Issue found in Armada 370: "No buffer space available" error during continuous ping

[Maggie Mae Roxas]

Hi Willy,
Good day.

BTW, here are some answers to your questions.

> In fact I don't know if you're running your own board or a "standard"
one (a mirabox or any NAS board).
We are using a "customized" one, not a "standard" one.
We based the design on Armada 370 RD Evaluation Board, but we used
Marvell 88E1512 as Ethernet PHY and Marvell 88F6707 as processor
instead of the ones in the Armada 370 RD (I think it uses Marvell
88E1310 as Ethernet PHY and Marvell 88F6W11 as processor).

> Because that could also be one of the
differences between what you observe on your side and our respective
experiences with our boards.
Acknowledged.

> Maggie, do you know if it is possible that for any reason your board
would not deliver an IRQ on Tx completion ? That could explain things.
> You can easily test reverting commit 4f3a4f701b just in case.
> If that's the case, then the next step will be to figure out how it is possible
that IRQs are disabled!
After reverting 4f3a4f701b, as I reported, issue does not happen anymore.
Please let me know how to "figure out how it is possible that IRQs are
disabled".

Also, what is the impact if I use this combination?
> - use 3.13.9 mvneta.c
> - apply cd71e246c16b30e3f396a85943d5f596202737ba
> - revert 4f3a4f701b59a3e4b5c8503ac3d905c0a326f922

Are there functionalities that won't work?

Thank you very much for your support.

Regards,
Maggie Roxas

On Sun, Jul 20, 2014 at 6:57 PM, Maggie Mae Roxas
<maggie.mae.roxas at gmail.com> wrote:
> Hi Willy, Thomas,
> Good day.
>
>> I'll try this one, specifcally this combination:
>> - use 3.13.9 mvneta.c
>> - apply cd71e246c16b30e3f396a85943d5f596202737ba
>> - revert 4f3a4f701b59a3e4b5c8503ac3d905c0a326f922
>
> This is to confirm that "No buffer space available" issue is resolved
> after I applied above combinations in in v3.13.9.
> Thanks a lot for the help!
>
> BTW, sorry for the late update - all our connections, electricity and
> stuff are unstable for the past week due to typhoon and we only got
> everything working just today.
>
> Thank you again for attending to our reports.
>
> Regards,
> Maggie Roxas
>
> On Thu, Jul 17, 2014 at 1:15 AM, Willy Tarreau <w at 1wt.eu> wrote:
>> Hi Maggie,
>>
>> On Wed, Jul 16, 2014 at 10:37:47PM -0700, Maggie Mae Roxas wrote:
>>> Hi Thomas, Willy,
>>> Good day.
>>>
>>> First of all, thanks again for looking further into this.
>>>
>>> > Maggie, do you know if it is possible that for any reason your board
>>> > would not deliver an IRQ on Tx completion ? That could explain things.
>>>
>>> I'm sorrry but I'm not really sure how to check this - all I know is
>>> that the difference in the working and not-working setup (both
>>> software and hardware) is the mvneta.c.
>>
>> In fact I don't know if you're running your own board or a "standard"
>> one (a mirabox or any NAS board). Because that could also be one of the
>> differences between what you observe on your side and our respective
>> experiences with our boards.
>>
>>> > You can easily test reverting commit 4f3a4f701b just in case. If that's
>>> > the case, then the next step will be to figure out how it is possible
>>> > that IRQs are disabled!
>>> I'll try this one, specifcally this combination:
>>> - use 3.13.9 mvneta.c
>>> - apply cd71e246c16b30e3f396a85943d5f596202737ba
>>> - revert 4f3a4f701b59a3e4b5c8503ac3d905c0a326f922
>>>
>>> I will update you the results within today until latest tomorrow.
>>
>> OK fine, thank you!
>>
>> Willy
>>