Messages in this thread Patch in this message | | | From | Konrad Rzeszutek Wilk <> | Subject | [PATCH 1/2] xen/p2m: Fix for 32-bit builds the "Reserve 8MB of _brk space for P2M" | Date | Thu, 16 Aug 2012 11:50:13 -0400 |
| |
The git commit 5bc6f9888db5739abfa0cae279b4b442e4db8049 xen/p2m: Reserve 8MB of _brk space for P2M leafs when populating back.
extended the _brk space to fit 1048576 PFNs. The math is that each P2M leaf can cover PAGE_SIZE/sizeof(unsigned long) PFNs. In 64-bit that means 512 PFNs, on 32-bit that is 1024. If on 64-bit machines we want to cover 4GB of PFNs, that means having enough for space to fit 1048576 unsigned longs.
On 64-bit: 1048576 * sizeof(unsigned long) (8) bytes = 8MB
On 32-bit: 1048576 * sizeof(unsigned long) (4) bytes = 4MB
We fix that by using the above mentioned math instead of predefined PMD_SIZE.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> --- arch/x86/xen/p2m.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c index b2e91d4..626c979 100644 --- a/arch/x86/xen/p2m.c +++ b/arch/x86/xen/p2m.c @@ -198,7 +198,8 @@ RESERVE_BRK(p2m_mid_identity, PAGE_SIZE * 2 * 3); * max we have is seen is 395979, but that does not mean it can't be more. * But some machines can have 3GB I/O holes even. So lets reserve enough * for 4GB of I/O and E820 holes. */ -RESERVE_BRK(p2m_populated, PMD_SIZE * 4); +RESERVE_BRK(p2m_populated, 1048576 * sizeof(unsigned long)); + static inline unsigned p2m_top_index(unsigned long pfn) { BUG_ON(pfn >= MAX_P2M_PFN); -- 1.7.7.6
| |