diff --git a/Makefile b/Makefile
index 23983a11..aa499028 100644
--- a/Makefile
+++ b/Makefile
@@ -329,6 +329,7 @@ TESTS	= $(BIN)/test_smoke $(BIN)/test_lex $(BIN)/test_parse $(BIN)/test_check \
 	  $(BIN)/test_checked_run \
 	  $(BIN)/test_f64cgen_run \
 	  $(BIN)/test_f64crossmod_run \
+	  $(BIN)/test_tuprecv_run \
 	  $(BIN)/test_floats_run \
 	  $(BIN)/test_bufio_run $(BIN)/test_random_run
 
@@ -1076,6 +1077,11 @@ $(BIN)/test_f64crossmod_run: test/wcc/953_f64crossmod_run.c $(BIN)/ww \
 		$(LIB)/libwwrt.a | $(BIN)
 	$(CC) $(CFLAGS) -o $@ $<
 
+$(BIN)/test_tuprecv_run: test/wcc/954_tuprecv_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/954_tuprecv_run.c b/test/wcc/954_tuprecv_run.c
new file mode 100644
index 00000000..1f064bf0
--- /dev/null
+++ b/test/wcc/954_tuprecv_run.c
@@ -0,0 +1,234 @@
+/*
+ * 954_tuprecv_run — runtime + byte-id regression net for #102: the 16B
+ * whole-tuple-from-call receive (`let t = mk()`, then field reads t.0 /
+ * t.1).
+ *
+ * Root: wwstage's cglet had NO 16B tuple-from-call receive branch, so a
+ * `let t = call()` whose callee returns a 2-eightbyte (16B) tuple fell
+ * through to the generic single-word store (`MOVQ AX, off(BP)`) and
+ * never spilled word1 (the DX eightbyte) — silent loss of t.1. cstage
+ * has the branch (cmd/w6c/cgen.c:6652: spill AX -> off+0 AND DX ->
+ * off+8). The fix mirrors it in cglet.
+ *
+ * NOTE on the diagnosis: this is NOT a tupstore cursor off-by-one, and
+ * it is NOT f64-specific. The destructure form `let (a,b) = mk()`
+ * (cgmlet + tupstore cursor) was already byte-id; only the whole-tuple
+ * N_LET receive dropped word1. An f64 element only made it visible
+ * first, but the all-integer (i64,i64) receive dropped DX too — so the
+ * "all-integer control" row below is ALSO a fix row (it diverges on
+ * master 60c3e51, byte-id after the fix), not a pure no-op control.
+ *
+ * THIS TEST MUST CATCH A WWSTAGE-ONLY DIVERGENCE. cstage is correct
+ * before and after, so a cstage-only probe is gate-blind. Each row
+ * carries BOTH dimensions (modelled on 953_f64crossmod_run):
+ *   (a) cstage `ww build` + run, asserting the exit code — pins the asm
+ *       both stages converge on as runtime-correct.
+ *   (b) w6c vs w6c_ww `.s` cmp — FAILS if the stages diverge. On master
+ *       60c3e51 (pre-fix) the three scalar-pair single-var rows each
+ *       diverge by exactly one dropped `MOVQ DX, -8(BP)`; post-fix all
+ *       are byte-identical.
+ *
+ * Two CONTROL rows pin that the fix touches nothing else: the
+ * destructure form `let (a,b) = mk()` (cgmlet + tupstore cursor) and a
+ * 32B wide/str-element single-var tuple stay byte-id both pre- and
+ * post-fix (the fix is in cglet/N_LET, gated on sz==16).
