* Re: scsi: fix sense_slab/bio swapping livelock
[not found] ` <Pine.LNX.4.64.0804071248110.30814@sbz-30.cs.Helsinki.FI>
@ 2008-04-07 10:07 ` Boaz Harrosh
2008-04-07 10:17 ` Pekka Enberg
0 siblings, 1 reply; 2+ messages in thread
From: Boaz Harrosh @ 2008-04-07 10:07 UTC (permalink / raw)
To: Pekka J Enberg
Cc: Hugh Dickins, James Bottomley, Andrew Morton, FUJITA Tomonori,
Jens Axboe, Christoph Lameter, Peter Zijlstra, Rafael J. Wysocki,
mpm, linux-scsi, linux-kernel
On Mon, Apr 07 2008 at 12:52 +0300, Pekka J Enberg <penberg@cs.helsinki.fi> wrote:
> Hi Boaz,
>
> On Mon, 7 Apr 2008, Boaz Harrosh wrote:
>> The slub behavior described above is disturbing. If I want a 128-byte kmalloc I
>> would use kmalloc. But if I want a dedicated kmem_cache of my own I take the trouble
>> to create one. As I understood it, a dedicated kmem_cache is somewhat growing but
>> lazy-shrinking and eventually maxes out to my usage of it. If I reserve one elemnt then
>> even when memory is low and caches are shrunk I have at least a page. But more then
>> In low memory condition, in a steady sate the cost of each allocation is kept low
>> because I have the pages for my self and I don't need to go grabbing global locks.
>> Sharing with other pools breaks that behavior. Perhaps we need a flag in kmem_cache
>> creation that says we do not want slab sharing (OK slub sharing in this case).
>
> I think you're better off using the page allocator then. SLOB, for
> example, doesn't guarantee you're the only user of a page for
> kmem_cache_alloc() either and I don't really see why it should as it tries
> to be as memory efficient as possible.
>
> Pekka
Please forgive my ignorance, but what is then the difference between kmem_cache_alloc()
and kmalloc?
would you not agree that sometimes we want to override that sharing of SLOBs?
Boaz
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: scsi: fix sense_slab/bio swapping livelock
2008-04-07 10:07 ` scsi: fix sense_slab/bio swapping livelock Boaz Harrosh
@ 2008-04-07 10:17 ` Pekka Enberg
0 siblings, 0 replies; 2+ messages in thread
From: Pekka Enberg @ 2008-04-07 10:17 UTC (permalink / raw)
To: Boaz Harrosh
Cc: Hugh Dickins, James Bottomley, Andrew Morton, FUJITA Tomonori,
Jens Axboe, Christoph Lameter, Peter Zijlstra, Rafael J. Wysocki,
mpm, linux-scsi, linux-kernel
Hi Boaz,
On Mon, Apr 7, 2008 at 1:07 PM, Boaz Harrosh <bharrosh@panasas.com> wrote:
> Please forgive my ignorance, but what is then the difference between kmem_cache_alloc()
> and kmalloc?
Constructors, user-defined alignment, and tighter object packing.
On Mon, Apr 7, 2008 at 1:07 PM, Boaz Harrosh <bharrosh@panasas.com> wrote:
> would you not agree that sometimes we want to override that sharing of SLOBs?
To be honest, I really don't quite understand your use-case. But as
far as I can tell, it hasn't never been explicitly guaranteed and most
certainly has not been true since the merging of SLOB and more
recently SLUB.
Pekka
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2008-04-07 10:18 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <200804062359.m36Nx3lA016774@hera.kernel.org>
[not found] ` <47F9EAD9.3060103@panasas.com>
[not found] ` <Pine.LNX.4.64.0804071248110.30814@sbz-30.cs.Helsinki.FI>
2008-04-07 10:07 ` scsi: fix sense_slab/bio swapping livelock Boaz Harrosh
2008-04-07 10:17 ` Pekka Enberg
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).