On 19 April 2016 at 07:13, Cedric BAIL <cedric.b...@free.fr> wrote:
> On Fri, Apr 15, 2016 at 6:17 PM, Carsten Haitzler <ras...@rasterman.com>
> wrote:
> > On Fri, 15 Apr 2016 13:40:41 -0700 Cedric BAIL <cedric.b...@free.fr>
> said:
> >> On Wed, Apr 13, 2016 at 4:37 PM, Carsten Haitzler <ras...@rasterman.com
> >
> >> wrote:
> >> > On Wed, 13 Apr 2016 11:16:58 -0700 Cedric BAIL <cedric.b...@free.fr>
> said:
> >> >> On Wed, Apr 13, 2016 at 3:47 AM, Carsten Haitzler <
> ras...@rasterman.com>
> >> >> wrote:
> >> >> > i'm wondering... shouldn't we make eo and eo base "threadsafe". ie
> a
> >> >> > basic eo object should have locks around things like:
> >> >>
> >> >> I am not to sure of the intended idea of making eo threadsafe for all
> >> >> of those function. If you are using the same object from two
> >> >> differents thread, it is pretty certain in my opinion that you will
> >> >> need to have lock to manage its use yourself or you will end up with
> >> >> crash. I don't think it is possible to build a threadsafe and safe to
> >> >> use API that won't require the user of that API to actually think
> >> >> about where to put lock. So if we advertise our API as threadsafe and
> >> >> the developer don't understand the limit of it (Like most Java
> >> >> developers for example), he will manage to have race condition and
> >> >> blame Eo for not being properly threadsafe. I believe it will result
> >> >> in a more complex use of eo, when can I rely on it being threadsafe,
> >> >> when can't I ? Which is why I don't think this is a good idea.
> >> >
> >> > you CANNOT build a threadsafe object ever if the base class isn't
> >> > threadsafe. i can NEVER do it. it's impossible to do.
> >>
> >> I do not believe it is possible to have a generic threadsafe object in
> >> C anyway. You need external synchronisation primitive or you will have
> >> race condition when passing parameter and returning value. Some
> >> primitive can be threadsafe, but the concept of threadsafety is
> >> unlikely to be broad.
> >
> > example:
> >
> > thread1:
> >
> > obj = eo_add(..);
> > eo_data_key_set(obj, "stuff", data);
> > send_msg(thread2, obj); // send msg to thread 2 to use obj
> > // some time later
> > for (;;) {
> > eo_data_key_set(obj, "otherstuff", data2);
> > sleep(1);
> > datax = eo_data_key_get(obj, "otherstuff");
> > eo_data_key_del(obj, "otherstuff");
> > }
> >
> > thread2:
> > wait_msg();
> > for (;;) {
> > data3 = eo_data_get_get(obj, "stuff");
> > process(data3);
> > }
>
> I guess, that sleep(1) is another synchronisation method in the
> example above (as is send_msg/wait_msg).
>
> > this above SHOULD work if obj was intended to be threadsafe. reality is
> we are
> > psetered day in and out to make efl threadsafe. all the time. reality is
> that
> > for gui studff this is nigh on impossible because of the model of
> keeping state
> > unrendered until going idle so changes need to be done within a lock of
> some
> > sort within the loop that owns the ui. most of efl is ui. but let's talk
> about
> > things like making the main loop an object so different threads can have
> > different loops. so you can do efl_loop_begin/end(mainloop); we have
> talked
> > about having inter-loop messaging similarly etc. this requires threadsafe
> > objects. if eo_base and below is not threadsafe then we are screwed in
> ever
> > being able to do this. you cannot have the above code which has no race
> > conditions at all because thread1 has guaranteed "stuff" is stored before
> > thread2 ever gets the message to use it and it doesnt ever mess with it
> again.
> > because keys are implemented using a list and a key lookup brings the
> key to
> > the front of the list (LRU) then 2 threads messing with the same list is
> going
> > to end up with a crash despite the programmer doing all the right things.
> > forcing them to create locks for every object separately to the object
> is just
> > poor poor poor design. that's another bit of data to remember, store,
> > delete/clean up when object is cleaned up and so on.
