Skip to content

G
golang
Commit 962dc4b4 authored by Michael Munday's avatar Michael Munday
Browse files
Options

cmd/compile: improve load/store merging on s390x 

This commit makes the process of load/store merging more incremental
for both big and little endian operations. It also adds support for
32-bit shifts (needed to merge 16- and 32-bit loads/stores).

In addition, the merging of little endian stores is now supported.
Little endian stores are now up to 30 times faster.

Change-Id: Iefdd81eda4a65b335f23c3ff222146540083ad9c
Reviewed-on: https://go-review.googlesource.com/29956
Run-TryBot: Michael Munday <munday@ca.ibm.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: 's avatarKeith Randall <khr@golang.org>
parent 41fa42b4
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 114 additions and 15 deletions
src/cmd/compile/internal/s390x/ssa.go
View file @ 962dc4b4
......@@ -368,6 +368,7 @@ func ssaGenValue(s *gc.SSAGenState, v *ssa.Value) {
p.To.Type = obj.TYPE_REG
p.To.Reg = v.Reg()
case ssa.OpS390XMOVBstore, ssa.OpS390XMOVHstore, ssa.OpS390XMOVWstore, ssa.OpS390XMOVDstore,
ssa.OpS390XMOVHBRstore, ssa.OpS390XMOVWBRstore, ssa.OpS390XMOVDBRstore,
ssa.OpS390XFMOVSstore, ssa.OpS390XFMOVDstore:
p := gc.Prog(v.Op.Asm())
p.From.Type = obj.TYPE_REG
......@@ -376,6 +377,7 @@ func ssaGenValue(s *gc.SSAGenState, v *ssa.Value) {
p.To.Reg = v.Args[0].Reg()
gc.AddAux(&p.To, v)
case ssa.OpS390XMOVBstoreidx, ssa.OpS390XMOVHstoreidx, ssa.OpS390XMOVWstoreidx, ssa.OpS390XMOVDstoreidx,
ssa.OpS390XMOVHBRstoreidx, ssa.OpS390XMOVWBRstoreidx, ssa.OpS390XMOVDBRstoreidx,
ssa.OpS390XFMOVSstoreidx, ssa.OpS390XFMOVDstoreidx:
r := v.Args[0].Reg()
i := v.Args[1].Reg()
......
src/cmd/compile/internal/ssa/gen/S390XOps.go
View file @ 962dc4b4
......@@ -133,6 +133,7 @@ func init() {
gpstore = regInfo{inputs: []regMask{ptrspsb, gpsp, 0}}
gpstoreconst = regInfo{inputs: []regMask{ptrspsb, 0}}
gpstoreidx = regInfo{inputs: []regMask{ptrsp, ptrsp, gpsp, 0}}
gpstorebr = regInfo{inputs: []regMask{ptrsp, gpsp, 0}}
gpmvc = regInfo{inputs: []regMask{ptrsp, ptrsp, 0}}
......@@ -322,26 +323,32 @@ func init() {
{name: "MOVWBRload", argLength: 2, reg: gpload, asm: "MOVWBR", aux: "SymOff", typ: "UInt32", clobberFlags: true, faultOnNilArg0: true}, // load 4 bytes from arg0+auxint+aux. arg1=mem. Reverse bytes.
{name: "MOVDBRload", argLength: 2, reg: gpload, asm: "MOVDBR", aux: "SymOff", typ: "UInt64", clobberFlags: true, faultOnNilArg0: true}, // load 8 bytes from arg0+auxint+aux. arg1=mem. Reverse bytes.
{name: "MOVBstore", argLength: 3, reg: gpstore, asm: "MOVB", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store byte in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVHstore", argLength: 3, reg: gpstore, asm: "MOVH", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 2 bytes in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVWstore", argLength: 3, reg: gpstore, asm: "MOVW", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 4 bytes in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVDstore", argLength: 3, reg: gpstore, asm: "MOVD", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 8 bytes in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVBstore", argLength: 3, reg: gpstore, asm: "MOVB", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store byte in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVHstore", argLength: 3, reg: gpstore, asm: "MOVH", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 2 bytes in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVWstore", argLength: 3, reg: gpstore, asm: "MOVW", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 4 bytes in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVDstore", argLength: 3, reg: gpstore, asm: "MOVD", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 8 bytes in arg1 to arg0+auxint+aux. arg2=mem
