go/src/cmd/compile/internal/gc/syntax.go

632 lines
20 KiB
Go
Raw Normal View History

// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// “Abstract” syntax representation.
package gc
import (
"cmd/compile/internal/syntax"
"cmd/internal/src"
)
// A Node is a single node in the syntax tree.
// Actually the syntax tree is a syntax DAG, because there is only one
// node with Op=ONAME for a given instance of a variable x.
// The same is true for Op=OTYPE and Op=OLITERAL.
type Node struct {
// Tree structure.
// Generic recursive walks should follow these fields.
Left *Node
Right *Node
Ninit Nodes
Nbody Nodes
List Nodes
Rlist Nodes
// most nodes
Type *Type
Orig *Node // original form, for printing, and tracking copies of ONAMEs
// func
Func *Func
// ONAME, OTYPE, OPACK, OLABEL, some OLITERAL
Name *Name
Sym *Sym // various
E interface{} // Opt or Val, see methods below
cmd/compile: add OSTRUCTKEY for keyed struct literals Previously, we used OKEY nodes to represent keyed struct literal elements. The field names were represented by an ONAME node, but this is clumsy because it's the only remaining case where ONAME was used to represent a bare identifier and not a variable. This CL introduces a new OSTRUCTKEY node op for use in struct literals. These ops instead store the field name in the node's own Sym field. This is similar in spirit to golang.org/cl/20890. Significant reduction in allocations for struct literal heavy code like package unicode: name old time/op new time/op delta Template 345ms ± 6% 341ms ± 6% ~ (p=0.141 n=29+28) Unicode 200ms ± 9% 184ms ± 7% -7.77% (p=0.000 n=29+30) GoTypes 1.04s ± 3% 1.05s ± 3% ~ (p=0.096 n=30+30) Compiler 4.47s ± 9% 4.49s ± 6% ~ (p=0.890 n=29+29) name old user-ns/op new user-ns/op delta Template 523M ±13% 516M ±17% ~ (p=0.400 n=29+30) Unicode 334M ±27% 314M ±30% ~ (p=0.093 n=30+30) GoTypes 1.53G ±10% 1.52G ±10% ~ (p=0.572 n=30+30) Compiler 6.28G ± 7% 6.34G ±11% ~ (p=0.300 n=30+30) name old alloc/op new alloc/op delta Template 44.5MB ± 0% 44.4MB ± 0% -0.35% (p=0.000 n=27+30) Unicode 39.2MB ± 0% 34.5MB ± 0% -11.79% (p=0.000 n=26+30) GoTypes 125MB ± 0% 125MB ± 0% -0.12% (p=0.000 n=29+30) Compiler 515MB ± 0% 515MB ± 0% -0.10% (p=0.000 n=29+30) name old allocs/op new allocs/op delta Template 426k ± 0% 424k ± 0% -0.39% (p=0.000 n=29+30) Unicode 374k ± 0% 323k ± 0% -13.67% (p=0.000 n=29+30) GoTypes 1.21M ± 0% 1.21M ± 0% -0.14% (p=0.000 n=29+29) Compiler 4.40M ± 0% 4.39M ± 0% -0.13% (p=0.000 n=29+30) Passes toolstash/buildall. Change-Id: Iba4ee765dd1748f67e52fcade1cd75c9f6e13fa9 Reviewed-on: https://go-review.googlesource.com/30974 Run-TryBot: Matthew Dempsky <mdempsky@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2016-10-12 15:48:18 -07:00
// Various. Usually an offset into a struct. For example:
// - ONAME nodes that refer to local variables use it to identify their stack frame position.
// - ODOT, ODOTPTR, and OINDREGSP use it to indicate offset relative to their base address.
cmd/compile: add OSTRUCTKEY for keyed struct literals Previously, we used OKEY nodes to represent keyed struct literal elements. The field names were represented by an ONAME node, but this is clumsy because it's the only remaining case where ONAME was used to represent a bare identifier and not a variable. This CL introduces a new OSTRUCTKEY node op for use in struct literals. These ops instead store the field name in the node's own Sym field. This is similar in spirit to golang.org/cl/20890. Significant reduction in allocations for struct literal heavy code like package unicode: name old time/op new time/op delta Template 345ms ± 6% 341ms ± 6% ~ (p=0.141 n=29+28) Unicode 200ms ± 9% 184ms ± 7% -7.77% (p=0.000 n=29+30) GoTypes 1.04s ± 3% 1.05s ± 3% ~ (p=0.096 n=30+30) Compiler 4.47s ± 9% 4.49s ± 6% ~ (p=0.890 n=29+29) name old user-ns/op new user-ns/op delta Template 523M ±13% 516M ±17% ~ (p=0.400 n=29+30) Unicode 334M ±27% 314M ±30% ~ (p=0.093 n=30+30) GoTypes 1.53G ±10% 1.52G ±10% ~ (p=0.572 n=30+30) Compiler 6.28G ± 7% 6.34G ±11% ~ (p=0.300 n=30+30) name old alloc/op new alloc/op delta Template 44.5MB ± 0% 44.4MB ± 0% -0.35% (p=0.000 n=27+30) Unicode 39.2MB ± 0% 34.5MB ± 0% -11.79% (p=0.000 n=26+30) GoTypes 125MB ± 0% 125MB ± 0% -0.12% (p=0.000 n=29+30) Compiler 515MB ± 0% 515MB ± 0% -0.10% (p=0.000 n=29+30) name old allocs/op new allocs/op delta Template 426k ± 0% 424k ± 0% -0.39% (p=0.000 n=29+30) Unicode 374k ± 0% 323k ± 0% -13.67% (p=0.000 n=29+30) GoTypes 1.21M ± 0% 1.21M ± 0% -0.14% (p=0.000 n=29+29) Compiler 4.40M ± 0% 4.39M ± 0% -0.13% (p=0.000 n=29+30) Passes toolstash/buildall. Change-Id: Iba4ee765dd1748f67e52fcade1cd75c9f6e13fa9 Reviewed-on: https://go-review.googlesource.com/30974 Run-TryBot: Matthew Dempsky <mdempsky@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2016-10-12 15:48:18 -07:00
// - OSTRUCTKEY uses it to store the named field's offset.
