[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] Re: [Xen-changelog] x86/64 e820-parse-and-map code should skip low 1GB now as that



On Thu, Apr 20, 2006 at 06:04:22PM +0000, Xen patchbot -3.0-testing wrote:
> # HG changeset patch
> # User kaf24@xxxxxxxxxxxxxxxxxxxx
> # Node ID fc7a39016601e9dd3161b91b294df1e9674ab383
> # Parent  d3db967d299a9193972a1e466e9a090358d1ec26
> x86/64 e820-parse-and-map code should skip low 1GB now as that
> is mapped by the boot code.
> 
> Signed-off-by: Keir Fraser <keir@xxxxxxxxxxxxx>
> 
> diff -r d3db967d299a -r fc7a39016601 xen/arch/x86/setup.c
> --- a/xen/arch/x86/setup.c    Thu Apr 20 17:10:51 2006 +0100
> +++ b/xen/arch/x86/setup.c    Thu Apr 20 17:11:12 2006 +0100
> @@ -321,7 +321,7 @@ void __init __start_xen(multiboot_info_t
>  #if defined (CONFIG_X86_64)
>          /*
>           * x86/64 maps all registered RAM. Points to note:
> -         *  1. The initial pagetable already maps low 64MB, so skip that.
> +         *  1. The initial pagetable already maps low 1GB, so skip that.
>           *  2. We must map *only* RAM areas, taking care to avoid I/O holes.
>           *     Failure to do this can cause coherency problems and deadlocks
>           *     due to cache-attribute mismatches (e.g., AMD/AGP Linux bug).
> @@ -329,13 +329,14 @@ void __init __start_xen(multiboot_info_t
>          {
>              /* Calculate page-frame range, discarding partial frames. */
>              unsigned long start, end;
> +            unsigned long init_mapped = 1UL << (30 - PAGE_SHIFT); /* 1GB */
>              start = PFN_UP(e820.map[i].addr);
>              end   = PFN_DOWN(e820.map[i].addr + e820.map[i].size);
>              /* Clip the range to above 64MB. */

The aboove comment should be killed too.

Thanks,
Jon

> -            if ( end < (64UL << (20-PAGE_SHIFT)) )
> +            if ( end < init_mapped )
>                  continue;
> -            if ( start < (64UL << (20-PAGE_SHIFT)) )
> -                start = 64UL << (20-PAGE_SHIFT);
> +            if ( start < init_mapped )
> +                start = init_mapped;
>              /* Request the mapping. */
>              map_pages_to_xen(
>                  PAGE_OFFSET + (start << PAGE_SHIFT),
> 
> _______________________________________________
> Xen-changelog mailing list
> Xen-changelog@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-changelog

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.