maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/X86/cmp.ll

468 lines
15 KiB
LLVM
Raw Normal View History

[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -show-mc-encoding | FileCheck %s
Rename a test-not-cmp.ll to cmp-test.ll; add a test. llvm-svn: 30243
2006-09-11 10:33:29 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
@d = global i8 0, align 1
filecheckize. llvm-svn: 116050
2010-10-08 13:02:29 +08:00
define i32 @test1(i32 %X, i32* %y) nounwind {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test1:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: cmpl $0, (%rsi) # encoding: [0x83,0x3e,0x00]
; CHECK-NEXT: je .LBB0_2 # encoding: [0x74,A]
; CHECK-NEXT: # fixup A - offset: 1, value: .LBB0_2-1, kind: FK_PCRel_1
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK-NEXT: # %bb.1: # %cond_true
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
; CHECK-NEXT: retq # encoding: [0xc3]
; CHECK-NEXT: .LBB0_2: # %ReturnBlock
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%tmp = load i32, i32* %y
%tmp.upgrd.1 = icmp eq i32 %tmp, 0
br i1 %tmp.upgrd.1, label %ReturnBlock, label %cond_true
Rename a test-not-cmp.ll to cmp-test.ll; add a test. llvm-svn: 30243
2006-09-11 10:33:29 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
cond_true:
ret i32 1
Rename a test-not-cmp.ll to cmp-test.ll; add a test. llvm-svn: 30243
2006-09-11 10:33:29 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
ReturnBlock:
ret i32 0
Rename a test-not-cmp.ll to cmp-test.ll; add a test. llvm-svn: 30243
2006-09-11 10:33:29 +08:00
}
filecheckize. llvm-svn: 116050
2010-10-08 13:02:29 +08:00
define i32 @test2(i32 %X, i32* %y) nounwind {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test2:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testl $536870911, (%rsi) # encoding: [0xf7,0x06,0xff,0xff,0xff,0x1f]
; CHECK-NEXT: # imm = 0x1FFFFFFF
; CHECK-NEXT: je .LBB1_2 # encoding: [0x74,A]
; CHECK-NEXT: # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK-NEXT: # %bb.1: # %cond_true
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
; CHECK-NEXT: retq # encoding: [0xc3]
; CHECK-NEXT: .LBB1_2: # %ReturnBlock
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%tmp = load i32, i32* %y
%tmp1 = shl i32 %tmp, 3
%tmp1.upgrd.2 = icmp eq i32 %tmp1, 0
br i1 %tmp1.upgrd.2, label %ReturnBlock, label %cond_true
Rename a test-not-cmp.ll to cmp-test.ll; add a test. llvm-svn: 30243
2006-09-11 10:33:29 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
cond_true:
ret i32 1
Rename a test-not-cmp.ll to cmp-test.ll; add a test. llvm-svn: 30243
2006-09-11 10:33:29 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
ReturnBlock:
ret i32 0
X86: Emit test instead of constant shift + compare if the shift result is unused. This allows us to compile return (mask & 0x8 ? a : b); into testb $8, %dil cmovnel %edx, %esi instead of andl $8, %edi shrl $3, %edi cmovnel %edx, %esi which we formed previously because dag combiner canonicalizes setcc of and into shift. llvm-svn: 207088
2014-04-24 16:15:31 +08:00
}
define i8 @test2b(i8 %X, i8* %y) nounwind {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test2b:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testb $31, (%rsi) # encoding: [0xf6,0x06,0x1f]
; CHECK-NEXT: je .LBB2_2 # encoding: [0x74,A]
; CHECK-NEXT: # fixup A - offset: 1, value: .LBB2_2-1, kind: FK_PCRel_1
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK-NEXT: # %bb.1: # %cond_true
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: movb $1, %al # encoding: [0xb0,0x01]
; CHECK-NEXT: retq # encoding: [0xc3]
; CHECK-NEXT: .LBB2_2: # %ReturnBlock
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%tmp = load i8, i8* %y
%tmp1 = shl i8 %tmp, 3
%tmp1.upgrd.2 = icmp eq i8 %tmp1, 0
br i1 %tmp1.upgrd.2, label %ReturnBlock, label %cond_true
X86: Emit test instead of constant shift + compare if the shift result is unused. This allows us to compile return (mask & 0x8 ? a : b); into testb $8, %dil cmovnel %edx, %esi instead of andl $8, %edi shrl $3, %edi cmovnel %edx, %esi which we formed previously because dag combiner canonicalizes setcc of and into shift. llvm-svn: 207088
2014-04-24 16:15:31 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
cond_true:
ret i8 1
X86: Emit test instead of constant shift + compare if the shift result is unused. This allows us to compile return (mask & 0x8 ? a : b); into testb $8, %dil cmovnel %edx, %esi instead of andl $8, %edi shrl $3, %edi cmovnel %edx, %esi which we formed previously because dag combiner canonicalizes setcc of and into shift. llvm-svn: 207088
2014-04-24 16:15:31 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
ReturnBlock:
ret i8 0
Rename a test-not-cmp.ll to cmp-test.ll; add a test. llvm-svn: 30243
2006-09-11 10:33:29 +08:00
}
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
define i64 @test3(i64 %x) nounwind {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test3:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
; CHECK-NEXT: sete %al # encoding: [0x0f,0x94,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
%t = icmp eq i64 %x, 0
%r = zext i1 %t to i64
ret i64 %r
}
define i64 @test4(i64 %x) nounwind {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test4:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
