Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Backport rv32 SRLI fix #157

Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 5 additions & 5 deletions llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -991,12 +991,12 @@ void RISCVDAGToDAGISel::Select(SDNode *Node) {
unsigned TrailingOnes = llvm::countr_one(Mask);
if (ShAmt >= TrailingOnes)
break;
// If the mask has 32 trailing ones, use SRLIW.
// If the mask has 32 trailing ones, use SRLI on RV32 or SRLIW on RV64.
if (TrailingOnes == 32) {
SDNode *SRLIW =
CurDAG->getMachineNode(RISCV::SRLIW, DL, VT, N0->getOperand(0),
CurDAG->getTargetConstant(ShAmt, DL, VT));
ReplaceNode(Node, SRLIW);
SDNode *SRLI = CurDAG->getMachineNode(
Subtarget->is64Bit() ? RISCV::SRLIW : RISCV::SRLI, DL, VT,
N0->getOperand(0), CurDAG->getTargetConstant(ShAmt, DL, VT));
ReplaceNode(Node, SRLI);
return;
}

Expand Down
21 changes: 21 additions & 0 deletions llvm/test/CodeGen/RISCV/aext.ll
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefix=RV32I %s
; RUN: llc -mtriple=riscv64 -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefix=RV64I %s

define i24 @aext(i32 %0) {
; RV32I-LABEL: aext:
; RV32I: # %bb.0:
; RV32I-NEXT: srli a0, a0, 8
; RV32I-NEXT: ret
;
; RV64I-LABEL: aext:
; RV64I: # %bb.0:
; RV64I-NEXT: srliw a0, a0, 8
; RV64I-NEXT: ret
%2 = and i32 %0, -256
%3 = lshr exact i32 %2, 8
%4 = trunc i32 %3 to i24
ret i24 %4
}