[squeak-dev] Process #suspend / #resume semantics

Eliot Miranda eliot.miranda at gmail.com
Thu Dec 30 19:24:16 UTC 2021


Hi Jaromir, Hi Craig, Hi All,

On Tue, Dec 28, 2021 at 3:32 PM Eliot Miranda <eliot.miranda at gmail.com>
wrote:

>
>
> On Tue, Dec 28, 2021 at 2:15 PM Eliot Miranda <eliot.miranda at gmail.com>
> wrote:
>
>>
>>
>> On Tue, Dec 28, 2021 at 1:53 PM <mail at jaromir.net> wrote:
>>
>>> Hi Eliot, all,
>>>
>>> this example shows Mutex's critical section can be entered multiple
>>> times:
>>>
>>
>> I know.  suspend is broken.  Please read my previous message fully and
>> carefully.  If I implement the second alternative then the example works
>> correctly.  In the siulator I get:
>>
>> {a Semaphore(a Process(75019) in [] in [] in UndefinedObject>>DoIt) .
>>  a Mutex(a Process(59775) in Mutex>>critical:) .
>>  a Process(75019) in [] in [] in UndefinedObject>>DoIt . false .
>>  a Process(59775) in Mutex>>critical: . false}
>>
>
> However, this comes at the cost of finding that the new terminate is
> broken.  If suspend does not remove a process from its list then a process
> terminated while waiting on a Semaphore remains on that semaphore.  So
> terminate must not only ensure that any critical sections are released, but
> that the process is removed from its list, if that list is a condition
> variable.  IMO this should happen very early on in terminate.  I'm running
> the second alternative but I have to filter out unrunnable processes in
> signal et al since suspend no longer removes from the list.
>

Having thought about it for a couple of days I now think that the second
alternative is the only rational approach.  This is that suspend always
removes a process from whatever list it is on, but if the list is not its
run queue, the process is backed up one bytecode to the send that invoked
the wait.  Hence if the process resumes it immediately progresses into the
wait again, leaving it exactly where it was if it hadn't been suspended.

Craig I hear your concern, but being able to (especially accidentally)
suspend a process and resume it and find it has progressed beyond whatever
condition variable it was waiting on is entirely unacceptable.

My first alternative, that suspend does not remove a process from its list
if a condition variable, breaks the existing code base.  For example a
typical pattern at start up is to suspend the old version of a process
waiting on a semaphore (e.g. the finalizationProcess) and start up a new
one.  The first alternative leaves the old process waiting on the semaphore.

Digression: VisualWorks also implements the second choice, but it isn't
obvious.  HPS, the VisualWorks VM, can only run jitted code; it has no
interpreter.  So backing up the pc to before the send is really difficult
to implement; it introducers arbitrarily many suspension points since the
send of the wait/enterCriticalSection et al can be preceded by an arbitrary
expression.  Instead, one additional suspension point is introduced,
complementing after a send, at a backward jump, and at method entry (after
frame build, before executing the first bytecode).  Here primitives can be
in one of two states, uncommitted, and committed.  Uncommitted primitives
after primitives in progress.  One can't actually see this state.  A
committed primitive has a frame/context allocated for its execution.  A
committed primitive may have completed (e.g. an FFI call is in progress,
waiting for a result) or is yet to start.  So HPS can back up a committed
primitive wait to the committed but uncompleted state. Hence
resuming reenters the wait state.  This is ok, but complex.

In Cog we have an interpreter and can easily convert a machine3 code frame
in to an interpreted frame, so backing up the process to the send that
invoked the wait/enterCriticalSection etc is fine.  I'll have a go at this
asap.


>
>
>>>     | s p1 p2 p3 m |
>>>     s := Semaphore new.
>>>     m := Mutex new.
>>>     p1 := [m critical: [s wait]] newProcess.
>>>     p1 resume.
>>>     p2 := [m critical: [s wait]] newProcess.
>>>     p2 resume.
>>>     Processor yield.
>>>     { p1. p1 suspend. p2. p2 suspend }.
>>>     p1 resume. p2 resume.
>>>     Processor yield.
>>>     { s. m. p1. p1 isTerminated. p2. p2 isTerminated. m isOwned. m
>>> instVarNamed: 'owner' }.
>>>     p3 := [m critical: [s wait]] newProcess.
>>>     p3 resume.
>>>     Processor yield.
>>>     { s. m. p1. p1 isTerminated. p2. p2 isBlocked. p3. p3 isBlocked. m
>>> isOwned. m instVarNamed: 'owner' }.
>>>
>>> I've just added a third process to your last example; p3 really enters
>>> the critical section and takes m's ownership despite the fact p2 is already
>>> waiting inside m's  critical section - because p2 managed to enter m
>>> withour taking m's ownership.
>>>
>>> Now we could repeat the procedure and keep adding processes inside the
>>> critical section indefinitely :) So I guess this really is a bug.
>>>
>>> Best,
>>>
>>> ~~~
>>> ^[^    Jaromir
>>>
>>> Sent from Squeak Inbox Talk
>>>
>>> On 2021-12-28T20:07:25+01:00, mail at jaromir.net wrote:
>>>
>>> > Hi Eliot,
>>> >
>>> > Thanks! Please see my comments below, it seems to me there may be a
>>> bug in the Mutex.
>>> >
>>> > ~~~
>>> > ^[^    Jaromir
>>> >
>>> > Sent from Squeak Inbox Talk
>>> >
>>> > On 2021-12-27T14:55:22-08:00, eliot.miranda at gmail.com wrote:
>>> >
>>> > > Hi Jaromir,
>>> > >
>>> > > On Mon, Dec 27, 2021 at 2:52 AM <mail at jaromir.net> wrote:
>>> > >
>>> > > > Hi all,
>>> > > >
>>> > > > What is the desirable semantics of resuming a previously suspended
>>> process?
>>> > > >
>>> > >
>>> > > That a process continue exactly as it had if it had not been
>>> suspended in
>>> > > the first place.  In this regard our suspend is hopelessly broken for
>>> > > processes that are waiting on condition variables. See below.
>>> > >
>>> > >
>>> > > >
>>> > > > #resume's comment says: "Allow the process that the receiver
>>> represents to
>>> > > > continue. Put the receiver in *line to become the activeProcess*."
>>> > > >
>>> > > > The side-effect of this is that a terminating process can get
>>> resumed
>>> > > > (unless suspendedContext is set to nil - see test
>>> KernelTests-jar.417 /
>>> > > > Inbox - which has the unfortunate side-effect of #isTerminated
>>> answer true
>>> > > > during termination).
>>> > > >
>>> > >
>>> > > But a process that is terminating should not be resumable.  This
>>> should be
>>> > > a non-issue.  If a process is terminating itself then it is the
>>> active
>>> > > process, it has nil as its suspendedContext, and Processor
>>> > > activeProcess resume always produces an error.. Any process that is
>>> not
>>> > > terminating itself can be made to fail by having the machinery set
>>> the
>>> > > suspendedContext to nil.
>>> > >
>>> >
>>> > Yes agreed, but unfortunately that's precisely what is not happening
>>> in the current and previous #terminate and what I'm proposing in
>>> Kernel-jar.1437 - to set the suspendedContext to nil during termination,
>>> even before calling #releaseCriticalSection.
>>> >
>>> > >
>>> > > > A similar side-effect: a process originally waiting on a semaphore
>>> and
>>> > > > then suspended can be resumed into the runnable state and get
>>> scheduled,
>>> > > > effectively escaping the semaphore wait.
>>> > > >
>>> > >
>>> > > Right,  This is the bug.  So for example
>>> > >     | s p |
>>> > >     s *:=* Semaphore new.
>>> > >     p *:=* [s wait] newProcess.
>>> > >     p resume.
>>> > >     Processor yield.
>>> > >     { p. p suspend }
>>> > >
>>> > > answers an Array of process p that is past the wait, and the
>>> semaphore, s.
>>> > > And
>>> > >
>>> > >     | s p |
>>> > >     s *:=* Semaphore new.
>>> > >     p *:=* [s wait] newProcess.
>>> > >     p resume.
>>> > >     Processor yield.
>>> > >     p suspend; resume.
>>> > >     Processor yield.
>>> > >     p isTerminated
>>> > >
>>> > > answers true, whereas in both cases the process should remain
>>> waiting on
>>> > > the semaphore.
>>> > >
>>> > > >
>>> > > > Is this an expected behavior or a bug?
>>> > > >
>>> > >
>>> > > IMO it is a dreadful bug.
>>> > >