// - OXCASE and OXFALL use it to validate the use of fallthrough.
// - Named OLITERALs use it to to store their ambient iota value.
// Possibly still more uses. If you find any, document them.
Xoffset int64
2016-12-15 17:17:01 -08:00
Pos src.XPos
Esc uint16 // EscXXX
Op Op
Ullman uint8 // sethi/ullman number
Addable bool // addressable
Etype EType // op for OASOP, etype for OTYPE, exclam for export, 6g saved reg, ChanDir for OTCHAN, for OINDEXMAP 1=LHS,0=RHS
Bounded bool // bounds check unnecessary
NonNil bool // guaranteed to be non-nil
Class Class // PPARAM, PAUTO, PEXTERN, etc
Embedded uint8 // ODCLFIELD embedded type
Colas bool // OAS resulting from :=
Diag bool // already printed error about this
Noescape bool // func arguments do not escape; TODO(rsc): move Noescape to Func struct (see CL 7360)
Walkdef uint8 // tracks state during typecheckdef; 2 == loop detected
Typecheck uint8 // tracks state during typechecking; 2 == loop detected
Local bool // type created in this file (see also Type.Local); TODO(gri): move this into flags
IsStatic bool // whether this Node will be converted to purely static data
Initorder uint8
Used bool // for variable/label declared and not used error
Isddd bool // is the argument variadic
Implicit bool
Addrtaken bool // address taken, even if not moved to heap
Assigned bool // is the variable ever assigned to
Likely int8 // likeliness of if statement
hasVal int8 // +1 for Val, -1 for Opt, 0 for not yet set
flags uint8 // TODO: store more bool fields in this flag field
}
// IsAutoTmp indicates if n was created by the compiler as a temporary,
// based on the setting of the .AutoTemp flag in n's Name.
func (n *Node) IsAutoTmp() bool {
if n == nil || n.Op != ONAME {
return false
}
return n.Name.AutoTemp
}
const (
hasBreak = 1 << iota
isClosureVar
isOutputParamHeapAddr
noInline // used internally by inliner to indicate that a function call should not be inlined; set for OCALLFUNC and OCALLMETH only
)
func (n *Node) HasBreak() bool {
return n.flags&hasBreak != 0
}
func (n *Node) SetHasBreak(b bool) {
if b {
n.flags |= hasBreak
} else {
n.flags &^= hasBreak
}
}
func (n *Node) isClosureVar() bool {
return n.flags&isClosureVar != 0
}
func (n *Node) setIsClosureVar(b bool) {
if b {
n.flags |= isClosureVar
} else {
n.flags &^= isClosureVar
}
}
func (n *Node) noInline() bool {
return n.flags&noInline != 0
}
func (n *Node) setNoInline(b bool) {
if b {
n.flags |= noInline
} else {
n.flags &^= noInline
}
}
func (n *Node) IsOutputParamHeapAddr() bool {
return n.flags&isOutputParamHeapAddr != 0
}
func (n *Node) setIsOutputParamHeapAddr(b bool) {
if b {
n.flags |= isOutputParamHeapAddr
} else {
n.flags &^= isOutputParamHeapAddr
}
}
// Val returns the Val for the node.
func (n *Node) Val() Val {
if n.hasVal != +1 {
return Val{}
}
return Val{n.E}
}
// SetVal sets the Val for the node, which must not have been used with SetOpt.
func (n *Node) SetVal(v Val) {
if n.hasVal == -1 {
Debug['h'] = 1
Dump("have Opt", n)
Fatalf("have Opt")
}
n.hasVal = +1
n.E = v.U
}
// Opt returns the optimizer data for the node.
func (n *Node) Opt() interface{} {
if n.hasVal != -1 {
return nil
}
return n.E
}
// SetOpt sets the optimizer data for the node, which must not have been used with SetVal.
// SetOpt(nil) is ignored for Vals to simplify call sites that are clearing Opts.
func (n *Node) SetOpt(x interface{}) {
if x == nil && n.hasVal >= 0 {
return
}
if n.hasVal == +1 {
Debug['h'] = 1
Dump("have Val", n)
Fatalf("have Val")
}
n.hasVal = -1
n.E = x
}
func (n *Node) Iota() int64 {
return n.Xoffset
}
func (n *Node) SetIota(x int64) {
n.Xoffset = x
}
// Name holds Node fields used only by named nodes (ONAME, OTYPE, OPACK, OLABEL, some OLITERAL).
type Name struct {
Pack *Node // real package for import . names
Pkg *Pkg // pkg for OPACK nodes
Heapaddr *Node // temp holding heap address of param (could move to Param?)
Defn *Node // initializing assignment
Curfn *Node // function for local variables
Param *Param // additional fields for ONAME, OTYPE
Decldepth int32 // declaration loop depth, increased for every loop or label
Vargen int32 // unique name for ONAME within a function. Function outputs are numbered starting at one.
