Add a "nop filler" pass to SPU.

Filling no-ops is done just before emitting of assembly,
when the instruction stream is final. No-ops are inserted
to align the instructions so the dual-issue of the pipeline
is utilized. This speeds up generated code with a minimum of 
1% on a select set of algorithms.

This pass may be redundant if the instruction scheduler and 
all subsequent passes that modify the instruction stream 
(prolog+epilog inserter, register scavenger, are there others?)
are made aware of the instruction alignments.

llvm-svn: 123226

1 files changed, 1 insertions, 0 deletions

diff --git a/llvm/lib/Target/CellSPU/SPU.h b/llvm/lib/Target/CellSPU/SPU.h
index 1f215113b40..72f84300b2c 100644
--- a/llvm/lib/Target/CellSPU/SPU.h
+++ b/llvm/lib/Target/CellSPU/SPU.h
@@ -23,6 +23,7 @@ namespace llvm {
   class formatted_raw_ostream;
 
   FunctionPass *createSPUISelDag(SPUTargetMachine &TM);
+  FunctionPass *createSPUNopFillerPass(SPUTargetMachine &tm);
 
   extern Target TheCellSPUTarget;
 }