[all-commits] [llvm/llvm-project] c63dbd: [RISCV] Custom lower (i32 (fptoui/fptosi X)).
Craig Topper via All-commits
all-commits at lists.llvm.org
Sat Jul 24 10:55:33 PDT 2021
Branch: refs/heads/main
Home: https://github.com/llvm/llvm-project
Commit: c63dbd850182797bc4b76124d08e1c320ab2365d
https://github.com/llvm/llvm-project/commit/c63dbd850182797bc4b76124d08e1c320ab2365d
Author: Craig Topper <craig.topper at sifive.com>
Date: 2021-07-24 (Sat, 24 Jul 2021)
Changed paths:
M llvm/lib/Target/RISCV/RISCVISelLowering.cpp
M llvm/lib/Target/RISCV/RISCVISelLowering.h
M llvm/lib/Target/RISCV/RISCVInstrInfoD.td
M llvm/lib/Target/RISCV/RISCVInstrInfoF.td
M llvm/lib/Target/RISCV/RISCVInstrInfoZfh.td
M llvm/test/CodeGen/RISCV/double-convert.ll
M llvm/test/CodeGen/RISCV/float-convert.ll
M llvm/test/CodeGen/RISCV/half-convert.ll
M llvm/test/CodeGen/RISCV/rv64d-double-convert.ll
M llvm/test/CodeGen/RISCV/rv64f-float-convert.ll
M llvm/test/CodeGen/RISCV/rv64f-half-convert.ll
Log Message:
-----------
[RISCV] Custom lower (i32 (fptoui/fptosi X)).
I stumbled onto a case where our (sext_inreg (assertzexti32 (fptoui X)), i32)
isel pattern can cause an fcvt.wu and fcvt.lu to be emitted if
the assertzexti32 has an additional user. If we add a one use check
it would just cause a fcvt.lu followed by a sext.w when only need
a fcvt.wu to satisfy both users.
To mitigate this I've added custom isel and new ISD opcodes for
fcvt.wu. This allows us to keep know it started life as a conversion
to i32 without needing to match multiple nodes. ComputeNumSignBits
has been taught that this new nodes produces 33 sign bits. To
prevent regressions when we need to zero extend the result of an
(i32 (fptoui X)), I've added a DAG combine to convert it to an
(i64 (fptoui X)) before type legalization. In most cases this would
happen in InstCombine, but a zero_extend can be created for function
returns or arguments.
To keep everything consistent I've added new nodes for fptosi as well.
Reviewed By: luismarques
Differential Revision: https://reviews.llvm.org/D106346
More information about the All-commits
mailing list