aboutsummaryrefslogtreecommitdiff
path: root/llvm/test/CodeGen/X86
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/test/CodeGen/X86')
-rw-r--r--llvm/test/CodeGen/X86/GlobalISel/legalize-memop-scalar-32.mir38
-rw-r--r--llvm/test/CodeGen/X86/GlobalISel/legalize-undef.mir3
-rw-r--r--llvm/test/CodeGen/X86/GlobalISel/regbankselect-x87.ll4
-rw-r--r--llvm/test/CodeGen/X86/GlobalISel/x86_64-irtranslator-struct-return.ll24
-rw-r--r--llvm/test/CodeGen/X86/embed-bitcode.ll13
-rw-r--r--llvm/test/CodeGen/X86/late-tail-dup-computed-goto.mir128
-rw-r--r--llvm/test/CodeGen/X86/swap.ll15
7 files changed, 180 insertions, 45 deletions
diff --git a/llvm/test/CodeGen/X86/GlobalISel/legalize-memop-scalar-32.mir b/llvm/test/CodeGen/X86/GlobalISel/legalize-memop-scalar-32.mir
index ba72c4f..bb7f921 100644
--- a/llvm/test/CodeGen/X86/GlobalISel/legalize-memop-scalar-32.mir
+++ b/llvm/test/CodeGen/X86/GlobalISel/legalize-memop-scalar-32.mir
@@ -10,18 +10,18 @@ body: |
bb.0:
; X32-LABEL: name: test_memop_s8tos32
; X32: [[DEF:%[0-9]+]]:_(p0) = IMPLICIT_DEF
- ; X32: [[LOAD:%[0-9]+]]:_(s8) = G_LOAD [[DEF]](p0) :: (load (s1))
- ; X32: [[LOAD1:%[0-9]+]]:_(s8) = G_LOAD [[DEF]](p0) :: (load (s8))
- ; X32: [[LOAD2:%[0-9]+]]:_(s16) = G_LOAD [[DEF]](p0) :: (load (s16))
- ; X32: [[LOAD3:%[0-9]+]]:_(s32) = G_LOAD [[DEF]](p0) :: (load (s32))
- ; X32: [[LOAD4:%[0-9]+]]:_(p0) = G_LOAD [[DEF]](p0) :: (load (p0))
- ; X32: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 1
- ; X32: [[AND:%[0-9]+]]:_(s8) = G_AND [[LOAD]], [[C]]
- ; X32: G_STORE [[AND]](s8), [[DEF]](p0) :: (store (s8))
- ; X32: G_STORE [[LOAD1]](s8), [[DEF]](p0) :: (store (s8))
- ; X32: G_STORE [[LOAD2]](s16), [[DEF]](p0) :: (store (s16))
- ; X32: G_STORE [[LOAD3]](s32), [[DEF]](p0) :: (store (s32))
- ; X32: G_STORE [[LOAD4]](p0), [[DEF]](p0) :: (store (p0))
+ ; X32-NEXT: [[LOAD:%[0-9]+]]:_(s8) = G_LOAD [[DEF]](p0) :: (load (s1))
+ ; X32-NEXT: [[LOAD1:%[0-9]+]]:_(s8) = G_LOAD [[DEF]](p0) :: (load (s8))
+ ; X32-NEXT: [[LOAD2:%[0-9]+]]:_(s16) = G_LOAD [[DEF]](p0) :: (load (s16))
+ ; X32-NEXT: [[LOAD3:%[0-9]+]]:_(s32) = G_LOAD [[DEF]](p0) :: (load (s32))
+ ; X32-NEXT: [[LOAD4:%[0-9]+]]:_(p0) = G_LOAD [[DEF]](p0) :: (load (p0))
+ ; X32-NEXT: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 1
+ ; X32-NEXT: [[AND:%[0-9]+]]:_(s8) = G_AND [[LOAD]], [[C]]
+ ; X32-NEXT: G_STORE [[AND]](s8), [[DEF]](p0) :: (store (s8))
