[PATCH] D97047: [Analysis][LoopVectorize] do not try to form reductions of pointers

[Sanjay Patel via Phabricator via llvm-commits]

spatel created this revision.
spatel added reviewers: nikic, lebedev.ri, bero, craig.topper, fhahn, dmgreen, RKSimon.
Herald added subscribers: hiraditya, mcrosier.
spatel requested review of this revision.
Herald added a project: LLVM.

This is a fix for https://llvm.org/PR49215 either before/after we make a verifier enhancement for vector reductions with D96904 <https://reviews.llvm.org/D96904>.

I'm not sure what the current thinking is for pointer math/logic in IR. We allow icmp on pointer values. Therefore, we match min/max patterns, so without this patch, the vectorizer could form a vector reduction from that sequence.

But the LangRef definitions for min/max and vector reduction intrinsics do not allow pointer types:
https://llvm.org/docs/LangRef.html#llvm-smax-intrinsic
https://llvm.org/docs/LangRef.html#llvm-vector-reduce-umax-intrinsic

So we would crash/assert at some point - either in IR verification, in the cost model, or in codegen. If we do want to allow this kind of transform, we will need to update the LangRef and all of those parts of the compiler.


https://reviews.llvm.org/D97047

Files:
  llvm/lib/Analysis/IVDescriptors.cpp
  llvm/test/Transforms/LoopVectorize/reduction-ptr.ll


Index: llvm/test/Transforms/LoopVectorize/reduction-ptr.ll
===================================================================
--- /dev/null
+++ llvm/test/Transforms/LoopVectorize/reduction-ptr.ll
@@ -0,0 +1,40 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s  -loop-vectorize -force-vector-width=4 -S | FileCheck %s
+
+target datalayout = "e-m:o-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
+
+; Reductions of pointer types are not supported.
+
+define void @PR49215(i32* %p, i32* %q) {
+; CHECK-LABEL: @PR49215(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT:    [[G:%.*]] = phi i32* [ [[P:%.*]], [[ENTRY]] ], [ [[UMIN:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT:    [[CMP2:%.*]] = icmp ult i32* [[Q:%.*]], [[G]]
+; CHECK-NEXT:    [[UMIN]] = select i1 [[CMP2]], i32* [[Q]], i32* [[G]]
+; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[IV_NEXT]], undef
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[LOOPEXIT:%.*]], label [[FOR_BODY]]
+; CHECK:       loopexit:
+; CHECK-NEXT:    [[UMIN_LCSSA:%.*]] = phi i32* [ [[UMIN]], [[FOR_BODY]] ]
+; CHECK-NEXT:    [[PHI_CAST:%.*]] = ptrtoint i32* [[UMIN_LCSSA]] to i64
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %for.body
+
+for.body:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+  %g = phi i32* [ %p, %entry ], [ %umin, %for.body ]
+  %cmp2 = icmp ult i32* %q, %g
+  %umin = select i1 %cmp2, i32* %q, i32* %g
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond = icmp eq i64 %iv.next, undef
+  br i1 %exitcond, label %loopexit, label %for.body
+
+loopexit:
+  %phi.cast = ptrtoint i32* %umin to i64
+  ret void
+}
Index: llvm/lib/Analysis/IVDescriptors.cpp
===================================================================
--- llvm/lib/Analysis/IVDescriptors.cpp
+++ llvm/lib/Analysis/IVDescriptors.cpp
@@ -198,6 +198,10 @@
   if (Phi->getNumIncomingValues() != 2)
     return false;
 
+  // Pointer min/max may be allowed, but it is not supported as a reduction op.
+  if (Phi->getType()->isPointerTy())
+    return false;
+
   // Reduction variables are only found in the loop header block.
   if (Phi->getParent() != TheLoop->getHeader())
     return false;


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D97047.324962.patch
Type: text/x-patch
Size: 2394 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210219/b0b76f1c/attachment.bin>