[llvm-commits] [llvm] r61778 - in /llvm/trunk: lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp test/CodeGen/X86/fold-pcmpeqd-0.ll

[Dan Gohman]

Author: djg
Date: Mon Jan  5 19:19:04 2009
New Revision: 61778

URL: http://llvm.org/viewvc/llvm-project?rev=61778&view=rev
Log:
Use a latency value of 0 for the artificial edges inserted by
AddPseudoTwoAddrDeps. This lets the scheduling infrastructure
avoid recalculating node heights. In very large testcases this
was a major bottleneck. Thanks to Roman Levenstein for finding
this!

As a side effect, fold-pcmpeqd-0.ll is now scheduled better
and it no longer requires spilling on x86-32.

Modified:
    llvm/trunk/lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp
    llvm/trunk/test/CodeGen/X86/fold-pcmpeqd-0.ll

Modified: llvm/trunk/lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp?rev=61778&r1=61777&r2=61778&view=diff

==============================================================================
--- llvm/trunk/lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp (original)
+++ llvm/trunk/lib/CodeGen/SelectionDAG/ScheduleDAGRRList.cpp Mon Jan  5 19:19:04 2009
@@ -1252,7 +1252,7 @@
             !scheduleDAG->IsReachable(SuccSU, SU)) {
           DOUT << "Adding a pseudo-two-addr edge from SU # " << SU->NodeNum
                << " to SU #" << SuccSU->NodeNum << "\n";
-          scheduleDAG->AddPred(SU, SDep(SuccSU, SDep::Order, /*Latency=*/1,
+          scheduleDAG->AddPred(SU, SDep(SuccSU, SDep::Order, /*Latency=*/0,
                                         /*Reg=*/0, /*isMustAlias=*/false,
                                         /*isArtificial=*/true));
         }

Modified: llvm/trunk/test/CodeGen/X86/fold-pcmpeqd-0.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/fold-pcmpeqd-0.ll?rev=61778&r1=61777&r2=61778&view=diff

==============================================================================
--- llvm/trunk/test/CodeGen/X86/fold-pcmpeqd-0.ll (original)
+++ llvm/trunk/test/CodeGen/X86/fold-pcmpeqd-0.ll Mon Jan  5 19:19:04 2009
@@ -1,4 +1,4 @@
-; RUN: llvm-as < %s | llc -mtriple=i386-apple-darwin   | not grep pcmpeqd
+; RUN: llvm-as < %s | llc -mtriple=i386-apple-darwin   | grep pcmpeqd | count 1
 ; RUN: llvm-as < %s | llc -mtriple=x86_64-apple-darwin | grep pcmpeqd | count 1
 
 ; This testcase shouldn't need to spill the -1 value,