From: Harald van Dijk <harald@gigawatt.nl>
To: Herbert Xu <herbert@gondor.apana.org.au>,
Andrej Shadura <andrew.shadura@collabora.co.uk>
Cc: dash@vger.kernel.org, Michael Biebl <biebl@debian.org>,
974705@bugs.debian.org
Subject: Re: Changes to job handling cause hangs in wait
Date: Tue, 1 Dec 2020 10:14:04 +0000 [thread overview]
Message-ID: <0e6bb3c5-96d9-18e1-e6bb-87cb34f85fab@gigawatt.nl> (raw)
In-Reply-To: <20201201060618.GA27281@gondor.apana.org.au>
On 01/12/2020 06:06, Herbert Xu wrote:
> On Tue, Dec 01, 2020 at 04:42:03PM +1100, Herbert Xu wrote:
>>
>> Nevermind, I see that the script has been modified to use bash.
>>
>> I can reproduce the problem now so it's all good.
>
> OK the problem is this:
>
> sh -c 'sleep 1d& exec $MYSHELL -c "sleep 1& wait"'
>
> You can replace MYSHELL with whatever shell you want to use.
>
> Essentially dash will now wait for all children, even ones that
> were created prior to its existence, however, bash only waits for
> children that it created directly.
>
> FWIW ksh exhibits the same behaviour as dash and I think there
> is nothing wrong with this.
POSIX says:
"If the wait utility is invoked with no operands, it shall wait
until all process IDs known to the invoking shell have terminated and
exit with a zero exit status."
I would say that child processes that were created before dash was
started do not have process IDs known to dash.
> So the problem is really in the parent of this shell, which appears
> to be bash:
>
> bash -c set -e; export USER=`id -nu`; . /etc/profile >/dev/null 2>&1 || true; . ~/.profile >/dev/null 2>&1 || true; buildtree="/tmp/autopkgtest-lxc.is4n6xxr/downtmp/build.f2G/real-tree"; mkdir -p -m 1777 -- "/tmp/autopkgtest-lxc.is4n6xxr/downtmp/timedated-artifacts"; export AUTOPKGTEST_ARTIFACTS="/tmp/autopkgtest-lxc.is4n6xxr/downtmp/timedated-artifacts"; export ADT_ARTIFACTS="$AUTOPKGTEST_ARTIFACTS"; mkdir -p -m 755 "/tmp/autopkgtest-lxc.is4n6xxr/downtmp/autopkgtest_tmp"; export AUTOPKGTEST_TMP="/tmp/autopkgtest-lxc.is4n6xxr/downtmp/autopkgtest_tmp"; export ADTTMP="$AUTOPKGTEST_TMP"; export DEBIAN_FRONTEND=noninteractive; export LANG=C.UTF-8; export DEB_BUILD_OPTIONS=parallel=2; unset LANGUAGE LC_CTYPE LC_NUMERIC LC_TIME LC_COLLATE LC_MONETARY LC_MESSAGES LC_PAPER LC_NAME LC_ADDRESS LC_TELEPHONE LC_MEASUREMENT LC_IDENTIFICATION LC_ALL;rm -f /tmp/autopkgtest_script_pid; set -C; echo $$ > /tmp/autopkgtest_script_pid; set +C; trap "rm -f /tmp/autopkgtest_script_pid" EXIT INT QUIT PIPE; cd "$buildtree"; export AUTOPKGTEST_NORMAL_USER=; export ADT_NORMAL_USER=; chmod +x /tmp/autopkgtest-lxc.is4n6xxr/downtmp/build.f2G/real-tree/debian/tests/timedated; touch /tmp/autopkgtest-lxc.is4n6xxr/downtmp/timedated-stdout /tmp/autopkgtest-lxc.is4n6xxr/downtmp/timedated-stderr; /tmp/autopkgtest-lxc.is4n6xxr/downtmp/build.f2G/real-tree/debian/tests/timedated 2> >(tee -a /tmp/autopkgtest-lxc.is4n6xxr/downtmp/timedated-stderr >&2) > >(tee -a /tmp/autopkgtest-lxc.is4n6xxr/downtmp/timedated-stdout);
>
> For some reason this is causing the final two tee's to be created
> as children of debian/tests/timedated rather than the bash shell.
This is because of the same optimisation that dash also has, where it
tries to avoid creating a subshell for the last command in a list when
it can just exec() without a fork() instead. A minimal example without
an explicit exec is
bash -c 'dash -c ": & wait" <(sleep 1d)'
Cheers,
Harald van Dijk
next prev parent reply other threads:[~2020-12-01 10:14 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-16 10:29 Changes to job handling cause hangs in wait Andrej Shadura
2020-11-17 3:33 ` Herbert Xu
2020-12-01 5:38 ` Herbert Xu
2020-12-01 5:42 ` Herbert Xu
2020-12-01 6:06 ` Herbert Xu
2020-12-01 6:56 ` Herbert Xu
2020-12-01 23:21 ` Michael Biebl
2020-12-01 23:26 ` Herbert Xu
2020-12-02 5:31 ` [PATCH] jobs: Only block in waitcmd on first run Herbert Xu
2020-12-03 12:27 ` Changes to job handling cause hangs in wait Michael Biebl
2020-12-07 3:55 ` Herbert Xu
2020-12-03 12:49 ` Michael Biebl
2020-12-01 10:14 ` Harald van Dijk [this message]
2020-12-01 10:34 ` Herbert Xu
2020-12-01 10:50 ` Harald van Dijk
2020-12-01 10:53 ` Herbert Xu
2020-12-01 10:55 ` Harald van Dijk
2020-12-01 10:56 ` Herbert Xu
2020-12-01 10:59 ` Harald van Dijk
2020-12-01 11:01 ` Herbert Xu
2020-12-01 13:04 ` Michael Biebl
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=0e6bb3c5-96d9-18e1-e6bb-87cb34f85fab@gigawatt.nl \
--to=harald@gigawatt.nl \
--cc=974705@bugs.debian.org \
--cc=andrew.shadura@collabora.co.uk \
--cc=biebl@debian.org \
--cc=dash@vger.kernel.org \
--cc=herbert@gondor.apana.org.au \
/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).