In tr(1), if we have already generated a given character class we
don't need to do it again.
Further, we don't need to keep all the memory we allocate for the set
of characters in the class. NCHARS + 1 is just an upper bound on the
length. We should return whatever we don't need.
This plugs a small leak. You can see the difference with time(1):
# Unpatched.
$ command time -l tr -d $(jot -s'' -b[:graph:] 20000) < /dev/null
0.11 real 0.09 user 0.02 sys
43632 maximum resident set size
0 average shared memory size
0 average unshared data size
0 average unshared stack size
10668 minor page faults
0 major page faults
0 swaps
0 block input operations
0 block output operations
0 messages sent
0 messages received
0 signals received
0 voluntary context switches
0 involuntary context switches
# Patched. Note the huge difference in maximum RSS.
$ command time -l tr -d $(jot -s'' -b[:graph:] 20000) < /dev/null
0.01 real 0.01 user 0.00 sys
1740 maximum resident set size
0 average shared memory size
0 average unshared data size
0 average unshared stack size
394 minor page faults
0 major page faults
0 swaps
0 block input operations
0 block output operations
0 messages sent
0 messages received
0 signals received
0 voluntary context switches
0 involuntary context switches
Personally I think using OOBCH as a sentinel value to mark the end of
the array is silly. We can trivially track the length of the set in
the CLASS struct. I'll save that for a later patch, though. That
involves changing code elsewhere in str.c.
While here, switch to an ANSI function definition.
ok?
Index: str.c
===================================================================
RCS file: /cvs/src/usr.bin/tr/str.c,v
retrieving revision 1.13
diff -u -p -r1.13 str.c
--- str.c 31 Oct 2021 21:34:16 -0000 1.13
+++ str.c 31 Oct 2021 22:12:44 -0000
@@ -164,24 +164,38 @@ static CLASS classes[] = {
};
static void
-genclass(s)
- STR *s;
+genclass(STR *s)
{
- int cnt, (*func)(int);
CLASS *cp, tmp;
- int *p;
+ size_t len;
+ int i;
tmp.name = (char *)s->str;
if ((cp = (CLASS *)bsearch(&tmp, classes, sizeof(classes) /
sizeof(CLASS), sizeof(CLASS), c_class)) == NULL)
errx(1, "unknown class %s", s->str);
- if ((cp->set = p = calloc(NCHARS + 1, sizeof(int))) == NULL)
- errx(1, "no memory for a class");
- for (cnt = 0, func = cp->func; cnt < NCHARS; ++cnt)
- if ((func)(cnt))
- *p++ = cnt;
- *p = OOBCH;
+ /*
+ * Generate the set of characters in the class if we haven't
+ * already done so.
+ */
+ if (cp->set == NULL) {
+ cp->set = reallocarray(NULL, NCHARS + 1, sizeof(*cp->set));
+ if (cp->set == NULL)
+ err(1, NULL);
+ len = 0;
+ for (i = 0; i < NCHARS; i++) {
+ if (cp->func(i)) {
+ cp->set[len] = i;
+ len++;
+ }
+ }
+ cp->set[len] = OOBCH;
+ len++;
+ cp->set = reallocarray(cp->set, len, sizeof(*cp->set));
+ if (cp->set == NULL)
+ err(1, NULL);
+ }
s->cnt = 0;
s->state = SET;
