Re: [PATCH RFC 1/2] Add polling support to pidfd

[Joel Fernandes]

On Fri, Apr 19, 2019 at 02:24:09PM -0700, Daniel Colascione wrote:
> On Fri, Apr 19, 2019 at 2:20 PM Joel Fernandes <joel@xxxxxxxxxxxxxxxxx> wrote:
> >
> > On Fri, Apr 19, 2019 at 10:57:11PM +0200, Christian Brauner wrote:
> > > On Fri, Apr 19, 2019 at 10:34 PM Daniel Colascione <dancol@xxxxxxxxxx> wrote:
> > > >
> > > > On Fri, Apr 19, 2019 at 12:49 PM Joel Fernandes <joel@xxxxxxxxxxxxxxxxx> wrote:
> > > > >
> > > > > On Fri, Apr 19, 2019 at 09:18:59PM +0200, Christian Brauner wrote:
> > > > > > On Fri, Apr 19, 2019 at 03:02:47PM -0400, Joel Fernandes wrote:
> > > > > > > On Thu, Apr 18, 2019 at 07:26:44PM +0200, Christian Brauner wrote:
> > > > > > > > On April 18, 2019 7:23:38 PM GMT+02:00, Jann Horn <jannh@xxxxxxxxxx> wrote:
> > > > > > > > >On Wed, Apr 17, 2019 at 3:09 PM Oleg Nesterov <oleg@xxxxxxxxxx> wrote:
> > > > > > > > >> On 04/16, Joel Fernandes wrote:
> > > > > > > > >> > On Tue, Apr 16, 2019 at 02:04:31PM +0200, Oleg Nesterov wrote:
> > > > > > > > >> > >
> > > > > > > > >> > > Could you explain when it should return POLLIN? When the whole
> > > > > > > > >process exits?
> > > > > > > > >> >
> > > > > > > > >> > It returns POLLIN when the task is dead or doesn't exist anymore,
> > > > > > > > >or when it
> > > > > > > > >> > is in a zombie state and there's no other thread in the thread
> > > > > > > > >group.
> > > > > > > > >>
> > > > > > > > >> IOW, when the whole thread group exits, so it can't be used to
> > > > > > > > >monitor sub-threads.
> > > > > > > > >>
> > > > > > > > >> just in case... speaking of this patch it doesn't modify
> > > > > > > > >proc_tid_base_operations,
> > > > > > > > >> so you can't poll("/proc/sub-thread-tid") anyway, but iiuc you are
> > > > > > > > >going to use
> > > > > > > > >> the anonymous file returned by CLONE_PIDFD ?
> > > > > > > > >
> > > > > > > > >I don't think procfs works that way. /proc/sub-thread-tid has
> > > > > > > > >proc_tgid_base_operations despite not being a thread group leader.
> > > > > > > > >(Yes, that's kinda weird.) AFAICS the WARN_ON_ONCE() in this code can
> > > > > > > > >be hit trivially, and then the code will misbehave.
> > > > > > > > >
> > > > > > > > >@Joel: I think you'll have to either rewrite this to explicitly bail
> > > > > > > > >out if you're dealing with a thread group leader, or make the code
> > > > > > > > >work for threads, too.
> > > > > > > >
> > > > > > > > The latter case probably being preferred if this API is supposed to be
> > > > > > > > useable for thread management in userspace.
> > > > > > >
> > > > > > > At the moment, we are not planning to use this for sub-thread management. I
> > > > > > > am reworking this patch to only work on clone(2) pidfds which makes the above
> > > > > >
> > > > > > Indeed and agreed.
> > > > > >
> > > > > > > discussion about /proc a bit unnecessary I think. Per the latest CLONE_PIDFD
> > > > > > > patches, CLONE_THREAD with pidfd is not supported.
> > > > > >
> > > > > > Yes. We have no one asking for it right now and we can easily add this
> > > > > > later.
> > > > > >
> > > > > > Admittedly I haven't gotten around to reviewing the patches here yet
> > > > > > completely. But one thing about using POLLIN. FreeBSD is using POLLHUP
> > > > > > on process exit which I think is nice as well. How about returning
> > > > > > POLLIN | POLLHUP on process exit?
> > > > > > We already do things like this. For example, when you proxy between
> > > > > > ttys. If the process that you're reading data from has exited and closed
> > > > > > it's end you still can't usually simply exit because it might have still
> > > > > > buffered data that you want to read.  The way one can deal with this
> > > > > > from  userspace is that you can observe a (POLLHUP | POLLIN) event and
