This is the mail archive of the
binutils@sourceware.cygnus.com
mailing list for the binutils project.
PATCH: Small fix for elfppc.sc
- To: binutils at sourceware dot cygnus dot com
- Subject: PATCH: Small fix for elfppc.sc
- From: Franz Sirl <Franz dot Sirl-kernel at lauterbach dot com>
- Date: Wed, 13 Oct 1999 00:36:42 +0200
- References: <37AF2370.F1A672D6@cygnus.co.uk> <4.2.1.10.19991012155633.00be7230@mail.lauterbach.com> <3803407C.1813F09@cygnus.co.uk>
Hi,
a patch on the gcc lists reminded me of this small patch I have in my sources
for a while. I discovered it while comparing elf.sc and elfppc.sc and Jonathan
assured me that it is correct :-).
Franz.
* elfppc.sc: Sort constructors
Index: ld/scripttempl/elfppc.sc
===================================================================
RCS file: /cvs/binutils/binutils/ld/scripttempl/elfppc.sc,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 elfppc.sc
--- elfppc.sc 1999/05/03 07:29:08 1.1.1.1
+++ elfppc.sc 1999/10/12 22:41:36
@@ -170,7 +170,7 @@ SECTIONS
*(.data)
${RELOCATING+*(.data.*)}
${RELOCATING+*(.gnu.linkonce.d*)}
- ${CONSTRUCTING+CONSTRUCTORS}
+ ${CONSTRUCTING+SORT(CONSTRUCTORS)}
}
.data1 ${RELOCATING-0} : { *(.data1) }
${RELOCATING+${OTHER_READWRITE_SECTIONS}}
Am Die, 12 Okt 1999 schrieb Jonathan Larmour:
>Franz Sirl wrote:
>>
>> At 15:24 12.10.99 , you wrote:
>> >Yes this makes perfect sense. But the correct place to send it to is
>> >binutils@sourceware.cygnus.com.
>>
>> Thanks for the clarification, so strictly speaking, is this a bug causing
>> possibly wrong behaviour of the code or is it more like a missed optimization?
>
>It is needed to correctly support -finit-priority. If your code doesn't use
>that (and it is disabled by default) then you wouldn't need it. If you *did*
>need it, it wouldn't work and would be a bug, not a missed optimization. So
>do submit it please :-).