; CHECK-NEXT: setle %al # encoding: [0x0f,0x9e,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
%t = icmp slt i64 %x, 1
%r = zext i1 %t to i64
ret i64 %r
}
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
define i32 @test5(double %A) nounwind {
; CHECK-LABEL: test5:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: ucomisd {{.*}}(%rip), %xmm0 # encoding: [0x66,0x0f,0x2e,0x05,A,A,A,A]
; CHECK-NEXT: # fixup A - offset: 4, value: {{\.LCPI.*}}-4, kind: reloc_riprel_4byte
; CHECK-NEXT: ja .LBB5_3 # encoding: [0x77,A]
; CHECK-NEXT: # fixup A - offset: 1, value: .LBB5_3-1, kind: FK_PCRel_1
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK-NEXT: # %bb.1: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: ucomisd {{.*}}(%rip), %xmm0 # encoding: [0x66,0x0f,0x2e,0x05,A,A,A,A]
; CHECK-NEXT: # fixup A - offset: 4, value: {{\.LCPI.*}}-4, kind: reloc_riprel_4byte
; CHECK-NEXT: jb .LBB5_3 # encoding: [0x72,A]
; CHECK-NEXT: # fixup A - offset: 1, value: .LBB5_3-1, kind: FK_PCRel_1
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK-NEXT: # %bb.2: # %bb12
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: movl $32, %eax # encoding: [0xb8,0x20,0x00,0x00,0x00]
; CHECK-NEXT: retq # encoding: [0xc3]
; CHECK-NEXT: .LBB5_3: # %bb8
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: jmp foo # TAILCALL
; CHECK-NEXT: # encoding: [0xeb,A]
; CHECK-NEXT: # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%tmp2 = fcmp ogt double %A, 1.500000e+02
%tmp5 = fcmp ult double %A, 7.500000e+01
%bothcond = or i1 %tmp2, %tmp5
br i1 %bothcond, label %bb8, label %bb12
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
bb8:
%tmp9 = tail call i32 (...) @foo() nounwind
ret i32 %tmp9
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
bb12:
ret i32 32
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
}
declare i32 @foo(...)
testcase that goes with r116053 llvm-svn: 116054
2010-10-08 13:12:30 +08:00
define i32 @test6() nounwind align 2 {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test6:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: cmpq $0, -{{[0-9]+}}(%rsp) # encoding: [0x48,0x83,0x7c,0x24,0xf8,0x00]
; CHECK-NEXT: je .LBB6_1 # encoding: [0x74,A]
; CHECK-NEXT: # fixup A - offset: 1, value: .LBB6_1-1, kind: FK_PCRel_1
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK-NEXT: # %bb.2: # %F
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
; CHECK-NEXT: .LBB6_1: # %T
; CHECK-NEXT: movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%A = alloca { i64, i64 }, align 8
%B = getelementptr inbounds { i64, i64 }, { i64, i64 }* %A, i64 0, i32 1
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%C = load i64, i64* %B
testcase that goes with r116053 llvm-svn: 116054
2010-10-08 13:12:30 +08:00
%D = icmp eq i64 %C, 0
br i1 %D, label %T, label %F
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
testcase that goes with r116053 llvm-svn: 116054
2010-10-08 13:12:30 +08:00
T:
ret i32 1
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
testcase that goes with r116053 llvm-svn: 116054
2010-10-08 13:12:30 +08:00
F:
ret i32 0
}
This is another case where instcombine demanded bits optimization created large immediates. Add dag combine logic to recover in case the large immediates doesn't fit in cmp immediate operand field. int foo(unsigned long l) { return (l>> 47) == 1; } we produce %shr.mask = and i64 %l, -140737488355328 %cmp = icmp eq i64 %shr.mask, 140737488355328 %conv = zext i1 %cmp to i32 ret i32 %conv which codegens to movq $0xffff800000000000,%rax andq %rdi,%rax movq $0x0000800000000000,%rcx cmpq %rcx,%rax sete %al movzbl %al,%eax ret TargetLowering::SimplifySetCC would transform (X & -256) == 256 -> (X >> 8) == 1 if the immediate fails the isLegalICmpImmediate() test. For x86, that's immediates which are not a signed 32-bit immediate. Based on a patch by Eli Friedman. PR10328 rdar://9758774 llvm-svn: 160346
2012-07-17 14:53:39 +08:00
define i32 @test7(i64 %res) nounwind {
Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. This was done with the following sed invocation to catch label lines demarking function boundaries: sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct. llvm-svn: 186258
2013-07-14 04:38:47 +08:00
; CHECK-LABEL: test7:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
; CHECK-NEXT: sete %al # encoding: [0x0f,0x94,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
For something like uint32_t hi(uint64_t res) { uint_32t hi = res >> 32; return !hi; } llvm IR looks like this: define i32 @hi(i64 %res) nounwind uwtable ssp { entry: %lnot = icmp ult i64 %res, 4294967296 %lnot.ext = zext i1 %lnot to i32 ret i32 %lnot.ext } The optimizer has optimize away the right shift and truncate but the resulting constant is too large to fit in the 32-bit immediate field. The resulting x86 code is worse as a result: movabsq $4294967296, %rax ## imm = 0x100000000 cmpq %rax, %rdi sbbl %eax, %eax andl $1, %eax This patch teaches the x86 lowering code to handle ult against a large immediate with trailing zeros. It will issue a right shift and a truncate followed by a comparison against a shifted immediate. shrq $32, %rdi testl %edi, %edi sete %al movzbl %al, %eax It also handles a ugt comparison against a large immediate with trailing bits set. i.e. X > 0x0ffffffff -> (X >> 32) >= 1 rdar://11866926 llvm-svn: 160312
