[IA64] Correct pernodesize calculation.
authorholt@sgi.com <holt@sgi.com>
Thu, 3 Apr 2008 20:17:12 +0000 (15:17 -0500)
committerTony Luck <tony.luck@intel.com>
Tue, 8 Apr 2008 20:50:58 +0000 (13:50 -0700)
A simple fix.  The existing pernodesize reservation is not taking into
account a second array of pg_data_t structures.  This is normally not
important because the PAGE_ALIGN macro reserves adequate space.

I made the compute_pernodesize steps in the same order as the fill_pernode
steps to make the correlation more clear.

Signed-off-by: Robin Holt <holt@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

No differences found