forked from OSchip/llvm-project
653 lines
16 KiB
LLVM
653 lines
16 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+bmi | FileCheck %s --check-prefixes=CHECK,X86,BMI1,X86-BMI1
|
|
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+bmi,+bmi2 | FileCheck %s --check-prefixes=CHECK,X86,BMI2,X86-BMI2
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+bmi | FileCheck %s --check-prefixes=CHECK,X64,BMI1,X64-BMI1
|
|
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+bmi,+bmi2 | FileCheck %s --check-prefixes=CHECK,X64,BMI2,X64-BMI2
|
|
|
|
define i32 @andn32(i32 %x, i32 %y) {
|
|
; X86-LABEL: andn32:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: andn32:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnl %esi, %edi, %eax
|
|
; X64-NEXT: retq
|
|
%tmp1 = xor i32 %x, -1
|
|
%tmp2 = and i32 %y, %tmp1
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i32 @andn32_load(i32 %x, i32* %y) {
|
|
; X86-LABEL: andn32_load:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: andnl (%eax), %ecx, %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: andn32_load:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnl (%rsi), %edi, %eax
|
|
; X64-NEXT: retq
|
|
%y1 = load i32, i32* %y
|
|
%tmp1 = xor i32 %x, -1
|
|
%tmp2 = and i32 %y1, %tmp1
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i64 @andn64(i64 %x, i64 %y) {
|
|
; X86-LABEL: andn64:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %ecx, %edx
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: andn64:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnq %rsi, %rdi, %rax
|
|
; X64-NEXT: retq
|
|
%tmp1 = xor i64 %x, -1
|
|
%tmp2 = and i64 %tmp1, %y
|
|
ret i64 %tmp2
|
|
}
|
|
|
|
; Don't choose a 'test' if an 'andn' can be used.
|
|
define i1 @andn_cmp(i32 %x, i32 %y) {
|
|
; X86-LABEL: andn_cmp:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: sete %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: andn_cmp:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnl %esi, %edi, %eax
|
|
; X64-NEXT: sete %al
|
|
; X64-NEXT: retq
|
|
%notx = xor i32 %x, -1
|
|
%and = and i32 %notx, %y
|
|
%cmp = icmp eq i32 %and, 0
|
|
ret i1 %cmp
|
|
}
|
|
|
|
; Recognize a disguised andn in the following 4 tests.
|
|
define i1 @and_cmp1(i32 %x, i32 %y) {
|
|
; X86-LABEL: and_cmp1:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: sete %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: and_cmp1:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnl %esi, %edi, %eax
|
|
; X64-NEXT: sete %al
|
|
; X64-NEXT: retq
|
|
%and = and i32 %x, %y
|
|
%cmp = icmp eq i32 %and, %y
|
|
ret i1 %cmp
|
|
}
|
|
|
|
define i1 @and_cmp2(i32 %x, i32 %y) {
|
|
; X86-LABEL: and_cmp2:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: setne %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: and_cmp2:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnl %esi, %edi, %eax
|
|
; X64-NEXT: setne %al
|
|
; X64-NEXT: retq
|
|
%and = and i32 %y, %x
|
|
%cmp = icmp ne i32 %and, %y
|
|
ret i1 %cmp
|
|
}
|
|
|
|
define i1 @and_cmp3(i32 %x, i32 %y) {
|
|
; X86-LABEL: and_cmp3:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: sete %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: and_cmp3:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnl %esi, %edi, %eax
|
|
; X64-NEXT: sete %al
|
|
; X64-NEXT: retq
|
|
%and = and i32 %x, %y
|
|
%cmp = icmp eq i32 %y, %and
|
|
ret i1 %cmp
|
|
}
|
|
|
|
define i1 @and_cmp4(i32 %x, i32 %y) {
|
|
; X86-LABEL: and_cmp4:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: setne %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: and_cmp4:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnl %esi, %edi, %eax
|
|
; X64-NEXT: setne %al
|
|
; X64-NEXT: retq
|
|
%and = and i32 %y, %x
|
|
%cmp = icmp ne i32 %y, %and
|
|
ret i1 %cmp
|
|
}
|
|
|
|
; A mask and compare against constant is ok for an 'andn' too
|
|
; even though the BMI instruction doesn't have an immediate form.
