[Intel-wired-lan] [PATCH net-next v3 5/8] igc: Avoid TX Hangs because long cycles

Kurt Kanzenbach kurt.kanzenbach at linutronix.de
Wed Jan 27 09:03:53 UTC 2021


On Tue Jan 26 2021, Vladimir Oltean wrote:
> On Fri, Jan 22, 2021 at 02:44:50PM -0800, Vinicius Costa Gomes wrote:
>> Avoid possible TX Hangs caused by using long Qbv cycles. In some
>> cases, using long cycles (more than 1 second) can cause transmissions
>> to be blocked for that time. As the TX Hang timeout is close to 1
>> second, we may need to reduce the cycle time to something more
>> reasonable: the value chosen is 1ms.
>> 
>> Signed-off-by: Vinicius Costa Gomes <vinicius.gomes at intel.com>
>> ---
>
> Don't you want this patch to go to 'net' and be backported?

I'm wondering about this patch as well. Is this fix related to frame
preemption? If I understand the code correctly the 1sec is a dummy cycle
and all queues are open. How should Tx hang then?

Thanks,
Kurt
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 832 bytes
Desc: not available
URL: <http://lists.osuosl.org/pipermail/intel-wired-lan/attachments/20210127/0faf3bba/attachment.asc>


More information about the Intel-wired-lan mailing list