[PATCH] [GNU] Implement --enable-new-dtags/--disable-new-dtags

Saleem Abdulrasool compnerd at compnerd.org
Mon Apr 6 18:51:58 PDT 2015


On Mon, Apr 6, 2015 at 6:22 PM, Ed Maste <emaste at freebsd.org> wrote:

> On 6 April 2015 at 17:23, Joerg Sonnenberger <joerg at britannica.bec.de>
> wrote:
> > On Mon, Apr 06, 2015 at 02:15:58PM -0700, Rui Ueyama wrote:
> >> Anyways, my point is that the change to the linker for this feature
> might
> >> be trivial, so it'd be easy to understand what is done here. But it
> would
> >> not help readers of the code understand why we want to do that. So I
> wanted
> >> to add a comment on that (a brief comment should suffice).
> >
> > I fully agree on that. First time I saw the option was like "WTF do they
> > need this?"
>
> Good point. Do you think we should go further and either just use
> DT_RUNPATH when we know the system's rtld supports it, or set both
> DT_RUNPATH and DT_RPATH in all cases?
>

No, setting both would mean that you still have DT_RPATH, which means that
LD_LIBRARY_PATH is no longer honored.  I think that would be detrimental to
the use of the flag in the first place.


> The "broken" DT_RPATH behaviour is as specified in gabi41.pdf, but I'm
> not sure that anyone really wants that. It seems a bit unfortunate to
> have the sensible behaviour hidden behind an option.
>

I suppose we could add yet another flag to change to the older behavior if
needed.


> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>



-- 
Saleem Abdulrasool
compnerd (at) compnerd (dot) org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150406/1087ae98/attachment.html>


More information about the llvm-commits mailing list