[RFC/commit] Getting rid of tm-rs6000.h (PROCESS_LINENUMBER_HOOK)

[Mark Kettenis]

> X-Spam-Check-By: sourceware.org
> Date: Tue, 18 Sep 2007 16:14:48 -0700
> From: Joel Brobecker <brobecker@adacore.com>
> Content-Disposition: inline
> Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm
> Sender: gdb-patches-owner@sourceware.org
> X-UTwente-MailScanner-Information: Scanned by MailScanner. Contact helpdesk@ITBE.utwente.nl for more information.
> X-UTwente-MailScanner: Found to be clean
> X-UTwente-MailScanner-From: gdb-patches-return-50837-m.m.kettenis=alumnus.utwente.nl@sourceware.org
> X-Spam-Status: No
> X-XS4ALL-DNSBL-Checked: mxdrop3.xs4all.nl checked 130.89.2.13 against DNS blacklists
> X-Virus-Scanned: by XS4ALL Virus Scanner
> X-XS4ALL-Spam-Score: 0.0 () DK_POLICY_SIGNSOME
> X-XS4ALL-Spam: NO
> X-CNFS-Analysis: v=1.0 c=1 a=5JwuhGrhyVAA:15 a=wfEPs4U+lFiAFA7pH+G9TA==:17 a=g1VYcl7afDL033UujJkA:9 a=oUFSmlPX7ia-zx2J8jsA:7 a=z6PKJkvbTO3t5axyC61YUgDJBJkA:4 a=b8hG5vVbyAkA:10 a=-y93oKWw6PMAJfIIPkkA:9 a=-0FndLuy2frk9IRV2UEA:7 a=zN-x5g51XQPJfcUlamfUbFjiEO0A:4 a=WZm9NgPvBkYA:10 a=5WZzfXpOq_gA:10 a=mDV3o1hIAAAA:8 a=vsAYgGIN1YtluYDjcCgA:9 a=al0t3wDx9O-uvKBlVngA:7 a=wYDCVDzEO6LVr6XnfH29QgUVgCgA:4 a=w2AolMDZV5cA:10 a=bClMUFNdNt6jSyT-7-0A:9 a=ZmXWUAE7x0GZgUYmYHgA:7 a=l2b_aCW9KmR25YviF7on3GO03rkA:4
> Envelope-To: mark.kettenis@xs4all.nl
> X-UIDL: 1190157346._smtp.mxdrop3.27993,S=16710
> 
> 
> --lrZ03NoBR/3+SXJZ
> Content-Type: text/plain; charset=us-ascii
> Content-Disposition: inline
> 
> Hello,
> 
> There is only one macro left in tm-rs6000.h, which is:
> 
>     /* In xcoff, we cannot process line numbers when we see them. This is
>        mainly because we don't know the boundaries of the include files. So,
>        we postpone that, and then enter and sort(?) the whole line table at
>        once, when we are closing the current symbol table in end_symtab(). */
>     
>     #define PROCESS_LINENUMBER_HOOK()       aix_process_linenos ()
>     extern void aix_process_linenos (void);
> 
> The AIX port is the only port that defines this macro, and it's only
> used in one place: buildsym.c:end_symtab(). I think the most appropriate
> way to define this notion is as a gdbarch method, set only in the case
> of AIX. This is what this patch does.

Hmm, if this is something that is purely for a specific object/debug format,
I think the gdbarch vector is the wrong place to put it.

Mark