881 lines
		
	
	
		
			37 KiB
		
	
	
	
		
			LLVM
		
	
	
	
			
		
		
	
	
			881 lines
		
	
	
		
			37 KiB
		
	
	
	
		
			LLVM
		
	
	
	
| ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 | |
| ; RUN: opt < %s -instcombine -S | FileCheck %s
 | |
| 
 | |
| declare void @use(i8)
 | |
| 
 | |
| define i32 @foo(i32 %a, i32 %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @foo(
 | |
| ; CHECK-NEXT:    [[E:%.*]] = icmp slt i32 [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[E]], i32 [[C:%.*]], i32 [[D:%.*]]
 | |
| ; CHECK-NEXT:    ret i32 [[TMP1]]
 | |
| ;
 | |
|   %e = icmp slt i32 %a, %b
 | |
|   %f = sext i1 %e to i32
 | |
|   %g = and i32 %c, %f
 | |
|   %h = xor i32 %f, -1
 | |
|   %i = and i32 %d, %h
 | |
|   %j = or i32 %g, %i
 | |
|   ret i32 %j
 | |
| }
 | |
| 
 | |
| define i32 @bar(i32 %a, i32 %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @bar(
 | |
| ; CHECK-NEXT:    [[E:%.*]] = icmp slt i32 [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[E]], i32 [[C:%.*]], i32 [[D:%.*]]
 | |
| ; CHECK-NEXT:    ret i32 [[TMP1]]
 | |
| ;
 | |
|   %e = icmp slt i32 %a, %b
 | |
|   %f = sext i1 %e to i32
 | |
|   %g = and i32 %c, %f
 | |
|   %h = xor i32 %f, -1
 | |
|   %i = and i32 %d, %h
 | |
|   %j = or i32 %i, %g
 | |
|   ret i32 %j
 | |
| }
 | |
| 
 | |
| define i32 @goo(i32 %a, i32 %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @goo(
 | |
| ; CHECK-NEXT:    [[T0:%.*]] = icmp slt i32 [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[T0]], i32 [[C:%.*]], i32 [[D:%.*]]
 | |
| ; CHECK-NEXT:    ret i32 [[TMP1]]
 | |
| ;
 | |
|   %t0 = icmp slt i32 %a, %b
 | |
|   %iftmp.0.0 = select i1 %t0, i32 -1, i32 0
 | |
|   %t1 = and i32 %iftmp.0.0, %c
 | |
|   %not = xor i32 %iftmp.0.0, -1
 | |
|   %t2 = and i32 %not, %d
 | |
|   %t3 = or i32 %t1, %t2
 | |
|   ret i32 %t3
 | |
| }
 | |
| 
 | |
| define i32 @poo(i32 %a, i32 %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @poo(
 | |
| ; CHECK-NEXT:    [[T0:%.*]] = icmp slt i32 [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[T3:%.*]] = select i1 [[T0]], i32 [[C:%.*]], i32 [[D:%.*]]
 | |
| ; CHECK-NEXT:    ret i32 [[T3]]
 | |
| ;
 | |
|   %t0 = icmp slt i32 %a, %b
 | |
|   %iftmp.0.0 = select i1 %t0, i32 -1, i32 0
 | |
|   %t1 = and i32 %iftmp.0.0, %c
 | |
|   %iftmp = select i1 %t0, i32 0, i32 -1
 | |
|   %t2 = and i32 %iftmp, %d
 | |
|   %t3 = or i32 %t1, %t2
 | |
|   ret i32 %t3
 | |
| }
 | |
| 
 | |
| ; PR32791 - https://bugs.llvm.org//show_bug.cgi?id=32791
 | |
| ; The 2nd compare/select are canonicalized, so CSE and another round of instcombine or some other pass will fold this.
 | |
| 
 | |
| define i32 @fold_inverted_icmp_preds(i32 %a, i32 %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @fold_inverted_icmp_preds(
 | |
| ; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[SEL1:%.*]] = select i1 [[CMP1]], i32 [[C:%.*]], i32 0
 | |
| ; CHECK-NEXT:    [[CMP2_NOT:%.*]] = icmp slt i32 [[A]], [[B]]
 | |
| ; CHECK-NEXT:    [[SEL2:%.*]] = select i1 [[CMP2_NOT]], i32 0, i32 [[D:%.*]]
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SEL1]], [[SEL2]]
 | |
| ; CHECK-NEXT:    ret i32 [[OR]]
 | |
| ;
 | |
|   %cmp1 = icmp slt i32 %a, %b
 | |
|   %sel1 = select i1 %cmp1, i32 %c, i32 0
 | |
|   %cmp2 = icmp sge i32 %a, %b
 | |
|   %sel2 = select i1 %cmp2, i32 %d, i32 0
 | |
|   %or = or i32 %sel1, %sel2
 | |
|   ret i32 %or
 | |
| }
 | |
| 
 | |
| ; The 2nd compare/select are canonicalized, so CSE and another round of instcombine or some other pass will fold this.
 | |
| 
 | |
| define i32 @fold_inverted_icmp_preds_reverse(i32 %a, i32 %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @fold_inverted_icmp_preds_reverse(
 | |
| ; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt i32 [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[SEL1:%.*]] = select i1 [[CMP1]], i32 0, i32 [[C:%.*]]
 | |
| ; CHECK-NEXT:    [[CMP2_NOT:%.*]] = icmp slt i32 [[A]], [[B]]
 | |
| ; CHECK-NEXT:    [[SEL2:%.*]] = select i1 [[CMP2_NOT]], i32 [[D:%.*]], i32 0
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SEL1]], [[SEL2]]
 | |
| ; CHECK-NEXT:    ret i32 [[OR]]
 | |
| ;
 | |
|   %cmp1 = icmp slt i32 %a, %b
 | |
|   %sel1 = select i1 %cmp1, i32 0, i32 %c
 | |
|   %cmp2 = icmp sge i32 %a, %b
 | |
|   %sel2 = select i1 %cmp2, i32 0, i32 %d
 | |
|   %or = or i32 %sel1, %sel2
 | |
|   ret i32 %or
 | |
| }
 | |
| 
 | |
| ; TODO: Should fcmp have the same sort of predicate canonicalization as icmp?
 | |
| 
 | |
| define i32 @fold_inverted_fcmp_preds(float %a, float %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @fold_inverted_fcmp_preds(
 | |
