[mlir] Removed TanHOp lowering from ConvertStandardToLLVM since there is no reasonabl...
authorMarcel Koester <marcel.koester@dfki.de>
Tue, 3 Mar 2020 10:07:14 +0000 (11:07 +0100)
committerMarcel Koester <marcel.koester@dfki.de>
Wed, 25 Mar 2020 15:43:45 +0000 (16:43 +0100)
commit2b529a396d73e793615ce2cca525d68f9f738566
tree8b7fb27aa2067de0f6749ee1ed754acac7e029a2
parent69def20542373faa660a914608403481e21fc668
[mlir] Removed TanHOp lowering from ConvertStandardToLLVM since there is no reasonable TanH representation in LLVM.

Summary: The current ConvertStandardToLLVM phase lowers the standard TanHOp to function calls to external tanh symbols. However, this leads to misunderstandings since these external symbols are not defined anywhere. This commit removes the TanHOp lowering functionality from ConvertStandardToLLVM, adapts the LowerGpuOpsToNVVMOps and LowerGpuOpsToROCDLOps passes and adjusts the affected test cases.

Reviewers: mravishankar, herhut

Subscribers: jholewinski, mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, nicolasvasilache, csigg, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, Joonsoo, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D75509
mlir/lib/Conversion/GPUCommon/OpToFuncCallLowering.h
mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
mlir/lib/Conversion/StandardToLLVM/StandardToLLVM.cpp
mlir/test/Conversion/StandardToLLVM/convert-to-llvmir.mlir