diff options
Diffstat (limited to 'docs/history/13-gap-fixes/code/src/compiler.zc')
| -rw-r--r-- | docs/history/13-gap-fixes/code/src/compiler.zc | 2638 |
1 files changed, 2638 insertions, 0 deletions
diff --git a/docs/history/13-gap-fixes/code/src/compiler.zc b/docs/history/13-gap-fixes/code/src/compiler.zc new file mode 100644 index 0000000..c6fa3cc --- /dev/null +++ b/docs/history/13-gap-fixes/code/src/compiler.zc @@ -0,0 +1,2638 @@ +// ============================================================================ +// compiler.zc: the resident zc compiler. it lives in the genome next to the +// runtime, compiles source straight to vm words in memory, and is visible to +// the programs it compiles: every function and global here can be called or +// patched by name, and parse words (immediate fns) run inside it. +// ============================================================================ + +// ---- compiler data region (fixed addresses, 0xA0100 .. 0xBE100) ---- +const IHN = 8192; +const R_IH = 0xA0100; // intern hash table +const R_POOL = 0xA2100; // interned strings: [symbol head] chars 0 +const R_POOLEND = 0xAC000; +const NS = 4096; // symbol table, one array per field +const R_SYM = 0xAC100; +const sname = 0xAC100, skind = 0xAD100, sval = 0xAE100, ssig = 0xAF100; +const sflag = 0xB0100, sfix = 0xB1100, snext = 0xB2100; +const NL = 512; // locals, one array per field +const lname = 0xB3100, lslot = 0xB3300, lhome = 0xB3500, lstate = 0xB3700; +const lowner = 0xB3900, lused = 0xB3B00, larr = 0xB3D00, lown = 0xB3F00; // lown: has owned an object +const NF = 8192; // patch nodes: a site and a link +const fsite = 0xB4100, fnext = 0xB6100; +const R_SNAP = 0xB8100; // analyzer snapshots +const R_QBUF = 0xBA100; // token buffer for quote and grab +const R_SRC = 0xBC100; // input source stack, 32 words per frame +const R_SEQ = 0xBC500; // operator code sequences +const bname = 0xBCD00, bprec = 0xBCD40, bseq = 0xBCD80, blen = 0xBCDC0; // binary operators +const uname = 0xBCE00, useq = 0xBCE20, ulen = 0xBCE40; // unary operators +const aname = 0xBCE60, abin = 0xBCEA0; // compound assignment +const kops = 0xBCF00; // every operator spelling the lexer knows +const R_CTX = 0xBD100; // saved compiler contexts for nested compiles +const R_LOOP = 0xBD900; // loop stack, 8 words per loop or switch +const R_ASM = 0xBDD00; // asm block labels + +// ---- token kinds, symbol kinds and flags ---- +const T_EOF = 0, T_NUM = 1, T_ID = 2, T_STR = 3, T_OP = 4; +const K_FN = 1, K_GVAR = 2, K_GARR = 3, K_CONST = 4, K_KW = 5, K_ENH = 6, K_TYPE = 7; +const F_DEF = 1, F_SIG = 2, F_UNTYPED = 4, F_STMT = 8, F_EXPR = 16, F_TOP = 32, F_FALLBACK = 64, F_GEN = 128, F_PRE = 256; +// variable states for the lifetime analyzer +const S_PLAIN = 0, S_M = 1, S_U = 2, S_D = 3, S_DEAD = 4, S_CHILD = 5, S_BORROW = 6; +// expression classes: plain word, a new object, or a bare variable (C_VAR + index) +const C_PLAIN = 0, C_FRESH = 1, C_VAR = 2; +// lvalue kinds +const LV_NONE = 0, LV_LOCAL = 1, LV_GLOBAL = 2, LV_ELEM = 3, LV_OTHER = 4; +// opcodes +const O_Q = 0, O_LD = 1, O_ST = 2, O_PICK = 3, O_DROP = 4, O_SWAP = 5, O_CALL = 6, O_JMP = 7; +const O_JZ = 8, O_RET = 9, O_ADD = 13, O_SUB = 14, O_LT = 20; +const XOP0 = 22; // first enhancement opcode +const SEQREF = -2; // in an operator sequence: the next word names a symbol + +// ---- state ---- +let nenh, pooltop, nsym, ubase, vislo, nfixu, fixfree; +let nbin, nun, nasg, nkop, seqtop; +let mnem = "q@!$_~cjzr,?.+-*&|^><%"; +let check_on = 1; // set to 0 to switch the analyzer off for what follows +let tk, tv, tglue, tline, line = 1, ch, chkind, chptr, chbase, qdone, isp = -1; +let tbuf[1100]; +let pk_tk, pk_tv, pq_n, pq_tk[4], pq_tv[4], pq_gl[4], pq_ln[4]; // up to 4 tokens of lookahead +let here, hdelta, hlimit, ptop; +let nloc, lbase, fdepth, fslots, fchk, fret, live, bxdepth, epichain, fframe, infn; +let lsp, loopbase, ctxsp, snaptop; +let lv, lvk, lvi, kw_expr, xcls, fty; +// interned spellings the parser compares against +let s_lp, s_rp, s_lb, s_rb, s_lk, s_rk, s_comma, s_semi, s_eq, s_arrow, s_inc, s_dec; +let s_star, s_amp, s_colon, s_hash, s_minus, s_andand, s_oror; +let s_dot, s_fn, s_var, s_const, s_else, s_as, s_unchecked, s_immediate, s_m, s_d, s_u; +let s_q, s_case, s_default, s_fovf, s_start, s_main, s_free, s_own, s_qnew, s_enhance, s_micro, s_vm; + +// ---- types: kinds, tables and parser state (see the types section) ---- +const TY_WORD = 0, TY_INT = 1, TY_CHAR = 2, TY_BOOL = 3, TY_VOID = 4, TY_PTR = 5, TY_ARR = 6; +const TY_FN = 7, TY_STRUCT = 8, TY_VAR = 9, TY_GEN = 10; +const R_TY = 0x9C000, R_TYEND = 0xA0000, NTL = 1024; // the first NTL words list every record +const ltype = 0xBE100, ltag = 0xBE300, stype = 0xBE500; // type per local, declared tag per local, type per symbol +let tytop, ntl, tword, tint, tchar, tbool, tvoid; +let tc_on; // the optional typechecker, off unless a program says typecheck on; +let xty, xzero, dry; // type of the last expression, was it a literal 0, and dry runs (typeof) +let tpn[8], ntp; // type parameters in scope +let dty, dtag; // result of dtype() +let s_lt, s_gt, s_shr, s_le, s_ge, s_eqeq, s_ne, s_plus, s_struct, s_type, s_typecheck, s_on, s_off, s_new, s_new_t; +let numops[12], nnumops; +let gbind[8], ngbind; // explicit type arguments for the next call: f<int>(...) + +// ---- errors ---- +// the compiler reports up to 20 errors before it stops. a syntax error +// (die) skips the token it happened on and sets panic, which skips ahead to +// the next ; or } (or the next top-level item) and keeps quiet until then. +// other errors (serr) just report: the parse is still in step. +let nerr, panic; +fn errat() { puts("zc: line "); putn(tline); puts(": "); nerr++; return 0; } +fn errend() { putc(10); if (nerr >= 20) { puts("zc: too many errors, stopping\n"); halt(); } return 0; } +fn giveup() { puts("zc: "); putn(nerr); if (nerr == 1) puts(" error\n"); else puts(" errors\n"); halt(); return 0; } +fn fatal(msg) { errat(); puts(msg); putc(10); halt(); return 0; } +fn serr(msg) { if (!panic && !dry) { errat(); puts(msg); errend(); } return 0; } +fn serrn(name, msg) { if (!panic && !dry) { errat(); puts(name); puts(": "); puts(msg); errend(); } return 0; } +fn die(msg) { serr(msg); panic = 1; if (tk == T_EOF) giveup(); next(); return 0; } +fn dien(name, msg) { serrn(name, msg); panic = 1; if (tk == T_EOF) giveup(); next(); return 0; } +// analyzer errors only count in checked, reachable code +fn cerr(msg) { if (fchk && live) serr(msg); return 0; } +fn cerrn(i, msg) { if (fchk && live) serrn(lname[i], msg); return 0; } + +// ---- interning ---- +fn hashs(s) { let h = 0; while (*s) { h = h * 31 + *s; s++; } return h & (IHN - 1); } +fn intern(s) +{ + let h = hashs(s), p, n; + while (1) { + p = R_IH[h]; + if (!p) break; + if (streq(p, s)) return p; + h = (h + 1) & (IHN - 1); + } + n = strlen(s); + if (pooltop + n + 2 >= R_POOLEND) fatal("string pool full"); + pooltop[0] = 0; + p = pooltop + 1; + wcopy(p, s, n + 1); + pooltop = pooltop + n + 2; + R_IH[h] = p; + return p; +} + +// ---- symbols ---- +// each interned name keeps the index (+1) of its newest symbol in p[-1], and +// each symbol links to the one it shadows, so lookup is one step. +fn addsym(name, kind, val) +{ + let i = nsym; + if (nsym >= NS) fatal("symbol table full"); + nsym++; + sname[i] = name; skind[i] = kind; sval[i] = val; ssig[i] = 0; sflag[i] = F_DEF; sfix[i] = 0; stype[i] = 0; + snext[i] = name[-1]; + name[-1] = i + 1; + return i; +} +// while an image builds, names below vislo are hidden, except keywords +fn lookup(name) +{ + let i = name[-1] - 1; + while (i >= 0) { + if (i >= vislo || skind[i] == K_KW || skind[i] == K_TYPE) return i; + i = snext[i] - 1; + } + return -1; +} +fn popsyms(to) { while (nsym > to) { nsym--; sname[nsym][-1] = snext[nsym]; } return 0; } +fn lookups(s) { return lookup(intern(s)); } + +// patch nodes +fn fixnew(site, link) +{ + let i = fixfree; + if (i) fixfree = fnext[i]; + else { nfixu++; i = nfixu; if (i >= NF) fatal("too many forward references"); } + fsite[i] = site; fnext[i] = link; + return i; +} +// patch every lit in a chain to point at v, and recycle the nodes +fn fixall(head, v) +{ + let n; + while (head) { setlit(fsite[head], v); n = fnext[head]; fnext[head] = fixfree; fixfree = head; head = n; } + return 0; +} + +// ---- code emission ---- +fn emit(w) { if (dry) return 0; if (here >= hlimit) fatal("program too big"); *(here + hdelta) = w; here++; return 0; } +fn lit(v) { emit(v + v + 1); return 0; } +fn op(k) { emit(k + k); return 0; } +fn setlit(at, v) { if (dry) return 0; *(at + hdelta) = v + v + 1; return 0; } +fn jmp(t) { lit(t); op(O_JMP); return 0; } +// a lit that will be patched later; returns its address +fn hole() { let a = here; lit(MW - 1); return a; } +fn fwdfn(name) { let s = addsym(name, K_FN, tline); sflag[s] = 0; return s; } // the line of the first use, until it's defined +fn fnref(s) +{ + if (sflag[s] & F_DEF) { lit(sval[s]); return 0; } + if (dry) return 0; + sfix[s] = fixnew(here, sfix[s]); + lit(MW - 1); // a jump to MW-1 halts if it ever runs unpatched + return 0; +} +fn callname(name) { let s = lookup(name); if (s < 0) s = fwdfn(name); fnref(s); op(O_CALL); return 0; } +fn laddr(k) { lit(ZFP); op(O_LD); if (k) { lit(k); op(O_ADD); } return 0; } +fn strlit(s) +{ + let j = hole(), a; + op(O_JMP); + a = here; + while (*s) { emit(*s); s++; } + emit(0); + setlit(j, here); + lit(a); + return 0; +} + +// ---- input sources ---- +// frame: 0 kind (0 stdin, 1 token list, 2 string) 1 object 2 index 3 end +// 4 transparent 5..7 saved token 8..10 saved char state 11 owned 12 saved line +// 13 saved token line 14 saved lookahead count 15..30 saved lookahead 31 saved char buffer +fn nextch() +{ + if (chkind == 0) { + if (qdone || eof()) { ch = -1; return 0; } + ch = getc() & 255; + if (ch == 96) { qdone = 1; ch = -1; return 0; } // a backtick ends the source + } else if (chkind == 3) { + ch = (chbase[chptr >> 2] >> ((chptr & 3) * 8)) & 255; + if (ch) chptr++; else ch = -1; + } else { + ch = *chptr & 255; + if (ch) chptr++; else ch = -1; + } + if (ch == 10) line++; + return 0; +} +fn src_push(kind, obj, trans, owned) +{ + let f, i; + isp++; + if (isp >= 32) fatal("sources nested too deeply"); + f = R_SRC + isp * 32; + f[0] = kind; f[1] = obj; f[2] = 0; f[3] = 0; f[4] = trans; f[11] = owned; + f[5] = tk; f[6] = tv; f[7] = tglue; f[8] = ch; f[9] = chkind; f[10] = chptr; + f[12] = line; f[13] = tline; f[14] = pq_n; f[31] = chbase; + i = 0; while (i < pq_n) { f[15 + i * 4] = pq_tk[i]; f[16 + i * 4] = pq_tv[i]; f[17 + i * 4] = pq_gl[i]; f[18 + i * 4] = pq_ln[i]; i++; } + pq_n = 0; + if (kind == 1) f[3] = obj[0]; + if (kind == 0) { chkind = 0; qdone = 0; line = 1; nextch(); } + if (kind == 2) { chkind = 2; chptr = obj; line = 1; nextch(); } + if (kind == 3) { chkind = 3; chbase = obj; chptr = 0; line = 1; nextch(); } + return 0; +} +fn src_pop() +{ + let f = R_SRC + isp * 32, i = 0; + tk = f[5]; tv = f[6]; tglue = f[7]; ch = f[8]; chkind = f[9]; chptr = f[10]; + line = f[12]; tline = f[13]; pq_n = f[14]; chbase = f[31]; + while (i < pq_n) { pq_tk[i] = f[15 + i * 4]; pq_tv[i] = f[16 + i * 4]; pq_gl[i] = f[17 + i * 4]; pq_ln[i] = f[18 + i * 4]; i++; } + isp--; + if (f[11]) free(f[1]); + return 0; +} +fn isopch(c) +{ + let s = "+-*/%=<>!&|^~?:.