| ; CHECK-NEXT:    [[CMP1:%.*]] = fcmp olt float [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[SEL1:%.*]] = select i1 [[CMP1]], i32 [[C:%.*]], i32 0
 | |
| ; CHECK-NEXT:    [[CMP2:%.*]] = fcmp uge float [[A]], [[B]]
 | |
| ; CHECK-NEXT:    [[SEL2:%.*]] = select i1 [[CMP2]], i32 [[D:%.*]], i32 0
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SEL1]], [[SEL2]]
 | |
| ; CHECK-NEXT:    ret i32 [[OR]]
 | |
| ;
 | |
|   %cmp1 = fcmp olt float %a, %b
 | |
|   %sel1 = select i1 %cmp1, i32 %c, i32 0
 | |
|   %cmp2 = fcmp uge float %a, %b
 | |
|   %sel2 = select i1 %cmp2, i32 %d, i32 0
 | |
|   %or = or i32 %sel1, %sel2
 | |
|   ret i32 %or
 | |
| }
 | |
| 
 | |
| ; The 2nd compare/select are canonicalized, so CSE and another round of instcombine or some other pass will fold this.
 | |
| 
 | |
| define <2 x i32> @fold_inverted_icmp_vector_preds(<2 x i32> %a, <2 x i32> %b, <2 x i32> %c, <2 x i32> %d) {
 | |
| ; CHECK-LABEL: @fold_inverted_icmp_vector_preds(
 | |
| ; CHECK-NEXT:    [[CMP1_NOT:%.*]] = icmp eq <2 x i32> [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[SEL1:%.*]] = select <2 x i1> [[CMP1_NOT]], <2 x i32> zeroinitializer, <2 x i32> [[C:%.*]]
 | |
| ; CHECK-NEXT:    [[CMP2:%.*]] = icmp eq <2 x i32> [[A]], [[B]]
 | |
| ; CHECK-NEXT:    [[SEL2:%.*]] = select <2 x i1> [[CMP2]], <2 x i32> [[D:%.*]], <2 x i32> zeroinitializer
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or <2 x i32> [[SEL1]], [[SEL2]]
 | |
| ; CHECK-NEXT:    ret <2 x i32> [[OR]]
 | |
| ;
 | |
|   %cmp1 = icmp ne <2 x i32> %a, %b
 | |
|   %sel1 = select <2 x i1> %cmp1, <2 x i32> %c, <2 x i32> <i32 0, i32 0>
 | |
|   %cmp2 = icmp eq <2 x i32> %a, %b
 | |
|   %sel2 = select <2 x i1> %cmp2, <2 x i32> %d, <2 x i32> <i32 0, i32 0>
 | |
|   %or = or <2 x i32> %sel1, %sel2
 | |
|   ret <2 x i32> %or
 | |
| }
 | |
| 
 | |
| define i32 @par(i32 %a, i32 %b, i32 %c, i32 %d) {
 | |
| ; CHECK-LABEL: @par(
 | |
| ; CHECK-NEXT:    [[T0:%.*]] = icmp slt i32 [[A:%.*]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[T0]], i32 [[C:%.*]], i32 [[D:%.*]]
 | |
| ; CHECK-NEXT:    ret i32 [[TMP1]]
 | |
| ;
 | |
|   %t0 = icmp slt i32 %a, %b
 | |
|   %iftmp.1.0 = select i1 %t0, i32 -1, i32 0
 | |
|   %t1 = and i32 %iftmp.1.0, %c
 | |
|   %not = xor i32 %iftmp.1.0, -1
 | |
|   %t2 = and i32 %not, %d
 | |
|   %t3 = or i32 %t1, %t2
 | |
|   ret i32 %t3
 | |
| }
 | |
| 
 | |
| ; In the following tests (8 commutation variants), verify that a bitcast doesn't get
 | |
| ; in the way of a select transform. These bitcasts are common in SSE/AVX and possibly
 | |
| ; other vector code because of canonicalization to i64 elements for vectors.
 | |
| 
 | |
| ; The fptosi instructions are included to avoid commutation canonicalization based on
 | |
| ; operator weight. Using another cast operator ensures that both operands of all logic
 | |
| ; ops are equally weighted, and this ensures that we're testing all commutation
 | |
| ; possibilities.
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap0(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap0(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %bc1, %sia
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %bc2, %sib
 | |
|   %or = or <2 x i64> %and1, %and2
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap1(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap1(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %bc1, %sia
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %bc2, %sib
 | |
|   %or = or <2 x i64> %and2, %and1
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap2(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap2(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %bc1, %sia
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %sib, %bc2
 | |
|   %or = or <2 x i64> %and1, %and2
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap3(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap3(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %bc1, %sia
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %sib, %bc2
 | |
|   %or = or <2 x i64> %and2, %and1
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap4(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap4(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %sia, %bc1
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %bc2, %sib
 | |
|   %or = or <2 x i64> %and1, %and2
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap5(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap5(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %sia, %bc1
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %bc2, %sib
 | |
|   %or = or <2 x i64> %and2, %and1