|
|
define i1 @and_cmp_const(i32 %x) {
|
|
; X86-LABEL: and_cmp_const:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: notl %eax
|
|
; X86-NEXT: andl $43, %eax
|
|
; X86-NEXT: sete %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: and_cmp_const:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: notl %edi
|
|
; X64-NEXT: andl $43, %edi
|
|
; X64-NEXT: sete %al
|
|
; X64-NEXT: retq
|
|
%and = and i32 %x, 43
|
|
%cmp = icmp eq i32 %and, 43
|
|
ret i1 %cmp
|
|
}
|
|
|
|
; But don't use 'andn' if the mask is a power-of-two.
|
|
define i1 @and_cmp_const_power_of_two(i32 %x, i32 %y) {
|
|
; X86-LABEL: and_cmp_const_power_of_two:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: btl %ecx, %eax
|
|
; X86-NEXT: setae %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: and_cmp_const_power_of_two:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: btl %esi, %edi
|
|
; X64-NEXT: setae %al
|
|
; X64-NEXT: retq
|
|
%shl = shl i32 1, %y
|
|
%and = and i32 %x, %shl
|
|
%cmp = icmp ne i32 %and, %shl
|
|
ret i1 %cmp
|
|
}
|
|
|
|
; Don't transform to 'andn' if there's another use of the 'and'.
|
|
define i32 @and_cmp_not_one_use(i32 %x) {
|
|
; X86-LABEL: and_cmp_not_one_use:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: andl $37, %ecx
|
|
; X86-NEXT: xorl %eax, %eax
|
|
; X86-NEXT: cmpl $37, %ecx
|
|
; X86-NEXT: sete %al
|
|
; X86-NEXT: addl %ecx, %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: and_cmp_not_one_use:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andl $37, %edi
|
|
; X64-NEXT: xorl %eax, %eax
|
|
; X64-NEXT: cmpl $37, %edi
|
|
; X64-NEXT: sete %al
|
|
; X64-NEXT: addl %edi, %eax
|
|
; X64-NEXT: retq
|
|
%and = and i32 %x, 37
|
|
%cmp = icmp eq i32 %and, 37
|
|
%ext = zext i1 %cmp to i32
|
|
%add = add i32 %and, %ext
|
|
ret i32 %add
|
|
}
|
|
|
|
; Verify that we're not transforming invalid comparison predicates.
|
|
define i1 @not_an_andn1(i32 %x, i32 %y) {
|
|
; X86-LABEL: not_an_andn1:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: andl %eax, %ecx
|
|
; X86-NEXT: cmpl %ecx, %eax
|
|
; X86-NEXT: setg %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: not_an_andn1:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andl %esi, %edi
|
|
; X64-NEXT: cmpl %edi, %esi
|
|
; X64-NEXT: setg %al
|
|
; X64-NEXT: retq
|
|
%and = and i32 %x, %y
|
|
%cmp = icmp sgt i32 %y, %and
|
|
ret i1 %cmp
|
|
}
|
|
|
|
define i1 @not_an_andn2(i32 %x, i32 %y) {
|
|
; X86-LABEL: not_an_andn2:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: andl %eax, %ecx
|
|
; X86-NEXT: cmpl %ecx, %eax
|
|
; X86-NEXT: setbe %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: not_an_andn2:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andl %esi, %edi
|
|
; X64-NEXT: cmpl %edi, %esi
|
|
; X64-NEXT: setbe %al
|
|
; X64-NEXT: retq
|
|
%and = and i32 %y, %x
|
|
%cmp = icmp ule i32 %y, %and
|
|
ret i1 %cmp
|
|
}
|
|
|
|
; Don't choose a 'test' if an 'andn' can be used.
