[GIT PULL] SLAB changes for v3.1-rc0

[GIT PULL] SLAB changes for v3.1-rc0

[Pekka Enberg]

Hi Linus,

Here's batch of slab/slub/slob changes accumulated over the past few months.
The biggest changes are alignment unification from Christoph Lameter and SLUB
debugging improvements from Ben Greear. Also notable is SLAB 'struct
kmem_cache' shrinkage from Eric Dumazet that helps large SMP systems.

Please note that the SLUB lockless slowpath patches will be sent in a separate
pull request.

                         Pekka

The following changes since commit 02f8c6aee8df3cdc935e9bdd4f2d020306035dbe:
   Linus Torvalds (1):
         Linux 3.0

are available in the git repository at:

   ssh://master.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6.git slab-for-linus

Ben Greear (2):
       slub: Enable backtrace for create/delete points
       slub: Add method to verify memory is not freed

Christoph Lameter (2):
       slab, slub, slob: Unify alignment definition
       slab allocators: Provide generic description of alignment defines

Eric Dumazet (1):
       slab: shrink sizeof(struct kmem_cache)

Hugh Dickins (1):
       slab: fix DEBUG_SLAB build

Marcin Slusarz (1):
       slub: reduce overhead of slub_debug

Pekka Enberg (1):
       SLUB: Fix missing <linux/stacktrace.h> include

Steven Rostedt (1):
       slob/lockdep: Fix gfp flags passed to lockdep

Tetsuo Handa (1):
       slab: fix DEBUG_SLAB warning

  include/linux/slab.h     |   20 +++++++++
  include/linux/slab_def.h |   52 ++++++-----------------
  include/linux/slob_def.h |   10 ----
  include/linux/slub_def.h |   23 ++++++----
  mm/slab.c                |   17 ++++----
  mm/slob.c                |    6 +++
  mm/slub.c                |  105 +++++++++++++++++++++++++++++++++++++++++++++-
  7 files changed, 164 insertions(+), 69 deletions(-)

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [GIT PULL] SLAB changes for v3.1-rc0

[Eric Dumazet]

Le vendredi 22 juillet 2011 A  11:08 +0300, Pekka Enberg a A(C)crit :
> Hi Linus,
> 
> Here's batch of slab/slub/slob changes accumulated over the past few months.
> The biggest changes are alignment unification from Christoph Lameter and SLUB
> debugging improvements from Ben Greear. Also notable is SLAB 'struct
> kmem_cache' shrinkage from Eric Dumazet that helps large SMP systems.
> 
> Please note that the SLUB lockless slowpath patches will be sent in a separate
> pull request.
> 
>                          Pekka

Hi Pekka

Could we also merge in 3.1 following "simple enough" patch ?

Thanks

[PATCH] slab: remove one NR_CPUS dependency

Reduce high order allocations in do_tune_cpucache() for some setups.
(NR_CPUS=4096 -> we need 64KB)

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Christoph Lameter <cl@linux.com>
CC: Pekka Enberg <penberg@kernel.org>
---
 mm/slab.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index 1e523ed..b80282a 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3934,7 +3934,7 @@ fail:
 
 struct ccupdate_struct {
 	struct kmem_cache *cachep;
-	struct array_cache *new[NR_CPUS];
+	struct array_cache *new[0];
 };
 
 static void do_ccupdate_local(void *info)
@@ -3956,7 +3956,8 @@ static int do_tune_cpucache(struct kmem_cache *cachep, int limit,
 	struct ccupdate_struct *new;
 	int i;
 
-	new = kzalloc(sizeof(*new), gfp);
+	new = kzalloc(sizeof(*new) + nr_cpu_ids * sizeof(struct array_cache *),
+		      gfp);
 	if (!new)
 		return -ENOMEM;
 


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [GIT PULL] SLAB changes for v3.1-rc0

[Pekka Enberg]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1035 bytes --]

Le vendredi 22 juillet 2011 A  11:08 +0300, Pekka Enberg a A(C)crit :
>> Here's batch of slab/slub/slob changes accumulated over the past few months.
>> The biggest changes are alignment unification from Christoph Lameter and SLUB
>> debugging improvements from Ben Greear. Also notable is SLAB 'struct
>> kmem_cache' shrinkage from Eric Dumazet that helps large SMP systems.

