation. | A. Unique TensorFlower | 2018-02-16 |
* | Automated g4 rollback of changelist 180000981 | A. Unique TensorFlower | 2018-01-02 |
* | Automated g4 rollback of changelist 179983419 | A. Unique TensorFlower | 2017-12-23 |
* | Adds FFT for XLA: CPU via Eigen, GPU via cuFFT. | A. Unique TensorFlower | 2017-12-22 |
* | [XLA] Add BitcastConvert HLO op to enable bitwise operations on | A. Unique TensorFlower | 2017-11-21 |
* | [XLA] Adding kConditional opcode that represents a conditional HLO instruction. | A. Unique TensorFlower | 2017-11-15 |
* | Change for asynchronous Send and Recv by splitting Send into {Send, SendDone} | HyoukJoong Lee | 2017-11-10 |
* | [XLA] Remove dead opcode kIndex. | Justin Lebar | 2017-10-30 |
* | [XLA:CPU] [XLA:GPU] Adds compiler support for C64 primitive type, including r... | A. Unique TensorFlower | 2017-10-27 |
* | [XLA] Remove dead kUpdate opcode. | Justin Lebar | 2017-10-25 |
* | [XLA] Add ShiftLeft, ShiftRightArithmetic, and ShiftRightLogical operators. | Peter Hawkins | 2017-10-13 |
* | [TF:XLA] Rename HloOpcode::kLogicalX to kX | A. Unique TensorFlower | 2017-10-09 |
* | Add vlogging of HloModule before and after fusion. | Mark Heffernan | 2017-10-04 |
* | [XLA] Make HloModule::computations() return raw pointers. | Justin Lebar | 2017-09-29 |
* | [XLA] Add support for QuantizeAndDequantizeV2. | Chris Leary | 2017-09-25 |
* | [XLA] Move ReusesOperandElements() fusion check into the CPU/GPU subclasses. | A. Unique TensorFlower | 2017-09-13 |
* | [XLA] Make some static functions in InstructionFusion members. | A. Unique TensorFlower | 2017-08-30 |
* | Minor bugfix to HloInstruction::MergeFusionInstruction, and allow Instruction... | A. Unique TensorFlower | 2017-08-27 |
* | Implement Batchnorm Inference by expanding them into smaller ops. | A. Unique TensorFlower | 2017-08-17 |
* | Implement dot fusion on the CPU backend for some simple cases. | A. Unique TensorFlower | 2017-08-14 |
* | Use XLA_VLOG_LINES() in literal_test_util to avoid truncation of large tensors. | A. Unique TensorFlower | 2017-07-31 |
* | [XLA] Finish normalizing fusion computations into standard computations | A. Unique TensorFlower | 2017-07-26 |
* | [XLA] Add support for sin(x) transcendental. | A. Unique TensorFlower | 2017-07-23 |
* | [BatchNorm] Skeleton code to implement BatchNormGrad | A. Unique TensorFlower | 2017-07-06 |
* | [XLA] Move HLO reachability into its own file and make update-able. | Mark Heffernan | 2017-06-29 |
* | [XLA] Several fixes to HLO reachability analysis. | Mark Heffernan | 2017-06-28 |
* | [XLA] Simplify the fusion heuristic | David Majnemer | 2017-06-26 |
* | [XLA] Disallow fuse X into Y if there are paths from X to Y which don't fuse | David Majnemer | 2017-06-26 |
* | Add plumbing for a ReducePrecision operation. | A. Unique TensorFlower | 2017-06-16 |
* | [XLA] Add a Cos unary operation that computes the elementwise cosine | Petros Mol | 2017-06-14 |
* | We believe a fused version of batch_norm_op can speed the algorithm up. This ... | A. Unique TensorFlower | 2017-06-13 |
* | [XLA] If an op has a single "large" operand, we want to fuse this op into som... | A. Unique TensorFlower | 2017-06-01 |
* | Don't try to fuse ops that require duplication unless we expect them to be fu... | A. Unique TensorFlower | 2017-05-18 |
* | Fusion uses the same cost model for all backends when it comes to | David Majnemer | 2017-03-29 |
* | [XLA] Move kPad from GpuElementalIrEmitter::MakeElementGenerator to Elemental... | David Majnemer | 2017-03-29 |
* | [XLA] Replace uses of std::set with std::vector. | Mark Heffernan | 2017-03-10 |
* | Merge changes from github. | Dandelion Mané | 2017-03-10 |
* | [XLA] Add an IsFinite operation that tests elementwise whether values are fin... | Peter Hawkins | 2017-02-24 |