Re: [Intel-wired-lan] [jkirsher/net-queue PATCH] i40e: Add programming descriptors to cleaned_count

2017-10-27 Thread Pavlos Parissis
On 28 October 2017 at 00:21, Paweł Staszewski  wrote:
> from today it is in net.git
>
> https://git.kernel.org/pub/scm/linux/kernel/git/davem/net.git/log/?qt=grep&q=i40e
>
> It will be later in net-next
>
>

I have to admit I am not that familiar with repo layout, I though
net-next holds what is going to be pushed to Linus's repo and
potentially some parts to linux-stable.

> Also can You please tell me what firmware You are using with Your nics ?
>

1.1752.0., but we use the firmware version that comes from HP, as we
use HP servers, see details here
https://communities.intel.com/message/501682#501682

> Those are X710 ?

Correct.

Cheers,
Pavlos


Re: [Intel-wired-lan] [jkirsher/net-queue PATCH] i40e: Add programming descriptors to cleaned_count

2017-10-27 Thread Paweł Staszewski

from today it is in net.git

https://git.kernel.org/pub/scm/linux/kernel/git/davem/net.git/log/?qt=grep&q=i40e

It will be later in net-next


Also can You please tell me what firmware You are using with Your nics ?

Those are X710 ?


Thanks

Paweł



W dniu 2017-10-27 o 23:20, Pavlos Parissis pisze:

On 23 October 2017 at 01:15, Paweł Staszewski  wrote:

Yes can confirm that after adding patch:

[jkirsher/net-queue PATCH] i40e: Add programming descriptors to
cleaned_count


There is no memleak.



Somehow this patch isn't present in the current net-next repo.
Shouldn't it be there?

Cheers,
Pavlos





Re: [Intel-wired-lan] [jkirsher/net-queue PATCH] i40e: Add programming descriptors to cleaned_count

2017-10-27 Thread Pavlos Parissis
On 23 October 2017 at 01:15, Paweł Staszewski  wrote:
> Yes can confirm that after adding patch:
>
> [jkirsher/net-queue PATCH] i40e: Add programming descriptors to
> cleaned_count
>
>
> There is no memleak.
>
>

Somehow this patch isn't present in the current net-next repo.
Shouldn't it be there?

Cheers,
Pavlos


Re: [jkirsher/net-queue PATCH] i40e: Add programming descriptors to cleaned_count

2017-10-22 Thread Paweł Staszewski

Yes can confirm that after adding patch:

[jkirsher/net-queue PATCH] i40e: Add programming descriptors to 
cleaned_count



There is no memleak.



W dniu 2017-10-22 o 20:01, Anders K. Pedersen | Cohaesio pisze:

On lør, 2017-10-21 at 18:12 -0700, Alexander Duyck wrote:

From: Alexander Duyck 

This patch updates the i40e driver to include programming descriptors
in
the cleaned_count. Without this change it becomes possible for us to
leak
memory as we don't trigger a large enough allocation when the time
comes to
allocate new buffers and we end up overwriting a number of rx_buffers
equal
to the number of programming descriptors we encountered.

Fixes: 0e626ff7ccbf ("i40e: Fix support for flow director programming
status")
Signed-off-by: Alexander Duyck 

This patch solves the remaining memory leak we've seen, so

Tested-by: Anders K. Pedersen 

Regards,
Anders




Re: [jkirsher/net-queue PATCH] i40e: Add programming descriptors to cleaned_count

2017-10-22 Thread Anders K. Pedersen | Cohaesio
On lør, 2017-10-21 at 18:12 -0700, Alexander Duyck wrote:
> From: Alexander Duyck 
> 
> This patch updates the i40e driver to include programming descriptors
> in
> the cleaned_count. Without this change it becomes possible for us to
> leak
> memory as we don't trigger a large enough allocation when the time
> comes to
> allocate new buffers and we end up overwriting a number of rx_buffers
> equal
> to the number of programming descriptors we encountered.
> 
> Fixes: 0e626ff7ccbf ("i40e: Fix support for flow director programming
> status")
> Signed-off-by: Alexander Duyck 

This patch solves the remaining memory leak we've seen, so

Tested-by: Anders K. Pedersen 

Regards,
Anders

[jkirsher/net-queue PATCH] i40e: Add programming descriptors to cleaned_count

2017-10-21 Thread Alexander Duyck
From: Alexander Duyck 

This patch updates the i40e driver to include programming descriptors in
the cleaned_count. Without this change it becomes possible for us to leak
memory as we don't trigger a large enough allocation when the time comes to
allocate new buffers and we end up overwriting a number of rx_buffers equal
to the number of programming descriptors we encountered.

Fixes: 0e626ff7ccbf ("i40e: Fix support for flow director programming status")
Signed-off-by: Alexander Duyck 
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c |1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c 
b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 368722defacb..d6d352a6e6ea 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2102,6 +2102,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, 
int budget)
 
if (unlikely(i40e_rx_is_programming_status(qword))) {
i40e_clean_programming_status(rx_ring, rx_desc, qword);
+   cleaned_count++;
continue;
}
size = (qword & I40E_RXD_QW1_LENGTH_PBUF_MASK) >>