Deprecating ntea

Christopher Faylor cgf-use-the-mailinglist-please@cygwin.com
Tue Feb 27 15:32:00 GMT 2007


On Tue, Feb 27, 2007 at 03:30:18PM +0100, Corinna Vinschen wrote:
>On Feb 27 09:03, Christopher Faylor wrote:
>> Off the top of my head things I'd like to fix would be:
>> 
>> signal mask
>> termios structure
>
>Wouldn't that be easily coverable by using the usual method of
>recognizing the application build?

That's how they are handled now.  I'd like to get rid of runtime checks.

>I did that for the new struct ifreq lately, for instance.  Or, as in
>case of the 32->64 bit transition, what about creating new entry points
>for new applications (sigaction vs.  sigaction64)?

The point of a cygwin2 would be to eliminate workarounds.

>> environ passing
>> user_data passing from program to cygwin
>
>These two I don't quite understand.  Wouldn't the method be just a
>change under the hood, invisible to the applications?

No.  There is code in the built-in stub which does stupid things with
the environment which, I assume, harkens back to when Steve Chamberlain
was trying to figure out Windows.

cgf



More information about the Cygwin-developers mailing list