[Interest] Signal and Slot mechansim in Qt4.8.x
Hi, I got a usecase where I had implemented in Qt4.8.x app on the assumption that slots execute one after another in the order they are connected. HMI response is slow when continuous data is pumped into my application from other device. Though we maintain separate threads(threads implementation being derived from Qthread) for reading from the ports and writing onto temporary buffers, HMI is still slow to respond while view switching. If multiple signals are emitted once, I would like to know if framework would take care of slots execution in parallel or in sequential order. Thanks and Regards, Ramakanth DISCLAIMER: This email (including any attachments) is intended for the sole use of the intended recipient/s and may contain material that is CONFIDENTIAL AND PRIVATE COMPANY INFORMATION. Any review or reliance by others or copying or distribution or forwarding of any or all of the contents in this message is STRICTLY PROHIBITED. If you are not the intended recipient, please contact the sender by email and delete all copies; your cooperation in this regard is appreciated. Disclaimer: This message and the information contained herein is proprietary and confidential and subject to the Tech Mahindra policy statement, you may review the policy at http://www.techmahindra.com/Disclaimer.html externally http://tim.techmahindra.com/tim/disclaimer.html internally within TechMahindra. ___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
Re: [Interest] Signal and Slot mechansim in Qt4.8.x
Hi, The order of the slots execution is undefined. You can not rely on it. Bo. Den 08-08-2014 10:39, Ramakanthreddy Kesireddy skrev: Hi, I got a usecase where I had implemented in Qt4.8.x app on the assumption that slots execute one after another in the order they are connected. HMI response is slow when continuous data is pumped into my application from other device. Though we maintain separate threads(threads implementation being derived from Qthread) for reading from the ports and writing onto temporary buffers, HMI is still slow to respond while view switching. If multiple signals are emitted once, I would like to know if framework would take care of slots execution in parallel or in sequential order. -- Viking Software Qt and C++ developers for hire http://www.vikingsoft.eu ___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
Re: [Interest] Signal and Slot mechansim in Qt4.8.x
Il 08/08/2014 10:51, Bo Thorsen ha scritto: The order of the slots execution is undefined. You can not rely on it. Generally speaking it *is* defined: slot activation follows the order of the connect() statements (*). In this specific case, there are multiple threads in the game, and the question doesn't elaborate enough on what's the actual structure of the connections (which threads emit which signals? what are those signals connected to? which connection types are sued?) to be able to know what's going on. (*) It has been finalized, by documenting it, around Qt 4.6. I guess simply too many people were relying on that behaviour (which was implemented, just not documented) to think to modify it at any time. HTH, -- Join us Oct 6-8 at BCC Berlin for Qt Developer Days 2014! Giuseppe D'Angelo | giuseppe.dang...@kdab.com | Software Engineer KDAB (UK) Ltd., a KDAB Group company Tel. UK +44-1738-450410, Sweden (HQ) +46-563-540090 KDAB - Qt Experts - Platform-independent software solutions smime.p7s Description: Firma crittografica S/MIME ___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
Re: [Interest] Signal and Slot mechansim in Qt4.8.x
Den 08-08-2014 12:15, Giuseppe D'Angelo skrev: Il 08/08/2014 10:51, Bo Thorsen ha scritto: The order of the slots execution is undefined. You can not rely on it. Generally speaking it *is* defined: slot activation follows the order of the connect() statements (*). Yes. In this specific case, there are multiple threads in the game, and the question doesn't elaborate enough on what's the actual structure of the connections (which threads emit which signals? what are those signals connected to? which connection types are sued?) to be able to know what's going on. It's the threads that kill this. He said that multiple threads are emitting at the same time (I'm assuming this means multiple cores on the target machine). I think it would be very dangerous to assume anything on the ordering of this. If he needs a strict order of this, he should implement a producer-consumer system for the messages instead. Bo. -- Viking Software Qt and C++ developers for hire http://www.vikingsoft.eu ___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
Re: [Interest] Signal and Slot mechansim in Qt4.8.x
Am 08.08.2014 um 10:39 schrieb Ramakanthreddy Kesireddy ramakanthreddy.kesire...@techmahindra.com: ... If multiple signals are emitted once, I would like to know if framework would take care of slots execution in parallel or in sequential order. For direct connections slots are always executed in the same thread from where the signal originated. As Bo already pointed out if you have several slots - directly - connected to a given signal, then they are sequentially processed - in /any/ (undefined/undocumented) order! If on the other hand those slots are indirectly connected (aka Queued Connection) then the slots are processed in their respective threads. So if you have N threads and each is connecting a distinct slot to a given signal with a queued connection (implying that each thread is running its own event queue), when this signal fires then the N slots execute in parallel (as far as the thread scheduler allows). Read more in the Qt docs, Signals and Slots Across Threads. Cheers, Oliver___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
Re: [Interest] Signal and Slot mechansim in Qt4.8.x
Am 08.08.2014 um 12:15 schrieb Giuseppe D'Angelo giuseppe.dang...@kdab.com: Il 08/08/2014 10:51, Bo Thorsen ha scritto: The order of the slots execution is undefined. You can not rely on it. Generally speaking it *is* defined: slot activation follows the order of the connect() statements (*). But, but...! ... (*) It has been finalized, by documenting it, around Qt 4.6. I guess simply too many people were relying on that behaviour I was just about to complain ;) Didn't realise that this behaviour has been made official now :) It might be okay if you have a /single/ class connecting to a given signal in a well-defined order. But as soon as slots of multiple classes are to be connected to a given signal in a well-defined order such a design would probably be doomed to fail rather sooner than later, IMHO. So I'm a bit surprised that people started to rely on the connection order (I remember faintly that some Qt docs even stated clearly that the order was undefined). But then again, nothing should surprise me anymore in software engineering (just frighten) ;) Cheers, Oliver ___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
Re: [Interest] Signal and Slot mechansim in Qt4.8.x
Den 08-08-2014 14:03, Till Oliver Knoll skrev: Am 08.08.2014 um 12:15 schrieb Giuseppe D'Angelo giuseppe.dang...@kdab.com: Il 08/08/2014 10:51, Bo Thorsen ha scritto: The order of the slots execution is undefined. You can not rely on it. Generally speaking it *is* defined: slot activation follows the order of the connect() statements (*). But, but...! ... (*) It has been finalized, by documenting it, around Qt 4.6. I guess simply too many people were relying on that behaviour I was just about to complain ;) Didn't realise that this behaviour has been made official now :) Yep, it says so on the internet (http://qt-project.org/doc/qt-5/signalsandslots.html) so it must be true :) It might be okay if you have a /single/ class connecting to a given signal in a well-defined order. But as soon as slots of multiple classes are to be connected to a given signal in a well-defined order such a design would probably be doomed to fail rather sooner than later, IMHO. As long as your application is single threaded, then the order is perfectly well defined. If you think of connections as implementations of the visitor pattern with one visitor object in a list of signals, then it's not hard to do this. Emitting a signal and the connected slot calls are exactly the same as calling any C++ function, so the order is easy. The most surprising thing to me is that it took until 4.6 before this was documented, since there doesn't seem to be any good reason not to implement connects as a list and call these one after the other. Bo. -- Viking Software Qt and C++ developers for hire http://www.vikingsoft.eu ___ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
