On Wed, Jun 30, 2010 at 11:37:10AM -0400, Rik van Riel wrote:
> @@ -2714,10 +2715,11 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
> if ((flags & FAULT_FLAG_WRITE) && reuse_swap_page(page)) {
> pte = maybe_mkwrite(pte_mkdirty(pte), vma);
> flags &= ~FAULT_FLAG_WRITE;
> + exclusive = 1;
> }
Agreed. it's something I considered doing too but I deferred until I
figure out all the mess with ksm-copy. I'll test your patch but it
should make no practical difference to my workload.
I also had a patch in my queue that changes this path which I deferred
as well for now.
====
Subject: set VM_FAULT_WRITE in do_swap_page
From: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Set the flag if do_swap_page is decowing the page the same way do_wp_page would
too.
Signed-off-by: Andrea Arcangeli <aarcange@xxxxxxxxxx>
---
diff --git a/mm/memory.c b/mm/memory.c
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2673,6 +2686,7 @@ static int do_swap_page(struct mm_struct
if ((flags & FAULT_FLAG_WRITE) && reuse_swap_page(page)) {
pte = maybe_mkwrite(pte_mkdirty(pte), vma);
flags &= ~FAULT_FLAG_WRITE;
+ ret |= VM_FAULT_WRITE;
}
flush_icache_page(vma, page);
set_pte_at(mm, address, page_table, pte);
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>