Funcdepth int32
Readonly bool
Captured bool // is the variable captured by a closure
Byval bool // is the variable captured by value or by reference
Needzero bool // if it contains pointers, needs to be zeroed on function entry
cmd/compile: recognize Syscall-like functions for liveness analysis Consider this code: func f(*int) func g() { p := new(int) f(p) } where f is an assembly function. In general liveness analysis assumes that during the call to f, p is dead in this frame. If f has retained p, p will be found alive in f's frame and keep the new(int) from being garbage collected. This is all correct and works. We use the Go func declaration for f to give the assembly function liveness information (the arguments are assumed live for the entire call). Now consider this code: func h1() { p := new(int) syscall.Syscall(1, 2, 3, uintptr(unsafe.Pointer(p))) } Here syscall.Syscall is taking the place of f, but because its arguments are uintptr, the liveness analysis and the garbage collector ignore them. Since p is no longer live in h once the call starts, if the garbage collector scans the stack while the system call is blocked, it will find no reference to the new(int) and reclaim it. If the kernel is going to write to *p once the call finishes, reclaiming the memory is a mistake. We can't change the arguments or the liveness information for syscall.Syscall itself, both for compatibility and because sometimes the arguments really are integers, and the garbage collector will get quite upset if it finds an integer where it expects a pointer. The problem is that these arguments are fundamentally untyped. The solution we have taken in the syscall package's wrappers in past releases is to insert a call to a dummy function named "use", to make it look like the argument is live during the call to syscall.Syscall: func h2() { p := new(int) syscall.Syscall(1, 2, 3, uintptr(unsafe.Pointer(p))) use(unsafe.Pointer(p)) } Keeping p alive during the call means that if the garbage collector scans the stack during the system call now, it will find the reference to p. Unfortunately, this approach is not available to users outside syscall, because 'use' is unexported, and people also have to realize they need to use it and do so. There is much existing code using syscall.Syscall without a 'use'-like function. That code will fail very occasionally in mysterious ways (see #13372). This CL fixes all that existing code by making the compiler do the right thing automatically, without any code modifications. That is, it takes h1 above, which is incorrect code today, and makes it correct code. Specifically, if the compiler sees a foreign func definition (one without a body) that has uintptr arguments, it marks those arguments as "unsafe uintptrs". If it later sees the function being called with uintptr(unsafe.Pointer(x)) as an argument, it arranges to mark x as having escaped, and it makes sure to hold x in a live temporary variable until the call returns, so that the garbage collector cannot reclaim whatever heap memory x points to. For now I am leaving the explicit calls to use in package syscall, but they can be removed early in a future cycle (likely Go 1.7). The rule has no effect on escape analysis, only on liveness analysis. Fixes #13372. Change-Id: I2addb83f70d08db08c64d394f9d06ff0a063c500 Reviewed-on: https://go-review.googlesource.com/18584 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2016-01-13 00:46:28 -05:00
Keepalive bool // mark value live across unknown assembly call
AutoTemp bool // is the variable a temporary (implies no dwarf info. reset if escapes to heap)
}
type Param struct {
Ntype *Node
cmd/compile: fix liveness computation for heap-escaped parameters The liveness computation of parameters generally was never correct, but forcing all parameters to be live throughout the function covered up that problem. The new SSA back end is too clever: even though it currently keeps the parameter values live throughout the function, it may find optimizations that mean the current values are not written back to the original parameter stack slots immediately or ever (for example if a parameter is set to nil, SSA constant propagation may replace all later uses of the parameter with a constant nil, eliminating the need to write the nil value back to the stack slot), so the liveness code must now track the actual operations on the stack slots, exposing these problems. One small problem in the handling of arguments is that nodarg can return ONAME PPARAM nodes with adjusted offsets, so that there are actually multiple *Node pointers for the same parameter in the instruction stream. This might be possible to correct, but not in this CL. For now, we fix this by using n.Orig instead of n when considering PPARAM and PPARAMOUT nodes. The major problem in the handling of arguments is general confusion in the liveness code about the meaning of PPARAM|PHEAP and PPARAMOUT|PHEAP nodes, especially as contrasted with PAUTO|PHEAP. The difference between these two is that when a local variable "moves" to the heap, it's really just allocated there to start with; in contrast, when an argument moves to the heap, the actual data has to be copied there from the stack at the beginning of the function, and when a result "moves" to the heap the value in the heap has to be copied back to the stack when the function returns This general confusion is also present in the SSA back end. The PHEAP bit worked decently when I first introduced it 7 years ago (!) in 391425ae. The back end did nothing sophisticated, and in particular there was no analysis at all: no escape analysis, no liveness analysis, and certainly no SSA back end. But the complications caused in the various downstream consumers suggest that this should be a detail kept mainly in the front end. This CL therefore eliminates both the PHEAP bit and even the idea of "heap variables" from the back ends. First, it replaces the PPARAM|PHEAP, PPARAMOUT|PHEAP, and PAUTO|PHEAP variable classes with the single PAUTOHEAP, a pseudo-class indicating a variable maintained on the heap and available by indirecting a local variable kept on the stack (a plain PAUTO). Second, walkexpr replaces all references to PAUTOHEAP variables with indirections of the corresponding PAUTO variable. The back ends and the liveness code now just see plain indirected variables. This may actually produce better code, but the real goal here is to eliminate these little-used and somewhat suspect code paths in the back end analyses. The OPARAM node type goes away too. A followup CL will do the same to PPARAMREF. I'm not sure that the back ends (SSA in particular) are handling those right either, and with the framework established in this CL that change is trivial and the result clearly more correct. Fixes #15747. Change-Id: I2770b1ce3cbc93981bfc7166be66a9da12013d74 Reviewed-on: https://go-review.googlesource.com/23393 Reviewed-by: Keith Randall <khr@golang.org> Run-TryBot: Russ Cox <rsc@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org>
2016-05-25 01:33:24 -04:00
// ONAME PAUTOHEAP
Stackcopy *Node // the PPARAM/PPARAMOUT on-stack slot (moved func params only)
// ONAME PPARAM
Field *Field // TFIELD in arg struct
// ONAME closure linkage
// Consider:
//
// func f() {
// x := 1 // x1
// func() {
// use(x) // x2
// func() {
// use(x) // x3
// --- parser is here ---
// }()
// }()
// }
//
// There is an original declaration of x and then a chain of mentions of x
// leading into the current function. Each time x is mentioned in a new closure,
// we create a variable representing x for use in that specific closure,
// since the way you get to x is different in each closure.
//
// Let's number the specific variables as shown in the code:
// x1 is the original x, x2 is when mentioned in the closure,
// and x3 is when mentioned in the closure in the closure.
//
// We keep these linked (assume N > 1):
//
// - x1.Defn = original declaration statement for x (like most variables)
// - x1.Innermost = current innermost closure x (in this case x3), or nil for none
// - x1.isClosureVar() = false
//
// - xN.Defn = x1, N > 1
// - xN.isClosureVar() = true, N > 1
// - x2.Outer = nil
// - xN.Outer = x(N-1), N > 2
//
//
// When we look up x in the symbol table, we always get x1.
// Then we can use x1.Innermost (if not nil) to get the x
// for the innermost known closure function,
// but the first reference in a closure will find either no x1.Innermost
// or an x1.Innermost with .Funcdepth < Funcdepth.