+ ; X32-NEXT: G_STORE [[LOAD1]](s8), [[DEF]](p0) :: (store (s8))
+ ; X32-NEXT: G_STORE [[LOAD2]](s16), [[DEF]](p0) :: (store (s16))
+ ; X32-NEXT: G_STORE [[LOAD3]](s32), [[DEF]](p0) :: (store (s32))
+ ; X32-NEXT: G_STORE [[LOAD4]](p0), [[DEF]](p0) :: (store (p0))
%0:_(p0) = IMPLICIT_DEF
%9:_(s1) = G_LOAD %0 :: (load (s1))
%1:_(s8) = G_LOAD %0 :: (load (s8))
@@ -46,13 +46,13 @@ body: |
; X32-LABEL: name: test_memop_s64
; X32: [[DEF:%[0-9]+]]:_(p0) = IMPLICIT_DEF
- ; X32: [[LOAD:%[0-9]+]]:_(s32) = G_LOAD [[DEF]](p0) :: (load (s32), align 8)
- ; X32: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 4
- ; X32: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[DEF]], [[C]](s32)
- ; X32: [[LOAD1:%[0-9]+]]:_(s32) = G_LOAD [[PTR_ADD]](p0) :: (load (s32) from unknown-address + 4)
- ; X32: G_STORE [[LOAD]](s32), [[DEF]](p0) :: (store (s32), align 8)
- ; X32: [[PTR_ADD1:%[0-9]+]]:_(p0) = G_PTR_ADD [[DEF]], [[C]](s32)
- ; X32: G_STORE [[LOAD1]](s32), [[PTR_ADD1]](p0) :: (store (s32) into unknown-address + 4)
+ ; X32-NEXT: [[LOAD:%[0-9]+]]:_(s32) = G_LOAD [[DEF]](p0) :: (load (s32), align 8)
+ ; X32-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 4
+ ; X32-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = nuw G_PTR_ADD [[DEF]], [[C]](s32)
+ ; X32-NEXT: [[LOAD1:%[0-9]+]]:_(s32) = G_LOAD [[PTR_ADD]](p0) :: (load (s32) from unknown-address + 4)
+ ; X32-NEXT: G_STORE [[LOAD]](s32), [[DEF]](p0) :: (store (s32), align 8)
+ ; X32-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p0) = nuw G_PTR_ADD [[DEF]], [[C]](s32)
+ ; X32-NEXT: G_STORE [[LOAD1]](s32), [[PTR_ADD1]](p0) :: (store (s32) into unknown-address + 4)
%0:_(p0) = IMPLICIT_DEF
%1:_(s64) = G_LOAD %0 :: (load (s64))
diff --git a/llvm/test/CodeGen/X86/GlobalISel/legalize-undef.mir b/llvm/test/CodeGen/X86/GlobalISel/legalize-undef.mir
index 8711d84..a7f2cbd 100644
--- a/llvm/test/CodeGen/X86/GlobalISel/legalize-undef.mir
+++ b/llvm/test/CodeGen/X86/GlobalISel/legalize-undef.mir
@@ -21,6 +21,7 @@ body: |
; X64-NEXT: G_STORE [[DEF3]](s32), [[DEF]](p0) :: (store (s32))
; X64-NEXT: [[DEF4:%[0-9]+]]:_(s64) = G_IMPLICIT_DEF
; X64-NEXT: G_STORE [[DEF4]](s64), [[DEF]](p0) :: (store (s64))
+ ;
; X32-LABEL: name: test_implicit_def
; X32: [[DEF:%[0-9]+]]:_(p0) = G_IMPLICIT_DEF
; X32-NEXT: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 0
@@ -35,7 +36,7 @@ body: |
