On 12/2/2014 3:51 PM, Or Gerlitz wrote:
On 12/2/2014 3:47 PM, Sagi Grimberg wrote:
Imagine 2 vectors are bound to the same core (IRQ affinity is not a
spread) or irqbalance is on.
I see.
We should be really moving to use blk_iopoll here and @ the initiator
too instead of all these local budgets
If we want blk_iopoll we need to move buffer allocation somewhere else
since it is soft-irq. I don't know if this is the best way to go here.
Sagi.
--
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
