On 04/26/2012 09:20 AM, David Rientjes wrote:
> On Thu, 26 Apr 2012, KAMEZAWA Hiroyuki wrote:
>
>>> Or do we instead do this:
>>>
>>> - some_function(foo, bar, GFP_NOIO);
>>> + old_gfp = set_current_gfp(GFP_NOIO);
>>> + some_function(foo, bar);
>>> + set_current_gfp(old_gfp);
>>>
>>> So the rule is "if the code was using an explicit GFP_foo then convert
>>> it to use set_current_gfp(). If the code was receiving a gfp_t
>>> variable from the caller then delete that arg".
>>>
>>> Or something like that. It's all too hopelessly impractical to bother
>>> discussing - 20 years too late!
>>>
>>>
>>> otoh, maybe a constrained version of this could be used to address the
>>> vmalloc() problem alone.
>>>
>>
>> Yes, I think it will be good start.
>>
>
> Maybe a per-thread_info variant of gfp_allowed_mask? So Andrew's
> set_current_gfp() becomes set_current_gfp_allowed() that does
>
> void set_current_gfp_allowed(gfp_t gfp_mask)
> {
> current->gfp_allowed = gfp_mask & gfp_allowed_mask;
> }
>
> and then the page allocator does
>
> gfp_mask &= current->gfp_allowed;
>
> rather than how it currently does
>
> gfp_mask &= gfp_allowed_mask;
>
> and then the caller of set_current_gfp_allowed() cleans up with
> set_current_gfp_allowed(__GFP_BITS_MASK).
Caller should restore old gfp_mask instead of __GFP_BITS_MASK in case of
nesting.And how do we care of atomic context?
I was about to add warning in __vmalloc internal if caller uses GFP_NOIO, GFP_NOFS, GFP_ATOMIC
with Nick's comment and let them make to fix it. But it seems Andrew doesn't agree.
Andrew, please tell me your opinion for fixing this problem.
--
Kind regards,
Minchan Kim
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>