diff --git a/Makefile b/Makefile index aa499028..26a7f24a 100644 --- a/Makefile +++ b/Makefile @@ -330,6 +330,7 @@ TESTS = $(BIN)/test_smoke $(BIN)/test_lex $(BIN)/test_parse $(BIN)/test_check \ $(BIN)/test_f64cgen_run \ $(BIN)/test_f64crossmod_run \ $(BIN)/test_tuprecv_run \ + $(BIN)/test_f64xmm_run \ $(BIN)/test_floats_run \ $(BIN)/test_bufio_run $(BIN)/test_random_run @@ -1082,6 +1083,11 @@ $(BIN)/test_tuprecv_run: test/wcc/954_tuprecv_run.c $(BIN)/ww \ $(LIB)/libwwrt.a | $(BIN) $(CC) $(CFLAGS) -o $@ $< +$(BIN)/test_f64xmm_run: test/wcc/955_f64xmm_run.c $(BIN)/ww \ + $(BIN)/w6c $(BIN)/w6c_ww $(BIN)/w6a $(BIN)/w6l \ + $(LIB)/libwwrt.a | $(BIN) + $(CC) $(CFLAGS) -o $@ $< + sizelint: @sh tools/sizelint diff --git a/test/wcc/955_f64xmm_run.c b/test/wcc/955_f64xmm_run.c new file mode 100644 index 00000000..0731a627 --- /dev/null +++ b/test/wcc/955_f64xmm_run.c @@ -0,0 +1,209 @@ +/* + * 955_f64xmm_run — runtime + byte-id regression net for #103: an f64 + * value failing to materialise in XMM (X0) before an SSE op. Two faces, + * same class, both GATE-BLIND (cstage cgen.c and wwstage cgenexpr.ww + * emitted byte-identical-but-wrong asm, so the 990-997 byte-id gates + * could never catch a reintroduction — only an executed-and-checked + * runtime probe can). + * + * FACE X — a no-decimal float-typed integer literal (`0f64`, `8f64`) + * is an N_INTLIT carrying float TYPE; the integer-immediate path + * stranded it in AX, so `n == 0f64` compared a stale X0 (true for + * all n) and `(8f64 * 10.0): i32` read garbage. Fixed by routing + * the float-typed N_INTLIT through the float-constant-in-X0 emit + * (cgen.c cgexpr_float / cgenexpr.ww cgfloatbits), plus the wwstage + * exprfloatkind N_INTLIT arm so the downstream f64->i32 cast emits + * CVTTSD2SI not MOVSXD (the #101 structural-vs-stamped asymmetry). + * FACE Z — a tuple positional f64 field read (`r.0`, r:(f64,i64)) + * loaded via the integer op into AX, so `r.0 == 0.0` was wrongly + * true. Fixed by a fld_isfloat branch -> MOVSD/MOVSS into X0 + * (cgen.c:5910 / cgenexpr.ww tuple arm), mirroring the struct-field + * float load at cgen.c:1462,1838 (the #96 pattern). + * + * Each row carries BOTH dimensions (like 953_f64crossmod_run): + * (a) cstage `ww build` + run, asserting the exit code. + * (b) w6c vs w6c_ww `.s` cmp — FAILS if the stages diverge. Both + * stages are fixed identically, so this stays byte-identical + * before and after; it catches one stage being fixed without the + * other. + */ +#include +#include +#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; +} + +struct row { const char *label; const char *src; int want_exit; }; + +static const struct row rows[] = { + /* FACE X false-case: g(8.0) must be 0. On the bug `n == 0f64` + * compares against a stale X0 (true for all n) -> 1. */ + { "x_cmp_false", + "package main;\n" + "fn g(n: f64) i32 = { if (n == 0f64) { return 1; }; return 0; };\n" + "export fn main() i32 = { return g(8.0); };\n", 0 }, + /* FACE X true-case: g(0.0) must still be 1 — the genuine equal + * case must survive the fix. */ + { "x_cmp_true", + "package main;\n" + "fn g(n: f64) i32 = { if (n == 0f64) { return 1; }; return 0; };\n" + "export fn main() i32 = { return g(0.0); };\n", 1 }, + /* FACE X arith: (8f64 * 10.0): i32 == 80. On the bug 8f64 never + * reaches X0, the MULSD reads stale X0 -> garbage. */ + { "x_arith", + "package main;\n" + "export fn main() i32 = { return (8f64 * 10.0): i32; };\n", 80 }, + /* FACE