Re: selftests: pidfd: pidfd_wait hangs on linux next kernel on all devices

[Christian Brauner]

On Thu, Sep 24, 2020 at 04:33:17PM +0200, Christian Brauner wrote:
> On Wed, Sep 23, 2020 at 07:52:05PM +0530, Naresh Kamboju wrote:
> > selftests: pidfd: pidfd_wait hangs on linux next kernel on x86_64,
> > i386 and arm64 Juno-r2
> > These devices are using NFS mounted rootfs.
> > I have tested pidfd testcases independently and all test PASS.
> > 
> > The Hang or exit from test run noticed when run by run_kselftest.sh
> > 
> > pidfd_wait.c:208:wait_nonblock:Expected sys_waitid(P_PIDFD, pidfd,
> > &info, WSTOPPED, NULL) (-1) == 0 (0)
> > wait_nonblock: Test terminated by assertion
> > 
> > metadata:
> >   git branch: master
> >   git repo: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
> >   git commit: e64997027d5f171148687e58b78c8b3c869a6158
> >   git describe: next-20200922
> >   make_kernelversion: 5.9.0-rc6
> >   kernel-config:
> > http://snapshots.linaro.org/openembedded/lkft/lkft/sumo/intel-core2-32/lkft/linux-next/865/config
> > 
> > Reported-by: Naresh Kamboju <naresh.kamboju@xxxxxxxxxx>
> 
> Thanks for reproting this. I'm taking a look now!

Ok, this is a simple race in the selftests, that I overlooked and which
is more likely to hit when there's a lot of processes running on the
system. Basically the child process hasn't SIGSTOPed itself yet but the
parent is already calling waitid() on a O_NONBLOCK pidfd. Since it
doesn't find a WSTOPPED process it returns -EAGAIN correctly.

The fix for this is to move the line where we're removing the O_NONBLOCK
property from the fd before the waitid() WSTOPPED call so we hang until
the child becomes stopped.

So I believe the fix is:

diff --git a/tools/testing/selftests/pidfd/pidfd_wait.c b/tools/testing/selftests/pidfd/pidfd_wait.c
index 4063d6f31fa4..be2943f072f6 100644
--- a/tools/testing/selftests/pidfd/pidfd_wait.c
+++ b/tools/testing/selftests/pidfd/pidfd_wait.c
@@ -205,6 +205,8 @@ TEST(wait_nonblock)
        ret = sys_waitid(P_PIDFD, pidfd, &info, WEXITED | WNOHANG, NULL);
        ASSERT_EQ(ret, 0);

+       ASSERT_EQ(fcntl(pidfd, F_SETFL, (flags & ~O_NONBLOCK)), 0);
+
        ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WSTOPPED, NULL), 0);
        ASSERT_EQ(info.si_signo, SIGCHLD);
        ASSERT_EQ(info.si_code, CLD_STOPPED);
@@ -212,8 +214,6 @@ TEST(wait_nonblock)

        ASSERT_EQ(sys_pidfd_send_signal(pidfd, SIGCONT, NULL, 0), 0);

-       ASSERT_EQ(fcntl(pidfd, F_SETFL, (flags & ~O_NONBLOCK)), 0);
-
        ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL), 0);
        ASSERT_EQ(info.si_signo, SIGCHLD);
        ASSERT_EQ(info.si_code, CLD_EXITED);

Christian