delorie.com/archives/browse.cgi   search  
Mail Archives: cygwin-developers/1998/05/02/06:45:54

From: cgf AT cygnus DOT com (Christopher Faylor)
Subject: RE: win32 configuration proposal
2 May 1998 06:45:54 -0700 :
Message-ID: <199805021335.JAA06243.cygnus.cygwin32.developers@kramden.cygnus.com>
To: ian AT cygnus DOT com, scox AT cygnus DOT com, sos AT buggy DOT prospect DOT com DOT ru
Cc: cygwin32-developers AT cygnus DOT com, gcc-local AT cygnus DOT com

>Ian Lance Taylor wrote:
>>    4. Mark ix86-mingw32 and ix86-cygwin32 as obsolescent configurations.
>> 
>>    Comments?
>> 
>> Just that I don't see any need to make the ix86-cygwin32 configuration
>> obsolescent.  Some people will always want to develop cygwin32
>> programs, and since we already have a configuration triplet for them I
>> don't see why we need to take it away and force them to use -mcygwin32
>> all the time.
>
>Will I have an ability to set default gcc behavior (cygwin32 or
>mingw32) with some environment variable (like MAKE_MODE for b19 make)?

That's a good question.  Otherwise we'll have to remember to change
the CFLAGS every time we build.

Another thing occurred to me, too.  If we start building applications
without the -mcygwin32 they will not respond to cygwin32 signals and
they will not inherit cygwin32 file descriptors.  That could cause
unexpected behavior.  For instance, sending a SIGTERM to the application
may not cause it to be terminated.

cgf

- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019