; X32-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[DEF4]](s64)
; X32-NEXT: G_STORE [[UV]](s32), [[DEF]](p0) :: (store (s32), align 8)
; X32-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 4
- ; X32-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[DEF]], [[C1]](s32)
+ ; X32-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = nuw G_PTR_ADD [[DEF]], [[C1]](s32)
; X32-NEXT: G_STORE [[UV1]](s32), [[PTR_ADD]](p0) :: (store (s32) into unknown-address + 4)
%5:_(p0) = G_IMPLICIT_DEF
%0:_(s1) = G_IMPLICIT_DEF
diff --git a/llvm/test/CodeGen/X86/GlobalISel/regbankselect-x87.ll b/llvm/test/CodeGen/X86/GlobalISel/regbankselect-x87.ll
index 99d458a..93c279e 100644
--- a/llvm/test/CodeGen/X86/GlobalISel/regbankselect-x87.ll
+++ b/llvm/test/CodeGen/X86/GlobalISel/regbankselect-x87.ll
@@ -164,12 +164,12 @@ define void @f5(ptr %a, ptr %b) {
; X86-NEXT: [[LOAD1:%[0-9]+]]:gpr(p0) = G_LOAD [[FRAME_INDEX1]](p0) :: (invariant load (p0) from %fixed-stack.0)
; X86-NEXT: [[LOAD2:%[0-9]+]]:gpr(s32) = G_LOAD [[LOAD]](p0) :: (load (s32) from %ir.a, align 8)
; X86-NEXT: [[C:%[0-9]+]]:gpr(s32) = G_CONSTANT i32 4
- ; X86-NEXT: [[PTR_ADD:%[0-9]+]]:gpr(p0) = G_PTR_ADD [[LOAD]], [[C]](s32)
+ ; X86-NEXT: [[PTR_ADD:%[0-9]+]]:gpr(p0) = nuw G_PTR_ADD [[LOAD]], [[C]](s32)
; X86-NEXT: [[COPY:%[0-9]+]]:gpr(p0) = COPY [[PTR_ADD]](p0)
; X86-NEXT: [[LOAD3:%[0-9]+]]:gpr(s32) = G_LOAD [[COPY]](p0) :: (load (s32) from %ir.a + 4, basealign 8)
; X86-NEXT: [[MV:%[0-9]+]]:gpr(s64) = G_MERGE_VALUES [[LOAD2]](s32), [[LOAD3]](s32)
; X86-NEXT: [[LOAD4:%[0-9]+]]:gpr(s32) = G_LOAD [[LOAD1]](p0) :: (load (s32) from %ir.b, align 8)
- ; X86-NEXT: [[PTR_ADD1:%[0-9]+]]:gpr(p0) = G_PTR_ADD [[LOAD1]], [[C]](s32)
+ ; X86-NEXT: [[PTR_ADD1:%[0-9]+]]:gpr(p0) = nuw G_PTR_ADD [[LOAD1]], [[C]](s32)
; X86-NEXT: [[LOAD5:%[0-9]+]]:gpr(s32) = G_LOAD [[PTR_ADD1]](p0) :: (load (s32) from %ir.b + 4, basealign 8)
; X86-NEXT: [[MV1:%[0-9]+]]:gpr(s64) = G_MERGE_VALUES [[LOAD4]](s32), [[LOAD5]](s32)
; X86-NEXT: [[COPY1:%[0-9]+]]:psr(s64) = COPY [[MV]](s64)
diff --git a/llvm/test/CodeGen/X86/GlobalISel/x86_64-irtranslator-struct-return.ll b/llvm/test/CodeGen/X86/GlobalISel/x86_64-irtranslator-struct-return.ll
index 171ccb2..40de4d0 100644
--- a/llvm/test/CodeGen/X86/GlobalISel/x86_64-irtranslator-struct-return.ll
+++ b/llvm/test/CodeGen/X86/GlobalISel/x86_64-irtranslator-struct-return.ll
