Tue Jul 13 19:11:35 2021 UTC ()
tests/lint: skip test for floating point overflow on alpha
On alpha and a few other platforms (see t_integration.sh), 'long double'
has 128 bit, which under IEEE 754 rules means a decimal exponent of up
to 4932.
(rillig)
diff -r1.4 -r1.5 src/tests/usr.bin/xlint/lint1/msg_142.c
diff -r1.2 -r1.3 src/tests/usr.bin/xlint/lint1/msg_142.exp
--- src/tests/usr.bin/xlint/lint1/msg_142.c 2021/04/05 01:35:34 1.4
+++ src/tests/usr.bin/xlint/lint1/msg_142.c 2021/07/13 19:11:35 1.5
@@ -1,6 +1,14 @@
-/* $NetBSD: msg_142.c,v 1.4 2021/04/05 01:35:34 rillig Exp $ */
+/* $NetBSD: msg_142.c,v 1.5 2021/07/13 19:11:35 rillig Exp $ */
# 3 "msg_142.c"
// Test for message: floating point overflow detected, op %s [142]
-double dbl = 1e100 * 1e100 * 1e100 * 1e100 * 1e100; /* expect: 142 *//* expect: 142 */
+/* lint1-only-if ldbl-64 */
+/*
+ * For 96-bit and 128-bit floating point numbers, a different number of
+ * multipliers is needed to produce an overflow.
+ */
+
+/* expect+2: warning: floating point overflow detected, op * [142] */
+/* expect+1: warning: floating point overflow detected, op * [142] */
+double dbl = 1e100 * 1e100 * 1e100 * 1e100 * 1e100;
--- src/tests/usr.bin/xlint/lint1/Attic/msg_142.exp 2021/01/08 21:25:03 1.2
+++ src/tests/usr.bin/xlint/lint1/Attic/msg_142.exp 2021/07/13 19:11:35 1.3
@@ -1,2 +1,2 @@
-msg_142.c(6): warning: floating point overflow detected, op * [142]
+msg_142.c(14): warning: floating point overflow detected, op * [142]
-msg_142.c(6): warning: floating point overflow detected, op * [142]
+msg_142.c(14): warning: floating point overflow detected, op * [142]