diff options
Diffstat (limited to 'llvm/test/Transforms')
7 files changed, 291 insertions, 144 deletions
diff --git a/llvm/test/Transforms/InstCombine/clamp-to-minmax.ll b/llvm/test/Transforms/InstCombine/clamp-to-minmax.ll index 7f32766..0ccaa9c 100644 --- a/llvm/test/Transforms/InstCombine/clamp-to-minmax.ll +++ b/llvm/test/Transforms/InstCombine/clamp-to-minmax.ll @@ -172,10 +172,8 @@ define float @clamp_negative_wrong_const(float %x) { ; Like @clamp_test_1 but both are min define float @clamp_negative_same_op(float %x) { ; CHECK-LABEL: @clamp_negative_same_op( -; CHECK-NEXT: [[INNER_CMP_INV:%.*]] = fcmp fast oge float [[X:%.*]], 2.550000e+02 -; CHECK-NEXT: [[INNER_SEL:%.*]] = select nnan ninf i1 [[INNER_CMP_INV]], float 2.550000e+02, float [[X]] -; CHECK-NEXT: [[OUTER_CMP:%.*]] = fcmp fast ult float [[X]], 1.000000e+00 -; CHECK-NEXT: [[R:%.*]] = select i1 [[OUTER_CMP]], float [[INNER_SEL]], float 1.000000e+00 +; CHECK-NEXT: [[OUTER_CMP_INV:%.*]] = fcmp fast oge float [[X:%.*]], 1.000000e+00 +; CHECK-NEXT: [[R:%.*]] = select nnan ninf i1 [[OUTER_CMP_INV]], float 1.000000e+00, float [[X]] ; CHECK-NEXT: ret float [[R]] ; %inner_cmp = fcmp fast ult float %x, 255.0 diff --git a/llvm/test/Transforms/InstCombine/select-gep.ll b/llvm/test/Transforms/InstCombine/select-gep.ll index dd8dffb..7181336 100644 --- a/llvm/test/Transforms/InstCombine/select-gep.ll +++ b/llvm/test/Transforms/InstCombine/select-gep.ll @@ -286,3 +286,35 @@ define <2 x ptr> @test7(<2 x ptr> %p1, i64 %idx, <2 x i1> %cc) { %select = select <2 x i1> %cc, <2 x ptr> %p1, <2 x ptr> %gep ret <2 x ptr> %select } + +define ptr @ptr_eq_replace_freeze1(ptr %p, ptr %q) { +; CHECK-LABEL: @ptr_eq_replace_freeze1( +; CHECK-NEXT: [[Q_FR:%.*]] = freeze ptr [[Q:%.*]] +; CHECK-NEXT: [[Q_FR1:%.*]] = freeze ptr [[Q1:%.*]] +; CHECK-NEXT: [[CMP:%.*]] = icmp eq ptr [[Q_FR]], [[Q_FR1]] +; CHECK-NEXT: [[SELECT:%.*]] = select i1 [[CMP]], ptr [[Q_FR]], ptr [[Q_FR1]] +; CHECK-NEXT: ret ptr [[SELECT]] +; + %p.fr = freeze ptr %p + %q.fr = freeze ptr %q + %cmp = icmp eq ptr %p.fr, %q.fr + %select = select i1 %cmp, ptr %p.fr, ptr %q.fr + ret ptr %select +} + +define ptr @ptr_eq_replace_freeze2(ptr %p, ptr %q) { +; CHECK-LABEL: @ptr_eq_replace_freeze2( +; CHECK-NEXT: [[P_FR:%.*]] = freeze ptr [[P:%.*]] +; CHECK-NEXT: [[P_FR1:%.*]] = freeze ptr [[P1:%.*]] +; CHECK-NEXT: [[CMP:%.*]] = icmp eq ptr [[P_FR1]], [[P_FR]] +; CHECK-NEXT: [[SELECT_V:%.*]] = select i1 [[CMP]], ptr [[P_FR1]], ptr [[P_FR]] +; CHECK-NEXT: [[SELECT:%.*]] = getelementptr i8, ptr [[SELECT_V]], i64 16 +; CHECK-NEXT: ret ptr [[SELECT]] +; + %gep1 = getelementptr i32, ptr %p, i64 4 + %gep2 = getelementptr i32, ptr %q, i64 4 + %cmp = icmp eq ptr %p, %q + %cmp.fr = freeze i1 %cmp + %select = select i1 %cmp.fr, ptr %gep1, ptr %gep2 + ret ptr %select +} diff --git a/llvm/test/Transforms/InstSimplify/domcondition.ll b/llvm/test/Transforms/InstSimplify/domcondition.ll index 43be5de..2893bb1 100644 --- a/llvm/test/Transforms/InstSimplify/domcondition.ll +++ b/llvm/test/Transforms/InstSimplify/domcondition.ll @@ -278,3 +278,210 @@ end: } declare void @foo(i32) + + +define i1 @simplify_fcmp_implied_by_dom_cond_range_true(float %x) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_range_true( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], 0.000000e+00 +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: ret i1 true +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, 0.0 + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp olt float %x, 1.0 + ret i1 %cmp2 + +if.else: + ret i1 false +} + +define i1 @simplify_fcmp_in_else_implied_by_dom_cond_range_true(float %x) { +; CHECK-LABEL: @simplify_fcmp_in_else_implied_by_dom_cond_range_true( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], 1.000000e+00 +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: ret i1 true +; CHECK: if.else: +; CHECK-NEXT: ret i1 true +; + %cmp = fcmp olt float %x, 1.0 + br i1 %cmp, label %if.then, label %if.else + +if.then: + ret i1 true + +if.else: + %cmp2 = fcmp uge float %x, 0.5 + ret i1 %cmp2 +} + +define i1 @simplify_fcmp_implied_by_dom_cond_range_false(float %x) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_range_false( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], 0.000000e+00 +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: ret i1 false +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, 0.0 + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp ogt float %x, 1.0 + ret i1 %cmp2 + +if.else: + ret i1 false +} + +define i1 @simplify_fcmp_implied_by_dom_cond_pred_true(float %x, float %y) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_pred_true( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], [[Y:%.*]] +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: ret i1 true +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, %y + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp ole float %x, %y + ret i1 %cmp2 + +if.else: + ret i1 false +} + +define i1 @simplify_fcmp_implied_by_dom_cond_pred_false(float %x, float %y) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_pred_false( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], [[Y:%.*]] +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: ret i1 false +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, %y + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp ogt float %x, %y + ret i1 %cmp2 + +if.else: + ret i1 false +} + +define i1 @simplify_fcmp_implied_by_dom_cond_pred_commuted(float %x, float %y) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_pred_commuted( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], [[Y:%.*]] +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: ret i1 true +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, %y + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp oge float %y, %x + ret i1 %cmp2 + +if.else: + ret i1 false +} + +; Negative tests + +define i1 @simplify_fcmp_implied_by_dom_cond_wrong_range(float %x) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_wrong_range( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], 0.000000e+00 +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: [[CMP2:%.