Re: fpu controlword altered on windows
Chris Robinson <[email protected]> Sat, 02 Jul 2011 03:54:31 -0700
| Newsgroups | gmane.comp.lib.openal |
|---|---|
| Message-ID | <2705962.GRjjnNya0l@kittycat> |
On Saturday, July 02, 2011 10:07:44 AM Xavier Bouchoux wrote: > nope. > the value returned by _controlfp() is the new value *after* applying the > change. So the saved value that is restored at function exit is actually > the changed value... > > http://msdn.microsoft.com/en-us/library/e9b52ceh%28VS.71%29.aspx I see. Going off of Wine's implementation, it returns the previous value. I'll have to talk to them about that. > yeah, that was a bit disturbing indeed. I had to double check the code > to see what enums I was supposed to use. > And most problematic, it means that when using the extension, you need > to querry for the extension in ALC and then expect there are new > enums/functions in both AL and ALC. (with no "correct" current context > for AL, as the enums are needed to create it) FWIW, the enums values are constant (or will be, when the extension is finished). You don't need to use al[c]GetEnumValue to get the actual value for an enum, except in cases where an extension abuses that to tell what exactly is supported (but even then, supported enums have a specific value; it's just a hacky way to query support). The loopback extension has a proper method to query supported formats, so you can just use pre-defined macros. _______________________________________________ Openal mailing list [email protected] http://opensource.creative.com/mailman/listinfo/openal