2012-07-17 03:35:43 +08:00
%lnot = icmp ult i64 %res, 4294967296
%lnot.ext = zext i1 %lnot to i32
ret i32 %lnot.ext
}
This is another case where instcombine demanded bits optimization created large immediates. Add dag combine logic to recover in case the large immediates doesn't fit in cmp immediate operand field. int foo(unsigned long l) { return (l>> 47) == 1; } we produce %shr.mask = and i64 %l, -140737488355328 %cmp = icmp eq i64 %shr.mask, 140737488355328 %conv = zext i1 %cmp to i32 ret i32 %conv which codegens to movq $0xffff800000000000,%rax andq %rdi,%rax movq $0x0000800000000000,%rcx cmpq %rcx,%rax sete %al movzbl %al,%eax ret TargetLowering::SimplifySetCC would transform (X & -256) == 256 -> (X >> 8) == 1 if the immediate fails the isLegalICmpImmediate() test. For x86, that's immediates which are not a signed 32-bit immediate. Based on a patch by Eli Friedman. PR10328 rdar://9758774 llvm-svn: 160346
2012-07-17 14:53:39 +08:00
define i32 @test8(i64 %res) nounwind {
Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. This was done with the following sed invocation to catch label lines demarking function boundaries: sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct. llvm-svn: 186258
2013-07-14 04:38:47 +08:00
; CHECK-LABEL: test8:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: cmpq $3, %rdi # encoding: [0x48,0x83,0xff,0x03]
; CHECK-NEXT: setb %al # encoding: [0x0f,0x92,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
For something like uint32_t hi(uint64_t res) { uint_32t hi = res >> 32; return !hi; } llvm IR looks like this: define i32 @hi(i64 %res) nounwind uwtable ssp { entry: %lnot = icmp ult i64 %res, 4294967296 %lnot.ext = zext i1 %lnot to i32 ret i32 %lnot.ext } The optimizer has optimize away the right shift and truncate but the resulting constant is too large to fit in the 32-bit immediate field. The resulting x86 code is worse as a result: movabsq $4294967296, %rax ## imm = 0x100000000 cmpq %rax, %rdi sbbl %eax, %eax andl $1, %eax This patch teaches the x86 lowering code to handle ult against a large immediate with trailing zeros. It will issue a right shift and a truncate followed by a comparison against a shifted immediate. shrq $32, %rdi testl %edi, %edi sete %al movzbl %al, %eax It also handles a ugt comparison against a large immediate with trailing bits set. i.e. X > 0x0ffffffff -> (X >> 32) >= 1 rdar://11866926 llvm-svn: 160312
2012-07-17 03:35:43 +08:00
%lnot = icmp ult i64 %res, 12884901888
%lnot.ext = zext i1 %lnot to i32
ret i32 %lnot.ext
}
This is another case where instcombine demanded bits optimization created large immediates. Add dag combine logic to recover in case the large immediates doesn't fit in cmp immediate operand field. int foo(unsigned long l) { return (l>> 47) == 1; } we produce %shr.mask = and i64 %l, -140737488355328 %cmp = icmp eq i64 %shr.mask, 140737488355328 %conv = zext i1 %cmp to i32 ret i32 %conv which codegens to movq $0xffff800000000000,%rax andq %rdi,%rax movq $0x0000800000000000,%rcx cmpq %rcx,%rax sete %al movzbl %al,%eax ret TargetLowering::SimplifySetCC would transform (X & -256) == 256 -> (X >> 8) == 1 if the immediate fails the isLegalICmpImmediate() test. For x86, that's immediates which are not a signed 32-bit immediate. Based on a patch by Eli Friedman. PR10328 rdar://9758774 llvm-svn: 160346
2012-07-17 14:53:39 +08:00
define i32 @test9(i64 %res) nounwind {
Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. This was done with the following sed invocation to catch label lines demarking function boundaries: sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct. llvm-svn: 186258
2013-07-14 04:38:47 +08:00
; CHECK-LABEL: test9:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: shrq $33, %rdi # encoding: [0x48,0xc1,0xef,0x21]
; CHECK-NEXT: sete %al # encoding: [0x0f,0x94,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
For something like uint32_t hi(uint64_t res) { uint_32t hi = res >> 32; return !hi; } llvm IR looks like this: define i32 @hi(i64 %res) nounwind uwtable ssp { entry: %lnot = icmp ult i64 %res, 4294967296 %lnot.ext = zext i1 %lnot to i32 ret i32 %lnot.ext } The optimizer has optimize away the right shift and truncate but the resulting constant is too large to fit in the 32-bit immediate field. The resulting x86 code is worse as a result: movabsq $4294967296, %rax ## imm = 0x100000000 cmpq %rax, %rdi sbbl %eax, %eax andl $1, %eax This patch teaches the x86 lowering code to handle ult against a large immediate with trailing zeros. It will issue a right shift and a truncate followed by a comparison against a shifted immediate. shrq $32, %rdi testl %edi, %edi sete %al movzbl %al, %eax It also handles a ugt comparison against a large immediate with trailing bits set. i.e. X > 0x0ffffffff -> (X >> 32) >= 1 rdar://11866926 llvm-svn: 160312
2012-07-17 03:35:43 +08:00
%lnot = icmp ult i64 %res, 8589934592
%lnot.ext = zext i1 %lnot to i32
ret i32 %lnot.ext
}