@@ -77,12 +77,12 @@ define { double, double } @test_return_d2(double %d.coerce0, double %d.coerce1)
; ALL-NEXT: [[FRAME_INDEX1:%[0-9]+]]:_(p0) = G_FRAME_INDEX %stack.1.d
; ALL-NEXT: G_STORE [[COPY]](s64), [[FRAME_INDEX1]](p0) :: (store (s64) into %ir.1)
; ALL-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 8
- ; ALL-NEXT: %5:_(p0) = nuw nusw G_PTR_ADD [[FRAME_INDEX1]], [[C1]](s64)
- ; ALL-NEXT: G_STORE [[COPY1]](s64), %5(p0) :: (store (s64) into %ir.2)
+ ; ALL-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = nuw nusw G_PTR_ADD [[FRAME_INDEX1]], [[C1]](s64)
+ ; ALL-NEXT: G_STORE [[COPY1]](s64), [[PTR_ADD]](p0) :: (store (s64) into %ir.2)
; ALL-NEXT: G_MEMCPY [[FRAME_INDEX]](p0), [[FRAME_INDEX1]](p0), [[C]](s64), 0 :: (store (s8) into %ir.3, align 8), (load (s8) from %ir.4, align 8)
; ALL-NEXT: [[LOAD:%[0-9]+]]:_(s64) = G_LOAD [[FRAME_INDEX]](p0) :: (dereferenceable load (s64) from %ir.5)
- ; ALL-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[FRAME_INDEX]], [[C1]](s64)
- ; ALL-NEXT: [[LOAD1:%[0-9]+]]:_(s64) = G_LOAD [[PTR_ADD]](p0) :: (dereferenceable load (s64) from %ir.5 + 8)
+ ; ALL-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p0) = nuw G_PTR_ADD [[FRAME_INDEX]], [[C1]](s64)
+ ; ALL-NEXT: [[LOAD1:%[0-9]+]]:_(s64) = G_LOAD [[PTR_ADD1]](p0) :: (dereferenceable load (s64) from %ir.5 + 8)
; ALL-NEXT: $xmm0 = COPY [[LOAD]](s64)
; ALL-NEXT: $xmm1 = COPY [[LOAD1]](s64)
; ALL-NEXT: RET 0, implicit $xmm0, implicit $xmm1
@@ -170,14 +170,14 @@ define { i64, i32 } @test_return_i3(i64 %i.coerce0, i32 %i.coerce1) {
; ALL-NEXT: [[FRAME_INDEX3:%[0-9]+]]:_(p0) = G_FRAME_INDEX %stack.3.tmp
; ALL-NEXT: G_STORE [[COPY]](s64), [[FRAME_INDEX2]](p0) :: (store (s64) into %ir.0, align 4)
; ALL-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 8
- ; ALL-NEXT: %7:_(p0) = nuw nusw G_PTR_ADD [[FRAME_INDEX2]], [[C1]](s64)
- ; ALL-NEXT: G_STORE [[COPY1]](s32), %7(p0) :: (store (s32) into %ir.1)
+ ; ALL-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = nuw nusw G_PTR_ADD [[FRAME_INDEX2]], [[C1]](s64)
+ ; ALL-NEXT: G_STORE [[COPY1]](s32), [[PTR_ADD]](p0) :: (store (s32) into %ir.1)
; ALL-NEXT: G_MEMCPY [[FRAME_INDEX1]](p0), [[FRAME_INDEX2]](p0), [[C]](s64), 0 :: (store (s8) into %ir.2, align 4), (load (s8) from %ir.3, align 4)
; ALL-NEXT: G_MEMCPY [[FRAME_INDEX]](p0), [[FRAME_INDEX1]](p0), [[C]](s64), 0 :: (store (s8) into %ir.4, align 4), (load (s8) from %ir.5, align 4)