// In that case, a new xN must be created, linked in with:
//
// xN.Defn = x1
// xN.Outer = x1.Innermost
// x1.Innermost = xN
//
// When we finish the function, we'll process its closure variables
// and find xN and pop it off the list using:
//
// x1 := xN.Defn
// x1.Innermost = xN.Outer
//
// We leave xN.Innermost set so that we can still get to the original
// variable quickly. Not shown here, but once we're
// done parsing a function and no longer need xN.Outer for the
// lexical x reference links as described above, closurebody
// recomputes xN.Outer as the semantic x reference link tree,
// even filling in x in intermediate closures that might not
// have mentioned it along the way to inner closures that did.
// See closurebody for details.
//
// During the eventual compilation, then, for closure variables we have:
//
// xN.Defn = original variable
// xN.Outer = variable captured in next outward scope
// to make closure where xN appears
//
// Because of the sharding of pieces of the node, x.Defn means x.Name.Defn
// and x.Innermost/Outer means x.Name.Param.Innermost/Outer.
Innermost *Node
Outer *Node
cmd/compile: add go:notinheap type pragma This adds a //go:notinheap pragma for declarations of types that must not be heap allocated. We ensure these rules by disallowing new(T), make([]T), append([]T), or implicit allocation of T, by disallowing conversions to notinheap types, and by propagating notinheap to any struct or array that contains notinheap elements. The utility of this pragma is that we can eliminate write barriers for writes to pointers to go:notinheap types, since the write barrier is guaranteed to be a no-op. This will let us mark several scheduler and memory allocator structures as go:notinheap, which will let us disallow write barriers in the scheduler and memory allocator much more thoroughly and also eliminate some problematic hybrid write barriers. This also makes go:nowritebarrierrec and go:yeswritebarrierrec much more powerful. Currently we use go:nowritebarrier all over the place, but it's almost never what you actually want: when write barriers are illegal, they're typically illegal for a whole dynamic scope. Partly this is because go:nowritebarrier has been around longer, but it's also because go:nowritebarrierrec couldn't be used in situations that had no-op write barriers or where some nested scope did allow write barriers. go:notinheap eliminates many no-op write barriers and go:yeswritebarrierrec makes it possible to opt back in to write barriers, so these two changes will let us use go:nowritebarrierrec far more liberally. This updates #13386, which is about controlling pointers from non-GC'd memory to GC'd memory. That would require some additional pragma (or pragmas), but could build on this pragma. Change-Id: I6314f8f4181535dd166887c9ec239977b54940bd Reviewed-on: https://go-review.googlesource.com/30939 Reviewed-by: Keith Randall <khr@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
2016-10-11 22:53:27 -04:00
// OTYPE
cmd/compile: add go:notinheap type pragma This adds a //go:notinheap pragma for declarations of types that must not be heap allocated. We ensure these rules by disallowing new(T), make([]T), append([]T), or implicit allocation of T, by disallowing conversions to notinheap types, and by propagating notinheap to any struct or array that contains notinheap elements. The utility of this pragma is that we can eliminate write barriers for writes to pointers to go:notinheap types, since the write barrier is guaranteed to be a no-op. This will let us mark several scheduler and memory allocator structures as go:notinheap, which will let us disallow write barriers in the scheduler and memory allocator much more thoroughly and also eliminate some problematic hybrid write barriers. This also makes go:nowritebarrierrec and go:yeswritebarrierrec much more powerful. Currently we use go:nowritebarrier all over the place, but it's almost never what you actually want: when write barriers are illegal, they're typically illegal for a whole dynamic scope. Partly this is because go:nowritebarrier has been around longer, but it's also because go:nowritebarrierrec couldn't be used in situations that had no-op write barriers or where some nested scope did allow write barriers. go:notinheap eliminates many no-op write barriers and go:yeswritebarrierrec makes it possible to opt back in to write barriers, so these two changes will let us use go:nowritebarrierrec far more liberally. This updates #13386, which is about controlling pointers from non-GC'd memory to GC'd memory. That would require some additional pragma (or pragmas), but could build on this pragma. Change-Id: I6314f8f4181535dd166887c9ec239977b54940bd Reviewed-on: https://go-review.googlesource.com/30939 Reviewed-by: Keith Randall <khr@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
2016-10-11 22:53:27 -04:00
//
// TODO: Should Func pragmas also be stored on the Name?
Pragma syntax.Pragma
Alias bool // node is alias for Ntype (only used when type-checking ODCLTYPE)
}
// Func holds Node fields used only with function-like nodes.
type Func struct {
Shortname *Sym
Enter Nodes // for example, allocate and initialize memory for escaping parameters
Exit Nodes
Cvars Nodes // closure params
Dcl []*Node // autodcl for this func/closure
Inldcl Nodes // copy of dcl for use in inlining
Closgen int
Outerfunc *Node // outer function (for closure)
FieldTrack map[*Sym]struct{}
Ntype *Node // signature
Top int // top context (Ecall, Eproc, etc)
Closure *Node // OCLOSURE <-> ODCLFUNC
Nname *Node
Inl Nodes // copy of the body for use in inlining
InlCost int32
Depth int32
Label int32 // largest auto-generated label in this function
2016-12-15 17:17:01 -08:00
Endlineno src.XPos
WBPos src.XPos // position of first write barrier
Pragma syntax.Pragma // go:xxx function annotations
Dupok bool // duplicate definitions ok
Wrapper bool // is method wrapper
Needctxt bool // function uses context register (has closure variables)
ReflectMethod bool // function calls reflect.Type.Method or MethodByName
IsHiddenClosure bool
NoFramePointer bool // Must not use a frame pointer for this function
}
type Op uint8
// Node ops.
