[llvm] [AArch64] Use a load instead of a store for inline stack probes (PR #170855)
Benjamin Lerman via llvm-commits
llvm-commits at lists.llvm.org
Mon Dec 8 03:06:02 PST 2025
https://github.com/qsr updated https://github.com/llvm/llvm-project/pull/170855
>From fe7f46674981754cf356ac2479a41e7943290457 Mon Sep 17 00:00:00 2001
From: Benjamin Lerman <qsr at google.com>
Date: Fri, 5 Dec 2025 14:57:08 +0100
Subject: [PATCH] [AArch64] Use a load instead of a store for inline stack
probes
Frequently, when big buffers are put on the stack we end up
with multiple virtual pages COW-mapped to single physical zero page.
Stack probes would unnecessarily trigger a COW on such pages. Avoid this
by using loads into the XZR.
---
.../Target/AArch64/AArch64FrameLowering.cpp | 18 ++++----
llvm/lib/Target/AArch64/AArch64InstrInfo.cpp | 6 +--
.../AArch64/AArch64PrologueEpilogue.cpp | 12 +++---
llvm/test/CodeGen/AArch64/sme-agnostic-za.ll | 4 +-
llvm/test/CodeGen/AArch64/sme-vg-to-stack.ll | 4 +-
.../AArch64/sme-za-lazy-save-buffer.ll | 8 ++--
.../AArch64/split-sve-stack-frame-layout.ll | 4 +-
.../test/CodeGen/AArch64/stack-probing-64k.ll | 38 ++++++++---------
.../CodeGen/AArch64/stack-probing-dynamic.ll | 24 +++++------
.../AArch64/stack-probing-last-in-block.mir | 4 +-
.../AArch64/stack-probing-no-scratch-reg.mir | 6 +--
.../AArch64/stack-probing-shrink-wrap.mir | 6 +--
.../test/CodeGen/AArch64/stack-probing-sve.ll | 20 ++++-----
llvm/test/CodeGen/AArch64/stack-probing.ll | 42 +++++++++----------
14 files changed, 99 insertions(+), 97 deletions(-)
diff --git a/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp b/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
index c2f5c0368a782..3e8ad4b2d71f4 100644
--- a/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
@@ -3753,9 +3753,9 @@ AArch64FrameLowering::inlineStackProbeLoopExactMultiple(
emitFrameOffset(*LoopMBB, LoopMBB->end(), DL, AArch64::SP, AArch64::SP,
StackOffset::getFixed(-ProbeSize), TII,
MachineInstr::FrameSetup);
- // STR XZR, [SP]
- BuildMI(*LoopMBB, LoopMBB->end(), DL, TII->get(AArch64::STRXui))
- .addReg(AArch64::XZR)
+ // LDR XZR, [SP]
+ BuildMI(*LoopMBB, LoopMBB->end(), DL, TII->get(AArch64::LDRXui))
+ .addDef(AArch64::XZR)
.addReg(AArch64::SP)
.addImm(0)
.setMIFlags(MachineInstr::FrameSetup);
@@ -3815,9 +3815,9 @@ void AArch64FrameLowering::inlineStackProbeFixed(
MachineInstr::FrameSetup, false, false, nullptr,
EmitAsyncCFI && !HasFP, CFAOffset);
CFAOffset += StackOffset::getFixed(ProbeSize);
- // STR XZR, [SP]
- BuildMI(*MBB, MBBI, DL, TII->get(AArch64::STRXui))
- .addReg(AArch64::XZR)
+ // LDR XZR, [SP]
+ BuildMI(*MBB, MBBI, DL, TII->get(AArch64::LDRXui))
+ .addDef(AArch64::XZR)
.addReg(AArch64::SP)
.addImm(0)
.setMIFlags(MachineInstr::FrameSetup);
@@ -3847,9 +3847,9 @@ void AArch64FrameLowering::inlineStackProbeFixed(
MachineInstr::FrameSetup, false, false, nullptr,
EmitAsyncCFI && !HasFP, CFAOffset);
if (ResidualSize > AArch64::StackProbeMaxUnprobedStack) {
- // STR XZR, [SP]
- BuildMI(*MBB, MBBI, DL, TII->get(AArch64::STRXui))
- .addReg(AArch64::XZR)
+ // LDR XZR, [SP]
+ BuildMI(*MBB, MBBI, DL, TII->get(AArch64::LDRXui))
+ .addDef(AArch64::XZR)
.addReg(AArch64::SP)
.addImm(0)
.setMIFlags(MachineInstr::FrameSetup);
diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp b/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
index 904577b8233d5..8f05e46b38fd3 100644
--- a/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
+++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
@@ -11140,9 +11140,9 @@ AArch64InstrInfo::probedStackAlloc(MachineBasicBlock::iterator MBBI,
.addMBB(ExitMBB)
.setMIFlags(Flags);
- // STR XZR, [SP]
- BuildMI(*LoopBodyMBB, LoopBodyMBB->end(), DL, TII->get(AArch64::STRXui))
- .addReg(AArch64::XZR)
+ // LDR XZR, [SP]
+ BuildMI(*LoopBodyMBB, LoopBodyMBB->end(), DL, TII->get(AArch64::LDRXui))
+ .addDef(AArch64::XZR)
.addReg(AArch64::SP)
.addImm(0)
.setMIFlags(Flags);
diff --git a/llvm/lib/Target/AArch64/AArch64PrologueEpilogue.cpp b/llvm/lib/Target/AArch64/AArch64PrologueEpilogue.cpp
index 965585f40571b..38a6e16324a35 100644
--- a/llvm/lib/Target/AArch64/AArch64PrologueEpilogue.cpp
+++ b/llvm/lib/Target/AArch64/AArch64PrologueEpilogue.cpp
@@ -555,9 +555,9 @@ void AArch64PrologueEmitter::allocateStackSpace(
// objects), we need to issue an extra probe, so these allocations start in
// a known state.
if (FollowupAllocs) {
- // STR XZR, [SP]
- BuildMI(MBB, MBBI, DL, TII->get(AArch64::STRXui))
- .addReg(AArch64::XZR)
+ // LDR XZR, [SP]
+ BuildMI(MBB, MBBI, DL, TII->get(AArch64::LDRXui))
+ .addDef(AArch64::XZR)
.addReg(AArch64::SP)
.addImm(0)
.setMIFlags(MachineInstr::FrameSetup);
@@ -590,9 +590,9 @@ void AArch64PrologueEmitter::allocateStackSpace(
}
if (FollowupAllocs || upperBound(AllocSize) + RealignmentPadding >
AArch64::StackProbeMaxUnprobedStack) {
- // STR XZR, [SP]
- BuildMI(MBB, MBBI, DL, TII->get(AArch64::STRXui))
- .addReg(AArch64::XZR)
+ // LDR XZR, [SP]
+ BuildMI(MBB, MBBI, DL, TII->get(AArch64::LDRXui))
+ .addDef(AArch64::XZR)
.addReg(AArch64::SP)
.addImm(0)
.setMIFlags(MachineInstr::FrameSetup);
diff --git a/llvm/test/CodeGen/AArch64/sme-agnostic-za.ll b/llvm/test/CodeGen/AArch64/sme-agnostic-za.ll
index 0906e10b551b7..194ccdd270271 100644
--- a/llvm/test/CodeGen/AArch64/sme-agnostic-za.ll
+++ b/llvm/test/CodeGen/AArch64/sme-agnostic-za.ll
@@ -365,7 +365,7 @@ define void @agnostic_za_buffer_alloc_with_stack_probes() nounwind "aarch64_za_s
; CHECK-NEXT: cmp sp, x19
; CHECK-NEXT: b.le .LBB7_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB7_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB7_1
; CHECK-NEXT: .LBB7_3:
; CHECK-NEXT: mov sp, x19
@@ -395,7 +395,7 @@ define void @agnostic_za_buffer_alloc_with_stack_probes() nounwind "aarch64_za_s
; CHECK-NEWLOWERING-NEXT: cmp sp, x19
; CHECK-NEWLOWERING-NEXT: b.le .LBB7_3
; CHECK-NEWLOWERING-NEXT: // %bb.2: // in Loop: Header=BB7_1 Depth=1
-; CHECK-NEWLOWERING-NEXT: str xzr, [sp]
+; CHECK-NEWLOWERING-NEXT: ldr xzr, [sp]
; CHECK-NEWLOWERING-NEXT: b .LBB7_1
; CHECK-NEWLOWERING-NEXT: .LBB7_3:
; CHECK-NEWLOWERING-NEXT: mov sp, x19
diff --git a/llvm/test/CodeGen/AArch64/sme-vg-to-stack.ll b/llvm/test/CodeGen/AArch64/sme-vg-to-stack.ll
index 06fb52ae10374..e23f754075abd 100644
--- a/llvm/test/CodeGen/AArch64/sme-vg-to-stack.ll
+++ b/llvm/test/CodeGen/AArch64/sme-vg-to-stack.ll
@@ -569,8 +569,8 @@ define void @vg_unwind_multiple_scratch_regs(ptr %out) #1 {
; CHECK-NEXT: .LBB4_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: cmp sp, x9
-; CHECK-NEXT: str xzr, [sp]
; CHECK-NEXT: b.ne .LBB4_1
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: .cfi_def_cfa_register wsp
@@ -630,8 +630,8 @@ define void @vg_unwind_multiple_scratch_regs(ptr %out) #1 {
; FP-CHECK-NEXT: .LBB4_1: // %entry
; FP-CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; FP-CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
+; FP-CHECK-NEXT: ldr xzr, [sp]
; FP-CHECK-NEXT: cmp sp, x9
-; FP-CHECK-NEXT: str xzr, [sp]
; FP-CHECK-NEXT: b.ne .LBB4_1
; FP-CHECK-NEXT: // %bb.2: // %entry
; FP-CHECK-NEXT: mov x8, sp
diff --git a/llvm/test/CodeGen/AArch64/sme-za-lazy-save-buffer.ll b/llvm/test/CodeGen/AArch64/sme-za-lazy-save-buffer.ll
index afd56d198d0d3..6e9d9108c85d1 100644
--- a/llvm/test/CodeGen/AArch64/sme-za-lazy-save-buffer.ll
+++ b/llvm/test/CodeGen/AArch64/sme-za-lazy-save-buffer.ll
@@ -68,7 +68,7 @@ define float @multi_bb_stpidr2_save_required_stackprobe(i32 %a, float %b, float
; CHECK: // %bb.0:
; CHECK-NEXT: stp x29, x30, [sp, #-16]! // 16-byte Folded Spill
; CHECK-NEXT: mov x29, sp
-; CHECK-NEXT: str xzr, [sp, #-16]!