; ALL-NEXT: G_MEMCPY [[FRAME_INDEX3]](p0), [[FRAME_INDEX]](p0), [[C]](s64), 0 :: (store (s8) into %ir.6, align 8), (load (s8) from %ir.7, align 4)
; ALL-NEXT: [[LOAD:%[0-9]+]]:_(s64) = G_LOAD [[FRAME_INDEX3]](p0) :: (dereferenceable load (s64) from %ir.tmp)
- ; ALL-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[FRAME_INDEX3]], [[C1]](s64)
- ; ALL-NEXT: [[LOAD1:%[0-9]+]]:_(s32) = G_LOAD [[PTR_ADD]](p0) :: (dereferenceable load (s32) from %ir.tmp + 8, align 8)
+ ; ALL-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p0) = nuw G_PTR_ADD [[FRAME_INDEX3]], [[C1]](s64)
+ ; ALL-NEXT: [[LOAD1:%[0-9]+]]:_(s32) = G_LOAD [[PTR_ADD1]](p0) :: (dereferenceable load (s32) from %ir.tmp + 8, align 8)
; ALL-NEXT: $rax = COPY [[LOAD]](s64)
; ALL-NEXT: $edx = COPY [[LOAD1]](s32)
; ALL-NEXT: RET 0, implicit $rax, implicit $edx
@@ -215,12 +215,12 @@ define { i64, i64 } @test_return_i4(i64 %i.coerce0, i64 %i.coerce1) {
; ALL-NEXT: [[FRAME_INDEX1:%[0-9]+]]:_(p0) = G_FRAME_INDEX %stack.1.i
; ALL-NEXT: G_STORE [[COPY]](s64), [[FRAME_INDEX1]](p0) :: (store (s64) into %ir.1, align 4)
; ALL-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 8
- ; ALL-NEXT: %5:_(p0) = nuw nusw G_PTR_ADD [[FRAME_INDEX1]], [[C1]](s64)
- ; ALL-NEXT: G_STORE [[COPY1]](s64), %5(p0) :: (store (s64) into %ir.2, align 4)
+ ; ALL-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = nuw nusw G_PTR_ADD [[FRAME_INDEX1]], [[C1]](s64)
+ ; ALL-NEXT: G_STORE [[COPY1]](s64), [[PTR_ADD]](p0) :: (store (s64) into %ir.2, align 4)
; ALL-NEXT: G_MEMCPY [[FRAME_INDEX]](p0), [[FRAME_INDEX1]](p0), [[C]](s64), 0 :: (store (s8) into %ir.3, align 4), (load (s8) from %ir.4, align 4)
; ALL-NEXT: [[LOAD:%[0-9]+]]:_(s64) = G_LOAD [[FRAME_INDEX]](p0) :: (dereferenceable load (s64) from %ir.5, align 4)
- ; ALL-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[FRAME_INDEX]], [[C1]](s64)
- ; ALL-NEXT: [[LOAD1:%[0-9]+]]:_(s64) = G_LOAD [[PTR_ADD]](p0) :: (dereferenceable load (s64) from %ir.5 + 8, align 4)
+ ; ALL-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p0) = nuw G_PTR_ADD [[FRAME_INDEX]], [[C1]](s64)
+ ; ALL-NEXT: [[LOAD1:%[0-9]+]]:_(s64) = G_LOAD [[PTR_ADD1]](p0) :: (dereferenceable load (s64) from %ir.5 + 8, align 4)
; ALL-NEXT: $rax = COPY [[LOAD]](s64)
; ALL-NEXT: $rdx = COPY [[LOAD1]](s64)
; ALL-NEXT: RET 0, implicit $rax, implicit $rdx
