Commit message (Expand) | Author | Age | |
---|---|---|---|
* | Adapt TensorFlow to LLVM API change from r325725 | 2018-02-22 | |
* | [XLA:CPU] Implement vectorized Log in LLVM IR | 2018-02-12 | |
* | Add debug flag to disable expensive LLVM optimization passes. | 2017-08-28 | |
* | CPU backend: support NEON intrinsics. | 2017-08-18 | |
* | In fast-math mode emit a tanh that has a faster min/max. | 2017-08-10 | |
* | Add a compiler interface to inspect LLVM IR. | 2017-08-09 | |
* | Add a configuration option for code size. | 2017-08-03 | |
* | Update Dataset API documentation. | 2017-07-27 | |
* | Add tfcompile support for --xla_dump_ir_to, to dump the LLVM IR. | 2017-07-13 | |
* | Make the CPU compiler dump optimized IR along with the unoptimized IR. | 2017-06-23 | |
* | Initial open-source release of XLA: Accelerated Linear Algebra. | 2017-01-09 |