Skip to content

Commit

Permalink
[Power9] Added support for the modsw, moduw, modsd, modud hardware in…
Browse files Browse the repository at this point in the history
…structions.

Note that if we need the result of both the divide and the modulo then we
compute the modulo based on the result of the divide and not using the new
hardware instruction.

Commit on behalf of STEFAN PINTILIE.
Differential Revision: https://reviews.llvm.org/D33940

llvm-svn: 305210
  • Loading branch information
Tony Jiang committed Jun 12, 2017
1 parent 4f01175 commit 30a49d1
Show file tree
Hide file tree
Showing 8 changed files with 351 additions and 5 deletions.
37 changes: 32 additions & 5 deletions llvm/lib/Target/PowerPC/PPCISelLowering.cpp
Expand Up @@ -204,11 +204,23 @@ PPCTargetLowering::PPCTargetLowering(const PPCTargetMachine &TM,
setOperationAction(ISD::FNEARBYINT, MVT::ppcf128, Expand);
setOperationAction(ISD::FREM, MVT::ppcf128, Expand);

// PowerPC has no SREM/UREM instructions
setOperationAction(ISD::SREM, MVT::i32, Expand);
setOperationAction(ISD::UREM, MVT::i32, Expand);
setOperationAction(ISD::SREM, MVT::i64, Expand);
setOperationAction(ISD::UREM, MVT::i64, Expand);
// PowerPC has no SREM/UREM instructions unless we are on P9
// On P9 we may use a hardware instruction to compute the remainder.
// The instructions are not legalized directly because in the cases where the
// result of both the remainder and the division is required it is more
// efficient to compute the remainder from the result of the division rather
// than use the remainder instruction.
if (Subtarget.isISA3_0()) {
setOperationAction(ISD::SREM, MVT::i32, Custom);
setOperationAction(ISD::UREM, MVT::i32, Custom);
setOperationAction(ISD::SREM, MVT::i64, Custom);
setOperationAction(ISD::UREM, MVT::i64, Custom);
} else {
setOperationAction(ISD::SREM, MVT::i32, Expand);
setOperationAction(ISD::UREM, MVT::i32, Expand);
setOperationAction(ISD::SREM, MVT::i64, Expand);
setOperationAction(ISD::UREM, MVT::i64, Expand);
}

// Don't use SMUL_LOHI/UMUL_LOHI or SDIVREM/UDIVREM to lower SREM/UREM.
setOperationAction(ISD::UMUL_LOHI, MVT::i32, Expand);
Expand Down Expand Up @@ -8394,6 +8406,18 @@ SDValue PPCTargetLowering::LowerINTRINSIC_VOID(SDValue Op,
return SDValue();
}

SDValue PPCTargetLowering::LowerREM(SDValue Op, SelectionDAG &DAG) const {
// Check for a DIV with the same operands as this REM.
for (auto UI : Op.getOperand(1)->uses()) {
if ((Op.getOpcode() == ISD::SREM && UI->getOpcode() == ISD::SDIV) ||
(Op.getOpcode() == ISD::UREM && UI->getOpcode() == ISD::UDIV))
if (UI->getOperand(0) == Op.getOperand(0) &&
UI->getOperand(1) == Op.getOperand(1))
return SDValue();
}
return Op;
}

SDValue PPCTargetLowering::LowerSIGN_EXTEND_INREG(SDValue Op,
SelectionDAG &DAG) const {
SDLoc dl(Op);
Expand Down Expand Up @@ -8862,6 +8886,9 @@ SDValue PPCTargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const {

case ISD::INTRINSIC_VOID:
return LowerINTRINSIC_VOID(Op, DAG);
case ISD::SREM:
case ISD::UREM:
return LowerREM(Op, DAG);
}
}

