[PATCH] bnx2fc: use kthread_create_on_node

[PATCH] bnx2fc: use kthread_create_on_node

[Eric Dumazet]

Since bnx2fc_percpu_thread_create() creates percpu kthread, it makes
sense to use kthread_create_on_node() to get proper NUMA affinity for
kthread stack.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: "James E.J. Bottomley" <JBottomley@parallels.com>
---
 drivers/scsi/bnx2fc/bnx2fc_fcoe.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
index a97aff3..9843434 100644
--- a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
+++ b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
@@ -2161,9 +2161,9 @@ static void bnx2fc_percpu_thread_create(unsigned int cpu)
 
 	p = &per_cpu(bnx2fc_percpu, cpu);
 
-	thread = kthread_create(bnx2fc_percpu_io_thread,
-				(void *)p,
-				"bnx2fc_thread/%d", cpu);
+	thread = kthread_create_on_node(bnx2fc_percpu_io_thread,
+					(void *)p, cpu_to_node(cpu),
+					"bnx2fc_thread/%d", cpu);
 	/* bind thread to the cpu */
 	if (likely(!IS_ERR(p->iothread))) {
 		kthread_bind(thread, cpu);

Re: [PATCH] bnx2fc: use kthread_create_on_node

[Bhanu Prakash Gollapudi]

On 7/29/2011 12:06 AM, Eric Dumazet wrote:
> Since bnx2fc_percpu_thread_create() creates percpu kthread, it makes
> sense to use kthread_create_on_node() to get proper NUMA affinity for
> kthread stack.
>
> Signed-off-by: Eric Dumazet<eric.dumazet@gmail.com>
> CC: "James E.J. Bottomley"<JBottomley@parallels.com>

Thanks Eric.

Acked-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com>

> ---
>   drivers/scsi/bnx2fc/bnx2fc_fcoe.c |    6 +++---
>   1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
> index a97aff3..9843434 100644
> --- a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
> +++ b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
> @@ -2161,9 +2161,9 @@ static void bnx2fc_percpu_thread_create(unsigned int cpu)
>
>   	p =&per_cpu(bnx2fc_percpu, cpu);
>
> -	thread = kthread_create(bnx2fc_percpu_io_thread,
> -				(void *)p,
> -				"bnx2fc_thread/%d", cpu);
> +	thread = kthread_create_on_node(bnx2fc_percpu_io_thread,
> +					(void *)p, cpu_to_node(cpu),
> +					"bnx2fc_thread/%d", cpu);
>   	/* bind thread to the cpu */
>   	if (likely(!IS_ERR(p->iothread))) {
>   		kthread_bind(thread, cpu);
>
>
>

Re: [PATCH] bnx2fc: use kthread_create_on_node

[Bhanu Prakash Gollapudi]

On 7/29/2011 12:22 PM, Bhanu Prakash Gollapudi wrote:
> On 7/29/2011 12:06 AM, Eric Dumazet wrote:
>> Since bnx2fc_percpu_thread_create() creates percpu kthread, it makes
>> sense to use kthread_create_on_node() to get proper NUMA affinity for
>> kthread stack.
>>
>> Signed-off-by: Eric Dumazet<eric.dumazet@gmail.com>
>> CC: "James E.J. Bottomley"<JBottomley@parallels.com>
>
> Thanks Eric.
>
> Acked-by: Bhanu Prakash Gollapudi<bprakash@broadcom.com>

James, I found a couple of old patches that were not applied, this being 
one and the other - http://marc.info/?l=linux-scsi&m=132561632116410&w=2.

Please let me know if you want the patches to be resubmitted, as they 
are a bit old now.

Thanks,
Bhanu

>
>> ---
>>    drivers/scsi/bnx2fc/bnx2fc_fcoe.c |    6 +++---
>>    1 file changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
>> index a97aff3..9843434 100644
>> --- a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
>> +++ b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
>> @@ -2161,9 +2161,9 @@ static void bnx2fc_percpu_thread_create(unsigned int cpu)
>>
>>    	p =&per_cpu(bnx2fc_percpu, cpu);
>>
>> -	thread = kthread_create(bnx2fc_percpu_io_thread,
>> -				(void *)p,
>> -				"bnx2fc_thread/%d", cpu);
>> +	thread = kthread_create_on_node(bnx2fc_percpu_io_thread,
>> +					(void *)p, cpu_to_node(cpu),
>> +					"bnx2fc_thread/%d", cpu);
>>    	/* bind thread to the cpu */
>>    	if (likely(!IS_ERR(p->iothread))) {
>>    		kthread_bind(thread, cpu);
>>
>>
>>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Re: [PATCH] bnx2fc: use kthread_create_on_node

[Joe Perches]

On Tue, 2012-05-29 at 17:26 -0700, Bhanu Prakash Gollapudi wrote:
> On 7/29/2011 12:22 PM, Bhanu Prakash Gollapudi wrote:
> > On 7/29/2011 12:06 AM, Eric Dumazet wrote:
> >> Since bnx2fc_percpu_thread_create() creates percpu kthread, it makes
> >> sense to use kthread_create_on_node() to get proper NUMA affinity for
> >> kthread stack.
> >>
> >> Signed-off-by: Eric Dumazet<eric.dumazet@gmail.com>
> >> CC: "James E.J. Bottomley"<JBottomley@parallels.com>
> >
> > Thanks Eric.
> >
> > Acked-by: Bhanu Prakash Gollapudi<bprakash@broadcom.com>
> 
> James, I found a couple of old patches that were not applied, this being 
> one and the other - http://marc.info/?l=linux-scsi&m=132561632116410&w=2.
> 
> Please let me know if you want the patches to be resubmitted, as they 
> are a bit old now.

Do tell if James wakes up from his slumber to
actually apply trivially correct but perhaps
what he considers unnecessary, bothersome patches.

Re: [PATCH] bnx2fc: use kthread_create_on_node

[James Bottomley]

On Tue, 2012-05-29 at 17:26 -0700, Bhanu Prakash Gollapudi wrote:
> On 7/29/2011 12:22 PM, Bhanu Prakash Gollapudi wrote:
> > On 7/29/2011 12:06 AM, Eric Dumazet wrote:
> >> Since bnx2fc_percpu_thread_create() creates percpu kthread, it makes
> >> sense to use kthread_create_on_node() to get proper NUMA affinity for
> >> kthread stack.
> >>
> >> Signed-off-by: Eric Dumazet<eric.dumazet@gmail.com>
> >> CC: "James E.J. Bottomley"<JBottomley@parallels.com>
> >
> > Thanks Eric.
> >
> > Acked-by: Bhanu Prakash Gollapudi<bprakash@broadcom.com>
> 
> James, I found a couple of old patches that were not applied, this being 
> one and the other - http://marc.info/?l=linux-scsi&m=132561632116410&w=2.
> 
> Please let me know if you want the patches to be resubmitted, as they 
> are a bit old now.

Yes, please.  And for future reference, if I ever miss anything you want
in, just fold it into your internal queue and resend for your next batch
(with the appropriate signoffs).

Thanks,

James