llvm-mirror/test/CodeGen/X86/fp-logic.ll
Sanjay Patel 7530f47511 [x86] replace integer logic ops with packed SSE FP logic ops
If we have an operand to a bitwise logic op that's already in
an XMM register and the result is going to be sent to an XMM
register, then use an SSE logic op to avoid moves between the
integer and vector register files.

Related commits:
http://reviews.llvm.org/rL248395
http://reviews.llvm.org/rL248399
http://reviews.llvm.org/rL248404
http://reviews.llvm.org/rL248409
http://reviews.llvm.org/rL248415

This should solve PR22428:
https://llvm.org/bugs/show_bug.cgi?id=22428

llvm-svn: 251378
2015-10-27 01:28:07 +00:00

265 lines
6.3 KiB
LLVM

; RUN: llc -mtriple=x86_64-unknown-unknown -mattr=sse2 < %s | FileCheck %s
; PR22428: https://llvm.org/bugs/show_bug.cgi?id=22428
; f1, f2, f3, and f4 should use an integer logic instruction.
; f9 and f10 should use an FP (SSE) logic instruction.
;
; f5, f6, f7, and f8 are less clear.
;
; For f5 and f6, we can save a register move by using an FP logic instruction,
; but we may need to calculate the relative costs of an SSE op vs. int op vs.
; scalar <-> SSE register moves.
;
; For f7 and f8, the SSE instructions don't take immediate operands, so if we
; use one of those, we either have to load a constant from memory or move the
; scalar immediate value from an integer register over to an SSE register.
; Optimizing for size may affect that decision. Also, note that there are no
; scalar versions of the FP logic ops, so if we want to fold a load into a
; logic op, we have to load or splat a 16-byte vector constant.
; 1 FP operand, 1 int operand, int result
define i32 @f1(float %x, i32 %y) {
; CHECK-LABEL: f1:
; CHECK: # BB#0:
; CHECK-NEXT: movd %xmm0, %eax
; CHECK-NEXT: andl %edi, %eax
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 %bc1, %y
ret i32 %and
}
; Swap operands of the logic op.
define i32 @f2(float %x, i32 %y) {
; CHECK-LABEL: f2:
; CHECK: # BB#0:
; CHECK-NEXT: movd %xmm0, %eax
; CHECK-NEXT: andl %edi, %eax
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 %y, %bc1
ret i32 %and
}
; 1 FP operand, 1 constant operand, int result
define i32 @f3(float %x) {
; CHECK-LABEL: f3:
; CHECK: # BB#0:
; CHECK-NEXT: movd %xmm0, %eax
; CHECK-NEXT: andl $1, %eax
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 %bc1, 1
ret i32 %and
}
; Swap operands of the logic op.
define i32 @f4(float %x) {
; CHECK-LABEL: f4:
; CHECK: # BB#0:
; CHECK-NEXT: movd %xmm0, %eax
; CHECK-NEXT: andl $2, %eax
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 2, %bc1
ret i32 %and
}
; 1 FP operand, 1 integer operand, FP result
define float @f5(float %x, i32 %y) {
; CHECK-LABEL: f5:
; CHECK: # BB#0:
; CHECK-NEXT: movd %xmm0, %eax
; CHECK-NEXT: andl %edi, %eax
; CHECK-NEXT: movd %eax, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 %bc1, %y
%bc2 = bitcast i32 %and to float
ret float %bc2
}
; Swap operands of the logic op.
define float @f6(float %x, i32 %y) {
; CHECK-LABEL: f6:
; CHECK: # BB#0:
; CHECK-NEXT: movd %xmm0, %eax
; CHECK-NEXT: andl %edi, %eax
; CHECK-NEXT: movd %eax, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 %y, %bc1
%bc2 = bitcast i32 %and to float
ret float %bc2
}
; 1 FP operand, 1 constant operand, FP result
define float @f7(float %x) {
; CHECK-LABEL: f7:
; CHECK: # BB#0:
; CHECK-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; CHECK-NEXT: andps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 %bc1, 3
%bc2 = bitcast i32 %and to float
ret float %bc2
}
; Swap operands of the logic op.
define float @f8(float %x) {
; CHECK-LABEL: f8:
; CHECK: # BB#0:
; CHECK-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; CHECK-NEXT: andps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 4, %bc1
%bc2 = bitcast i32 %and to float
ret float %bc2
}
; 2 FP operands, int result
define i32 @f9(float %x, float %y) {
; CHECK-LABEL: f9:
; CHECK: # BB#0:
; CHECK-NEXT: andps %xmm1, %xmm0
; CHECK-NEXT: movd %xmm0, %eax
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%bc2 = bitcast float %y to i32
%and = and i32 %bc1, %bc2
ret i32 %and
}
; 2 FP operands, FP result
define float @f10(float %x, float %y) {
; CHECK-LABEL: f10:
; CHECK: # BB#0:
; CHECK-NEXT: andps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%bc2 = bitcast float %y to i32
%and = and i32 %bc1, %bc2
%bc3 = bitcast i32 %and to float
ret float %bc3
}
define float @or(float %x, float %y) {
; CHECK-LABEL: or:
; CHECK: # BB#0:
; CHECK-NEXT: orps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%bc2 = bitcast float %y to i32
%and = or i32 %bc1, %bc2
%bc3 = bitcast i32 %and to float
ret float %bc3
}
define float @xor(float %x, float %y) {
; CHECK-LABEL: xor:
; CHECK: # BB#0:
; CHECK-NEXT: xorps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%bc2 = bitcast float %y to i32
%and = xor i32 %bc1, %bc2
%bc3 = bitcast i32 %and to float
ret float %bc3
}
define float @f7_or(float %x) {
; CHECK-LABEL: f7_or:
; CHECK: # BB#0:
; CHECK-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; CHECK-NEXT: orps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = or i32 %bc1, 3
%bc2 = bitcast i32 %and to float
ret float %bc2
}
define float @f7_xor(float %x) {
; CHECK-LABEL: f7_xor:
; CHECK: # BB#0:
; CHECK-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; CHECK-NEXT: xorps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = xor i32 %bc1, 3
%bc2 = bitcast i32 %and to float
ret float %bc2
}
; Make sure that doubles work too.
define double @doubles(double %x, double %y) {
; CHECK-LABEL: doubles:
; CHECK: # BB#0:
; CHECK-NEXT: andpd %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast double %x to i64
%bc2 = bitcast double %y to i64
%and = and i64 %bc1, %bc2
%bc3 = bitcast i64 %and to double
ret double %bc3
}
define double @f7_double(double %x) {
; CHECK-LABEL: f7_double:
; CHECK: # BB#0:
; CHECK-NEXT: movsd {{.*#+}} xmm1 = mem[0],zero
; CHECK-NEXT: andpd %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast double %x to i64
%and = and i64 %bc1, 3
%bc2 = bitcast i64 %and to double
ret double %bc2
}
; Grabbing the sign bit is a special case that could be handled
; by movmskps/movmskpd, but if we're not shifting it over, then
; a simple FP logic op is cheaper.
define float @movmsk(float %x) {
; CHECK-LABEL: movmsk:
; CHECK: # BB#0:
; CHECK-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; CHECK-NEXT: andps %xmm1, %xmm0
; CHECK-NEXT: retq
%bc1 = bitcast float %x to i32
%and = and i32 %bc1, 2147483648
%bc2 = bitcast i32 %and to float
ret float %bc2
}