+; CHECK-NEXT: ldr xzr, [sp, #-16]!
; CHECK-NEXT: .cfi_def_cfa w29, 16
; CHECK-NEXT: .cfi_offset w30, -8
; CHECK-NEXT: .cfi_offset w29, -16
@@ -80,7 +80,7 @@ define float @multi_bb_stpidr2_save_required_stackprobe(i32 %a, float %b, float
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.le .LBB2_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB2_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB2_1
; CHECK-NEXT: .LBB2_3:
; CHECK-NEXT: mov sp, x9
@@ -113,7 +113,7 @@ define float @multi_bb_stpidr2_save_required_stackprobe(i32 %a, float %b, float
; CHECK-NEWLOWERING: // %bb.0:
; CHECK-NEWLOWERING-NEXT: stp x29, x30, [sp, #-16]! // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT: mov x29, sp
-; CHECK-NEWLOWERING-NEXT: str xzr, [sp, #-16]!
+; CHECK-NEWLOWERING-NEXT: ldr xzr, [sp, #-16]!
; CHECK-NEWLOWERING-NEXT: .cfi_def_cfa w29, 16
; CHECK-NEWLOWERING-NEXT: .cfi_offset w30, -8
; CHECK-NEWLOWERING-NEXT: .cfi_offset w29, -16
@@ -127,7 +127,7 @@ define float @multi_bb_stpidr2_save_required_stackprobe(i32 %a, float %b, float
; CHECK-NEWLOWERING-NEXT: cmp sp, x9
; CHECK-NEWLOWERING-NEXT: b.le .LBB2_3
; CHECK-NEWLOWERING-NEXT: // %bb.2: // in Loop: Header=BB2_1 Depth=1
-; CHECK-NEWLOWERING-NEXT: str xzr, [sp]
+; CHECK-NEWLOWERING-NEXT: ldr xzr, [sp]
; CHECK-NEWLOWERING-NEXT: b .LBB2_1
; CHECK-NEWLOWERING-NEXT: .LBB2_3:
; CHECK-NEWLOWERING-NEXT: mov sp, x9
diff --git a/llvm/test/CodeGen/AArch64/split-sve-stack-frame-layout.ll b/llvm/test/CodeGen/AArch64/split-sve-stack-frame-layout.ll
index 492f73e8a0d2a..c237248085df3 100644
--- a/llvm/test/CodeGen/AArch64/split-sve-stack-frame-layout.ll
+++ b/llvm/test/CodeGen/AArch64/split-sve-stack-frame-layout.ll
@@ -770,7 +770,7 @@ define void @zpr_and_ppr_local_stack_probing(<vscale x 16 x i1> %pred, <vscale x
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: sub sp, sp, #2848
; CHECK-NEXT: addvl sp, sp, #-2
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: .cfi_escape 0x0f, 0x09, 0x8f, 0xb0, 0x16, 0x92, 0x2e, 0x00, 0x40, 0x1e, 0x22 // sp + 2864 + 16 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: add x8, sp, #2848
@@ -1102,3 +1102,5 @@ define void @sve_locals_zpr_ppr_csr_vla(i64 %n, <vscale x 16 x i1> %pred, <vscal
store volatile <vscale x 16 x i8> %vector, ptr %zpr_local
ret void
}
+;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
+; CHECK-FRAMELAYOUT: {{.*}}
diff --git a/llvm/test/CodeGen/AArch64/stack-probing-64k.ll b/llvm/test/CodeGen/AArch64/stack-probing-64k.ll
index 5f833e32fb8cb..4217a9096affb 100644
--- a/llvm/test/CodeGen/AArch64/stack-probing-64k.ll
+++ b/llvm/test/CodeGen/AArch64/stack-probing-64k.ll
@@ -13,7 +13,7 @@ define void @static_65536(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 65552
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #16, lsl #12 // =65536
@@ -37,7 +37,7 @@ define void @static_65552(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 65552
-; CHECK-NEXT: str xzr, [sp], #-16
+; CHECK-NEXT: ldr xzr, [sp], #-16
; CHECK-NEXT: .cfi_def_cfa_offset 65568
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
@@ -64,7 +64,7 @@ define void @static_66560(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 65552
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1024
; CHECK-NEXT: .cfi_def_cfa_offset 66576
; CHECK-NEXT: mov x8, sp
@@ -92,10 +92,10 @@ define void @static_66576(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 65552
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1040
