Re: [PATCH 1/3] clean up setup_per_zone_pages_min

From: Minchan Kim
Date: Wed May 20 2009 - 06:53:41 EST


On Wed, May 20, 2009 at 7:47 PM, Mel Gorman <mel@xxxxxxxxx> wrote:
> On Wed, May 20, 2009 at 07:30:45PM +0900, Minchan Kim wrote:
>> On Wed, 20 May 2009 11:21:29 +0100
>> Mel Gorman <mel@xxxxxxxxx> wrote:
>>
>> > On Wed, May 20, 2009 at 06:58:03PM +0900, Minchan Kim wrote:
>> > > Hi, Mel.
>> > >
>> > > On Wed, 20 May 2009 09:54:16 +0100
>> > > Mel Gorman <mel@xxxxxxxxx> wrote:
>> > >
>> > > > On Wed, May 20, 2009 at 04:18:53PM +0900, Minchan Kim wrote:
>> > > > >
>> > > > > Mel changed zone->pages_[high/low/min] with zone->watermark array.
>> > > > > So, setup_per_zone_pages_min also have to be changed.
>> > > > >
>> > > >
>> > > > Just to be clear, this is a function renaming to match the new zone
>> > > > field name, not something I missed. As the function changes min, low and
>> > > > max, a better name might have been setup_per_zone_watermarks but whether
>> > >
>> > > At first, I thouht, too. But It's handle of min_free_kbytes.
>> > > Documentation said, it's to compute a watermark[WMARK_MIN].
>> > > I think many people already used that knob to contorl pages_min to keep the
>> > > low pages.
>> >
>> > Which documentation?
>>
>> Documentation/sysctl/vm.txt - min_free_kbytes.
>>
>
> That documentation states
>
> ====
> This is used to force the Linux VM to keep a minimum number
> of kilobytes free. ÂThe VM uses this number to compute a pages_min
> value for each lowmem zone in the system. ÂEach lowmem zone gets
> a number of reserved free pages based proportionally on its size.
> ====
>
> This is true. It just happens in the implementation that sets pages_min
> (or it's renamed value) also sets the low and high watermarks are also set
> based on the value of the minimum value. It doesn't need to be updated.
>

Okay.
I will modify function name with setup_per_zone_watermarks at next version.
Thanks for careful review, Mel.


--
Kinds regards,
Minchan Kim
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/