From: Matthias Braun <matze@braunis.de>
Date: Tue, 16 Jun 2015 18:22:26 +0000 (+0000)
Subject: TargetRegisterInfo: Make the concept of imprecise lane masks explicit
X-Git-Url: http://plrg.eecs.uci.edu/git/?p=oota-llvm.git;a=commitdiff_plain;h=ff881427ec2336a561d63d25498741e980b18358

TargetRegisterInfo: Make the concept of imprecise lane masks explicit

LaneMasks as given by getSubRegIndexLaneMask() have a limited number of
of bits, so for targets with more than 31 disjunct subregister there may
be cases where:

   getSubReg(Reg,A) does not overlap getSubReg(Reg,B)

but we still have

   (getSubRegIndexLaneMask(A) & getSubRegIndexLaneMask(B)) != 0.

I had hoped to keep this an implementation detail of the tablegen but as
my next commit shows we can avoid unnecessary imp-defs operands if we
know that the lane masks in use are precise.

This is in preparation to http://reviews.llvm.org/D10470.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239837 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/include/llvm/Target/TargetRegisterInfo.h b/include/llvm/Target/TargetRegisterInfo.h
index 121b8a23252..1fd99a31993 100644
--- a/include/llvm/Target/TargetRegisterInfo.h
+++ b/include/llvm/Target/TargetRegisterInfo.h
@@ -373,6 +373,19 @@ public:
     return SubRegIndexLaneMasks[SubIdx];
   }
 
+  /// Returns true if the given lane mask is imprecise.
+  ///
+  /// LaneMasks as given by getSubRegIndexLaneMask() have a limited number of
+  /// bits, so for targets with more than 31 disjunct subregister indices there
+  /// may be cases where:
+  ///    getSubReg(Reg,A) does not overlap getSubReg(Reg,B)
+  /// but we still have
+  ///    (getSubRegIndexLaneMask(A) & getSubRegIndexLaneMask(B)) != 0.
+  /// This function returns true in those cases.
+  static bool isImpreciseLaneMask(unsigned LaneMask) {
+    return LaneMask & 0x80000000u;
+  }
+
   /// The lane masks returned by getSubRegIndexLaneMask() above can only be
   /// used to determine if sub-registers overlap - they can't be used to
   /// determine if a set of sub-registers completely cover another
diff --git a/lib/CodeGen/RegisterCoalescer.cpp b/lib/CodeGen/RegisterCoalescer.cpp
index e513a4f1ccf..7b7c7502172 100644
--- a/lib/CodeGen/RegisterCoalescer.cpp
+++ b/lib/CodeGen/RegisterCoalescer.cpp
@@ -2633,7 +2633,8 @@ bool RegisterCoalescer::joinVirtRegs(CoalescerPair &CP) {
       // "overflow bit" 32. As a workaround we drop all subregister ranges
       // which means we loose some precision but are back to a well defined
       // state.
-      assert((CP.getNewRC()->getLaneMask() & 0x80000000u)
+      assert(TargetRegisterInfo::isImpreciseLaneMask(
+             CP.getNewRC()->getLaneMask())
              && "SubRange merge should only fail when merging into bit 32.");
       DEBUG(dbgs() << "\tSubrange join aborted!\n");
       LHS.clearSubRanges();