const (
OXXX = Op(iota)
// names
ONAME // var, const or func name
ONONAME // unnamed arg or return value: f(int, string) (int, error) { etc }
OTYPE // type name
OPACK // import
OLITERAL // literal
// expressions
OADD // Left + Right
OSUB // Left - Right
OOR // Left | Right
OXOR // Left ^ Right
OADDSTR // +{List} (string addition, list elements are strings)
OADDR // &Left
OANDAND // Left && Right
OAPPEND // append(List)
OARRAYBYTESTR // Type(Left) (Type is string, Left is a []byte)
OARRAYBYTESTRTMP // Type(Left) (Type is string, Left is a []byte, ephemeral)
OARRAYRUNESTR // Type(Left) (Type is string, Left is a []rune)
OSTRARRAYBYTE // Type(Left) (Type is []byte, Left is a string)
OSTRARRAYBYTETMP // Type(Left) (Type is []byte, Left is a string, ephemeral)
OSTRARRAYRUNE // Type(Left) (Type is []rune, Left is a string)
OAS // Left = Right or (if Colas=true) Left := Right
OAS2 // List = Rlist (x, y, z = a, b, c)
OAS2FUNC // List = Rlist (x, y = f())
OAS2RECV // List = Rlist (x, ok = <-c)
OAS2MAPR // List = Rlist (x, ok = m["foo"])
OAS2DOTTYPE // List = Rlist (x, ok = I.(int))
OASOP // Left Etype= Right (x += y)
OASWB // Left = Right (with write barrier)
OCALL // Left(List) (function call, method call or type conversion)
OCALLFUNC // Left(List) (function call f(args))
OCALLMETH // Left(List) (direct method call x.Method(args))
OCALLINTER // Left(List) (interface method call x.Method(args))
OCALLPART // Left.Right (method expression x.Method, not called)
OCAP // cap(Left)
OCLOSE // close(Left)
OCLOSURE // func Type { Body } (func literal)
OCMPIFACE // Left Etype Right (interface comparison, x == y or x != y)
OCMPSTR // Left Etype Right (string comparison, x == y, x < y, etc)
OCOMPLIT // Right{List} (composite literal, not yet lowered to specific form)
OMAPLIT // Type{List} (composite literal, Type is map)
OSTRUCTLIT // Type{List} (composite literal, Type is struct)
OARRAYLIT // Type{List} (composite literal, Type is array)
OSLICELIT // Type{List} (composite literal, Type is slice)
OPTRLIT // &Left (left is composite literal)
OCONV // Type(Left) (type conversion)
OCONVIFACE // Type(Left) (type conversion, to interface)
OCONVNOP // Type(Left) (type conversion, no effect)
OCOPY // copy(Left, Right)
ODCL // var Left (declares Left of type Left.Type)
// Used during parsing but don't last.
ODCLFUNC // func f() or func (r) f()
ODCLFIELD // struct field, interface field, or func/method argument/return value.
ODCLCONST // const pi = 3.14
ODCLTYPE // type Int int or type Int = int
ODELETE // delete(Left, Right)
ODOT // Left.Sym (Left is of struct type)
ODOTPTR // Left.Sym (Left is of pointer to struct type)
ODOTMETH // Left.Sym (Left is non-interface, Right is method name)
ODOTINTER // Left.Sym (Left is interface, Right is method name)
OXDOT // Left.Sym (before rewrite to one of the preceding)
ODOTTYPE // Left.Right or Left.Type (.Right during parsing, .Type once resolved)
ODOTTYPE2 // Left.Right or Left.Type (.Right during parsing, .Type once resolved; on rhs of OAS2DOTTYPE)
OEQ // Left == Right
ONE // Left != Right
OLT // Left < Right
OLE // Left <= Right
OGE // Left >= Right
OGT // Left > Right
OIND // *Left
OINDEX // Left[Right] (index of array or slice)
OINDEXMAP // Left[Right] (index of map)
cmd/compile: stop adding implicit OKEY nodes Keys are uncommon in array and slice literals, and normalizing OARRAYLIT and OSLICELIT nodes to always use OKEY ends up not reducing complexity much. Instead, only create OKEY nodes to represent explicit keys, and recalculate implicit keys when/where necessary. Fixes #15350. name old time/op new time/op delta Template 299ms ± 9% 299ms ±12% ~ (p=0.694 n=28+30) Unicode 165ms ± 7% 162ms ± 9% ~ (p=0.084 n=27+27) GoTypes 950ms ± 9% 963ms ± 5% ~ (p=0.301 n=30+29) Compiler 4.23s ± 7% 4.17s ± 7% ~ (p=0.057 n=29+27) name old user-ns/op new user-ns/op delta Template 389M ±15% 400M ±12% ~ (p=0.202 n=30+29) Unicode 246M ±21% 232M ±22% -5.76% (p=0.006 n=28+29) GoTypes 1.34G ± 8% 1.34G ± 7% ~ (p=0.775 n=28+30) Compiler 5.91G ± 6% 5.87G ± 7% ~ (p=0.298 n=28+29) name old alloc/op new alloc/op delta Template 41.2MB ± 0% 41.2MB ± 0% ~ (p=0.085 n=30+30) Unicode 34.0MB ± 0% 31.5MB ± 0% -7.28% (p=0.000 n=30+29) GoTypes 121MB ± 0% 121MB ± 0% ~ (p=0.657 n=30+30) Compiler 511MB ± 0% 511MB ± 0% -0.01% (p=0.001 n=29+29) name old allocs/op new allocs/op delta Template 390k ± 0% 390k ± 0% ~ (p=0.225 n=30+29) Unicode 318k ± 0% 293k ± 0% -8.03% (p=0.000 n=30+29) GoTypes 1.16M ± 0% 1.16M ± 0% ~ (p=0.745 n=30+30) Compiler 4.35M ± 0% 4.35M ± 0% ~ (p=0.105 n=30+30) Change-Id: I6310739a0bfdb54f1ab8a460b2c03615ad1ff5bc Reviewed-on: https://go-review.googlesource.com/32221 Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com> Run-TryBot: Matthew Dempsky <mdempsky@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org>
2016-10-27 02:02:30 -07:00
OKEY // Left:Right (key:value in struct/array/map literal)
OSTRUCTKEY // Sym:Left (key:value in struct literal, after type checking)
OLEN // len(Left)
OMAKE // make(List) (before type checking converts to one of the following)
OMAKECHAN // make(Type, Left) (type is chan)
OMAKEMAP // make(Type, Left) (type is map)
OMAKESLICE // make(Type, Left, Right) (type is slice)
OMUL // Left * Right
ODIV // Left / Right
OMOD // Left % Right
OLSH // Left << Right
ORSH // Left >> Right
OAND // Left & Right
OANDNOT // Left &^ Right
ONEW // new(Left)
ONOT // !Left
OCOM // ^Left
OPLUS // +Left
OMINUS // -Left
OOROR // Left || Right
OPANIC // panic(Left)
OPRINT // print(List)
OPRINTN // println(List)
OPAREN // (Left)
OSEND // Left <- Right
OSLICE // Left[List[0] : List[1]] (Left is untypechecked or slice)
OSLICEARR // Left[List[0] : List[1]] (Left is array)
OSLICESTR // Left[List[0] : List[1]] (Left is string)
