This is the mail archive of the mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: cygwin gcc 3.4 and cygwin

I thought I had a legitimate concern and question, not one that deserved
"just" a sarcastic response.

It would be easy to accendentally release things for Cygwin that are ABI
incompatible with Cygwin's gcc.

Why do we persist this way?  I would be happy to do the necessary leg work
to make vanilla gcc the same as Cygwin gcc.

With Redhat's influence on the free software world, I would think,
mistakenly, I guess, that Cygwin local patches would be short-lived,
migrating relatively quickly back to the official sources.  What is wrong
with this assumption?

Just trying to understand and help out, not cause problems or insult.


Christopher Faylor wrote:

FWIW, I build cygwin itself with an unpatched version of gcc several times
a day.

Brian Ford wrote:

Gee.  I hope Cygwin, and anything else you compile with that compiler
for Cygwin, does not have structures containing doubles.  Without
MASK_ALIGN_DOUBLE in TARGET_SUBTARGET_DEFAULT of gcc/config/i386/cygwin.h,
the standard Cygwin compiler and vanilla gcc are ABI incompatible.

Doesn't this seem bad?

Christopher Faylor wrote:

Oh, it seems horrific.  Now I won't be able to sleep at night.  Thanks a

Brian Ford
Senior Realtime Software Engineer
VITAL - Visual Simulation Systems
FlightSafety International
Phone: 314-551-8460
Fax:   314-551-8444

Unsubscribe info:
Bug reporting:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]