*]] = fcmp olt float [[X]], -1.000000e+00 +; CHECK-NEXT: ret i1 [[CMP2]] +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, 0.0 + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp olt float %x, -1.0 + ret i1 %cmp2 + +if.else: + ret i1 false +} + +define i1 @simplify_fcmp_implied_by_dom_cond_range_mismatched_operand(float %x, float %y) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_range_mismatched_operand( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], 0.000000e+00 +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: [[CMP2:%.*]] = fcmp olt float [[Y:%.*]], 1.000000e+00 +; CHECK-NEXT: ret i1 [[CMP2]] +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, 0.0 + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp olt float %y, 1.0 + ret i1 %cmp2 + +if.else: + ret i1 false +} + +define i1 @simplify_fcmp_implied_by_dom_cond_wrong_pred(float %x, float %y) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_wrong_pred( +; CHECK-NEXT: [[CMP:%.*]] = fcmp ole float [[X:%.*]], [[Y:%.*]] +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: [[CMP2:%.*]] = fcmp olt float [[X]], [[Y]] +; CHECK-NEXT: ret i1 [[CMP2]] +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp ole float %x, %y + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp olt float %x, %y + ret i1 %cmp2 + +if.else: + ret i1 false +} + +define i1 @simplify_fcmp_implied_by_dom_cond_pred_mismatched_operand(float %x, float %y, float %z) { +; CHECK-LABEL: @simplify_fcmp_implied_by_dom_cond_pred_mismatched_operand( +; CHECK-NEXT: [[CMP:%.*]] = fcmp olt float [[X:%.*]], [[Y:%.*]] +; CHECK-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]] +; CHECK: if.then: +; CHECK-NEXT: [[CMP2:%.*]] = fcmp ole float [[X]], [[Z:%.*]] +; CHECK-NEXT: ret i1 [[CMP2]] +; CHECK: if.else: +; CHECK-NEXT: ret i1 false +; + %cmp = fcmp olt float %x, %y + br i1 %cmp, label %if.then, label %if.else + +if.then: + %cmp2 = fcmp ole float %x, %z + ret i1 %cmp2 + +if.else: + ret i1 false +} diff --git a/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll b/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll index f5329cf..8784873 100644 --- a/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll +++ b/llvm/test/Transforms/LoopVectorize/X86/replicating-load-store-costs.ll @@ -454,132 +454,6 @@ exit: ret void } -declare i1 @cond() - -define double @test_load_used_by_other_load_scev(ptr %ptr.a, ptr %ptr.b, ptr %ptr.c) { -; I64-LABEL: define double @test_load_used_by_other_load_scev( -; I64-SAME: ptr [[PTR_A:%.*]], ptr [[PTR_B:%.*]], ptr [[PTR_C:%.*]]) { -; I64-NEXT: [[ENTRY:.*]]: -; I64-NEXT: br label %[[OUTER_LOOP:.*]] -; I64: [[OUTER_LOOP_LOOPEXIT:.*]]: -; I64-NEXT: br label %[[OUTER_LOOP]] -; I64: [[OUTER_LOOP]]: -; I64-NEXT: [[ACCUM:%.*]] = phi double [ 0.000000e+00, %[[ENTRY]] ], [ [[TMP29:%.*]], %[[OUTER_LOOP_LOOPEXIT]] ] -; I64-NEXT: [[COND:%.*]] = call i1 @cond() -; I64-NEXT: br i1 [[COND]], label %[[INNER_LOOP_PREHEADER:.*]], label %[[EXIT:.*]] -; I64: [[INNER_LOOP_PREHEADER]]: -; I64-NEXT: br label %[[VECTOR_PH:.*]] -; I64: [[VECTOR_PH]]: -; I64-NEXT: br label %[[VECTOR_BODY:.*]] -; I64: [[VECTOR_BODY]]: -; I64-NEXT: [[TMP0:%.*]] = add i64 0, 1 -; I64-NEXT: [[TMP1:%.*]] = add i64 1, 1 -; I64-NEXT: [[TMP2:%.*]] = getelementptr i8, ptr [[PTR_C]], i64 [[TMP0]] -; I64-NEXT: [[TMP3:%.*]] = getelementptr i8, ptr [[PTR_C]], i64 [[TMP1]] -; I64-NEXT: [[TMP4:%.*]] = getelementptr i64, ptr [[PTR_A]], i64 [[TMP0]] -; I64-NEXT: [[TMP5:%.