[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).

llvm-svn: 187723

1 files changed, 3 insertions, 0 deletions

diff --git a/llvm/lib/Target/SystemZ/SystemZTargetMachine.cpp b/llvm/lib/Target/SystemZ/SystemZTargetMachine.cpp
index 2bacc2bc..856183c 100644
--- a/llvm/lib/Target/SystemZ/SystemZTargetMachine.cpp
+++ b/llvm/lib/Target/SystemZ/SystemZTargetMachine.cpp
@@ -82,6 +82,9 @@ bool SystemZPassConfig::addPreEmitPass() {
   // CC values (while still being worthwhile) and others that happen to make
   // the CC result more useful than it was originally.
   //
+  // Another reason is that we only want to use BRANCH ON COUNT in cases
+  // where we know that the count register is not going to be spilled.
+  //
   // Doing it so late makes it more likely that a register will be reused
   // between the comparison and the branch, but it isn't clear whether
   // preventing that would be a win or not.