Re: [RFC][PATCH] Manage jbd allocations from its own slabs

[Badari Pulavarty <pbadari@us.ibm.com>]

Christoph Hellwig wrote:
> On Wed, Aug 23, 2006 at 04:08:15PM -0700, Badari Pulavarty wrote:
>   
> > Hi,
> >
> > Here is the fix to "bh: Ensure bh fits within a page" problem
> > caused by JBD.
> >
> > BTW, I realized that this problem can happen only with 1k, 2k
> > filesystems - as 4k, 8k allocations disable slab debug 
> > automatically. But for completeness, I created slabs for those
> > also.
> >
> > What do you think ? I ran basic tests and things are fine.
> >     
>
> Why can't you just use alloc_page?  I bet the whole slab overhead
> eats more memory than what's wasted when using alloc_pages.  Especially
> as the typical usecase is a 4k blocks filesystem with 4k pagesize
> where the overhead of alloc_page is non-existant.
>   

Yes. That was what proposed earlier. But for 1k, 2k allocations we end 
up wasting whole page.
Isn't it ? Thats why I created right sized slabs and disable slab-debug. 
I guess, I can do this
only for 1k, 2k filesystems and directly use alloc_page() for 4k and 8k 
- but that would make
code ugly and also it doesn't handle cases for bigger base pagesize 
systems (64k power).

Thanks,
Badari


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