*]] = getelementptr i64, ptr [[PTR_A]], i64 [[TMP1]] -; I64-NEXT: [[TMP6:%.*]] = load i64, ptr [[TMP4]], align 8 -; I64-NEXT: [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8 -; I64-NEXT: [[TMP8:%.*]] = getelementptr double, ptr [[PTR_B]], i64 [[TMP6]] -; I64-NEXT: [[TMP9:%.*]] = getelementptr double, ptr [[PTR_B]], i64 [[TMP7]] -; I64-NEXT: [[TMP10:%.*]] = load double, ptr [[PTR_A]], align 8 -; I64-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x double> poison, double [[TMP10]], i64 0 -; I64-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x double> [[BROADCAST_SPLATINSERT]], <2 x double> poison, <2 x i32> zeroinitializer -; I64-NEXT: [[TMP11:%.*]] = fadd <2 x double> [[BROADCAST_SPLAT]], zeroinitializer -; I64-NEXT: [[TMP12:%.*]] = getelementptr i8, ptr [[TMP2]], i64 8 -; I64-NEXT: [[TMP13:%.*]] = getelementptr i8, ptr [[TMP3]], i64 8 -; I64-NEXT: [[TMP14:%.*]] = load double, ptr [[TMP12]], align 8 -; I64-NEXT: [[TMP15:%.*]] = load double, ptr [[TMP13]], align 8 -; I64-NEXT: [[TMP16:%.*]] = insertelement <2 x double> poison, double [[TMP14]], i32 0 -; I64-NEXT: [[TMP17:%.*]] = insertelement <2 x double> [[TMP16]], double [[TMP15]], i32 1 -; I64-NEXT: [[TMP18:%.*]] = fmul <2 x double> [[TMP11]], zeroinitializer -; I64-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <2 x double> poison, double [[ACCUM]], i64 0 -; I64-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <2 x double> [[BROADCAST_SPLATINSERT1]], <2 x double> poison, <2 x i32> zeroinitializer -; I64-NEXT: [[TMP19:%.*]] = shufflevector <2 x double> [[BROADCAST_SPLAT2]], <2 x double> [[TMP18]], <2 x i32> <i32 1, i32 2> -; I64-NEXT: [[TMP20:%.*]] = fmul <2 x double> [[TMP17]], zeroinitializer -; I64-NEXT: [[TMP21:%.*]] = fadd <2 x double> [[TMP20]], zeroinitializer -; I64-NEXT: [[TMP22:%.*]] = fadd <2 x double> [[TMP21]], splat (double 1.000000e+00) -; I64-NEXT: [[TMP23:%.*]] = load double, ptr [[TMP8]], align 8 -; I64-NEXT: [[TMP24:%.*]] = load double, ptr [[TMP9]], align 8 -; I64-NEXT: [[TMP25:%.*]] = insertelement <2 x double> poison, double [[TMP23]], i32 0 -; I64-NEXT: [[TMP26:%.*]] = insertelement <2 x double> [[TMP25]], double [[TMP24]], i32 1 -; I64-NEXT: [[TMP27:%.*]] = fdiv <2 x double> [[TMP26]], [[TMP22]] -; I64-NEXT: [[TMP28:%.*]] = fsub <2 x double> [[TMP19]], [[TMP27]] -; I64-NEXT: br label %[[MIDDLE_BLOCK:.*]] -; I64: [[MIDDLE_BLOCK]]: -; I64-NEXT: [[TMP29]] = extractelement <2 x double> [[TMP28]], i32 1 -; I64-NEXT: br label %[[OUTER_LOOP_LOOPEXIT]] -; I64: [[EXIT]]: -; I64-NEXT: ret double [[ACCUM]] -; -; I32-LABEL: define double @test_load_used_by_other_load_scev( -; I32-SAME: ptr [[PTR_A:%.*]], ptr [[PTR_B:%.*]], ptr [[PTR_C:%.*]]) { -; I32-NEXT: [[ENTRY:.*]]: -; I32-NEXT: br label %[[OUTER_LOOP:.*]] -; I32: [[OUTER_LOOP]]: -; I32-NEXT: [[ACCUM:%.*]] = phi double [ 0.000000e+00, %[[ENTRY]] ], [ [[RESULT:%.*]], %[[INNER_LOOP:.*]] ] -; I32-NEXT: [[COND:%.*]] = call i1 @cond() -; I32-NEXT: br i1 [[COND]], label %[[INNER_LOOP]], label %[[EXIT:.*]] -; I32: [[INNER_LOOP]]: -; I32-NEXT: [[IV:%.*]] = phi i64 [ 0, %[[OUTER_LOOP]] ], [ [[IV_NEXT:%.*]], %[[INNER_LOOP]] ] -; I32-NEXT: [[ACCUM_INNER:%.*]] = phi double [ [[ACCUM]], %[[OUTER_LOOP]] ], [ [[MUL1:%.