OSLICE3 // Left[List[0] : List[1] : List[2]] (Left is untypedchecked or slice)
OSLICE3ARR // Left[List[0] : List[1] : List[2]] (Left is array)
ORECOVER // recover()
ORECV // <-Left
ORUNESTR // Type(Left) (Type is string, Left is rune)
OSELRECV // Left = <-Right.Left: (appears as .Left of OCASE; Right.Op == ORECV)
OSELRECV2 // List = <-Right.Left: (apperas as .Left of OCASE; count(List) == 2, Right.Op == ORECV)
OIOTA // iota
OREAL // real(Left)
OIMAG // imag(Left)
OCOMPLEX // complex(Left, Right)
OALIGNOF // unsafe.Alignof(Left)
OOFFSETOF // unsafe.Offsetof(Left)
OSIZEOF // unsafe.Sizeof(Left)
// statements
OBLOCK // { List } (block of code)
OBREAK // break
OCASE // case Left or List[0]..List[1]: Nbody (select case after processing; Left==nil and List==nil means default)
OXCASE // case List: Nbody (select case before processing; List==nil means default)
OCONTINUE // continue
ODEFER // defer Left (Left must be call)
OEMPTY // no-op (empty statement)
OFALL // fallthrough (after processing)
OXFALL // fallthrough (before processing)
OFOR // for Ninit; Left; Right { Nbody }
OGOTO // goto Left
OIF // if Ninit; Left { Nbody } else { Rlist }
OLABEL // Left:
OPROC // go Left (Left must be call)
ORANGE // for List = range Right { Nbody }
ORETURN // return List
OSELECT // select { List } (List is list of OXCASE or OCASE)
OSWITCH // switch Ninit; Left { List } (List is a list of OXCASE or OCASE)
OTYPESW // List = Left.(type) (appears as .Left of OSWITCH)
// types
OTCHAN // chan int
OTMAP // map[string]int
OTSTRUCT // struct{}
OTINTER // interface{}
OTFUNC // func()
OTARRAY // []int, [8]int, [N]int or [...]int
// misc
ODDD // func f(args ...int) or f(l...) or var a = [...]int{0, 1, 2}.
ODDDARG // func f(args ...int), introduced by escape analysis.
OINLCALL // intermediary representation of an inlined call.
OEFACE // itable and data words of an empty-interface value.
OITAB // itable word of an interface value.
OIDATA // data word of an interface value in Left
OSPTR // base pointer of a slice or string.
OCLOSUREVAR // variable reference at beginning of closure function
OCFUNC // reference to c function pointer (not go func value)
OCHECKNIL // emit code to ensure pointer/interface not nil
OVARKILL // variable is dead
cmd/compile: recognize Syscall-like functions for liveness analysis Consider this code: func f(*int) func g() { p := new(int) f(p) } where f is an assembly function. In general liveness analysis assumes that during the call to f, p is dead in this frame. If f has retained p, p will be found alive in f's frame and keep the new(int) from being garbage collected. This is all correct and works. We use the Go func declaration for f to give the assembly function liveness information (the arguments are assumed live for the entire call). Now consider this code: func h1() { p := new(int) syscall.Syscall(1, 2, 3, uintptr(unsafe.Pointer(p))) } Here syscall.Syscall is taking the place of f, but because its arguments are uintptr, the liveness analysis and the garbage collector ignore them. Since p is no longer live in h once the call starts, if the garbage collector scans the stack while the system call is blocked, it will find no reference to the new(int) and reclaim it. If the kernel is going to write to *p once the call finishes, reclaiming the memory is a mistake. We can't change the arguments or the liveness information for syscall.Syscall itself, both for compatibility and because sometimes the arguments really are integers, and the garbage collector will get quite upset if it finds an integer where it expects a pointer. The problem is that these arguments are fundamentally untyped. The solution we have taken in the syscall package's wrappers in past releases is to insert a call to a dummy function named "use", to make it look like the argument is live during the call to syscall.Syscall: func h2() { p := new(int) syscall.Syscall(1, 2, 3, uintptr(unsafe.Pointer(p))) use(unsafe.Pointer(p)) } Keeping p alive during the call means that if the garbage collector scans the stack during the system call now, it will find the reference to p. Unfortunately, this approach is not available to users outside syscall, because 'use' is unexported, and people also have to realize they need to use it and do so. There is much existing code using syscall.Syscall without a 'use'-like function. That code will fail very occasionally in mysterious ways (see #13372). This CL fixes all that existing code by making the compiler do the right thing automatically, without any code modifications. That is, it takes h1 above, which is incorrect code today, and makes it correct code. Specifically, if the compiler sees a foreign func definition (one without a body) that has uintptr arguments, it marks those arguments as "unsafe uintptrs". If it later sees the function being called with uintptr(unsafe.Pointer(x)) as an argument, it arranges to mark x as having escaped, and it makes sure to hold x in a live temporary variable until the call returns, so that the garbage collector cannot reclaim whatever heap memory x points to. For now I am leaving the explicit calls to use in package syscall, but they can be removed early in a future cycle (likely Go 1.7). The rule has no effect on escape analysis, only on liveness analysis. Fixes #13372. Change-Id: I2addb83f70d08db08c64d394f9d06ff0a063c500 Reviewed-on: https://go-review.googlesource.com/18584 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2016-01-13 00:46:28 -05:00
OVARLIVE // variable is alive
OINDREGSP // offset plus indirect of REGSP, such as 8(SP).
// arch-specific opcodes
OCMP // compare: ACMP.
ODEC // decrement: ADEC.
OINC // increment: AINC.
OEXTEND // extend: ACWD/ACDQ/ACQO.
