public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* gcc-3.4.0 patches for 2.4.27?
@ 2004-04-28 13:36 Mikael Pettersson
  2004-04-28 14:14 ` Marcelo Tosatti
  2004-05-01 14:00 ` Denis Vlasenko
  0 siblings, 2 replies; 3+ messages in thread
From: Mikael Pettersson @ 2004-04-28 13:36 UTC (permalink / raw)
  To: marcelo.tosatti; +Cc: linux-kernel

Marcelo,

Will you accept patches allowing gcc-3.4.0 to compile
2.4.27 or not? I can understand if you want to be
conservative and not change _anything_ if you don't have to.

gcc-3.4.0 errors out in 2.4.27-pre1 due to (a) inconsistent
FASTCALL declarations, (b) uninlinable inlines, and (c)
-funit-at-a-time which seems to leave unresolved calls to
strcpy() around [gcc optimises sprintf "%s" to strcpy()].
There are also tons of warnings due to cast-as-lvalue
and "+m" asm() constraints.

I currently have a 40KB+ patch for 2.4.27-pre1 which works
for me on i386, UP and SMP. The changes are all backports
from 2.6. I'll do x86_64 and ppc(32) in a few days.

http://www.csd.uu.se/~mikpe/linux/patches/2.4/patch-gcc340-fixes-2.4.27-pre1
is the location of the current version.

/Mikael

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

* Re: gcc-3.4.0 patches for 2.4.27?
  2004-04-28 13:36 gcc-3.4.0 patches for 2.4.27? Mikael Pettersson
@ 2004-04-28 14:14 ` Marcelo Tosatti
  2004-05-01 14:00 ` Denis Vlasenko
  1 sibling, 0 replies; 3+ messages in thread
From: Marcelo Tosatti @ 2004-04-28 14:14 UTC (permalink / raw)
  To: Mikael Pettersson; +Cc: linux-kernel

On Wed, Apr 28, 2004 at 03:36:47PM +0200, Mikael Pettersson wrote:
> Marcelo,
> 
> Will you accept patches allowing gcc-3.4.0 to compile
> 2.4.27 or not? I can understand if you want to be
> conservative and not change _anything_ if you don't have to.
> 
> gcc-3.4.0 errors out in 2.4.27-pre1 due to (a) inconsistent
> FASTCALL declarations, (b) uninlinable inlines, and (c)
> -funit-at-a-time which seems to leave unresolved calls to
> strcpy() around [gcc optimises sprintf "%s" to strcpy()].
> There are also tons of warnings due to cast-as-lvalue
> and "+m" asm() constraints.
> 
> I currently have a 40KB+ patch for 2.4.27-pre1 which works
> for me on i386, UP and SMP. The changes are all backports
> from 2.6. I'll do x86_64 and ppc(32) in a few days.
> 
> http://www.csd.uu.se/~mikpe/linux/patches/2.4/patch-gcc340-fixes-2.4.27-pre1
> is the location of the current version.

Hi Mikael,

The patch looks alright for me, but I better see reviews from 
other people.

Since v2.4 is in feature freeze, I have a reason not to apply it.

I'm not sure. Lets wait a while and see.

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

* Re: gcc-3.4.0 patches for 2.4.27?
  2004-04-28 13:36 gcc-3.4.0 patches for 2.4.27? Mikael Pettersson
  2004-04-28 14:14 ` Marcelo Tosatti
@ 2004-05-01 14:00 ` Denis Vlasenko
  1 sibling, 0 replies; 3+ messages in thread
From: Denis Vlasenko @ 2004-05-01 14:00 UTC (permalink / raw)
  To: Mikael Pettersson, marcelo.tosatti; +Cc: linux-kernel

On Wednesday 28 April 2004 16:36, Mikael Pettersson wrote:
> Marcelo,
>
> Will you accept patches allowing gcc-3.4.0 to compile
> 2.4.27 or not? I can understand if you want to be
> conservative and not change _anything_ if you don't have to.
>
> gcc-3.4.0 errors out in 2.4.27-pre1 due to (a) inconsistent
> FASTCALL declarations, (b) uninlinable inlines, and (c)
> -funit-at-a-time which seems to leave unresolved calls to
> strcpy() around [gcc optimises sprintf "%s" to strcpy()].

Instead of disabling -funit-at-a-time, maybe hunt down
and replace sprintf("%s")? Or provide non-inlined
instance of strcpy() to link against, that won't waste much space.
--
vda


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

end of thread, other threads:[~2004-05-01 14:00 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-04-28 13:36 gcc-3.4.0 patches for 2.4.27? Mikael Pettersson
2004-04-28 14:14 ` Marcelo Tosatti
2004-05-01 14:00 ` Denis Vlasenko

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox