Re: [PATCH][5/?] count writeback pages in nr_scanned

[Nick Piggin <nickpiggin@yahoo.com.au>]

Andrew Morton wrote:
> Nick Piggin <nickpiggin@yahoo.com.au> wrote:
> 
>>Rik van Riel wrote:
>>
>>>On Thu, 6 Jan 2005, Nick Piggin wrote:
>>>
>>>
>>>>I think what Andrea is worried about is that blk_congestion_wait is 
>>>>fairly vague, and can be a source of instability in the scanning 
>>>>implementation.
>>>
>>>
>>>The recent OOM kill problem has been happening:
>>>1) with cache pressure on lowmem only, due to a block device write
>>>2) with no block congestion at all
>>>3) with pretty much all pageable lowmme pages in writeback state
>>>
>>>It appears the VM has trouble dealing with the situation where
>>>there is no block congestion to wait on...
>>>
>>
>>Try, together with your nr_scanned patch, to replace blk_congestion_wait
>>with io_schedule_timeout.
> 
> 
> Why?  Is the nr_scanned fix insufficient?
> 

I thought it sounded like he implied that nr_scanned was insufficient
(otherwise he might have said "to wait on ... but my patch fixes it").