Hi Alex,
How about my attached patch.
The size of a fragment which has EFI_MEMORY_RUNTIME is so small that
the fragment looks being ignored always. Is it necessary for xen?
I think it just should be passed to dom0.
I'll try your patch also.
Thanks,
Kouya
Alex Williamson writes:
> Hi Kouya,
>
> Here's a different approach that might be more clean. Newer upstream
> kernels have changed EFI MDT parsing so that it doesn't change the
> original EFI table. Here's a reference to the upstream changeset:
>
> http://www.kernel.org/hg/linux-2.6/?cs=fb781f6d3e81
>
> The attached patch is a port of this to xen. Does this improve things
> on the PRIMEQUEST? The nice part about this is that it doesn't add any
> new Xen specific code to the linux-xen files. Thanks,
>
> Alex
diff -r 25483d9b55d4 xen/arch/ia64/linux-xen/efi.c
--- a/xen/arch/ia64/linux-xen/efi.c Wed May 31 16:07:47 2006 -0600
+++ b/xen/arch/ia64/linux-xen/efi.c Thu Jun 08 12:03:41 2006 +0900
@@ -328,6 +328,8 @@ efi_memmap_walk (efi_freemem_callback_t
continue;
#ifdef XEN
+ if (md->attribute & EFI_MEMORY_RUNTIME)
+ continue;
// this works around a problem in the ski bootloader
{
extern long running_on_sim;
_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel
|