Skip to content

[AArch64] Add SUBS(CSEL) fold from brcond. #142103

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.

Sign up for GitHub

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

Merged
merged 1 commit into from
Jun 6, 2025
Merged

[AArch64] Add SUBS(CSEL) fold from brcond. #142103

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
13 changes: 13 additions & 0 deletions llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24534,6 +24534,19 @@ static SDValue performBRCONDCombine(SDNode *N,
if (CC != AArch64CC::EQ && CC != AArch64CC::NE)
return SDValue();

// Fold away brcond(NE, cmp(csel(1, 0, CC, Cmp), 1)) -> brcond(~CC, Cmp)
if (isCMP(Cmp) && CC == AArch64CC::NE && isOneConstant(Cmp.getOperand(1))) {
SDValue CSel = Cmp.getOperand(0);
auto CSelCC = getCSETCondCode(CSel);
if (CSelCC) {
SDLoc DL(N);
return DAG.getNode(
N->getOpcode(), DL, N->getVTList(), Chain, Dest,
DAG.getConstant(getInvertedCondCode(*CSelCC), DL, MVT::i32),
CSel.getOperand(3));
}
}

unsigned CmpOpc = Cmp.getOpcode();
if (CmpOpc != AArch64ISD::ADDS && CmpOpc != AArch64ISD::SUBS)
return SDValue();
Expand Down
35 changes: 15 additions & 20 deletions llvm/test/CodeGen/AArch64/Atomics/aarch64-atomicrmw-lsfe.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -451,11 +451,10 @@ define dso_local bfloat @atomicrmw_fsub_bfloat_aligned_monotonic(ptr %ptr, bfloa
; -O0-LABEL: atomicrmw_fsub_bfloat_aligned_monotonic:
; -O0: add w8, w8, w9
; -O0: add w8, w8, w9
; -O0: ldaxrh w9, [x11]
; -O0: cmp w9, w8, uxth
; -O0: ldaxrh w8, [x11]
; -O0: cmp w8, w9, uxth
; -O0: stlxrh w10, w12, [x11]
; -O0: subs w8, w9, w8, uxth
; -O0: subs w8, w8, #1
; -O0: subs w9, w8, w9, uxth
;
; -O1-LABEL: atomicrmw_fsub_bfloat_aligned_monotonic:
; -O1: ldxrh w9, [x0]
Expand All @@ -470,11 +469,10 @@ define dso_local bfloat @atomicrmw_fsub_bfloat_aligned_acquire(ptr %ptr, bfloat
; -O0-LABEL: atomicrmw_fsub_bfloat_aligned_acquire:
; -O0: add w8, w8, w9
; -O0: add w8, w8, w9
; -O0: ldaxrh w9, [x11]
; -O0: cmp w9, w8, uxth
; -O0: ldaxrh w8, [x11]
; -O0: cmp w8, w9, uxth
; -O0: stlxrh w10, w12, [x11]
; -O0: subs w8, w9, w8, uxth
; -O0: subs w8, w8, #1
; -O0: subs w9, w8, w9, uxth
;
; -O1-LABEL: atomicrmw_fsub_bfloat_aligned_acquire:
; -O1: ldaxrh w9, [x0]
Expand All @@ -489,11 +487,10 @@ define dso_local bfloat @atomicrmw_fsub_bfloat_aligned_release(ptr %ptr, bfloat
; -O0-LABEL: atomicrmw_fsub_bfloat_aligned_release:
; -O0: add w8, w8, w9
; -O0: add w8, w8, w9
; -O0: ldaxrh w9, [x11]
; -O0: cmp w9, w8, uxth
; -O0: ldaxrh w8, [x11]
; -O0: cmp w8, w9, uxth
; -O0: stlxrh w10, w12, [x11]
; -O0: subs w8, w9, w8, uxth
; -O0: subs w8, w8, #1
; -O0: subs w9, w8, w9, uxth
;
; -O1-LABEL: atomicrmw_fsub_bfloat_aligned_release:
; -O1: ldxrh w9, [x0]
Expand All @@ -508,11 +505,10 @@ define dso_local bfloat @atomicrmw_fsub_bfloat_aligned_acq_rel(ptr %ptr, bfloat
; -O0-LABEL: atomicrmw_fsub_bfloat_aligned_acq_rel:
; -O0: add w8, w8, w9
; -O0: add w8, w8, w9
; -O0: ldaxrh w9, [x11]
; -O0: cmp w9, w8, uxth
; -O0: ldaxrh w8, [x11]
; -O0: cmp w8, w9, uxth
; -O0: stlxrh w10, w12, [x11]
; -O0: subs w8, w9, w8, uxth
; -O0: subs w8, w8, #1
; -O0: subs w9, w8, w9, uxth
;
; -O1-LABEL: atomicrmw_fsub_bfloat_aligned_acq_rel:
; -O1: ldaxrh w9, [x0]
Expand All @@ -527,11 +523,10 @@ define dso_local bfloat @atomicrmw_fsub_bfloat_aligned_seq_cst(ptr %ptr, bfloat
; -O0-LABEL: atomicrmw_fsub_bfloat_aligned_seq_cst:
; -O0: add w8, w8, w9
; -O0: add w8, w8, w9
; -O0: ldaxrh w9, [x11]
; -O0: cmp w9, w8, uxth
; -O0: ldaxrh w8, [x11]
; -O0: cmp w8, w9, uxth
; -O0: stlxrh w10, w12, [x11]
; -O0: subs w8, w9, w8, uxth
; -O0: subs w8, w8, #1
; -O0: subs w9, w8, w9, uxth
;
; -O1-LABEL: atomicrmw_fsub_bfloat_aligned_seq_cst:
; -O1: ldaxrh w9, [x0]
Expand Down
Loading
Loading