; CHECK-NEXT: .cfi_def_cfa_offset 66592
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #16, lsl #12 // =65536
@@ -121,10 +121,10 @@ define void @static_132096(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 65552
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 131088
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1024
; CHECK-NEXT: .cfi_def_cfa_offset 132112
; CHECK-NEXT: mov x8, sp
@@ -152,21 +152,21 @@ define void @static_327664(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 65552
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 131088
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 196624
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT: .cfi_def_cfa_offset 262160
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #15, lsl #12 // =61440
; CHECK-NEXT: .cfi_def_cfa_offset 323600
; CHECK-NEXT: sub sp, sp, #4080
; CHECK-NEXT: .cfi_def_cfa_offset 327680
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #79, lsl #12 // =323584
@@ -195,7 +195,7 @@ define void @static_327680(ptr %out) #0 {
; CHECK-NEXT: .LBB6_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.ne .LBB6_1
; CHECK-NEXT: // %bb.2: // %entry
@@ -227,7 +227,7 @@ define void @static_328704(ptr %out) #0 {
; CHECK-NEXT: .LBB7_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.ne .LBB7_1
; CHECK-NEXT: // %bb.2: // %entry
@@ -263,14 +263,14 @@ define void @static_328720(ptr %out) #0 {
; CHECK-NEXT: .LBB8_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: sub sp, sp, #16, lsl #12 // =65536
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.ne .LBB8_1
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: .cfi_def_cfa_register wsp
; CHECK-NEXT: sub sp, sp, #1040
; CHECK-NEXT: .cfi_def_cfa_offset 328736
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #80, lsl #12 // =327680
@@ -309,7 +309,7 @@ define void @static_16_align_131072(ptr %out) #0 {
; CHECK-NEXT: b.le .LBB9_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB9_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB9_1
; CHECK-NEXT: .LBB9_3: // %entry
; CHECK-NEXT: mov sp, x9
@@ -343,7 +343,7 @@ define void @static_16_align_8192(ptr %out) #0 {
; CHECK-NEXT: sub x9, sp, #1, lsl #12 // =4096
; CHECK-NEXT: sub x9, x9, #4080
; CHECK-NEXT: and sp, x9, #0xffffffffffffe000
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: mov sp, x29
@@ -373,7 +373,7 @@ define void @static_32752_align_32k(ptr %out) #0 {
; CHECK-NEXT: sub x9, sp, #7, lsl #12 // =28672
; CHECK-NEXT: sub x9, x9, #4080
; CHECK-NEXT: and sp, x9, #0xffffffffffff8000
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: mov sp, x29
diff --git a/llvm/test/CodeGen/AArch64/stack-probing-dynamic.ll b/llvm/test/CodeGen/AArch64/stack-probing-dynamic.ll
index 6bc8bccac7472..e69174472e662 100644
--- a/llvm/test/CodeGen/AArch64/stack-probing-dynamic.ll
+++ b/llvm/test/CodeGen/AArch64/stack-probing-dynamic.ll
@@ -24,7 +24,7 @@ define void @dynamic(i64 %size, ptr %out) #0 {
; CHECK-NEXT: cmp sp, x8
; CHECK-NEXT: b.le .LBB0_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB0_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB0_1
; CHECK-NEXT: .LBB0_3:
; CHECK-NEXT: mov sp, x8
@@ -56,7 +56,7 @@ define void @dynamic_fixed(i64 %size, ptr %out1, ptr %out2) #0 {
; CHECK-NEXT: .cfi_def_cfa w29, 16
; CHECK-NEXT: .cfi_offset w30, -8
; CHECK-NEXT: .cfi_offset w29, -16
-; CHECK-NEXT: str xzr, [sp, #-64]!
+; CHECK-NEXT: ldr xzr, [sp, #-64]!