>>> > > > If a bug, should a suspended process somehow remember its previous
>>> state
>>> > > > and/or queue and return to the same one if resumed?
>>> > > >
>>> > >
>>> > > IMO the primitive should back up the process to the
>>> > > wait/primitiveEnterCriticalSection. This is trivial to implement in
>>> the
>>> > > image, but is potentially non-atomic.  It is perhaps tricky to
>>> implement in
>>> > > the VM, but will be atomic.
>>> > >
>>> > > Sorry if I'm missing something :)
>>> > > >
>>> > >
>>> > > You're not missing anything :-)  Here's another example that answers
>>> two
>>> > > processes which should both block but if resumed both make progress.
>>> > >
>>> > >     | s p1 p2 m |
>>> > >     s *:=* Semaphore new.
>>> > >     m *:=* Mutex new.
>>> > >     p1 *:=* [m critical: [s wait]] newProcess.
>>> > >     p1 resume.
>>> > >     p2 *:=* [m critical: [s wait]] newProcess.
>>> > >     p2 resume.
>>> > >     Processor yield.
>>> > >     { p1. p1 suspend. p2. p2 suspend }
>>> > >
>>> > > p1 enters the mutex's critical section, becoming the mutex's owner.
>>> p2 then
>>> > > blocks attempting to enter m's critical section.  Let's resume these
>>> two,
>>> > > and examine the semaphore and mutex:
>>> > >
>>> > >     | s p1 p2 m |
>>> > >     s *:=* Semaphore new.
>>> > >     m *:=* Mutex new.
>>> > >     p1 *:=* [m critical: [s wait]] newProcess.
>>> > >     p1 resume.
>>> > >     p2 *:=* [m critical: [s wait]] newProcess.
>>> > >     p2 resume.
>>> > >     Processor yield.
>>> > >     { p1. p1 suspend. p2. p2 suspend }.
>>> > >     p1 resume. p2 resume.
>>> > >     Processor yield.
>>> > >     { s. m. p1. p1 isTerminated. p2. p2 isTerminated }
>>> > >
>>> > > In this case the end result for p2 is accidentally correct. It ends
>>> up
>>> > > waiting on s within m's critical section. But p1 ends up
>>> terminated.  IMO
>>> > > the correct result is that p1 remains waiting on s, and is still the
>>> owner
>>> > > of m, and p2 remains blocked trying to take ownership of m.
>>> > >
>>> >
>>> > Perfect example! My naive expectation was when a process inside a
>>> critical section gets suspended the Mutex gets unlocked but that's
>>> apparently wrong :)
>>> >
>>> > But still, there's something wrong with the example: If p1 resumes it
>>> releases m's ownership and terminates, then p2 takes over and proceeds
>>> inside the critical section and gets blocked at the semaphore. I'd expect
>>> p2 would become the owner of the Mutex m BUT it's not! There's no owner
>>> while p2 is sitting at the semaphore. Try:
>>> >
>>> >     | s p1 p2 m |
>>> >     s := Semaphore new.
>>> >     m := Mutex new.
>>> >     p1 := [m critical: [s wait]] newProcess.
>>> >     p1 resume.
>>> >     p2 := [m critical: [s wait]] newProcess.
>>> >     p2 resume.
>>> >     Processor yield.
>>> >     { p1. p1 suspend. p2. p2 suspend }.
>>> >     p1 resume. p2 resume.
>>> >     Processor yield.
>>> >     { s. m. p1. p1 isTerminated. p2. p2 isTerminated. m isOwned. m
>>> instVarNamed: 'owner' }
>>> >
>>> > It seems to me that when p2 gets suspended it is stopped somewhere
>>> inside #primitiveEnterCriticalSection before the owner is set and when it
>>> gets resumed it is placed into the runnable queue with the pc pointing
>>> right behind the primitive and so when it runs it just continues inside
>>> #critical and get blocked at the semaphore, all without having the
>>> ownership.
>>> >
>>> > Is this interpretation right? It would mean Mutex's critical section
>>> can be entered twice via this mechanism...
>>> >
>>> > Cuis does set the ownership to p2 in this example.
>>> >
>>> > Thanks again,
>>> >
>>> > Jaromir
>>> > >
>>> > > >
>>> > > > Best,
>>> > > > ~~~
>>> > > > ^[^    Jaromir
>>> > > >
>>> > > > Sent from Squeak Inbox Talk
>>> > > >
>>> > >
>>> > > _,,,^..^,,,_
>>> > > best, Eliot
>>> > > -------------- next part --------------
>>> > > An HTML attachment was scrubbed...
>>> > > URL: <
>>> http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20211227/8719df13/attachment.html
>>> >
>>> > >
>>> > >
>>> >
>>> >
>>>
>>
>>
>> --
>> _,,,^..^,,,_
>> best, Eliot
>>
>
>
> --
> _,,,^..^,,,_
> best, Eliot
>


-- 
_,,,^..^,,,_
best, Eliot
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20211230/060f87d7/attachment.html>


More information about the Squeak-dev mailing list