This is the mail archive of the binutils@sources.redhat.com mailing list for the binutils 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] |
On Mon, May 17, 2004 at 08:41:50PM -0700, H. J. Lu wrote: > On Mon, May 17, 2004 at 06:06:05PM -0700, H. J. Lu wrote: > > On Mon, May 17, 2004 at 02:24:23PM -0700, H. J. Lu wrote: > > > On Sat, May 15, 2004 at 11:09:47AM +0930, Alan Modra wrote: > > > > On Fri, May 14, 2004 at 02:07:37PM -0700, H. J. Lu wrote: > > > > > When there are mixed COMDAT and linkonce inputs, linker doesn't handle > > > > > them gracefully: > > > > > > > > Ewww. Should we even try? I understand that such a patch might be > > > > useful while gcc is emitting both comdat and linkonce, but once you've > > > > completed the change to comdat it shouldn't be necessary. Also, I'm not > > > > really happy with where you have added this code. At least, it is the > > > > wrong place to be discarding duplicate sections. That ought to happen > > > > in ldlang.c:section_already_linked. > > > > > > > > > > This patch implements it. > > > > > > > > > > Here is an update. I should skip checking members of section groups > > for already linked section. > > > > Another update. Should I skip COMDAT group members on the already > linked list? It may happen if a COMDAT group member has the same > section name as a linkonce section. I fixed a small memory leak. H.J.
Attachment:
bfd-comdat-linkonce-6.patch
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |