* [2.6.15-rc2-mm1] Disabled flatmem on ppc32? (ARCH=powerpc)
@ 2005-11-27 1:46 Kyle Moffett
2005-11-27 13:43 ` Andy Whitcroft
0 siblings, 1 reply; 3+ messages in thread
From: Kyle Moffett @ 2005-11-27 1:46 UTC (permalink / raw)
To: LKML Kernel; +Cc: Andrew Morton, Benjamin Herrenschmidt, Andy Whitcroft
There is a Kconfig problem for ppc32 in the latest -mm kernel. It
seems that somehow the Kconfig logic for selecting memory models
under ARCH=powerpc doesn't quite get it right for standard flatmem
ppc32 systems. When I look at the memory model selection, I only see
sparsemem, whereas on a normal -rc2 kernel, I can see both flatmem
and sparsemem. This somehow triggers a #error where the number of
reserved bits is less than the number necessary for the sparsemem
layout (because we're on a 32-bit arch without the address space for
sparsemem?).
Cheers,
Kyle Moffett
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [2.6.15-rc2-mm1] Disabled flatmem on ppc32? (ARCH=powerpc)
2005-11-27 1:46 [2.6.15-rc2-mm1] Disabled flatmem on ppc32? (ARCH=powerpc) Kyle Moffett
@ 2005-11-27 13:43 ` Andy Whitcroft
2005-11-27 14:32 ` Kyle Moffett
0 siblings, 1 reply; 3+ messages in thread
From: Andy Whitcroft @ 2005-11-27 13:43 UTC (permalink / raw)
To: Kyle Moffett
Cc: LKML Kernel, Andrew Morton, Benjamin Herrenschmidt,
Anton Blanchard
Kyle Moffett wrote:
> There is a Kconfig problem for ppc32 in the latest -mm kernel. It
> seems that somehow the Kconfig logic for selecting memory models under
> ARCH=powerpc doesn't quite get it right for standard flatmem ppc32
> systems. When I look at the memory model selection, I only see
> sparsemem, whereas on a normal -rc2 kernel, I can see both flatmem and
> sparsemem. This somehow triggers a #error where the number of reserved
> bits is less than the number necessary for the sparsemem layout
> (because we're on a 32-bit arch without the address space for sparsemem?).
I suspect we have neglected to add back the default for FLATMEM on 32
bit when we allowed FLATMEM to be disabled for 64 bit. Will go and look
at it. Thanks for the report.
-apw
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [2.6.15-rc2-mm1] Disabled flatmem on ppc32? (ARCH=powerpc)
2005-11-27 13:43 ` Andy Whitcroft
@ 2005-11-27 14:32 ` Kyle Moffett
0 siblings, 0 replies; 3+ messages in thread
From: Kyle Moffett @ 2005-11-27 14:32 UTC (permalink / raw)
To: Andy Whitcroft
Cc: LKML Kernel, Andrew Morton, Benjamin Herrenschmidt,
Anton Blanchard
On Nov 27, 2005, at 08:43:40, Andy Whitcroft wrote:
> Kyle Moffett wrote:
>
>
>> There is a Kconfig problem for ppc32 in the latest -mm kernel. It
>> seems that somehow the Kconfig logic for selecting memory models
>> under ARCH=powerpc doesn't quite get it right for standard
>> flatmem ppc32 systems. When I look at the memory model
>> selection, I only see sparsemem, whereas on a normal -rc2 kernel,
>> I can see both flatmem and sparsemem. This somehow triggers a
>> #error where the number of reserved bits is less than the number
>> necessary for the sparsemem layout (because we're on a 32-bit arch
>> without the address space for sparsemem?).
>>
>
> I suspect we have neglected to add back the default for FLATMEM on
> 32 bit when we allowed FLATMEM to be disabled for 64 bit. Will go
> and look at it. Thanks for the report.
>
When you get it figured out, could you CC the patch to me? I'd like
to do some testing -mm1 and this is a bit of a showstopper :-D. Thanks!
Cheers,
Kyle Moffett
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2005-11-27 14:32 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-11-27 1:46 [2.6.15-rc2-mm1] Disabled flatmem on ppc32? (ARCH=powerpc) Kyle Moffett
2005-11-27 13:43 ` Andy Whitcroft
2005-11-27 14:32 ` Kyle Moffett
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.