This is another case where instcombine demanded bits optimization created large immediates. Add dag combine logic to recover in case the large immediates doesn't fit in cmp immediate operand field. int foo(unsigned long l) { return (l>> 47) == 1; } we produce %shr.mask = and i64 %l, -140737488355328 %cmp = icmp eq i64 %shr.mask, 140737488355328 %conv = zext i1 %cmp to i32 ret i32 %conv which codegens to movq $0xffff800000000000,%rax andq %rdi,%rax movq $0x0000800000000000,%rcx cmpq %rcx,%rax sete %al movzbl %al,%eax ret TargetLowering::SimplifySetCC would transform (X & -256) == 256 -> (X >> 8) == 1 if the immediate fails the isLegalICmpImmediate() test. For x86, that's immediates which are not a signed 32-bit immediate. Based on a patch by Eli Friedman. PR10328 rdar://9758774 llvm-svn: 160346
2012-07-17 14:53:39 +08:00
define i32 @test10(i64 %res) nounwind {
Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. This was done with the following sed invocation to catch label lines demarking function boundaries: sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct. llvm-svn: 186258
2013-07-14 04:38:47 +08:00
; CHECK-LABEL: test10:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
; CHECK-NEXT: setne %al # encoding: [0x0f,0x95,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
For something like uint32_t hi(uint64_t res) { uint_32t hi = res >> 32; return !hi; } llvm IR looks like this: define i32 @hi(i64 %res) nounwind uwtable ssp { entry: %lnot = icmp ult i64 %res, 4294967296 %lnot.ext = zext i1 %lnot to i32 ret i32 %lnot.ext } The optimizer has optimize away the right shift and truncate but the resulting constant is too large to fit in the 32-bit immediate field. The resulting x86 code is worse as a result: movabsq $4294967296, %rax ## imm = 0x100000000 cmpq %rax, %rdi sbbl %eax, %eax andl $1, %eax This patch teaches the x86 lowering code to handle ult against a large immediate with trailing zeros. It will issue a right shift and a truncate followed by a comparison against a shifted immediate. shrq $32, %rdi testl %edi, %edi sete %al movzbl %al, %eax It also handles a ugt comparison against a large immediate with trailing bits set. i.e. X > 0x0ffffffff -> (X >> 32) >= 1 rdar://11866926 llvm-svn: 160312
2012-07-17 03:35:43 +08:00
%lnot = icmp uge i64 %res, 4294967296
%lnot.ext = zext i1 %lnot to i32
ret i32 %lnot.ext
}
This is another case where instcombine demanded bits optimization created large immediates. Add dag combine logic to recover in case the large immediates doesn't fit in cmp immediate operand field. int foo(unsigned long l) { return (l>> 47) == 1; } we produce %shr.mask = and i64 %l, -140737488355328 %cmp = icmp eq i64 %shr.mask, 140737488355328 %conv = zext i1 %cmp to i32 ret i32 %conv which codegens to movq $0xffff800000000000,%rax andq %rdi,%rax movq $0x0000800000000000,%rcx cmpq %rcx,%rax sete %al movzbl %al,%eax ret TargetLowering::SimplifySetCC would transform (X & -256) == 256 -> (X >> 8) == 1 if the immediate fails the isLegalICmpImmediate() test. For x86, that's immediates which are not a signed 32-bit immediate. Based on a patch by Eli Friedman. PR10328 rdar://9758774 llvm-svn: 160346
2012-07-17 14:53:39 +08:00
define i32 @test11(i64 %l) nounwind {
Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. This was done with the following sed invocation to catch label lines demarking function boundaries: sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct. llvm-svn: 186258
2013-07-14 04:38:47 +08:00
; CHECK-LABEL: test11:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: shrq $47, %rdi # encoding: [0x48,0xc1,0xef,0x2f]
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: cmpq $1, %rdi # encoding: [0x48,0x83,0xff,0x01]
; CHECK-NEXT: sete %al # encoding: [0x0f,0x94,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
This is another case where instcombine demanded bits optimization created large immediates. Add dag combine logic to recover in case the large immediates doesn't fit in cmp immediate operand field. int foo(unsigned long l) { return (l>> 47) == 1; } we produce %shr.mask = and i64 %l, -140737488355328 %cmp = icmp eq i64 %shr.mask, 140737488355328 %conv = zext i1 %cmp to i32 ret i32 %conv which codegens to movq $0xffff800000000000,%rax andq %rdi,%rax movq $0x0000800000000000,%rcx cmpq %rcx,%rax sete %al movzbl %al,%eax ret TargetLowering::SimplifySetCC would transform (X & -256) == 256 -> (X >> 8) == 1 if the immediate fails the isLegalICmpImmediate() test. For x86, that's immediates which are not a signed 32-bit immediate. Based on a patch by Eli Friedman. PR10328 rdar://9758774 llvm-svn: 160346
2012-07-17 14:53:39 +08:00
%shr.mask = and i64 %l, -140737488355328
%cmp = icmp eq i64 %shr.mask, 140737488355328
%conv = zext i1 %cmp to i32
ret i32 %conv
}
Teach SimplifySetCC that comparing AssertZext i1 against a constant 1 can be rewritten as a compare against a constant 0 with the opposite condition. llvm-svn: 170495
2012-12-19 14:12:28 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
define i32 @test12() ssp uwtable {
Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. This was done with the following sed invocation to catch label lines demarking function boundaries: sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct. llvm-svn: 186258
2013-07-14 04:38:47 +08:00
; CHECK-LABEL: test12:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: pushq %rax # encoding: [0x50]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-NEXT: .cfi_def_cfa_offset 16
