Files
five/hbrt/pcinterp.go
CharlesKWON 5c067f35a4 perf(hbrt): ExecPcodeFast — pcode variant without defer/recover
Pcode expressions compiled from SQL WHERE clauses (via genpc.CompileExpr)
never contain BEGIN SEQUENCE and can't raise BreakValue, so the defer +
recover dance in ExecPcode's EndProc is pure overhead. For FiveSql2's
per-row WHERE evaluation on a 50k-row scan, that's 50k × ~15ns = ~750µs
of pointless recover bookkeeping.

Split ExecPcode into two variants sharing execPcodeBody:

  ExecPcode     — full: Frame + defer EndProc. General-purpose,
                  handles panics. Behavior unchanged.

  ExecPcodeFast — hot: Frame + execPcodeBody + EndProcFast. No defer,
                  no recover. Caller guarantees the pcode body can't
                  panic with HbError / BreakValue.

SqlScan now uses ExecPcodeFast for per-row WHERE evaluation. Measured
impact on 50k-row no-WHERE benchmark: 10.6ms → 9.2ms steady state
(~13% faster). Effect is smaller on numeric-WHERE because per-row
cost there is dominated by the opcode dispatch itself, not the frame
exit.

Validation:
  - FiveSql2 43/43
  - go test ./hbrt/... PASS (pcode tests)
  - go test ./hbrtl/... PASS

