Re: Linux-2.6.13-rc7

From: Al Viro
Date: Wed Aug 24 2005 - 14:13:03 EST


On Wed, Aug 24, 2005 at 11:43:51AM -0700, Paul Jackson wrote:
> Al Viro wrote:
> > ... breaks ppc64 since there we have node_to_cpumask() done as inlined
> > function, not a macro. So we get __first_cpu(&node_to_cpumask(...),...),
> > with obvious consequences.
>
> I sent a patch for this a few hours ago, thanks to Paul Mackerras's report:
>
> [PATCH 2.6.13-rc6] cpu_exclusive sched domains build fix
>
> It just makes a local copy of the cpumask_t in a local variable on the stack.
>
> I'm still a couple of hours from actually verifying that ppc64 builds with
> this - due to unrelated confusions on my end. Perhaps you or Mackerras will
> report in first, to verify if this patch works as advertised.

It does, no (build) regressions. BTW, tree is not far from allmodconfig
buildable on a bunch of targets now - yesterday pile of fixes was about
half of the set needed for that. Most of the remaining stuff is for
m68k (and applies both to Linus' tree and m68k CVS); I'll send that today
and if Geert ACKs them, we will be _very_ close to having 2.6.13 build
out of the box on the following set:
alpha, amd64, arm (RPC and versatile being tracked), i386, ia64, m32r,
m68k (!SUN3), ppc (6xx, 44x, chestnut being tracked), ppc64, sparc,
sparc64, s390, s390x, uml-i386, uml-amd64.

All of these - with allmodconfig, alpha, amd64 and i386 being tracked
separately as SMP and UP. Missing targets:
frv: need newer toolchain on build box
mips, parisc: need out-of-tree patches
v850, m68knommu: gcc gives ICE on attempt to build cross-toolchain
h8300: binutils in FC4 doesn't know what to do with that target,
have not tried that on sarge yet.
sh, sh64: need kernel headers that would make glibc happy enough
to build libc headers for that puppy; I don't have them
cris, xtensa: haven't looked into those
arm26: needs gcc3 since gcc4 had dropped that target; I might take
a look into that on a sarge-based build box someday.

sun3 is seriously broken and I doubt that we'll see any takers for testing
2.6 on those anyway ;-)

A bunch of arm and ppc subarchitectures are not covered yet - I can add those
to build setup, just give me a list in order of preference. Or ask me how
to set up a cross-build farm of your own...
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/