@#$"; + while (*s) { if (*s == c) return 1; s++; } + return 0; +} +// is buf[0..n) followed by c the start of some operator the lexer knows? +fn opprefix(n, c) +{ + let i = 0, k, j; + while (i < nkop) { + k = kops[i]; + j = 0; + while (j < n && k[j] == tbuf[j]) j++; + if (j == n && k[n] == c) return 1; + i++; + } + return 0; +} +fn isdig(c) { return c >= '0' && c <= '9'; } +fn isal(c) { return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || c == '_'; } +fn hexv(c) +{ + if (isdig(c)) return c - '0'; + if (c >= 'a' && c <= 'f') return c - 'a' + 10; + if (c >= 'A' && c <= 'F') return c - 'A' + 10; + return -1; +} +fn escch() +{ + let c = ch; + nextch(); + if (c != 92) return c; + c = ch; + nextch(); + if (c == 'n') return 10; + if (c == 't') return 9; + if (c == 'r') return 13; + if (c == '0') return 0; + if (c == 'e') return 27; + return c; +} +fn oplex(n) +{ + while (ch >= 0 && isopch(ch) && opprefix(n, ch)) { tbuf[n] = ch; n++; nextch(); } + tbuf[n] = 0; + tk = T_OP; tv = intern(tbuf); + return 0; +} +fn lex() +{ + let n = 0, v, dg; + tglue = 1; + while (1) { + while (ch == 32 || ch == 9 || ch == 10 || ch == 13) { nextch(); tglue = 0; } + if (ch != '/') break; + nextch(); + if (ch == '/') { while (ch >= 0 && ch != 10) nextch(); tglue = 0; continue; } + if (ch == '*') { + nextch(); + while (ch >= 0) { if (ch == '*') { nextch(); if (ch == '/') break; } else nextch(); } + nextch(); tglue = 0; continue; + } + tline = line; tbuf[0] = '/'; oplex(1); return 0; + } + tline = line; + if (ch < 0) { tk = T_EOF; tv = 0; return 0; } + if (isdig(ch)) { + v = 0; + if (ch == '0') { + nextch(); + if (ch == 'x' || ch == 'X') { + nextch(); + while ((dg = hexv(ch)) >= 0) { v = v * 16 + dg; nextch(); } + tk = T_NUM; tv = v; return 0; + } + } + while (isdig(ch)) { v = v * 10 + ch - '0'; nextch(); } + tk = T_NUM; tv = v; return 0; + } + if (isal(ch)) { + while (isal(ch) || isdig(ch)) { if (n < 1000) { tbuf[n] = ch; n++; } nextch(); } + tbuf[n] = 0; + tk = T_ID; tv = intern(tbuf); return 0; + } + if (ch == 39) { nextch(); tv = escch(); if (ch != 39) die("bad character literal"); nextch(); tk = T_NUM; return 0; } + if (ch == 34) { + nextch(); + while (ch >= 0 && ch != 34) { v = escch(); if (n < 1000) { tbuf[n] = v; n++; } } + nextch(); + tbuf[n] = 0; + tk = T_STR; tv = intern(tbuf); return 0; + } + tbuf[0] = ch; + nextch(); + if (isopch(tbuf[0])) return oplex(1); + tbuf[1] = 0; + tk = T_OP; tv = intern(tbuf); + return 0; +} +// one token of lookahead: peek() fills pk_*, next() takes it first +fn next() +{ + let i = 1; + if (pq_n) { + tk = pq_tk[0]; tv = pq_tv[0]; tglue = pq_gl[0]; tline = pq_ln[0]; + while (i < pq_n) { pq_tk[i - 1] = pq_tk[i]; pq_tv[i - 1] = pq_tv[i]; pq_gl[i - 1] = pq_gl[i]; pq_ln[i - 1] = pq_ln[i]; i++; } + pq_n--; + return 0; + } + return next0(); +} +// the k-th token after the current one (k from 0 to 3), into pk_tk and pk_tv +fn peekat(k) +{ + let a, b, c, e; + while (pq_n <= k) { + a = tk; b = tv; c = tglue; e = tline; + next0(); + pq_tk[pq_n] = tk; pq_tv[pq_n] = tv; pq_gl[pq_n] = tglue; pq_ln[pq_n] = tline; pq_n++; + tk = a; tv = b; tglue = c; tline = e; + } + pk_tk = pq_tk[k]; pk_tv = pq_tv[k]; + return 0; +} +fn peek() { return peekat(0); } +fn next0() +{ + let f, q, i; + f = R_SRC + isp * 32; + if (f[0] != 1) { lex(); return 0; } + i = f[2]; + if (i < f[3]) { q = f[1]; tk = q[1 + i * 3]; tv = q[2 + i * 3]; tline = q[3 + i * 3]; tglue = 0; f[2] = i + 1; return 0; } + if (!f[4]) { tk = T_EOF; tv = 0; return 0; } + src_pop(); // a fed list ran out: continue with the token that followed it + return 0; +} +fn is_op(s) { return tk == T_OP && tv == s; } +fn is_id(s) { return tk == T_ID && tv == s; } +fn accept(s) { if ((tk == T_OP || tk == T_ID) && tv == s) { next(); return 1; } return 0; } +fn expect(s) { if (!accept(s)) { if (!panic && !dry) { errat(); puts("expected "); puts(s); errend(); } panic = 1; if (tk == T_EOF) giveup(); } return 0; } +fn name() +{ + let n = tv; + if (tk != T_ID) die("expected a name"); + next(); + return n; +} + +// ---- operator tables ---- +// a sequence is written like asm: one-char mnemonics, numbers, and names +// (resolved when the operator is used, so image builds stay self-contained) +fn seqasm(s) +{ + let start = seqtop, n, neg, v, k, c; + while (1) { + while (*s == 32) s++; + if (!*s) break; + n = 0; + while (s[n] && s[n] != 32) { tbuf[n] = s[n]; n++; } + tbuf[n] = 0; + s = s + n; + neg = tbuf[0] == '-' && isdig(tbuf[1]); + if (isdig(tbuf[neg])) { + v = 0; k = neg; + while (tbuf[k]) { v = v * 10 + tbuf[k] - '0'; k++; } + if (neg) v = 0 - v; + R_SEQ[seqtop] = v + v + 1; seqtop++; + } else if (n == 1 && (c = mnemidx(tbuf[0])) >= 0) { + R_SEQ[seqtop] = c + c; seqtop++; + } else { + R_SEQ[seqtop] = SEQREF; R_SEQ[seqtop + 1] = intern(tbuf); seqtop = seqtop + 2; + } + } + return start; +} +fn mnemidx(c) { let i = 0; while (mnem[i]) { if (mnem[i] == c) return i; i++; } return -1; } +fn emitseq(a, n) +{ + let e = a + n, s; + while (a < e) { + if (R_SEQ[a] == SEQREF) { + s = lookup(R_SEQ[a + 1]); + if (s < 0) s = fwdfn(R_SEQ[a + 1]); + if (skind[s] == K_FN) fnref(s); else if (skind[s] == K_ENH) op(XOP0 + sval[s]); else lit(sval[s]); + a = a + 2; + } else { emit(R_SEQ[a]); a++; } + } + return 0; +} +fn addkop(p) { let i = 0; while (i < nkop) { if (kops[i] == p) return 0; i++; } kops[nkop] = p; nkop++; return 0; } +fn binfind(p) { let i = nbin; while (i) { i--; if (bname[i] == p) return i; } return -1; } +fn unfind(p) { let i = nun; while (i) { i--; if (uname[i] == p) return i; } return -1; } +fn asgfind(p) { let i = nasg; while (i) { i--; if (aname[i] == p) return i; } return -1; } +// add or replace a binary operator. code is a sequence string, or 0 for the +// built-in && (prec 2) and || (prec 1) behaviour, chosen by -1 / -2 in prec. +fn add_binop(nm, prec, code) +{ + let p = intern(nm), i = binfind(p), a, n; + if (i < 0) { i = nbin; nbin++; } + bname[i] = p; bprec[i] = prec; + if (prec < 0) { bprec[i] = 0 - prec; blen[i] = prec; bseq[i] = 0; addkop(p); return i; } + bseq[i] = seqasm(code); blen[i] = seqtop - bseq[i]; + addkop(p); + // and its compound assignment, x op= y + n = strlen(nm); + wcopy(tbuf + 1000, nm, n); + tbuf[1000 + n] = '='; tbuf[1001 + n] = 0; + a = intern(tbuf + 1000); + if (prec >= 8 || prec == 5 || prec == 4 || prec == 3) { + if (asgfind(a) < 0) { aname[nasg] = a; nasg++; } + abin[asgfind(a)] = i; + addkop(a); + } + return i; +} +fn add_unop(nm, code) +{ + let p = intern(nm), i = unfind(p); + if (i < 0) { i = nun; nun++; } + uname[i] = p; useq[i] = seqasm(code); ulen[i] = seqtop - useq[i]; + addkop(p); + return i; +} +fn add_kw(nm, handler, flags) { let s = addsym(intern(nm), K_KW, handler); sflag[s] = F_DEF | flags; return s; } + +// ---- locals and the analyzer's view of them ---- +fn findloc(n) { let i = nloc; while (i > lbase) { i--; if (lname[i] == n) return i; } return -1; } +fn newloc(n) +{ + let i = nloc; + if (nloc >= NL) fatal("too many locals"); + nloc++; + lname[i] = n; lslot[i] = fslots; lhome[i] = fdepth; lstate[i] = S_PLAIN; + lowner[i] = 0; lused[i] = 0; larr[i] = 0; lown[i] = 0; ltype[i] = tword; ltag[i] = 0; + return i; +} +fn lroot(i) { while (lstate[i] == S_CHILD) i = lowner[i]; return i; } +fn tracked(i) { return lstate[i] != S_PLAIN; } +// a use that doesn't give the object away +fn touch(i) +{ + let r; + lused[i] = 1; + if (!fchk) return 0; + r = lroot(i); + lused[r] = 1; + if (lstate[r] == S_DEAD) { cerrn(i, "used after it was given away"); return 0; } + // a d object that shows up in a deeper block is given to that block + if (lstate[r] == S_D && lhome[r] < fdepth) { lhome[r] = fdepth; lstate[r] = S_M; lown[r] = 1; } + return 0; +} +fn plainuse(c) +{ + if (c == C_FRESH) cerr("a new object can't be used as a plain value; bind it to a name first"); + if (c >= C_VAR) touch(c - C_VAR); + return C_PLAIN; +} +fn freevar(i) { laddr(lslot[i]); op(O_LD); callname(s_free); op(O_DROP); return 0; } +// leaving every block deeper than or equal to dep: free what is owned, and +// complain about what should have left already. final means the block really +// ends here rather than an early exit. +fn exit_vars(dep, final) +{ + let i = nloc, st; + if (!fchk || !live) return 0; + while (i > lbase) { + i--; + if (lhome[i] < dep) continue; + st = lstate[i]; + if (st == S_M) freevar(i); else if (st == S_U) cerrn(i, "u object must leave this block (block value, assignment to the enclosing block, or return)"); + else if (st == S_D) cerrn(i, "d object was never given away"); + } + return 0; +} +// an m object has to be used somewhere in the block that owns it, on any path +fn unused_check(from) +{ + while (from < nloc) { if (fchk && lown[from] && !lused[from]) serrn(lname[from], "m object is never used"); from++; } + return 0; +} +fn block_close(save) +{ + let i = lbase; + unused_check(save); + exit_vars(fdepth, 1); + while (i < save) { if (lhome[i] >= fdepth) { lstate[i] = S_DEAD; lhome[i] = fdepth - 1; } i++; } + nloc = save; + return 0; +} + +// snapshots: [lbase] [nloc] then state, owner, home per local +fn snap_push() +{ + let b = snaptop, i = lbase, p; + p = R_SNAP + b; + p[0] = lbase; p[1] = nloc; + p = p + 2; + while (i < nloc) { p[0] = lstate[i]; p[1] = lowner[i]; p[2] = lhome[i]; p = p + 3; i++; } + snaptop = p - R_SNAP; + if (snaptop >= 8000) fatal("nesting too deep for the analyzer"); + return b; +} +fn snap_pop(b) { snaptop = b; return 0; } +fn snap_restore(b) +{ + let p = R_SNAP + b, i = p[0], e = p[1]; + if (e > nloc) e = nloc; + p = p + 2; + while (i < e) { lstate[i] = p[0]; lowner[i] = p[1]; lhome[i] = p[2]; p = p + 3; i++; } + return 0; +} +fn empty(s) { return s == S_PLAIN || s == S_DEAD; } +// compare the current states with a snapshot: -1 if they agree, else the +// first local that differs. plain and dead agree and merge to dead. +fn snap_cmp(b) +{ + let p = R_SNAP + b, i = p[0], e = p[1], s; + if (e > nloc) e = nloc; + p = p + 2; + while (i < e) { + s = lstate[i]; + if (empty(s) && empty(p[0])) { if (p[0] == S_DEAD) lstate[i] = S_DEAD; } + else if (s != p[0] || lhome[i] != p[2] || (s == S_CHILD && lowner[i] != p[1])) return i; + p = p + 3; i++; + } + return -1; +} +fn statename(s) +{ + if (s == S_PLAIN) return "empty"; + if (s == S_M) return "m"; + if (s == S_U) return "u"; + if (s == S_D) return "d"; + if (s == S_DEAD) return "given away"; + if (s == S_CHILD) return "owned by a container"; + return "borrowed"; +} +fn merge_err(i, sa, what) +{ + if (!fchk || panic || dry) return 0; + errat(); puts(lname[i]); puts(": "); puts(what); + puts(" (one path leaves it "); puts(statename(sa)); puts(", another "); puts(statename(lstate[i])); puts(")"); + errend(); + return 0; +} +// after an if: the current state is the else path, sa is the then path + +// ---- joins ---- +// where control flow comes back together (after an if, a loop, a switch or +// ?:) every variable has to agree across the paths that reach the join. one +// exception: a variable that became d, or was given away, on some path is dead +// after the join, and the paths where it's still alive free it on the way in +// (fixups). u objects and borrows don't get this treatment: a u object still +// has to leave, and a borrow never changes. +const R_EXIT = 0xBF500, R_EXITEND = 0xBFF00; // exit records: the states at each break +let extop, jps[64], njps; // the paths being joined: [lbase, nloc, triples] +fn jst(p, i) { if (i >= p[1]) return S_PLAIN; return p[2 + 3 * (i - p[0])]; } +fn jow(p, i) { return p[3 + 3 * (i - p[0])]; } +fn jhm(p, i) { return p[4 + 3 * (i - p[0])]; } +// the states at a break, with the jump to patch: [site] then a snapshot +fn exit_rec(site) +{ + let p = R_EXIT + extop, i = lbase, q; + if (extop + 3 + 3 * (nloc - lbase) >= R_EXITEND - R_EXIT) fatal("too many breaks for the analyzer"); + p[0] = site; p[1] = lbase; p[2] = nloc; + q = p + 3; + while (i < nloc) { q[0] = lstate[i]; q[1] = lowner[i]; q[2] = lhome[i]; q = q + 3; i++; } + extop = q - R_EXIT; + return p; +} +fn jadd(p) { if (njps >= 64) fatal("too many paths for the analyzer"); jps[njps] = p; njps++; return 0; } +fn jsame(p, q, i) +{ + let s = jst(p, i), t = jst(q, i); + if (empty(s) && empty(t)) return 1; + return s == t && jhm(p, i) == jhm(q, i) && (s != S_CHILD || jow(p, i) == jow(q, i)); +} +// join the paths in jps over the locals [lbase, n0) into the current states +fn join(n0, what) +{ + let i = lbase, j, p, s, s0, dead, okdead, bad; + if (!njps) return 0; + while (i < n0) { + p = jps[0]; s0 = jst(p, i); + dead = 0; okdead = 1; bad = -1; + j = 0; + while (j < njps) { + s = jst(jps[j], i); + if (s == S_DEAD || s == S_D) dead = 1; + if (s == S_U || s == S_BORROW) okdead = 0; + if (!jsame(p, jps[j], i)) bad = j; + j++; + } + lstate[i] = s0; lowner[i] = jow(p, i); lhome[i] = jhm(p, i); + if (bad < 0) { + if (empty(s0)) { j = 0; while (j < njps) { if (jst(jps[j], i) == S_DEAD) lstate[i] = S_DEAD; j++; } } + } else if (dead && okdead) lstate[i] = S_DEAD; + else if (fchk && !panic && !dry) { + errat(); puts(lname[i]); puts(": "); puts(what); + puts(" (one path leaves it "); puts(statename(s0)); puts(", another "); puts(statename(jst(jps[bad], i))); puts(")"); + errend(); + } + i++; + } + return 0; +} +// on path p: count (and with doit, emit) the frees for what the join made dead but p still owns +fn fixups(p, n0, doit) +{ + let i = lbase, n = 0, s; + if (!fchk) return 0; + while (i < n0) { + s = jst(p, i); + if (lstate[i] == S_DEAD && (s == S_M || s == S_D)) { n++; if (doit) freevar(i); } + i++; + } + return n; +} +// a jump back to the top of a loop: whatever the loop owns has to be as it was +fn backedge(s0) +{ + let p = R_SNAP + s0, i = p[0], e = p[1], s; + if (!fchk || !live) return 0; + if (e > nloc) e = nloc; + while (i < e) { + s = lstate[i]; + if (empty(s) && empty(jst(p, i))) { if (jst(p, i) == S_DEAD) lstate[i] = S_DEAD; } + else if (s != jst(p, i) || lhome[i] != jhm(p, i) || (s == S_CHILD && lowner[i] != jow(p, i))) { + if (s == S_DEAD || s == S_D) cerrn(i, "is given away inside a loop that comes back around to it; give it away on a path that leaves the loop"); + else if (!panic && !dry) { + errat(); puts(lname[i]); puts(": changes state inside a loop; each pass has to leave it as it found it (it went from "); + puts(statename(jst(p, i))); puts(" to "); puts(statename(s)); puts(")"); errend(); + } + } + i++; + } + return 0; +} +// joins two paths that end at the same point, like the branches of if/else: +// path a jumped (through the hole at ja) and path b falls through to here +fn join2(sa, livea, ja, livebv, n0, what) +{ + let sb, nt, cj; + sb = snap_push(); + njps = 0; if (livea) jadd(R_SNAP + sa); if (livebv) jadd(R_SNAP + sb); + join(n0, what); + if (livebv) fixups(R_SNAP + sb, n0, 1); + nt = 0; if (livea) nt = fixups(R_SNAP + sa, n0, 0); + if (nt) { cj = hole(); op(O_JMP); setlit(ja, here); fixups(R_SNAP + sa, n0, 1); setlit(cj, here); } + else setlit(ja, here); + live = livea || livebv; + return 0; +} + +// ---- analyzer rules for the places values flow ---- +// recast: x as u / x as d / x as m +fn recast(c, t) +{ + let i, st; + if (!fchk) return c; + if (c < C_VAR) { cerr("only a named owned object can be recast"); return c; } + i = c - C_VAR; + st = lstate[i]; + lused[i] = 1; + if (st == S_CHILD) cerrn(i, "belongs to a container; recast the container instead"); + else if (st == S_BORROW) cerrn(i, "is borrowed and can't be recast"); + else if (st == S_PLAIN) cerrn(i, "isn't an owned object"); + else if (st == S_DEAD) cerrn(i, "used after it was given away"); + else if (t == S_M && st != S_M) cerrn(i, "can't be recast back to m"); + else lstate[i] = t; + return c; +} +// an argument for a parameter tagged t (0 plain, 1 m, 2 d) +fn argcheck(c, t) +{ + let i, st; + if (!fchk) return 0; + if (c == C_FRESH) { if (t != 2) cerr("a new object can only go straight to a d parameter; bind it to a name first"); return 0; } + if (c < C_VAR || !tracked(c - C_VAR)) return 0; + i = c - C_VAR; + st = lstate[i]; + if (lstate[lroot(i)] == S_DEAD) { cerrn(i, "used after it was given away"); return 0; } + if (t <= 1) { // untagged parameters are m: a lend + if (st == S_D) { cerrn(i, "is d; pass it to a d parameter (lending it would lose it)"); return 0; } + touch(i); + return 0; + } + if (st == S_D) { lstate[i] = S_DEAD; lused[i] = 1; return 0; } + if (st == S_M) cerrn(i, "is m; write d*x to give it away"); + else if (st == S_U) cerrn(i, "is u; write d*x to give it away instead of returning it"); + else if (st == S_CHILD) cerrn(i, "belongs to a container and can't be given away alone"); + else cerrn(i, "is borrowed and can't be given away"); + return 0; +} +fn emitown(p) { laddr(lslot[p]); op(O_LD); callname(s_own); return 0; } +// a store of class rc into an lvalue of kind k (local or container index i) +fn assigncheck(k, i, rc) +{ + let st, j, r, js; + if (!fchk) return C_PLAIN; + j = -1; + if (rc >= C_VAR && tracked(rc - C_VAR)) j = rc - C_VAR; + if (j >= 0 && lstate[lroot(j)] == S_DEAD) { cerrn(j, "used after it was given away"); return C_PLAIN; } + if (k == LV_LOCAL) { + st = lstate[i]; + if (j < 0 && rc != C_FRESH) { + if (!empty(st)) cerrn(i, "still owns an object; overwriting it would leak it"); + else lstate[i] = S_PLAIN; + return C_PLAIN; + } + if (!empty(st)) { cerrn(i, "still owns an object; overwriting it would leak it"); return C_PLAIN; } + if (rc == C_FRESH) { lstate[i] = S_M; lused[i] = 0; lown[i] = 1; if (ltag[i]) lstate[i] = ltag[i]; return C_PLAIN; } + js = lstate[j]; + if (js == S_U && lhome[i] == lhome[j] - 1) { // handing a u object up one block + lstate[i] = S_M; lused[i] = 0; lown[i] = 1; lstate[j] = S_DEAD; lused[j] = 1; if (ltag[i]) lstate[i] = ltag[i]; + return C_PLAIN; + } + if (js == S_U) cerrn(j, "is u; it can only be handed to a variable of the enclosing block, or returned"); + else cerrn(j, "is an owned object and can't be copied into another variable"); + return C_PLAIN; + } + if (k == LV_ELEM) { + r = lroot(i); + st = lstate[r]; + touch(i); + if (rc == C_FRESH) { emitown(i); return C_PLAIN; } + if (j < 0) return C_PLAIN; + js = lstate[j]; + if (js == S_CHILD) cerrn(j, "already belongs to a container"); + else if (js == S_BORROW) cerrn(j, "is borrowed and can't be stored"); + else if (st == S_BORROW) { + if (js == S_U) cerrn(j, "is u; it has to leave through the block above, not into a borrowed object"); + else { lstate[j] = S_DEAD; lused[j] = 1; emitown(i); } // storing it gives it away + } else if (js != st) cerrn(j, "its tag doesn't match its container's (both have to be m, u or d)"); + else if (lhome[j] < lhome[r]) cerrn(j, "lives in an outer block than the container it's stored into"); + else { lstate[j] = S_CHILD; lowner[j] = i; lused[j] = 1; emitown(i); } + return C_PLAIN; + } + // a global, or memory the analyzer doesn't track: the collector takes it + if (j >= 0) { + if (lstate[j] == S_D) { lstate[j] = S_DEAD; lused[j] = 1; } + else cerrn(j, "must be d to be stored in a global or in untracked memory (the collector takes it over)"); + } + return C_PLAIN; +} +fn retcheck(c) +{ + let i; + if (!fchk) return 0; + i = -1; + if (c >= C_VAR && tracked(c - C_VAR)) i = c - C_VAR; + if (i >= 0 && lstate[lroot(i)] == S_DEAD) { cerrn(i, "used after it was given away"); return 0; } + if (fret) { + if (i >= 0) { + if (lstate[i] == S_U) { lstate[i] = S_DEAD; lused[i] = 1; } + else cerrn(i, "has to be u to be returned; write 'return u*x'"); + } + return 0; + } + if (c == C_FRESH) cerr("returns a new object; declare the function with -> .u"); + else if (i >= 0) cerrn(i, "is owned and can't be returned from a function without -> .u"); + return 0; +} +// the last expression of a block expression +fn handup(c) +{ + let i; + if (!fchk) return C_PLAIN; + if (c == C_FRESH) return C_FRESH; + if (c < C_VAR || !tracked(c - C_VAR)) return C_PLAIN; + i = c - C_VAR; + if (lstate[i] == S_U && lhome[i] == fdepth) { lstate[i] = S_DEAD; lused[i] = 1; return C_FRESH; } + if (lstate[i] == S_U) cerrn(i, "only a u object of this block can be its value"); + else cerrn(i, "has to be u to be handed up as the block's value"); + return C_PLAIN; +} +fn discard(c) +{ + if (c == C_FRESH) cerr("a new object (or a u result) can't be thrown away"); + else if (c >= C_VAR) touch(c - C_VAR); + return 0; +} + +// ---- types ---- +// types are a compiler primitive: records in the type area, hash-consed so +// the same type is always the same pointer and comparing types is comparing +// words. untyped code is typed word, which goes with anything. the records +// stay in memory after compiling, so programs can inspect them. +// t[0] kind t[1] size in words t[2] name t[3] n t[4] a t[5] b t[6] c t[7..] extra +// ptr: a = element arr: a = element, n = length +// fn: a = return type, n = parameter count, b = signature (u m d tags), extra = parameter types +// struct: n = field count (-1 until laid out), a = fields (name, type, offset), +// b = generic template, c = type arguments [n, t1, t2 ...] +// var: a type parameter, n = its index +// gen: a generic struct: n = parameter count, a = fields (name, type), b = field count, c = parameter names + + +fn talloc(n) { let p = tytop; if (tytop + n >= R_TYEND) fatal("the type area is full"); tytop = tytop + n; wfill(p, 0, n); return p; } +fn tnew(kind, size, nm, extra) +{ + let t = talloc(7 + extra); + t[0] = kind; t[1] = size; t[2] = nm; + if (ntl >= NTL) fatal("too many types"); + R_TY[ntl] = t; ntl++; + return t; +} +fn tptr(e) +{ + let i = 0, t; + while (i < ntl) { t = R_TY[i]; if (t[0] == TY_PTR && t[4] == e) return t; i++; } + t = tnew(TY_PTR, 1, 0, 0); t[4] = e; + return t; +} +fn tarr(e, n) +{ + let i = 0, t; + while (i < ntl) { t = R_TY[i]; if (t[0] == TY_ARR && t[4] == e && t[3] == n) return t; i++; } + t = tnew(TY_ARR, n * tsz(e), 0, 0); t[4] = e; t[3] = n; + return t; +} +fn tfn(ret, np, ps, sig) +{ + let i = 0, j, t; + while (i < ntl) { + t = R_TY[i]; + if (t[0] == TY_FN && t[4] == ret && t[3] == np && t[5] == sig) { + j = 0; while (j < np && t[7 + j] == ps[j]) j++; + if (j == np) return t; + } + i++; + } + t = tnew(TY_FN, 1, 0, np); t[4] = ret; t[3] = np; t[5] = sig; + wcopy(t + 7, ps, np); + return t; +} +fn tvar(i, nm) +{ + let k = 0, t; + while (k < ntl) { t = R_TY[k]; if (t[0] == TY_VAR && t[3] == i) return t; k++; } + t = tnew(TY_VAR, 1, nm, 0); t[3] = i; + return t; +} +fn tinst(g, args, n) +{ + let i = 0, j, t, v; + while (i < ntl) { + t = R_TY[i]; + if (t[0] == TY_STRUCT && t[5] == g) { + v = t[6]; j = 0; while (j < n && v[1 + j] == args[j]) j++; + if (j == n) return t; + } + i++; + } + v = talloc(n + 1); v[0] = n; wcopy(v + 1, args, n); + t = tnew(TY_STRUCT, 0, g[2], 0); t[3] = -1; t[5] = g; t[6] = v; + return t; +} +// lay out a generic instance on