> > > > > > you keep on reading until you only observe a POLLHUP without a POLLIN
> > > > > > event at which point you know you have read
> > > > > > all data.
> > > > > > I like the semantics for pidfds as well as it would indicate:
> > > > > > - POLLHUP -> process has exited
> > > > > > - POLLIN  -> information can be read
> > > > >
> > > > > Actually I think a bit different about this, in my opinion the pidfd should
> > > > > always be readable (we would store the exit status somewhere in the future
> > > > > which would be readable, even after task_struct is dead). So I was thinking
> > > > > we always return EPOLLIN.  If process has not exited, then it blocks.
> > > >
> > > > ITYM that a pidfd polls as readable *once a task exits* and stays
> > > > readable forever. Before a task exit, a poll on a pidfd should *not*
> > > > yield POLLIN and reading that pidfd should *not* complete immediately.
> > > > There's no way that, having observed POLLIN on a pidfd, you should
> > > > ever then *not* see POLLIN on that pidfd in the future --- it's a
> > > > one-way transition from not-ready-to-get-exit-status to
> > > > ready-to-get-exit-status.
> > >
> > > What do you consider interesting state transitions? A listener on a pidfd
> > > in epoll_wait() might be interested if the process execs for example.
> > > That's a very valid use-case for e.g. systemd.
> > > We can't use EPOLLIN for that too otherwise you'd need to to waitid(_WNOHANG)
> > > to check whether an exit status can be read which is not nice and then you
> > > multiplex different meanings on the same bit.
> > > I would prefer if the exit status can only be read from the parent which is
> > > clean and the least complicated semantics, i.e. Linus waitid() idea.
> > > EPOLLIN on a pidfd could very well mean that data can be read via
> > > a read() on the pidfd *other* than the exit status. The read could e.g.
> > > give you a lean struct that indicates the type of state transition: NOTIFY_EXIT,
> > > NOTIFY_EXEC, etc.. This way we are not bound to a specific poll event indicating
> > > a specific state.
> > > Though there's a case to be made that EPOLLHUP could indicate process exit
> > > and EPOLLIN a state change + read().
> >
> > According to Linus, POLLHUP usually indicates that something is readable:
> 
> I don't think Linus said that POLLHUP means readable. He did say that
> it usually doesn't make sense to set POLLHUP without POLLIN, but
> that's not the same as POLLHUP indicating readability.

Ok, fair enough.

> > https://lkml.org/lkml/2019/4/18/1181
> > "So generally a HUP condition should mean that POLLIN and POLLOUT also
> > get set. Not because there's any actual _data_ to be read, but simply
> > because the read will not block."
> >
> > I feel the future state changes such as for NOTIFY_EXEC can easily be
> > implemented on top of this patch.
> >
> > Just for the exit notification purposes, the states are:
> > if process has exit_state == 0, block.
> > if process is zombie/dead but not reaped, then return POLLIN
> > if process is reaped, then return POLLIN | POLLHUP
> 
> Setting POLLHUP when the process is reaped is harmless, but I don't
> think it's useful. I can't think of a reason that anyone would care.

We can also outright remove it. Oleg seemed to not mind it, in fact he said
it may be useful to indicate the reap status so at least I am inclined to
leave it in.

> You can't block and wait on reaping, so you could only busy-wait, and
> you can look for ESRCH on any proc file today to detect reaping. I'd

proc file reading is racy though. We shouldn't even talk about that since the
point of pidfd is to avoid such raw "pid" related races.

> rather keep POLLHUP available for some other use than use it to signal
> whether a process is reaped.

Like what? I feel it is semantically better than POLLERR.

thanks,

 - Joel