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 Wed, Jun 30, 2004 at 11:17:47PM +0930, Alan Modra wrote: > On Tue, Jun 29, 2004 at 11:13:32AM -0700, H. J. Lu wrote: > > On Tue, Jun 29, 2004 at 12:56:06AM -0700, H. J. Lu wrote: > > > On Mon, Jun 28, 2004 at 06:22:42PM -0700, H. J. Lu wrote: > > > > section_already_linked in ldlang.c has some code to deal with PE's > > > > comdat, which isn't use by any other targets. Also there is > > > > bfd_comdat_info in bfd_section for it. I suggest we move > > > > bfd_comdat_info to PE target and move section_already_linked into > > > > bfd target vector. We can have a special section_already_linked > > > > function for PE and a generic one for others. With those changes, we > > > > can remove PE specifc code from linker as well as bfd_section. Any > > > > comments? > > > This is the complete patch. With this patch, I withdraw the section > > identifier patch: > > I like it, except that it seems to me that you could leave > section_already_linked where it is, and just provide a bfd xvec > predicate. That would save some code duplication, which helps > in maintaining the linker. > The problem is that piece of code can be very much format specific. I have a follow-on patch to fix http://sources.redhat.com/bugzilla/show_bug.cgi?id=246 My patch needs to access ELF backend. To do it in the linker, I will need to add another field to bfd xvec and we have to do more checking on input to see if it is ELF. I think the code duplication is very minimum here. Here is an updated patch to fix a few problems. H.J.
Attachment:
binutils-comdat-4.patch
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |