Re: [PATCH]: Clean up of __alloc_pages

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

Paul Jackson wrote:
>>> 2) The can_try_harder flag values were driving me nuts.
>>
>>Please instead use a second argument 'gfp_high', which will nicely
>>match zone_watermark_ok, and use that consistently when converting
>>__alloc_pages code to use get_page_from_freelist. Ie. keep current
>>behaviour.
> 
> 
> Well ... I still don't understand what you're suggesting, so I
> guess I will have to wait for an actual patch incorporating it.
> 

See how can_try_harder and gfp_high is used currently. They
are simple boolean values and are easily derived from parameters
passed into __alloc_pages.

> Are you also objecting to converting "can_try_harder" to an
> enum, and getting the values in order of desperation?  If so,
> I don't why you object.
> 

Because then to get current behaviour you would have to add
branches to get the correct enum value.

> And there is still the issue that I don't think cpuset constraints
> should be applied in the last attempt before oom_killing for
> GFP_ATOMIC requests.
> 

Sure.

-- 
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com