cgen: type-key wwstage struct-layout at 3 sites via stamped tinfo (#21)

wwstage cgen resolved struct LAYOUT by bare-leaf name (structlookup /
structparamsize) at three caller-side sites — the by-value arg push
(cgenutil), the let-receive copy width (cgenstmt), and the field-read
offset (cgenexpr). Under a cross-module same-leaf collision (two modules
each exporting a `pair`, 16B vs 24B) the name lookup first-matches the
WRONG type, so the push dropped the 2nd eightbyte, the receive over-copied,
and the field read the wrong offset. cstage type-keys off the stamped
tinfo and is correct; this aligns wwstage UP to it (ww-only change).

Route all three sites through the stamped node.type_ via a new
structabisizetn(*tinfo) accessor (push + receive) and tichase(type_).fields
(field-read, structlookupchain removed). One commit (rule-11 carve-out):
the collision drives all three at once and no per-site fixture isolates, so
it cannot bisect-split. A scoped slice of the #209/#211 name-keyed-cgen
cluster retirement; the cgdot *struct-ptr/global and let-copy siblings stay
name-keyed and are filed (#31).

New table-driven test 793_xmod_struct_argpush_collide_run (4 scenarios:
push/recv/field over 16B and 12B tails) reddens under a revert of the three
cgen files. Full make test green (336 passed); make sizelint clean.
This commit is contained in:
2026-06-29 11:45:08 +09:00
parent 7b9488706b
commit 5ae6e3419e
5 changed files with 435 additions and 69 deletions

