Re: [RFC PATCH 0/2] On-demand Filesystem Initialisation

Re: [RFC PATCH 0/2] On-demand Filesystem Initialisation

[Dave Chinner]

On Sun, Jun 01, 2008 at 03:51:53PM +0100, Tom Spink wrote:
> 
> (resend to include CCs)

What cc's? Still no xfs cc on it. I added it to this reply....

> This (short) patch series is another RFC for the patch that introduces on-demand
> filesystem initialisation.  In addition to the original infrastructure
> implementation (with clean-ups), it changes XFS to use this new infrastructure.
> 
> I wrote a toy filesystem (testfs) to simulate scheduling/allocation delays and
> to torture the mount/unmount cycles.  I didn't manage to deadlock the system
> in my tests.  XFS also works as expected aswell, in that the global threads
> are not created until an XFS filesystem is mounted for the first time.  When the
> last XFS filesystem is unmounted, the threads go away.
> 
> Please let me know what you think!

Why even bother? This is why we have /modular/ kernels - if you're
not using XFS then don't load it and you won't see those pesky
threads. That'll save on a bunch of memory as well because the xfs
module ain't small (>480k on i386)....

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

Re: [RFC PATCH 0/2] On-demand Filesystem Initialisation

[Tom Spink]

2008/6/2 Dave Chinner <david@fromorbit.com>:
> On Sun, Jun 01, 2008 at 03:51:53PM +0100, Tom Spink wrote:
>>
>> (resend to include CCs)
>
> What cc's? Still no xfs cc on it. I added it to this reply....
>
>> This (short) patch series is another RFC for the patch that introduces on-demand
>> filesystem initialisation.  In addition to the original infrastructure
>> implementation (with clean-ups), it changes XFS to use this new infrastructure.
>>
>> I wrote a toy filesystem (testfs) to simulate scheduling/allocation delays and
>> to torture the mount/unmount cycles.  I didn't manage to deadlock the system
>> in my tests.  XFS also works as expected aswell, in that the global threads
>> are not created until an XFS filesystem is mounted for the first time.  When the
>> last XFS filesystem is unmounted, the threads go away.
>>
>> Please let me know what you think!
>
> Why even bother? This is why we have /modular/ kernels - if you're
> not using XFS then don't load it and you won't see those pesky
> threads. That'll save on a bunch of memory as well because the xfs
> module ain't small (>480k on i386)....

Yeah, absolutely.  But if the filesystem is built-in, you can't unload it.

> Cheers,
>
> Dave.

Thanks for taking a look, anyway!

-- 
Tom Spink