+ */
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.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;
+}
+
+struct row { const char *label; const char *src; int want_exit; };
+
+static const struct row rows[] = {
+	/* mixed (f64, i64): f=2.5 -> i32 2, i=7 -> i32 7, 2+7 = 9. The
+	 * f64 element rides its eightbyte in AX at receive and is re-read
+	 * from X0 at field-read; the i64 (t.1) rides DX — the dropped
+	 * word. */
+	{ "f64_i64",
+	  "package main;\n"
+	  "fn mk() (f64, i64) = { return (2.5, 7); };\n"
+	  "export fn main() i32 = {\n"
+	  "\tlet t = mk();\n"
+	  "\tlet f: f64 = t.0;\n"
+	  "\tlet i: i64 = t.1;\n"
+	  "\treturn (f: i32) + (i: i32);\n"
+	  "};\n", 9 },
+	/* order-swap (i64, f64): f64 in the 2nd slot. i=7, f=2.5 -> 2,
+	 * 7+2 = 9. Confirms the spill is position-agnostic (both words go
+	 * over the integer cursor regardless of which holds the f64). */
+	{ "i64_f64",
+	  "package main;\n"
+	  "fn mk() (i64, f64) = { return (7, 2.5); };\n"
+	  "export fn main() i32 = {\n"
+	  "\tlet t = mk();\n"
+	  "\tlet i: i64 = t.0;\n"
+	  "\tlet f: f64 = t.1;\n"
+	  "\treturn (i: i32) + (f: i32);\n"
+	  "};\n", 9 },
+	/* all-integer (i64, i64): a=5, b=7, 5+7 = 12. NOT a no-op control
+	 * — pre-fix wwstage dropped DX here too (proves the fix is not
+	 * f64-gated). */
+	{ "i64_i64",
+	  "package main;\n"
+	  "fn mk() (i64, i64) = { return (5, 7); };\n"
+	  "export fn main() i32 = {\n"
+	  "\tlet t = mk();\n"
+	  "\tlet a: i64 = t.0;\n"
+	  "\tlet b: i64 = t.1;\n"
+	  "\treturn (a: i32) + (b: i32);\n"
+	  "};\n", 12 },
+	/* CONTROL — destructure form `let (f,i) = mk()` (cgmlet + tupstore
+	 * cursor, a different and already-correct path; bootstrap relies on
+	 * it, which is why 990-997 pass). The fix is in cglet/N_LET only, so
+	 * this stays byte-id pre- and post-fix. f=2.5->2, i=7, 2+7 = 9. */
+	{ "ctl_destr",
+	  "package main;\n"
+	  "fn mk() (f64, i64) = { return (2.5, 7); };\n"
+	  "export fn main() i32 = {\n"
+	  "\tlet (f, i) = mk();\n"
+	  "\treturn (f: i32) + (i: i32);\n"
+	  "};\n", 9 },
+	/* CONTROL — wide/str element single-var (i64, str) is a 32B tuple
+	 * handled by the separate 32B receive branch; the fix's sz==16 gate
+	 * excludes it, so it stays byte-id pre- and post-fix. t.0 = 7. */
+	{ "ctl_str",
+	  "package main;\n"
+	  "fn mk() (i64, str) = { return (7, \"hi\"); };\n"
+	  "export fn main() i32 = {\n"
+	  "\tlet t: (i64, str) = mk();\n"
+	  "\treturn t.0: i32;\n"
+	  "};\n", 7 },
+	{ 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, "tuprecv: 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/wwtup_%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 in a scratch dir so the output binary
+		 * (and any intermediates) land there. */
+		char tmpdir[64];
+		snprintf(tmpdir, sizeof tmpdir, "/tmp/wwtup_%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.
+		 * FAILS on the pre-fix wwstage divergence (dropped MOVQ DX). */
+		char cs_s[64], ws_s[64];
+		snprintf(cs_s, sizeof cs_s, "/tmp/wwtup_%d_%d_cs.s",
+		    getpid(), i);
+		snprintf(ws_s, sizeof ws_s, "/tmp/wwtup_%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 tuple-receive tests failed\n",
+		    fail, n);
+		return 1;
+	}
+	printf("tuprecv: %d/%d ok (cstage run + cs==ww byte-id)\n", n, n);
+	return 0;
+}