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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123226 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Target/CellSPU/SPUTargetMachine.cpp b/lib/Target/CellSPU/SPUTargetMachine.cpp
index 3423c69..3ed7361 100644
--- a/lib/Target/CellSPU/SPUTargetMachine.cpp
+++ b/lib/Target/CellSPU/SPUTargetMachine.cpp
@@ -59,3 +59,12 @@
PM.add(createSPUISelDag(*this));
return false;
}
+
+// passes to run just before printing the assembly
+bool SPUTargetMachine::
+addPreEmitPass(PassManagerBase &PM, CodeGenOpt::Level OptLevel)
+{
+ //align instructions with nops/lnops for dual issue
+ PM.add(createSPUNopFillerPass(*this));
+ return true;
+}