[patch] Drop support for non-cfi capable assemblers

Hal Finkel hfinkel at anl.gov
Sun Mar 23 14:17:20 PDT 2014


----- Original Message -----
> From: "Brad Smith" <brad at comstyle.com>
> To: "Hal Finkel" <hfinkel at anl.gov>
> Cc: "Rafael Espíndola" <rafael.espindola at gmail.com>, "llvm-commits" <llvm-commits at cs.uiuc.edu>
> Sent: Sunday, March 23, 2014 3:31:59 PM
> Subject: Re: [patch] Drop support for non-cfi capable assemblers
> 
> On 23/03/14 4:29 PM, Hal Finkel wrote:
> > ----- Original Message -----
> >> From: "Brad Smith" <brad at comstyle.com>
> >> To: "Rafael Espíndola" <rafael.espindola at gmail.com>,
> >> "llvm-commits" <llvm-commits at cs.uiuc.edu>
> >> Sent: Sunday, March 23, 2014 3:18:29 PM
> >> Subject: Re: [patch] Drop support for non-cfi capable assemblers
> >>
> >> On 21/03/14 2:40 PM, Rafael Espíndola wrote:
> >>> Having the MCStreamer interface feed back to itself via
> >>> EmitDwarfAdvanceLineAddr and EmitDwarfAdvanceFrameAddr to support
> >>> non-cfi assemblers is one of my favorite hacks, but it might no
> >>> longer
> >>> be necessary.
> >>>
> >>> In llvm, MC itself supports cfi since the end of 2010 (support
> >>> started
> >>> in r119972, is reported in the 2.9 release notes).
> >>>
> >>> In binutils the support has been around for much longer. It looks
> >>> like
> >>> support started to be added in May 2003. It is available in 2.15
> >>> 4000400 (31-Aug-2011, 2.14 is from 12-Jun-2003).
> >>
> >> Not unless IAS is made default on PowerPC and SPARC.
> >
> > Really? Do you support configurations on which this is relevant?
> 
> Yes.

Could we activate the integrated assembler by default on these systems? I believe that in general, while we're missing instruction definitions for a number of instructions supported by newer assemblers, we should have definitions for essentially all instructions supported by assemblers old enough not to have CFI support.

 -Hal

> 
> 
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
> 
> 

-- 
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory




More information about the llvm-commits mailing list