lkml.org 
[lkml]   [2011]   [May]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
DateWed, 25 May 2011 16:37:50 -0700 (PDT)
FromDavid Rientjes <>
SubjectRe: [patch 6/6] slab: only define SLAB_CACHE_DMA for CONFIG_ZONE_DMA
On Wed, 25 May 2011, Heiko Carstens wrote:

> Ok, now I see what you want. But please let's don't add an
> #ifdef CONFIG_ZONE_DMA
> to the dasd driver. Instead just re-add ZONE_DMA to 31-bit s390. Everything
> would be in ZONE_DMA again and ZONE_NORMAL would be empty. Doesn't matter
> if we have an additional zone, since 31-bit support isn't important anymore.
>

Ok, that's fine. Many other architectures do not allow this to be
disabled, so we can add s390 to the list.

> So I could add the following patch to the s390 tree, if wanted:
>

Sure! Please remove the CONFIG_ZONE_DMA from
arch/s390/appldata/appldata_mem.c as well and then add my:

Acked-by: David Rientjes <rientjes@google.com>

once you sign-it-off.

> ---
> arch/s390/Kconfig | 2 +-
> arch/s390/mm/init.c | 2 --
> 2 files changed, 1 insertion(+), 3 deletions(-)
>
> --- a/arch/s390/Kconfig
> +++ b/arch/s390/Kconfig
> @@ -2,7 +2,7 @@ config MMU
> def_bool y
>
> config ZONE_DMA
> - def_bool y if 64BIT
> + def_bool y
>
> config LOCKDEP_SUPPORT
> def_bool y
> --- a/arch/s390/mm/init.c
> +++ b/arch/s390/mm/init.c
> @@ -119,9 +119,7 @@ void __init paging_init(void)
> sparse_memory_present_with_active_regions(MAX_NUMNODES);
> sparse_init();
> memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
> -#ifdef CONFIG_ZONE_DMA
> max_zone_pfns[ZONE_DMA] = PFN_DOWN(MAX_DMA_ADDRESS);
> -#endif
> max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
> free_area_init_nodes(max_zone_pfns);
> fault_init();
>


\
 
 \ /
  Last update: 2011-05-26 01:41    [from the cache]
©2003-2010