diff --git a/Makefile b/Makefile index 7862f865..2faa3e42 100644 --- a/Makefile +++ b/Makefile @@ -449,6 +449,7 @@ TESTS = $(BIN)/test_smoke $(BIN)/test_lex $(BIN)/test_parse $(BIN)/test_check \ $(BIN)/test_hex_run $(BIN)/test_utf8_run $(BIN)/test_bytes_run \ $(BIN)/test_path_run \ $(BIN)/test_letshadow_run \ + $(BIN)/test_strglobeq_run \ $(BIN)/test_decimal_run $(BIN)/test_strconv_int_run \ $(BIN)/test_stof_run $(BIN)/test_ftos_run \ $(BIN)/test_memio_run $(BIN)/test_temp_run $(BIN)/test_getopt_run \ @@ -2002,6 +2003,10 @@ $(BIN)/test_letshadow_run: test/wcc/989_letshadow_run.c $(BIN)/ww $(BIN)/w6c \ $(BIN)/w6a $(BIN)/w6l $(LIB)/libwwrt.a | $(BIN) $(CC) $(CFLAGS) -o $@ $< +$(BIN)/test_strglobeq_run: test/wcc/989_strglobeq_run.c $(BIN)/ww $(BIN)/w6c \ + $(BIN)/w6a $(BIN)/w6l $(LIB)/libwwrt.a | $(BIN) + $(CC) $(CFLAGS) -o $@ $< + $(BIN)/test_ascii_run: test/wcc/904_ascii_run.c $(BIN)/ww $(BIN)/w6c \ $(BIN)/w6a $(BIN)/w6l $(LIB)/libwwrt.a | $(BIN) $(CC) $(CFLAGS) -o $@ $< diff --git a/cmd/w6c/cgen.c b/cmd/w6c/cgen.c index db8cec2a..7561f60f 100644 --- a/cmd/w6c/cgen.c +++ b/cmd/w6c/cgen.c @@ -4568,10 +4568,23 @@ cgexpr(Cg *c, Node *n, Local *locals) /* Push rhs (len, then ptr top) */ if (n->rhs->kind == N_IDENT) { int off = localfind(locals, n->rhs->str); - ins2(c, A_MOVQ, amem(D_BP, off + 8), areg(D_AX)); - ins1(c, A_PUSHQ, areg(D_AX)); - ins2(c, A_MOVQ, amem(D_BP, off), areg(D_AX)); - ins1(c, A_PUSHQ, areg(D_AX)); + /* #154: localfind→0 for a module global, but the + * str header lives at name(SB), not BP+0. Mirror + * #148's slice global branch (cgen.c:9082): LEAQ + * the symbol into a base reg, push len then ptr. */ + if (off == 0 && let_islet(n->rhs->str)) { + ins2(c, A_LEAQ, masym(c, n->rhs->str), + areg(D_BX)); + ins2(c, A_MOVQ, amem(D_BX, 8), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + ins2(c, A_MOVQ, amem(D_BX, 0), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + } else { + ins2(c, A_MOVQ, amem(D_BP, off + 8), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + ins2(c, A_MOVQ, amem(D_BP, off), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + } } else { cgexpr(c, n->rhs, locals); /* AX=ptr, BX=len */ ins1(c, A_PUSHQ, areg(D_BX)); @@ -4580,10 +4593,21 @@ cgexpr(Cg *c, Node *n, Local *locals) /* Push lhs */ if (n->lhs->kind == N_IDENT) { int off = localfind(locals, n->lhs->str); - ins2(c, A_MOVQ, amem(D_BP, off + 8), areg(D_AX)); - ins1(c, A_PUSHQ, areg(D_AX)); - ins2(c, A_MOVQ, amem(D_BP, off), areg(D_AX)); - ins1(c, A_PUSHQ, areg(D_AX)); + /* #154: see the rhs branch above — a module-global + * str ident lives at name(SB), not BP+0. */ + if (off == 0 && let_islet(n->lhs->str)) { + ins2(c, A_LEAQ, masym(c, n->lhs->str), + areg(D_BX)); + ins2(c, A_MOVQ, amem(D_BX, 8), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + ins2(c, A_MOVQ, amem(D_BX, 0), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + } else { + ins2(c, A_MOVQ, amem(D_BP, off + 8), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + ins2(c, A_MOVQ, amem(D_BP, off), areg(D_AX)); + ins1(c, A_PUSHQ, areg(D_AX)); + } } else { cgexpr(c, n->lhs, locals); ins1(c, A_PUSHQ, areg(D_BX)); diff --git a/test/wcc/989_strglobeq.ww b/test/wcc/989_strglobeq.ww new file mode 100644 index 00000000..b5c0a4ab --- /dev/null +++ b/test/wcc/989_strglobeq.ww @@ -0,0 +1,67 @@ +// strglobeq — #154 regression pin. `str == str` / `str != str` delegate to +// rt_streq, and cgen's str-compare arm had an N_IDENT fast-path that ALWAYS +// read the header off BP+localfind(name). For a module-GLOBAL str ident +// localfind→0, so it loaded saved-BP/retaddr garbage instead of name(SB) — +// a silent cstage miscompile (the header lives at name(SB), not the frame). +// Fix mirrors #148's slice global branch: LEAQ name(SB), load ptr/len off it. +// +// Table-driven: each row is {input, expected}; the loop feeds every input +// through the five compile-time-distinct comparison shapes (the bug is per +// SHAPE in cgen, so the shapes are separate fns the rows drive). Shapes cover +// const-global AND let-global operands, ident on RHS AND on LHS, == AND !