Patch for new DW_OP

[Tambay, Pallavi]

Hi Nick,

Thanks for applying the patch.
PGI does not plan to submit this extension to the DWARF committee
because
the function of the DW_OP_PGI_omp_thread_num operation is
implementation-defined, i.e, it is dependent on the PGI implementation
of OpenMP.

Pallavi

-----Original Message-----
From: Nick Clifton [mailto:nickc@redhat.com] 
Sent: Monday, March 03, 2008 2:21 AM
To: Tambay, Pallavi
Cc: binutils@sourceware.org; Meissner, Michael; rajagopal, dwarak;
steven.nakamoto@pgroup.com
Subject: Re: Patch for new DW_OP

Hi Pallavi,

> This patch adds the following dwarf operation as a PGI
> (STMicroelectronics) extension.

As a matter of interest have PGI submitted this extension to the DWARF 
committee for consideration ?

> elf/ChangeLog
> 2008-02-29  Pallavi Tambay  <pallavi.tambay@amd.com>
> 
> 	* dwarf2.h: (enum dwarf_location_atom): Add new DW_OP,
> 	DW_OP_PGI_omp_thread_num.

Approved and applied.

I also applied the attached patch to allow objdump and readelf to decode

this op.

Cheers
   Nick

binutils/ChangeLog
2008-03-03  Nick Clifton  <nickc@redhat.com>

	* dwarf.c (decode_location_expression): Handle
	DW_OP_PGI_omp_thread_num.