Recently, Shaohua reported that MADV_FREE is much slower than
MADV_DONTNEED in his MADV_FREE bomb test. The reason is many of
applications went to stall with direct reclaim since kswapd's
reclaim speed isn't fast than applications's allocation speed
so that it causes lots of stall and lock contention.
This patch throttles MADV_FREEing so it works only if there
are enough pages in the system which will not trigger backgroud/
direct reclaim. Otherwise, MADV_FREE falls back to MADV_DONTNEED
because there is no point to delay freeing if we know system
is under memory pressure.
When I test the patch on my 3G machine + 12 CPU + 8G swap,
test: 12 processes
loop = 5;
mmap(512M);
while (loop--) {
memset(512M);
madvise(MADV_FREE or MADV_DONTNEED);
}
1) dontneed: 6.78user 234.09system 0:48.89elapsed
2) madvfree: 6.03user 401.17system 1:30.67elapsed
3) madvfree + this ptach: 5.68user 113.42system 0:36.52elapsed
It's clearly win.
Reported-by: Shaohua Li <shli@xxxxxxxxxx>
Signed-off-by: Minchan Kim <minchan@xxxxxxxxxx>
---
mm/madvise.c | 13 +++++++++++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/mm/madvise.c b/mm/madvise.c
index 6d0fcb8921c2..81bb26ecf064 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -523,8 +523,17 @@ madvise_vma(struct vm_area_struct *vma, struct vm_area_struct **prev,
* XXX: In this implementation, MADV_FREE works like
* MADV_DONTNEED on swapless system or full swap.
*/
- if (get_nr_swap_pages() > 0)
- return madvise_free(vma, prev, start, end);
+ if (get_nr_swap_pages() > 0) {
+ unsigned long threshold;
+ /*
+ * If we have trobule with memory pressure(ie,
+ * under high watermark), free pages instantly.
+ */
+ threshold = min_free_kbytes >> (PAGE_SHIFT - 10);
+ threshold = threshold + (threshold >> 1);
+ if (nr_free_pages() > threshold)
+ return madvise_free(vma, prev, start, end);
+ }
/* passthrough */
case MADV_DONTNEED:
return madvise_dontneed(vma, prev, start, end);
--
1.9.1
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>