another re_alloc() bit
realloc(NULL, size) is legitimate usage and nhrealloc() shouldn't log a "< 0x00000000 __FILE__ __LINE__" entry for it. heaputil would complain about freeing Null.
This commit is contained in:
@@ -164,9 +164,10 @@ nhrealloc(
|
||||
* allocation rather than make a new one */
|
||||
|
||||
if (newptr != oldptr) {
|
||||
/* realloc() freed oldptr */
|
||||
(void) fprintf(heaplog, "%c%5s %s %4d %s\n", '<', "",
|
||||
fmt_ptr((genericptr_t) oldptr), line, file);
|
||||
/* if oldptr wasn't Null, realloc() freed it */
|
||||
if (oldptr)
|
||||
(void) fprintf(heaplog, "%c%5s %s %4d %s\n", '<', "",
|
||||
fmt_ptr((genericptr_t) oldptr), line, file);
|
||||
op = '>'; /* new allocation rather than size-change of old one */
|
||||
}
|
||||
(void) fprintf(heaplog, "%c%5u %s %4d %s\n", op, newlth,
|
||||
|
||||
Reference in New Issue
Block a user