|
|
define i1 @andn_cmp_swap_ops(i64 %x, i64 %y) {
|
|
; X86-LABEL: andn_cmp_swap_ops:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %ecx, %ecx
|
|
; X86-NEXT: andnl {{[0-9]+}}(%esp), %eax, %eax
|
|
; X86-NEXT: orl %ecx, %eax
|
|
; X86-NEXT: sete %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: andn_cmp_swap_ops:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: andnq %rsi, %rdi, %rax
|
|
; X64-NEXT: sete %al
|
|
; X64-NEXT: retq
|
|
%notx = xor i64 %x, -1
|
|
%and = and i64 %y, %notx
|
|
%cmp = icmp eq i64 %and, 0
|
|
ret i1 %cmp
|
|
}
|
|
|
|
; Use a 'test' (not an 'and') because 'andn' only works for i32/i64.
|
|
define i1 @andn_cmp_i8(i8 %x, i8 %y) {
|
|
; X86-LABEL: andn_cmp_i8:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movb {{[0-9]+}}(%esp), %al
|
|
; X86-NEXT: notb %al
|
|
; X86-NEXT: testb %al, {{[0-9]+}}(%esp)
|
|
; X86-NEXT: sete %al
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: andn_cmp_i8:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: notb %sil
|
|
; X64-NEXT: testb %sil, %dil
|
|
; X64-NEXT: sete %al
|
|
; X64-NEXT: retq
|
|
%noty = xor i8 %y, -1
|
|
%and = and i8 %x, %noty
|
|
%cmp = icmp eq i8 %and, 0
|
|
ret i1 %cmp
|
|
}
|
|
|
|
declare i32 @llvm.x86.bmi.bextr.32(i32, i32)
|
|
|
|
define i32 @bextr32(i32 %x, i32 %y) {
|
|
; X86-LABEL: bextr32:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: bextrl %eax, {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: bextr32:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: bextrl %esi, %edi, %eax
|
|
; X64-NEXT: retq
|
|
%tmp = tail call i32 @llvm.x86.bmi.bextr.32(i32 %x, i32 %y)
|
|
ret i32 %tmp
|
|
}
|
|
|
|
define i32 @bextr32_load(i32* %x, i32 %y) {
|
|
; X86-LABEL: bextr32_load:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: bextrl %eax, (%ecx), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: bextr32_load:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: bextrl %esi, (%rdi), %eax
|
|
; X64-NEXT: retq
|
|
%x1 = load i32, i32* %x
|
|
%tmp = tail call i32 @llvm.x86.bmi.bextr.32(i32 %x1, i32 %y)
|
|
ret i32 %tmp
|
|
}
|
|
|
|
define i32 @bextr32b(i32 %x) uwtable ssp {
|
|
; X86-LABEL: bextr32b:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl $3076, %eax # imm = 0xC04
|
|
; X86-NEXT: bextrl %eax, {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: bextr32b:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: movl $3076, %eax # imm = 0xC04
|
|
; X64-NEXT: bextrl %eax, %edi, %eax
|
|
; X64-NEXT: retq
|
|
%1 = lshr i32 %x, 4
|
|
%2 = and i32 %1, 4095
|
|
ret i32 %2
|
|
}
|
|
|
|
; Make sure we still use AH subreg trick to extract 15:8
|
|
define i32 @bextr32_subreg(i32 %x) uwtable ssp {
|
|
; X86-LABEL: bextr32_subreg:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movzbl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: bextr32_subreg:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: movl %edi, %eax
|
|
; X64-NEXT: movzbl %ah, %eax
|
|
; X64-NEXT: retq
|
|
%1 = lshr i32 %x, 8
|
|
%2 = and i32 %1, 255
|
|
ret i32 %2
|
|
}
|
|
|
|
define i32 @bextr32b_load(i32* %x) uwtable ssp {
|
|
; X86-LABEL: bextr32b_load:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movl $3076, %ecx # imm = 0xC04
|
|
; X86-NEXT: bextrl %ecx, (%eax), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: bextr32b_load:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: movl $3076, %eax # imm = 0xC04
|
|
; X64-NEXT: bextrl %eax, (%rdi), %eax
|
|
; X64-NEXT: retq
|
|
%1 = load i32, i32* %x
|
|
%2 = lshr i32 %1, 4
|
|
