[RFC PATCH, i386]: Use %r15 for REAL_PIC_OFFSET_TABLE_REGNUM on x86_64

[Mike Frysinger]

On Tuesday 25 December 2012 14:12:09 Uros Bizjak wrote:
> On Tue, Dec 25, 2012 at 6:54 AM, Mike Frysinger <vapier@gentoo.org> wrote:
> > On Monday 24 December 2012 17:26:47 Leif Ekblad wrote:
> >> In the case of cpuid, the code is hardly performance sensitive, and
> >> probably runs only at startup. An alternative solution for the broken
> >> code here is to move the result from rbx to another register, and to
> >> save/restore rbx. Currently, this is the only place in libgcc and
> >> newlib affected by this problem.
> > 
> > it's not a question of performance.  i can't remember how many various
> > projects i've had to tweak the inline asm code to work with __PIC__
> > (either because it's going into shared library code or it's being built
> > as a PIE). Andi's point is now we have to redo all of that work a 2nd
> > time and handle two different cases depending on gcc version ?  it'd be
> > a _lot_ better if gcc were intelligent and end users didn't have to code
> > crap like stuffing %ebx somewhere temporarily.
> 
> Please note that we are not talking about 32bit code, where this would
> make a difference, but for 64bit targets with -mcmodel=medium and
> -mcmodel=large exclusively. The default x64_64 -mcmodel=small doesn't
> use PIC register, other code models are rarely used, so I sincerely
> doubt that any %rbx workarounds were needed in the past for x86_64.

i'm aware.  the comment still applies.  you're breaking asm code that used to 
work because the gcc inline asm code isn't intelligent enough (currently) to 
transparently handle the PIC register for the user.
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20121225/36883b4a/attachment.sig>