>
> As I have stated in my email, this is just making the internal kind of
> thread safe, but you are still opening huge synchronization issue that
> have to be addressed outside of the object. And this is exactly what
> your example show. So, this means that enabling internal thread safety
> per object, depend on that said object being useful in a thread or
> not. So why would we need that in the base object ?
>
> Remember we can always do in an inherited function :
>
> bob()
> {
> lock();
> super(bob());
> unlock();
> return ;
> }
>
> This will provide exactly the same kind of internal thread safety you
> are advocating without having to pay for its cost for object that
> don't need it.
>
> [snip]
>
> >> > i'm not saying to make all of efl threadsafe. but base class and core
> eo
> >> > stuff like the object table have to be to allow someone to build
> threadsafe
> >> > objects on top.
> >> >
> >> > we already want to make the main loop "thread safe" in that one
> thread will
> >> >
> >> > mainloop.begin();
> >> > or
> >> > efl_loop_beging(mainloop);
> >>
> >> This call has no synchronisation issue as long as eo internal code to
> >> resolve function is threadsafe, because it doesn't influence the
> >> parameter, nor the return value and that's all we need and can do !
> >> Oh, and eo_ref/unref should already be threadsafe, but anyway in the
> >> example above, the mainloop should be refcounted when a thread get
> >> created and unref when it get destroyed. Otherwise the above code will
> >> always be race.
> >
> > no - see above. if we dont lock simple things like key_data_set/get/del
> then
> > the object data structure internals get messed up. i'm not talking about
> the
> > input params or returns. i'm talking about the object internal state is
> > entirely thead-unsafe as 2 threads doing 2 key data ops at the same time
> on
> > the same obj will screw eachother up as they mess with the same data
> struct. it
> > is IMPOSSIBLE to build threadsafe objects on top of eo no matter how hard
> > you try if eo_base itself doesnt even do the simplest attempts to be
> > threadsafe. we need to make ref++ and ref-- atomic (threadsafe) in
> addition to
> > other things like key data, and so on to even make this possible
>
> See above, I don't see why it is impossible. And as stated before,
> eo_ref/eo_unref along with the code doing the function pointer
> resolution need to be threadsafe and that's the only thing that is
> needed. I remember a talk a long time ago with Tom and that was a
> solved case in my mind. Maybe with eo4 it did change, but it
> shouldn't.
>
> [snip]
>
> >> That's why I am really not convinced at all that we should increase
> >> the cost of our base object for pseudo thread safety as at the end it
> >> is doable for specific case in the object themself and in a less
> >> confusing way.
> >
> > then people are left to add locks and do super every base call and they
> CANNOt
> > do this to ref++ and ref--. btw. even then for things like callback_call
> you
> > CANNOt because you have to unlock when you call the cb func as you lose
> scope
> > control so overriding cant do it without a complete re-implementation.
>
> Clearly callback_call is in the hot path, paying the price of any kind
> of lock there will have a severe impact on our performance and the
> benefit is still far from obvious to me. As you say, we can easily
> reimplement the callback_call for handling threadsafety as part of an
> inherited object. Of course in all logic, if we really need a
> internally thread safe object, we can have another object in Eo that
> inherit from Eo_Base that will be able to recycle the internal easily
> without really duplicating code. It makes sense if we ever have object
> that provide internal thread safety to expose them in a specific
> object hierarchy so that it is obvious to developers what they can use
> safely. Otherwise they will confused (I mean more than by all the
> problem they will create by using threads).
>
> Also all of this is very hyptothetical. We do not have any object that
> is useful in another thread right now. Maybe this is something to add
> when we do actually have a use case.
>
>
To sum things up:
We need:
- eo_ref/eu_unref should be atomic
- call_resolve should be thread safe
- eo_data_scope_get() should be safe (I believe it is)
Cedric proposes that Eo.Base remains unsafe but instead we have
Eo.Thread_Safe.Base that implements all of Eo.Base methods in a thread safe
manner.
Then only the rare objects that require thread safety can inherit from
Eo.Thread_Safe.Base and they are "safe" at the base level.
Eo.Thread_Safe.Base can also provide lock(), unlock(), ...
UI objects can not be safe (they could even "abort()" when called from
another thread, a la Android - in debug mode).
Best regards,
--
Jean-Philippe André
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
_______________________________________________
enlightenment-devel mailing list
enlightenment-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