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap6(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap6(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %sia, %bc1
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %sib, %bc2
 | |
|   %or = or <2 x i64> %and1, %and2
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_swap7(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_swap7(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[SIA]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[SIB]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[CMP:%.*]], <4 x i32> [[TMP1]], <4 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %sia, %bc1
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %sib, %bc2
 | |
|   %or = or <2 x i64> %and2, %and1
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <2 x i64> @bitcast_select_multi_uses(<4 x i1> %cmp, <2 x i64> %a, <2 x i64> %b) {
 | |
| ; CHECK-LABEL: @bitcast_select_multi_uses(
 | |
| ; CHECK-NEXT:    [[SEXT:%.*]] = sext <4 x i1> [[CMP:%.*]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[BC1:%.*]] = bitcast <4 x i32> [[SEXT]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[AND1:%.*]] = and <2 x i64> [[BC1]], [[A:%.*]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[SEXT]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[BC2:%.*]] = xor <2 x i64> [[TMP1]], <i64 -1, i64 -1>
 | |
| ; CHECK-NEXT:    [[AND2:%.*]] = and <2 x i64> [[BC2]], [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or <2 x i64> [[AND2]], [[AND1]]
 | |
| ; CHECK-NEXT:    [[ADD:%.*]] = add <2 x i64> [[AND2]], [[BC2]]
 | |
| ; CHECK-NEXT:    [[SUB:%.*]] = sub <2 x i64> [[OR]], [[ADD]]
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[SUB]]
 | |
| ;
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %bc1 = bitcast <4 x i32> %sext to <2 x i64>
 | |
|   %and1 = and <2 x i64> %a, %bc1
 | |
|   %neg = xor <4 x i32> %sext, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %bc2 = bitcast <4 x i32> %neg to <2 x i64>
 | |
|   %and2 = and <2 x i64> %b, %bc2
 | |
|   %or = or <2 x i64> %and2, %and1
 | |
|   %add = add <2 x i64> %and2, %bc2
 | |
|   %sub = sub <2 x i64> %or, %add
 | |
|   ret <2 x i64> %sub
 | |
| }
 | |
| 
 | |
| define i1 @bools(i1 %a, i1 %b, i1 %c) {
 | |
| ; CHECK-LABEL: @bools(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[C:%.*]], i1 [[B:%.*]], i1 [[A:%.*]]
 | |
| ; CHECK-NEXT:    ret i1 [[TMP1]]
 | |
| ;
 | |
|   %not = xor i1 %c, -1
 | |
|   %and1 = and i1 %not, %a
 | |
|   %and2 = and i1 %c, %b
 | |
|   %or = or i1 %and1, %and2
 | |
|   ret i1 %or
 | |
| }
 | |
| 
 | |
| define i1 @bools_logical(i1 %a, i1 %b, i1 %c) {
 | |
| ; CHECK-LABEL: @bools_logical(
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = select i1 [[C:%.*]], i1 [[B:%.*]], i1 [[A:%.*]]
 | |
| ; CHECK-NEXT:    ret i1 [[OR]]
 | |
| ;
 | |
|   %not = xor i1 %c, -1
 | |
|   %and1 = select i1 %not, i1 %a, i1 false
 | |
|   %and2 = select i1 %c, i1 %b, i1 false
 | |
|   %or = select i1 %and1, i1 true, i1 %and2
 | |
|   ret i1 %or
 | |
| }
 | |
| 
 | |
| ; Form a select if we know we can replace 2 simple logic ops.
 | |
| 
 | |
| define i1 @bools_multi_uses1(i1 %a, i1 %b, i1 %c) {
 | |
| ; CHECK-LABEL: @bools_multi_uses1(
 | |
| ; CHECK-NEXT:    [[NOT:%.*]] = xor i1 [[C:%.*]], true
 | |
| ; CHECK-NEXT:    [[AND1:%.*]] = and i1 [[NOT]], [[A:%.*]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[C]], i1 [[B:%.*]], i1 [[A]]
 | |
| ; CHECK-NEXT:    [[XOR:%.*]] = xor i1 [[TMP1]], [[AND1]]
 | |
| ; CHECK-NEXT:    ret i1 [[XOR]]
 | |
| ;
 | |
|   %not = xor i1 %c, -1
 | |
|   %and1 = and i1 %not, %a
 | |
|   %and2 = and i1 %c, %b
 | |
|   %or = or i1 %and1, %and2
 | |
|   %xor = xor i1 %or, %and1
 | |
|   ret i1 %xor
 | |
| }
 | |
| 
 | |
| define i1 @bools_multi_uses1_logical(i1 %a, i1 %b, i1 %c) {
 | |
| ; CHECK-LABEL: @bools_multi_uses1_logical(
 | |
| ; CHECK-NEXT:    [[NOT:%.*]] = xor i1 [[C:%.*]], true
 | |
| ; CHECK-NEXT:    [[AND1:%.*]] = select i1 [[NOT]], i1 [[A:%.*]], i1 false
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = select i1 [[C]], i1 [[B:%.*]], i1 [[A]]
 | |
| ; CHECK-NEXT:    [[XOR:%.*]] = xor i1 [[OR]], [[AND1]]
 | |
| ; CHECK-NEXT:    ret i1 [[XOR]]
 | |
| ;
 | |
|   %not = xor i1 %c, -1
 | |
|   %and1 = select i1 %not, i1 %a, i1 false
 | |
|   %and2 = select i1 %c, i1 %b, i1 false
 | |
|   %or = select i1 %and1, i1 true, i1 %and2
 | |
|   %xor = xor i1 %or, %and1
 | |
|   ret i1 %xor
 | |
| }
 | |
| 
 | |
| ; Don't replace a cheap logic op with a potentially expensive select
 | |
| ; unless we can also eliminate one of the other original ops.
 | |
| 
 | |
| define i1 @bools_multi_uses2(i1 %a, i1 %b, i1 %c) {
 | |
