From: David Kastrup <dak@gnu.org>
To: git@vger.kernel.org
Cc: David Kastrup <dak@gnu.org>
Subject: [PATCH] blame.c: prepare_lines should not call xrealloc for every line
Date: Wed, 12 Feb 2014 15:27:24 +0100 [thread overview]
Message-ID: <1392215244-26785-1-git-send-email-dak@gnu.org> (raw)
Making a single preparation run for counting the lines will avoid memory
fragmentation. Also, fix the allocated memory size which was wrong
when sizeof(int *) != sizeof(int), and would have been too small
for sizeof(int *) < sizeof(int), admittedly unlikely.
Signed-off-by: David Kastrup <dak@gnu.org>
---
Since there was no feedback after the last defense/explanation of the
coding choices, the code rewritten by this patch was much more awful,
and the kind of style requests (fixed already in the last iteration)
are not actually heeded by the core developers themselves, I have no
idea whether this patch will be dropped just like the last one.
As opposed to the last try, this incorporates a suggestion from Jeff
to change sizeof(type) to sizeof(expression) which is not helping much
since the types of lineno and sb->lineno still need to be changed in
sync.
It also fiddles cosmetically with the code layout of the loops.
builtin/blame.c | 46 +++++++++++++++++++++++++++++++---------------
1 file changed, 31 insertions(+), 15 deletions(-)
diff --git a/builtin/blame.c b/builtin/blame.c
index e44a6bb..1aefedf 100644
--- a/builtin/blame.c
+++ b/builtin/blame.c
@@ -1772,25 +1772,41 @@ static int prepare_lines(struct scoreboard *sb)
{
const char *buf = sb->final_buf;
unsigned long len = sb->final_buf_size;
- int num = 0, incomplete = 0, bol = 1;
+ const char *end = buf + len;
+ const char *p;
+ int *lineno;
+ int num = 0, incomplete = 0;
- if (len && buf[len-1] != '\n')
- incomplete++; /* incomplete line at the end */
- while (len--) {
- if (bol) {
- sb->lineno = xrealloc(sb->lineno,
- sizeof(int *) * (num + 1));
- sb->lineno[num] = buf - sb->final_buf;
- bol = 0;
- }
- if (*buf++ == '\n') {
+ for (p = buf;;) {
+ p = memchr(p, '\n', end - p);
+ if (p) {
+ p++;
num++;
- bol = 1;
+ continue;
}
+ break;
}
- sb->lineno = xrealloc(sb->lineno,
- sizeof(int *) * (num + incomplete + 1));
- sb->lineno[num + incomplete] = buf - sb->final_buf;
+
+ if (len && end[-1] != '\n')
+ incomplete++; /* incomplete line at the end */
+
+ sb->lineno = xmalloc(sizeof(*sb->lineno) * (num + incomplete + 1));
+ lineno = sb->lineno;
+
+ *lineno++ = 0;
+ for (p = buf;;) {
+ p = memchr(p, '\n', end - p);
+ if (p) {
+ p++;
+ *lineno++ = p - buf;
+ continue;
+ }
+ break;
+ }
+
+ if (incomplete)
+ *lineno++ = len;
+
sb->num_lines = num + incomplete;
return sb->num_lines;
}
--
1.8.3.2
next reply other threads:[~2014-02-12 14:30 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-12 14:27 David Kastrup [this message]
2014-02-12 19:36 ` [PATCH] blame.c: prepare_lines should not call xrealloc for every line Junio C Hamano
-- strict thread matches above, loose matches on Subject: below --
2014-02-04 21:46 David Kastrup
2014-02-04 21:40 David Kastrup
2014-02-04 20:06 David Kastrup
2014-02-04 20:10 ` David Kastrup
2014-02-04 20:49 ` Junio C Hamano
2014-02-04 21:00 ` Junio C Hamano
2014-02-04 21:09 ` David Kastrup
2014-02-04 22:28 ` Philip Oakley
2014-02-04 22:48 ` Philip Oakley
2014-02-04 20:24 ` Junio C Hamano
2014-02-04 20:52 ` David Kastrup
2014-02-04 21:03 ` Junio C Hamano
2014-02-04 21:11 ` David Kastrup
2014-02-04 21:41 ` Junio C Hamano
2014-02-04 21:27 ` David Kastrup
2014-02-04 21:44 ` Junio C Hamano
2014-02-04 21:48 ` David Kastrup
2014-02-04 22:06 ` Junio C Hamano
2014-02-05 8:39 ` David Kastrup
2014-02-05 20:39 ` Junio C Hamano
2014-02-06 0:34 ` David Kastrup
2014-02-06 10:29 ` David Kastrup
2014-02-05 9:22 ` David Kastrup
2014-02-05 20:34 ` Junio C Hamano
2014-02-05 23:45 ` David Kastrup
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1392215244-26785-1-git-send-email-dak@gnu.org \
--to=dak@gnu.org \
--cc=git@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).