Files
ww/test/lang/dotbase_hijack_test.ww
Hojun-Cho a6b74e46c4 test: migrate global value r/w family-5 to test/lang @test, retire C twins (fold-2)
Continue fold-2: migrate the global-value read/store/addr-of family (non-tagged)
from bespoke build+run C twins to test/lang @test, retiring each twin in the
same commit. Runtime coverage MOVES from $(TESTS) to test-lang (T1 runs+asserts
via `ww test`) + test-lang-byteid (T2 keeps cs==ww .s byte-id); the $(TESTS)
headline drops 3 (421->418). All asserts are primitive int comparisons; fresh
globals are zeroed, so a store that misses the symbol reads back 0 and FAILS.

  989_globptrfield_run.c  -> glob_ptr_field_test.ww  (scalar field store through a module-global *struct pointer loads the pointer from gp(SB), not saved BP; runtime gp=&backing form, static-init is #48-blocked)
  989_globstructret_run.c -> glob_struct_ret_test.ww (returning a module-global struct ident by value copies g's bytes; rsz 16 + rsz 24, each field asserted)
  989_dotbasehijack_run.c -> dotbase_hijack_test.ww  (indexing an [N]T field of a module-global struct addresses the struct field, not an unrelated global sharing the field name; +typed-inner control +colliding-global-intact pin)

glob_ptr_field keeps only the .c's offset-0 row: an added non-zero-offset row
(gp.g=9) surfaced a cs!=ww divergence (cstage folds the offset into the store
displacement `MOVQ AX,8(BX)`; wwstage emits `ADDQ $8,BX; MOVQ AX,(BX)`) — a
latent global-*struct-pointer field-store divergence beyond the .c's coverage,
reported for the backlog, not carried here. Bump LANGBYTEID_EXPECTED_MIN 35->38.
2026-06-22 12:31:23 +09:00

46 lines
1.8 KiB
Plaintext

// dotbase_hijack_test — indexing an `[N]T` field of a module-GLOBAL struct
// (`gs.fld[i]`) must address gs's field, not be hijacked by an unrelated
// module-global that shares the FIELD's name, migrated from
// test/wcc/989_dotbasehijack_run.c (F8-c4, report-item #20). wwstage's
// dotbaseaddr #128b probe did `letvartnode(base.str)` (base.str = the DOT's
// FIELD name) whenever the inner ident wasn't a local; for `gs.fld` with gs a
// module-global struct, the inner is global so the probe fired and, finding the
// unrelated global `fld: [2]i64`, emitted `LEAQ fld(SB)` — the WRONG symbol
// (returning 9 not 7). cstage gates the probe on a NULL/ty_err inner type (a
// module qualifier has no struct type), so a typed-struct inner resolves the
// field off the struct. The fix gates the wwstage probe the same way; the .s is
// byte-identical. The colliding global is read directly (global_fld_intact) so
// the name clash is genuine; no_collide is the typed-inner control the
// probe-skip must keep correct.
package dotbase_hijack_test;
type s = struct { pad: i64, fld: [2]i64 };
type s2 = struct { pad: i64, arr: [2]i64 };
let gs: s = s { pad = 0, fld = [7, 8] };
let gs2: s2 = s2 { pad = 0, arr = [5, 6] };
let fld: [2]i64 = [9, 10];
@test fn collide_0() void = {
// pre-fix wwstage read fld[0] (== 9) off the colliding global.
assert(gs.fld[0]: i32 == 7);
};
@test fn collide_1() void = {
// pre-fix wwstage read fld[1] (== 10) off the colliding global.
assert(gs.fld[1]: i32 == 8);
};
@test fn no_collide() void = {
// typed-struct inner with NO name collision — the probe-skip must not
// break the ordinary global-struct array-field index.
assert(gs2.arr[0]: i32 == 5);
};
@test fn global_fld_intact() void = {
// the colliding global is a genuine distinct symbol, not gs.fld.
assert(fld[0]: i32 == 9);
assert(fld[1]: i32 == 10);
};