%3 = and i32 %2, 4095
|
|
ret i32 %3
|
|
}
|
|
|
|
; PR34042
|
|
define i32 @bextr32c(i32 %x, i16 zeroext %y) {
|
|
; X86-LABEL: bextr32c:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movswl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: bextrl %eax, {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: bextr32c:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: bextrl %esi, %edi, %eax
|
|
; X64-NEXT: retq
|
|
%tmp0 = sext i16 %y to i32
|
|
%tmp1 = tail call i32 @llvm.x86.bmi.bextr.32(i32 %x, i32 %tmp0)
|
|
ret i32 %tmp1
|
|
}
|
|
|
|
define i32 @non_bextr32(i32 %x) {
|
|
; X86-LABEL: non_bextr32:
|
|
; X86: # %bb.0: # %entry
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: shrl $2, %eax
|
|
; X86-NEXT: andl $111, %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: non_bextr32:
|
|
; X64: # %bb.0: # %entry
|
|
; X64-NEXT: shrl $2, %edi
|
|
; X64-NEXT: andl $111, %edi
|
|
; X64-NEXT: movl %edi, %eax
|
|
; X64-NEXT: retq
|
|
entry:
|
|
%shr = lshr i32 %x, 2
|
|
%and = and i32 %shr, 111
|
|
ret i32 %and
|
|
}
|
|
|
|
define i32 @blsi32(i32 %x) {
|
|
; X86-LABEL: blsi32:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: blsil {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsi32:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsil %edi, %eax
|
|
; X64-NEXT: retq
|
|
%tmp = sub i32 0, %x
|
|
%tmp2 = and i32 %x, %tmp
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i32 @blsi32_load(i32* %x) {
|
|
; X86-LABEL: blsi32_load:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: blsil (%eax), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsi32_load:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsil (%rdi), %eax
|
|
; X64-NEXT: retq
|
|
%x1 = load i32, i32* %x
|
|
%tmp = sub i32 0, %x1
|
|
%tmp2 = and i32 %x1, %tmp
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i64 @blsi64(i64 %x) {
|
|
; X86-LABEL: blsi64:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: pushl %esi
|
|
; X86-NEXT: .cfi_def_cfa_offset 8
|
|
; X86-NEXT: .cfi_offset %esi, -8
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %esi
|
|
; X86-NEXT: xorl %edx, %edx
|
|
; X86-NEXT: movl %ecx, %eax
|
|
; X86-NEXT: negl %eax
|
|
; X86-NEXT: sbbl %esi, %edx
|
|
; X86-NEXT: andl %esi, %edx
|
|
; X86-NEXT: andl %ecx, %eax
|
|
; X86-NEXT: popl %esi
|
|
; X86-NEXT: .cfi_def_cfa_offset 4
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsi64:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsiq %rdi, %rax
|
|
; X64-NEXT: retq
|
|
%tmp = sub i64 0, %x
|
|
%tmp2 = and i64 %tmp, %x
|
|
ret i64 %tmp2
|
|
}
|
|
|
|
define i32 @blsmsk32(i32 %x) {
|
|
; X86-LABEL: blsmsk32:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: blsmskl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsmsk32:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsmskl %edi, %eax
|
|
; X64-NEXT: retq
|
|
%tmp = sub i32 %x, 1
|
|
%tmp2 = xor i32 %x, %tmp
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i32 @blsmsk32_load(i32* %x) {
|
|
; X86-LABEL: blsmsk32_load:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: blsmskl (%eax), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsmsk32_load:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsmskl (%rdi), %eax
|
|
; X64-NEXT: retq
|
|
%x1 = load i32, i32* %x
|
|
%tmp = sub i32 %x1, 1
|
|
%tmp2 = xor i32 %x1, %tmp
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i64 @blsmsk64(i64 %x) {
|
|
; X86-LABEL: blsmsk64:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: pushl %esi
|
|
