[llvm] 41f430a - [X86] Don't fold very large offsets into addr displacements during ISel (#121678)
via llvm-commits
llvm-commits at lists.llvm.org
Fri Jan 17 05:09:04 PST 2025
Author: Wesley Wiser
Date: 2025-01-17T20:09:00+07:00
New Revision: 41f430a48db992477534b65b288b47d487c4797d
URL: https://github.com/llvm/llvm-project/commit/41f430a48db992477534b65b288b47d487c4797d
DIFF: https://github.com/llvm/llvm-project/commit/41f430a48db992477534b65b288b47d487c4797d.diff
LOG: [X86] Don't fold very large offsets into addr displacements during ISel (#121678)
Doing so can cause the resulting displacement after frame layout to
become inexpressible (or cause over/underflow currently during frame
layout).
Fixes the error reported in
https://github.com/llvm/llvm-project/pull/101840#issuecomment-2306975944.
Added:
llvm/test/CodeGen/X86/dag-large-offset.ll
Modified:
llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
llvm/test/CodeGen/X86/xor-lea.ll
Removed:
################################################################################
diff --git a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
index 9b340a778b36ad..84bcdae520885e 100644
--- a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
+++ b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
@@ -1800,10 +1800,10 @@ void X86DAGToDAGISel::emitFunctionEntryCode() {
emitSpecialCodeForMain();
}
-static bool isDispSafeForFrameIndex(int64_t Val) {
- // On 64-bit platforms, we can run into an issue where a frame index
+static bool isDispSafeForFrameIndexOrRegBase(int64_t Val) {
+ // We can run into an issue where a frame index or a register base
// includes a displacement that, when added to the explicit displacement,
- // will overflow the displacement field. Assuming that the frame index
+ // will overflow the displacement field. Assuming that the
// displacement fits into a 31-bit integer (which is only slightly more
// aggressive than the current fundamental assumption that it fits into
// a 32-bit integer), a 31-bit disp should always be safe.
@@ -1831,7 +1831,7 @@ bool X86DAGToDAGISel::foldOffsetIntoAddress(uint64_t Offset,
// In addition to the checks required for a register base, check that
// we do not try to use an unsafe Disp with a frame index.
if (AM.BaseType == X86ISelAddressMode::FrameIndexBase &&
- !isDispSafeForFrameIndex(Val))
+ !isDispSafeForFrameIndexOrRegBase(Val))
return true;
// In ILP32 (x32) mode, pointers are 32 bits and need to be zero-extended to
// 64 bits. Instructions with 32-bit register addresses perform this zero
@@ -1849,10 +1849,14 @@ bool X86DAGToDAGISel::foldOffsetIntoAddress(uint64_t Offset,
// to get an address size override to be emitted. However, this
// pseudo-register is not part of any register class and therefore causes
// MIR verification to fail.
- if (Subtarget->isTarget64BitILP32() && !isUInt<31>(Val) &&
+ if (Subtarget->isTarget64BitILP32() &&
+ !isDispSafeForFrameIndexOrRegBase((uint32_t)Val) &&
!AM.hasBaseOrIndexReg())
return true;
- }
+ } else if (AM.hasBaseOrIndexReg() && !isDispSafeForFrameIndexOrRegBase(Val))
+ // For 32-bit X86, make sure the displacement still isn't close to the
+ // expressible limit.
+ return true;
AM.Disp = Val;
return false;
}
@@ -2553,7 +2557,7 @@ bool X86DAGToDAGISel::matchAddressRecursively(SDValue N, X86ISelAddressMode &AM,
case ISD::FrameIndex:
if (AM.BaseType == X86ISelAddressMode::RegBase &&
AM.Base_Reg.getNode() == nullptr &&
- (!Subtarget->is64Bit() || isDispSafeForFrameIndex(AM.Disp))) {
+ (!Subtarget->is64Bit() || isDispSafeForFrameIndexOrRegBase(AM.Disp))) {
AM.BaseType = X86ISelAddressMode::FrameIndexBase;
AM.Base_FrameIndex = cast<FrameIndexSDNode>(N)->getIndex();
return false;
diff --git a/llvm/test/CodeGen/X86/dag-large-offset.ll b/llvm/test/CodeGen/X86/dag-large-offset.ll
new file mode 100644
index 00000000000000..2774a939931539
--- /dev/null
+++ b/llvm/test/CodeGen/X86/dag-large-offset.ll
@@ -0,0 +1,47 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc < %s -mtriple=i386 --frame-pointer=all | FileCheck %s
+
+; ISel will try to fold pointer arithmetic into the address displacement. However, we don't
+; want to do that if the offset is very close to the expressible limit because the final frame
+; layout may push it over/under the limit.
+
+define i32 @foo(i1 %b) #0 {
+; CHECK-LABEL: foo:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: pushl %ebp
+; CHECK-NEXT: .cfi_def_cfa_offset 8
+; CHECK-NEXT: .cfi_offset %ebp, -8
+; CHECK-NEXT: movl %esp, %ebp
+; CHECK-NEXT: .cfi_def_cfa_register %ebp
+; CHECK-NEXT: subl $8, %esp
+; CHECK-NEXT: movl __stack_chk_guard, %eax
+; CHECK-NEXT: movl %eax, -4(%ebp)
+; CHECK-NEXT: testb $1, 8(%ebp)
+; CHECK-NEXT: jne .LBB0_1
+; CHECK-NEXT: # %bb.2: # %entry
+; CHECK-NEXT: xorl %eax, %eax
+; CHECK-NEXT: jmp .LBB0_3
+; CHECK-NEXT: .LBB0_1:
+; CHECK-NEXT: movl $-2147483647, %eax # imm = 0x80000001
+; CHECK-NEXT: leal -5(%ebp,%eax), %eax
+; CHECK-NEXT: .LBB0_3: # %entry
+; CHECK-NEXT: movl __stack_chk_guard, %ecx
+; CHECK-NEXT: cmpl -4(%ebp), %ecx
+; CHECK-NEXT: jne .LBB0_5
+; CHECK-NEXT: # %bb.4: # %entry
+; CHECK-NEXT: addl $8, %esp
+; CHECK-NEXT: popl %ebp
+; CHECK-NEXT: .cfi_def_cfa %esp, 4
+; CHECK-NEXT: retl
+; CHECK-NEXT: .LBB0_5: # %entry
+; CHECK-NEXT: .cfi_def_cfa %ebp, 8
+; CHECK-NEXT: calll __stack_chk_fail
+entry:
+ %a = alloca i8, align 1
+ %0 = ptrtoint ptr %a to i32
+ %sub = add i32 %0, -2147483647
+ %retval.0 = select i1 %b, i32 %sub, i32 0
+ ret i32 %retval.0
+}
+
+attributes #0 = { sspreq }
diff --git a/llvm/test/CodeGen/X86/xor-lea.ll b/llvm/test/CodeGen/X86/xor-lea.ll
index 10e9525a2706a3..d50752e48d293f 100644
--- a/llvm/test/CodeGen/X86/xor-lea.ll
+++ b/llvm/test/CodeGen/X86/xor-lea.ll
@@ -327,7 +327,8 @@ define i32 @xor_shl_sminval_i32(i32 %x) {
; X86-LABEL: xor_shl_sminval_i32:
; X86: # %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT: leal -2147483648(,%eax,8), %eax
+; X86-NEXT: movl $-2147483648, %ecx # imm = 0x80000000
+; X86-NEXT: leal (%ecx,%eax,8), %eax
; X86-NEXT: retl
;
; X64-LABEL: xor_shl_sminval_i32:
More information about the llvm-commits
mailing list