| ; CHECK-LABEL: @bools_multi_uses2(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[C:%.*]], i1 [[B:%.*]], i1 [[A:%.*]]
 | |
| ; CHECK-NEXT:    ret i1 [[TMP1]]
 | |
| ;
 | |
|   %not = xor i1 %c, -1
 | |
|   %and1 = and i1 %not, %a
 | |
|   %and2 = and i1 %c, %b
 | |
|   %or = or i1 %and1, %and2
 | |
|   %add = add i1 %and1, %and2
 | |
|   %and3 = and i1 %or, %add
 | |
|   ret i1 %and3
 | |
| }
 | |
| 
 | |
| define i1 @bools_multi_uses2_logical(i1 %a, i1 %b, i1 %c) {
 | |
| ; CHECK-LABEL: @bools_multi_uses2_logical(
 | |
| ; CHECK-NEXT:    [[NOT:%.*]] = xor i1 [[C:%.*]], true
 | |
| ; CHECK-NEXT:    [[AND1:%.*]] = select i1 [[NOT]], i1 [[A:%.*]], i1 false
 | |
| ; CHECK-NEXT:    [[AND2:%.*]] = select i1 [[C]], i1 [[B:%.*]], i1 false
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = select i1 [[C]], i1 [[B]], i1 [[A]]
 | |
| ; CHECK-NEXT:    [[ADD:%.*]] = xor i1 [[AND1]], [[AND2]]
 | |
| ; CHECK-NEXT:    [[AND3:%.*]] = select i1 [[OR]], i1 [[ADD]], i1 false
 | |
| ; CHECK-NEXT:    ret i1 [[AND3]]
 | |
| ;
 | |
|   %not = xor i1 %c, -1
 | |
|   %and1 = select i1 %not, i1 %a, i1 false
 | |
|   %and2 = select i1 %c, i1 %b, i1 false
 | |
|   %or = select i1 %and1, i1 true, i1 %and2
 | |
|   %add = add i1 %and1, %and2
 | |
|   %and3 = select i1 %or, i1 %add, i1 false
 | |
|   ret i1 %and3
 | |
| }
 | |
| 
 | |
| define <4 x i1> @vec_of_bools(<4 x i1> %a, <4 x i1> %b, <4 x i1> %c) {
 | |
| ; CHECK-LABEL: @vec_of_bools(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select <4 x i1> [[C:%.*]], <4 x i1> [[B:%.*]], <4 x i1> [[A:%.*]]
 | |
| ; CHECK-NEXT:    ret <4 x i1> [[TMP1]]
 | |
| ;
 | |
|   %not = xor <4 x i1> %c, <i1 true, i1 true, i1 true, i1 true>
 | |
|   %and1 = and <4 x i1> %not, %a
 | |
|   %and2 = and <4 x i1> %b, %c
 | |
|   %or = or <4 x i1> %and2, %and1
 | |
|   ret <4 x i1> %or
 | |
| }
 | |
| 
 | |
| define i4 @vec_of_casted_bools(i4 %a, i4 %b, <4 x i1> %c) {
 | |
| ; CHECK-LABEL: @vec_of_casted_bools(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i4 [[B:%.*]] to <4 x i1>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i4 [[A:%.*]] to <4 x i1>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[C:%.*]], <4 x i1> [[TMP1]], <4 x i1> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i1> [[TMP3]] to i4
 | |
| ; CHECK-NEXT:    ret i4 [[TMP4]]
 | |
| ;
 | |
|   %not = xor <4 x i1> %c, <i1 true, i1 true, i1 true, i1 true>
 | |
|   %bc1 = bitcast <4 x i1> %not to i4
 | |
|   %bc2 = bitcast <4 x i1> %c to i4
 | |
|   %and1 = and i4 %a, %bc1
 | |
|   %and2 = and i4 %bc2, %b
 | |
|   %or = or i4 %and1, %and2
 | |
|   ret i4 %or
 | |
| }
 | |
| 
 | |
| ; Inverted 'and' constants mean this is a select which is canonicalized to a shuffle.
 | |
| 
 | |
| define <4 x i32> @vec_sel_consts(<4 x i32> %a, <4 x i32> %b) {
 | |
| ; CHECK-LABEL: @vec_sel_consts(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i32> [[A:%.*]], <4 x i32> [[B:%.*]], <4 x i32> <i32 0, i32 5, i32 6, i32 3>
 | |
| ; CHECK-NEXT:    ret <4 x i32> [[TMP1]]
 | |
| ;
 | |
|   %and1 = and <4 x i32> %a, <i32 -1, i32 0, i32 0, i32 -1>
 | |
|   %and2 = and <4 x i32> %b, <i32 0, i32 -1, i32 -1, i32 0>
 | |
|   %or = or <4 x i32> %and1, %and2
 | |
|   ret <4 x i32> %or
 | |
| }
 | |
| 
 | |
| define <3 x i129> @vec_sel_consts_weird(<3 x i129> %a, <3 x i129> %b) {
 | |
| ; CHECK-LABEL: @vec_sel_consts_weird(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <3 x i129> [[A:%.*]], <3 x i129> [[B:%.*]], <3 x i32> <i32 0, i32 4, i32 2>
 | |
| ; CHECK-NEXT:    ret <3 x i129> [[TMP1]]
 | |
| ;
 | |
|   %and1 = and <3 x i129> %a, <i129 -1, i129 0, i129 -1>
 | |
|   %and2 = and <3 x i129> %b, <i129 0, i129 -1, i129 0>
 | |
|   %or = or <3 x i129> %and2, %and1
 | |
|   ret <3 x i129> %or
 | |
| }
 | |
| 
 | |
| ; The mask elements must be inverted for this to be a select.
 | |
| 
 | |
| define <4 x i32> @vec_not_sel_consts(<4 x i32> %a, <4 x i32> %b) {
 | |
| ; CHECK-LABEL: @vec_not_sel_consts(
 | |
| ; CHECK-NEXT:    [[AND1:%.*]] = and <4 x i32> [[A:%.*]], <i32 -1, i32 0, i32 0, i32 0>
 | |
| ; CHECK-NEXT:    [[AND2:%.*]] = and <4 x i32> [[B:%.*]], <i32 0, i32 -1, i32 0, i32 -1>
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or <4 x i32> [[AND1]], [[AND2]]
 | |
| ; CHECK-NEXT:    ret <4 x i32> [[OR]]
 | |
| ;
 | |
|   %and1 = and <4 x i32> %a, <i32 -1, i32 0, i32 0, i32 0>
 | |
|   %and2 = and <4 x i32> %b, <i32 0, i32 -1, i32 0, i32 -1>
 | |
|   %or = or <4 x i32> %and1, %and2
 | |
|   ret <4 x i32> %or
 | |
| }
 | |
| 
 | |
| define <4 x i32> @vec_not_sel_consts_undef_elts(<4 x i32> %a, <4 x i32> %b) {
 | |
| ; CHECK-LABEL: @vec_not_sel_consts_undef_elts(
 | |
| ; CHECK-NEXT:    [[AND1:%.*]] = and <4 x i32> [[A:%.*]], <i32 -1, i32 undef, i32 0, i32 0>