first use +fn tresolve(t) +{ + let g, f, n, i, fs, off = 0, ft, bs[8]; + if (t[0] != TY_STRUCT || t[3] >= 0) return t; + if (t[3] == -2) { serr("a struct can't contain itself by value"); t[3] = 0; return t; } + t[3] = -2; + g = t[5]; n = g[5]; f = g[4]; + wfill(bs, 0, 8); wcopy(bs, t[6] + 1, t[6][0]); + fs = talloc(n * 3 + 1); + i = 0; + while (i < n) { + ft = subst(f[i + i + 1], bs); + fs[i * 3] = f[i + i]; fs[i * 3 + 1] = ft; fs[i * 3 + 2] = off; + off = off + tsz(ft); + i++; + } + t[4] = fs; t[1] = off; t[3] = n; + return t; +} +fn tsz(t) { if (t[0] == TY_STRUCT && t[3] < 0) tresolve(t); return t[1]; } +// replace type parameters with what they were bound to (unbound ones become word) +fn subst(t, bs) +{ + let k = t[0], i, n, v[24], vec; + if (k == TY_VAR) { if (bs[t[3]]) return bs[t[3]]; return tword; } + if (k == TY_PTR) return tptr(subst(t[4], bs)); + if (k == TY_ARR) return tarr(subst(t[4], bs), t[3]); + if (k == TY_FN) { n = t[3]; i = 0; while (i < n) { v[i] = subst(t[7 + i], bs); i++; } return tfn(subst(t[4], bs), n, v, t[5]); } + if (k == TY_STRUCT && t[5]) { vec = t[6]; n = vec[0]; i = 0; while (i < n) { v[i] = subst(vec[1 + i], bs); i++; } return tinst(t[5], v, n); } + return t; +} +fn isnum(t) { return t[0] == TY_INT || t[0] == TY_CHAR || t[0] == TY_BOOL; } +fn isagg(t) { return t[0] == TY_ARR || t[0] == TY_STRUCT; } +fn isptr(t) { return t[0] == TY_PTR || t[0] == TY_ARR; } +fn decay(t) { if (t[0] == TY_ARR) return tptr(t[4]); return t; } +fn telem(t) { if (t[0] == TY_PTR || t[0] == TY_ARR) return t[4]; return tword; } +// can a value of type src go where dst is expected? +fn compat(dst, src) +{ + if (dst == src || dst == tword || src == tword) return 1; + if (isnum(dst) && isnum(src)) return 1; + src = decay(src); + if (dst[0] == TY_PTR && src[0] == TY_PTR) { + if (dst[4] == tword || src[4] == tword || dst[4] == tvoid || src[4] == tvoid) return 1; + return dst[4] == src[4]; + } + if ((dst[0] == TY_PTR || dst[0] == TY_FN) && xzero) return 1; // a literal 0 is a null pointer + return dst == src; +} +fn unify(p, a, bs) +{ + let i, n, pv, av; + if (p[0] == TY_VAR) { i = p[3]; if (!bs[i]) { bs[i] = a; return 1; } return compat(bs[i], a); } + a = decay(a); + if (p[0] == TY_PTR && a[0] == TY_PTR) { if (a[4] == tword) return 1; return unify(p[4], a[4], bs); } + if (p[0] == TY_STRUCT && p[5] && a[0] == TY_STRUCT && a[5] == p[5]) { + pv = p[6]; av = a[6]; n = pv[0]; i = 0; + while (i < n) { if (!unify(pv[1 + i], av[1 + i], bs)) return 0; i++; } + return 1; + } + return compat(p, a); +} +fn tprint(t) +{ + let k = t[0], i, v; + if (k == TY_PTR) { putc('*'); return tprint(t[4]); } + if (k == TY_ARR) { putc('['); putn(t[3]); putc(']'); return tprint(t[4]); } + if (k == TY_FN) { + puts("fn("); + i = 0; + while (i < t[3]) { + if (i) puts(", "); + tprint(t[7 + i]); + if (i < 12 && tparamtag(t, i) == 1) puts(".m"); + if (i < 12 && tparamtag(t, i) == 2) puts(".d"); + i++; + } + putc(')'); + if (t[4] != tword || (t[5] & 1)) { puts(" -> "); if (t[4] != tword) tprint(t[4]); if (t[5] & 1) puts(".u"); } + return 0; + } + if (t[2]) puts(t[2]); else puts("?"); + if (k == TY_STRUCT && t[5]) { + v = t[6]; putc('<'); + i = 0; while (i < v[0]) { if (i) puts(", "); tprint(v[1 + i]); i++; } + putc('>'); + } + return 0; +} +fn tcerr(what, want, got) +{ + if (!tc_on || !live || dry || panic) return 0; + errat(); puts("type error: "); puts(what); + puts(": expected "); tprint(want); puts(", got "); tprint(got); + errend(); + return 0; +} +fn tcneed(ok, what, got) +{ + if (ok || !tc_on || !live || dry || panic) return 0; + errat(); puts("type error: "); puts(what); puts(", got "); tprint(got); + errend(); + return 0; +} +fn tpfind(n) { let i = ntp; while (i) { i--; if (tpn[i] == n) return i; } return -1; } +fn istypename(n) +{ + let s; + if (tpfind(n) >= 0) return 1; + if (findloc(n) >= 0) return 0; + s = lookup(n); + return s >= 0 && skind[s] == K_TYPE; +} +// the closing > of a type argument list; splits >> in two +fn closeangle() +{ + if (is_op(s_shr)) { tv = s_gt; return 0; } + expect(s_gt); + return 0; +} +fn ptype() +{ + let n, s, t, np = 0, ps[24], ret, args[8], na = 0, i; + if (accept(s_star)) return tptr(ptype()); + if (accept(s_lk)) { n = cexpr(); expect(s_rk); t = ptype(); if (n < 0) serr("a negative array length"); return tarr(t, n); } + if (is_id(s_fn)) { + next(); expect(s_lp); + if (!is_op(s_rp)) { while (1) { if (np >= 24) die("too many parameters"); ps[np] = ptype(); np++; if (!accept(s_comma)) break; } } + expect(s_rp); + ret = tword; + i = 0; + if (accept(s_arrow)) { if (!is_op(s_dot)) ret = ptype(); if (accept(s_dot)) { if (!is_id(s_u)) die("the return tag is .u"); next(); i = 1; } } + return tfn(ret, np, ps, i + np + np); + } + if (tk != T_ID) { die("expected a type"); return tword; } + n = tv; + i = tpfind(n); + if (i >= 0) { next(); return tvar(i, n); } + s = lookup(n); + if (s < 0 || skind[s] != K_TYPE) { dien(n, "isn't a type"); return tword; } + next(); + t = sval[s]; + if (t[0] != TY_GEN) return t; + expect(s_lt); + while (1) { + if (na >= 8) die("too many type arguments"); + args[na] = ptype(); + if (tsz(args[na]) != 1) serr("type arguments have to be one word (a pointer, a number, or a parameter)"); + na++; + if (!accept(s_comma)) break; + } + closeangle(); + if (na != t[3]) serrn(n, "wrong number of type arguments"); + return tinst(t, args, na); +} +// does a declaration start here? (a type, or a bare .tag) +fn tstart() +{ + let k; + if (tk == T_OP) { + if (tv == s_lk || tv == s_dot) return 1; + if (tv == s_star) { // *T name, **T name ...: look past the stars + k = 0; + while (k < 3) { peekat(k); if (!(pk_tk == T_OP && pk_tv == s_star)) break; k++; } + peekat(k); + return (pk_tk == T_ID && (istypename(pk_tv) || pk_tv == s_fn)) || (pk_tk == T_OP && pk_tv == s_lk); + } + return 0; + } + if (tk != T_ID) return 0; + return tv == s_fn || istypename(tv); +} +// [type] [.tag] +fn dtype() +{ + dty = 0; dtag = 0; + if (!is_op(s_dot)) dty = ptype(); + if (accept(s_dot)) { + if (tv == s_m) dtag = S_M; else if (tv == s_u) dtag = S_U; else if (tv == s_d) dtag = S_D; else die("a tag is .u, .m or .d"); + next(); + } + return 0; +} +// struct name { type field, ... } or struct name<A, B> { ... } +fn kw_struct() +{ + let n, t, s, nf = 0, buf = R_QBUF, i, fs, off = 0, gen = 0, ft; + next(); + n = name(); + if (is_op(s_lt)) { // generic + next(); ntp = 0; + while (1) { if (ntp >= 8) die("too many type parameters"); tpn[ntp] = name(); ntp++; if (!accept(s_comma)) break; } + closeangle(); + gen = 1; + t = tnew(TY_GEN, 0, n, 0); t[3] = ntp; + fs = talloc(ntp); wcopy(fs, tpn, ntp); t[6] = fs; + } else { t = tnew(TY_STRUCT, 0, n, 0); t[3] = -2; } // -2 while its fields are read + s = addsym(n, K_TYPE, t); + expect(s_lb); + while (!is_op(s_rb)) { + ft = ptype(); + while (1) { + if (nf >= 1000) fatal("too many fields"); + buf[nf + nf] = name(); buf[nf + nf + 1] = ft; nf++; + if (!accept(s_comma)) break; + } + expect(s_semi); + } + next(); + if (gen) { + fs = talloc(nf + nf); wcopy(fs, buf, nf + nf); + t[4] = fs; t[5] = nf; + ntp = 0; + } else { + fs = talloc(nf * 3 + 1); + i = 0; + while (i < nf) { + ft = buf[i + i + 1]; + if (ft == t) { serr("a struct can't contain itself by value; use a pointer"); ft = tword; } + fs[i * 3] = buf[i + i]; fs[i * 3 + 1] = ft; fs[i * 3 + 2] = off; + off = off + tsz(ft); + i++; + } + t[4] = fs; t[1] = off; t[3] = nf; + } + accept(s_semi); + return s; +} +// type name = T; +fn kw_type() { let n; next(); n = name(); expect(s_eq); addsym(n, K_TYPE, ptype()); expect(s_semi); return 0; } +// typecheck on; / typecheck off; +fn kw_typecheck() +{ + next(); + if (is_id(s_on)) tc_on = 1; else if (is_id(s_off)) tc_on = 0; else die("typecheck on or typecheck off"); + next(); expect(s_semi); + return 0; +} +// the field of a struct type, or -1 +fn tfindfield(t, n) { let i = 0; tresolve(t); while (i < t[3]) { if (t[4][i * 3] == n) return i; i++; } return -1; } +// sizeof(T or expr), typeof(T or expr): the expression form is compiled dry, so it never runs +fn typearg() +{ + let t, sb, c; + expect(s_lp); + if (tstart() && !is_op(s_dot)) t = ptype(); + else { + sb = snap_push(); dry++; + c = expr(); rval(); + t = xty; + dry--; snap_restore(sb); snap_pop(sb); + } + expect(s_rp); + return t; +} +fn kw_sizeof() { let t; next(); t = typearg(); lit(tsz(t)); xty = tint; return 0; } +fn kw_typeof() { let t; next(); t = typearg(); lit(t); xty = tword; return 0; } +fn types_init() +{ + tytop = R_TY + NTL; ntl = 0; + tword = tnew(TY_WORD, 1, intern("word"), 0); + tint = tnew(TY_INT, 1, intern("int"), 0); + tchar = tnew(TY_CHAR, 1, intern("char"), 0); + tbool = tnew(TY_BOOL, 1, intern("bool"), 0); + tvoid = tnew(TY_VOID, 0, intern("void"), 0); + addsym(tword[2], K_TYPE, tword); addsym(tint[2], K_TYPE, tint); addsym(tchar[2], K_TYPE, tchar); + addsym(tbool[2], K_TYPE, tbool); addsym(tvoid[2], K_TYPE, tvoid); + s_lt = intern("<"); s_gt = intern(">"); s_shr = intern(">>"); s_le = intern("<="); s_ge = intern(">="); + s_eqeq = intern("=="); s_ne = intern("!="); s_plus = intern("+"); + s_struct = intern("struct"); s_type = intern("type"); s_typecheck = intern("typecheck"); + s_on = intern("on"); s_off = intern("off"); s_new = intern("new"); s_new_t = intern("new_t"); + numops[0] = intern("*"); numops[1] = intern("/"); numops[2] = intern("%"); numops[3] = intern("&"); + numops[4] = intern("|"); numops[5] = intern("^"); numops[6] = intern("<<"); numops[7] = s_shr; nnumops = 8; + return 0; +} +fn isnumop(o) { let i = 0; while (i < nnumops) { if (numops[i] == o) return 1; i++; } return 0; } +// the type of a binary operation, checking it when the typechecker is on +fn bintype(b, lt, rt) +{ + let o = bname[b]; + lt = decay(lt); rt = decay(rt); + if (blen[b] < 0) { tcneed(!isagg(lt) && !isagg(rt), "&& and || need scalars", lt); return tbool; } + if (o == s_eqeq || o == s_ne) { if (!compat(lt, rt) && !compat(rt, lt)) tcerr("comparison", lt, rt); return tbool; } + if (o == s_lt || o == s_gt || o == s_le || o == s_ge) { + if (!