On Wed, 20 Mar 2013 16:31:58 +0200
Horia Geanta wrote:
> This reverts commit 891104ed008e8646c7860fe5bc70b0aac55dcc6c.
>
> Current IPsec ESN implementation for authencesn(cbc(aes), hmac(sha))
> (separate encryption and integrity algorithms) does not conform
> to RFC4303.
>
> ICV is generated by
On Wed, 20 Mar 2013 16:31:38 +0200
Horia Geanta wrote:
> This reverts commit e763eb699be723fb41af818118068c6b3afdaf8d.
>
> Current IPsec ESN implementation for authencesn(cbc(aes), hmac(sha))
> (separate encryption and integrity algorithms) does not conform
> to RFC4303.
>
> ICV is generated by
On Tue, Mar 19, 2013 at 05:00:19PM -0700, Tejun Heo wrote:
> and also available in the following git branch.
>
> git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq.git review-numa
Branch rebased on top of the current wq/for-3.10 with updated patches.
The new comit ID is 9555fbc12d786a9eae7cf770
Currently, an unbound workqueue has single current, or first, pwq
(pool_workqueue) to which all new work items are queued. This often
isn't optimal on NUMA machines as workers may jump around across node
boundaries and work items get assigned to workers without any regard
to NUMA affinity.
This p
Use kmem_cache_alloc_node() with @pool->node instead of
kmem_cache_zalloc() when allocating a pool_workqueue so that it's
allocated on the same node as the associated worker_pool. As there's
no no kmem_cache_zalloc_node(), move zeroing to init_pwq().
This was suggested by Lai Jiangshan.
Signed-o
Currently, an unbound workqueue has single current, or first, pwq
(pool_workqueue) to which all new work items are queued. This often
isn't optimal on NUMA machines as workers may jump around across node
boundaries and work items get assigned to workers without any regard
to NUMA affinity.
This p
On Wed, Mar 20, 2013 at 11:48 PM, Tejun Heo wrote:
> On Wed, Mar 20, 2013 at 11:43:30PM +0800, Lai Jiangshan wrote:
>> > + for_each_node(node)
>> > +
>> > BUG_ON(!alloc_cpumask_var_node(&wq_numa_possible_cpumask[node],
>> > + GFP_KE
On Wed, Mar 20, 2013 at 11:52:03PM +0800, Lai Jiangshan wrote:
> > +static struct pool_workqueue *alloc_unbound_pwq(struct workqueue_struct
> > *wq,
> > + const struct workqueue_attrs *attrs)
> > +{
> > + struct worker_pool *pool;
> > + struct pool
On Wed, Mar 20, 2013 at 8:00 AM, Tejun Heo wrote:
> Break init_and_link_pwq() into init_pwq() and link_pwq() and move
> unbound-workqueue specific handling into apply_workqueue_attrs().
> Also, factor out unbound pool and pool_workqueue allocation into
> alloc_unbound_pwq().
>
> This reorganizatio
On Wed, Mar 20, 2013 at 11:43:30PM +0800, Lai Jiangshan wrote:
> > + for_each_node(node)
> > +
> > BUG_ON(!alloc_cpumask_var_node(&wq_numa_possible_cpumask[node],
> > + GFP_KERNEL, node));
> > + for_each_possible_cpu(cpu) {
> >
On Wed, Mar 20, 2013 at 8:00 AM, Tejun Heo wrote:
> Unbound workqueues are going to be NUMA-affine. Add wq_numa_tbl_len
> and wq_numa_possible_cpumask[] in preparation. The former is the
> highest NUMA node ID + 1 and the latter is masks of possibles CPUs for
> each NUMA node.
>
> This patch onl
On Wed, Mar 20, 2013 at 8:26 AM, Lai Jiangshan wrote:
>> for_eahc_node(node)
>> if (pwq_tbl[node] != dfl_pwq)
>> kfree(pwq_tbl[node]);
>> kfree(dfl_pwq);
>
> I also missed.
> we still need to put_unbound_pool() before free(pwq)
Yeap, we do.
