Re: [RFC][PATCH] flexible array implementation v4

[Benjamin Blum <bblum@google.com>]

On Thu, Jul 23, 2009 at 8:08 PM, KAMEZAWA
Hiroyuki<kamezawa.hiroyu@jp.fujitsu.com> wrote:
> On Fri, 24 Jul 2009 11:02:28 +0800
> Li Zefan <lizf@cn.fujitsu.com> wrote:
>
>> Dave Hansen wrote:
>> > Remaining issues:
>> > - How should we deal with out-of-range indexes, especially
>> >   in flex_array_get() which returns void*?  ERR_PTR()?
>> >   BUG_ON()?  return NULL (current behavior)?
>> > - Should care be taken not to allow a flex_array_get() to
>> >   an index where no flex_array_put() was done?
>> > - Should we decay further than just packing things into the
>> >   'base' page?  Should we actually kmalloc() less than a
>> >   page at times when it will fit?
>> >
>>
>> I sugguest find some candidate users and see how this flex_array
>> fits them.
>>
> Hmm, can't we rewrite cgroup->tasks file using this ?
> I'll try some if I have time.
>
> -Kame
>
>
>

I can do that; it would go nicely on top of my procs file patch
series, but probably wants to wait until after both patch series have
been accepted (see http://lkml.org/lkml/2009/7/23/330 )