From: Jan Stancek <jstancek@redhat.com>
To: ltp@lists.linux.it
Subject: [LTP] [PATCH] tst_test: Allow to set timeout from test setup()
Date: Wed, 3 Aug 2016 11:33:35 -0400 (EDT) [thread overview]
Message-ID: <1951597278.861795.1470238415053.JavaMail.zimbra@redhat.com> (raw)
In-Reply-To: <20160803150655.GB25589@rei>
----- Original Message -----
> From: "Cyril Hrubis" <chrubis@suse.cz>
> To: "Jan Stancek" <jstancek@redhat.com>
> Cc: ltp@lists.linux.it
> Sent: Wednesday, 3 August, 2016 5:06:55 PM
> Subject: Re: [PATCH] tst_test: Allow to set timeout from test setup()
>
> Hi!
> > void tst_run_tcases(int argc, char *argv[], struct tst_test *self)
> > @@ -706,16 +715,9 @@ void tst_run_tcases(int argc, char *argv[], struct
> > tst_test *self)
> >
> > do_setup(argc, argv);
> >
> > - if (tst_test->timeout)
> > - tst_set_timeout(tst_test->timeout);
> > - else
> > - tst_set_timeout(300);
> > -
> > SAFE_SIGNAL(SIGALRM, alarm_handler);
> > SAFE_SIGNAL(SIGUSR1, heartbeat_handler);
> >
> > - alarm(results->timeout);
> > -
>
> After this change the tst_test->setup() runs without a timeout, doesn't it?
>
> I think that it will be safer to have it run with either with a default
> timeout or with a test->timeout. And that is the whole point of setting
> the timeout twice in case that we call tst_set_timeout() in the test
> setup().
Good point, so setting timeout twice seems unavoidable.
But I'm still thinking about actions needed to set timeout
that are now not part of tst_set_timeout(). Would it make sense
to bring those in?
diff --git a/lib/tst_test.c b/lib/tst_test.c
index 72d46969a51d..9eb1393415fa 100644
--- a/lib/tst_test.c
+++ b/lib/tst_test.c
@@ -636,8 +636,6 @@ static void testrun(void)
do_test_setup();
- kill(getppid(), SIGUSR1);
-
if (duration > 0)
stop_time = get_time_ms() + (unsigned long long)(duration * 1000);
@@ -694,6 +692,11 @@ void tst_set_timeout(unsigned int timeout)
tst_res(TINFO, "Timeout per run is %uh %02um %02us",
results->timeout/3600, (results->timeout%3600)/60,
results->timeout % 60);
+
+ if (getpid() == lib_pid)
+ alarm(results->timeout);
+ else
+ kill(getppid(), SIGUSR1);
}
void tst_run_tcases(int argc, char *argv[], struct tst_test *self)
@@ -706,16 +709,14 @@ void tst_run_tcases(int argc, char *argv[], struct tst_test *self)
do_setup(argc, argv);
+ SAFE_SIGNAL(SIGALRM, alarm_handler);
+ SAFE_SIGNAL(SIGUSR1, heartbeat_handler);
+
if (tst_test->timeout)
tst_set_timeout(tst_test->timeout);
else
tst_set_timeout(300);
- SAFE_SIGNAL(SIGALRM, alarm_handler);
- SAFE_SIGNAL(SIGUSR1, heartbeat_handler);
-
- alarm(results->timeout);
-
test_pid = fork();
if (test_pid < 0)
tst_brk(TBROK | TERRNO, "fork()");
next prev parent reply other threads:[~2016-08-03 15:33 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-03 13:53 [LTP] [PATCH] tst_test: Allow to set timeout from test setup() Cyril Hrubis
2016-08-03 14:38 ` Jan Stancek
2016-08-03 15:06 ` Cyril Hrubis
2016-08-03 15:33 ` Jan Stancek [this message]
2016-08-03 15:34 ` Cyril Hrubis
2016-08-03 15:44 ` Jan Stancek
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=1951597278.861795.1470238415053.JavaMail.zimbra@redhat.com \
--to=jstancek@redhat.com \
--cc=ltp@lists.linux.it \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.