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: callq test12b # encoding: [0xe8,A,A,A,A]
[llvm-mc] - Produce R_X86_64_PLT32 for "call/jmp foo". For instructions like call foo and jmp foo patch changes relocation produced from R_X86_64_PC32 to R_X86_64_PLT32. Relocation can be used as a marker for 32-bit PC-relative branches. Linker will reduce PLT32 relocation to PC32 if function is defined locally. Differential revision: https://reviews.llvm.org/D43383 llvm-svn: 325569
2018-02-20 18:17:57 +08:00
; CHECK-NEXT: # fixup A - offset: 1, value: test12b-4, kind: reloc_branch_4byte_pcrel
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testb %al, %al # encoding: [0x84,0xc0]
; CHECK-NEXT: je .LBB12_2 # encoding: [0x74,A]
; CHECK-NEXT: # fixup A - offset: 1, value: .LBB12_2-1, kind: FK_PCRel_1
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK-NEXT: # %bb.1: # %T
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
; CHECK-NEXT: popq %rcx # encoding: [0x59]
Correct dwarf unwind information in function epilogue This patch aims to provide correct dwarf unwind information in function epilogue for X86. It consists of two parts. The first part inserts CFI instructions that set appropriate cfa offset and cfa register in emitEpilogue() in X86FrameLowering. This part is X86 specific. The second part is platform independent and ensures that: * CFI instructions do not affect code generation (they are not counted as instructions when tail duplicating or tail merging) * Unwind information remains correct when a function is modified by different passes. This is done in a late pass by analyzing information about cfa offset and cfa register in BBs and inserting additional CFI directives where necessary. Added CFIInstrInserter pass: * analyzes each basic block to determine cfa offset and register are valid at its entry and exit * verifies that outgoing cfa offset and register of predecessor blocks match incoming values of their successors * inserts additional CFI directives at basic block beginning to correct the rule for calculating CFA Having CFI instructions in function epilogue can cause incorrect CFA calculation rule for some basic blocks. This can happen if, due to basic block reordering, or the existence of multiple epilogue blocks, some of the blocks have wrong cfa offset and register values set by the epilogue block above them. CFIInstrInserter is currently run only on X86, but can be used by any target that implements support for adding CFI instructions in epilogue. Patch by Violeta Vukobrat. Differential Revision: https://reviews.llvm.org/D42848 llvm-svn: 330706
2018-04-24 18:32:08 +08:00
; CHECK-NEXT: .cfi_def_cfa_offset 8
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: retq # encoding: [0xc3]
; CHECK-NEXT: .LBB12_2: # %F
Correct dwarf unwind information in function epilogue This patch aims to provide correct dwarf unwind information in function epilogue for X86. It consists of two parts. The first part inserts CFI instructions that set appropriate cfa offset and cfa register in emitEpilogue() in X86FrameLowering. This part is X86 specific. The second part is platform independent and ensures that: * CFI instructions do not affect code generation (they are not counted as instructions when tail duplicating or tail merging) * Unwind information remains correct when a function is modified by different passes. This is done in a late pass by analyzing information about cfa offset and cfa register in BBs and inserting additional CFI directives where necessary. Added CFIInstrInserter pass: * analyzes each basic block to determine cfa offset and register are valid at its entry and exit * verifies that outgoing cfa offset and register of predecessor blocks match incoming values of their successors * inserts additional CFI directives at basic block beginning to correct the rule for calculating CFA Having CFI instructions in function epilogue can cause incorrect CFA calculation rule for some basic blocks. This can happen if, due to basic block reordering, or the existence of multiple epilogue blocks, some of the blocks have wrong cfa offset and register values set by the epilogue block above them. CFIInstrInserter is currently run only on X86, but can be used by any target that implements support for adding CFI instructions in epilogue. Patch by Violeta Vukobrat. Differential Revision: https://reviews.llvm.org/D42848 llvm-svn: 330706
2018-04-24 18:32:08 +08:00
; CHECK-NEXT: .cfi_def_cfa_offset 16
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: movl $2, %eax # encoding: [0xb8,0x02,0x00,0x00,0x00]
; CHECK-NEXT: popq %rcx # encoding: [0x59]
