On 12/2/2014 12:41 PM, Sagi Grimberg wrote:
On 12/2/2014 12:29 PM, Or Gerlitz wrote:
On 12/1/2014 7:50 PM, Sagi Grimberg wrote:
In case the CQ is packed with completions, we can't just hog the CPU
forever
but we are polling from a work queue, not from hard or soft irq, right?
Right, No danger of getting soft-lockup. The problem here is you have 2
(or more) CQs mapped to the same vector<->core. The serialization of the
work-queue will cause a possible starvation of other CQs.
But we are using per-device and not per-connection CQs, and each CQ uses
a different vector.
--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
