Re: [Laptopkernel-devel] non-compile problem with gcc 3.3

[Anders Karlsson]

Good Afternoon,

On Tue, 2003-06-03 at 14:37, Sebastian Henschel wrote:
> Received at 2003-05-31 / 15:36 by Anders Karlsson:

[snip]

> > I thought there was someone that figured out that it was swsusp that
> > modified the file that breaks gcc. 
> 
> it really seems like that. if i disable swsusp, it compiles fine (though
> it was only vanilla rc6 + acpi + swsusp).

I have not yet been able to test out the laptop patch as I only have GCC
3.3. I am running 2.4.21-rc6-ac1 at the moment, trying to get a Thinkpad
X31 working properly.

In the -rc6-ac1 patch there was a refresh of ACPI, although if I try and
boot the X31 using that, it will hang the kernel in the boot sequence
doing something on the PCI bus. (Not got the full error message.)

> > You might want to report back to the
> > gcc team that the compiler breaks on the code and you might also want to
> 
> they already know that... in a way... see my first post, the bug report
> on gcc-bugzilla ist just the very essence of the error. :)

At least they know about it. I noticed a post on LKML requesting the
next -rc kernel using inline if GCC >= 3.3 is detected. Might that help
in this instance?

> > Either way, the right answer it to get things working with gcc 3.3 as
> > that compiler is not going to go away, quite the opposite. :)
> > 
> > Anyone disagree?
> 
> well, i suppose gcc 3.3.1 should be released with a fix as soon as
> possible. :)

Would be nice. :) Not to mention down right handy. ;-)

Regards,

-- 
Anders Karlsson <address@hidden>
Trudheim Technology Limited

signature.asc
Description: This is a digitally signed message part