mirror of
https://github.com/nim-lang/Nim.git
synced 2026-09-02 03:43:41 +00:00
`for i in k..<n.len: ... n[i] ...` is the dominant shape for walking a `PNode`'s children in the code generator: 41 such loops across the cgen files, and 777 indexed node accesses in total. It reads worse than iterating, it bounds-checks every subscript, and it is quadratic the moment the backend reads children off a NIF `Cursor` rather than a materialised tree (a child is `firstSon` plus one `skip` per preceding sibling, and `skip` steps over a whole subtree). 31 of the 41 are converted: * 20 to `sons`/`sonsFrom` — the index only ever subscripted `n`. * 9 to `isons`, which now takes a `start` index (defaulting to 0, so its eight existing call sites are unchanged). These genuinely need `i`: a parallel index into the routine's `PType` (`typ.n[i]`, `typ[i]`), a `needTmp[i-1]` lookup, an `i == field.position` test, `$i` in a generated struct name, or the index passed straight to `genOtherArg`. * 2 to `sonsFrom` with a variable start (`firstParam`, `offset`). `sonsFrom` is new, next to `sons`/`isons` in astdef. The remaining 10 are deliberate. Eight are not `PNode` at all — `varargs[Snippet]`, `seq[PSym]`, `string`, and `PType`, where `sons` is a `proc ...: var TTypeSeq` rather than an iterator, so a blind rewrite would compile into something quite different. Two iterate `0..<it.len-1`, excluding the last child, which no iterator expresses cleanly. Pure refactor, and verified as one: all 219 generated `.c` files of a 219-module program and the linked binary are byte-identical to the parent commit. That is the bar that matters here, because the index arithmetic (`i-1`, `i == position`, `$i`) is the easy thing to get wrong. It also caught a real slip on the way: `genFieldCheck` reassigns its loop variable, which a `for` binding cannot do. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
136 lines
5.3 KiB
Nim
136 lines
5.3 KiB
Nim
#
|
|
#
|
|
# The Nim Compiler
|
|
# (c) Copyright 2020 Andreas Rumpf
|
|
#
|
|
# See the file "copying.txt", included in this
|
|
# distribution, for details about the copyright.
|
|
#
|
|
|
|
# included from cgen.nim
|
|
|
|
## Code specialization instead of the old, incredibly slow 'genericReset'
|
|
## implementation.
|
|
|
|
proc specializeResetT(p: BProc, accessor: Rope, typ: PType)
|
|
|
|
proc specializeResetN(p: BProc, accessor: Rope, n: PNode;
|
|
typ: PType) =
|
|
if n == nil: return
|
|
case n.kind
|
|
of nkRecList:
|
|
for it in sons(n):
|
|
specializeResetN(p, accessor, it, typ)
|
|
of nkRecCase:
|
|
if (n[0].kind != nkSym): internalError(p.config, n.info, "specializeResetN")
|
|
let disc = n[0].sym
|
|
if disc.loc.snippet == "": fillObjectFields(p.module, typ)
|
|
if disc.loc.t == nil:
|
|
internalError(p.config, n.info, "specializeResetN()")
|
|
let discField = dotField(accessor, disc.loc.snippet)
|
|
p.s(cpsStmts).addSwitchStmt(discField):
|
|
for branch in sonsFrom(n, 1):
|
|
assert branch.kind in {nkOfBranch, nkElse}
|
|
var caseBuilder: SwitchCaseBuilder
|
|
p.s(cpsStmts).addSwitchCase(caseBuilder):
|
|
if branch.kind == nkOfBranch:
|
|
genCaseRange(p, branch, caseBuilder)
|
|
else:
|
|
p.s(cpsStmts).addCaseElse(caseBuilder)
|
|
do:
|
|
specializeResetN(p, accessor, lastSon(branch), typ)
|
|
p.s(cpsStmts).addBreak()
|
|
specializeResetT(p, discField, disc.loc.t)
|
|
of nkSym:
|
|
let field = n.sym
|
|
if field.typ.kind == tyVoid: return
|
|
if field.loc.snippet == "": fillObjectFields(p.module, typ)
|
|
if field.loc.t == nil:
|
|
internalError(p.config, n.info, "specializeResetN()")
|
|