(lt == tword || rt == tword || (isnum(lt) && isnum(rt)) || (lt[0] == TY_PTR && rt[0] == TY_PTR))) tcerr("comparison", lt, rt); + return tbool; + } + if (o == s_plus) { + if (lt[0] == TY_PTR && (isnum(rt) || rt == tword)) return lt; + if (rt[0] == TY_PTR && isnum(lt)) return rt; + if (lt == tword || rt == tword) return tword; + if (isnum(lt) && isnum(rt)) return tint; + tcerr("+", lt, rt); return tword; + } + if (o == s_minus) { + if (lt[0] == TY_PTR && rt[0] == TY_PTR) return tint; + if (lt[0] == TY_PTR && (isnum(rt) || rt == tword)) return lt; + if (lt == tword || rt == tword) return tword; + if (isnum(lt) && isnum(rt)) return tint; + tcerr("-", lt, rt); return tword; + } + if (isnumop(o)) { + if (lt == tword || rt == tword) return tword; + if (isnum(lt) && isnum(rt)) return tint; + if (isnum(lt)) lt = rt; + tcneed(0, "arithmetic needs numbers", lt); return tword; + } + return tword; // operators a program added +} + +// ---- reflection ---- +// for programs (and parse words) looking at types at runtime +fn tkind(t) { return t[0]; } +fn tsize(t) { return tsz(t); } +fn tname(t) { return t[2]; } +fn tlen(t) { return t[3]; } +fn tnfields(t) { tresolve(t); return t[3]; } +fn tfname(t, i) { tresolve(t); return t[4][i * 3]; } +fn tftype(t, i) { tresolve(t); return t[4][i * 3 + 1]; } +fn tfoff(t, i) { tresolve(t); return t[4][i * 3 + 2]; } +fn tret(t) { return t[4]; } +fn tnparams(t) { return t[3]; } +fn tparam(t, i) { return t[7 + i]; } +fn tparamtag(t, i) { return ((t[5] >> 6) >> (i + i)) & 3; } // 0 untagged (m), 1 .m, 2 .d +fn tretu(t) { return t[5] & 1; } +fn tnargs(t) { if (t[0] != TY_STRUCT || !t[5]) return 0; return t[6][0]; } +fn targ(t, i) { return t[6][1 + i]; } +// the type of a function, by address. untyped functions get one made from their signature +fn sigof(f) +{ + let i = nsym, s = -1, ws[24], n, k; + while (i) { i--; if (skind[i] == K_FN && sval[i] == f && (sflag[i] & F_DEF)) { s = i; break; } } + if (s < 0) return 0; + if (stype[s]) return stype[s]; + n = (ssig[s] >> 1) & 31; k = 0; + while (k < n) { ws[k] = tword; k++; } + return tfn(tword, n, ws, ssig[s]); +} +fn sigofs(nm) { let s = lookups(nm); if (s < 0 || skind[s] != K_FN) return 0; return sigof(sval[s]); } + +// ---- expressions ---- +// every expression leaves its value on the stack, returns its analyzer class, +// and sets xty to its type. lv means the stack holds an address that hasn't +// been loaded yet; structs and arrays are never loaded, their address is +// their value. +fn rval() { if (lv) { if (!isagg(xty)) op(O_LD); lv = 0; } return 0; } +fn kwcall(s, e) +{ + let f = sval[s], old = kw_expr, c; + kw_expr = e; xcls = C_PLAIN; xty = tword; xzero = 0; + f(); + kw_expr = old; + c = xcls; + lv = 0; lvk = LV_NONE; + return c; +} +fn stmtonly() { if (kw_expr) serr("this can't be used inside an expression"); return 0; } +// the arguments of a call. ft is the callee's type (or 0), bs the generic bindings +fn args(tags, ft, bs) +{ + let np = 0, c, pt; + expect(s_lp); + if (!is_op(s_rp)) { + while (1) { + c = expr(); rval(); + argcheck(c, (tags >> (np + np)) & 3); + if (ft && np < ft[3]) { pt = ft[7 + np]; if (!unify(pt, xty, bs)) tcerr("argument", subst(pt, bs), xty); } + np++; + if (!accept(s_comma)) break; + } + } + expect(s_rp); + return np; +} +// new(T) and new(T, count): a typed object, its type stamped into the header +fn newtyped() +{ + let t, c; + t = ptype(); + if (accept(s_comma)) { + c = expr(); rval(); plainuse(c); + if (tsz(t) != 1) { lit(tsz(t)); op(15); } + lit(tarr(t, 0)); + } else { lit(tsz(t)); lit(t); } + expect(s_rp); + callname(s_new_t); + lv = 0; lvk = LV_NONE; xty = tptr(t); xzero = 0; + return C_FRESH; +} +fn call(s) +{ + let np, tags = 0, ret = 0, ft = stype[s], bs[8]; + if (sname[s] == s_new && skind[s] == K_FN) { + peek(); + if (pk_tk == T_OP && (pk_tv == s_star || pk_tv == s_lk)) { next(); return newtyped(); } + if (pk_tk == T_ID && (istypename(pk_tv) || pk_tv == s_fn)) { next(); return newtyped(); } + } + if (sflag[s] & F_SIG) { tags = ssig[s] >> 6; ret = ssig[s] & 1; } + else sflag[s] = sflag[s] | F_UNTYPED; + wfill(bs, 0, 8); + if (ngbind) { wcopy(bs, gbind, ngbind); ngbind = 0; } + np = args(tags, ft, bs); + if ((sflag[s] & F_SIG) && np != ((ssig[s] >> 1) & 31)) serrn(sname[s], "wrong number of arguments"); + fnref(s); op(O_CALL); + lv = 0; lvk = LV_NONE; xzero = 0; + xty = tword; + if (ft) xty = subst(ft[4], bs); + if (ret) return C_FRESH; + return C_PLAIN; +} +// an enhancement used like a function: the arguments, then its one opcode +fn enhcall(s) +{ + let nin = ssig[s] & 15, nout = (ssig[s] >> 4) & 15, n = 0, c; + expect(s_lp); + if (!is_op(s_rp)) { while (1) { c = expr(); rval(); plainuse(c); n++; if (!accept(s_comma)) break; } } + expect(s_rp); + if (n != nin) serrn(sname[s], "wrong number of inputs"); + lv = 0; lvk = LV_NONE; xty = tword; xzero = 0; + if (sflag[s] & F_FALLBACK) { fnref(sfix[s] - 1); op(O_CALL); return C_PLAIN; } + op(XOP0 + sval[s]); + if (nout == 0) lit(0); + if (nout > 1) serrn(sname[s], "leaves several values; use it from asm"); + return C_PLAIN; +} +fn primary() +{ + let n, i, s, c; + lv = 0; lvk = LV_NONE; xzero = 0; + if (tk == T_NUM) { lit(tv); xzero = tv == 0; xty = tint; next(); return C_PLAIN; } + if (tk == T_STR) { strlit(tv); next(); xty = tptr(tchar); return C_PLAIN; } + if (tk == T_OP) { + if (tv == s_lp) { next(); c = expr(); expect(s_rp); return c; } + if (tv == s_lb) return blockexpr(); + die("expected an expression"); lit(0); xty = tword; return C_PLAIN; + } + if (tk != T_ID) { die("expected an expression"); lit(0); xty = tword; return C_PLAIN; } + n = tv; + i = findloc(n); + if (i >= 0) { + next(); + if (!fchk && tracked(i)) serrn(n, "is tracked by the analyzer and can't be used in unchecked code"); + laddr(lslot[i]); + xty = ltype[i]; + if (larr[i]) return C_PLAIN; + lv = 1; lvk = LV_LOCAL; lvi = i; + if (tracked(i)) return C_VAR + i; + return C_PLAIN; + } + s = lookup(n); + if (s >= 0 && skind[s] == K_KW) { + if (!(sflag[s] & F_EXPR)) dien(n, "can't be used inside an expression"); + return kwcall(s, 1); + } + if (s >= 0 && skind[s] == K_TYPE) { dien(n, "is a type, not a value"); lit(0); xty = tword; return C_PLAIN; } + next(); + xty = tword; + if (s >= 0 && skind[s] == K_CONST) { lit(sval[s]); xty = tint; xzero = sval[s] == 0; return C_PLAIN; } + if (s >= 0 && skind[s] == K_GVAR) { lit(sval[s]); if (stype[s]) xty = stype[s]; lv = 1; lvk = LV_GLOBAL; return C_PLAIN; } + if (s >= 0 && skind[s] == K_GARR) { lit(sval[s]); if (stype[s]) xty = stype[s]; return C_PLAIN; } + if (s >= 0 && skind[s] == K_ENH) return enhcall(s); + if (s < 0) s = fwdfn(n); + if (is_op(s_lt) && (sflag[s] & F_GEN)) { // f<int, *char>(...) + next(); ngbind = 0; + while (1) { if (ngbind >= 8) die("too many type arguments"); gbind[ngbind] = ptype(); ngbind++; if (!accept(s_comma)) break; } + closeangle(); + if (!is_op(s_lp)) die("expected ( after the type arguments"); + } + if (is_op(s_lp)) return call(s); + fnref(s); + if (stype[s]) xty = stype[s]; + return C_PLAIN; +} +// ++ and -- on the address at the top of the stack +fn incdec(post, dlt) +{ + if (!lv) die("++ and -- need something to assign to"); + if (lvk == LV_LOCAL && fchk && tracked(lvi)) cerrn(lvi, "owns an object and can't be incremented"); + tcneed(isnum(xty) || xty[0] == TY_PTR || xty == tword, "++ and -- need a number or a pointer", xty); + lv = 0; lvk = LV_NONE; xzero = 0; + lit(0); op(O_PICK); op(O_LD); + if (post) { lit(0); op(O_PICK); lit(dlt); op(O_ADD); lit(2); op(O_PICK); op(O_ST); } + else { lit(dlt); op(O_ADD); lit(0); op(O_PICK); lit(2); op(O_PICK); op(O_ST); } + op(O_SWAP); op(O_DROP); + return C_PLAIN; +} +// p.x and p->x: a field of a struct, or of the struct a pointer points at +fn field(c) +{ + let n, st, i, b = -1; + next(); + n = name(); + rval(); + if (c == C_FRESH) cerr("use a field of a new object only after binding it to a name"); + else if (c >= C_VAR) { touch(c - C_VAR); b = c - C_VAR; } + st = xty; + if (st[0] == TY_PTR) st = st[4]; + if (st[0] != TY_STRUCT) { + if (!panic && !dry) { errat(); putc('.'); puts(n); puts(": that's a "); tprint(xty); puts(", not a struct or a pointer to one"); errend(); } + xty = tword; lv = 1; lvk = LV_OTHER; return C_PLAIN; + } + i = tfindfield(st, n); + if (i < 0) { + if (!panic && !dry) { errat(); tprint(st); puts(" has no field "); puts(n); errend(); } + xty = tword; lv = 1; lvk = LV_OTHER; return C_PLAIN; + } + if (st[4][i * 3 + 2]) { lit(st[4][i * 3 + 2]); op(O_ADD); } + xty = st[4][i * 3 + 1]; + lv = 1; xzero = 0; + if (b >= 0) { lvk = LV_ELEM; lvi = b; } else lvk = LV_OTHER; + return C_PLAIN; +} +fn postfix() +{ + let c = primary(), b, n, t, et, ft, bs[8]; + while (1) { + if (tk == T_ID && tv == s_as) { // e as T: same bits, another type + next(); t = ptype(); rval(); xty = t; lvk = LV_NONE; + continue; + } + if (tk != T_OP) return c; + if (tv == s_lk) { + next(); + rval(); + b = -1; + if (c == C_FRESH) cerr("index a new object only after binding it to a name"); + else if (c >= C_VAR) { touch(c - C_VAR); b = c - C_VAR; } + tcneed(isptr(xty) || xty == tword, "only pointers and arrays can be indexed", xty); + et = telem(xty); + c = expr(); rval(); plainuse(c); + tcneed(isnum(xty) || xty == tword, "an index has to be a number", xty); + expect(s_rk); + if (tsz(et) != 1) { lit(tsz(et)); op(15); } + op(O_ADD); + lv = 1; xty = et; xzero = 0; + if (b >= 0) { lvk = LV_ELEM; lvi = b; } else lvk = LV_OTHER; + c = C_PLAIN; + continue; + } + if (tv == s_dot || tv == s_arrow) { c = field(c); continue; } + if (tv == s_lp) { // call through a value + rval(); plainuse(c); + ft = 0; + if (xty[0] == TY_FN) ft = xty; + else tcneed(xty == tword, "only functions can be called", xty); + wfill(bs, 0, 8); + n = args(0, ft, bs); + lit(n); op(O_PICK); op(O_CALL); op(O_SWAP); op(O_DROP); + lv = 0; lvk = LV_NONE; c = C_PLAIN; xzero = 0; + xty = tword; + if (ft) xty = subst(ft[4], bs); + continue; + } + if (tv == s_inc) { next(); c = incdec(1, 1); continue; } + if (tv == s_dec) { next(); c = incdec(1, -1); continue; } + return c; + } + return c; +} +fn unary() +{ + let c, un, t; + if (tk == T_ID && (tv == s_u || tv == s_m || tv == s_d)) { // u*x, d*x, m*x: recast + t = S_M; if (tv == s_u) t = S_U; if (tv == s_d) t = S_D; + next(); expect(s_star); + c = unary(); rval(); + lvk = LV_NONE; + return recast(c, t); + } + if (tk == T_OP) { + if (tv == s_star) { + next(); c = unary(); rval(); plainuse(c); + tcneed(xty[0] == TY_PTR || xty == tword, "only pointers can be dereferenced", xty); + xty = telem(xty); + lv = 1; lvk = LV_OTHER; xzero = 0; + return C_PLAIN; + } + if (tv == s_amp) { + next(); c = unary(); + if (!lv && !isagg(xty)) die("& needs something addressable"); + if (c >= C_VAR) touch(c - C_VAR); + lv = 0; lvk = LV_NONE; xzero = 0; + xty = tptr(xty); + return C_PLAIN; + } + if (tv == s_inc) { next(); unary(); return incdec(0, 1); } + if (tv == s_dec) { next(); unary(); return incdec(0, -1); } + un = unfind(tv); + if (un >= 0) { + next(); c = unary(); rval(); plainuse(c); + emitseq(useq[un], ulen[un]); + if (un <= 1) xty = tbool; + else if (xty != tword) { tcneed(isnum(xty), "this operator needs a number", xty); xty = tint; } + lv = 0; lvk = LV_NONE; xzero = 0; + return C_PLAIN; + } + } + return postfix(); +} +fn bin(minp) +{ + let c = unary(), b, p, a, j, lt; + while (tk == T_OP) { + b = binfind(tv); + if (b < 0) return c; + p = bprec[b]; + if (p < minp) return c; + next(); rval(); plainuse(c); + lt = xty; + if (blen[b] == -1) { // && : 0 if the left side is 0, else the right side as 0 or 1 + a = hole(); op(O_JZ); + c = bin(p + 1); rval(); plainuse(c); + emitseq(useq[1], ulen[1]); // !! normalizes + j = hole(); op(O_JMP); + setlit(a, here); lit(0); setlit(j, here); + } else if (blen[b] == -2) { // || : 1 if the left side isn't 0, else the right side as 0 or 1 + a = hole(); op(O_JZ); + lit(1); j = hole(); op(O_JMP); + setlit(a, here); + c = bin(p + 1); rval(); plainuse(c); + emitseq(useq[1], ulen[1]); + setlit(j, here); + } else { + c = bin(p + 1); rval(); plainuse(c); + emitseq(bseq[b], blen[b]); + } + xty = bintype(b, lt, xty); + lv = 0; lvk = LV_NONE; c = C_PLAIN; xzero = 0; + } + return c; +} +// [addr v] -> [v], storing v at addr +fn storekeep() { op(O_SWAP); lit(1); op(O_PICK); op(O_SWAP); op(O_ST); return 0; } +// c ? a : b +fn ternary(c) +{ + let a, b, s0, sa, livea, live0 = live, n0 = nloc, ca, cb, ta; + rval(); plainuse(c); + next(); + a = hole(); op(O_JZ); + s0 = snap_push(); + ca = expr(); rval(); ta = xty; + if (ca != C_FRESH) plainuse(ca); + livea = live; sa = snap_push(); + b = hole(); op(O_JMP); + expect(s_colon); + setlit(a, here); + snap_restore(s0); live = live0; + cb = expr(); rval(); + if (cb != C_FRESH) plainuse(cb); + join2(sa, livea, b, live, n0, "ends the two sides of ?: in different states"); + snap_pop(s0); + if (!compat(ta, xty) && !compat(xty, ta)) tcerr("the two sides of ?:", ta, xty); + if (ta != tword) xty = ta; + lv = 0; lvk = LV_NONE; xzero = 0; + if (ca == C_FRESH && cb == C_FRESH) return C_FRESH; + if (ca == C_FRESH || cb == C_FRESH) cerr("both sides of ?: have to be new objects, or neither"); + return C_PLAIN; +} +fn expr() +{ + let c = bin(1), k, i, a, rc, lt; + if (tk != T_OP) return c; + if (tv == s_q) return ternary(c); + if (tv == s_eq) { + if (!lv) die("this can't be assigned to"); + k = lvk; i = lvi; lt = xty; + if (isagg(lt)) serr("a whole struct or array can't be assigned; assign its fields"); + next(); + rc = expr(); rval(); + if (!compat(lt, xty)) tcerr("assignment", lt, xty); + storekeep(); + lv = 0; lvk = LV_NONE; xty = lt; xzero = 0; + return assigncheck(k, i, rc); + } + a = asgfind(tv); + if (a >= 0) { + if (!lv) die("this can't be assigned to"); + if (lvk == LV_LOCAL && fchk && tracked(lvi)) cerrn(lvi, "owns an object; compound assignment would lose it"); + lt = xty; + next(); + lit(0); op(O_PICK); op(O_LD); + lv = 0; + rc = expr(); rval(); plainuse(rc); + emitseq(bseq[abin[a]], blen[abin[a]]); + k = bintype(abin[a], lt, xty); + if (!compat(lt, k)) tcerr("compound assignment", lt, k); + storekeep(); + lv = 0; lvk = LV_NONE; xty = lt; xzero = 0; + return C_PLAIN; + } + return c; +} +// does the current token start a statement rather than an expression? +fn stmtstart() +{ + let s; + if (tk == T_OP) return tv == s_semi || tv == s_lb || tstart(); + if (tk != T_ID || findloc(tv) >= 0) return 0; + if (tstart()) return 1; + s = lookup(tv); + return s >= 0 && skind[s] == K_KW && !(sflag[s] & F_EXPR); +} +// { stmts; tail } as an expression: its value is the tail, or 0 +fn blockexpr() +{ + let save = nloc, c = C_PLAIN, tail = 0, t = tint; + next(); + fdepth++; bxdepth++; + while (!is_op(s_rb)) { + if (tk == T_EOF) die("missing }"); + if (stmtstart()) { stmt(); continue; } + c = expr(); rval(); + if (accept(s_semi)) { op(O_DROP); discard(c); c = C_PLAIN; continue; } + if (!is_op(s_rb)) die("expected ; or } after an expression"); + tail = 1; t = xty; + } + next(); + if (tail) c = handup(c); else lit(0); + block_close(save); + fdepth--; bxdepth--; + lv = 0; lvk = LV_NONE; xty = t; xzero = 0; + return c; +} + +// ---- constant expressions ---- +fn cprim() +{ + let v, s; + if (tk == T_NUM) { v = tv; next(); return v; } + if (accept(s_minus)) return 0 - cprim(); + if (tk == T_OP && tv == uname[2]) { next(); return ~cprim(); } + if (accept(s_lp)) { v = cexpr(); expect(s_rp); return v; } + if (tk == T_ID) { + s = lookup(tv); + if (s >= 0 && (skind[s] == K_CONST || skind[s] == K_GARR || (skind[s] == K_FN && (sflag[s] & F_DEF)) || skind[s] == K_GVAR)) { next(); return sval[s]; } + } + die("expected a constant"); + return 0; +} +fn cexpr() +{ + let v = cprim(), o; + while (tk == T_OP) { + o = tv; + if (o == bname[3]) { next(); v = v + cprim(); } + else if (o == bname[4]) { next(); v = v - cprim(); } + else if (o == bname[0]) { next(); v = v * cprim(); } + else if (o == bname[15]) { next(); v = v | cprim(); } + else if (o == bname[13]) { next(); v = v & cprim(); } + else if (o == bname[5]) { next(); v = v << cprim(); } + else if (o == bname[6]) { next(); v = v >> cprim(); } + else return v; + } + return v; +} + +// ---- statements ---- +// after a syntax error: skip to the end of the statement +fn resync() +{ + while (tk != T_EOF && !is_op(s_semi) && !is_op(s_rb)) next(); + if (is_op(s_semi)) next(); + panic = 0; + return 0; +} +fn stmt() +{ + let s, c; + if (panic) return resync(); + if (tk == T_ID && findloc(tv) < 0) { + s = lookup(tv); + if (s >= 0 && skind[s] == K_KW && (sflag[s] & F_STMT)) { kwcall(s, 0); return 0; } + } + if (tstart()) { localdecl(1); return 0; } + if (tk == T_OP) { + if (tv == s_semi) { next(); return 0; } + if (tv == s_lb) { block(); return 0; } + } + c = expr(); rval(); op(O_DROP); + expect(s_semi); + discard(c); + return 0; +} +fn block() +{ + let save = nloc; + expect(s_lb); + fdepth++; + while (!is_op(s_rb)) { if (tk == T_EOF) die("missing }"); stmt(); } + block_close(save); // before the }, so its line is the one reported + next(); + fdepth--; + return 0; +} +// the body of if/while/for: its own block, even without braces +fn substmt() +{ + let save = nloc; + if (is_op(s_lb)) return block(); + fdepth++; + stmt(); + block_close(save); + fdepth--; + return 0; +} +// let a = e, b[10]; or, typed: int a = 1, b; *node.m p = new(node); .u r; +// let infers pointer, struct and function types from the initializer and +// keeps numbers as plain words. +fn localdecl(typed) +{ + let n, i, c, k, t = 0, tg = 0; + if (typed) { dtype(); t = dty; tg = dtag; if (t == tvoid) serr("a variable can't be void"); } + while (1) { + n = name(); + i = newloc(n); + ltag[i] = tg; + if (accept(s_lk)) { + k = cexpr(); expect(s_rk); + if (!t) t = tword; + larr[i] = 1; ltype[i] = tarr(t, k); fslots = fslots + tsz(ltype[i]); + } else if (t && isagg(t)) { larr[i] = 1; ltype[i] = t; fslots = fslots + tsz(t); } + else { fslots++; if (t) ltype[i] = t; } + if (accept(s_eq)) { + if (larr[i]) serr("an array or a struct can't be initialized with ="); + c = expr(); rval(); laddr(lslot[i]); op(O_ST); + if (t) { if (!compat(t, xty)) tcerr("initializer", t, xty); } + else if (!isnum(xty)) ltype[i] = xty; + assigncheck(LV_LOCAL, i, c); + } + if (!accept(s_comma)) break; + } + expect(s_semi); + return 0; +} +fn kw_var() { stmtonly(); next(); return localdecl(0); } +fn kw_if() +{ + let a, b, c, s0, sa, live0, livea, n0 = nloc, nt, cj; + stmtonly(); + next(); + expect(s_lp); c = expr(); rval(); plainuse(c); expect(s_rp); + a = hole(); op(O_JZ); + live0 = live; + s0 = snap_push(); + substmt(); + livea = live; sa = snap_push(); + if (is_id(s_else)) { + next(); + b = hole(); op(O_JMP); + setlit(a, here); + snap_restore(s0); live = live0; + substmt(); + join2(sa, livea, b, live, n0, "ends the branches of an if in different states"); + } else { + // the path that skips the then branch still has the states from before it + njps = 0; if (livea) jadd(R_SNAP + sa); if (live0) jadd(R_SNAP + s0); + join(n0, "ends the branches of an if in different states"); + if (livea) fixups(R_SNAP + sa, n0, 1); + nt = 0; if (live0) nt = fixups(R_SNAP + s0, n0, 0); + if (nt) { cj = hole(); op(O_JMP); setlit(a, here); fixups(R_SNAP + s0, n0, 1); setlit(cj, here); } + else setlit(a, here); + live = livea || live0; + } + snap_pop(s0); + return 0; +} +// loop frame: 0 continue target 1 break chain 2 depth 3 snapshot 4 block-expression depth +// loop frame: 0 continue target (-1 for a switch) 2 depth 3 snapshot at the top +// 4 block-expression depth 5 locals at the start 6 its first exit record +fn loop_push(cont, s0) +{ + let f = R_LOOP + lsp * 8; + if (lsp >= 60) fatal("loops nested too deeply"); + f[0] = cont; f[1] = 0; f[2] = fdepth; f[3] = s0; f[4] = bxdepth; f[5] = nloc; f[6] = extop; + lsp++; + return 0; +} +// the end of a loop or switch: join its exits (the fall-out path sc through +// the jump at a, if there is one, and every break) and lay out each path's +// fixups on its way to the end +fn loop_end(a, sc) +{ + let f, p, e, n0, outs = 0; + lsp--; + f = R_LOOP + lsp * 8; + n0 = f[5]; + njps = 0; + if (sc >= 0) jadd(R_SNAP + sc); + p = R_EXIT + f[6]; e = R_EXIT + extop; + while (p < e) { jadd(p + 1); p = p + 3 + 3 * (p[2] - p[1]); } + join(n0, "leaves the loop in different states on different paths"); + live = njps > 0; + p = R_EXIT + f[6]; + while (p < e) { + if (fixups(p + 1, n0, 0)) { setlit(p[0], here); fixups(p + 1, n0, 1); outs = fixnew(hole(), outs); op(O_JMP); } + else outs = fixnew(p[0], outs); + p = p + 3 + 3 * (p[2] - p[1]); + } + if (a >= 0) setlit(a, here); + if (sc >= 0) fixups(R_SNAP + sc, n0, 1); + fixall(outs, here); + extop = f[6]; + return 0; +} +fn kw_while() +{ + let c0 = here, a, s0, sc = -1, c, inf = 0; + stmtonly(); + next(); + s0 = snap_push(); + expect(s_lp); + if (tk == T_NUM && tv) { peek(); if (pk_tk == T_OP && pk_tv == s_rp) inf = 1; } // while (1): no way out but break + c = expr(); rval(); plainuse(c); expect(s_rp); + a = hole(); op(O_JZ); + if (!inf && live) sc = snap_push(); + loop_push(c0, s0); + substmt(); + backedge(s0); + jmp(c0); + loop_end(a, sc); + snap_pop(s0); + return 0; +} +fn kw_for() +{ + let save = nloc, s0, sc = -1, c0, a, b, st, c, inf = 0; + stmtonly(); + next(); + expect(s_lp); + fdepth++; + if (is_id(s_var)) kw_var(); + else if (tstart()) localdecl(1); + else { if (!is_op(s_semi)) { c = expr(); rval(); op(O_DROP); discard(c); } expect(s_semi); } + s0 = snap_push(); + c0 = here; + if (!is_op(s_semi)) { + if (tk == T_NUM && tv) { peek(); if (pk_tk == T_OP && pk_tv == s_semi) inf = 1; } + c = expr(); rval(); plainuse(c); + } else { lit(1); inf = 1; } + expect(s_semi); + a = hole(); op(O_JZ); + if (!inf && live) sc = snap_push(); + b = hole(); op(O_JMP); + st = here; + if (!is_op(s_rp)) { c = expr(); rval(); op(O_DROP); discard(c); } + expect(s_rp); + jmp(c0); + setlit(b, here); + loop_push(st, s0); + substmt(); + backedge(s0); + jmp(st); + loop_end(a, sc); + snap_pop(s0); + block_close(save); + fdepth--; + return 0; +} +fn kw_break() +{ + let f; + stmtonly(); + next(); expect(s_semi); + if (lsp <= loopbase) { serr("break outside a loop or a switch"); return 0; } + f = R_LOOP + (lsp - 1) * 8; + if (f[4] != bxdepth) { serr("break can't leave a block expression"); return 0; } + exit_vars(f[2] + 1, 0); + if (live) exit_rec(hole()); else hole(); + op(O_JMP); + live = 0; + return 0; +} +fn kw_continue() +{ + let j = lsp - 1, f; + stmtonly(); + next(); expect(s_semi); + while (j >= loopbase && R_LOOP[j * 8] == -1) j--; // past any switch + if (j < loopbase) { serr("continue outside a loop"); return 0; } + f = R_LOOP + j * 8; + if (f[4] != bxdepth) { serr("continue can't leave a block expression"); return 0; } + backedge(f[3]); + exit_vars(f[2] + 1, 0); + jmp(f[0]); + live = 0; + return 0; +} +// switch (e) { case 1, 2: ... case 'x': ... default: ... } +// cases don't fall through: each one ends where the next begins, and break +// leaves the switch early. case labels are constants. +fn casebody(s0, live0) +{ + let save = nloc; + snap_restore(s0); live = live0; + fdepth++; + while (!is_op(s_rb) && !is_id(s_case) && !is_id(s_default)) { if (tk == T_EOF) { die("missing }"); break; } stmt(); } + block_close(save); + fdepth--; + if (live) { exit_rec(hole()); op(O_JMP); } // the end of a case leaves the switch + live = 0; + return 0; +} +fn kw_switch() +{ + let save = nloc, tmp, c, s0, live0, defent = -1, skip, nxt, bj, v, a = -1; + stmtonly(); + next(); + expect(s_lp); c = expr(); rval(); plainuse(c); expect(s_rp); + fdepth++; + tmp = newloc(0); fslots++; // a hidden local holds the value + laddr(lslot[tmp]); op(O_ST); + live0 = live; + s0 = snap_push(); + loop_push(-1, s0); + expect(s_lb); + while (!is_op(s_rb)) { + if (tk == T_EOF) { die("missing }"); break; } + if (is_id(s_case)) { + next(); + bj = 0; + while (1) { + v = cexpr(); + laddr(lslot[tmp]); op(O_LD); lit(v); op(O_SUB); + bj = fixnew(hole(), bj); op(O_JZ); + if (!accept(s_comma)) break; + } + expect(s_colon); + nxt = hole(); op(O_JMP); + fixall(bj, here); + casebody(s0, live0); + setlit(nxt, here); + } else