 | |
| ; CHECK-NEXT:    [[AND2:%.*]] = and <4 x i32> [[B:%.*]], <i32 0, i32 -1, i32 0, i32 undef>
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or <4 x i32> [[AND1]], [[AND2]]
 | |
| ; CHECK-NEXT:    ret <4 x i32> [[OR]]
 | |
| ;
 | |
|   %and1 = and <4 x i32> %a, <i32 -1, i32 undef, i32 0, i32 0>
 | |
|   %and2 = and <4 x i32> %b, <i32 0, i32 -1, i32 0, i32 undef>
 | |
|   %or = or <4 x i32> %and1, %and2
 | |
|   ret <4 x i32> %or
 | |
| }
 | |
| 
 | |
| ; The inverted constants may be operands of xor instructions.
 | |
| 
 | |
| define <4 x i32> @vec_sel_xor(<4 x i32> %a, <4 x i32> %b, <4 x i1> %c) {
 | |
| ; CHECK-LABEL: @vec_sel_xor(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = xor <4 x i1> [[C:%.*]], <i1 false, i1 true, i1 true, i1 true>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = select <4 x i1> [[TMP1]], <4 x i32> [[A:%.*]], <4 x i32> [[B:%.*]]
 | |
| ; CHECK-NEXT:    ret <4 x i32> [[TMP2]]
 | |
| ;
 | |
|   %mask = sext <4 x i1> %c to <4 x i32>
 | |
|   %mask_flip1 = xor <4 x i32> %mask, <i32 -1, i32 0, i32 0, i32 0>
 | |
|   %not_mask_flip1 = xor <4 x i32> %mask, <i32 0, i32 -1, i32 -1, i32 -1>
 | |
|   %and1 = and <4 x i32> %not_mask_flip1, %a
 | |
|   %and2 = and <4 x i32> %mask_flip1, %b
 | |
|   %or = or <4 x i32> %and1, %and2
 | |
|   ret <4 x i32> %or
 | |
| }
 | |
| 
 | |
| ; Allow the transform even if the mask values have multiple uses because
 | |
| ; there's still a net reduction of instructions from removing the and/and/or.
 | |
| 
 | |
| define <4 x i32> @vec_sel_xor_multi_use(<4 x i32> %a, <4 x i32> %b, <4 x i1> %c) {
 | |
| ; CHECK-LABEL: @vec_sel_xor_multi_use(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = xor <4 x i1> [[C:%.*]], <i1 true, i1 false, i1 false, i1 false>
 | |
| ; CHECK-NEXT:    [[MASK_FLIP1:%.*]] = sext <4 x i1> [[TMP1]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = xor <4 x i1> [[C]], <i1 false, i1 true, i1 true, i1 true>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[TMP2]], <4 x i32> [[A:%.*]], <4 x i32> [[B:%.*]]
 | |
| ; CHECK-NEXT:    [[ADD:%.*]] = add <4 x i32> [[TMP3]], [[MASK_FLIP1]]
 | |
| ; CHECK-NEXT:    ret <4 x i32> [[ADD]]
 | |
| ;
 | |
|   %mask = sext <4 x i1> %c to <4 x i32>
 | |
|   %mask_flip1 = xor <4 x i32> %mask, <i32 -1, i32 0, i32 0, i32 0>
 | |
|   %not_mask_flip1 = xor <4 x i32> %mask, <i32 0, i32 -1, i32 -1, i32 -1>
 | |
|   %and1 = and <4 x i32> %not_mask_flip1, %a
 | |
|   %and2 = and <4 x i32> %mask_flip1, %b
 | |
|   %or = or <4 x i32> %and1, %and2
 | |
|   %add = add <4 x i32> %or, %mask_flip1
 | |
|   ret <4 x i32> %add
 | |
| }
 | |
| 
 | |
| ; The 'ashr' guarantees that we have a bitmask, so this is select with truncated condition.
 | |
| 
 | |
| define i32 @allSignBits(i32 %cond, i32 %tval, i32 %fval) {
 | |
| ; CHECK-LABEL: @allSignBits(
 | |
| ; CHECK-NEXT:    [[ISNEG1:%.*]] = icmp slt i32 [[COND:%.*]], 0
 | |
| ; CHECK-NEXT:    [[A1:%.*]] = select i1 [[ISNEG1]], i32 [[TVAL:%.*]], i32 0
 | |
| ; CHECK-NEXT:    [[ISNEG:%.*]] = icmp slt i32 [[COND]], 0
 | |
| ; CHECK-NEXT:    [[A2:%.*]] = select i1 [[ISNEG]], i32 0, i32 [[FVAL:%.*]]
 | |
| ; CHECK-NEXT:    [[SEL:%.*]] = or i32 [[A1]], [[A2]]
 | |
| ; CHECK-NEXT:    ret i32 [[SEL]]
 | |
| ;
 | |
|   %bitmask = ashr i32 %cond, 31
 | |
|   %not_bitmask = xor i32 %bitmask, -1
 | |
|   %a1 = and i32 %tval, %bitmask
 | |
|   %a2 = and i32 %not_bitmask, %fval
 | |
|   %sel = or i32 %a1, %a2
 | |
|   ret i32 %sel
 | |
| }
 | |
| 
 | |
| define <4 x i8> @allSignBits_vec(<4 x i8> %cond, <4 x i8> %tval, <4 x i8> %fval) {
 | |
| ; CHECK-LABEL: @allSignBits_vec(
 | |
| ; CHECK-NEXT:    [[ISNEG1:%.*]] = icmp slt <4 x i8> [[COND:%.*]], zeroinitializer
 | |
| ; CHECK-NEXT:    [[A1:%.*]] = select <4 x i1> [[ISNEG1]], <4 x i8> [[TVAL:%.*]], <4 x i8> zeroinitializer
 | |
| ; CHECK-NEXT:    [[ISNEG:%.*]] = icmp slt <4 x i8> [[COND]], zeroinitializer
 | |
| ; CHECK-NEXT:    [[A2:%.*]] = select <4 x i1> [[ISNEG]], <4 x i8> zeroinitializer, <4 x i8> [[FVAL:%.*]]
 | |
| ; CHECK-NEXT:    [[SEL:%.*]] = or <4 x i8> [[A2]], [[A1]]
 | |
| ; CHECK-NEXT:    ret <4 x i8> [[SEL]]
 | |
| ;
 | |
|   %bitmask = ashr <4 x i8> %cond, <i8 7, i8 7, i8 7, i8 7>
 | |
|   %not_bitmask = xor <4 x i8> %bitmask, <i8 -1, i8 -1, i8 -1, i8 -1>
 | |
|   %a1 = and <4 x i8> %tval, %bitmask
 | |
|   %a2 = and <4 x i8> %fval, %not_bitmask
 | |
|   %sel = or <4 x i8> %a2, %a1
 | |
|   ret <4 x i8> %sel
 | |
| }
 | |
| 
 | |
| ; Negative test - make sure that bitcasts from FP do not cause a crash.
 | |
| 
 | |
| define <2 x i64> @fp_bitcast(<4 x i1> %cmp, <2 x double> %a, <2 x double> %b) {
 | |
| ; CHECK-LABEL: @fp_bitcast(