; CHECK-NEXT: add x9, x0, #15
; CHECK-NEXT: mov x8, sp
; CHECK-DAG: sub x10, x29, #64
@@ -69,7 +69,7 @@ define void @dynamic_fixed(i64 %size, ptr %out1, ptr %out2) #0 {
; CHECK-NEXT: cmp sp, x8
; CHECK-NEXT: b.le .LBB1_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB1_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB1_1
; CHECK-NEXT: .LBB1_3:
; CHECK-NEXT: mov sp, x8
@@ -109,7 +109,7 @@ define void @dynamic_align_64(i64 %size, ptr %out) #0 {
; CHECK-NEXT: and sp, x9, #0xffffffffffffffc0
; CHECK-NEXT: add x9, x0, #15
; CHECK-NEXT: mov x8, sp
-; CHECK-DAG: str xzr, [sp]
+; CHECK-DAG: ldr xzr, [sp]
; CHECK-DAG: and x9, x9, #0xfffffffffffffff0
; CHECK-NOT: INVALID_TO_BREAK_UP_CHECK_DAG
; CHECK-DAG: mov x19, sp
@@ -120,7 +120,7 @@ define void @dynamic_align_64(i64 %size, ptr %out) #0 {
; CHECK-NEXT: cmp sp, x8
; CHECK-NEXT: b.le .LBB2_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB2_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB2_1
; CHECK-NEXT: .LBB2_3:
; CHECK-NEXT: mov sp, x8
@@ -163,7 +163,7 @@ define void @dynamic_align_8192(i64 %size, ptr %out) #0 {
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.le .LBB3_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB3_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB3_1
; CHECK-NEXT: .LBB3_3:
; CHECK-NEXT: mov sp, x9
@@ -180,7 +180,7 @@ define void @dynamic_align_8192(i64 %size, ptr %out) #0 {
; CHECK-NEXT: cmp sp, x8
; CHECK-NEXT: b.le .LBB3_6
; CHECK-NEXT: // %bb.5: // in Loop: Header=BB3_4 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB3_4
; CHECK-NEXT: .LBB3_6:
; CHECK-NEXT: mov sp, x8
@@ -220,7 +220,7 @@ define void @dynamic_64k_guard(i64 %size, ptr %out) #0 "stack-probe-size"="65536
; CHECK-NEXT: cmp sp, x8
; CHECK-NEXT: b.le .LBB4_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB4_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB4_1
; CHECK-NEXT: .LBB4_3:
; CHECK-NEXT: mov sp, x8
@@ -264,13 +264,13 @@ define void @no_reserved_call_frame(i64 %n) #0 {
; CHECK-NEXT: b.le .LBB5_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB5_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB5_1
; CHECK-NEXT: .LBB5_3: // %entry
; CHECK-NEXT: mov sp, x0
; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1104
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: bl callee_stack_args
; CHECK-NEXT: add sp, sp, #1104
; CHECK-NEXT: mov sp, x29
@@ -300,8 +300,8 @@ define void @reserved_call_frame(i64 %n) #0 {
; CHECK-NEXT: .cfi_offset w30, -24
; CHECK-NEXT: .cfi_offset w29, -32
; CHECK-NEXT: sub sp, sp, #1504
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: add x0, sp, #1104
-; CHECK-NEXT: str xzr, [sp]
; CHECK-NEXT: bl callee_stack_args
; CHECK-NEXT: add sp, sp, #1504
; CHECK-NEXT: .cfi_def_cfa wsp, 32
@@ -343,7 +343,7 @@ define void @dynamic_sve(i64 %size, ptr %out) #0 "target-features"="+sve" {
; CHECK-NEXT: cmp sp, x8
; CHECK-NEXT: b.le .LBB7_3
; CHECK-NEXT: // %bb.2: // in Loop: Header=BB7_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB7_1
; CHECK-NEXT: .LBB7_3:
; CHECK-NEXT: mov sp, x8
diff --git a/llvm/test/CodeGen/AArch64/stack-probing-last-in-block.mir b/llvm/test/CodeGen/AArch64/stack-probing-last-in-block.mir
index 25184b0c0b56a..deba15ba37798 100644
--- a/llvm/test/CodeGen/AArch64/stack-probing-last-in-block.mir
+++ b/llvm/test/CodeGen/AArch64/stack-probing-last-in-block.mir
@@ -96,7 +96,7 @@ body: |
; CHECK-NEXT: liveins: $x9
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: $sp = frame-setup SUBXri $sp, 1, 12
- ; CHECK-NEXT: frame-setup STRXui $xzr, $sp, 0
+ ; CHECK-NEXT: $xzr = frame-setup LDRXui $sp, 0
; CHECK-NEXT: $xzr = frame-setup SUBSXrx64 $sp, $x9, 24, implicit-def $nzcv
; CHECK-NEXT: frame-setup Bcc 1, %bb.3, implicit $nzcv
; CHECK-NEXT: {{ $}}
@@ -106,7 +106,7 @@ body: |
; CHECK-NEXT: frame-setup CFI_INSTRUCTION def_cfa_register $wsp
; CHECK-NEXT: $sp = frame-setup SUBXri $sp, 3392, 0
; CHECK-NEXT: frame-setup CFI_INSTRUCTION def_cfa_offset 200016
- ; CHECK-NEXT: frame-setup STRXui $xzr, $sp, 0