if (is_id(s_default)) { + next(); expect(s_colon); + if (defent >= 0) serr("a switch has one default"); + skip = hole(); op(O_JMP); // only reached when nothing matched + defent = here; + casebody(s0, live0); + setlit(skip, here); + } else die("expected case or default"); + } + next(); + snap_restore(s0); live = live0; + if (defent >= 0) { jmp(defent); loop_end(-1, -1); } + else { a = hole(); op(O_JMP); loop_end(a, s0); } // nothing matched: out with the states from the top + snap_pop(s0); + block_close(save); + fdepth--; + return 0; +} +fn kw_return() +{ + let c; + stmtonly(); + next(); + if (!infn) serr("return outside a function"); + if (bxdepth) serr("return can't leave a block expression"); + if (is_op(s_semi)) { lit(0); c = C_PLAIN; } + else { + c = expr(); rval(); + if (fty == tvoid) tcerr("return value", tvoid, xty); + else if (fty && !compat(fty, xty)) tcerr("return value", fty, xty); + } + expect(s_semi); + retcheck(c); + exit_vars(1, 0); + epichain = fixnew(here, epichain); + lit(MW - 1); op(O_JMP); + live = 0; + return 0; +} +fn kw_unchecked() +{ + let old = fchk; + stmtonly(); + next(); + fchk = 0; + substmt(); + fchk = old; + return 0; +} + +// ---- asm blocks: raw vm words, one per token ---- +// numbers push, -n pushes a negative number, #n emits a raw word, one-char +// tokens from the mnemonic table are opcodes, names push the address of a +// variable or function, :name defines a label for this block. +fn asmblock() +{ + let nl = 0, nr = 0, i, k, c, s, ln = R_ASM, la = R_ASM + 64, rn = R_ASM + 128, ra = R_ASM + 384; + expect(s_lb); + while (!is_op(s_rb)) { + if (tk == T_EOF) die("unterminated asm block"); + if (tk == T_NUM) { lit(tv); next(); continue; } + if (tk == T_STR) { strlit(tv); next(); continue; } + if (is_op(s_hash)) { next(); if (tk != T_NUM) die("# needs a number"); emit(tv); next(); continue; } + if (is_op(s_minus)) { next(); if (tk == T_NUM && tglue) { lit(0 - tv); next(); } else op(O_SUB); continue; } + if (is_op(s_colon)) { next(); if (nl >= 64) fatal("too many asm labels"); ln[nl] = name(); la[nl] = here; nl++; continue; } + if (tv[0] && !tv[1] && (c = mnemidx(tv[0])) >= 0) { op(c); next(); continue; } + if (tk != T_ID) die("bad token in asm"); + i = findloc(tv); + if (i >= 0) { lused[i] = 1; laddr(lslot[i]); next(); continue; } + k = -1; i = 0; + while (i < nl) { if (ln[i] == tv) k = i; i++; } + if (k >= 0) { lit(la[k]); next(); continue; } + s = lookup(tv); + if (s >= 0 && skind[s] == K_ENH) { op(XOP0 + sval[s]); next(); continue; } + if (s >= 0 && skind[s] != K_FN && skind[s] != K_KW) { lit(sval[s]); next(); continue; } + if (s >= 0 && skind[s] == K_FN) { fnref(s); next(); continue; } + if (nr >= 250) fatal("too many asm label references"); + rn[nr] = tv; ra[nr] = hole(); nr++; next(); // a label further down this block + } + next(); + i = 0; + while (i < nr) { + k = -1; c = 0; + while (c < nl) { if (ln[c] == rn[i]) k = c; c++; } + if (k < 0) serrn(rn[i], "unknown name in asm"); + setlit(ra[i], la[k]); + i++; + } + return 0; +} +fn kw_asm() { next(); asmblock(); xcls = C_PLAIN; return 0; } + +// ---- quote: code as data ---- +// a token list is an object: [n] then kind, value, line for each token. names, +// strings and operators are interned pointers. +fn grab() -> .u +{ + let n = 0, depth = 0, q; + if (!is_op(s_lb)) die("expected {"); + next(); + while (1) { + if (tk == T_EOF) die("unterminated quote"); + if (is_op(s_lb)) depth++; + if (is_op(s_rb)) { if (!depth) break; depth--; } + if (n >= 2700) fatal("quote too long"); + R_QBUF[n * 3] = tk; R_QBUF[n * 3 + 1] = tv; R_QBUF[n * 3 + 2] = tline; n++; + next(); + } + next(); + q = new(n * 3 + 1); + q[0] = n; + wcopy(q + 1, R_QBUF, n * 3); + return u*q; +} +unchecked fn qnew(src) -> .u { let n = src[0] * 3 + 1, q = new(n); wcopy(q, src, n); return q; } +fn qdup(.m q) -> .u { let n = q[0] * 3 + 1, r = new(n); wcopy(r, q, n); return u*r; } +fn wcopyto(.m q, n) { let i = 0; while (i < n) { emit(q[i]); i++; } return 0; } +fn kw_quote() +{ + let q, j, a, n; + next(); + q = grab(); + j = hole(); op(O_JMP); + a = here; + n = q[0] * 3 + 1; + wcopyto(q, n); + free(d*q); + setlit(j, here); + lit(a); + callname(s_qnew); + xcls = C_FRESH; + return 0; +} +// tokens of a string, as a token list +fn tokens(s) -> .u +{ + let n = 0, q; + src_push(2, s, 0, 0); + next(); + while (tk != T_EOF) { + if (n >= 2700) fatal("too many tokens"); + R_QBUF[n * 3] = tk; R_QBUF[n * 3 + 1] = tv; R_QBUF[n * 3 + 2] = tline; n++; + next(); + } + src_pop(); + q = new(n * 3 + 1); + q[0] = n; + wcopy(q + 1, R_QBUF, n * 3); + return u*q; +} +// feed a token list into the input: the parser reads it next, then carries +// on where it was. the list is freed once it has been read. +unchecked fn feed(.d q) { src_push(1, q, 1, 1); next(); return 0; } + +// ---- reading a unit and scanning it for signatures ---- +// a unit from stdin is read whole first (4 chars to a word), then scanned +// once for fn signatures, so a function's tags are known before its first +// call even when it's defined further down. +unchecked fn readsrc() +{ + let cap = 4090, b = new(4090), n = 0, c, nb, w; + while (!eof()) { + c = getc() & 255; + if (c == 96) break; + if ((n >> 2) + 2 >= cap) { nb = new(cap + cap + 6); wcopy(nb, b, cap); free(b); b = nb; cap = cap + cap + 6; } + w = n >> 2; + if (!(n & 3)) b[w] = 0; + b[w] = b[w] | (c << ((n & 3) * 8)); + n++; + } + b[n >> 2] = b[n >> 2] & (pow2[(n & 3) * 8] - 1); // the 0 at the end + if ((n & 3) == 0) b[n >> 2] = 0; + b[(n >> 2) + 1] = 0; + return b; +} +// skip a <...> group (type parameters) +fn pskipangle() +{ + let dp = 0; + while (tk != T_EOF) { + if (is_op(s_lt)) dp++; + else if (is_op(s_gt)) dp--; + else if (is_op(s_shr)) dp = dp - 2; + next(); + if (dp <= 0) return 0; + } + return 0; +} +// at the name after fn: read the signature's tags and record them +fn prescan_fn() +{ + let n = tv, np = 0, tags = 0, t = 0, ret = 0, dp, s; + if (tk != T_ID) return 0; + next(); + if (is_op(s_lt)) pskipangle(); + if (!is_op(s_lp)) return 0; + next(); + if (!is_op(s_rp)) { + np = 1; dp = 1; + while (tk != T_EOF) { + if (is_op(s_lp)) dp++; + else if (is_op(s_rp)) { dp--; if (!dp) break; } + else if (dp == 1 && is_op(s_comma)) { if (np <= 12) tags = tags + (t << (np + np - 2)); np++; t = 0; } + else if (dp == 1 && is_op(s_dot)) { next(); if (is_id(s_m)) t = 1; else if (is_id(s_d)) t = 2; continue; } + next(); + } + if (np <= 12) tags = tags + (t << (np + np - 2)); + } + next(); + if (accept(s_arrow)) { + while (tk != T_EOF && !is_op(s_lb) && !is_op(s_semi)) { + if (is_op(s_dot)) { next(); if (is_id(s_u)) ret = 1; continue; } + next(); + } + } + s = lookup(n); + if (s < 0 || s < ubase || skind[s] != K_FN || (sflag[s] & F_DEF)) s = fwdfn(n); + if (!(sflag[s] & F_SIG)) { ssig[s] = ret + np + np + tags * 64; sflag[s] = sflag[s] | F_SIG | F_PRE; } + return 0; +} +fn prescan(kind, obj) +{ + let depth = 0, sp = panic, sn = nerr; + dry++; + src_push(kind, obj, 0, 0); + next(); + while (tk != T_EOF) { + if (is_op(s_lb)) depth++; + else if (is_op(s_rb)) depth--; + else if (depth <= 0 && is_id(s_fn)) { next(); prescan_fn(); continue; } + next(); + } + src_pop(); + dry--; panic = sp; nerr = sn; + return 0; +} +// compile a whole unit from stdin +fn unit_stdin() +{ + let src = readsrc(); + prescan(3, src); + src_push(3, src, 0, 0); + next(); + unit_top(); + unit_end(ubase); + src_pop(); + free(src); + return 0; +} + +// ---- nested compiles ---- +// compile() and load() can run while another function is being compiled (from +// a parse word). the function context is saved, the new code is emitted in +// place with a jump around it, and then everything is put back. +fn ctx_push() +{ + let f = R_CTX + ctxsp * 24; + if (ctxsp >= 30) fatal("compiles nested too deeply"); + ctxsp++; + f[0] = nloc; f[1] = lbase; f[2] = fdepth; f[3] = fslots; f[4] = fchk; f[5] = fret; f[6] = live; + f[7] = bxdepth; f[8] = epichain; f[9] = fframe; f[10] = infn; f[11] = lsp; f[12] = loopbase; + f[13] = lv; f[14] = lvk; f[15] = lvi; f[16] = kw_expr; f[17] = xcls; f[18] = snaptop; + f[19] = fty; f[20] = ntp; f[21] = xty; wcopy(f + 22, tpn, 2); + return 0; +} +fn ctx_pop() +{ + let f; + ctxsp--; + f = R_CTX + ctxsp * 24; + nloc = f[0]; lbase = f[1]; fdepth = f[2]; fslots = f[3]; fchk = f[4]; fret = f[5]; live = f[6]; + bxdepth = f[7]; epichain = f[8]; fframe = f[9]; infn = f[10]; lsp = f[11]; loopbase = f[12]; + lv = f[13]; lvk = f[14]; lvi = f[15]; kw_expr = f[16]; xcls = f[17]; snaptop = f[18]; + fty = f[19]; ntp = f[20]; xty = f[21]; wcopy(tpn, f + 22, 2); + return 0; +} + +// ---- functions and the top level ---- +fn fn_begin(chk, ret) +{ + lbase = nloc; fdepth = 1; fslots = 0; fchk = chk && check_on; fret = ret; fty = 0; + live = 1; bxdepth = 0; epichain = 0; infn = 1; loopbase = lsp; lv = 0; lvk = LV_NONE; + return 0; +} +// fp -= frame size, then check the frame stack: past FBOT, reset fp and stop +fn prologue() +{ + let k; + lit(ZFP); op(O_LD); fframe = hole(); op(O_SUB); lit(ZFP); op(O_ST); + lit(ZFP); op(O_LD); lit(FBOT); op(O_LT); k = hole(); op(O_JZ); + lit(FTOP); lit(ZFP); op(O_ST); callname(s_fovf); + setlit(k, here); + return 0; +} +fn epilogue() +{ + lit(0); + fixall(epichain, here); + lit(ZFP); op(O_LD); lit(fslots); op(O_ADD); lit(ZFP); op(O_ST); op(O_RET); + setlit(fframe, fslots); + infn = 0; + nloc = lbase; + return 0; +} +fn body() +{ + expect(s_lb); + while (!is_op(s_rb)) { if (tk == T_EOF) die("missing } at the end of a function"); stmt(); } + unused_check(lbase); + if (fchk && live) { exit_vars(1, 1); if (fret) cerr("a -> .u function has to end with a return"); } + next(); + return 0; +} +// fn name<T>(int a, *node.m n, .d x, b) -> *list<T>.u { ... } +fn fndef(chk) +{ + let n, s, np = 0, tags = 0, t, ret = 0, sig, i, pn[24], pt[24], pty[24], rty = 0, typed = 0, ft; + next(); + n = name(); + ntp = 0; + if (is_op(s_lt)) { // type parameters + next(); + while (1) { if (ntp >= 8) die("too many type parameters"); tpn[ntp] = name(); ntp++; if (!accept(s_comma)) break; } + closeangle(); + typed = 1; + } + s = lookup(n); + if (s < 0 || s < ubase || skind[s] != K_FN || (sflag[s] & F_DEF)) { s = addsym(n, K_FN, 0); sflag[s] = 0; } + expect(s_lp); + if (!is_op(s_rp)) { + while (1) { + t = 0; + pty[np] = tword; + if (!is_op(s_dot) && ((tk == T_OP && (tv == s_star || tv == s_lk)) || (tk == T_ID && (tv == s_fn || istypename(tv))))) { + pty[np] = ptype(); typed = 1; + if (isagg(pty[np])) serr("structs and arrays are passed by pointer"); + } + if (accept(s_dot)) { // .m lends, .d takes over + if (tv == s_m) t = 1; else if (tv == s_d) t = 2; else die("a parameter tag is .m or .d"); + next(); + } + pn[np] = name(); + if (np >= 12 && t) die("only the first 12 parameters can be tagged"); + pt[np] = t; + if (np < 12) tags = tags + (t << (np + np)); + np++; + if (np >= 24) die("too many parameters"); + if (!accept(s_comma)) break; + } + } + expect(s_rp); + if (accept(s_arrow)) { + if (!is_op(s_dot)) { rty = ptype(); typed = 1; } + if (accept(s_dot)) { if (!is_id(s_u)) die("the return tag is .u"); next(); ret = 1; } + } + sig = ret + np + np + tags * 64; + if ((sflag[s] & F_SIG) && !