specializeResetT(p, dotField(accessor, field.loc.snippet), field.loc.t)
|
|
else: internalError(p.config, n.info, "specializeResetN()")
|
|
|
|
proc specializeResetT(p: BProc, accessor: Rope, typ: PType) =
|
|
if typ == nil: return
|
|
|
|
case typ.kind
|
|
of tyGenericInst, tyGenericBody, tyTypeDesc, tyAlias, tyDistinct, tyInferred,
|
|
tySink, tyOwned:
|
|
specializeResetT(p, accessor, skipModifier(typ))
|
|
of tyArray:
|
|
let arraySize = lengthOrd(p.config, typ.indexType)
|
|
var i: TLoc = getTemp(p, getSysType(p.module.g.graph, unknownLineInfo, tyInt))
|
|
p.s(cpsStmts).addForRangeExclusive(i.snippet, cIntValue(0), cIntValue(arraySize)):
|
|
specializeResetT(p, subscript(accessor, i.snippet), typ.elementType)
|
|
of tyObject:
|
|
var x = typ.baseClass
|
|
if x != nil: x = x.skipTypes(skipPtrs)
|
|
specializeResetT(p, accessor.parentObj(p.module), x)
|
|
if typ.n != nil:
|
|
if typ.sym != nil and sfImportc in typ.sym.flags:
|
|
# imported C struct, nimZeroMem
|
|
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimZeroMem"),
|
|
cCast(ptrType(CPointer), cAddr(accessor)),
|
|
cSizeof(getTypeDesc(p.module, typ)))
|
|
else:
|
|
specializeResetN(p, accessor, typ.n, typ)
|
|
if isCaseObj(typ.n):
|
|
# The active branch was released above. Clear the complete object so
|
|
# stale bytes from overlapping branches cannot be traced by the GC.
|
|
# type
|
|
# Foo = object
|
|
# case kind: bool
|
|
# of true:
|
|
# a: ref Bar # 8 bytes (pointer)
|
|
# of false:
|
|
# b: int # 4 bytes
|
|
# specializeResetT for b emits accessor.b = 0 — writes 4 bytes
|
|
# But the union is 8 bytes wide (sized by the largest branch)
|
|
# The remaining 4 bytes where a used to live are untouched
|
|
# Those stale bytes could contain a heap pointer the GC traces → crash
|
|
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimZeroMem"),
|
|
cCast(CPointer, cAddr(accessor)),
|
|
cSizeof(getTypeDesc(p.module, typ)))
|
|
of tyTuple:
|
|
let typ = getUniqueType(typ)
|
|
for i, a in typ.ikids:
|
|
specializeResetT(p, dotField(accessor, "Field" & $i), a)
|
|
|
|
of tyString, tyRef, tySequence:
|
|
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "unsureAsgnRef"),
|
|
cCast(ptrType(CPointer), cAddr(accessor)),
|
|
NimNil)
|
|
|
|
of tyProc:
|
|
if typ.callConv == ccClosure:
|
|
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "unsureAsgnRef"),
|
|
cCast(ptrType(CPointer), cAddr(dotField(accessor, "ClE_0"))),
|
|
NimNil)
|
|
p.s(cpsStmts).addFieldAssignment(accessor, "ClP_0", NimNil)
|
|
else:
|
|
p.s(cpsStmts).addAssignment(accessor, NimNil)
|
|
of tyChar, tyBool, tyEnum, tyRange, tyInt..tyUInt64:
|
|
p.s(cpsStmts).addAssignment(accessor, cIntValue(0))
|
|
of tyCstring, tyPointer, tyPtr, tyVar, tyLent:
|
|
p.s(cpsStmts).addAssignment(accessor, NimNil)
|
|
of tySet:
|
|
case mapSetType(p.config, typ)
|
|
of ctArray:
|
|
let t = getTypeDesc(p.module, typ)
|
|
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimZeroMem"),
|
|
accessor,
|
|
cSizeof(t))
|
|
of ctInt8, ctInt16, ctInt32, ctInt64:
|
|
p.s(cpsStmts).addAssignment(accessor, cIntValue(0))
|
|
else:
|
|
raiseAssert "unexpected set type kind"
|
|
of tyNone, tyEmpty, tyNil, tyUntyped, tyTyped, tyGenericInvocation,
|
|
tyGenericParam, tyOrdinal, tyOpenArray, tyForward, tyVarargs,
|
|
tyUncheckedArray, tyError, tyBuiltInTypeClass, tyUserTypeClass,
|
|
tyUserTypeClassInst, tyCompositeTypeClass, tyAnd, tyOr, tyNot,
|
|
tyAnything, tyStatic, tyFromExpr, tyConcept, tyVoid, tyIterable:
|
|
discard
|
|
|
|
proc specializeReset(p: BProc, a: TLoc) =
|
|
specializeResetT(p, rdLoc(a), a.t)
|