=, +// and a length>1 global ("/usr") so the LEN word — not just the ptr — is read +// off name(SB) on both sub-sites. signalled = row*10+shape pinpoints failures. +// +// Run with `out/bin/ww run test/wcc/989_strglobeq.ww`; exit 0 = all pass. + +package main; + +import os; + +const csep: str = "/"; +let lsep: str = "/"; +const longsep: str = "/usr"; + +// rhs-ident global (p == g): const, let, len>1, and !=. +fn eqr_const(p: str) bool = { return p == csep; }; +fn eqr_let(p: str) bool = { return p == lsep; }; +fn eqr_long(p: str) bool = { return p == longsep; }; +fn ner_const(p: str) bool = { return p != csep; }; + +// lhs-ident global (g == p): const and len>1. +fn eql_const(p: str) bool = { return csep == p; }; +fn eql_long(p: str) bool = { return longsep == p; }; + +type row = struct { + in: str, + eqsep: bool, // in == "/" + eqlong: bool, // in == "/usr" +}; + +let signalled: i32 = 0; +fn fail() void = { os.exit(signalled + 10); }; + +export fn main() i32 = { + let rows: [_]row = [ + row { in = "/", eqsep = true, eqlong = false }, + row { in = "foo", eqsep = false, eqlong = false }, + row { in = "/usr", eqsep = false, eqlong = true }, + ]; + // len() stamps i32 on cstage (#26); the index must match for the bound. + for (let i: i32 = 0; i < len(rows); i += 1) { + let r = rows[i]; + signalled = i * 10 + 1; + if (eqr_const(r.in) != r.eqsep) { fail(); }; + signalled = i * 10 + 2; + if (eqr_let(r.in) != r.eqsep) { fail(); }; + signalled = i * 10 + 3; + if (eql_const(r.in) != r.eqsep) { fail(); }; + signalled = i * 10 + 4; + if (ner_const(r.in) != !r.eqsep) { fail(); }; + signalled = i * 10 + 5; + if (eqr_long(r.in) != r.eqlong) { fail(); }; + signalled = i * 10 + 6; + if (eql_long(r.in) != r.eqlong) { fail(); }; + }; + return 0; +}; diff --git a/test/wcc/989_strglobeq_run.c b/test/wcc/989_strglobeq_run.c new file mode 100644 index 00000000..44bc84c4 --- /dev/null +++ b/test/wcc/989_strglobeq_run.c @@ -0,0 +1,51 @@ +/* + * 989_strglobeq_run — #154 regression pin. Compile + run the strglobeq + * fixture under the C-side `ww run` driver (cstage w6c) and assert exit 0. + * + * The bug (str== fast-path read the global str header off BP+0 instead of + * name(SB)) is a CSTAGE silent miscompile; byte-id 990-997 can't see a + * runtime-value miscompile, so only this value check catches it. Same + * thin-wrapper shape as 989_letshadow_run. + */ +#include +#include +#include +#include + +static int +runwait(const char *cmd) +{ + int rc = system(cmd); + if (rc == -1) return -1; + if (WIFEXITED(rc)) return WEXITSTATUS(rc); + return 1; +} + +int +main(void) +{ + const char *bin = getenv("BIN"); + if (!bin) bin = "out/bin"; + char absbin[1024]; + if (bin[0] != '/') { + char cwd[1024]; + if (getcwd(cwd, sizeof cwd) == NULL) return 1; + snprintf(absbin, sizeof absbin, "%s/%s", cwd, bin); + bin = absbin; + } + char cwd[1024]; + if (getcwd(cwd, sizeof cwd) == NULL) return 1; + + const char *src = "test/wcc/989_strglobeq.ww"; + char path[1024], cmd[2048]; + snprintf(path, sizeof path, "%s/%s", cwd, src); + snprintf(cmd, sizeof cmd, "%s/ww run %s", bin, path); + int rc = runwait(cmd); + if (rc != 0) { + fprintf(stderr, "strglobeq_run FAIL: %s exited %d " + "(row %d miscompiled — #154)\n", src, rc, rc - 10); + return 1; + } + printf("strglobeq_run: %s ok\n", src); + return 0; +}