This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: Backed it out (Improvements to section to segment mapping)
- To: Nick Clifton <nickc at cygnus dot com>
- Subject: Re: Backed it out (Improvements to section to segment mapping)
- From: "H . J . Lu" <hjl at valinux dot com>
- Date: Thu, 3 Aug 2000 12:55:18 -0700
- Cc: ian at zembu dot com, binutils at sourceware dot cygnus dot com
- References: <200008031944.MAA31609@elmo.cygnus.com>
On Thu, Aug 03, 2000 at 12:44:21PM -0700, Nick Clifton wrote:
> Hi H.J.,
>
> : I backed out your patch since it breaks stripping dyamic binaries
> : under Linux/ia32.
>
> Please do not do this. If there is a problem with a checked in patch
> please report it to the list. If we (those interested in binutils)
> then decide that the problem cannot be fixed, or that the patch was
> basically wrong, then the patch can be backed out, but not before.
>
I agree with you in general if a problem doesn't effect most people
in normal use. However this particular bug seems to affect all ELF
targets. The problem is I spent almost a day trying to figure out why
my machine stopped working. Since I changed quite a few things from
glibc, kernel to gcc, it took me a while to track it down to binutils.
I'd like to see a semi stable binutls in CVS. That is why I also
submitted a testcase. If one wants one particular feature, it doesn't
mean everyone who works on binutls should also suffer.
> As it turns out my patch was basically OK, it just had a bug in it. I
> have now fixed this bug and restored the code (with the bug fix
> applied).
>
Thanks.
--
H.J. Lu (hjl@gnu.org)