From: Jason Riedy <ejr@EECS.Berkeley.EDU>
To: git@vger.kernel.org
Subject: Re: Solaris cloning woes partly diagnosed
Date: Sun, 02 Apr 2006 12:52:35 -0700 [thread overview]
Message-ID: <824.1144007555@lotus.CS.Berkeley.EDU> (raw)
In-Reply-To: <Pine.LNX.4.64.0604021159110.3050@g5.osdl.org>
And Linus Torvalds writes:
-
- so it really really looks like fgets() would have problems with a SIGALRM
- coming in and doesn't just re-try on EINTR. Can Solaris stdio _really_ be
- that broken? (Yeah, yeah, it may be "conforming". It's also so incredibly
- programmer-unfriendly that it's not even funny)
Yes, it is that broken. I haven't encountered the problem
consistently in git myself, so I can't tell you if the patch
works. Google finds similar reports and patches for BOINC, ruby,
and a few other projects.
Solaris folks will say you should be using sigaction with
SA_RESTART. IIRC, SA_RESTART isn't guaranteed to be there
or work, but all the systems I deal with right now have it.
So an alternate patch for this one use is appended... Other
uses of signal could be changed to sigaction, too. And
progress_update "should" be sig_atomic_t.
Passes the pack-objects tests, but I can't make the problem
happen on demand. (I have seen it occur before, but never
during make test, and I'd not tracked it down...)
Jason
----
diff --git a/pack-objects.c b/pack-objects.c
index ccfaa5f..1faa0bb 100644
--- a/pack-objects.c
+++ b/pack-objects.c
@@ -877,10 +877,21 @@ static int try_delta(struct unpacked *cu
return 0;
}
-static void progress_interval(int signum)
+static void progress_interval(int);
+
+static void setup_progress_signal(void)
+{
+ struct sigaction sa;
+ sa.sa_handler = progress_interval;
+ sigemptyset(&sa.sa_mask);
+ sa.sa_flags = SA_RESTART;
+ sigaction(SIGALRM, &sa, NULL);
+}
+
+void progress_interval(int signum)
{
- signal(SIGALRM, progress_interval);
progress_update = 1;
+ setup_progress_signal();
}
static void find_deltas(struct object_entry **list, int window, int depth)
@@ -1094,7 +1105,7 @@ int main(int argc, char **argv)
v.it_interval.tv_sec = 1;
v.it_interval.tv_usec = 0;
v.it_value = v.it_interval;
- signal(SIGALRM, progress_interval);
+ setup_progress_signal();
setitimer(ITIMER_REAL, &v, NULL);
fprintf(stderr, "Generating pack...\n");
}
next prev parent reply other threads:[~2006-04-02 19:52 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-04-02 10:41 Solaris cloning woes partly diagnosed Junio C Hamano
2006-04-02 18:33 ` Linus Torvalds
2006-04-02 19:10 ` Jason Riedy
2006-04-02 19:22 ` Linus Torvalds
2006-04-02 19:18 ` Linus Torvalds
2006-04-02 19:52 ` Jason Riedy [this message]
2006-04-02 20:28 ` Linus Torvalds
2006-04-02 20:31 ` [PATCH 2/2] pack-objects: be incredibly anal about stdio semantics Linus Torvalds
2006-04-02 21:09 ` Junio C Hamano
2006-04-02 21:21 ` Linus Torvalds
2006-04-02 22:12 ` Jason Riedy
2006-04-02 22:58 ` Linus Torvalds
2006-04-02 22:29 ` [PATCH] Use sigaction and SA_RESTART in read-tree.c; add option in Makefile Jason Riedy
2006-04-03 1:02 ` Linus Torvalds
2006-04-03 4:20 ` Junio C Hamano
2006-04-03 4:40 ` Linus Torvalds
2006-04-03 3:06 ` Solaris cloning woes partly diagnosed Linus Torvalds
2006-04-04 18:21 ` H. Peter Anvin
2006-04-04 8:47 ` [RFH] Solaris cloning woes Junio C Hamano
2006-04-04 18:53 ` Jason Riedy
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=824.1144007555@lotus.CS.Berkeley.EDU \
--to=ejr@eecs.berkeley.edu \
--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).