View File

@@ -0,0 +1,301 @@
/*
* 793_xmod_struct_argpush_collide_run — project #21/#224 runtime + byte-id
* net. An inferred-let struct LOCAL whose type leaf collides with a
* DIFFERENT module's same-leaf struct miscompiled in wwstage at THREE
* caller-side register-ABI sites, all keyed off the name-resolved
* (collision-prone) struct instead of the checker-STAMPED tinfo:
*
* - push (cgenutil.ww pushargsrev): the by-value struct call-arg
* COUNT came from name-keyed structparamsize(lc.tnode). On the
* collision lc.tnode is a bare leaf structlookup mis-resolves to the
* foreign same-leaf struct (>16B → 0), so the fast path was skipped
* and the arg dropped to a scalar single-PUSHQ — the 2nd eightbyte
* (the sub-8 tail) was lost. [the original #21 site]
* - recv (cgenstmt.ww cgletbody): `let s = m1.mk()` sized the
* register receive via structlookup(tn.str) → the foreign struct's
* word count, OVER-copying a 24B foreign over a 16B local (spilling
* CX into a neighbour slot).
* - field (cgenexpr.ww cgdot direct-struct arm): `s.<field>` resolved
* the offset via structlookupchain(tn) → the foreign struct's layout,
* reading the field at the WRONG offset + wrong load-op. [#224, filed
* not-fixed by 784/#223]
*
* cstage is correct at all three — it keys on args[i]->type / lu->size /
* t->fields (the checker-STAMPED type), never a name re-lookup. Fix
* (wwstage-only align-UP, #21/#224, ONE commit): route all three off the
* stamped tinfo — structabisizetn(arg/tn.type_) for the push+recv COUNT,
* tichase(dotlhs.type_).fields (tfield) for the field offset+type.
*
* GATE-BLIND in the bootstrap (same class as 784/#223): the trigger needs
* a SECOND module contributing a same-leaf struct of a DIFFERENT size, so
* the name-keyed lookup mis-resolves. A single-module synthetic resolves
* correctly (structlookup's same-module pass) and stays byte-id — it does
* NOT redden. Hence this 2-module probe.
*
* row | m1.pair (local) | m2.pair (foreign) | site(s) | exit
* -----------------+----------------------+-------------------+---------------+-----
* recvpush_t16 | {u64,u16} 16B | 24B | recv + push | 107
* fieldread_t16 | {u64,u16} 16B | 24B | recv + field | 107
* combined_t16 | {u64,u16} 16B | 24B | recv+field+push| 114
* recvpush_t12 | {u32,u32,u32} 12B | 24B | recv + push | 6
*
* cstage `ww build --sep` + run pins runtime; the wwstage binary is run
* too (the bug WAS a wrong wwstage runtime value); raw cs.s vs ww.s cmp
* over the driver-produced per-package asm pins rule-10 byte-id (the
* discriminating net — pre-fix __root.s diverges at the push/recv/field).
*/
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <sys/stat.h>
#include <sys/wait.h>
static int
runwait(const char *cmd)
{
int rc = system(cmd);
if (rc == -1) return -1;
if (WIFEXITED(rc)) return WEXITSTATUS(rc);
return -1;
}
static int
slurp_eq(const char *a, const char *b)
{
FILE *fa = fopen(a, "rb");
FILE *fb = fopen(b, "rb");
if (!fa || !fb) { if (fa) fclose(fa); if (fb) fclose(fb); return -1; }
int rc = 0;
for (;;) {
int ca = fgetc(fa);
int cb = fgetc(fb);
if (ca != cb) { rc = -1; break; }
if (ca == EOF) break;
}
fclose(fa); fclose(fb);
return rc;
}
struct file { const char *name; const char *src; };
struct scenario {
const char *label;
const struct file *files; /* name==NULL terminates */
int want_exit;
};
/* m2.pair is a 24B struct sharing the leaf `pair` with m1.pair — the
* name-keyed lookup's any-module fallback mis-resolves m1.pair to it. */
static const char m2_24[] =
"package m2;\n"
"export type pair = struct { hi: u64, mid: u64, lo: u64 };\n"
"export fn use2(p: pair) i32 = { return (p.hi + p.mid + p.lo): i32; };\n";
/* ---- recvpush_t16: inferred-let recv + by-value push (the #21 site) -- */
static const struct file recvpush_t16_files[] = {
{ "m1.ww",
"package m1;\n"
"export type pair = struct { hi: u64, lo: u16 };\n"
"export fn mk() pair = { return pair { hi = 100: u64, lo = 7: u16 }; };\n"
"export fn consume(p: pair) i32 = { return (p.hi + (p.lo: u64)): i32; };\n" },
{ "m2.ww", m2_24 },
{ "main.ww",
"package main;\n"
"import m1;\n"
"import m2;\n"
"fn main() i32 = {\n"
" let dummy: m2.pair;\n"
" dummy.hi = 0: u64;\n"
" let s = m1.mk();\n"
" return m1.consume(s);\n"
"};\n" },
{ NULL, NULL }
};
/* ---- fieldread_t16: inferred-let recv + direct field read (#224) ----- */
static const struct file fieldread_t16_files[] = {
{ "m1.ww",
"package m1;\n"
"export type pair = struct { hi: u64, lo: u16 };\n"
"export fn mk() pair = { return pair { hi = 100: u64, lo = 7: u16 }; };\n" },
{ "m2.ww", m2_24 },
{ "main.ww",
"package main;\n"
"import m1;\n"
"import m2;\n"
"fn main() i32 = {\n"
" let dummy: m2.pair;\n"
" dummy.hi = 0: u64;\n"
" let s = m1.mk();\n"
" return (s.hi + (s.lo: u64)): i32;\n"
"};\n" },
{ NULL, NULL }
};