Expand Down
1 change: 1 addition & 0 deletions llvm/lib/Target/PowerPC/PPCISelLowering.h
Expand Up @@ -918,6 +918,7 @@ namespace llvm {
SDValue LowerEXTRACT_VECTOR_ELT(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerINTRINSIC_WO_CHAIN(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerINTRINSIC_VOID(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerREM(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerSCALAR_TO_VECTOR(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerSIGN_EXTEND_INREG(SDValue Op, SelectionDAG &DAG) const;
SDValue LowerMUL(SDValue Op, SelectionDAG &DAG) const;
Expand Down
10 changes: 10 additions & 0 deletions llvm/lib/Target/PowerPC/PPCInstr64Bit.td
Expand Up @@ -683,6 +683,16 @@ def DIVDE : XOForm_1<31, 425, 0, (outs g8rc:$rT), (ins g8rc:$rA, g8rc:$rB),
"divde $rT, $rA, $rB", IIC_IntDivD,
[(set i64:$rT, (int_ppc_divde g8rc:$rA, g8rc:$rB))]>,
isPPC64, Requires<[HasExtDiv]>;

let Predicates = [IsISA3_0] in {
def MODSD : XForm_8<31, 777, (outs g8rc:$rT), (ins g8rc:$rA, g8rc:$rB),
"modsd $rT, $rA, $rB", IIC_IntDivW,
[(set i64:$rT, (srem i64:$rA, i64:$rB))]>;
def MODUD : XForm_8<31, 265, (outs g8rc:$rT), (ins g8rc:$rA, g8rc:$rB),
"modud $rT, $rA, $rB", IIC_IntDivW,
[(set i64:$rT, (urem i64:$rA, i64:$rB))]>;
}

let Defs = [CR0] in
def DIVDEo : XOForm_1<31, 425, 0, (outs g8rc:$rT), (ins g8rc:$rA, g8rc:$rB),
"divde. $rT, $rA, $rB", IIC_IntDivD,
Expand Down
8 changes: 8 additions & 0 deletions llvm/lib/Target/PowerPC/PPCInstrInfo.td
Expand Up @@ -2544,6 +2544,14 @@ let Uses = [RM] in {
"mffs. $rT", IIC_IntMFFS, []>, isDOT;
}

let Predicates = [IsISA3_0] in {
def MODSW : XForm_8<31, 779, (outs gprc:$rT), (ins gprc:$rA, gprc:$rB),
"modsw $rT, $rA, $rB", IIC_IntDivW,
[(set i32:$rT, (srem i32:$rA, i32:$rB))]>;
def MODUW : XForm_8<31, 267, (outs gprc:$rT), (ins gprc:$rA, gprc:$rB),
"moduw $rT, $rA, $rB", IIC_IntDivW,
[(set i32:$rT, (urem i32:$rA, i32:$rB))]>;
}

let PPC970_Unit = 1, hasSideEffects = 0 in { // FXU Operations.
// XO-Form instructions. Arithmetic instructions that can set overflow bit
Expand Down
263 changes: 263 additions & 0 deletions llvm/test/CodeGen/PowerPC/ppc64-P9-mod.ll
@@ -0,0 +1,263 @@
; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr9 -verify-machineinstrs | FileCheck %s
; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr9 -verify-machineinstrs | FileCheck %s
; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr8 | FileCheck %s -check-prefix=CHECK-PWR8 -implicit-check-not mod[us][wd]

@mod_resultsw = common local_unnamed_addr global i32 0, align 4
@mod_resultud = common local_unnamed_addr global i64 0, align 8
@div_resultsw = common local_unnamed_addr global i32 0, align 4
@mod_resultuw = common local_unnamed_addr global i32 0, align 4
@div_resultuw = common local_unnamed_addr global i32 0, align 4
@div_resultsd = common local_unnamed_addr global i64 0, align 8
@mod_resultsd = common local_unnamed_addr global i64 0, align 8
@div_resultud = common local_unnamed_addr global i64 0, align 8

