[SystemZ] Use BRCT and BRCTG to eliminate add-&-compare sequences

This patch just uses a peephole test for "add; compare; branch" sequences
within a single block.  The IR optimizers already convert loops to
decrement-and-branch-on-nonzero form in some cases, so even this
simplistic test triggers many times during a clang bootstrap and
projects/test-suite run.  It looks like there are still cases where we
need to more strongly prefer branches on nonzero though.  E.g. I saw a
case where a loop that started out with a check for 0 ended up with a
check for -1.  I'll try to look at that sometime.

I ended up adding the Reference class because MachineInstr::readsRegister()
doesn't check for subregisters (by design, as far as I could tell).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187723 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 9 insertions, 1 deletions

diff --git a/lib/Target/SystemZ/SystemZInstrInfo.h b/lib/Target/SystemZ/SystemZInstrInfo.h
index 1392745672..b12b471a4d 100644
--- a/lib/Target/SystemZ/SystemZInstrInfo.h
+++ b/lib/Target/SystemZ/SystemZInstrInfo.h
@@ -69,7 +69,15 @@ namespace SystemZII {
 
     // An instruction that peforms a 64-bit signed comparison and branches
     // on the result.
-    BranchCG
+    BranchCG,
+
+    // An instruction that decrements a 32-bit register and branches if
+    // the result is nonzero.
+    BranchCT,
+
+    // An instruction that decrements a 64-bit register and branches if
+    // the result is nonzero.
+    BranchCTG
   };
   // Information about a branch instruction.
   struct Branch {