*]], %[[INNER_LOOP]] ] -; I32-NEXT: [[IDX_PLUS1:%.*]] = add i64 [[IV]], 1 -; I32-NEXT: [[GEP_C:%.*]] = getelementptr i8, ptr [[PTR_C]], i64 [[IDX_PLUS1]] -; I32-NEXT: [[GEP_A_I64:%.*]] = getelementptr i64, ptr [[PTR_A]], i64 [[IDX_PLUS1]] -; I32-NEXT: [[LOAD_IDX:%.*]] = load i64, ptr [[GEP_A_I64]], align 8 -; I32-NEXT: [[GEP_B:%.*]] = getelementptr double, ptr [[PTR_B]], i64 [[LOAD_IDX]] -; I32-NEXT: [[LOAD_A:%.*]] = load double, ptr [[PTR_A]], align 8 -; I32-NEXT: [[ADD1:%.*]] = fadd double [[LOAD_A]], 0.000000e+00 -; I32-NEXT: [[GEP_C_OFFSET:%.*]] = getelementptr i8, ptr [[GEP_C]], i64 8 -; I32-NEXT: [[LOAD_C:%.*]] = load double, ptr [[GEP_C_OFFSET]], align 8 -; I32-NEXT: [[MUL1]] = fmul double [[ADD1]], 0.000000e+00 -; I32-NEXT: [[MUL2:%.*]] = fmul double [[LOAD_C]], 0.000000e+00 -; I32-NEXT: [[ADD2:%.*]] = fadd double [[MUL2]], 0.000000e+00 -; I32-NEXT: [[ADD3:%.*]] = fadd double [[ADD2]], 1.000000e+00 -; I32-NEXT: [[LOAD_B:%.*]] = load double, ptr [[GEP_B]], align 8 -; I32-NEXT: [[DIV:%.*]] = fdiv double [[LOAD_B]], [[ADD3]] -; I32-NEXT: [[RESULT]] = fsub double [[ACCUM_INNER]], [[DIV]] -; I32-NEXT: [[IV_NEXT]] = add i64 [[IV]], 1 -; I32-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[IV]], 1 -; I32-NEXT: br i1 [[EXITCOND]], label %[[OUTER_LOOP]], label %[[INNER_LOOP]] -; I32: [[EXIT]]: -; I32-NEXT: ret double [[ACCUM]] -; -entry: - br label %outer.loop - -outer.loop: - %accum = phi double [ 0.0, %entry ], [ %result, %inner.loop ] - %cond = call i1 @cond() - br i1 %cond, label %inner.loop, label %exit - -inner.loop: - %iv = phi i64 [ 0, %outer.loop ], [ %iv.next, %inner.loop ] - %accum.inner = phi double [ %accum, %outer.loop ], [ %mul1, %inner.loop ] - %idx.plus1 = add i64 %iv, 1 - %gep.c = getelementptr i8, ptr %ptr.c, i64 %idx.plus1 - %gep.a.i64 = getelementptr i64, ptr %ptr.a, i64 %idx.plus1 - %load.idx = load i64, ptr %gep.a.i64, align 8 - %gep.b = getelementptr double, ptr %ptr.b, i64 %load.idx - %load.a = load double, ptr %ptr.a, align 8 - %add1 = fadd double %load.a, 0.000000e+00 - %gep.c.offset = getelementptr i8, ptr %gep.c, i64 8 - %load.c = load double, ptr %gep.c.offset, align 8 - %mul1 = fmul double %add1, 0.000000e+00 - %mul2 = fmul double %load.c, 0.000000e+00 - %add2 = fadd double %mul2, 0.000000e+00 - %add3 = fadd double %add2, 1.000000e+00 - %load.b = load double, ptr %gep.b, align 8 - %div = fdiv double %load.b, %add3 - %result = fsub double %accum.inner, %div - %iv.next = add i64 %iv, 1 - %exitcond = icmp eq i64 %iv, 1 - br i1 %exitcond, label %outer.loop, label %inner.loop - -exit: - ret double %accum -} - attributes #0 = { "target-cpu"="znver2" } !0 = distinct !{!0, !1} diff --git a/llvm/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll b/llvm/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll index f9e79ca..bee6b37 100644 --- a/llvm/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll +++ b/llvm/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll @@ -1565,14 +1565,14 @@ end: ; lookup (since i3 can only hold values in the range of explicit ; values) and simultaneously trying to generate a branch to deal with ; the fact that we have holes in the range. -define i32 @covered_switch_with_bit_tests(i3) { +define i32 @covered_switch_with_bit_tests(i3) !prof !0 { ; CHECK-LABEL: @covered_switch_with_bit_tests( ; CHECK-NEXT: entry: ; CHECK-NEXT: [[SWITCH_TABLEIDX:%.