{name: "MOVHBRstore", argLength: 3, reg: gpstorebr, asm: "MOVHBR", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 2 bytes in arg1 to arg0+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVWBRstore", argLength: 3, reg: gpstorebr, asm: "MOVWBR", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 4 bytes in arg1 to arg0+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVDBRstore", argLength: 3, reg: gpstorebr, asm: "MOVDBR", aux: "SymOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true}, // store 8 bytes in arg1 to arg0+auxint+aux. arg2=mem. Reverse bytes.
{name: "MVC", argLength: 3, reg: gpmvc, asm: "MVC", aux: "SymValAndOff", typ: "Mem", clobberFlags: true, faultOnNilArg0: true, faultOnNilArg1: true}, // arg0=destptr, arg1=srcptr, arg2=mem, auxint=size,off
// indexed loads/stores
// TODO(mundaym): add sign-extended indexed loads
{name: "MOVBZloadidx", argLength: 3, reg: gploadidx, asm: "MOVBZ", aux: "SymOff", clobberFlags: true}, // load a byte from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVHZloadidx", argLength: 3, reg: gploadidx, asm: "MOVHZ", aux: "SymOff", clobberFlags: true}, // load 2 bytes from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVWZloadidx", argLength: 3, reg: gploadidx, asm: "MOVWZ", aux: "SymOff", clobberFlags: true}, // load 4 bytes from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVDloadidx", argLength: 3, reg: gploadidx, asm: "MOVD", aux: "SymOff", clobberFlags: true}, // load 8 bytes from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVHBRloadidx", argLength: 3, reg: gploadidx, asm: "MOVHBR", aux: "SymOff", clobberFlags: true}, // load 2 bytes from arg0+arg1+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVWBRloadidx", argLength: 3, reg: gploadidx, asm: "MOVWBR", aux: "SymOff", clobberFlags: true}, // load 4 bytes from arg0+arg1+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVDBRloadidx", argLength: 3, reg: gploadidx, asm: "MOVDBR", aux: "SymOff", clobberFlags: true}, // load 8 bytes from arg0+arg1+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVBstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVB", aux: "SymOff", clobberFlags: true}, // store byte in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVHstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVH", aux: "SymOff", clobberFlags: true}, // store 2 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVWstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVW", aux: "SymOff", clobberFlags: true}, // store 4 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVDstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVD", aux: "SymOff", clobberFlags: true}, // store 8 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVBZloadidx", argLength: 3, reg: gploadidx, asm: "MOVBZ", aux: "SymOff", clobberFlags: true}, // load a byte from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVHZloadidx", argLength: 3, reg: gploadidx, asm: "MOVHZ", aux: "SymOff", clobberFlags: true}, // load 2 bytes from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVWZloadidx", argLength: 3, reg: gploadidx, asm: "MOVWZ", aux: "SymOff", clobberFlags: true}, // load 4 bytes from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVDloadidx", argLength: 3, reg: gploadidx, asm: "MOVD", aux: "SymOff", clobberFlags: true}, // load 8 bytes from arg0+arg1+auxint+aux. arg2=mem
{name: "MOVHBRloadidx", argLength: 3, reg: gploadidx, asm: "MOVHBR", aux: "SymOff", clobberFlags: true}, // load 2 bytes from arg0+arg1+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVWBRloadidx", argLength: 3, reg: gploadidx, asm: "MOVWBR", aux: "SymOff", clobberFlags: true}, // load 4 bytes from arg0+arg1+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVDBRloadidx", argLength: 3, reg: gploadidx, asm: "MOVDBR", aux: "SymOff", clobberFlags: true}, // load 8 bytes from arg0+arg1+auxint+aux. arg2=mem. Reverse bytes.
