Displaying 4 results from an estimated 4 matches for "compacton".
Did you mean:
compaction
2016 Mar 23
1
[PATCH v2 13/18] mm/compaction: support non-lru movable page migration
...g the page. Trying
> > trylock_page() in this case could cause a problem.
>
> I don't get it. Why do you think trylock_page could cause a problem?
> Could you elaborate it more?
Okay. Consider following sequence.
CPU-A CPU-B
check PageMovable() in compacton
... free the page
... allocate the page for other usecase
... (maybe for file cache or slub)
get unless 0 in isolate_movable_page()
trylock (success)
(try) lock! failed!
In this case, so...
2016 Mar 23
1
[PATCH v2 13/18] mm/compaction: support non-lru movable page migration
...g the page. Trying
> > trylock_page() in this case could cause a problem.
>
> I don't get it. Why do you think trylock_page could cause a problem?
> Could you elaborate it more?
Okay. Consider following sequence.
CPU-A CPU-B
check PageMovable() in compacton
... free the page
... allocate the page for other usecase
... (maybe for file cache or slub)
get unless 0 in isolate_movable_page()
trylock (success)
(try) lock! failed!
In this case, so...
2016 Mar 22
2
[PATCH v2 13/18] mm/compaction: support non-lru movable page migration
On Mon, Mar 21, 2016 at 03:31:02PM +0900, Minchan Kim wrote:
> We have allowed migration for only LRU pages until now and it was
> enough to make high-order pages. But recently, embedded system(e.g.,
> webOS, android) uses lots of non-movable pages(e.g., zram, GPU memory)
> so we have seen several reports about troubles of small high-order
> allocation. For fixing the problem, there
2016 Mar 22
2
[PATCH v2 13/18] mm/compaction: support non-lru movable page migration
On Mon, Mar 21, 2016 at 03:31:02PM +0900, Minchan Kim wrote:
> We have allowed migration for only LRU pages until now and it was
> enough to make high-order pages. But recently, embedded system(e.g.,
> webOS, android) uses lots of non-movable pages(e.g., zram, GPU memory)
> so we have seen several reports about troubles of small high-order
> allocation. For fixing the problem, there