diff --git a/llvm/test/Transforms/SimpleLoopUnswitch/AMDGPU/uniform-unswitch.ll b/llvm/test/Transforms/SimpleLoopUnswitch/AMDGPU/uniform-unswitch.ll new file mode 100644 index 0000000000000..8d1b0c7177be4 --- /dev/null +++ b/llvm/test/Transforms/SimpleLoopUnswitch/AMDGPU/uniform-unswitch.ll @@ -0,0 +1,77 @@ +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 3 +; RUN: opt -mtriple=amdgcn-- -passes='default' -S %s | FileCheck %s + +; Check that loop unswitch happened and condition hoisted out of the loop. +; Condition is uniform so even targets with divergence should perform unswitching. + +; This fails with the new pass manager: +; https://bugs.llvm.org/show_bug.cgi?id=48819 +; The correct behaviour (allow uniform non-trivial branches to be +; unswitched on all targets) requires access to the function-level +; divergence analysis from a loop transform, which is currently not +; supported in the new pass manager. + +; SHOULDBE-LABEL: {{^}}define amdgpu_kernel void @uniform_unswitch +; SHOULDBE: entry: +; SHOULDBE-NEXT: [[LOOP_COND:%[a-z0-9]+]] = icmp +; SHOULDBE-NEXT: [[IF_COND:%[a-z0-9]+]] = icmp eq i32 %x, 123456 +; SHOULDBE-NEXT: and i1 [[LOOP_COND]], [[IF_COND]] +; SHOULDBE-NEXT: br i1 + +define amdgpu_kernel void @uniform_unswitch(ptr nocapture %out, i32 %n, i32 %x) { +; CHECK-LABEL: define amdgpu_kernel void @uniform_unswitch +; CHECK-SAME: (ptr nocapture writeonly [[OUT:%.*]], i32 [[N:%.*]], i32 [[X:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] { +; CHECK-NEXT: entry: +; CHECK-NEXT: [[OUT_GLOBAL:%.*]] = addrspacecast ptr [[OUT]] to ptr addrspace(1) +; CHECK-NEXT: [[CMP6:%.*]] = icmp sgt i32 [[N]], 0 +; CHECK-NEXT: br i1 [[CMP6]], label [[FOR_BODY_LR_PH:%.*]], label [[FOR_COND_CLEANUP:%.*]] +; CHECK: for.body.lr.ph: +; CHECK-NEXT: [[CMP1:%.*]] = icmp eq i32 [[X]], 123456 +; CHECK-NEXT: br label [[FOR_BODY:%.*]] +; CHECK: for.cond.cleanup: +; CHECK-NEXT: ret void +; CHECK: for.body: +; CHECK-NEXT: [[I_07:%.*]] = phi i32 [ 0, [[FOR_BODY_LR_PH]] ], [ [[INC:%.*]], [[FOR_INC:%.*]] ] +; CHECK-NEXT: br i1 [[CMP1]], label [[IF_THEN:%.*]], label [[FOR_INC]] +; CHECK: if.then: +; CHECK-NEXT: [[TMP0:%.*]] = zext i32 [[I_07]] to i64 +; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr addrspace(1) [[OUT_GLOBAL]], i64 [[TMP0]] +; CHECK-NEXT: store i32 [[I_07]], ptr addrspace(1) [[ARRAYIDX]], align 4 +; CHECK-NEXT: br label [[FOR_INC]] +; CHECK: for.inc: +; CHECK-NEXT: [[INC]] = add nuw nsw i32 [[I_07]], 1 +; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INC]], [[N]] +; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]] +; +entry: + %cmp6 = icmp sgt i32 %n, 0 + br i1 %cmp6, label %for.body.lr.ph, label %for.cond.cleanup + +for.body.lr.ph: ; preds = %entry + %cmp1 = icmp eq i32 %x, 123456 + br label %for.body + +for.cond.cleanup.loopexit: ; preds = %for.inc + br label %for.cond.cleanup + +for.cond.cleanup: ; preds = %for.cond.cleanup.loopexit, %entry + ret void + +for.body: ; preds = %for.inc, %for.body.lr.ph + %i.07 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.inc ] + br i1 %cmp1, label %if.then, label %for.inc + +if.then: ; preds = %for.body + %arrayidx = getelementptr inbounds i32, ptr %out, i32 %i.07 + store i32 %i.07, ptr %arrayidx, align 4 + br label %for.inc + +for.inc: ; preds = %for.body, %if.then + %inc = add nuw nsw i32 %i.07, 1 + %exitcond = icmp eq i32 %inc, %n + br i1 %exitcond, label %for.cond.cleanup.loopexit, label %for.body +} + +declare i32 @llvm.amdgcn.workitem.id.x() #0 + +attributes #0 = { nounwind readnone }