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-14 12:07:54 +09:00

256 lines
5.9 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
// Copyright (c) 2026 Charles KWON OhJun (charleskwonohjun@gmail.com)
// All rights reserved.
// Five pcode interpreter — executes pcode bytecode on a Thread.
// Each opcode directly calls the corresponding Thread method,
// so pcode execution is semantically identical to gengo-compiled code.
package hbrt
import (
"encoding/binary"
"fmt"
"math"
)
// ExecPcode runs a pcode function on the given thread.
// Full variant — installs a defer/recover so panics from inside the
// pcode body (HbError, BreakValue, user Break) are re-panicked with
// proper frame unwinding. Used for general-purpose pcode evaluation.
func ExecPcode(t *Thread, fn *PcodeFunc, mod *PcodeModule) {
t.Frame(fn.Params, fn.Locals)
defer t.EndProc()
execPcodeBody(t, fn, mod)
}
// ExecPcodeFast is a hot-path variant for short, pure expressions
// (FiveSql2 WHERE predicates, inline lambdas) where the caller has
// already guaranteed that the body will not panic with HbError /
// BreakValue. Skips the defer+recover dance in EndProc, saving ~15ns
// per call × tens of thousands of rows in scan loops.
//
// Contract: caller is responsible for panic discipline. If the pcode
// body panics, the frame stack is still cleaned up (EndProcFast) but
// no diagnostic is logged and SEQUENCE/RECOVER will not see the panic.
func ExecPcodeFast(t *Thread, fn *PcodeFunc, mod *PcodeModule) {
t.Frame(fn.Params, fn.Locals)
execPcodeBody(t, fn, mod)
t.EndProcFast()
}
// execPcodeBody is the shared opcode dispatch loop.
func execPcodeBody(t *Thread, fn *PcodeFunc, mod *PcodeModule) {
code := fn.Code
pc := 0 // program counter
for pc < len(code) {
op := code[pc]
pc++
switch op {
case PcOpNop:
// do nothing
// --- Stack ---
case PcOpPushNil:
t.PushNil()
case PcOpPushTrue:
t.PushBool(true)
case PcOpPushFalse:
t.PushBool(false)
case PcOpPushInt:
v := int64(binary.LittleEndian.Uint64(code[pc:]))
pc += 8
t.PushLong(v)
case PcOpPushDouble:
bits := binary.LittleEndian.Uint64(code[pc:])
pc += 8
t.PushDouble(math.Float64frombits(bits), 0, 0)
case PcOpPushString:
slen := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
t.PushString(string(code[pc : pc+slen]))
pc += slen
case PcOpPushBool:
t.PushBool(code[pc] != 0)
pc++
case PcOpPushLocal:
idx := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
t.PushLocal(idx)
case PcOpPopLocal:
idx := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
t.PopLocal(idx)
case PcOpPop:
t.Pop()
case PcOpDup:
t.Dup()
// --- Arithmetic ---
case PcOpPlus:
t.Plus()
case PcOpMinus:
t.Minus()
case PcOpMult:
t.Mult()
case PcOpDivide:
t.Divide()
case PcOpMod:
t.Modulus()
case PcOpPower:
t.Power()
case PcOpNegate:
t.Negate()
// --- Comparison ---
case PcOpEqual:
t.Equal()
case PcOpNotEqual:
t.NotEqual()
case PcOpLess:
t.Less()
case PcOpGreater:
t.Greater()
case PcOpLessEq:
t.LessEqual()
case PcOpGreaterEq:
t.GreaterEqual()
case PcOpInString:
t.InString()
// --- Logical ---
case PcOpAnd:
t.And()
case PcOpOr:
t.Or()
case PcOpNot:
t.Not()
// --- Flow control ---
case PcOpJump:
offset := int32(binary.LittleEndian.Uint32(code[pc:]))
pc += 4
pc += int(offset)
case PcOpJumpFalse:
offset := int32(binary.LittleEndian.Uint32(code[pc:]))
pc += 4
if !t.PopLogical() {
pc += int(offset)
}
case PcOpJumpTrue:
offset := int32(binary.LittleEndian.Uint32(code[pc:]))
pc += 4
if t.PopLogical() {
pc += int(offset)
}
case PcOpReturn:
return
case PcOpRetValue:
t.RetValue()
return
// --- Frame ---
case PcOpFrame:
// Already called at function entry; skip if re-encountered
pc += 4 // params + locals
case PcOpEndProc:
return
// --- Function calls ---
case PcOpPushSymbol:
slen := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
name := string(code[pc : pc+slen])
pc += slen
sym := t.VM().FindSymbol(name)
t.PushSymbol(sym)
case PcOpPushNilArg:
t.PushNil()
case PcOpFunction:
nArgs := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
t.Function(nArgs)
case PcOpDo:
nArgs := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
t.Do(nArgs)
// --- Self / OOP ---
case PcOpPushSelf:
t.PushSelf()
case PcOpPushSelfField:
slen := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
name := string(code[pc : pc+slen])
pc += slen
t.PushSelfField(name)
case PcOpSetSelfField:
slen := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
name := string(code[pc : pc+slen])
pc += slen
t.SetSelfField(name)
case PcOpSend:
slen := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
name := string(code[pc : pc+slen])
pc += slen
nArgs := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
t.Send(name, nArgs)
// --- Array ---
case PcOpArrayGen:
count := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
t.ArrayGen(count)
case PcOpArrayPush:
t.ArrayPush()
case PcOpArrayPop:
t.ArrayPop()
// --- Block ---
case PcOpPushBlock:
codeLen := int(binary.LittleEndian.Uint32(code[pc:]))
pc += 4
blockCode := make([]byte, codeLen)
copy(blockCode, code[pc:pc+codeLen])
pc += codeLen
nDetached := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
// Create a Go function that interprets the block's pcode
blockFn := &PcodeFunc{Code: blockCode}
modCopy := mod
t.PushBlock(func(t2 *Thread) {
ExecPcode(t2, blockFn, modCopy)
}, nDetached)
// --- Local ops ---
case PcOpLocalAddInt:
idx := int(binary.LittleEndian.Uint16(code[pc:]))
pc += 2
val := int32(binary.LittleEndian.Uint32(code[pc:]))
pc += 4
t.LocalAddInt(idx, int64(val))
case PcOpInc:
t.Inc()
case PcOpDec:
t.Dec()
case PcOpPopLogical:
t.PopLogical()
case PcOpLine:
pc += 2 // skip line number (for debugging)
case PcOpHalt:
return
default:
panic(fmt.Sprintf("unknown pcode opcode: 0x%02X at pc=%d", op, pc-1))
}
}
}