From: Jeff King <peff@peff.net>
To: Michael Haggerty <mhagger@alum.mit.edu>
Cc: William Duclot <william.duclot@ensimag.grenoble-inp.fr>,
Junio C Hamano <gitster@pobox.com>,
git@vger.kernel.org, antoine.queru@ensimag.grenoble-inp.fr,
francois.beutin@ensimag.grenoble-inp.fr,
Johannes.Schindelin@gmx.de, mh@glandium.org
Subject: [PATCH] send-pack: use buffered I/O to talk to pack-objects
Date: Wed, 8 Jun 2016 15:42:16 -0400 [thread overview]
Message-ID: <20160608194216.GA3731@sigill.intra.peff.net> (raw)
In-Reply-To: <20160608191918.GB19572@sigill.intra.peff.net>
On Wed, Jun 08, 2016 at 03:19:18PM -0400, Jeff King wrote:
> That made me wonder if we could repeatedly reuse a buffer attached to
> the file descriptor. And indeed, isn't that what stdio is? The whole
> reason this buffer exists is because we are using a direct descriptor
> write. If we switched this function to use fprintf(), we'd avoid the
> whole buffer question, have a fixed cap on our memory use (since we just
> flush anytime the buffer is full) _and_ we'd reduce the number of
> write syscalls we're making by almost a factor of 100.
So all of this strbuf discussion aside, I think it is worth doing
something like this for this particular case.
-- >8 --
Subject: send-pack: use buffered I/O to talk to pack-objects
We start a pack-objects process and then write all of the
positive and negative sha1s to it over a pipe. We do so by
formatting each item into a fixed-size buffer and then
writing each individually. This has two drawbacks:
1. There's some manual computation of the buffer size,
which is not immediately obvious is correct (though it
is).
2. We write() once per sha1, which means a lot more system
calls than are necessary.
We can solve both by wrapping the pipe descriptor in a stdio
handle; this is the same technique used by upload-pack when
serving fetches.
Note that we can also simplify and improve the error
handling here. The original detected a single write error
and broke out of the loop (presumably to avoid writing the
error message over and over), but never actually acted on
seeing an error; we just fed truncated input and took
whatever pack-objects returned.
In practice, this probably didn't matter, as the likely
errors would be caused by pack-objects dying (and we'd
probably just die with SIGPIPE anyway). But we can easily
make this simpler and more robust; the stdio handle keeps an
error flag, which we can check at the end.
Signed-off-by: Jeff King <peff@peff.net>
---
send-pack.c | 33 ++++++++++++++++-----------------
1 file changed, 16 insertions(+), 17 deletions(-)
diff --git a/send-pack.c b/send-pack.c
index 37ee04e..299d303 100644
--- a/send-pack.c
+++ b/send-pack.c
@@ -36,18 +36,15 @@ int option_parse_push_signed(const struct option *opt,
die("bad %s argument: %s", opt->long_name, arg);
}
-static int feed_object(const unsigned char *sha1, int fd, int negative)
+static void feed_object(const unsigned char *sha1, FILE *fh, int negative)
{
- char buf[42];
-
if (negative && !has_sha1_file(sha1))
- return 1;
+ return;
- memcpy(buf + negative, sha1_to_hex(sha1), 40);
if (negative)
- buf[0] = '^';
- buf[40 + negative] = '\n';
- return write_or_whine(fd, buf, 41 + negative, "send-pack: send refs");
+ putc('^', fh);
+ fputs(sha1_to_hex(sha1), fh);
+ putc('\n', fh);
}
/*
@@ -73,6 +70,7 @@ static int pack_objects(int fd, struct ref *refs, struct sha1_array *extra, stru
NULL,
};
struct child_process po = CHILD_PROCESS_INIT;
+ FILE *po_in;
int i;
i = 4;
@@ -97,21 +95,22 @@ static int pack_objects(int fd, struct ref *refs, struct sha1_array *extra, stru
* We feed the pack-objects we just spawned with revision
* parameters by writing to the pipe.
*/
+ po_in = xfdopen(po.in, "w");
for (i = 0; i < extra->nr; i++)
- if (!feed_object(extra->sha1[i], po.in, 1))
- break;
+ feed_object(extra->sha1[i], po_in, 1);
while (refs) {
- if (!is_null_oid(&refs->old_oid) &&
- !feed_object(refs->old_oid.hash, po.in, 1))
- break;
- if (!is_null_oid(&refs->new_oid) &&
- !feed_object(refs->new_oid.hash, po.in, 0))
- break;
+ if (!is_null_oid(&refs->old_oid))
+ feed_object(refs->old_oid.hash, po_in, 1);
+ if (!is_null_oid(&refs->new_oid))
+ feed_object(refs->new_oid.hash, po_in, 0);
refs = refs->next;
}
- close(po.in);
+ fflush(po_in);
+ if (ferror(po_in))
+ die_errno("error writing to pack-objects");
+ fclose(po_in);
if (args->stateless_rpc) {
char *buf = xmalloc(LARGE_PACKET_MAX);
--
2.9.0.rc2.138.geb72a36
next prev parent reply other threads:[~2016-06-08 19:42 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-06 15:13 [PATCH V2 0/3] strbuf: improve API William Duclot
2016-06-06 15:13 ` [PATCH V2 1/3] strbuf: add tests William Duclot
2016-06-06 16:11 ` Matthieu Moy
2016-06-07 8:44 ` Johannes Schindelin
2016-06-06 15:13 ` [PATCH V2 2/3] pretty.c: rename strbuf_wrap() function William Duclot
2016-06-06 16:12 ` Matthieu Moy
2016-06-07 9:04 ` Johannes Schindelin
2016-06-06 15:13 ` [PATCH V2 3/3] strbuf: allow to use preallocated memory William Duclot
2016-06-06 16:17 ` Matthieu Moy
2016-06-06 17:19 ` Junio C Hamano
2016-06-06 20:39 ` William Duclot
2016-06-06 22:44 ` Junio C Hamano
2016-06-06 22:58 ` Jeff King
2016-06-06 23:24 ` Junio C Hamano
2016-06-06 23:25 ` Junio C Hamano
2016-06-06 23:30 ` Jeff King
2016-06-07 9:06 ` William Duclot
2016-06-07 18:10 ` Junio C Hamano
2016-06-08 16:20 ` Michael Haggerty
2016-06-08 18:07 ` Junio C Hamano
2016-06-08 19:19 ` Jeff King
2016-06-08 19:42 ` Jeff King [this message]
2016-06-09 12:10 ` [PATCH] send-pack: use buffered I/O to talk to pack-objects Matthieu Moy
2016-06-09 14:34 ` Ramsay Jones
2016-06-09 17:12 ` Jeff King
2016-06-09 22:40 ` Ramsay Jones
2016-06-09 16:40 ` Junio C Hamano
2016-06-09 17:14 ` Jeff King
2016-06-09 17:22 ` Matthieu Moy
2016-06-08 19:48 ` [PATCH V2 3/3] strbuf: allow to use preallocated memory Junio C Hamano
2016-06-08 19:52 ` Jeff King
2016-06-08 23:05 ` Junio C Hamano
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
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160608194216.GA3731@sigill.intra.peff.net \
--to=peff@peff.net \
--cc=Johannes.Schindelin@gmx.de \
--cc=antoine.queru@ensimag.grenoble-inp.fr \
--cc=francois.beutin@ensimag.grenoble-inp.fr \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=mh@glandium.org \
--cc=mhagger@alum.mit.edu \
--cc=william.duclot@ensimag.grenoble-inp.fr \
/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.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
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 read-only IMAP folder(s) and NNTP newsgroup(s).