autoconf
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Solving the config.h nightmare ?


From: Daniel Reed
Subject: Re: Solving the config.h nightmare ?
Date: Sun, 23 Apr 2006 04:26:05 -0400 (EDT)

On Sat, 22 Apr 2006, Russ Allbery wrote:
Ralf Wildenhues <address@hidden> writes:
* Russ Allbery wrote on Sun, Apr 23, 2006 at 03:58:52AM CEST:
Oh, the ones that can conflict should not appear in the installed header
_at all_, but rather be recomputed by the Autoconf macro that goes along
with the package, to use by the client package.  (At least for the
generic ones I'd recommend that; for the other ones renaming is fine.)
Hm... I'm not sure that's a good idea.

Once you've compiled a library, the Autoconf results are now part of the
ABI, particularly such things as the size of data types.  If something
changes on the system to change the Autoconf results and a package
recomputes those values with different results, but you don't recompile
the library, you now get ABI breakage and bus errors and segfaults galore.

Not if the API has been abstracted adequately. Data types that are used internally should never *need* to be used as part of the API, and hence their changing definitions between library build time and dependent software build time should not affect ABI in any way.

--
Daniel Reed <address@hidden>      http://shell.n.ml.org/n/        
http://naim.n.ml.org/
wc: write error: Broken pipe




reply via email to

[Prev in Thread] Current Thread [Next in Thread]