; X86-NEXT: .cfi_def_cfa_offset 8
|
|
; X86-NEXT: .cfi_offset %esi, -8
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %esi
|
|
; X86-NEXT: movl %ecx, %eax
|
|
; X86-NEXT: addl $-1, %eax
|
|
; X86-NEXT: movl %esi, %edx
|
|
; X86-NEXT: adcl $-1, %edx
|
|
; X86-NEXT: xorl %ecx, %eax
|
|
; X86-NEXT: xorl %esi, %edx
|
|
; X86-NEXT: popl %esi
|
|
; X86-NEXT: .cfi_def_cfa_offset 4
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsmsk64:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsmskq %rdi, %rax
|
|
; X64-NEXT: retq
|
|
%tmp = sub i64 %x, 1
|
|
%tmp2 = xor i64 %tmp, %x
|
|
ret i64 %tmp2
|
|
}
|
|
|
|
define i32 @blsr32(i32 %x) {
|
|
; X86-LABEL: blsr32:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: blsrl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsr32:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsrl %edi, %eax
|
|
; X64-NEXT: retq
|
|
%tmp = sub i32 %x, 1
|
|
%tmp2 = and i32 %x, %tmp
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i32 @blsr32_load(i32* %x) {
|
|
; X86-LABEL: blsr32_load:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: blsrl (%eax), %eax
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsr32_load:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsrl (%rdi), %eax
|
|
; X64-NEXT: retq
|
|
%x1 = load i32, i32* %x
|
|
%tmp = sub i32 %x1, 1
|
|
%tmp2 = and i32 %x1, %tmp
|
|
ret i32 %tmp2
|
|
}
|
|
|
|
define i64 @blsr64(i64 %x) {
|
|
; X86-LABEL: blsr64:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: pushl %esi
|
|
; X86-NEXT: .cfi_def_cfa_offset 8
|
|
; X86-NEXT: .cfi_offset %esi, -8
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %esi
|
|
; X86-NEXT: movl %ecx, %eax
|
|
; X86-NEXT: addl $-1, %eax
|
|
; X86-NEXT: movl %esi, %edx
|
|
; X86-NEXT: adcl $-1, %edx
|
|
; X86-NEXT: andl %ecx, %eax
|
|
; X86-NEXT: andl %esi, %edx
|
|
; X86-NEXT: popl %esi
|
|
; X86-NEXT: .cfi_def_cfa_offset 4
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsr64:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsrq %rdi, %rax
|
|
; X64-NEXT: retq
|
|
%tmp = sub i64 %x, 1
|
|
%tmp2 = and i64 %tmp, %x
|
|
ret i64 %tmp2
|
|
}
|
|
|
|
; PR35792 - https://bugs.llvm.org/show_bug.cgi?id=35792
|
|
|
|
define i64 @blsr_disguised_constant(i64 %x) {
|
|
; X86-LABEL: blsr_disguised_constant:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: blsrl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: movzwl %ax, %eax
|
|
; X86-NEXT: xorl %edx, %edx
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsr_disguised_constant:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: blsrl %edi, %eax
|
|
; X64-NEXT: movzwl %ax, %eax
|
|
; X64-NEXT: retq
|
|
%a1 = and i64 %x, 65535
|
|
%a2 = add i64 %x, 65535
|
|
%r = and i64 %a1, %a2
|
|
ret i64 %r
|
|
}
|
|
|
|
; The add here used to get shrunk, but the and did not thus hiding the blsr pattern.
|
|
; We now use the knowledge that upper bits of the shift guarantee the and result has 0s in the upper bits to reduce it too.
|
|
define i64 @blsr_disguised_shrunk_add(i64 %x) {
|
|
; X86-LABEL: blsr_disguised_shrunk_add:
|
|
; X86: # %bb.0:
|
|
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
|
|
; X86-NEXT: shrl $16, %eax
|
|
; X86-NEXT: blsrl %eax, %eax
|
|
; X86-NEXT: xorl %edx, %edx
|
|
; X86-NEXT: retl
|
|
;
|
|
; X64-LABEL: blsr_disguised_shrunk_add:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: shrq $48, %rdi
|
|
; X64-NEXT: blsrl %edi, %eax
|
|
; X64-NEXT: retq
|
|
%a = lshr i64 %x, 48
|
|
%b = add i64 %a, -1
|
|
%c = and i64 %b, %a
|
|
ret i64 %c
|
|
}
|