Z: tuple positional f64 field compare. r.0 == 0.0 with + * r = (8.0, 0) must be false -> 9. On the bug r.0 loads into AX + * (integer op), `== 0.0` reads stale X0 -> wrongly true -> 5. */ + { "z_tuple_field", + "package main;\n" + "fn norm(n: f64) (f64, i64) = { return (n, 0); };\n" + "export fn main() i32 = {\n" + " const r = norm(8.0);\n" + " if (r.0 == 0.0) { return 5; };\n" + " return 9;\n" + "};\n", 9 }, + /* CONTROL: the let-bound spelling (`const m = r.0; m == 0.0`) was + * already correct (the let-init is float-aware) and must STAY + * correct -> 9. Guards against the FACE-Z fix over- or + * under-reaching. */ + { "z_letbound_control", + "package main;\n" + "fn norm(n: f64) (f64, i64) = { return (n, 0); };\n" + "export fn main() i32 = {\n" + " const r = norm(8.0);\n" + " const m = r.0;\n" + " if (m == 0.0) { return 5; };\n" + " return 9;\n" + "};\n", 9 }, + { NULL, NULL, 0 } +}; + +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; +} + +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 w6c[1100], w6c_ww[1100]; + snprintf(w6c, sizeof w6c, "%s/w6c", bin); + snprintf(w6c_ww, sizeof w6c_ww, "%s/w6c_ww", bin); + if (access(w6c_ww, X_OK) != 0) { + fprintf(stderr, "f64xmm: w6c_ww missing — cannot run the " + "cs==ww byte-id gate (the whole point of this test)\n"); + return 1; + } + + int n = 0, fail = 0; + for (int i = 0; rows[i].src; i++, n++) { + char src[64]; + snprintf(src, sizeof src, "/tmp/wwf64m_%d_%d.ww", getpid(), i); + FILE *f = fopen(src, "wb"); + if (f == NULL) { fail++; continue; } + fputs(rows[i].src, f); + fclose(f); + + /* (a) cstage build + run. */ + char tmpdir[64]; + snprintf(tmpdir, sizeof tmpdir, "/tmp/wwf64m_%d_d_%d", + getpid(), i); + mkdir(tmpdir, 0755); + + char cmd[2048]; + snprintf(cmd, sizeof cmd, "cd %s && %s/ww build %s", + tmpdir, bin, src); + if (runwait(cmd) != 0) { + fprintf(stderr, "row[%s]: cstage build failed\n", + rows[i].label); + fail++; + unlink(src); rmdir(tmpdir); + continue; + } + + char outbin[128]; + const char *base = strrchr(src, '/'); + base = base ? base + 1 : src; + snprintf(outbin, sizeof outbin, "%s/%s", tmpdir, base); + char *dot = strrchr(outbin, '.'); + if (dot && strcmp(dot, ".ww") == 0) *dot = '\0'; + + int got = runwait(outbin); + if (got != rows[i].want_exit) { + fprintf(stderr, "row[%s]: cstage exit %d, want %d\n", + rows[i].label, got, rows[i].want_exit); + fail++; + } + unlink(outbin); rmdir(tmpdir); + + /* (b) cs==ww byte-id gate: emit .s from both stages, cmp. */ + char cs_s[64], ws_s[64]; + snprintf(cs_s, sizeof cs_s, "/tmp/wwf64m_%d_%d_cs.s", + getpid(), i); + snprintf(ws_s, sizeof ws_s, "/tmp/wwf64m_%d_%d_ww.s", + getpid(), i); + + snprintf(cmd, sizeof cmd, "%s -o %s %s 2>/dev/null", + w6c, cs_s, src); + if (runwait(cmd) != 0) { + fprintf(stderr, "row[%s]: w6c failed\n", rows[i].label); + fail++; unlink(src); continue; + } + snprintf(cmd, sizeof cmd, "%s -o %s %s 2>/dev/null", + w6c_ww, ws_s, src); + if (runwait(cmd) != 0) { + fprintf(stderr, "row[%s]: w6c_ww failed\n", + rows[i].label); + fail++; unlink(src); unlink(cs_s); continue; + } + if (slurp_eq(cs_s, ws_s) != 0) { + fprintf(stderr, + "row[%s]: cstage/wwstage .s DIFFER (rule-10 " + "byte-id violation)\n", rows[i].label); + fail++; + } + unlink(src); unlink(cs_s); unlink(ws_s); + } + + if (fail) { + fprintf(stderr, "%d/%d f64 xmm-materialise tests failed\n", + fail, n); + return 1; + } + printf("f64xmm: %d/%d ok (cstage run + cs==ww byte-id)\n", n, n); + return 0; +}