OHMUL // high mul: AMUL/AIMUL for unsigned/signed (OMUL uses AIMUL for both).
ORROTC // right rotate-carry: ARCR.
ORETJMP // return to other function
cmd/internal/gc: move cgen, regalloc, et al to portable code This CL moves the bulk of the code that has been copy-and-pasted since the initial 386 port back into a shared place, cutting 5 copies to 1. The motivation here is not cleanup per se but instead to reduce the cost of introducing changes in shared concepts like regalloc or general expression evaluation. For example, a change after this one will implement x.(*T) without a call into the runtime. This CL makes that followup work 5x easier. The single copy still has more special cases for architecture details than I'd like, but having them called out explicitly like this at least opens the door to generalizing the conditions and smoothing out the distinctions in the future. This is a LARGE CL. I started by trying to pull in one function at a time in a sequence of CLs and it became clear that everything was so interrelated that it had to be moved as a whole. Apologies for the size. It is not clear how many more releases this code will matter for; eventually it will be replaced by Keith's SSA work. But as noted above, the deduplication was necessary to reduce the cost of working on the current code while we have it. Passes tests on amd64, 386, arm, and ppc64le. Can build arm64 binaries but not tested there. Being able to build binaries means it is probably very close. Change-Id: I735977f04c0614f80215fb12966dfe9bbd1f5861 Reviewed-on: https://go-review.googlesource.com/7853 Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2015-03-18 17:26:36 -04:00
OPS // compare parity set (for x86 NaN check)
cmd/internal/gc, cmd/6g: generate boolean values without jumps Use SETcc instructions instead of Jcc to generate boolean values. This generates shorter, jump-free code, which may in turn enable other peephole optimizations. For example, given func f(i, j int) bool { return i == j } Before "".f t=1 size=32 value=0 args=0x18 locals=0x0 0x0000 00000 (x.go:3) TEXT "".f(SB), $0-24 0x0000 00000 (x.go:3) FUNCDATA $0, gclocals·b4c25e9b09fd0cf9bb429dcefe91c353(SB) 0x0000 00000 (x.go:3) FUNCDATA $1, gclocals·33cdeccccebe80329f1fdbee7f5874cb(SB) 0x0000 00000 (x.go:4) MOVQ "".i+8(FP), BX 0x0005 00005 (x.go:4) MOVQ "".j+16(FP), BP 0x000a 00010 (x.go:4) CMPQ BX, BP 0x000d 00013 (x.go:4) JEQ 21 0x000f 00015 (x.go:4) MOVB $0, "".~r2+24(FP) 0x0014 00020 (x.go:4) RET 0x0015 00021 (x.go:4) MOVB $1, "".~r2+24(FP) 0x001a 00026 (x.go:4) JMP 20 After "".f t=1 size=32 value=0 args=0x18 locals=0x0 0x0000 00000 (x.go:3) TEXT "".f(SB), $0-24 0x0000 00000 (x.go:3) FUNCDATA $0, gclocals·b4c25e9b09fd0cf9bb429dcefe91c353(SB) 0x0000 00000 (x.go:3) FUNCDATA $1, gclocals·33cdeccccebe80329f1fdbee7f5874cb(SB) 0x0000 00000 (x.go:4) MOVQ "".i+8(FP), BX 0x0005 00005 (x.go:4) MOVQ "".j+16(FP), BP 0x000a 00010 (x.go:4) CMPQ BX, BP 0x000d 00013 (x.go:4) SETEQ "".~r2+24(FP) 0x0012 00018 (x.go:4) RET regexp benchmarks, best of 12 runs: benchmark old ns/op new ns/op delta BenchmarkNotOnePassShortB 782 733 -6.27% BenchmarkLiteral 180 171 -5.00% BenchmarkNotLiteral 2855 2721 -4.69% BenchmarkMatchHard_32 2672 2557 -4.30% BenchmarkMatchHard_1K 80182 76732 -4.30% BenchmarkMatchEasy1_32M 76440180 73304748 -4.10% BenchmarkMatchEasy1_32K 68798 66350 -3.56% BenchmarkAnchoredLongMatch 482 465 -3.53% BenchmarkMatchEasy1_1M 2373042 2292692 -3.39% BenchmarkReplaceAll 2776 2690 -3.10% BenchmarkNotOnePassShortA 1397 1360 -2.65% BenchmarkMatchClass_InRange 3842 3742 -2.60% BenchmarkMatchEasy0_32 125 122 -2.40% BenchmarkMatchEasy0_32K 11414 11164 -2.19% BenchmarkMatchEasy0_1K 668 654 -2.10% BenchmarkAnchoredShortMatch 260 255 -1.92% BenchmarkAnchoredLiteralShortNonMatch 164 161 -1.83% BenchmarkOnePassShortB 623 612 -1.77% BenchmarkOnePassShortA 801 788 -1.62% BenchmarkMatchClass 4094 4033 -1.49% BenchmarkMatchEasy0_32M 14078800 13890704 -1.34% BenchmarkMatchHard_32K 4095844 4045820 -1.22% BenchmarkMatchEasy1_1K 1663 1643 -1.20% BenchmarkMatchHard_1M 131261708 129708215 -1.18% BenchmarkMatchHard_32M 4210112412 4169292003 -0.97% BenchmarkMatchMedium_32K 2460752 2438611 -0.90% BenchmarkMatchEasy0_1M 422914 419672 -0.77% BenchmarkMatchMedium_1M 78581121 78040160 -0.69% BenchmarkMatchMedium_32M 2515287278 2498464906 -0.67% BenchmarkMatchMedium_32 1754 1746 -0.46% BenchmarkMatchMedium_1K 52105 52106 +0.00% BenchmarkAnchoredLiteralLongNonMatch 185 185 +0.00% BenchmarkMatchEasy1_32 107 107 +0.00% BenchmarkOnePassLongNotPrefix 505 505 +0.00% BenchmarkOnePassLongPrefix 147 147 +0.00% The godoc binary is ~0.12% smaller after this CL. Updates #5729. toolstash -cmp passes for all architectures other than amd64 and amd64p32. Other architectures can be done in follow-up CLs. Change-Id: I0e167e259274b722958567fc0af83a17ca002da7 Reviewed-on: https://go-review.googlesource.com/2284 Reviewed-by: Russ Cox <rsc@golang.org>
