[PowerPC] Add Itineraries of IIC_IntRotateDI for P7/P8

When doing some instruction scheduling work, we noticed some missing itineraries.
Before we switch to machine scheduler, those missing itineraries might not have impact to actually scheduling, 
because we can still get same latency due to default values.

With machine scheduler, however, itineraries will have impact to scheduling.
eg: NumMicroOps will default to be 0 if there is NO itineraries for specific instruction class.
And most of the instruction class with itineraries will have NumMicroOps default to 1.

This will has impact on the count of RetiredMOps, affects the Pending/Available Queue, 
then causing different scheduling or suboptimal scheduling further.

Patch by jsji (Jinsong Ji)
Differential Revision: https://reviews.llvm.org/D51506

llvm-svn: 341293
This commit is contained in:
QingShan Zhang 2018-09-03 03:14:29 +00:00
parent 17e709b66a
commit c2b6c547dc
6 changed files with 81 additions and 73 deletions

View File

@ -126,6 +126,10 @@ def P7Itineraries : ProcessorItineraries<
P7_DU3, P7_DU4], 0>,
InstrStage<1, [P7_FX1, P7_FX2]>],
[1, 1, 1]>,
InstrItinData<IIC_IntRotateDI , [InstrStage<1, [P7_DU1, P7_DU2,
P7_DU3, P7_DU4], 0>,
InstrStage<1, [P7_FX1, P7_FX2]>],
[1, 1, 1]>,
InstrItinData<IIC_IntShift , [InstrStage<1, [P7_DU1, P7_DU2,
P7_DU3, P7_DU4], 0>,
InstrStage<1, [P7_FX1, P7_FX2]>],

View File

@ -102,6 +102,10 @@ def P8Itineraries : ProcessorItineraries<
P8_DU4, P8_DU5, P8_DU6], 0>,
InstrStage<1, [P8_FXU1, P8_FXU2]>],
[1, 1, 1]>,
InstrItinData<IIC_IntRotateDI , [InstrStage<1, [P8_DU1, P8_DU2, P8_DU3,
P8_DU4, P8_DU5, P8_DU6], 0>,
InstrStage<1, [P8_FXU1, P8_FXU2]>],
[1, 1, 1]>,
InstrItinData<IIC_IntShift , [InstrStage<1, [P8_DU1, P8_DU2, P8_DU3,
P8_DU4, P8_DU5, P8_DU6], 0>,
InstrStage<1, [P8_FXU1, P8_FXU2]>],

View File

@ -33,20 +33,20 @@ define signext i32 @main() {
;
; CHECK-P7-LABEL: main:
; CHECK-P7: li 3, -32477
; CHECK-P7: lis 4, 0
; CHECK-P7: lis 5, 0
; CHECK-P7: li 7, 0
; CHECK-P7: li 5, 234
; CHECK-P7: sth 3, 46(1)
; CHECK-P7: ori 4, 4, 33059
; CHECK-P7: rlwinm 3, 6, 3, 27, 27
; CHECK-P7: li 6, 234
; CHECK-P7: ori 5, 5, 33059
; CHECK-P7: rlwinm 3, 4, 3, 27, 27
; CHECK-P7: ori 7, 7, 65535
; CHECK-P7: sync
; CHECK-P7: slw 6, 6, 3
; CHECK-P7: slw 8, 5, 3
; CHECK-P7: slw 9, 4, 3
; CHECK-P7: rldicr 4, 6, 0, 61
; CHECK-P7: slw 5, 7, 3
; CHECK-P7: and 7, 8, 5
; CHECK-P7: and 8, 9, 5
; CHECK-P7: rldicr 4, 4, 0, 61
; CHECK-P7: and 7, 6, 5
; CHECK-P7: and 8, 8, 5
; CHECK-P7: .LBB0_1: # %L.entry
; CHECK-P7: lwarx 9, 0, 4
; CHECK-P7: and 6, 9, 5

View File

@ -36,8 +36,8 @@ define i37 @fshl_i37(i37 %x, i37 %y, i37 %z) {
; CHECK: # %bb.0:
; CHECK-NEXT: lis 6, -8857
; CHECK-NEXT: clrldi 5, 5, 27
; CHECK-NEXT: clrldi 4, 4, 27
; CHECK-NEXT: ori 6, 6, 51366
; CHECK-NEXT: clrldi 4, 4, 27
; CHECK-NEXT: sldi 6, 6, 32
; CHECK-NEXT: oris 6, 6, 3542
; CHECK-NEXT: ori 6, 6, 31883