diff --git a/llvm/test/CodeGen/X86/embed-bitcode.ll b/llvm/test/CodeGen/X86/embed-bitcode.ll
index 0d66ba8..d4af954 100644
--- a/llvm/test/CodeGen/X86/embed-bitcode.ll
+++ b/llvm/test/CodeGen/X86/embed-bitcode.ll
@@ -1,10 +1,23 @@
; RUN: llc -filetype=obj -mtriple=x86_64 %s -o %t
; RUN: llvm-readelf -S %t | FileCheck %s
+; RUN: llc -filetype=obj -mtriple=x86_64-pc-windows-msvc %s -o %t
+; RUN: llvm-readobj -S %t | FileCheck %s --check-prefix=COFF
; CHECK: .text PROGBITS 0000000000000000 [[#%x,OFF:]] 000000 00 AX 0
; CHECK-NEXT: .llvmbc PROGBITS 0000000000000000 [[#%x,OFF:]] 000004 00 0
; CHECK-NEXT: .llvmcmd PROGBITS 0000000000000000 [[#%x,OFF:]] 000005 00 0
+; COFF: Name: .llvmbc (2E 6C 6C 76 6D 62 63 00)
+; COFF: Characteristics [
+; COFF-NEXT: IMAGE_SCN_ALIGN_1BYTES
+; COFF-NEXT: IMAGE_SCN_MEM_DISCARDABLE
+; COFF-NEXT: ]
+; COFF: Name: .llvmcmd (2E 6C 6C 76 6D 63 6D 64)
+; COFF: Characteristics [
+; COFF-NEXT: IMAGE_SCN_ALIGN_1BYTES
+; COFF-NEXT: IMAGE_SCN_MEM_DISCARDABLE
+; COFF-NEXT: ]
+
@llvm.embedded.module = private constant [4 x i8] c"BC\C0\DE", section ".llvmbc", align 1
@llvm.cmdline = private constant [5 x i8] c"-cc1\00", section ".llvmcmd", align 1
@llvm.compiler.used = appending global [2 x ptr] [ptr @llvm.embedded.module, ptr @llvm.cmdline], section "llvm.metadata"
diff --git a/llvm/test/CodeGen/X86/late-tail-dup-computed-goto.mir b/llvm/test/CodeGen/X86/late-tail-dup-computed-goto.mir
new file mode 100644
index 0000000..e272e7e
--- /dev/null
+++ b/llvm/test/CodeGen/X86/late-tail-dup-computed-goto.mir
@@ -0,0 +1,128 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5
+# RUN: llc -mtriple=x86_64-unknown-linux-gnu -run-pass=tailduplication -tail-dup-pred-size=1 -tail-dup-succ-size=1 %s -o - | FileCheck %s
+#
+# Check that only the computed gotos are duplicated aggressively.
+#
+--- |
+ @computed_goto.dispatch = constant [5 x ptr] [ptr null, ptr blockaddress(@computed_goto, %bb1), ptr blockaddress(@computed_goto, %bb2), ptr blockaddress(@computed_goto, %bb3), ptr blockaddress(@computed_goto, %bb4)]
+ declare i64 @f0()
+ declare i64 @f1()
+ declare i64 @f2()
+ declare i64 @f3()
+ declare i64 @f4()
+ declare i64 @f5()
+ define void @computed_goto() {
+ start:
+ ret void
+ bb1:
+ ret void
+ bb2:
+ ret void
+ bb3:
+ ret void
+ bb4:
+ ret void
+ }
+ define void @jump_table() { ret void }
+ define void @jump_table_pic() { ret void }
+...