{name: "MOVBstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVB", aux: "SymOff", clobberFlags: true}, // store byte in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVHstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVH", aux: "SymOff", clobberFlags: true}, // store 2 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVWstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVW", aux: "SymOff", clobberFlags: true}, // store 4 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVDstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVD", aux: "SymOff", clobberFlags: true}, // store 8 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem
{name: "MOVHBRstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVHBR", aux: "SymOff", clobberFlags: true}, // store 2 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem. Reverse bytes.
{name: "MOVWBRstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVWBR", aux: "SymOff", clobberFlags: true}, // store 4 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem. Reverse bytes.
{name: "MOVDBRstoreidx", argLength: 4, reg: gpstoreidx, asm: "MOVDBR", aux: "SymOff", clobberFlags: true}, // store 8 bytes in arg2 to arg0+arg1+auxint+aux. arg3=mem. Reverse bytes.
// For storeconst ops, the AuxInt field encodes both
// the value to store and an address offset of the store.
......
src/cmd/compile/internal/ssa/opGen.go
View file @ 962dc4b4
......@@ -1359,6 +1359,9 @@ const (
OpS390XMOVHstore
OpS390XMOVWstore
OpS390XMOVDstore
OpS390XMOVHBRstore
OpS390XMOVWBRstore
OpS390XMOVDBRstore
OpS390XMVC
OpS390XMOVBZloadidx
OpS390XMOVHZloadidx
......@@ -1371,6 +1374,9 @@ const (
OpS390XMOVHstoreidx
OpS390XMOVWstoreidx
OpS390XMOVDstoreidx
OpS390XMOVHBRstoreidx
OpS390XMOVWBRstoreidx
OpS390XMOVDBRstoreidx
OpS390XMOVBstoreconst
OpS390XMOVHstoreconst
OpS390XMOVWstoreconst
......@@ -17188,6 +17194,48 @@ var opcodeTable = [...]opInfo{
},
},
},
{
name: "MOVHBRstore",
auxType: auxSymOff,
argLen: 3,
clobberFlags: true,
faultOnNilArg0: true,
asm: s390x.AMOVHBR,
reg: regInfo{
inputs: []inputInfo{
{0, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{1, 37887}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
},
},
},
{
name: "MOVWBRstore",
auxType: auxSymOff,
argLen: 3,
clobberFlags: true,
faultOnNilArg0: true,
asm: s390x.AMOVWBR,
reg: regInfo{
inputs: []inputInfo{
{0, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{1, 37887}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
},
},
},
{
name: "MOVDBRstore",
auxType: auxSymOff,
argLen: 3,
clobberFlags: true,
faultOnNilArg0: true,
asm: s390x.AMOVDBR,
reg: regInfo{
inputs: []inputInfo{
{0, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{1, 37887}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
},
},
},
{
name: "MVC",
auxType: auxSymValAndOff,
......@@ -17371,6 +17419,48 @@ var opcodeTable = [...]opInfo{
},
},
},
{
name: "MOVHBRstoreidx",
auxType: auxSymOff,
argLen: 4,
clobberFlags: true,
asm: s390x.AMOVHBR,
reg: regInfo{
inputs: []inputInfo{
{0, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{1, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{2, 37887}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
},
},
},
{
name: "MOVWBRstoreidx",
auxType: auxSymOff,
argLen: 4,
clobberFlags: true,
asm: s390x.AMOVWBR,
reg: regInfo{
inputs: []inputInfo{
{0, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{1, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{2, 37887}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
},
},
},
{
name: "MOVDBRstoreidx",
auxType: auxSymOff,
argLen: 4,
clobberFlags: true,
asm: s390x.AMOVDBR,
reg: regInfo{
inputs: []inputInfo{
{0, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{1, 37886}, // R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
{2, 37887}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R12 SP
},
},
},
{
name: "MOVBstoreconst",
auxType: auxSymValAndOff,
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment