[AMDGPU/MemOpsCluster] Let mem ops clustering logic also consider number of clustered...
authorhsmahesha <mahesha.comp@gmail.com>
Mon, 1 Jun 2020 17:20:29 +0000 (22:50 +0530)
committerhsmahesha <mahesha.comp@gmail.com>
Mon, 1 Jun 2020 17:22:34 +0000 (22:52 +0530)
commit0ed2c046362e2248eaf3d81e235115b28d4af262
treeaf072e6d557b36129a564bb389bceebab2a885f8
parentee9a251caf1d785798c3602d473c3d2d84180d50
[AMDGPU/MemOpsCluster] Let mem ops clustering logic also consider number of clustered bytes

Summary:
While clustering mem ops, AMDGPU target needs to consider number of clustered bytes
to decide on max number of mem ops that can be clustered. This patch adds support to pass
number of clustered bytes to target mem ops clustering logic.

Reviewers: foad, rampitec, arsenm, vpykhtin, javedabsar

Reviewed By: foad

Subscribers: MatzeB, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, javed.absar, kerbowa, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D80545
14 files changed:
llvm/include/llvm/CodeGen/TargetInstrInfo.h
llvm/lib/CodeGen/MachineScheduler.cpp
llvm/lib/CodeGen/TargetInstrInfo.cpp
llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
llvm/lib/Target/AArch64/AArch64InstrInfo.h
llvm/lib/Target/AMDGPU/SIInsertHardClauses.cpp
llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
llvm/lib/Target/AMDGPU/SIInstrInfo.h
llvm/lib/Target/Hexagon/HexagonInstrInfo.cpp
llvm/lib/Target/Hexagon/HexagonInstrInfo.h
llvm/lib/Target/Lanai/LanaiInstrInfo.cpp
llvm/lib/Target/Lanai/LanaiInstrInfo.h
llvm/lib/Target/X86/X86InstrInfo.cpp
llvm/lib/Target/X86/X86InstrInfo.h