Re: [Bug 514579] Re: gcc-4.4 cross build for m68k on i486 fails
On Fri, 18 Jun 2010, Thorsten Glaser wrote:
> (Donʼt Cc: me please, I read the debian-68k@ list.)
>
> Finn Thain dixit:
>
> >Hopefully, all those targets available to the etch-m68k compilers. Not
> >sure whether fidoa was one of those?
>
> Hrm. As I wrote in <Pine.BSM.4.64L.1004191952070.24486@herc.mirbsd.org>
> it didnʼt build (due to inline asm in glibc (2.7) headers).
By "etch-m68k compilers", I meant gcc 4.1.1-21. It looks like gcc-4.1.1-21
doesn't offer fido support, so I would disable fido support if that will
fix the build.
> >You'll see that I've requested that the patch "m68k-allow-gnu99.diff"
> >be dropped from both 4.4 and 4.5. I hope this does not harm your work
> >on gcc? 4.3 at least, won't be affected.
>
> I donʼt think so, but it might affect systems with libc-2.7 still
> installed
I've never heard of any. Etch-m68k only offered glibc-2.3.6. I think
glibc-2.5 may have been used on some buildd machines. If 2.5 can't be used
with gcc-4.4 (without "m68k-allow-gnu99.diff") then why not simply abandon
glibc 2.5, and replace it with cross-compiled glibc-2.11 packages?
> (although I can probably circumvent that by passing non-default CFLAGS
> everywhere). Unless you have a newer (e)glibc, even when without TLS
> support, to get rid of that?
A "lenny-m68k" doesn't really interest me. Likewise glibc newer than 2.3.6
but lacking NPTL (i.e. 2.7). A lot of work went into eglibc for m68k
around 2.10 and later. I think we should focus on squeeze/sid and leave
linuxthreads behind.
Finn
>
> bye,
> //mirabilos
>
Reply to: