> On Mar 4, 2016, at 5:28 PM, Andrzej Telszewski <[email protected]> wrote: > > On 04/03/16 01:05, Bubke Marco wrote: >> Can you file a bug report? > > But is it really a bug?
bugreports.qt.io is used for all kinds of issues, be it bugs, suggestions, feature requests, ... Br, Eike > Have a look at the attached image. > > If I add -Wno-padded, then the warning disappears. > > In my opinion, adding -Wno-padded to the default flags would be sufficient. > >> >> -- >> Sent from cellphone, sorry for the typos >> >> >> >> On March 3, 2016 22:22:26 Andrzej Telszewski <[email protected]> wrote: >> >>> On 03/03/16 21:55, Bubke Marco wrote: >>>> You don't have to silence the warning, it is a warning not a error. And >>>> you can change that in the settings. >>>> >>>> I think it quite informative that >>>> >>>> bool x, y, z; >>>> int w; >>>> char c; >>>> >>>> Is not optimal. If you don't like that disable it in the settings. >>> >>> I do think that having this kind of information is useful, but in my >>> opinion -Wno-padded should be default, and turning on the warning should >>> be left to those who need it. I personally added -Wno-padded to the >>> clang model settings. >>> >>> Say you have a struct like that: >>> struct nopad >>> { >>> bool flag1; >>> bool flag2; >>> }; >>> >>> It is perfectly fine structure and you don't have to pad it, yet you get >>> the warning. I would understand, if the warning told you to re-arrange >>> the members, but it simply warns you about what is the default compiler >>> behavior, that it will add 2 or 6 bytes of padding. This has been the >>> default compiler behavior for ages, so I don't see the point of giving >>> the warning by default;) In addition, the padding depends on the >>> architecture you're compiling for, so you would have to take it into >>> account. >>> >>> I'm not against the warning, I'm just saying it shouldn't be enabled by >>> default. >>> >>> My real structure is like that: >>> typedef struct >>> { >>> string name; >>> bool can_span; >>> } info_key_t; >>> >>> There is nothing sensible I can change to make the warning disappear. >>> >>>> >>>> -- >>>> Sent from cellphone, sorry for the typos >>>> >>>> >>>> On March 3, 2016 20:44:28 Andrzej Telszewski <[email protected]> wrote: >>>> >>>>> On 03/03/16 20:23, Bubke Marco wrote: >>>>>> Hi, we will add more options in 4.0 to the code model, so you can >>>>>> choose different sets. >>>>>> >>>>>> Anyway I find the padding warning quite useful because it shows that >>>>>> your data structures are sub optimal. >>>>>> >>>>> >>>>> Nope, it does not tell you it's suboptimal. It tells you that compiler >>>>> will add the additional padding/alignment bytes. >>>>> >>>>> Having structure with say 3 bool's and nothing more is perfectly fine. >>>>> In such a case, to silence the warning, I would have to add some dummy >>>>> fields, which most of the time is not necessary. >>>>> >>>>>> -- >>>>>> Sent from cellphone, sorry for the typos >>>>>> >>>>>> >>>>>> >>>>>> On March 3, 2016 19:30:31 Andrzej Telszewski <[email protected]> >>>>>> wrote: >>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> Qt Creator: 3.6.0 >>>>>>> >>>>>>> Could _-Wno-padded_ flag be added to the Clang Code Model flags? >>>>>>> I don't believe that warning about padding should be the default >>>>>>> behavior. >>>>>>> >>>>>>> -- >>>>>>> Best regards, >>>>>>> Andrzej Telszewski >>>>>>> _______________________________________________ >>>>>>> Qt-creator mailing list >>>>>>> [email protected] >>>>>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>>>> >>>>> >>>>> >>>>> -- >>>>> Best regards, >>>>> Andrzej Telszewski >>>> >>>> -- >>>> Sent from cellphone, sorry for the typos >>>> _______________________________________________ >>>> Qt-creator mailing list >>>> [email protected] >>>> http://lists.qt-project.org/mailman/listinfo/qt-creator >>>> >>> >>> >>> -- >>> Pozdrawiam, >>> Best regards, >>> Andrzej Telszewski >> . >> > > > -- > Pozdrawiam, > Best regards, > Andrzej Telszewski > <struct_padding.png>_______________________________________________ > Qt-creator mailing list > [email protected] > http://lists.qt-project.org/mailman/listinfo/qt-creator -- Eike Ziller, Principle Software Engineer - The Qt Company GmbH The Qt Company GmbH, Rudower Chaussee 13, D-12489 Berlin Geschäftsführer: Mika Pälsi, Juha Varelius, Tuula Haataja Sitz der Gesellschaft: Berlin, Registergericht: Amtsgericht Charlottenburg, HRB 144331 B _______________________________________________ Qt-creator mailing list [email protected] http://lists.qt-project.org/mailman/listinfo/qt-creator