 | |
| ; CHECK-NEXT:    [[SIA:%.*]] = fptosi <2 x double> [[A:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[SIB:%.*]] = fptosi <2 x double> [[B:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[BC1:%.*]] = bitcast <2 x double> [[A]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[AND1:%.*]] = and <2 x i64> [[SIA]], [[BC1]]
 | |
| ; CHECK-NEXT:    [[BC2:%.*]] = bitcast <2 x double> [[B]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[AND2:%.*]] = and <2 x i64> [[SIB]], [[BC2]]
 | |
| ; CHECK-NEXT:    [[OR:%.*]] = or <2 x i64> [[AND2]], [[AND1]]
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[OR]]
 | |
| ;
 | |
|   %sia = fptosi <2 x double> %a to <2 x i64>
 | |
|   %sib = fptosi <2 x double> %b to <2 x i64>
 | |
|   %bc1 = bitcast <2 x double> %a to <2 x i64>
 | |
|   %and1 = and <2 x i64> %sia, %bc1
 | |
|   %bc2 = bitcast <2 x double> %b to <2 x i64>
 | |
|   %and2 = and <2 x i64> %sib, %bc2
 | |
|   %or = or <2 x i64> %and2, %and1
 | |
|   ret <2 x i64> %or
 | |
| }
 | |
| 
 | |
| define <4 x i32> @computesignbits_through_shuffles(<4 x float> %x, <4 x float> %y, <4 x float> %z) {
 | |
| ; CHECK-LABEL: @computesignbits_through_shuffles(
 | |
| ; CHECK-NEXT:    [[CMP:%.*]] = fcmp ole <4 x float> [[X:%.*]], [[Y:%.*]]
 | |
| ; CHECK-NEXT:    [[SEXT:%.*]] = sext <4 x i1> [[CMP]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[S1:%.*]] = shufflevector <4 x i32> [[SEXT]], <4 x i32> undef, <4 x i32> <i32 0, i32 0, i32 1, i32 1>
 | |
| ; CHECK-NEXT:    [[S2:%.*]] = shufflevector <4 x i32> [[SEXT]], <4 x i32> undef, <4 x i32> <i32 2, i32 2, i32 3, i32 3>
 | |
| ; CHECK-NEXT:    [[SHUF_OR1:%.*]] = or <4 x i32> [[S1]], [[S2]]
 | |
| ; CHECK-NEXT:    [[S3:%.*]] = shufflevector <4 x i32> [[SHUF_OR1]], <4 x i32> undef, <4 x i32> <i32 0, i32 0, i32 1, i32 1>
 | |
| ; CHECK-NEXT:    [[S4:%.*]] = shufflevector <4 x i32> [[SHUF_OR1]], <4 x i32> undef, <4 x i32> <i32 2, i32 2, i32 3, i32 3>
 | |
| ; CHECK-NEXT:    [[SHUF_OR2:%.*]] = or <4 x i32> [[S3]], [[S4]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = trunc <4 x i32> [[SHUF_OR2]] to <4 x i1>
 | |
| ; CHECK-NEXT:    [[DOTV:%.*]] = select <4 x i1> [[TMP1]], <4 x float> [[Z:%.*]], <4 x float> [[X]]
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x float> [[DOTV]] to <4 x i32>
 | |
| ; CHECK-NEXT:    ret <4 x i32> [[TMP2]]
 | |
| ;
 | |
|   %cmp = fcmp ole <4 x float> %x, %y
 | |
|   %sext = sext <4 x i1> %cmp to <4 x i32>
 | |
|   %s1 = shufflevector <4 x i32> %sext, <4 x i32> undef, <4 x i32> <i32 0, i32 0, i32 1, i32 1>
 | |
|   %s2 = shufflevector <4 x i32> %sext, <4 x i32> undef, <4 x i32> <i32 2, i32 2, i32 3, i32 3>
 | |
|   %shuf_or1 = or <4 x i32> %s1, %s2
 | |
|   %s3 = shufflevector <4 x i32> %shuf_or1, <4 x i32> undef, <4 x i32> <i32 0, i32 0, i32 1, i32 1>
 | |
|   %s4 = shufflevector <4 x i32> %shuf_or1, <4 x i32> undef, <4 x i32> <i32 2, i32 2, i32 3, i32 3>
 | |
|   %shuf_or2 = or <4 x i32> %s3, %s4
 | |
|   %not_or2 = xor <4 x i32> %shuf_or2, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %xbc = bitcast <4 x float> %x to <4 x i32>
 | |
|   %zbc = bitcast <4 x float> %z to <4 x i32>
 | |
|   %and1 = and <4 x i32> %not_or2, %xbc
 | |
|   %and2 = and <4 x i32> %shuf_or2, %zbc
 | |
|   %sel = or <4 x i32> %and1, %and2
 | |
|   ret <4 x i32> %sel
 | |
| }
 | |
| 
 | |
| define <4 x i32> @computesignbits_through_two_input_shuffle(<4 x i32> %x, <4 x i32> %y, <4 x i1> %cond1, <4 x i1> %cond2) {
 | |
| ; CHECK-LABEL: @computesignbits_through_two_input_shuffle(
 | |
| ; CHECK-NEXT:    [[SEXT1:%.*]] = sext <4 x i1> [[COND1:%.*]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[SEXT2:%.*]] = sext <4 x i1> [[COND2:%.*]] to <4 x i32>
 | |
| ; CHECK-NEXT:    [[COND:%.*]] = shufflevector <4 x i32> [[SEXT1]], <4 x i32> [[SEXT2]], <4 x i32> <i32 0, i32 2, i32 4, i32 6>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = trunc <4 x i32> [[COND]] to <4 x i1>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = select <4 x i1> [[TMP1]], <4 x i32> [[Y:%.*]], <4 x i32> [[X:%.*]]
 | |
| ; CHECK-NEXT:    ret <4 x i32> [[TMP2]]
 | |
| ;
 | |
|   %sext1 = sext <4 x i1> %cond1 to <4 x i32>
 | |
|   %sext2 = sext <4 x i1> %cond2 to <4 x i32>
 | |
|   %cond = shufflevector <4 x i32> %sext1, <4 x i32> %sext2, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
 | |
|   %notcond = xor <4 x i32> %cond, <i32 -1, i32 -1, i32 -1, i32 -1>
 | |
|   %and1 = and <4 x i32> %notcond, %x
 | |
|   %and2 = and <4 x i32> %cond, %y
 | |
|   %sel = or <4 x i32> %and1, %and2
 | |
|   ret <4 x i32> %sel
 | |
| }
 | |
| 
 | |
| ; Bitcast of condition from narrow source element type can be converted to select.
 | |