*]] = sub i3 [[TMP0:%.*]], -4 ; CHECK-NEXT: [[SWITCH_MASKINDEX:%.*]] = zext i3 [[SWITCH_TABLEIDX]] to i8 ; CHECK-NEXT: [[SWITCH_SHIFTED:%.*]] = lshr i8 -61, [[SWITCH_MASKINDEX]] ; CHECK-NEXT: [[SWITCH_LOBIT:%.*]] = trunc i8 [[SWITCH_SHIFTED]] to i1 -; CHECK-NEXT: br i1 [[SWITCH_LOBIT]], label [[SWITCH_LOOKUP:%.*]], label [[L6:%.*]] +; CHECK-NEXT: br i1 [[SWITCH_LOBIT]], label [[SWITCH_LOOKUP:%.*]], label [[L6:%.*]], !prof [[PROF1:![0-9]+]] ; CHECK: switch.lookup: ; CHECK-NEXT: [[TMP1:%.*]] = zext i3 [[SWITCH_TABLEIDX]] to i64 ; CHECK-NEXT: [[SWITCH_GEP:%.*]] = getelementptr inbounds [8 x i32], ptr @switch.table.covered_switch_with_bit_tests, i64 0, i64 [[TMP1]] @@ -1588,7 +1588,7 @@ entry: i3 -4, label %l5 i3 3, label %l1 i3 2, label %l1 - ] + ], !prof !1 l1: br label %l2 @@ -2425,3 +2425,10 @@ return: %res = phi i1 [ 0, %bb0 ], [ 1, %bb1 ] ret i1 %res } + +!0 = !{!"function_entry_count", i32 10} +!1 = !{!"branch_weights", i32 3, i32 5, i32 7, i32 11, i32 13} +;. +; CHECK: [[META0:![0-9]+]] = !{!"function_entry_count", i32 10} +; CHECK: [[PROF1]] = !{!"branch_weights", i32 36, i32 3} +;. diff --git a/llvm/test/Transforms/SimplifyCFG/rangereduce.ll b/llvm/test/Transforms/SimplifyCFG/rangereduce.ll index 17d65a4..d1fba91 100644 --- a/llvm/test/Transforms/SimplifyCFG/rangereduce.ll +++ b/llvm/test/Transforms/SimplifyCFG/rangereduce.ll @@ -1,15 +1,22 @@ -; NOTE: Assertions have been autogenerated by utils/update_test_checks.py +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals ; RUN: opt < %s -passes=simplifycfg -simplifycfg-require-and-preserve-domtree=1 -switch-to-lookup -S | FileCheck %s ; RUN: opt < %s -passes='simplifycfg<switch-to-lookup>' -S | FileCheck %s target datalayout = "e-n32" -define i32 @test1(i32 %a) { +;. +; CHECK: @switch.table.test1 = private unnamed_addr constant [4 x i32] [i32 11984, i32 1143, i32 99783, i32 99783], align 4 +; CHECK: @switch.table.test3 = private unnamed_addr constant [3 x i32] [i32 11984, i32 1143, i32 99783], align 4 +; CHECK: @switch.table.test6 = private unnamed_addr constant [4 x i32] [i32 99783, i32 99783, i32 1143, i32 11984], align 4 +; CHECK: @switch.table.test8 = private unnamed_addr constant [5 x i32] [i32 11984, i32 1143, i32 99783, i32 8867, i32 99783], align 4 +; CHECK: @switch.table.test9 = private unnamed_addr constant [8 x i32] [i32 99783, i32 8867, i32 99783, i32 8867, i32 8867, i32 8867, i32 11984, i32 1143], align 4 +;. +define i32 @test1(i32 %a) !prof !0 { ; CHECK-LABEL: @test1( ; CHECK-NEXT: [[TMP1:%.*]] = sub i32 [[A:%.*]], 97 ; CHECK-NEXT: [[TMP2:%.*]] = call i32 @llvm.fshl.i32(i32 [[TMP1]], i32 [[TMP1]], i32 30) ; CHECK-NEXT: [[TMP3:%.*]] = icmp ult i32 [[TMP2]], 4 -; CHECK-NEXT: br i1 [[TMP3]], label [[SWITCH_LOOKUP:%.*]], label [[COMMON_RET:%.*]] +; CHECK-NEXT: br i1 [[TMP3]], label [[SWITCH_LOOKUP:%.*]], label [[COMMON_RET:%.*]], !prof [[PROF1:![0-9]+]] ; CHECK: switch.lookup: ; CHECK-NEXT: [[TMP4:%.*]] = zext nneg i32 [[TMP2]] to i64 ; CHECK-NEXT: [[SWITCH_GEP:%.