Re: Xen-unstable panic: FATAL PAGE FAULT

[Keir Fraser <keir.fraser@eu.citrix.com>]

On 31/08/2010 17:22, "Jan Beulich" <JBeulich@novell.com> wrote:

>> Where is even that constraint ensured in the code? I can't see it (I would
>> have assumed that pfn_pdx_hole_setup() would be ensuring it).
> 
> That's somewhat implicit: srat_parse_regions() gets passed an
> address that is at least BOOTSTRAP_DIRECTMAP_END (i.e. 4G).
> Thus srat_parse_regions() starts off with a mask with the lower
> 32 bits all set (only more bits can get set subsequently). Thus
> the earliest zero bit pfn_pdx_hole_setup() can find is bit 20
> (due to the >> PAGE_SHIFT in the invocation). Consequently
> the smallest chunk where arithmetic is valid really is 4Gb, not
> 256Mb as I first wrote.

Well, that's a bit too implicit for me. How about we initialise 'j' to
MAX_ORDER in pfn_pdx_hole_setup() with a comment about supporting page_info
pointer arithmetic within allocatable multi-page regions?

Something like the appended (but with a code comment)?

 -- Keir

--- a/xen/arch/x86/x86_64/mm.c    Mon Aug 30 14:59:12 2010 +0100
+++ b/xen/arch/x86/x86_64/mm.c    Tue Aug 31 17:34:34 2010 +0100
@@ -165,7 +165,8 @@
 {
     unsigned int i, j, bottom_shift, hole_shift;
 
-    for ( hole_shift = bottom_shift = j = 0; ; )
+    hole_shift = bottom_shift = 0;
+    for ( j = MAX_ORDER-1; ; )
     {
         i = find_next_zero_bit(&mask, BITS_PER_LONG, j);
         j = find_next_bit(&mask, BITS_PER_LONG, i);