[Bug binutils/11133] gc-sections deletes sections with __start/__stop reference

[hjl dot tools at gmail dot com]

------- Additional Comments From hjl dot tools at gmail dot com  2010-01-07 
13:47 -------
(In reply to comment #7)
> Imagine someone who wants an array of functions (perhaps constructor like) or
> some other object used in their app.  They set up weak references in a named
> section, then access the array using the __start_* symbol.  The array itself
> isn't referenced so references there don't cause the functions to be kept.
> Oh..  My array won't be kept either.  I didn't think of that earlier..  Hmm, 
> the
> named section would need to be a debug section, and then it doesn't matter if
> there is a change in the way __start_* affects --gc-sections.
> 
> So my concern that HJ's idea might affect some existing app is probably not
> valid.  I still don't particularly like a reference to __start_somesection
> meaning that we should keep all "somesection" input sections.  It would be
> better if there were a more explicit way to mark input sections.  I also note
> that HJ's patch is incorrect since marking a section with gc_mark and 
> returning
> NULL from gc_mark_hook will result in no mark phase recursion for that 
> section.

I will try a different approach.



-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hjl dot tools at gmail dot
                   |                            |com


http://sourceware.org/bugzilla/show_bug.cgi?id=11133

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.