; Function Attrs: norecurse nounwind
define void @modulo_sw(i32 signext %a, i32 signext %b) local_unnamed_addr {
entry:
%rem = srem i32 %a, %b
store i32 %rem, i32* @mod_resultsw, align 4
ret void
; CHECK-LABEL: modulo_sw
; CHECK: modsw {{[0-9]+}}, 3, 4
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_sw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind readnone
define zeroext i32 @modulo_uw(i32 zeroext %a, i32 zeroext %b) local_unnamed_addr {
entry:
%rem = urem i32 %a, %b
ret i32 %rem
; CHECK-LABEL: modulo_uw
; CHECK: moduw {{[0-9]+}}, 3, 4
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_uw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind readnone
define i64 @modulo_sd(i64 %a, i64 %b) local_unnamed_addr {
entry:
%rem = srem i64 %a, %b
ret i64 %rem
; CHECK-LABEL: modulo_sd
; CHECK: modsd {{[0-9]+}}, 3, 4
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_sd
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
define void @modulo_ud(i64 %a, i64 %b) local_unnamed_addr {
entry:
%rem = urem i64 %a, %b
store i64 %rem, i64* @mod_resultud, align 8
ret void
; CHECK-LABEL: modulo_ud
; CHECK: modud {{[0-9]+}}, 3, 4
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_ud
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
define void @modulo_div_sw(i32 signext %a, i32 signext %b) local_unnamed_addr {
entry:
%rem = srem i32 %a, %b
store i32 %rem, i32* @mod_resultsw, align 4
%div = sdiv i32 %a, %b
store i32 %div, i32* @div_resultsw, align 4
ret void
; CHECK-LABEL: modulo_div_sw
; CHECK-NOT: modsw
; CHECK: div
; CHECK-NOT: modsw
; CHECK: mull
; CHECK-NOT: modsw
; CHECK: sub
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_div_sw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
define void @modulo_div_abc_sw(i32 signext %a, i32 signext %b, i32 signext %c) local_unnamed_addr {
entry:
%rem = srem i32 %a, %c
store i32 %rem, i32* @mod_resultsw, align 4
%div = sdiv i32 %b, %c
store i32 %div, i32* @div_resultsw, align 4
ret void
; CHECK-LABEL: modulo_div_abc_sw
; CHECK: modsw {{[0-9]+}}, 3, 5
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_div_abc_sw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
define void @modulo_div_uw(i32 zeroext %a, i32 zeroext %b) local_unnamed_addr {
entry:
%rem = urem i32 %a, %b
store i32 %rem, i32* @mod_resultuw, align 4
%div = udiv i32 %a, %b
store i32 %div, i32* @div_resultuw, align 4
ret void
; CHECK-LABEL: modulo_div_uw
; CHECK-NOT: modsw
; CHECK: div
; CHECK-NOT: modsw
; CHECK: mull
; CHECK-NOT: modsw
; CHECK: sub
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_div_uw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
define void @modulo_div_swuw(i32 signext %a, i32 signext %b) local_unnamed_addr {
entry:
%rem = srem i32 %a, %b
store i32 %rem, i32* @mod_resultsw, align 4
%div = udiv i32 %a, %b
store i32 %div, i32* @div_resultsw, align 4
ret void
; CHECK-LABEL: modulo_div_swuw
; CHECK: modsw {{[0-9]+}}, 3, 4
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_div_swuw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
define void @modulo_div_udsd(i64 %a, i64 %b) local_unnamed_addr {
entry:
%rem = urem i64 %a, %b
store i64 %rem, i64* @mod_resultud, align 8
%div = sdiv i64 %a, %b
store i64 %div, i64* @div_resultsd, align 8
ret void
; CHECK-LABEL: modulo_div_udsd
; CHECK: modud {{[0-9]+}}, 3, 4
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_div_udsd
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
define void @modulo_const32_sw(i32 signext %a) local_unnamed_addr {
entry:
%rem = srem i32 %a, 32
store i32 %rem, i32* @mod_resultsw, align 4
ret void
; CHECK-LABEL: modulo_const32_sw
; CHECK-NOT: modsw
; CHECK: srawi
; CHECK-NOT: modsw
; CHECK: addze
; CHECK-NOT: modsw
; CHECK: slwi
; CHECK-NOT: modsw
; CHECK: subf
; CHECK-NOT: modsw
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_const32_sw
; CHECK-PWR8: srawi
; CHECK-PWR8: addze
; CHECK-PWR8: slwi
; CHECK-PWR8: subf
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind readnone
define signext i32 @modulo_const3_sw(i32 signext %a) local_unnamed_addr {
entry:
%rem = srem i32 %a, 3
ret i32 %rem
; CHECK-LABEL: modulo_const3_sw
; CHECK-NOT: modsw
; CHECK: mull
; CHECK-NOT: modsw
; CHECK: sub
; CHECK-NOT: modsw
; CHECK: blr
; CHECK-PWR8-LABEL: modulo_const3_sw
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind readnone
define signext i32 @const2_modulo_sw(i32 signext %a) local_unnamed_addr {
entry:
%rem = srem i32 2, %a
ret i32 %rem
; CHECK-LABEL: const2_modulo_sw
; CHECK: modsw {{[0-9]+}}, {{[0-9]+}}, 3
; CHECK: blr
; CHECK-PWR8-LABEL: const2_modulo_sw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}