| 
 | |
| define <2 x i64> @bitcast_vec_cond(<16 x i1> %cond, <2 x i64> %c, <2 x i64> %d) {
 | |
| ; CHECK-LABEL: @bitcast_vec_cond(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[D:%.*]] to <16 x i8>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[C:%.*]] to <16 x i8>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <16 x i1> [[COND:%.*]], <16 x i8> [[TMP1]], <16 x i8> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <16 x i8> [[TMP3]] to <2 x i64>
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[TMP4]]
 | |
| ;
 | |
|   %s = sext <16 x i1> %cond to <16 x i8>
 | |
|   %t9 = bitcast <16 x i8> %s to <2 x i64>
 | |
|   %nott9 = xor <2 x i64> %t9, <i64 -1, i64 -1>
 | |
|   %t11 = and <2 x i64> %nott9, %c
 | |
|   %t12 = and <2 x i64> %t9, %d
 | |
|   %r = or <2 x i64> %t11, %t12
 | |
|   ret <2 x i64> %r
 | |
| }
 | |
| 
 | |
| ; Negative test - bitcast of condition from wide source element type cannot be converted to select.
 | |
| 
 | |
| define <8 x i3> @bitcast_vec_cond_commute1(<3 x i1> %cond, <8 x i3> %pc, <8 x i3> %d) {
 | |
| ; CHECK-LABEL: @bitcast_vec_cond_commute1(
 | |
| ; CHECK-NEXT:    [[C:%.*]] = mul <8 x i3> [[PC:%.*]], [[PC]]
 | |
| ; CHECK-NEXT:    [[S:%.*]] = sext <3 x i1> [[COND:%.*]] to <3 x i8>
 | |
| ; CHECK-NEXT:    [[T9:%.*]] = bitcast <3 x i8> [[S]] to <8 x i3>
 | |
| ; CHECK-NEXT:    [[NOTT9:%.*]] = xor <8 x i3> [[T9]], <i3 -1, i3 -1, i3 -1, i3 -1, i3 -1, i3 -1, i3 -1, i3 -1>
 | |
| ; CHECK-NEXT:    [[T11:%.*]] = and <8 x i3> [[C]], [[NOTT9]]
 | |
| ; CHECK-NEXT:    [[T12:%.*]] = and <8 x i3> [[T9]], [[D:%.*]]
 | |
| ; CHECK-NEXT:    [[R:%.*]] = or <8 x i3> [[T11]], [[T12]]
 | |
| ; CHECK-NEXT:    ret <8 x i3> [[R]]
 | |
| ;
 | |
|   %c = mul <8 x i3> %pc, %pc ; thwart complexity-based canonicalization
 | |
|   %s = sext <3 x i1> %cond to <3 x i8>
 | |
|   %t9 = bitcast <3 x i8> %s to <8 x i3>
 | |
|   %nott9 = xor <8 x i3> %t9, <i3 -1, i3 -1, i3 -1, i3 -1, i3 -1, i3 -1, i3 -1, i3 -1>
 | |
|   %t11 = and <8 x i3> %c, %nott9
 | |
|   %t12 = and <8 x i3> %t9, %d
 | |
|   %r = or <8 x i3> %t11, %t12
 | |
|   ret <8 x i3> %r
 | |
| }
 | |
| 
 | |
| define <2 x i16> @bitcast_vec_cond_commute2(<4 x i1> %cond, <2 x i16> %pc, <2 x i16> %pd) {
 | |
| ; CHECK-LABEL: @bitcast_vec_cond_commute2(
 | |
| ; CHECK-NEXT:    [[C:%.*]] = mul <2 x i16> [[PC:%.*]], [[PC]]
 | |
| ; CHECK-NEXT:    [[D:%.*]] = mul <2 x i16> [[PD:%.*]], [[PD]]
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i16> [[D]] to <4 x i8>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i16> [[C]] to <4 x i8>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[COND:%.*]], <4 x i8> [[TMP1]], <4 x i8> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i8> [[TMP3]] to <2 x i16>
 | |
| ; CHECK-NEXT:    ret <2 x i16> [[TMP4]]
 | |
| ;
 | |
|   %c = mul <2 x i16> %pc, %pc ; thwart complexity-based canonicalization
 | |
|   %d = mul <2 x i16> %pd, %pd ; thwart complexity-based canonicalization
 | |
|   %s = sext <4 x i1> %cond to <4 x i8>
 | |
|   %t9 = bitcast <4 x i8> %s to <2 x i16>
 | |
|   %nott9 = xor <2 x i16> %t9, <i16 -1, i16 -1>
 | |
|   %t11 = and <2 x i16> %c, %nott9
 | |
|   %t12 = and <2 x i16> %d, %t9
 | |
|   %r = or <2 x i16> %t11, %t12
 | |
|   ret <2 x i16> %r
 | |
| }
 | |
| 
 | |
| ; Condition doesn't have to be a bool vec - just all signbits.
 | |
| 
 | |
| define <2 x i16> @bitcast_vec_cond_commute3(<4 x i8> %cond, <2 x i16> %pc, <2 x i16> %pd) {
 | |
| ; CHECK-LABEL: @bitcast_vec_cond_commute3(
 | |
| ; CHECK-NEXT:    [[C:%.*]] = mul <2 x i16> [[PC:%.*]], [[PC]]
 | |
| ; CHECK-NEXT:    [[D:%.*]] = mul <2 x i16> [[PD:%.*]], [[PD]]
 | |
| ; CHECK-NEXT:    [[DOTNOT:%.*]] = icmp sgt <4 x i8> [[COND:%.*]], <i8 -1, i8 -1, i8 -1, i8 -1>
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i16> [[D]] to <4 x i8>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i16> [[C]] to <4 x i8>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <4 x i1> [[DOTNOT]], <4 x i8> [[TMP2]], <4 x i8> [[TMP1]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i8> [[TMP3]] to <2 x i16>
 | |
| ; CHECK-NEXT:    ret <2 x i16> [[TMP4]]
 | |
| ;
 | |
|   %c = mul <2 x i16> %pc, %pc ; thwart complexity-based canonicalization
 | |
|   %d = mul <2 x i16> %pd, %pd ; thwart complexity-based canonicalization
 | |
|   %s = ashr <4 x i8> %cond, <i8 7, i8 7, i8 7, i8 7>
 | |
|   %t9 = bitcast <4 x i8> %s to <2 x i16>
 | |
|   %nott9 = xor <2 x i16> %t9, <i16 -1, i16 -1>
 | |
|   %t11 = and <2 x i16> %c, %nott9
 | |
|   %t12 = and <2 x i16> %d, %t9
 | |
|   %r = or <2 x i16> %t11, %t12
 | |
|   ret <2 x i16> %r
 | |
| }
 | |
| 
 | |
| ; Don't crash on invalid type for compute signbits.