+ ; CHECK-NEXT: $xzr = frame-setup LDRXui $sp, 0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: bb.1.loop:
; CHECK-NEXT: successors: %bb.1(0x7c000000), %bb.2(0x04000000)
diff --git a/llvm/test/CodeGen/AArch64/stack-probing-no-scratch-reg.mir b/llvm/test/CodeGen/AArch64/stack-probing-no-scratch-reg.mir
index a9c9b5ff60e45..05d97770d5aee 100644
--- a/llvm/test/CodeGen/AArch64/stack-probing-no-scratch-reg.mir
+++ b/llvm/test/CodeGen/AArch64/stack-probing-no-scratch-reg.mir
@@ -45,7 +45,7 @@ body: |
; CHECK-LABEL: name: f
; CHECK: bb.0.entry:
; CHECK-NEXT: successors: %bb.3(0x80000000)
- ; CHECK-NEXT: liveins: $w0, $lr
+ ; CHECK-NEXT: liveins: $w0, $lr, $fp
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: early-clobber $sp = frame-setup STPXpre killed $fp, killed $lr, $sp, -2 :: (store (s64) into %stack.2), (store (s64) into %stack.1)
; CHECK-NEXT: $x9 = frame-setup SUBXri $sp, 36, 12
@@ -55,7 +55,7 @@ body: |
; CHECK-NEXT: liveins: $x0, $x1, $x2, $x3, $x4, $x5, $x6, $x7, $x8, $x9, $x10, $x11, $x12, $x13, $x14, $x15, $x16, $x17, $x18, $x19, $x20, $x21, $x22, $x23, $x25, $x27, $x28
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: $sp = frame-setup SUBXri $sp, 1, 12
- ; CHECK-NEXT: frame-setup STRXui $xzr, $sp, 0
+ ; CHECK-NEXT: $xzr = frame-setup LDRXui $sp, 0
; CHECK-NEXT: $xzr = frame-setup SUBSXrx64 $sp, $x9, 24, implicit-def $nzcv
; CHECK-NEXT: frame-setup Bcc 1, %bb.3, implicit $nzcv
; CHECK-NEXT: {{ $}}
@@ -64,7 +64,7 @@ body: |
; CHECK-NEXT: liveins: $x0, $x1, $x2, $x3, $x4, $x5, $x6, $x7, $x8, $x10, $x11, $x12, $x13, $x14, $x15, $x16, $x17, $x18, $x19, $x20, $x21, $x22, $x23, $x25, $x27, $x28
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: $sp = frame-setup SUBXri $sp, 2544, 0
- ; CHECK-NEXT: frame-setup STRXui $xzr, $sp, 0
+ ; CHECK-NEXT: $xzr = frame-setup LDRXui $sp, 0
; CHECK-NEXT: $x9 = IMPLICIT_DEF
; CHECK-NEXT: dead $wzr = SUBSWri killed renamable $w0, 1, 0, implicit-def $nzcv
; CHECK-NEXT: Bcc 12, %bb.2, implicit $nzcv
diff --git a/llvm/test/CodeGen/AArch64/stack-probing-shrink-wrap.mir b/llvm/test/CodeGen/AArch64/stack-probing-shrink-wrap.mir
index 985ec35213970..e02adfa29e6d4 100644
--- a/llvm/test/CodeGen/AArch64/stack-probing-shrink-wrap.mir
+++ b/llvm/test/CodeGen/AArch64/stack-probing-shrink-wrap.mir
@@ -47,7 +47,7 @@ body: |
; CHECK-LABEL: name: f
; CHECK: bb.0.entry:
; CHECK-NEXT: successors: %bb.3(0x80000000)
- ; CHECK-NEXT: liveins: $w0, $lr
+ ; CHECK-NEXT: liveins: $w0, $lr, $fp
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: early-clobber $sp = frame-setup STPXpre killed $fp, killed $lr, $sp, -2 :: (store (s64) into %stack.2), (store (s64) into %stack.1)
; CHECK-NEXT: $x9 = frame-setup SUBXri $sp, 36, 12
@@ -57,7 +57,7 @@ body: |
; CHECK-NEXT: liveins: $w0, $x9
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: $sp = frame-setup SUBXri $sp, 1, 12
- ; CHECK-NEXT: frame-setup STRXui $xzr, $sp, 0
+ ; CHECK-NEXT: $xzr = frame-setup LDRXui $sp, 0
; CHECK-NEXT: $xzr = frame-setup SUBSXrx64 $sp, $x9, 24, implicit-def $nzcv
; CHECK-NEXT: frame-setup Bcc 1, %bb.3, implicit $nzcv
; CHECK-NEXT: {{ $}}
@@ -66,7 +66,7 @@ body: |
; CHECK-NEXT: liveins: $w0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: $sp = frame-setup SUBXri $sp, 2544, 0
- ; CHECK-NEXT: frame-setup STRXui $xzr, $sp, 0
+ ; CHECK-NEXT: $xzr = frame-setup LDRXui $sp, 0
; CHECK-NEXT: dead $wzr = SUBSWri killed renamable $w0, 1, 0, implicit-def $nzcv
; CHECK-NEXT: Bcc 12, %bb.2, implicit $nzcv
; CHECK-NEXT: B %bb.1
diff --git a/llvm/test/CodeGen/AArch64/stack-probing-sve.ll b/llvm/test/CodeGen/AArch64/stack-probing-sve.ll
index 79cf6d708c3e0..3f6d5dcaed8d0 100644
--- a/llvm/test/CodeGen/AArch64/stack-probing-sve.ll
+++ b/llvm/test/CodeGen/AArch64/stack-probing-sve.ll
@@ -64,7 +64,7 @@ define void @sve_16_vector(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: addvl sp, sp, #-16
; CHECK-NEXT: .cfi_escape 0x0f, 0x0a, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x80, 0x01, 0x1e, 0x22 // sp + 16 + 128 * VG