(sflag[s] & F_PRE)) { if (ssig[s] != sig) serrn(n, "doesn't match its earlier declaration"); } + else if ((sflag[s] & F_UNTYPED) && (tags || ret)) serrn(n, "was called before its tags were declared; put a prototype above the first call"); + ssig[s] = sig; + sflag[s] = (sflag[s] | F_SIG) - (sflag[s] & F_PRE); + if (ntp) sflag[s] = sflag[s] | F_GEN; + if (typed) { + if (!rty) rty = tword; + ft = tfn(rty, np, pty, sig); + if (stype[s] && stype[s] != ft) serrn(n, "doesn't match its earlier declaration"); + stype[s] = ft; + } + if (accept(s_semi)) { ntp = 0; return s; } + if (sflag[s] & F_DEF) serrn(n, "is defined twice"); + sval[s] = here; + sflag[s] = sflag[s] | F_DEF; + fixall(sfix[s], here); sfix[s] = 0; + fn_begin(chk, ret); + if (typed) fty = rty; + prologue(); + i = 0; + while (i < np) { + t = newloc(pn[i]); fslots++; + ltype[t] = pty[i]; + if (fchk && pt[i] == 1) lstate[t] = S_BORROW; + if (fchk && pt[i] == 2) { lstate[t] = S_M; lown[t] = 1; } + i++; + } + i = np; + while (i) { i--; laddr(i); op(O_ST); } + body(); + epilogue(); + ntp = 0; + return s; +} +// let g = 1, buf[10]; int g = 3; [10]int buf; point origin; +fn gdecl(typed) +{ + let n, a, k, t = 0, s; + if (typed) { dtype(); t = dty; if (dtag) serr("tags go on locals and parameters, not globals"); if (t == tvoid) serr("a variable can't be void"); } + else next(); + while (1) { + n = name(); + if (accept(s_lk)) { + k = cexpr(); expect(s_rk); + s = addsym(n, K_GARR, here); + if (!t) t = tword; + stype[s] = tarr(t, k); + k = tsz(stype[s]); + while (k) { emit(0); k--; } + } else if (t && isagg(t)) { + s = addsym(n, K_GARR, here); stype[s] = t; + k = tsz(t); + while (k) { emit(0); k--; } + } else { + a = here; + s = addsym(n, K_GVAR, a); + if (t) stype[s] = t; + emit(0); + if (accept(s_eq)) { + if (tk == T_STR) { *(a + hdelta) = here; k = tv; while (*k) { emit(*k); k++; } emit(0); next(); } + else *(a + hdelta) = cexpr(); + } + } + if (!accept(s_comma)) break; + } + expect(s_semi); + return 0; +} +fn gconst() +{ + let n; + next(); + while (1) { n = name(); expect(s_eq); addsym(n, K_CONST, cexpr()); if (!accept(s_comma)) break; } + expect(s_semi); + return 0; +} +// enhance name(a, b) [-> n] micro "recipe" [else fn]; +// enhance name(a, b) [-> n] vm { asm } [else fn]; +// lays out the descriptor the install opcode reads, and wires it in right +// away (an image wires its own in when it boots). +fn enhdef() +{ + let n, np = 0, nout = 1, dsc, na, pay, len = 0, kind = 0, i, p, s, slot, r, pn[8]; + next(); + n = name(); + expect(s_lp); + if (!is_op(s_rp)) { while (1) { if (np == 8) die("an enhancement takes at most 8 inputs"); pn[np] = name(); np++; if (!accept(s_comma)) break; } } + expect(s_rp); + if (accept(s_arrow)) { if (tk != T_NUM || tv < 0 || tv > 8) die("expected the number of outputs"); nout = tv; next(); } + dsc = here; i = 0; while (i < 7) { emit(0); i++; } + na = here; i = 0; while (i < np) { emit(0); i++; } + i = 0; + while (i < np) { *(na + i + hdelta) = here; p = pn[i]; while (*p) { emit(*p); p++; } emit(0); i++; } + if (accept(s_micro)) { + pay = here; + if (tk != T_STR) die("expected the recipe as a string"); + while (tk == T_STR) { p = tv; while (*p) { emit(*p); p++; } next(); } + emit(0); + } else if (accept(s_vm)) { pay = here; asmblock(); op(O_RET); len = here - pay; kind = 1; } + else die("expected micro or vm"); + slot = nenh; + if (slot >= 512) fatal("all 512 enhancement slots are taken"); + nenh++; + p = dsc + hdelta; + p[0] = slot; p[1] = kind; p[2] = np; p[3] = nout; p[4] = pay; p[5] = len; p[6] = na; + s = addsym(n, K_ENH, slot); + ssig[s] = np + nout * 16 + dsc * 256; + if (accept(s_else)) { r = lookup(tv); if (r < 0) r = fwdfn(tv); next(); sfix[s] = r + 1; } + expect(s_semi); + if (!hdelta) { + r = install(dsc); + if (r < 0) { + if (!sfix[s]) serrn(n, "the enhancement didn't install (see install's codes in core.c)"); + sflag[s] = sflag[s] | F_FALLBACK; + } + } + return 0; +} +// after a syntax error at the top level: skip to something that starts a declaration +fn topsync() +{ + let depth = 0; + while (tk != T_EOF) { + if (is_op(s_lb)) depth++; + else if (is_op(s_rb)) { depth--; if (depth < 0) depth = 0; } + else if (!depth && tk == T_ID && (tv == s_fn || tv == s_var || tv == s_const || tv == s_struct || tv == s_type + || tv == s_enhance || tv == s_typecheck || tv == s_unchecked || tv == s_immediate)) break; + next(); + } + panic = 0; + return 0; +} +fn topitem() +{ + let chk = 1, imm = 0, s, k; + if (panic) return topsync(); + if (is_id(s_unchecked)) { + next(); + if (accept(s_semi)) { check_on = 0; return 0; } // unchecked; switches the analyzer off from here on + chk = 0; + } + if (is_id(s_immediate)) { next(); imm = 1; } + if (is_id(s_fn)) { + s = fndef(chk); + if (imm) { + if (hdelta) serr("parse words can't be defined while building an image"); + k = addsym(sname[s], K_KW, sval[s]); + sflag[k] = F_DEF | F_STMT | F_EXPR | F_TOP; + } + return 0; + } + if (!chk || imm) { die("expected fn"); return 0; } + if (is_id(s_var)) return gdecl(0); + if (is_id(s_struct)) { kw_struct(); return 0; } + if (is_id(s_type)) return kw_type(); + if (is_id(s_typecheck)) return kw_typecheck(); + if (is_id(s_enhance)) return enhdef(); + if (is_id(s_const)) return gconst(); + if (tk == T_ID) { + s = lookup(tv); + if (s >= 0 && skind[s] == K_KW && (sflag[s] & F_TOP)) { kwcall(s, 0); return 0; } + } + if (tstart()) return gdecl(1); + die("expected fn, let, const, struct, type or a declaration"); + return 0; +} +fn unit_top() { while (tk != T_EOF) topitem(); return 0; } +fn unit_end(from) +{ + let i = from; + panic = 0; + while (i < nsym) { + if (skind[i] == K_FN && !(sflag[i] & F_DEF) && sfix[i]) { tline = sval[i]; serrn(sname[i], "is used but never defined"); } + i++; + } + if (nerr) giveup(); + return 0; +} + +// compile a token list as the body of a new function, or (top) as top-level +// declarations. returns the entry address of the function. +unchecked fn compile_list(q, top) +{ + let hop = -1, entry = 0, from = nsym; + ctx_push(); + if (infn) { hop = hole(); op(O_JMP); } + if (top) prescan(1, q); + src_push(1, q, 0, 0); + next(); + if (top) unit_top(); + else { + entry = here; + fn_begin(1, 0); + prologue(); + while (tk != T_EOF) stmt(); + unused_check(lbase); + if (fchk && live) exit_vars(1, 1); + epilogue(); + } + src_pop(); + unit_end(from); + if (hop >= 0) setlit(hop, here); + if (here > ptop) ptop = here; + ctx_pop(); + return entry; +} +fn compile(q) { return compile_list(q, 0); } +fn load(q) { compile_list(q, 1); return 0; } +fn eval(q) { let f = compile_list(q, 0); return f(); } +fn evals(s) { let q = tokens(s), r = eval(q); free(d*q); return r; } + +// ---- symbol blob ---- +// an image ends with its own symbol table so the genome knows its names: +// kind, value, signature, name chars, 0 ... then a 0 kind. +fn blob_emit(from) +{ + let i = from, p; + while (i < nsym) { + if ((skind[i] >= K_FN && skind[i] <= K_CONST) || skind[i] == K_ENH) { + emit(skind[i]); emit(sval[i]); emit(ssig[i]); + p = sname[i]; + while (*p) { emit(*p); p++; } + emit(0); + } + i++; + } + emit(0); + return 0; +} +fn blob_load(p) +{ + let s, n; + while (*p) { + n = intern(p + 3); + s = addsym(n, p[0], p[1]); + ssig[s] = p[2]; + sflag[s] = F_DEF | F_SIG; + if (p[0] == K_ENH) { if (p[1] >= nenh) nenh = p[1] + 1; install(p[2] >> 8); } + p = p + 3 + strlen(p + 3) + 1; + } + return 0; +} +// compile the rest of stdin as a genome image that will run at org, writing +// it to dst. only the source's own names are visible while it compiles. +// returns the image length in words. +fn build_image(org, dst) +{ + let sh = here, sd = hdelta, sl = hlimit, sub = ubase, svl = vislo, sns = nsym, sen = nenh, n; + ubase = nsym; vislo = nsym; nenh = 0; + here = org; hdelta = dst - org; hlimit = org + 0x30000; + lit(FTOP); lit(ZFP); op(O_ST); fnref(fwdfn(s_start)); op(O_JMP); emit(0); emit(0); + unit_stdin(); + *(dst + 5) = here; + blob_emit(ubase); + *(dst + 6) = here; + n = here - org; + popsyms(sns); + ubase = sub; vislo = svl; here = sh; hdelta = sd; hlimit = sl; nenh = sen; + return n; +} + +// ---- setup ---- +fn gc_roots() +{ + markrange(0, ptop); + if (!hdelta && here > ptop) markrange(ptop, here); // globals of a unit still compiling + markrange(GORG, GORG[6]); + markrange(R_SRC, R_SRC + 1024); + markrange(*ZFP, FTOP); + markrange(DSLO, DSHI); + return 0; +} +fn zc_init() +{ + pooltop = R_POOL; + wfill(R_IH, 0, IHN); + nsym = 0; nfixu = 0; fixfree = 0; nbin = 0; nun = 0; nasg = 0; nkop = 0; seqtop = 0; + lsp = 0; ctxsp = 0; snaptop = 0; nloc = 0; lbase = 0; + s_lp = intern("("); s_rp = intern(")"); s_lb = intern("{"); s_rb = intern("}"); + s_lk = intern("["); s_rk = intern("]"); s_comma = intern(","); s_semi = intern(";"); + s_eq = intern("="); s_arrow = intern("->"); s_inc = intern("++"); s_dec = intern("--"); + s_star = intern("*"); s_amp = intern("&"); s_colon = intern(":"); s_hash = intern("#"); + s_minus = intern("-"); s_dot = intern("."); s_q = intern("?"); s_case = intern("case"); s_default = intern("default"); s_andand = intern("&&"); s_oror = intern("||"); + s_fn = intern("fn"); s_var = intern("let"); s_const = intern("const"); s_else = intern("else"); + s_as = intern("as"); s_unchecked = intern("unchecked"); s_immediate = intern("immediate"); + s_m = intern("m"); s_d = intern("d"); s_u = intern("u"); + s_start = intern("start"); s_main = intern("main"); s_free = intern("free"); + s_own = intern("own"); s_qnew = intern("qnew"); s_fovf = intern("frame_overflow"); + s_enhance = intern("enhance"); s_micro = intern("micro"); s_vm = intern("vm"); + return 0; +} +// operators and keywords. these are plain tables in memory: a program can +// add operators, change precedences or replace keywords with its own. +fn zc_tables() +{ + add_unop("!", "0 $ 0 ~ - | 31 > 1 +"); // index 0 + add_unop("!!", "0 $ 0 ~ - | 31 > 0 ~ -"); // index 1: normalize to 0 or 1 + add_unop("~", "-1 ^"); // index 2 + add_unop("-", "0 ~ -"); + add_binop("*", 10, "*"); // the first few indexes are used by cexpr + add_binop("/", 10, "idiv"); // native divide, wired in by rt.zc + add_binop("%", 10, "imod"); + add_binop("+", 9, "+"); // 3 + add_binop("-", 9, "-"); // 4 + add_binop("<<", 8, "ishl"); // 5 + add_binop(">>", 8, ">"); // 6 + add_binop("<", 7, "<"); + add_binop(">", 7, "~ <"); + add_binop("<=", 7, "~ < 1 ^"); + add_binop(">=", 7, "< 1 ^"); + add_binop("==", 6, "- 0 $ 0 ~ - | 31 > 1 +"); + add_binop("!=", 6, "- 0 $ 0 ~ - | 31 > 0 ~ -"); + add_binop("&", 5, "&"); // 13 + add_binop("^", 4, "^"); + add_binop("|", 3, "|"); // 15 + add_binop("&&", -1, 0); + add_binop("||", -2, 0); + addkop(s_eq); addkop(s_arrow); addkop(s_inc); addkop(s_dec); + add_kw("let", kw_var, F_STMT); + add_kw("if", kw_if, F_STMT); + add_kw("while", kw_while, F_STMT); + add_kw("for", kw_for, F_STMT); + add_kw("switch", kw_switch, F_STMT); + add_kw("break", kw_break, F_STMT); + add_kw("continue", kw_continue, F_STMT); + add_kw("return", kw_return, F_STMT); + add_kw("unchecked", kw_unchecked, F_STMT); + add_kw("asm", kw_asm, F_STMT | F_EXPR); + add_kw("quote", kw_quote, F_EXPR); + add_kw("sizeof", kw_sizeof, F_EXPR); + add_kw("typeof", kw_typeof, F_EXPR); + return 0; +} +fn start() +{ + let s, f; + heap_init(); + zc_init(); + blob_load(GORG[5]); + zc_tables(); + types_init(); + here = 0; hdelta = 0; hlimit = HEAP_LO; ptop = 0; + ubase = nsym; vislo = 0; + unit_stdin(); + if (here > ptop) ptop = here; + s = lookup(s_main); + if (s >= 0 && skind[s] == K_FN && (sflag[s] & F_DEF)) { f = sval[s]; f(); } + halt(); + return 0; +} |