*]] = getelementptr inbounds [4 x i32], ptr @switch.table.test1, i64 0, i64 [[TMP4]] @@ -24,7 +31,7 @@ define i32 @test1(i32 %a) { i32 101, label %two i32 105, label %three i32 109, label %three - ] + ], !prof !1 def: ret i32 8867 @@ -310,3 +317,12 @@ three: ret i32 99783 } +!0 = !{!"function_entry_count", i32 100} +!1 = !{!"branch_weights", i32 5, i32 7, i32 11, i32 13, i32 17} +;. +; CHECK: attributes #[[ATTR0:[0-9]+]] = { optsize } +; CHECK: attributes #[[ATTR1:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) } +;. +; CHECK: [[META0:![0-9]+]] = !{!"function_entry_count", i32 100} +; CHECK: [[PROF1]] = !{!"branch_weights", i32 48, i32 5} +;. diff --git a/llvm/test/Transforms/SimplifyCFG/switch_create.ll b/llvm/test/Transforms/SimplifyCFG/switch_create.ll index 18c4ade..ef5aee6 100644 --- a/llvm/test/Transforms/SimplifyCFG/switch_create.ll +++ b/llvm/test/Transforms/SimplifyCFG/switch_create.ll @@ -1,4 +1,4 @@ -; NOTE: Assertions have been autogenerated by utils/update_test_checks.py +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals ; RUN: opt -S -passes=simplifycfg -simplifycfg-require-and-preserve-domtree=1 -switch-range-to-icmp < %s | FileCheck %s ; RUN: opt -S -data-layout="p:32:32-p1:16:16" -passes=simplifycfg -simplifycfg-require-and-preserve-domtree=1 -switch-range-to-icmp < %s | FileCheck -check-prefix=CHECK -check-prefix=DL %s @@ -6,12 +6,12 @@ declare void @foo1() declare void @foo2() -define void @test1(i32 %V) { +define void @test1(i32 %V) !prof !0 { ; CHECK-LABEL: @test1( ; CHECK-NEXT: switch i32 [[V:%.*]], label [[F:%.*]] [ ; CHECK-NEXT: i32 17, label [[T:%.*]] ; CHECK-NEXT: i32 4, label [[T]] -; CHECK-NEXT: ] +; CHECK-NEXT: ], !prof [[PROF1:![0-9]+]] ; CHECK: common.ret: ; CHECK-NEXT: ret void ; CHECK: T: @@ -24,7 +24,7 @@ define void @test1(i32 %V) { %C1 = icmp eq i32 %V, 4 ; <i1> [#uses=1] %C2 = icmp eq i32 %V, 17 ; <i1> [#uses=1] %CN = or i1 %C1, %C2 ; <i1> [#uses=1] - br i1 %CN, label %T, label %F + br i1 %CN, label %T, label %F, !prof !1 T: ; preds = %0 call void @foo1( ) ret void @@ -116,12 +116,12 @@ F: ; preds = %0 ret void } -define void @test2(i32 %V) { +define void @test2(i32 %V) !prof !0 { ; CHECK-LABEL: @test2( ; CHECK-NEXT: switch i32 [[V:%.*]], label [[T:%.*]] [ ; CHECK-NEXT: i32 17, label [[F:%.*]] ; CHECK-NEXT: i32 4, label [[F]] -; CHECK-NEXT: ] +; CHECK-NEXT: ], !prof [[PROF2:![0-9]+]] ; CHECK: common.ret: ; CHECK-NEXT: ret void ; CHECK: T: @@ -134,7 +134,7 @@ define void @test2(i32 %V) { %C1 = icmp ne i32 %V, 4 ; <i1> [#uses=1] %C2 = icmp ne i32 %V, 17 ; <i1> [#uses=1] %CN = and i1 %C1, %C2 ; <i1> [#uses=1] - br i1 %CN, label %T, label %F + br i1 %CN, label %T, label %F, !prof !1 T: ; preds = %0 call void @foo1( ) ret void @@ -1313,3 +1313,16 @@ if.then: if.end: ret void } + +!0 = !{!"function_entry_count", i32 100} +!1 = !{!"branch_weights", i32 6, i32 10} +;. +; DL: attributes #[[ATTR0:[0-9]+]] = { noredzone nounwind ssp } +; DL: attributes #[[ATTR1:[0-9]+]] = { nounwind } +; DL: attributes #[[ATTR2]] = { noredzone nounwind } +; DL: attributes #[[ATTR3]] = { noredzone } +;. +; DL: [[META0:![0-9]+]] = !{!"function_entry_count", i32 100} +; DL: [[PROF1]] = !{!"branch_weights", i32 10, i32 3, i32 3} +; DL: [[PROF2]] = !{!"branch_weights", i32 6, i32 5, i32 5} +;. |