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: addvl sp, sp, #16
; CHECK-NEXT: .cfi_def_cfa wsp, 16
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
@@ -111,7 +111,7 @@ define void @sve_17_vector(ptr %out) #0 {
; CHECK-NEXT: b.le .LBB3_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB3_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB3_1
; CHECK-NEXT: .LBB3_3: // %entry
; CHECK-NEXT: mov sp, x9
@@ -218,7 +218,7 @@ define void @sve_16v_csr(<vscale x 4 x float> %a) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: addvl sp, sp, #-16
; CHECK-NEXT: .cfi_escape 0x0f, 0x0a, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x80, 0x01, 0x1e, 0x22 // sp + 16 + 128 * VG
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: str z23, [sp] // 16-byte Folded Spill
; CHECK-NEXT: str z22, [sp, #1, mul vl] // 16-byte Folded Spill
; CHECK-NEXT: str z21, [sp, #2, mul vl] // 16-byte Folded Spill
@@ -347,7 +347,7 @@ define void @sve_16v_1p_csr(<vscale x 4 x float> %a) #0 {
; CHECK-NEXT: b.le .LBB9_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB9_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB9_1
; CHECK-NEXT: .LBB9_3: // %entry
; CHECK-NEXT: mov sp, x9
@@ -463,7 +463,7 @@ define void @sve_1_vector_4096_arr(ptr %out) #0 {
; CHECK-NEXT: b.le .LBB11_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB11_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB11_1
; CHECK-NEXT: .LBB11_3: // %entry
; CHECK-NEXT: mov sp, x9
@@ -512,7 +512,7 @@ define void @sve_1_vector_16_arr_align_8192(ptr %out) #0 {
; CHECK-NEXT: b.le .LBB12_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB12_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB12_1
; CHECK-NEXT: .LBB12_3: // %entry
; CHECK-NEXT: mov sp, x9
@@ -553,7 +553,7 @@ define void @sve_1024_64k_guard(ptr %out) #0 "stack-probe-size"="65536" {
; CHECK-NEXT: .cfi_escape 0x0f, 0x0a, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x80, 0x0e, 0x1e, 0x22 // sp + 16 + 1792 * VG
; CHECK-NEXT: addvl sp, sp, #-32
; CHECK-NEXT: .cfi_escape 0x0f, 0x0a, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x80, 0x10, 0x1e, 0x22 // sp + 16 + 2048 * VG
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: addvl sp, sp, #31
; CHECK-NEXT: .cfi_escape 0x0f, 0x0a, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x88, 0x0e, 0x1e, 0x22 // sp + 16 + 1800 * VG
; CHECK-NEXT: addvl sp, sp, #31
@@ -612,7 +612,7 @@ define void @sve_1028_64k_guard(ptr %out) #0 "stack-probe-size"="65536" {
; CHECK-NEXT: b.le .LBB14_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB14_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB14_1
; CHECK-NEXT: .LBB14_3: // %entry
; CHECK-NEXT: mov sp, x9
@@ -657,7 +657,7 @@ define void @sve_5_vector(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: addvl sp, sp, #-5
; CHECK-NEXT: .cfi_escape 0x0f, 0x09, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x28, 0x1e, 0x22 // sp + 16 + 40 * VG
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: addvl sp, sp, #5
; CHECK-NEXT: .cfi_def_cfa wsp, 16
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
@@ -683,7 +683,7 @@ define void @sve_unprobed_area(<vscale x 4 x float> %a, i32 %n) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: addvl sp, sp, #-4
; CHECK-NEXT: .cfi_escape 0x0f, 0x09, 0x8f, 0x10, 0x92, 0x2e, 0x00, 0x11, 0x20, 0x1e, 0x22 // sp + 16 + 32 * VG
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: str p9, [sp, #7, mul vl] // 2-byte Spill
; CHECK-NEXT: str z10, [sp, #1, mul vl] // 16-byte Folded Spill
; CHECK-NEXT: str z9, [sp, #2, mul vl] // 16-byte Folded Spill
diff --git a/llvm/test/CodeGen/AArch64/stack-probing.ll b/llvm/test/CodeGen/AArch64/stack-probing.ll
index 3e0eaf1340aea..c2d9f5ff2c2ce 100644
--- a/llvm/test/CodeGen/AArch64/stack-probing.ll
+++ b/llvm/test/CodeGen/AArch64/stack-probing.ll
@@ -78,7 +78,7 @@ define void @static_1040(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #1040
; CHECK-NEXT: .cfi_def_cfa_offset 1056