 | |
| 
 | |
| define <2 x i64> @bitcast_fp_vec_cond(<2 x double> %s, <2 x i64> %c, <2 x i64> %d) {
 | |
| ; CHECK-LABEL: @bitcast_fp_vec_cond(
 | |
| ; CHECK-NEXT:    [[T9:%.*]] = bitcast <2 x double> [[S:%.*]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[NOTT9:%.*]] = xor <2 x i64> [[T9]], <i64 -1, i64 -1>
 | |
| ; CHECK-NEXT:    [[T11:%.*]] = and <2 x i64> [[NOTT9]], [[C:%.*]]
 | |
| ; CHECK-NEXT:    [[T12:%.*]] = and <2 x i64> [[T9]], [[D:%.*]]
 | |
| ; CHECK-NEXT:    [[R:%.*]] = or <2 x i64> [[T11]], [[T12]]
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[R]]
 | |
| ;
 | |
|   %t9 = bitcast <2 x double> %s to <2 x i64>
 | |
|   %nott9 = xor <2 x i64> %t9, <i64 -1, i64 -1>
 | |
|   %t11 = and <2 x i64> %nott9, %c
 | |
|   %t12 = and <2 x i64> %t9, %d
 | |
|   %r = or <2 x i64> %t11, %t12
 | |
|   ret <2 x i64> %r
 | |
| }
 | |
| 
 | |
| ; Wider source type would be ok except poison could leak across elements.
 | |
| 
 | |
| define <2 x i64> @bitcast_int_vec_cond(i1 %b, <2 x i64> %c, <2 x i64> %d) {
 | |
| ; CHECK-LABEL: @bitcast_int_vec_cond(
 | |
| ; CHECK-NEXT:    [[S:%.*]] = sext i1 [[B:%.*]] to i128
 | |
| ; CHECK-NEXT:    [[T9:%.*]] = bitcast i128 [[S]] to <2 x i64>
 | |
| ; CHECK-NEXT:    [[NOTT9:%.*]] = xor <2 x i64> [[T9]], <i64 -1, i64 -1>
 | |
| ; CHECK-NEXT:    [[T11:%.*]] = and <2 x i64> [[NOTT9]], [[C:%.*]]
 | |
| ; CHECK-NEXT:    [[T12:%.*]] = and <2 x i64> [[T9]], [[D:%.*]]
 | |
| ; CHECK-NEXT:    [[R:%.*]] = or <2 x i64> [[T11]], [[T12]]
 | |
| ; CHECK-NEXT:    ret <2 x i64> [[R]]
 | |
| ;
 | |
|   %s = sext i1 %b to i128
 | |
|   %t9 = bitcast i128 %s to <2 x i64>
 | |
|   %nott9 = xor <2 x i64> %t9, <i64 -1, i64 -1>
 | |
|   %t11 = and <2 x i64> %nott9, %c
 | |
|   %t12 = and <2 x i64> %t9, %d
 | |
|   %r = or <2 x i64> %t11, %t12
 | |
|   ret <2 x i64> %r
 | |
| }
 | |
| 
 | |
| ; Converting integer logic ops to vector select is allowed.
 | |
| 
 | |
| define i64 @bitcast_int_scalar_cond(<2 x i1> %b, i64 %c, i64 %d) {
 | |
| ; CHECK-LABEL: @bitcast_int_scalar_cond(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i64 [[D:%.*]] to <2 x i32>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i64 [[C:%.*]] to <2 x i32>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <2 x i1> [[B:%.*]], <2 x i32> [[TMP1]], <2 x i32> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP3]] to i64
 | |
| ; CHECK-NEXT:    ret i64 [[TMP4]]
 | |
| ;
 | |
|   %s = sext <2 x i1> %b to <2 x i32>
 | |
|   %t9 = bitcast <2 x i32> %s to i64
 | |
|   %nott9 = xor i64 %t9, -1
 | |
|   %t11 = and i64 %nott9, %c
 | |
|   %t12 = and i64 %t9, %d
 | |
|   %r = or i64 %t11, %t12
 | |
|   ret i64 %r
 | |
| }
 | |
| 
 | |
| ; Peek through bitcasts and sexts to find negated bool condition.
 | |
| 
 | |
| define <1 x i6> @bitcast_sext_cond(<2 x i1> %cmp, <1 x i6> %a, <1 x i6> %b) {
 | |
| ; CHECK-LABEL: @bitcast_sext_cond(
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i6> [[A:%.*]] to <2 x i3>
 | |
| ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <1 x i6> [[B:%.*]] to <2 x i3>
 | |
| ; CHECK-NEXT:    [[TMP3:%.*]] = select <2 x i1> [[CMP:%.*]], <2 x i3> [[TMP1]], <2 x i3> [[TMP2]]
 | |
| ; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i3> [[TMP3]] to <1 x i6>
 | |
| ; CHECK-NEXT:    ret <1 x i6> [[TMP4]]
 | |
| ;
 | |
|   %sext = sext <2 x i1> %cmp to <2 x i3>
 | |
|   %bc1 = bitcast <2 x i3> %sext to <1 x i6>
 | |
|   %neg = xor <2 x i1> %cmp, <i1 -1, i1 -1>
 | |
|   %sext2 = sext <2 x i1> %neg to <2 x i3>
 | |
|   %bc2 = bitcast <2 x i3> %sext2 to <1 x i6>
 | |
|   %and1 = and <1 x i6> %bc1, %a
 | |
|   %and2 = and <1 x i6> %bc2, %b
 | |
|   %or = or <1 x i6> %and1, %and2
 | |
|   ret <1 x i6> %or
 | |
| }
 | |
| 
 | |
| ; Extra uses may prevent other transforms from creating the canonical patterns.
 | |
| 
 | |
| define i8 @sext_cond_extra_uses(i1 %cmp, i8 %a, i8 %b) {
 | |
| ; CHECK-LABEL: @sext_cond_extra_uses(
 | |
| ; CHECK-NEXT:    [[NEG:%.*]] = xor i1 [[CMP:%.*]], true
 | |
| ; CHECK-NEXT:    [[SEXT1:%.*]] = sext i1 [[CMP]] to i8
 | |
| ; CHECK-NEXT:    call void @use(i8 [[SEXT1]])
 | |
| ; CHECK-NEXT:    [[SEXT2:%.*]] = sext i1 [[NEG]] to i8
 | |
| ; CHECK-NEXT:    call void @use(i8 [[SEXT2]])
 | |
| ; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[CMP]], i8 [[A:%.*]], i8 [[B:%.*]]
 | |
| ; CHECK-NEXT:    ret i8 [[TMP1]]
 | |
| ;
 | |
|   %neg = xor i1 %cmp, -1
 | |
|   %sext1 = sext i1 %cmp to i8
 | |
|   call void @use(i8 %sext1)
 | |
|   %sext2 = sext i1 %neg to i8
 | |
|   call void @use(i8 %sext2)
 | |
|   %and1 = and i8 %sext1, %a
 | |
|   %and2 = and i8 %sext2, %b
 | |
|   %or = or i8 %and1, %and2
 | |
|   ret i8 %or
 | |
| }
 |