On 06/14/2016 11:54 PM, Guilherme G. Piccoli wrote:
On 06/14/2016 04:58 PM, Christoph Hellwig wrote:
I take this opportunity to ask you something, since I'm working in a
related code in a specific driver - sorry in advance if my question is
silly or if I misunderstood your code.
The function irq_create_affinity_mask() below deals with the case in
which we have nr_vecs < num_online_cpus(); in this case, wouldn't be a
good idea to trying distribute the vecs among cores?
Example: if we have 128 online cpus, 8 per core (meaning 16 cores) and
64 vecs, I guess would be ideal to distribute 4 vecs _per core_, leaving
4 CPUs in each core without vecs.
Hello Christoph and Guilherme,
I would also like to see irq_create_affinity_mask() modified such that
it implements Guilherme's algorithm. I think blk-mq requests should be
processed by a CPU core from the NUMA node from which the request has
been submitted. With the proposed algorithm, if the number of MSI-X
vectors is less than or equal to the number of CPU cores of a single
node, all interrupt vectors will be assigned to the first NUMA node.
Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