+---
+name: computed_goto
+alignment: 1
+exposesReturnsTwice: false
+legalized: false
+regBankSelected: false
+selected: false
+failedISel: false
+tracksRegLiveness: true
+body: |
+ ; CHECK-LABEL: name: computed_goto
+ ; CHECK: bb.0:
+ ; CHECK-NEXT: successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: CALL64pcrel32 target-flags(x86-plt) @f0, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:gr64 = COPY $rax
+ ; CHECK-NEXT: [[COPY1:%[0-9]+]]:gr64_nosp = COPY [[COPY]]
+ ; CHECK-NEXT: [[COPY2:%[0-9]+]]:gr64_nosp = COPY [[COPY1]]
+ ; CHECK-NEXT: JMP64m $noreg, 8, [[COPY2]], @computed_goto.dispatch, $noreg
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: bb.1.bb1 (ir-block-address-taken %ir-block.bb1):
+ ; CHECK-NEXT: successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: CALL64pcrel32 target-flags(x86-plt) @f1, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ ; CHECK-NEXT: [[COPY3:%[0-9]+]]:gr64 = COPY $rax
+ ; CHECK-NEXT: [[COPY1:%[0-9]+]]:gr64_nosp = COPY [[COPY3]]
+ ; CHECK-NEXT: [[COPY2:%[0-9]+]]:gr64_nosp = COPY [[COPY1]]
+ ; CHECK-NEXT: JMP64m $noreg, 8, [[COPY2]], @computed_goto.dispatch, $noreg
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: bb.2.bb2 (ir-block-address-taken %ir-block.bb2):
+ ; CHECK-NEXT: successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: CALL64pcrel32 target-flags(x86-plt) @f2, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ ; CHECK-NEXT: [[COPY4:%[0-9]+]]:gr64 = COPY $rax
+ ; CHECK-NEXT: [[COPY1:%[0-9]+]]:gr64_nosp = COPY [[COPY4]]
+ ; CHECK-NEXT: [[COPY2:%[0-9]+]]:gr64_nosp = COPY [[COPY1]]
+ ; CHECK-NEXT: JMP64m $noreg, 8, [[COPY2]], @computed_goto.dispatch, $noreg
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: bb.3.bb3 (ir-block-address-taken %ir-block.bb3):
+ ; CHECK-NEXT: successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: CALL64pcrel32 target-flags(x86-plt) @f3, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ ; CHECK-NEXT: [[COPY5:%[0-9]+]]:gr64 = COPY $rax
+ ; CHECK-NEXT: [[COPY1:%[0-9]+]]:gr64_nosp = COPY [[COPY5]]
+ ; CHECK-NEXT: [[COPY2:%[0-9]+]]:gr64_nosp = COPY [[COPY1]]
+ ; CHECK-NEXT: JMP64m $noreg, 8, [[COPY2]], @computed_goto.dispatch, $noreg
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: bb.4.bb4 (ir-block-address-taken %ir-block.bb4):
+ ; CHECK-NEXT: successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: CALL64pcrel32 target-flags(x86-plt) @f4, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ ; CHECK-NEXT: [[COPY6:%[0-9]+]]:gr64 = COPY $rax
+ ; CHECK-NEXT: [[COPY1:%[0-9]+]]:gr64_nosp = COPY [[COPY6]]
+ ; CHECK-NEXT: [[COPY2:%[0-9]+]]:gr64_nosp = COPY [[COPY1]]
+ ; CHECK-NEXT: JMP64m $noreg, 8, [[COPY2]], @computed_goto.dispatch, $noreg
+ bb.0:
+ successors: %bb.5(0x80000000)
+
+ CALL64pcrel32 target-flags(x86-plt) @f0, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ %0:gr64 = COPY $rax
+ %6:gr64_nosp = COPY %0
+ JMP_1 %bb.5
+
+ bb.1.bb1 (ir-block-address-taken %ir-block.bb1):
+ successors: %bb.5(0x80000000)
+
+ CALL64pcrel32 target-flags(x86-plt) @f1, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ %1:gr64 = COPY $rax
+ %6:gr64_nosp = COPY %1
+ JMP_1 %bb.5
+
+ bb.2.bb2 (ir-block-address-taken %ir-block.bb2):
+ successors: %bb.5(0x80000000)
+
+ CALL64pcrel32 target-flags(x86-plt) @f2, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ %2:gr64 = COPY $rax
+ %6:gr64_nosp = COPY %2
+ JMP_1 %bb.5
+
+ bb.3.bb3 (ir-block-address-taken %ir-block.bb3):
+ successors: %bb.5(0x80000000)
+
+ CALL64pcrel32 target-flags(x86-plt) @f3, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ %3:gr64 = COPY $rax
+ %6:gr64_nosp = COPY %3
+ JMP_1 %bb.5
+
+ bb.4.bb4 (ir-block-address-taken %ir-block.bb4):
+ successors: %bb.5(0x80000000)
+
+ CALL64pcrel32 target-flags(x86-plt) @f4, csr_64, implicit $rsp, implicit $ssp, implicit-def $rsp, implicit-def $ssp, implicit-def $rax
+ %4:gr64 = COPY $rax
+ %6:gr64_nosp = COPY %4
+
+ bb.5:
+ successors: %bb.1(0x20000000), %bb.2(0x20000000), %bb.3(0x20000000), %bb.4(0x20000000)
+
+ %5:gr64_nosp = COPY %6
+ JMP64m $noreg, 8, %5, @computed_goto.dispatch, $noreg
+...