View File

@ -71,49 +71,49 @@ define i64 @ReverseBits64(i64 %n) {
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lis 4, -21846
; CHECK-NEXT: lis 5, 21845
; CHECK-NEXT: lis 6, -13108
; CHECK-NEXT: lis 7, 13107
; CHECK-NEXT: sldi 8, 3, 1
; CHECK-NEXT: rldicl 3, 3, 63, 1
; CHECK-NEXT: lis 7, -13108
; CHECK-NEXT: lis 8, 13107
; CHECK-NEXT: ori 4, 4, 43690
; CHECK-NEXT: ori 5, 5, 21845
; CHECK-NEXT: ori 6, 6, 52428
; CHECK-NEXT: ori 7, 7, 13107
; CHECK-NEXT: ori 7, 7, 52428
; CHECK-NEXT: ori 8, 8, 13107
; CHECK-NEXT: sldi 4, 4, 32
; CHECK-NEXT: sldi 5, 5, 32
; CHECK-NEXT: oris 4, 4, 43690
; CHECK-NEXT: oris 5, 5, 21845
; CHECK-NEXT: sldi 6, 3, 1
; CHECK-NEXT: rldicl 3, 3, 63, 1
; CHECK-NEXT: ori 4, 4, 43690
; CHECK-NEXT: ori 5, 5, 21845
; CHECK-NEXT: and 4, 8, 4
; CHECK-NEXT: and 3, 3, 5
; CHECK-NEXT: sldi 5, 6, 32
; CHECK-NEXT: sldi 6, 7, 32
; CHECK-NEXT: lis 7, 3855
; CHECK-NEXT: or 3, 3, 4
; CHECK-NEXT: oris 4, 5, 52428
; CHECK-NEXT: oris 5, 6, 13107
; CHECK-NEXT: lis 6, -3856
; CHECK-NEXT: ori 7, 7, 3855
; CHECK-NEXT: sldi 8, 3, 2
; CHECK-NEXT: ori 4, 4, 52428
; CHECK-NEXT: rldicl 3, 3, 62, 2
; CHECK-NEXT: ori 5, 5, 13107
; CHECK-NEXT: ori 6, 6, 61680
; CHECK-NEXT: and 4, 8, 4
; CHECK-NEXT: and 3, 3, 5
; CHECK-NEXT: sldi 5, 6, 32
; CHECK-NEXT: sldi 6, 7, 32
; CHECK-NEXT: or 3, 3, 4
; CHECK-NEXT: oris 4, 5, 61680
; CHECK-NEXT: oris 5, 6, 3855
; CHECK-NEXT: sldi 6, 3, 4
; CHECK-NEXT: ori 4, 4, 61680
; CHECK-NEXT: rldicl 3, 3, 60, 4
; CHECK-NEXT: ori 5, 5, 3855
; CHECK-NEXT: sldi 7, 7, 32
; CHECK-NEXT: sldi 8, 8, 32
; CHECK-NEXT: and 4, 6, 4
; CHECK-NEXT: and 3, 3, 5
; CHECK-NEXT: lis 5, -3856
; CHECK-NEXT: oris 6, 7, 52428
; CHECK-NEXT: oris 7, 8, 13107
; CHECK-NEXT: or 3, 3, 4
; CHECK-NEXT: lis 4, 3855
; CHECK-NEXT: ori 5, 5, 61680
; CHECK-NEXT: ori 6, 6, 52428
; CHECK-NEXT: ori 7, 7, 13107
; CHECK-NEXT: ori 4, 4, 3855
; CHECK-NEXT: sldi 8, 3, 2
; CHECK-NEXT: rldicl 3, 3, 62, 2
; CHECK-NEXT: and 6, 8, 6
; CHECK-NEXT: and 3, 3, 7
; CHECK-NEXT: sldi 5, 5, 32
; CHECK-NEXT: sldi 4, 4, 32
; CHECK-NEXT: or 3, 3, 6
; CHECK-NEXT: oris 5, 5, 61680
; CHECK-NEXT: oris 4, 4, 3855
; CHECK-NEXT: sldi 6, 3, 4
; CHECK-NEXT: ori 5, 5, 61680
; CHECK-NEXT: ori 4, 4, 3855
; CHECK-NEXT: rldicl 3, 3, 60, 4
; CHECK-NEXT: and 5, 6, 5
; CHECK-NEXT: and 3, 3, 4
; CHECK-NEXT: or 3, 3, 5
; CHECK-NEXT: rldicl 4, 3, 32, 32
; CHECK-NEXT: rlwinm 5, 3, 24, 0, 31
; CHECK-NEXT: rlwinm 6, 4, 24, 0, 31