; Function Attrs: norecurse nounwind
; FIXME On power 9 this test will still produce modsw because the divide is in
; a different block than the remainder. Due to the nature of the SDAG we cannot
; see the div in the other block.
define void @blocks_modulo_div_sw(i32 signext %a, i32 signext %b, i32 signext %c) local_unnamed_addr {
entry:
%div = sdiv i32 %a, %b
store i32 %div, i32* @div_resultsw, align 4
%cmp = icmp sgt i32 %c, 0
br i1 %cmp, label %if.then, label %if.end

if.then: ; preds = %entry
%rem = srem i32 %a, %b
store i32 %rem, i32* @mod_resultsw, align 4
br label %if.end

if.end: ; preds = %if.then, %entry
ret void
; CHECK-LABEL: blocks_modulo_div_sw
; CHECK: div
; CHECK: modsw {{[0-9]+}}, 3, 4
; CHECK: blr
; CHECK-PWR8-LABEL: blocks_modulo_div_sw
; CHECK-PWR8: div
; CHECK-PWR8: mull
; CHECK-PWR8: sub
; CHECK-PWR8: blr
}


12 changes: 12 additions & 0 deletions llvm/test/MC/Disassembler/PowerPC/ppc64-encoding.txt
Expand Up @@ -352,6 +352,18 @@
# CHECK: divweu. 2, 3, 4
0x7c 0x43 0x23 0x17

# CHECK: modsw 2, 3, 4
0x7c 0x43 0x26 0x16

# CHECK: moduw 2, 3, 4
0x7c 0x43 0x22 0x16

# CHECK: modsd 2, 3, 4
0x7c 0x43 0x26 0x12

# CHECK: modud 2, 3, 4
0x7c 0x43 0x22 0x12

# CHECK: mulld 2, 3, 4
0x7c 0x43 0x21 0xd2

Expand Down
12 changes: 12 additions & 0 deletions llvm/test/MC/Disassembler/PowerPC/ppc64le-encoding.txt
Expand Up @@ -349,6 +349,18 @@
# CHECK: divweu. 2, 3, 4
0x17 0x23 0x43 0x7c

# CHECK: modsw 2, 3, 4
0x16 0x26 0x43 0x7c

# CHECK: moduw 2, 3, 4
0x16 0x22 0x43 0x7c

# CHECK: modsd 2, 3, 4
0x12 0x26 0x43 0x7c

# CHECK: modud 2, 3, 4
0x12 0x22 0x43 0x7c

# CHECK: mulld 2, 3, 4
0xd2 0x21 0x43 0x7c

Expand Down

0 comments on commit 30a49d1

Please sign in to comment.