public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
* About passing -mtune and -march options from GCC to AS
@ 2009-05-26 16:52 Valdimir Volynsky
  2009-05-27  0:33 ` Dave Korn
  0 siblings, 1 reply; 3+ messages in thread
From: Valdimir Volynsky @ 2009-05-26 16:52 UTC (permalink / raw)
  To: binutils

Please, read this thread 
http://gcc.gnu.org/ml/gcc-patches/2009-05/threads.html#01556
What's your opinion?

Сheers,
Vladimir
---
Professional hosting for everyone - http://www.host.ru

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: About passing -mtune and -march options from GCC to AS
  2009-05-26 16:52 About passing -mtune and -march options from GCC to AS Valdimir Volynsky
@ 2009-05-27  0:33 ` Dave Korn
  2009-06-15 14:32   ` Nick Clifton
  0 siblings, 1 reply; 3+ messages in thread
From: Dave Korn @ 2009-05-27  0:33 UTC (permalink / raw)
  To: Valdimir Volynsky; +Cc: binutils

Valdimir Volynsky wrote:
> Please, read this thread
> http://gcc.gnu.org/ml/gcc-patches/2009-05/threads.html#01556
> What's your opinion?
> 
> Сheers,
> Vladimir

  To sum up the discussion a bit for those who don't want to follow the link,
the question is how to communicate the options like -march and -mtune to the
assembler, without breaking anything if it turns out not to be GAS.  The ideas
suggested have included an environment variable (which as I understand it we
are trying to avoid as a general matter of policy), adding a magic comment or
ident string that GAS could recognize and other assemblers ignore, or deciding
whether or not to do so based on the outcome of autoconf tests and
--with-(gnu-)as at configure time.  Arguments for or against any particular
style of doing it, or new and better suggestions, would be eagerly received.

    cheers,
      DaveK

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: About passing -mtune and -march options from GCC to AS
  2009-05-27  0:33 ` Dave Korn
@ 2009-06-15 14:32   ` Nick Clifton
  0 siblings, 0 replies; 3+ messages in thread
From: Nick Clifton @ 2009-06-15 14:32 UTC (permalink / raw)
  To: Dave Korn, Valdimir Volynsky; +Cc: binutils

Hi Dave, Hi Vladimir,

>   To sum up the discussion a bit for those who don't want to follow the link,
> the question is how to communicate the options like -march and -mtune to the
> assembler, without breaking anything if it trns out not to be GAS.  The ideas
> suggested have included an environment variable (which as I understand it we
> are trying to avoid as a general matter of policy), adding a magic comment or
> ident string that GAS could recognize and other assemblers ignore, or deciding
> whether or not to do so based on the outcome of autoconf tests and
> --with-(gnu-)as at configure time.  Arguments for or against any particular
> style of doing it, or new and better suggestions, would be eagerly received.

My suggestion: use autoconf.  This is the method that is already in use 
for solving other, similar situations.  Plus it has the least hack-ish 
feel to it.

Cheers
   Nick


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2009-06-15 14:32 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-05-26 16:52 About passing -mtune and -march options from GCC to AS Valdimir Volynsky
2009-05-27  0:33 ` Dave Korn
2009-06-15 14:32   ` Nick Clifton

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).