diff --git a/llvm/test/CodeGen/X86/swap.ll b/llvm/test/CodeGen/X86/swap.ll
index 1dc454dd..3330403 100644
--- a/llvm/test/CodeGen/X86/swap.ll
+++ b/llvm/test/CodeGen/X86/swap.ll
@@ -113,21 +113,17 @@ define dso_local void @onealloc_readback_1(ptr nocapture %a, ptr nocapture %b) l
;
; AA-LABEL: onealloc_readback_1:
; AA: # %bb.0: # %entry
-; AA-NEXT: vmovups (%rdi), %xmm0
-; AA-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
; AA-NEXT: vmovups (%rsi), %xmm0
; AA-NEXT: vmovups %xmm0, (%rdi)
; AA-NEXT: retq
entry:
%alloc = alloca [16 x i8], i8 2, align 1
%part1 = getelementptr inbounds [16 x i8], ptr %alloc, i64 1, i64 0
- call void @llvm.lifetime.start.p0(i64 16, ptr nonnull %part1)
- call void @llvm.lifetime.start.p0(i64 16, ptr nonnull %alloc)
+ call void @llvm.lifetime.start.p0(i64 32, ptr nonnull %alloc)
call void @llvm.memcpy.p0.p0.i64(ptr nonnull align 1 %part1, ptr align 1 %a, i64 16, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr nonnull align 1 %alloc, ptr align 1 %b, i64 16, i1 false)
- call void @llvm.lifetime.end.p0(i64 16, ptr nonnull %part1)
tail call void @llvm.memcpy.p0.p0.i64(ptr nonnull align 1 %a, ptr align 1 %alloc, i64 16, i1 false)
- call void @llvm.lifetime.end.p0(i64 16, ptr nonnull %alloc)
+ call void @llvm.lifetime.end.p0(i64 32, ptr nonnull %alloc)
ret void
}
@@ -144,19 +140,16 @@ define dso_local void @onealloc_readback_2(ptr nocapture %a, ptr nocapture %b) l
; AA-LABEL: onealloc_readback_2:
; AA: # %bb.0: # %entry
; AA-NEXT: vmovups (%rsi), %xmm0
-; AA-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
; AA-NEXT: vmovups %xmm0, (%rdi)
; AA-NEXT: retq
entry:
%alloc = alloca [16 x i8], i8 2, align 1
%part2 = getelementptr inbounds [16 x i8], ptr %alloc, i64 1, i64 0
- call void @llvm.lifetime.start.p0(i64 16, ptr nonnull %alloc)
- call void @llvm.lifetime.start.p0(i64 16, ptr nonnull %part2)
+ call void @llvm.lifetime.start.p0(i64 32, ptr nonnull %alloc)
call void @llvm.memcpy.p0.p0.i64(ptr nonnull align 1 %alloc, ptr align 1 %a, i64 16, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr nonnull align 1 %part2, ptr align 1 %b, i64 16, i1 false)
- call void @llvm.lifetime.end.p0(i64 16, ptr nonnull %alloc)
tail call void @llvm.memcpy.p0.p0.i64(ptr nonnull align 1 %a, ptr align 1 %part2, i64 16, i1 false)
- call void @llvm.lifetime.end.p0(i64 16, ptr nonnull %part2)
+ call void @llvm.lifetime.end.p0(i64 32, ptr nonnull %alloc)
ret void
}
generated by cgit v1.1 at 2025-07-29 18:42:37 +0000