Correct dwarf unwind information in function epilogue This patch aims to provide correct dwarf unwind information in function epilogue for X86. It consists of two parts. The first part inserts CFI instructions that set appropriate cfa offset and cfa register in emitEpilogue() in X86FrameLowering. This part is X86 specific. The second part is platform independent and ensures that: * CFI instructions do not affect code generation (they are not counted as instructions when tail duplicating or tail merging) * Unwind information remains correct when a function is modified by different passes. This is done in a late pass by analyzing information about cfa offset and cfa register in BBs and inserting additional CFI directives where necessary. Added CFIInstrInserter pass: * analyzes each basic block to determine cfa offset and register are valid at its entry and exit * verifies that outgoing cfa offset and register of predecessor blocks match incoming values of their successors * inserts additional CFI directives at basic block beginning to correct the rule for calculating CFA Having CFI instructions in function epilogue can cause incorrect CFA calculation rule for some basic blocks. This can happen if, due to basic block reordering, or the existence of multiple epilogue blocks, some of the blocks have wrong cfa offset and register values set by the epilogue block above them. CFIInstrInserter is currently run only on X86, but can be used by any target that implements support for adding CFI instructions in epilogue. Patch by Violeta Vukobrat. Differential Revision: https://reviews.llvm.org/D42848 llvm-svn: 330706
2018-04-24 18:32:08 +08:00
; CHECK-NEXT: .cfi_def_cfa_offset 8
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%tmp1 = call zeroext i1 @test12b()
br i1 %tmp1, label %T, label %F
Teach SimplifySetCC that comparing AssertZext i1 against a constant 1 can be rewritten as a compare against a constant 0 with the opposite condition. llvm-svn: 170495
2012-12-19 14:12:28 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
T:
Teach SimplifySetCC that comparing AssertZext i1 against a constant 1 can be rewritten as a compare against a constant 0 with the opposite condition. llvm-svn: 170495
2012-12-19 14:12:28 +08:00
ret i32 1
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
F:
Teach SimplifySetCC that comparing AssertZext i1 against a constant 1 can be rewritten as a compare against a constant 0 with the opposite condition. llvm-svn: 170495
2012-12-19 14:12:28 +08:00
ret i32 2
}
declare zeroext i1 @test12b()
X86: Emit test instead of constant shift + compare if the shift result is unused. This allows us to compile return (mask & 0x8 ? a : b); into testb $8, %dil cmovnel %edx, %esi instead of andl $8, %edi shrl $3, %edi cmovnel %edx, %esi which we formed previously because dag combiner canonicalizes setcc of and into shift. llvm-svn: 207088
2014-04-24 16:15:31 +08:00
define i32 @test13(i32 %mask, i32 %base, i32 %intra) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test13:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testb $8, %dil # encoding: [0x40,0xf6,0xc7,0x08]
; CHECK-NEXT: cmovnel %edx, %esi # encoding: [0x0f,0x45,0xf2]
; CHECK-NEXT: movl %esi, %eax # encoding: [0x89,0xf0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
X86: Emit test instead of constant shift + compare if the shift result is unused. This allows us to compile return (mask & 0x8 ? a : b); into testb $8, %dil cmovnel %edx, %esi instead of andl $8, %edi shrl $3, %edi cmovnel %edx, %esi which we formed previously because dag combiner canonicalizes setcc of and into shift. llvm-svn: 207088
2014-04-24 16:15:31 +08:00
%and = and i32 %mask, 8
%tobool = icmp ne i32 %and, 0
%cond = select i1 %tobool, i32 %intra, i32 %base
ret i32 %cond
}
X86: Don't transform shifts into ands when the sign bit is tested. Should unbreak MultiSource/Benchmarks/mediabench/g721/g721encode/encode. llvm-svn: 207145
2014-04-25 04:51:37 +08:00
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
define i32 @test14(i32 %mask, i32 %base, i32 %intra) {
; CHECK-LABEL: test14:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: shrl $7, %edi # encoding: [0xc1,0xef,0x07]
; CHECK-NEXT: cmovnsl %edx, %esi # encoding: [0x0f,0x49,0xf2]
; CHECK-NEXT: movl %esi, %eax # encoding: [0x89,0xf0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
X86: Don't transform shifts into ands when the sign bit is tested. Should unbreak MultiSource/Benchmarks/mediabench/g721/g721encode/encode. llvm-svn: 207145
2014-04-25 04:51:37 +08:00
%s = lshr i32 %mask, 7
%tobool = icmp sgt i32 %s, -1
%cond = select i1 %tobool, i32 %intra, i32 %base
ret i32 %cond
}
X86: Don't turn shifts into ands if there's another use that may not check for equality. Fixes PR19964. llvm-svn: 210371
2014-06-07 05:08:55 +08:00
; PR19964
define zeroext i1 @test15(i32 %bf.load, i32 %n) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test15:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: shrl $16, %edi # encoding: [0xc1,0xef,0x10]
; CHECK-NEXT: sete %cl # encoding: [0x0f,0x94,0xc1]
; CHECK-NEXT: cmpl %esi, %edi # encoding: [0x39,0xf7]
; CHECK-NEXT: setae %al # encoding: [0x0f,0x93,0xc0]
; CHECK-NEXT: orb %cl, %al # encoding: [0x08,0xc8]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
X86: Don't turn shifts into ands if there's another use that may not check for equality. Fixes PR19964. llvm-svn: 210371
2014-06-07 05:08:55 +08:00
%bf.lshr = lshr i32 %bf.load, 16
%cmp2 = icmp eq i32 %bf.lshr, 0
%cmp5 = icmp uge i32 %bf.lshr, %n
%.cmp5 = or i1 %cmp2, %cmp5
ret i1 %.cmp5
}
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
define i8 @test16(i16 signext %L) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test16:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testw %di, %di # encoding: [0x66,0x85,0xff]
; CHECK-NEXT: setns %al # encoding: [0x0f,0x99,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%lshr = lshr i16 %L, 15
