Re: [PATCH 3/3] CMA: Add cma_alloc_counter to make cma_alloc work better if it meet busy range

2014-12-29 Thread Joonsoo Kim
On Thu, Dec 25, 2014 at 05:43:28PM +0800, Hui Zhu wrote: > In [1], Joonsoo said that cma_alloc_counter is useless because pageblock > is isolated. > But if alloc_contig_range meet a busy range, it will undo_isolate_page_range > before goto try next range. At this time, __rmqueue_cma can begin

Re: [PATCH 3/3] CMA: Add cma_alloc_counter to make cma_alloc work better if it meet busy range

2014-12-29 Thread Joonsoo Kim
On Thu, Dec 25, 2014 at 05:43:28PM +0800, Hui Zhu wrote: In [1], Joonsoo said that cma_alloc_counter is useless because pageblock is isolated. But if alloc_contig_range meet a busy range, it will undo_isolate_page_range before goto try next range. At this time, __rmqueue_cma can begin allocd

[PATCH 3/3] CMA: Add cma_alloc_counter to make cma_alloc work better if it meet busy range

2014-12-25 Thread Hui Zhu
In [1], Joonsoo said that cma_alloc_counter is useless because pageblock is isolated. But if alloc_contig_range meet a busy range, it will undo_isolate_page_range before goto try next range. At this time, __rmqueue_cma can begin allocd CMA memory from the range. So I add cma_alloc_counter let

[PATCH 3/3] CMA: Add cma_alloc_counter to make cma_alloc work better if it meet busy range

2014-12-25 Thread Hui Zhu
In [1], Joonsoo said that cma_alloc_counter is useless because pageblock is isolated. But if alloc_contig_range meet a busy range, it will undo_isolate_page_range before goto try next range. At this time, __rmqueue_cma can begin allocd CMA memory from the range. So I add cma_alloc_counter let