This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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: stale fixincludes fixes


* Bruce Korb wrote on Thu, Sep 10, 2009 at 11:35:48PM CEST:
> On Thu, Sep 10, 2009 at 12:49 PM, Ralf Wildenhues wrote:
> > That calls for a cleanup sweep of inclhack.def. ?OK for this followup
> > patch? ?This time tested on i686-pc-linux-gnu only, so I'm driving blind
> > as to whether any of the removed hacks could have fixed a system that is
> > still supported.
> 
> If you are sure the code is obsolete, then just remove it.  I trust you.  :)
> If you have any doubts, just surround it with:
> 
> #ifdef OBSOLETE_CODE
> ..........
> #endif /* OBSOLETE_CODE */

I have checked old Solaris and AIX systems and codesearch, and haven't
found any headers that could match, so I've gone ahead and removed the
code now.

* Bruce Korb wrote on Thu, Sep 10, 2009 at 11:26:07PM CEST:
> One of the things that could easily be done is to fiddle the code a bit
> to record which fixes fire and ask the person building GCC to let some
> obscure email address know that a particular fix fired on a particular
> platform, perhaps marking some "obviously in use" fixes with a "don't
> bother" flag.  If some fix hasn't been reported in after several years,
> it might be a reasonable candidate for removal.  Or, mark the fixes we
> think may be obsolete and emit a message to the GCC builder, "Your
> build required the XXX fix and we're considering obsoleting that fix.
> Please notify xxx@whatever.xxx that you still need this fix."

Yes, that could be a good strategy, too.

Thanks,
Ralf


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