From: Eric Wong <e@80x24.org>
To: meta@public-inbox.org
Subject: [PATCH v2] drop redundant calls to DS->Reset
Date: Sun, 26 Nov 2023 21:08:01 +0000 [thread overview]
Message-ID: <20231126210802.M17016@dcvr> (raw)
In-Reply-To: <20231126021105.408573-8-e@80x24.org>
Reset gets called on END{} anyways to workaround DBI lifetime
problems, so there's no need to call it near exit. We can't
replace calls to POSIX::_exit with `exit' to force END{} to
run just yet, as there are still some lingering destruction
ordering problems on newer DBI and or Perls.
---
Dropping POSIX::_exit was too aggressive and caused segfaults
on my FreeBSD 13.2 VM.
lib/PublicInbox/CodeSearchIdx.pm | 1 -
lib/PublicInbox/Daemon.pm | 1 -
lib/PublicInbox/IO.pm | 6 ++----
3 files changed, 2 insertions(+), 6 deletions(-)
diff --git a/lib/PublicInbox/CodeSearchIdx.pm b/lib/PublicInbox/CodeSearchIdx.pm
index 3764f13e..bb1d698b 100644
--- a/lib/PublicInbox/CodeSearchIdx.pm
+++ b/lib/PublicInbox/CodeSearchIdx.pm
@@ -1292,7 +1292,6 @@ sub cidx_run { # main entry point
local @PublicInbox::DS::post_loop_do = (\&shards_active);
PublicInbox::DS::event_loop($MY_SIG, $SIGSET);
- PublicInbox::DS->Reset;
$self->lock_release(!!$NCHANGE);
show_json($self);
}
diff --git a/lib/PublicInbox/Daemon.pm b/lib/PublicInbox/Daemon.pm
index f33f6f17..a2c1ed6e 100644
--- a/lib/PublicInbox/Daemon.pm
+++ b/lib/PublicInbox/Daemon.pm
@@ -703,7 +703,6 @@ sub run {
local %POST_ACCEPT;
daemon_loop();
- PublicInbox::DS->Reset;
# ->DESTROY runs when $for_destroy goes out-of-scope
}
diff --git a/lib/PublicInbox/IO.pm b/lib/PublicInbox/IO.pm
index 6593dcdf..5654f3b0 100644
--- a/lib/PublicInbox/IO.pm
+++ b/lib/PublicInbox/IO.pm
@@ -15,10 +15,8 @@ use Errno qw(EINTR EAGAIN);
sub waitcb { # awaitpid callback
my ($pid, $errref, $cb, @args) = @_;
- $errref //= \my $workaround_await_pids_clobbered;
- $$errref = $?; # sets .cerr for _close
+ $$errref = $? if $errref; # sets .cerr for _close
$cb->($pid, @args) if $cb; # may clobber $?
- $? = $$errref;
}
sub attach_pid {
@@ -52,7 +50,7 @@ sub close {
} else { # wait synchronously
my $w = awaitpid($reap->[1]);
}
- $? ? '' : $ret; # use $?, AWAIT_PIDS may be cleared on ->Reset (FIXME?)
+ $? ? '' : $ret;
}
sub DESTROY {
prev parent reply other threads:[~2023-11-26 21:10 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-26 2:10 [PATCH 0/7] more I/O + process reliability and cleanups Eric Wong
2023-11-26 2:10 ` [PATCH 1/7] xap_helper_cxx: do not copy xap_helper.h source Eric Wong
2023-11-26 2:11 ` [PATCH 2/7] xap_client: attach PID to the IO object Eric Wong
2023-11-26 2:11 ` [PATCH 3/7] xap_client: pass arguments to top-level xap_helper Eric Wong
2023-11-26 2:11 ` [PATCH 4/7] xap_helper: allow PI_NO_CXX to disable C++ in more places Eric Wong
2023-11-26 2:11 ` [PATCH 5/7] git: move rbuf handling to PublicInbox::IO Eric Wong
2023-11-26 2:11 ` [PATCH 6/7] git: improve coupling with {sock} and {inflight} fields Eric Wong
2023-11-26 2:11 ` [PATCH 7/7] drop redundant calls to DS->Reset Eric Wong
2023-11-26 21:08 ` Eric Wong [this message]
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: https://public-inbox.org/README
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20231126210802.M17016@dcvr \
--to=e@80x24.org \
--cc=meta@public-inbox.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 read-only IMAP folder(s) and NNTP newsgroup(s).