On Wed, Jun 01, 2016 at 02:41:51PM -0700, Andrew Morton wrote:
On Wed, 1 Jun 2016 08:21:09 +0900 Minchan Kim minchan@kernel.org wrote:
Recently, I got many reports about perfermance degradation in embedded system(Android mobile phone, webOS TV and so on) and easy fork fail.
The problem was fragmentation caused by zram and GPU driver mainly. With memory pressure, their pages were spread out all of pageblock and it cannot be migrated with current compaction algorithm which supports only LRU pages. In the end, compaction cannot work well so reclaimer shrinks all of working set pages. It made system very slow and even to fail to fork easily which requires order-[2 or 3] allocations.
Other pain point is that they cannot use CMA memory space so when OOM kill happens, I can see many free pages in CMA area, which is not memory efficient. In our product which has big CMA memory, it reclaims zones too exccessively to allocate GPU and zram page although there are lots of free space in CMA so system becomes very slow easily.
But this isn't presently implemented for GPU drivers or for CMA, yes?
What's the story there?
Broken (out-of-tree) drivers that don't allocate their gpu stuff correctly. There's piles of drivers that get_user_page all over the place but then fail to timely get off these pages again. The fix is to get off those pages again (either by unpinning timely, or registering an mmu_notifier if the driver wants to keep the pages pinned indefinitely, as a caching optimization).
At least that's my guess, and iirc it was confirmed first time around this series showed up. -Daniel