diff mbox series

[v2,04/14] use bits_mask() for predefined_max()

Message ID 20181212164554.43133-5-luc.vanoostenryck@gmail.com (mailing list archive)
State Superseded, archived
Headers show
Series predefined macros for intmax_t/intptr_t/... | expand

Commit Message

Luc Van Oostenryck Dec. 12, 2018, 4:45 p.m. UTC 
Creating a bit mask using '(1 << n) - 1' is undefined if
n is as big as the width of an int.

Use the safe helper bits_mask() to create the mask/value
for predefined_max().

Note: predefined_max() is currently correct only for signed types.

Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
---
 lib.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
diff mbox series

Patch

diff --git a/lib.c b/lib.c
index db33e58b2..ad583da93 100644
--- a/lib.c
+++ b/lib.c
@@ -45,6 +45,7 @@ 
 #include "linearize.h"
 #include "target.h"
 #include "version.h"
+#include "bits.h"
 
 int verbose, optimize_level, optimize_size, preprocessing;
 int die_if_error = 0;
@@ -1162,7 +1163,7 @@  static void predefined_width(const char *name, unsigned bits)
 
 static void predefined_max(const char *name, const char *suffix, unsigned bits)
 {
-	unsigned long long max = (1ULL << (bits - 1 )) - 1;
+	unsigned long long max = bits_mask(bits - 1);
 	char buf[32];
 
 	snprintf(buf, sizeof(buf), "__%s_MAX__", name);