maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/mlir/test/Conversion
History
George Mitenkov de3ad5bb09 [MLIR][SPIRVToLLVM] Enhanced conversion for execution mode 
This patch introduces a new conversion pattern for `spv.ExecutionMode`.
`spv.ExecutionMode` may contain important information about the entry
point, which we want to preserve. For example, `LocalSize` provides
information about the work-group size that can be reused. Hence, the
pattern for entry-point ops changes to the following:
- `spv.EntryPoint` is still simply removed
- Info from `spv.ExecutionMode` is used to create a global struct variable,
  which looks like:

  ```
  struct {
    int32_t executionMode;
    int32_t values[];          // optional values
  };
  ```

Reviewed By: mravishankar

Differential Revision: https://reviews.llvm.org/D89989
 		2020-11-10 18:33:54 +03:00
..
AVX512ToLLVM [mlir][AVX512] Start a primitive AVX512 dialect 2020-03-20 14:11:57 -04:00
AffineToStandard [mlir][AffineToStandard] Make LowerAffine pass Op-agnostic. 2020-07-27 12:14:17 -07:00
AsyncToLLVM [MLIR] Change FuncOp assembly syntax to print visibility inline instead of in attrib dict. 2020-11-09 11:08:08 -08:00
GPUCommon [mlir][gpu] Fix leaked stream and module when lowering gpu.launch_func to runtime calls. 2020-10-29 08:40:51 +01:00
GPUToCUDA [mlir] switch the modeling of LLVM types to use the new mechanism 2020-08-04 14:29:25 +02:00
GPUToNVVM MLIR: add SinOp Lowering to __nv_sinf and __nv_sin 2020-10-28 14:15:26 +01:00
GPUToROCDL Added std.floor operation to match std.ceil 2020-08-18 10:25:32 -07:00
GPUToROCm [mlir] switch the modeling of LLVM types to use the new mechanism 2020-08-04 14:29:25 +02:00
GPUToSPIRV [mlir][gpu] NFC: switch occurrences of gpu.launch_func to custom format. 2020-10-22 17:27:19 +02:00
GPUToVulkan [mlir][gpu] NFC: switch occurrences of gpu.launch_func to custom format. 2020-10-22 17:27:19 +02:00
LinalgToSPIRV [mlir][Linalg] Uniformize linalg.generic with named ops. 2020-09-22 04:13:22 -04:00
LinalgToVector [mlir][AsmPrinter] Refactor printing to only print aliases for attributes/types that will exist in the output. 2020-11-09 21:54:47 -08:00
OpenMPToLLVM [OpenMP][MLIR] Conversion pattern for OpenMP to LLVM 2020-08-27 19:32:15 +01:00
PDLToPDLInterp [mlir] Add a conversion pass between PDL and the PDL Interpreter Dialect 2020-10-26 18:01:06 -07:00
SCFToGPU [mlir] Use affine dim instead of symbol in SCFToGPU lowering. 2020-10-20 11:56:34 +02:00
SCFToStandard [mlir] Add a simpler lowering pattern for WhileOp representing a do-while loop 2020-11-04 09:43:13 +01:00
SPIRVToLLVM [MLIR][SPIRVToLLVM] Enhanced conversion for execution mode 2020-11-10 18:33:54 +03:00
ShapeToStandard [mlir] Add partial lowering of shape.cstr_broadcastable. 2020-11-03 09:57:23 +01:00
StandardToLLVM [MLIR] Support `global_memref` and `get_global_memref` in standard -> LLVM conversion. 2020-11-09 10:54:21 -08:00
StandardToSPIRV [spirv] Fix legalize standard to spir-v for transfer ops 2020-10-21 13:56:01 -07:00
VectorToLLVM [llvm][mlir] Promote the experimental reduction intrinsics to be first class intrinsics. 2020-10-07 10:36:44 -07:00
VectorToROCDL [mlir] switch the modeling of LLVM types to use the new mechanism 2020-08-04 14:29:25 +02:00
VectorToSCF [mlir][AsmPrinter] Refactor printing to only print aliases for attributes/types that will exist in the output. 2020-11-09 21:54:47 -08:00
VectorToSPIRV [mlir][spirv] Add VectorInsertDynamicOp and vector.insertelement lowering 2020-11-10 09:49:12 +01:00