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

169 lines
3.6 KiB
LLVM
Raw Normal View History

testcase that goes with r116053 llvm-svn: 116054
2010-10-08 13:12:30 +08:00
; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -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
filecheckize. llvm-svn: 116050
2010-10-08 13:02:29 +08:00
define i32 @test1(i32 %X, i32* %y) nounwind {
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
%tmp = load i32* %y ; <i32> [#uses=1]
%tmp.upgrd.1 = icmp eq i32 %tmp, 0 ; <i1> [#uses=1]
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
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
cond_true: ; preds = %0
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
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
ReturnBlock: ; preds = %0
ret i32 0
filecheckize. llvm-svn: 116050
2010-10-08 13:02:29 +08:00
; CHECK: test1:
; CHECK: cmpl $0, (%rsi)
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 {
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
%tmp = load i32* %y ; <i32> [#uses=1]
%tmp1 = shl i32 %tmp, 3 ; <i32> [#uses=1]
%tmp1.upgrd.2 = icmp eq i32 %tmp1, 0 ; <i1> [#uses=1]
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
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
cond_true: ; preds = %0
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
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
ReturnBlock: ; preds = %0
ret i32 0
filecheckize. llvm-svn: 116050
2010-10-08 13:02:29 +08:00
; CHECK: test2:
; CHECK: movl (%rsi), %eax
; CHECK: shll $3, %eax
; CHECK: testl %eax, %eax
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 {
%t = icmp eq i64 %x, 0
%r = zext i1 %t to i64
ret i64 %r
; CHECK: test3:
; CHECK: testq %rdi, %rdi
; CHECK: sete %al
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias (movzx/movsx) because they give more information. Revert that part of the patch. llvm-svn: 129498
2011-04-14 09:46:37 +08:00
; CHECK: movzbl %al, %eax
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
; CHECK: ret
}
define i64 @test4(i64 %x) nounwind {
%t = icmp slt i64 %x, 1
%r = zext i1 %t to i64
ret i64 %r
; CHECK: test4:
; CHECK: testq %rdi, %rdi
; CHECK: setle %al
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias (movzx/movsx) because they give more information. Revert that part of the patch. llvm-svn: 129498
2011-04-14 09:46:37 +08:00
; CHECK: movzbl %al, %eax
merge tests llvm-svn: 116051
2010-10-08 13:04:58 +08:00
; CHECK: ret
}
define i32 @test5(double %A) nounwind {
entry:
%tmp2 = fcmp ogt double %A, 1.500000e+02; <i1> [#uses=1]
%tmp5 = fcmp ult double %A, 7.500000e+01; <i1> [#uses=1]
%bothcond = or i1 %tmp2, %tmp5; <i1> [#uses=1]
br i1 %bothcond, label %bb8, label %bb12
bb8:; preds = %entry
%tmp9 = tail call i32 (...)* @foo( ) nounwind ; <i32> [#uses=1]
ret i32 %tmp9
bb12:; preds = %entry
ret i32 32
; CHECK: test5:
; CHECK: ucomisd LCPI4_0(%rip), %xmm0
; CHECK: ucomisd LCPI4_1(%rip), %xmm0
}
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 {
%A = alloca {i64, i64}, align 8
%B = getelementptr inbounds {i64, i64}* %A, i64 0, i32 1
%C = load i64* %B
%D = icmp eq i64 %C, 0
br i1 %D, label %T, label %F
T:
ret i32 1
F:
ret i32 0
; CHECK: test6:
; CHECK: cmpq $0, -8(%rsp)
; CHECK: encoding: [0x48,0x83,0x7c,0x24,0xf8,0x00]
}
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
; rdar://11866926
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 {
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
entry:
; CHECK: test7:
; CHECK-NOT: movabsq
; CHECK: shrq $32, %rdi
Implement r160312 as target indepedenet dag combine. llvm-svn: 160354
2012-07-17 16:31:11 +08:00
; CHECK: testq %rdi, %rdi
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
; CHECK: sete
%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 {
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
entry:
; CHECK: test8:
; CHECK-NOT: movabsq
; CHECK: shrq $32, %rdi
Implement r160312 as target indepedenet dag combine. llvm-svn: 160354
2012-07-17 16:31:11 +08:00
; CHECK: cmpq $3, %rdi
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 {
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
entry:
; CHECK: test9:
; CHECK-NOT: movabsq
; CHECK: shrq $33, %rdi
Implement r160312 as target indepedenet dag combine. llvm-svn: 160354
2012-07-17 16:31:11 +08:00
; CHECK: testq %rdi, %rdi
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
; CHECK: sete
%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 {
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
entry:
; CHECK: test10:
; CHECK-NOT: movabsq
; CHECK: shrq $32, %rdi
Implement r160312 as target indepedenet dag combine. llvm-svn: 160354
2012-07-17 16:31:11 +08:00
; CHECK: testq %rdi, %rdi
; CHECK: setne
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
; rdar://9758774
define i32 @test11(i64 %l) nounwind {
entry:
; CHECK: test11:
; CHECK-NOT: movabsq
; CHECK-NOT: andq
; CHECK: shrq $47, %rdi
; CHECK: cmpq $1, %rdi
%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
define i32 @test12() uwtable ssp {
; CHECK: test12:
; CHECK: testb
%1 = call zeroext i1 @test12b()
br i1 %1, label %2, label %3
; <label>:2 ; preds = %0
ret i32 1
; <label>:3 ; preds = %0
ret i32 2
}
declare zeroext i1 @test12b()