On Mon, 25 Jul 2011, Eric Dumazet wrote:
>> Please note that the SLUB lockless slowpath patches will be sent in a separate
>> pull request.
>>
>>                          Pekka
>
> Hi Pekka
>
> Could we also merge in 3.1 following "simple enough" patch ?
>
> Thanks
>
> [PATCH] slab: remove one NR_CPUS dependency
>
> Reduce high order allocations in do_tune_cpucache() for some setups.
> (NR_CPUS=4096 -> we need 64KB)
>
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
> Acked-by: Christoph Lameter <cl@linux.com>
> CC: Pekka Enberg <penberg@kernel.org>

Sorry for the delay. I applied the patch and will send it to Linus within 
few days.

 			Pekka

Re: [GIT PULL] SLAB changes for v3.1-rc0

[Pekka Enberg]

On Fri, Jul 22, 2011 at 11:08 AM, Pekka Enberg <penberg@kernel.org> wrote:
> Please note that the SLUB lockless slowpath patches will be sent in a
> separate pull request.

Christoph, your debugging fix has been in linux-next for few days now
and no problem have been reported. I'm considering sending the series
to Linus. What do you think?

                        Pekka

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [GIT PULL] SLAB changes for v3.1-rc0

[Christoph Lameter]

On Thu, 28 Jul 2011, Pekka Enberg wrote:

> On Fri, Jul 22, 2011 at 11:08 AM, Pekka Enberg <penberg@kernel.org> wrote:
> > Please note that the SLUB lockless slowpath patches will be sent in a
> > separate pull request.
>
> Christoph, your debugging fix has been in linux-next for few days now
> and no problem have been reported. I'm considering sending the series
> to Linus. What do you think?

Do not know of any other new issues so ok.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [GIT PULL] SLAB changes for v3.1-rc0

[David Rientjes]

On Thu, 28 Jul 2011, Pekka Enberg wrote:

> Christoph, your debugging fix has been in linux-next for few days now
> and no problem have been reported. I'm considering sending the series
> to Linus. What do you think?
> 

I ran slub/lockless through some stress testing and it seems to be quite 
stable on my testing cluster.  There is about a 2.3% performance 
improvement with the lockless slowpath on the netperf benchmark with 
various thread counts on my 16-core 64GB Opterons, so I'd recommend it to 
be merged into 3.1.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [GIT PULL] SLAB changes for v3.1-rc0

[Christoph Lameter]

On Thu, 28 Jul 2011, David Rientjes wrote:

> On Thu, 28 Jul 2011, Pekka Enberg wrote:
>
> > Christoph, your debugging fix has been in linux-next for few days now
> > and no problem have been reported. I'm considering sending the series
> > to Linus. What do you think?
> >
>
> I ran slub/lockless through some stress testing and it seems to be quite
> stable on my testing cluster.  There is about a 2.3% performance
> improvement with the lockless slowpath on the netperf benchmark with
> various thread counts on my 16-core 64GB Opterons, so I'd recommend it to
> be merged into 3.1.

Great. Could you also test the next stage of patches (not yet even in
Pekka's tree) where we add a per cpu cache of partial allocated slab
pages? This decreases the per node lock contention further. I can repost
the set if the old one does not work for you. Shows significant
improvement here as well.



--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [GIT PULL] SLAB changes for v3.1-rc0

[Pekka Enberg]

> On Thu, 28 Jul 2011, David Rientjes wrote:
>
>> On Thu, 28 Jul 2011, Pekka Enberg wrote:
>>
>>> Christoph, your debugging fix has been in linux-next for few days now
>>> and no problem have been reported. I'm considering sending the series
>>> to Linus. What do you think?
>>>
>>
>> I ran slub/lockless through some stress testing and it seems to be quite
>> stable on my testing cluster.  There is about a 2.3% performance
>> improvement with the lockless slowpath on the netperf benchmark with
>> various thread counts on my 16-core 64GB Opterons, so I'd recommend it to
>> be merged into 3.1.

On Fri, 29 Jul 2011, Christoph Lameter wrote:
> Great. Could you also test the next stage of patches (not yet even in
> Pekka's tree) where we add a per cpu cache of partial allocated slab
> pages? This decreases the per node lock contention further. I can repost
> the set if the old one does not work for you. Shows significant
> improvement here as well.

They don't apply so please resend them.

 			Pekka

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>