On Wed, Mar 20, 2013 at 11:05 PM, Tejun Heo wrote:
> On Wed, Mar 20, 2013 at 11:03:53PM +0800, Lai Jiangshan wrote:
>> > +enomem:
>> > + free_workqueue_attrs(tmp_attrs);
>> > + if (pwq_tbl) {
>> > + for_each_node(node)
>> > + kfree(pwq_tbl[node]);
>>
On Wed, Mar 20, 2013 at 11:03:53PM +0800, Lai Jiangshan wrote:
> > +enomem:
> > + free_workqueue_attrs(tmp_attrs);
> > + if (pwq_tbl) {
> > + for_each_node(node)
> > + kfree(pwq_tbl[node]);
>
> It will free the dfl_pwq multi times.
Oops, you're righ
On Wed, Mar 20, 2013 at 8:00 AM, Tejun Heo wrote:
> Currently, an unbound workqueue has single current, or first, pwq
> (pool_workqueue) to which all new work items are queued. This often
> isn't optimal on NUMA machines as workers may jump around across node
> boundaries and work items get assig
On Wed, Mar 20, 2013 at 11:08:29PM +0900, JoonSoo Kim wrote:
> 2013/3/20 Tejun Heo :
> > Unbound workqueues are going to be NUMA-affine. Add wq_numa_tbl_len
> > and wq_numa_possible_cpumask[] in preparation. The former is the
> > highest NUMA node ID + 1 and the latter is masks of possibles CPUs
This reverts commit e763eb699be723fb41af818118068c6b3afdaf8d.
Current IPsec ESN implementation for authencesn(cbc(aes), hmac(sha))
(separate encryption and integrity algorithms) does not conform
to RFC4303.
ICV is generated by hashing the sequence
SPI, SeqNum-High, SeqNum-Low, IV, Payload
instead
This reverts commit 891104ed008e8646c7860fe5bc70b0aac55dcc6c.
Current IPsec ESN implementation for authencesn(cbc(aes), hmac(sha))
(separate encryption and integrity algorithms) does not conform
to RFC4303.
ICV is generated by hashing the sequence
SPI, SeqNum-High, SeqNum-Low, IV, Payload
instead
On 3/20/2013 4:15 PM, Horia Geanta wrote:
This reverts commit e763eb699be723fb41af818118068c6b3afdaf8d
(upstream: 891104ed008e8646c7860fe5bc70b0aac55dcc6c).
This line is not needed.
Will resend this with commit message updated.
Sorry for the noise.
--
To unsubscribe from this list: send the
On 3/20/2013 4:15 PM, Horia Geanta wrote:
This reverts commit 891104ed008e8646c7860fe5bc70b0aac55dcc6c
(upstream: 891104ed008e8646c7860fe5bc70b0aac55dcc6c).
This line is not needed.
Will resend this with commit message updated.
Sorry for the noise.
--
To unsubscribe from this list: send the l
Fixes a race on driver init with registering algorithms where the
driver status flag wasn't being set before self testing started.
Added the cra_alignmask field for CBC and ECB modes.
Fixed a bug in GCM where AES block size was being used instead of
authsize.
Signed-off-by: Kent Yoder
---
This reverts commit e763eb699be723fb41af818118068c6b3afdaf8d
(upstream: 891104ed008e8646c7860fe5bc70b0aac55dcc6c).
Current IPsec ESN implementation for authencesn(cbc(aes), hmac(sha))
(separate encryption and integrity algorithms) does not conform
to RFC4303.
ICV is generated by hashing the seque
This reverts commit 891104ed008e8646c7860fe5bc70b0aac55dcc6c
(upstream: 891104ed008e8646c7860fe5bc70b0aac55dcc6c).
Current IPsec ESN implementation for authencesn(cbc(aes), hmac(sha))
(separate encryption and integrity algorithms) does not conform
to RFC4303.
ICV is generated by hashing the seque
2013/3/20 Tejun Heo :
> Unbound workqueues are going to be NUMA-affine. Add wq_numa_tbl_len
> and wq_numa_possible_cpumask[] in preparation. The former is the
> highest NUMA node ID + 1 and the latter is masks of possibles CPUs for
> each NUMA node.
It is better to move this code to topology.c o
(off-topic)
Hi, tj,
I think a0265a7f5161b6cb55e82b71edb236bbe0d9b3ae(tj/for-3.10) is
wrong direction, if workqueue_freezing is used only in
freeze_workqueues_begin()/thaw_workqueues(), which means it can be
removed or it is bug which is needed to be fixed. BUT
a0265a7f5161b6cb55e82b71edb236bbe0d
25 matches
Mail list logo
