[PATCH] D70066: [MacroFusion] Limit the max fused number as 2 to reduce the dependency
qshanz via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Nov 26 21:13:36 PST 2019
steven.zhang updated this revision to Diff 231175.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D70066/new/
https://reviews.llvm.org/D70066
Files:
llvm/lib/CodeGen/MacroFusion.cpp
llvm/test/CodeGen/AArch64/macro-fusion.ll
Index: llvm/test/CodeGen/AArch64/macro-fusion.ll
===================================================================
--- llvm/test/CodeGen/AArch64/macro-fusion.ll
+++ llvm/test/CodeGen/AArch64/macro-fusion.ll
@@ -1,21 +1,18 @@
; REQUIRES: asserts
; RUN: llc < %s -mtriple=aarch64-linux-gnu -mattr=+fuse-arith-logic -verify-misched -debug-only=machine-scheduler 2>&1 > /dev/null | FileCheck %s
-; Verify that, the macro-fusion creates the necessary dependencies between SUs.
+; Verify that, the macro-fusion creates the necessary dependencies between SUs and
+; only 2 SU's are fused at most.
define signext i32 @test(i32 signext %a, i32 signext %b, i32 signext %c, i32 signext %d) {
entry:
; CHECK: ********** MI Scheduling **********
; CHECK-LABEL: %bb.0 entry
; CHECK: Macro fuse: SU([[SU4:[0-9]+]]) - SU([[SU5:[0-9]+]])
; CHECK: Bind SU([[SU1:[0-9]+]]) - SU([[SU4]])
-; CHECK: Macro fuse: SU([[SU5]]) - SU([[SU6:[0-9]+]])
-; CHECK: Bind SU([[SU0:[0-9]+]]) - SU([[SU5]])
-; CHECK: SU([[SU0]]): %{{[0-9]+}}:gpr32 = COPY $w3
+; CHECK-NOT: Macro fuse:
; CHECK: SU([[SU1]]): %{{[0-9]+}}:gpr32 = COPY $w2
; CHECK: SU([[SU4]]): %{{[0-9]+}}:gpr32 = nsw ADDWrr
; CHECK: SU([[SU5]]): %{{[0-9]+}}:gpr32 = nsw ADDWrr
-; CHECK: SU([[SU6]]): %{{[0-9]+}}:gpr32 = nsw SUBWrr
-
%add = add nsw i32 %b, %a
%add1 = add nsw i32 %add, %c
%sub = sub nsw i32 %add1, %d
Index: llvm/lib/CodeGen/MacroFusion.cpp
===================================================================
--- llvm/lib/CodeGen/MacroFusion.cpp
+++ llvm/lib/CodeGen/MacroFusion.cpp
@@ -32,10 +32,27 @@
static cl::opt<bool> EnableMacroFusion("misched-fusion", cl::Hidden,
cl::desc("Enable scheduling for macro fusion."), cl::init(true));
+namespace {
+
static bool isHazard(const SDep &Dep) {
return Dep.getKind() == SDep::Anti || Dep.getKind() == SDep::Output;
}
+static SUnit *getPredClusterSU(const SUnit &SU) {
+ for (const SDep &SI : SU.Preds)
+ if (SI.isCluster())
+ return SI.getSUnit();
+
+ return nullptr;
+}
+
+static bool hasLessThanNumFused(const SUnit &SU, unsigned FuseLimit) {
+ unsigned Num = 1;
+ const SUnit *CurrentSU = &SU;
+ while ((CurrentSU = getPredClusterSU(*CurrentSU)) && Num < FuseLimit) Num ++;
+ return Num < FuseLimit;
+}
+
static bool fuseInstructionPair(ScheduleDAGInstrs &DAG, SUnit &FirstSU,
SUnit &SecondSU) {
// Check that neither instr is already paired with another along the edge
@@ -110,8 +127,6 @@
return true;
}
-namespace {
-
/// Post-process the DAG to create cluster edges between instrs that may
/// be fused by the processor into a single operation.
class MacroFusion : public ScheduleDAGMutation {
@@ -161,8 +176,10 @@
if (DepSU.isBoundaryNode())
continue;
+ // Only back-2-back macro fusion is supported now.
const MachineInstr *DepMI = DepSU.getInstr();
- if (!shouldScheduleAdjacent(TII, ST, DepMI, AnchorMI))
+ if (!hasLessThanNumFused(DepSU, 2) ||
+ !shouldScheduleAdjacent(TII, ST, DepMI, AnchorMI))
continue;
if (fuseInstructionPair(DAG, DepSU, AnchorSU))
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D70066.231175.patch
Type: text/x-patch
Size: 3114 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191127/95a4de6e/attachment-0001.bin>
More information about the llvm-commits
mailing list