2015-04-08 09:54:15 -07:00
OPC // compare parity clear (for x86 NaN check)
cmd/internal/gc: inline runtime.getg This more closely restores what the old C runtime did. (In C, g was an 'extern register' with the same effective implementation as in this CL.) On a late 2012 MacBookPro10,2, best of 5 old vs best of 5 new: benchmark old ns/op new ns/op delta BenchmarkBinaryTree17 4981312777 4463426605 -10.40% BenchmarkFannkuch11 3046495712 3006819428 -1.30% BenchmarkFmtFprintfEmpty 89.3 79.8 -10.64% BenchmarkFmtFprintfString 284 262 -7.75% BenchmarkFmtFprintfInt 282 262 -7.09% BenchmarkFmtFprintfIntInt 480 448 -6.67% BenchmarkFmtFprintfPrefixedInt 382 358 -6.28% BenchmarkFmtFprintfFloat 529 486 -8.13% BenchmarkFmtManyArgs 1849 1773 -4.11% BenchmarkGobDecode 12835963 11794385 -8.11% BenchmarkGobEncode 10527170 10288422 -2.27% BenchmarkGzip 436109569 438422516 +0.53% BenchmarkGunzip 110121663 109843648 -0.25% BenchmarkHTTPClientServer 81930 85446 +4.29% BenchmarkJSONEncode 24638574 24280603 -1.45% BenchmarkJSONDecode 93022423 85753546 -7.81% BenchmarkMandelbrot200 4703899 4735407 +0.67% BenchmarkGoParse 5319853 5086843 -4.38% BenchmarkRegexpMatchEasy0_32 151 151 +0.00% BenchmarkRegexpMatchEasy0_1K 452 453 +0.22% BenchmarkRegexpMatchEasy1_32 131 132 +0.76% BenchmarkRegexpMatchEasy1_1K 761 722 -5.12% BenchmarkRegexpMatchMedium_32 228 224 -1.75% BenchmarkRegexpMatchMedium_1K 63751 64296 +0.85% BenchmarkRegexpMatchHard_32 3188 3238 +1.57% BenchmarkRegexpMatchHard_1K 95396 96756 +1.43% BenchmarkRevcomp 661587262 687107364 +3.86% BenchmarkTemplate 108312598 104008540 -3.97% BenchmarkTimeParse 453 459 +1.32% BenchmarkTimeFormat 475 441 -7.16% The garbage benchmark from the benchmarks subrepo gets 2.6% faster as well. Change-Id: I320aeda332db81012688b26ffab23f6581c59cfa Reviewed-on: https://go-review.googlesource.com/8460 Reviewed-by: Rick Hudson <rlh@golang.org> Run-TryBot: Rick Hudson <rlh@golang.org> Reviewed-by: Austin Clements <austin@google.com>
2015-04-03 12:23:28 -04:00
OGETG // runtime.getg() (read g pointer)
OEND
)
// Nodes is a pointer to a slice of *Node.
// For fields that are not used in most nodes, this is used instead of
// a slice to save space.
type Nodes struct{ slice *[]*Node }
// Slice returns the entries in Nodes as a slice.
// Changes to the slice entries (as in s[i] = n) will be reflected in
// the Nodes.
func (n Nodes) Slice() []*Node {
if n.slice == nil {
return nil
}
return *n.slice
}
// Len returns the number of entries in Nodes.
func (n Nodes) Len() int {
if n.slice == nil {
return 0
}
return len(*n.slice)
}
// Index returns the i'th element of Nodes.
// It panics if n does not have at least i+1 elements.
func (n Nodes) Index(i int) *Node {
return (*n.slice)[i]
}
// First returns the first element of Nodes (same as n.Index(0)).
// It panics if n has no elements.
func (n Nodes) First() *Node {
return (*n.slice)[0]
}
// Second returns the second element of Nodes (same as n.Index(1)).
// It panics if n has fewer than two elements.
func (n Nodes) Second() *Node {
return (*n.slice)[1]
}
// Set sets n to a slice.
// This takes ownership of the slice.
func (n *Nodes) Set(s []*Node) {
if len(s) == 0 {
n.slice = nil
} else {
// Copy s and take address of t rather than s to avoid
// allocation in the case where len(s) == 0 (which is
// over 3x more common, dynamically, for make.bash).
t := s
n.slice = &t
}
}
// Set1 sets n to a slice containing a single node.
func (n *Nodes) Set1(node *Node) {
n.slice = &[]*Node{node}
}
// Set2 sets n to a slice containing two nodes.
func (n *Nodes) Set2(n1, n2 *Node) {
n.slice = &[]*Node{n1, n2}
}
// MoveNodes sets n to the contents of n2, then clears n2.
func (n *Nodes) MoveNodes(n2 *Nodes) {
n.slice = n2.slice
n2.slice = nil
}
// SetIndex sets the i'th element of Nodes to node.
// It panics if n does not have at least i+1 elements.
func (n Nodes) SetIndex(i int, node *Node) {
(*n.slice)[i] = node
}
// Addr returns the address of the i'th element of Nodes.
// It panics if n does not have at least i+1 elements.
func (n Nodes) Addr(i int) **Node {
return &(*n.slice)[i]
}
// Append appends entries to Nodes.
// If a slice is passed in, this will take ownership of it.
func (n *Nodes) Append(a ...*Node) {
if len(a) == 0 {
return
}
if n.slice == nil {
n.slice = &a
} else {
*n.slice = append(*n.slice, a...)
}
}
// Prepend prepends entries to Nodes.
// If a slice is passed in, this will take ownership of it.
func (n *Nodes) Prepend(a ...*Node) {
if len(a) == 0 {
return
}
if n.slice == nil {
n.slice = &a
} else {
*n.slice = append(a, *n.slice...)
}
}
// AppendNodes appends the contents of *n2 to n, then clears n2.
func (n *Nodes) AppendNodes(n2 *Nodes) {
switch {
case n2.slice == nil:
case n.slice == nil:
n.slice = n2.slice
default:
*n.slice = append(*n.slice, *n2.slice...)
}
n2.slice = nil
}