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
%trunc = trunc i16 %lshr to i8
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
%not = xor i8 %trunc, 1
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
ret i8 %not
}
define i8 @test17(i32 %L) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test17:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testl %edi, %edi # encoding: [0x85,0xff]
; CHECK-NEXT: setns %al # encoding: [0x0f,0x99,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%lshr = lshr i32 %L, 31
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
%trunc = trunc i32 %lshr to i8
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
%not = xor i8 %trunc, 1
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
ret i8 %not
}
define i8 @test18(i64 %L) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test18:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
; CHECK-NEXT: setns %al # encoding: [0x0f,0x99,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%lshr = lshr i64 %L, 63
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
%trunc = trunc i64 %lshr to i8
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
%not = xor i8 %trunc, 1
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
ret i8 %not
}
define zeroext i1 @test19(i32 %L) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test19:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: testl %edi, %edi # encoding: [0x85,0xff]
; CHECK-NEXT: setns %al # encoding: [0x0f,0x99,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%lshr = lshr i32 %L, 31
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
%trunc = trunc i32 %lshr to i1
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
%not = xor i1 %trunc, true
[X86] Emit more efficient >= comparisons against 0 We don't do a great job with >= 0 comparisons against zero when the result is used as an i8. Given something like: void f(long long LL, bool *B) { *B = LL >= 0; } We used to generate: shrq $63, %rdi xorb $1, %dil movb %dil, (%rsi) Now we generate: testq %rdi, %rdi setns (%rsi) Differential Revision: http://reviews.llvm.org/D12136 llvm-svn: 245498
2015-08-20 04:51:40 +08:00
ret i1 %not
}
[X86 Target Lowering] Merged ICMP test. llvm-svn: 273995
2016-06-28 14:25:38 +08:00
; This test failed due to incorrect handling of "shift + icmp" sequence
define void @test20(i32 %bf.load, i8 %x1, i8* %b_addr) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test20:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: testl $16777215, %edi # encoding: [0xf7,0xc7,0xff,0xff,0xff,0x00]
; CHECK-NEXT: # imm = 0xFFFFFF
; CHECK-NEXT: setne %al # encoding: [0x0f,0x95,0xc0]
; CHECK-NEXT: movzbl %sil, %ecx # encoding: [0x40,0x0f,0xb6,0xce]
; CHECK-NEXT: addl %eax, %ecx # encoding: [0x01,0xc1]
; CHECK-NEXT: setne (%rdx) # encoding: [0x0f,0x95,0x02]
; CHECK-NEXT: testl $16777215, %edi # encoding: [0xf7,0xc7,0xff,0xff,0xff,0x00]
; CHECK-NEXT: # imm = 0xFFFFFF
; CHECK-NEXT: setne {{.*}}(%rip) # encoding: [0x0f,0x95,0x05,A,A,A,A]
; CHECK-NEXT: # fixup A - offset: 3, value: d-4, kind: reloc_riprel_4byte
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
[X86 Target Lowering] Merged ICMP test. llvm-svn: 273995
2016-06-28 14:25:38 +08:00
%bf.shl = shl i32 %bf.load, 8
%bf.ashr = ashr exact i32 %bf.shl, 8
%tobool4 = icmp ne i32 %bf.ashr, 0
%conv = zext i1 %tobool4 to i32
%conv6 = zext i8 %x1 to i32
%add = add nuw nsw i32 %conv, %conv6
%tobool7 = icmp ne i32 %add, 0
%frombool = zext i1 %tobool7 to i8
store i8 %frombool, i8* %b_addr, align 1
%tobool14 = icmp ne i32 %bf.shl, 0
%frombool15 = zext i1 %tobool14 to i8
store i8 %frombool15, i8* @d, align 1
ret void
[X86] Replace AND+IMM64 with SRL/SHL Emit SHRQ/SHLQ instead of ANDQ with a 64 bit constant mask if the result is unused and the mask has only higher/lower bits set. For example, with this patch LLVM emits shrq $41, %rdi je instead of movabsq $0xFFFFFE0000000000, %rcx testq %rcx, %rdi je This reduces number of instructions, code size and register pressure. The transformation is applied only for cases where the mask cannot be encoded as an immediate value within TESTQ instruction. Differential Revision: https://reviews.llvm.org/D28198 llvm-svn: 291806
2017-01-13 03:54:27 +08:00
}
define i32 @test21(i64 %val) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test21:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: shrq $41, %rdi # encoding: [0x48,0xc1,0xef,0x29]
; CHECK-NEXT: setne %al # encoding: [0x0f,0x95,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%and = and i64 %val, -2199023255552
[X86] Replace AND+IMM64 with SRL/SHL Emit SHRQ/SHLQ instead of ANDQ with a 64 bit constant mask if the result is unused and the mask has only higher/lower bits set. For example, with this patch LLVM emits shrq $41, %rdi je instead of movabsq $0xFFFFFE0000000000, %rcx testq %rcx, %rdi je This reduces number of instructions, code size and register pressure. The transformation is applied only for cases where the mask cannot be encoded as an immediate value within TESTQ instruction. Differential Revision: https://reviews.llvm.org/D28198 llvm-svn: 291806
2017-01-13 03:54:27 +08:00
%cmp = icmp ne i64 %and, 0
%ret = zext i1 %cmp to i32
ret i32 %ret
}
; AND-to-SHR transformation is enabled for eq/ne condition codes only.
define i32 @test22(i64 %val) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test22:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%and = and i64 %val, -2199023255552
[X86] Replace AND+IMM64 with SRL/SHL Emit SHRQ/SHLQ instead of ANDQ with a 64 bit constant mask if the result is unused and the mask has only higher/lower bits set. For example, with this patch LLVM emits shrq $41, %rdi je instead of movabsq $0xFFFFFE0000000000, %rcx testq %rcx, %rdi je This reduces number of instructions, code size and register pressure. The transformation is applied only for cases where the mask cannot be encoded as an immediate value within TESTQ instruction. Differential Revision: https://reviews.llvm.org/D28198 llvm-svn: 291806
2017-01-13 03:54:27 +08:00
%cmp = icmp ult i64 %and, 0
%ret = zext i1 %cmp to i32
ret i32 %ret
}
define i32 @test23(i64 %val) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test23:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: testq $-1048576, %rdi # encoding: [0x48,0xf7,0xc7,0x00,0x00,0xf0,0xff]
; CHECK-NEXT: # imm = 0xFFF00000
; CHECK-NEXT: setne %al # encoding: [0x0f,0x95,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%and = and i64 %val, -1048576
[X86] Replace AND+IMM64 with SRL/SHL Emit SHRQ/SHLQ instead of ANDQ with a 64 bit constant mask if the result is unused and the mask has only higher/lower bits set. For example, with this patch LLVM emits shrq $41, %rdi je instead of movabsq $0xFFFFFE0000000000, %rcx testq %rcx, %rdi je This reduces number of instructions, code size and register pressure. The transformation is applied only for cases where the mask cannot be encoded as an immediate value within TESTQ instruction. Differential Revision: https://reviews.llvm.org/D28198 llvm-svn: 291806
2017-01-13 03:54:27 +08:00
%cmp = icmp ne i64 %and, 0
%ret = zext i1 %cmp to i32
ret i32 %ret
}
define i32 @test24(i64 %val) {
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
; CHECK-LABEL: test24:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0: # %entry
[x86] remove unnecessary OS specifier from test llvm-svn: 313986
2017-09-22 22:38:57 +08:00
; CHECK-NEXT: xorl %eax, %eax # encoding: [0x31,0xc0]
; CHECK-NEXT: shlq $16, %rdi # encoding: [0x48,0xc1,0xe7,0x10]
; CHECK-NEXT: setne %al # encoding: [0x0f,0x95,0xc0]
; CHECK-NEXT: retq # encoding: [0xc3]
[x86] NFC: Clean up two tests and generate precise checks for them. Mostly this involved giving unnamed values names and running the IR through `opt` to re-format it but merging in any important comments in the original. I then deleted pointless comments and inlined the function attributes for ease of reading and editting. All of this is to make it much easier to see the instructions being generated here and evaluate any updates to the tests. llvm-svn: 311634
2017-08-24 15:38:36 +08:00
entry:
%and = and i64 %val, 281474976710655
[X86] Replace AND+IMM64 with SRL/SHL Emit SHRQ/SHLQ instead of ANDQ with a 64 bit constant mask if the result is unused and the mask has only higher/lower bits set. For example, with this patch LLVM emits shrq $41, %rdi je instead of movabsq $0xFFFFFE0000000000, %rcx testq %rcx, %rdi je This reduces number of instructions, code size and register pressure. The transformation is applied only for cases where the mask cannot be encoded as an immediate value within TESTQ instruction. Differential Revision: https://reviews.llvm.org/D28198 llvm-svn: 291806
2017-01-13 03:54:27 +08:00
%cmp = icmp ne i64 %and, 0
%ret = zext i1 %cmp to i32
ret i32 %ret
}