MW-Subsystem refactoring (Task #107)

Everything about development and the OpenMW source code.
corristo
Posts: 495
Joined: 12 Aug 2011, 08:29

Re: MW-Subsystem refactoring (Task #107)

Post by corristo »

We're using cmake, so Visual Studio projects just generated by cmake.

And I think everyone using Linux/OS X/any other *nix knows about -j key for make to compile several files in parallel.
Does Visual Studio only allow building projects in parallel?
User avatar
gus
Posts: 390
Joined: 11 Aug 2011, 15:41

Re: MW-Subsystem refactoring (Task #107)

Post by gus »

Thanks a lot! it greatly reduced compilation time on windows (more than 45% i think).
Does Visual Studio only allow building projects in parallel?
No. But the file generated by CMake doesn't build anything in parallel.
User avatar
Zini
Posts: 5538
Joined: 06 Aug 2011, 15:16

Re: MW-Subsystem refactoring (Task #107)

Post by Zini »

That is just the default setting, right? At least on the command line this can be adjusted on a case-by-case basis via the -j switch (very useful when you often do several tasks in parallel and want to control how much CPU power the OpenMW build process consumes). I assume MSVC must have something similar.
User avatar
gus
Posts: 390
Joined: 11 Aug 2011, 15:41

Re: MW-Subsystem refactoring (Task #107)

Post by gus »

That is just the default setting, right?
Yes. Changing it is very easy.
User avatar
gus
Posts: 390
Joined: 11 Aug 2011, 15:41

Re: MW-Subsystem refactoring (Task #107)

Post by gus »

Some of your changes with the dialogue manager are causing problems on windows (hope i post in the right topic).
Visual Studio is complaining in every dialogue related file because std::deque use a non defined type 'MWDialogue::StampedJournalEntry'.

I think the problems come from line 28 of journal.h but i'm not sure.
User avatar
Zini
Posts: 5538
Joined: 06 Aug 2011, 15:16

Re: MW-Subsystem refactoring (Task #107)

Post by Zini »

Got notified about this problem yesterday by another MSVC user. I used to complain about gcc error messages, but now I am convinced that MSVC's error messages suck even more. I wasn't able to pinpoint the problem and had to resolve to the brute force method (including the headers in the interface header instead of using forward declarations). The fix is in master now.
User avatar
scrawl
Posts: 2152
Joined: 18 Feb 2012, 11:51

Re: MW-Subsystem refactoring (Task #107)

Post by scrawl »

Could you merge decouple branch into master, before I start rewriting the input system?

Edit: Just merged it locally in my branch, should be fine I guess
Post Reply