[PATCH] D26097: Don't claim the udiv created in BypassSlowDivision is exact.

[Justin Lebar via llvm-commits]

jlebar created this revision.
jlebar added a reviewer: tra.
jlebar added a subscriber: llvm-commits.
Herald added a subscriber: jholewinski.

In BypassSlowDivision's short-dividend path, we would create e.g.

  udiv exact i32 %a, %b

"exact" here means that we are asserting that %a is a multiple of %b.
But we have no reason to believe this must be true -- this is just a
bug, as far as I can tell.


https://reviews.llvm.org/D26097

Files:
  llvm/lib/Transforms/Utils/BypassSlowDivision.cpp
  llvm/test/Transforms/CodeGenPrepare/NVPTX/bypass-slow-div-not-exact.ll


Index: llvm/test/Transforms/CodeGenPrepare/NVPTX/bypass-slow-div-not-exact.ll
===================================================================
--- /dev/null
+++ llvm/test/Transforms/CodeGenPrepare/NVPTX/bypass-slow-div-not-exact.ll
@@ -0,0 +1,16 @@
+; RUN: opt -S -codegenprepare < %s | FileCheck %s
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+target triple = "nvptx64-nvidia-cuda"
+
+; Check that the smaller-width division that the BypassSlowDivision pass
+; creates is not marked as "exact" (that is, it doesn't claim that the
+; numerator is a multiple of the denominator).
+;
+; CHECK-LABEL: @test
+define void @test(i64 %a, i64 %b, i64* %retptr) {
+  ; CHECK: udiv i32
+  %d = sdiv i64 %a, %b
+  store i64 %d, i64* %retptr
+  ret void
+}
Index: llvm/lib/Transforms/Utils/BypassSlowDivision.cpp
===================================================================
--- llvm/lib/Transforms/Utils/BypassSlowDivision.cpp
+++ llvm/lib/Transforms/Utils/BypassSlowDivision.cpp
@@ -120,8 +120,7 @@
                                                  BypassType);
 
   // udiv/urem because optimization only handles positive numbers
-  Value *ShortQuotientV = FastBuilder.CreateExactUDiv(ShortDividendV,
-                                                      ShortDivisorV);
+  Value *ShortQuotientV = FastBuilder.CreateUDiv(ShortDividendV, ShortDivisorV);
   Value *ShortRemainderV = FastBuilder.CreateURem(ShortDividendV,
                                                   ShortDivisorV);
   Value *FastQuotientV = FastBuilder.CreateCast(Instruction::ZExt,


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D26097.76242.patch
Type: text/x-patch
Size: 1570 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161028/1aa7709f/attachment.bin>