/* ---- combined_t16: recv + field + push in one fn -------------------- */
static const struct file combined_t16_files[] = {
{ "m1.ww",
"package m1;\n"
"export type pair = struct { hi: u64, lo: u16 };\n"
"export fn mk() pair = { return pair { hi = 100: u64, lo = 7: u16 }; };\n"
"export fn consume(p: pair) i32 = { return (p.hi + (p.lo: u64)): i32; };\n" },
{ "m2.ww", m2_24 },
{ "main.ww",
"package main;\n"
"import m1;\n"
"import m2;\n"
"fn main() i32 = {\n"
" let dummy: m2.pair;\n"
" dummy.hi = 0: u64;\n"
" let s = m1.mk();\n"
" let x: i32 = (s.lo: i32);\n"
" return m1.consume(s) + x;\n"
"};\n" },
{ NULL, NULL }
};
/* ---- recvpush_t12: a maxalign-4 12B sub-8-tail tail shape ----------- */
static const struct file recvpush_t12_files[] = {
{ "m1.ww",
"package m1;\n"
"export type pair = struct { a: u32, b: u32, c: u32 };\n"
"export fn mk() pair = { return pair { a = 1: u32, b = 2: u32, c = 3: u32 }; };\n"
"export fn consume(p: pair) i32 = { return (p.a + p.b + p.c): i32; };\n" },
{ "m2.ww", m2_24 },
{ "main.ww",
"package main;\n"
"import m1;\n"
"import m2;\n"
"fn main() i32 = {\n"
" let dummy: m2.pair;\n"
" dummy.hi = 0: u64;\n"
" let s = m1.mk();\n"
" return m1.consume(s);\n"
"};\n" },
{ NULL, NULL }
};
static const struct scenario scenarios[] = {
{ "recvpush_t16", recvpush_t16_files, 107 },
{ "fieldread_t16", fieldread_t16_files, 107 },
{ "combined_t16", combined_t16_files, 114 },
{ "recvpush_t12", recvpush_t12_files, 6 },
};
static int
run_scenario(const char *cdrv, const char *wdrv, const struct scenario *sc)
{
/* Private fixture dir — the driver's srcd-first import search can't
* pick up a polluting same-name file (mirrors 784/#215). */
char dir[] = "/tmp/ww793_XXXXXX";
if (mkdtemp(dir) == NULL) {
fprintf(stderr, "793[%s]: mkdtemp failed\n", sc->label);
return -1;
}
char path[1024], cmd[4096];
int rc = 0;
for (int i = 0; sc->files[i].name; i++) {
snprintf(path, sizeof path, "%s/%s", dir, sc->files[i].name);
FILE *f = fopen(path, "wb");
if (!f) { fprintf(stderr, "793[%s]: write %s\n", sc->label,
sc->files[i].name); rc = -1; goto done; }
fputs(sc->files[i].src, f);
fclose(f);
}
/* cstage --sep build + run pins runtime; pin WW_PKGCACHE under the
* scratch dir so out/.pkgcache is untouched. */
snprintf(cmd, sizeof cmd,
"cd %s && WW_PKGCACHE=%s/pkgc_c %s build --sep -I %s -o %s/main %s/main.ww",
dir, dir, cdrv, dir, dir, dir);
if (runwait(cmd) != 0) {
fprintf(stderr, "793[%s]: cstage build failed\n", sc->label);
rc = -1; goto done;
}
snprintf(path, sizeof path, "%s/main", dir);
int gotc = runwait(path);
if (gotc != sc->want_exit) {
fprintf(stderr, "793[%s]: cstage exit %d, want %d\n",
sc->label, gotc, sc->want_exit);
rc = -1;
}
/* wwstage --sep build + run: the bug WAS a wrong wwstage runtime
* value (a dropped/over-copied field), so run the wwstage binary too. */
snprintf(cmd, sizeof cmd,
"cd %s && WW_PKGCACHE=%s/pkgc_w %s build --sep -I %s -o %s/mainww %s/main.ww",
dir, dir, wdrv, dir, dir, dir);
if (runwait(cmd) != 0) {
fprintf(stderr, "793[%s]: ww_ww build failed\n", sc->label);
rc = -1; goto done;
}
snprintf(path, sizeof path, "%s/mainww", dir);
int gotw = runwait(path);
if (gotw != sc->want_exit) {
fprintf(stderr, "793[%s]: wwstage exit %d, want %d\n",
sc->label, gotw, sc->want_exit);
rc = -1;
}
/* Byte-id net (the discriminator): per-package asm under
* <stem>.sepwork/<pkg>.s; concat (sorted glob, identical set both
* stages) for the compare. Pre-fix __root.s diverges. */
char cs_s[1024], ws_s[1024];
snprintf(cs_s, sizeof cs_s, "%s/all_cs.s", dir);
snprintf(ws_s, sizeof ws_s, "%s/all_ww.s", dir);
snprintf(cmd, sizeof cmd, "cat %s/main.sepwork/*.s > %s 2>/dev/null",
dir, cs_s); if (system(cmd)) {}
snprintf(cmd, sizeof cmd, "cat %s/mainww.sepwork/*.s > %s 2>/dev/null",
dir, ws_s); if (system(cmd)) {}
if (slurp_eq(cs_s, ws_s) != 0) {
fprintf(stderr, "793[%s]: cs.s/ww.s DIFFER (rule-10 byte-id "
"violation — #21/#224 regression)\n", sc->label);
rc = -1;
}
done:
snprintf(cmd, sizeof cmd, "rm -rf %s", dir);
(void)runwait(cmd);
return rc;
}
int
main(void)
{
const char *bin = getenv("BIN");
if (!bin) bin = "out/bin";
char absbin[2048];
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 cdrv[2100], wdrv[2100];
snprintf(cdrv, sizeof cdrv, "%s/ww", bin);
snprintf(wdrv, sizeof wdrv, "%s/ww_ww", bin);
if (access(wdrv, X_OK) != 0) {
fprintf(stderr, "793: ww_ww missing — cannot run the cs==ww "
"byte-id gate (the whole point of this test)\n");
return 1;
}
int n = (int)(sizeof scenarios / sizeof scenarios[0]);
int fail = 0;
for (int i = 0; i < n; i++) {
if (run_scenario(cdrv, wdrv, &scenarios[i]) != 0)
fail++;
}
if (fail) {
fprintf(stderr, "793 xmod_struct_argpush_collide: %d/%d "
"scenarios failed\n", fail, n);
return 1;
}
printf("xmod_struct_argpush_collide: %d/%d ok (cstage+wwstage run + "
"cs==ww byte-id)\n", n, n);
return 0;
}