Stephan Herhut 1a44a1f9dd Drop the implementation of the mlir_gpu plugin prototype in XLA. ...

This leaves some passes and the kernel_lowering in place for now. Those are pending removal,, too.

PiperOrigin-RevId: 348621555
Change-Id: I4ff2ed11fb646bf76ceaa780956c6257d89f54ae

2020-12-22 06:05:01 -08:00

..

experimental/conv_emitter

Merge pull request #45166 from ROCmSoftwarePlatform:google_upstream_rocm_add_remove_no_rocm_tag

2020-12-21 00:32:48 -08:00

tests/passes

Drop the implementation of the mlir_gpu plugin prototype in XLA.

2020-12-22 06:05:01 -08:00

BUILD

Drop the implementation of the mlir_gpu plugin prototype in XLA.

2020-12-22 06:05:01 -08:00

kernel_lowering.cc

Use MLIR target configuration for lowering to ROCDL.

2020-12-16 07:04:41 -08:00

kernel_lowering.h

[mlir][NFC] Replace usages of Function.h and Module.h with BuiltinOps.h

2020-12-01 13:23:48 -08:00

passes.cc

Use mlir::OpState::operator->() to get to methods of mlir::Operation.

2020-12-21 02:16:31 -08:00

passes.h

Delete MoveScalarComputationsIntoGpuLaunchPass, as this is now available upstream in MLIR core.

2020-10-12 03:42:43 -07:00

passes.td

Delete MoveScalarComputationsIntoGpuLaunchPass, as this is now available upstream in MLIR core.

2020-10-12 03:42:43 -07:00

xla_mlir_gpu_opt.cc

Expose mlir_gpu passes to test tool and add a first test.

2020-09-24 00:48:45 -07:00