[PATCH] Optionally disable the BadCFGConflict check in MachineBlockPlacement.cpp
Daniel Jasper
djasper at google.com
Wed Jan 14 08:36:22 PST 2015
Hi chandlerc,
Some benchmarks have shown that this could lead to a potential performance benefit.
A possible explanation. In diamond-shaped CFGs (A followed by either B or C both followed by D), putting B and C both in between A and D leads to the code being less dense than it could be. Always either B or C have to be skipped increasing the chance of cache misses etc. Moving either B or C to after D might be beneficial on average.
In the long run, but we should probably do a better job of analyzing the basic block and branch probabilities to move the correct one of B or C to after D. But even if we don't use this in the long run, it is a good baseline for benchmarking.
http://reviews.llvm.org/D6969
Files:
lib/CodeGen/MachineBlockPlacement.cpp
test/CodeGen/X86/code_placement_bad_cfg_check.ll
Index: lib/CodeGen/MachineBlockPlacement.cpp
===================================================================
--- lib/CodeGen/MachineBlockPlacement.cpp
+++ lib/CodeGen/MachineBlockPlacement.cpp
@@ -60,6 +60,12 @@
"blocks in the function."),
cl::init(0), cl::Hidden);
+static cl::opt<bool> NoBadCFGConflictCheck(
+ "no-bad-cfg-conflict-check",
+ cl::desc("Don't check whether a hot successor has a more important "
+ "predecessor."),
+ cl::init(false), cl::Hidden);
+
// FIXME: Find a good default for this flag and remove the flag.
static cl::opt<unsigned>
ExitBlockBias("block-placement-exit-block-bias",
@@ -374,29 +380,31 @@
continue;
}
- // Make sure that a hot successor doesn't have a globally more important
- // predecessor.
- BlockFrequency CandidateEdgeFreq
- = MBFI->getBlockFreq(BB) * SuccProb * HotProb.getCompl();
- bool BadCFGConflict = false;
- for (MachineBasicBlock::pred_iterator PI = (*SI)->pred_begin(),
- PE = (*SI)->pred_end();
- PI != PE; ++PI) {
- if (*PI == *SI || (BlockFilter && !BlockFilter->count(*PI)) ||
- BlockToChain[*PI] == &Chain)
+ if (!NoBadCFGConflictCheck) {
+ // Make sure that a hot successor doesn't have a globally more important
+ // predecessor.
+ BlockFrequency CandidateEdgeFreq =
+ MBFI->getBlockFreq(BB) * SuccProb * HotProb.getCompl();
+ bool BadCFGConflict = false;
+ for (MachineBasicBlock::pred_iterator PI = (*SI)->pred_begin(),
+ PE = (*SI)->pred_end();
+ PI != PE; ++PI) {
+ if (*PI == *SI || (BlockFilter && !BlockFilter->count(*PI)) ||
+ BlockToChain[*PI] == &Chain)
+ continue;
+ BlockFrequency PredEdgeFreq =
+ MBFI->getBlockFreq(*PI) * MBPI->getEdgeProbability(*PI, *SI);
+ if (PredEdgeFreq >= CandidateEdgeFreq) {
+ BadCFGConflict = true;
+ break;
+ }
+ }
+ if (BadCFGConflict) {
+ DEBUG(dbgs() << " " << getBlockName(*SI) << " -> " << SuccProb
+ << " (prob) (non-cold CFG conflict)\n");
continue;
- BlockFrequency PredEdgeFreq
- = MBFI->getBlockFreq(*PI) * MBPI->getEdgeProbability(*PI, *SI);
- if (PredEdgeFreq >= CandidateEdgeFreq) {
- BadCFGConflict = true;
- break;
}
}
- if (BadCFGConflict) {
- DEBUG(dbgs() << " " << getBlockName(*SI) << " -> " << SuccProb
- << " (prob) (non-cold CFG conflict)\n");
- continue;
- }
}
DEBUG(dbgs() << " " << getBlockName(*SI) << " -> " << SuccProb
Index: test/CodeGen/X86/code_placement_bad_cfg_check.ll
===================================================================
--- /dev/null
+++ test/CodeGen/X86/code_placement_bad_cfg_check.ll
@@ -0,0 +1,25 @@
+; RUN: llc -mcpu=corei7 -mtriple=x86_64-linux -no-bad-cfg-conflict-check < %s | FileCheck %s
+
+;CHECK: callq h
+;CHECK: ret
+;CHECK: callq
+define void @foo(i32 %t) nounwind readnone ssp uwtable {
+ %cmp = icmp eq i32 %t, 0
+ br i1 %cmp, label %if.then, label %if.else
+
+if.then:
+ call void @f()
+ br label %if.end
+
+if.else:
+ call void @g()
+ br label %if.end
+
+if.end:
+ call void @h()
+ ret void
+}
+
+declare void @f()
+declare void @g()
+declare void @h()
EMAIL PREFERENCES
http://reviews.llvm.org/settings/panel/emailpreferences/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D6969.18160.patch
Type: text/x-patch
Size: 3556 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150114/0add88ef/attachment.bin>
More information about the llvm-commits
mailing list