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #1040
@@ -102,7 +102,7 @@ define void @static_4096(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 4112
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #1, lsl #12 // =4096
@@ -126,7 +126,7 @@ define void @static_4112(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 4112
-; CHECK-NEXT: str xzr, [sp], #-16
+; CHECK-NEXT: ldr xzr, [sp], #-16
; CHECK-NEXT: .cfi_def_cfa_offset 4128
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
@@ -153,7 +153,7 @@ define void @static_5120(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 4112
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1024
; CHECK-NEXT: .cfi_def_cfa_offset 5136
; CHECK-NEXT: mov x8, sp
@@ -181,10 +181,10 @@ define void @static_5136(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 4112
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1040
; CHECK-NEXT: .cfi_def_cfa_offset 5152
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #1, lsl #12 // =4096
@@ -210,10 +210,10 @@ define void @static_9216(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 4112
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 8208
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1024
; CHECK-NEXT: .cfi_def_cfa_offset 9232
; CHECK-NEXT: mov x8, sp
@@ -241,19 +241,19 @@ define void @static_20464(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 4112
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 8208
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 12304
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
; CHECK-NEXT: .cfi_def_cfa_offset 16400
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: sub sp, sp, #4080
; CHECK-NEXT: .cfi_def_cfa_offset 20480
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #4, lsl #12 // =16384
@@ -282,7 +282,7 @@ define void @static_20480(ptr %out) #0 {
; CHECK-NEXT: .LBB10_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.ne .LBB10_1
; CHECK-NEXT: // %bb.2: // %entry
@@ -314,7 +314,7 @@ define void @static_21504(ptr %out) #0 {
; CHECK-NEXT: .LBB11_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.ne .LBB11_1
; CHECK-NEXT: // %bb.2: // %entry
@@ -350,14 +350,14 @@ define void @static_21520(ptr %out) #0 {
; CHECK-NEXT: .LBB12_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: sub sp, sp, #1, lsl #12 // =4096
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: cmp sp, x9
; CHECK-NEXT: b.ne .LBB12_1
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: .cfi_def_cfa_register wsp
; CHECK-NEXT: sub sp, sp, #1040
; CHECK-NEXT: .cfi_def_cfa_offset 21536
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: add sp, sp, #5, lsl #12 // =20480
@@ -396,7 +396,7 @@ define void @static_16_align_8192(ptr %out) #0 {
; CHECK-NEXT: b.le .LBB13_3
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: // in Loop: Header=BB13_1 Depth=1
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: b .LBB13_1
; CHECK-NEXT: .LBB13_3: // %entry
; CHECK-NEXT: mov sp, x9
@@ -429,7 +429,7 @@ define void @static_16_align_2048(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub x9, sp, #2032
; CHECK-NEXT: and sp, x9, #0xfffffffffffff800
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: mov sp, x29
@@ -458,7 +458,7 @@ define void @static_2032_align_2048(ptr %out) #0 {
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: sub x9, sp, #2032
; CHECK-NEXT: and sp, x9, #0xfffffffffffff800
-; CHECK-NEXT: str xzr, [sp]
+; CHECK-NEXT: ldr xzr, [sp]
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
; CHECK-NEXT: mov sp, x29
@@ -485,7 +485,7 @@ define void @static_9232(ptr %out) uwtable(async) {
; CHECK-NEXT: .cfi_def_cfa_offset 8208
; CHECK-NEXT: sub sp, sp, #800
; CHECK-NEXT: .cfi_def_cfa_offset 9008
-; CHECK-NEXT: str xzr, [sp], #-240
+; CHECK-NEXT: ldr xzr, [sp], #-240
; CHECK-NEXT: .cfi_def_cfa_offset 9248
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: str x8, [x0]
More information about the llvm-commits
mailing list