This is the mail archive of the binutils@sourceware.cygnus.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]

Re: All those old patches.


Geoff Keating wrote:
> 
> > Date: Tue, 23 May 2000 11:22:59 +0200
> > From: Momchil Velikov <velco@fadata.bg>
> 
> > > Why is the comment
> > >
> > > -      /* Both of these sections are (technically) created by the user
> > > -        putting data in them, so they shouldn't be marked
> > > -        SEC_LINKER_CREATED.
> > >
> > > wrong?
> >
> > They ARE linker created, aren't they ? And LINKER puts data in them ?
> 
> No, the user puts data in them, by writing things like
> 
>     .section ".sdata"
> somelabel:
>     .long 1234
> 
> and the sections need not exist if the user never does this.

I don't think so. The linker puts data in them, namely the implicit
pointers for R_PPC_EMB_SDAI16 and R_PPC_EMB_SDA2I16 relocs.
Setting the SEC_LINKER_CREATED flag would cause
_bfd_elf_create_linker_section
to create a new section despite the eventual existance of a user section
with the same name. I think "Both of these sections" means THESE two
sections, not any other section with the same name. 

This is how I understand the things and that's why I changed the
comment.

Regards,
-velco

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]