* bfd assertion bug found
@ 1995-11-27 20:37 David S. Miller
1995-11-27 23:04 ` Ian Lance Taylor
0 siblings, 1 reply; 3+ messages in thread
From: David S. Miller @ 1995-11-27 20:37 UTC (permalink / raw)
To: gas2
It seems that the linker interprets arguments differently in 2.6,
if I specify '-N' to the linker in 2.5.2 it produces a static final
object, in 2.6 it sets config.dynamic_link to true. This is what
causes trouble when relocations are being scanned.
Later,
David S. Miller
davem@caip.rutgers.edu
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: bfd assertion bug found
1995-11-27 20:37 bfd assertion bug found David S. Miller
@ 1995-11-27 23:04 ` Ian Lance Taylor
1995-11-28 0:08 ` David S. Miller
0 siblings, 1 reply; 3+ messages in thread
From: Ian Lance Taylor @ 1995-11-27 23:04 UTC (permalink / raw)
To: davem; +Cc: gas2
Date: Mon, 27 Nov 1995 23:37:29 -0500
From: "David S. Miller" <davem@caip.rutgers.edu>
It seems that the linker interprets arguments differently in 2.6,
if I specify '-N' to the linker in 2.5.2 it produces a static final
object, in 2.6 it sets config.dynamic_link to true. This is what
causes trouble when relocations are being scanned.
It sounds like a bug that -N does not cause a static link. However, I
do not understand how it could produce the symptoms you are reporting.
I have no trouble doing a dynamic link with -N--the final output does
not run, since the dynamic information is misplaced, but the link does
succeed without any assertion errors. I think you must be seeing two
different problems.
Ian
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: bfd assertion bug found
1995-11-27 23:04 ` Ian Lance Taylor
@ 1995-11-28 0:08 ` David S. Miller
0 siblings, 0 replies; 3+ messages in thread
From: David S. Miller @ 1995-11-28 0:08 UTC (permalink / raw)
To: ian; +Cc: gas2
Date: Tue, 28 Nov 95 02:04:04 EST
From: Ian Lance Taylor <ian@cygnus.com>
Cc: gas2@cygnus.com
Date: Mon, 27 Nov 1995 23:37:29 -0500
From: "David S. Miller" <davem@caip.rutgers.edu>
It seems that the linker interprets arguments differently in 2.6,
if I specify '-N' to the linker in 2.5.2 it produces a static final
object, in 2.6 it sets config.dynamic_link to true. This is what
causes trouble when relocations are being scanned.
It sounds like a bug that -N does not cause a static link. However, I
do not understand how it could produce the symptoms you are reporting.
I have no trouble doing a dynamic link with -N--the final output does
not run, since the dynamic information is misplaced, but the link does
succeed without any assertion errors. I think you must be seeing two
different problems.
And whats even stranger is that I used binutils-2.6 until today with
no problems, but I made quite some extensive changes to my sources
before I did the first build which failed, I must be tickling an
obscure bug of some sort.
Later,
David S. Miller
davem@caip.rutgers.edu
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~1995-11-28 0:08 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1995-11-27 20:37 bfd assertion bug found David S. Miller
1995-11-27 23:04 ` Ian Lance Taylor
1995-11-28 0:08 ` David S. Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).