View File

@ -47,49 +47,49 @@ define i64 @testBitReverseIntrinsicI64(i64 %arg) {
; CHECK: # %bb.0:
; CHECK-NEXT: lis 4, -21846
; CHECK-NEXT: lis 5, 21845
; CHECK-NEXT: lis 6, -13108
; CHECK-NEXT: lis 7, 13107
; CHECK-NEXT: sldi 8, 3, 1
; CHECK-NEXT: rldicl 3, 3, 63, 1
; CHECK-NEXT: lis 7, -13108
; CHECK-NEXT: lis 8, 13107
; CHECK-NEXT: ori 4, 4, 43690
; CHECK-NEXT: ori 5, 5, 21845
; CHECK-NEXT: ori 6, 6, 52428
; CHECK-NEXT: ori 7, 7, 13107
; CHECK-NEXT: ori 7, 7, 52428
; CHECK-NEXT: ori 8, 8, 13107
; CHECK-NEXT: sldi 4, 4, 32
; CHECK-NEXT: sldi 5, 5, 32
; CHECK-NEXT: oris 4, 4, 43690
; CHECK-NEXT: oris 5, 5, 21845
; CHECK-NEXT: sldi 6, 3, 1
; CHECK-NEXT: rldicl 3, 3, 63, 1
; CHECK-NEXT: ori 4, 4, 43690
; CHECK-NEXT: ori 5, 5, 21845
; CHECK-NEXT: and 4, 8, 4
; CHECK-NEXT: and 3, 3, 5
; CHECK-NEXT: sldi 5, 6, 32
; CHECK-NEXT: sldi 6, 7, 32
; CHECK-NEXT: lis 7, 3855
; CHECK-NEXT: or 3, 3, 4
; CHECK-NEXT: oris 4, 5, 52428
; CHECK-NEXT: oris 5, 6, 13107
; CHECK-NEXT: lis 6, -3856
; CHECK-NEXT: ori 7, 7, 3855
; CHECK-NEXT: sldi 8, 3, 2
; CHECK-NEXT: ori 4, 4, 52428
; CHECK-NEXT: rldicl 3, 3, 62, 2
; CHECK-NEXT: ori 5, 5, 13107
; CHECK-NEXT: ori 6, 6, 61680
; CHECK-NEXT: and 4, 8, 4
; CHECK-NEXT: and 3, 3, 5
; CHECK-NEXT: sldi 5, 6, 32
; CHECK-NEXT: sldi 6, 7, 32
; CHECK-NEXT: or 3, 3, 4
; CHECK-NEXT: oris 4, 5, 61680
; CHECK-NEXT: oris 5, 6, 3855
; CHECK-NEXT: sldi 6, 3, 4
; CHECK-NEXT: ori 4, 4, 61680
; CHECK-NEXT: rldicl 3, 3, 60, 4
; CHECK-NEXT: ori 5, 5, 3855
; CHECK-NEXT: sldi 7, 7, 32
; CHECK-NEXT: sldi 8, 8, 32
; CHECK-NEXT: and 4, 6, 4
; CHECK-NEXT: and 3, 3, 5
; CHECK-NEXT: lis 5, -3856
; CHECK-NEXT: oris 6, 7, 52428
; CHECK-NEXT: oris 7, 8, 13107
; CHECK-NEXT: or 3, 3, 4
; CHECK-NEXT: lis 4, 3855
; CHECK-NEXT: ori 5, 5, 61680
; CHECK-NEXT: ori 6, 6, 52428
; CHECK-NEXT: ori 7, 7, 13107
; CHECK-NEXT: ori 4, 4, 3855
; CHECK-NEXT: sldi 8, 3, 2
; CHECK-NEXT: rldicl 3, 3, 62, 2
; CHECK-NEXT: and 6, 8, 6
; CHECK-NEXT: and 3, 3, 7
; CHECK-NEXT: sldi 5, 5, 32
; CHECK-NEXT: sldi 4, 4, 32
; CHECK-NEXT: or 3, 3, 6
; CHECK-NEXT: oris 5, 5, 61680
; CHECK-NEXT: oris 4, 4, 3855
; CHECK-NEXT: sldi 6, 3, 4
; CHECK-NEXT: ori 5, 5, 61680
; CHECK-NEXT: ori 4, 4, 3855
; CHECK-NEXT: rldicl 3, 3, 60, 4
; CHECK-NEXT: and 5, 6, 5
; CHECK-NEXT: and 3, 3, 4
; CHECK-NEXT: or 3, 3, 5
; CHECK-NEXT: rldicl 4, 3, 32, 32
; CHECK-NEXT: rlwinm 5, 3, 24, 0, 31
; CHECK-NEXT: rlwinm 6, 4, 24, 0, 31