svn commit: r210217 - head/sys/kern

[Bruce Evans]

On Wed, 21 Jul 2010, Ivan Voras wrote:

> On 21 July 2010 06:18, Bruce Evans <brde at optusnet.com.au> wrote:
>> On Mon, 19 Jul 2010, John Baldwin wrote:
>>
>>>> Log:
>>>> Â In keeping with the Age-of-the-fruitbat theme, scale up hirunningspace
>>>> on
>>>> Â machines which can clearly afford the memory.
>>>>
>>>> Â This is a somewhat conservative version of the patch - more fine tuning
>>>> may be
>>>> Â necessary.
>>>>
>>>> Â Idea from: Thread on hackers@
>>>> Â Discussed with: alc
>>
>> Sorry I didn't look at the thread, but I wonder if you should increase
>> lorunningspace similarly.
>
> The previous ratio of lorunningspace to hirunningspace was 1/2 - is
> this still a good target?

I don't know if the ratio is more important than difference.  Maybe neither
is very important once the difference is not very small.

> It does seem like there would be more benefitial to hang these
> variables per mount-point or something similar but I'm content that
> they are tunable and that the new values help high-end machines,
> probably in cooperation with tagged (NCQ-like) IO.

So the high end machine owners are less capable of tuning? :-)  This
might be the case even if they also have higher end money and support,
since the higher end is bleeding edge.

Bruce