Messages in this thread Patch in this message | | | Date | Wed, 19 Apr 2017 17:53:09 -0400 | From | Keith Busch <> | Subject | Re: irq/affinity: Fix extra vecs calculation |
| |
On Wed, Apr 19, 2017 at 12:53:44PM -0700, Andrei Vagin wrote: > On Wed, Apr 19, 2017 at 01:03:59PM -0400, Keith Busch wrote: > > If it's a divide by 0 as your last link indicates, that must mean there > > are possible nodes, but have no CPUs, and those should be skipped. If > > that's the case, the following should fix it, but I'm going to do some > > more qemu testing with various CPU topologies to confirm. > > I printed variables from my test host, I think this can help to > investigate the issue: > > irq_create_affinity_masks:116: vecs_to_assign 0 ncpus 2 extra_vecs 2 vecs_per_node 0 affv 2 curvec 2 nodes 1
That explains a lot. This setup wants 2 "pre_vectors", but I didn't know that was even a thing. This should fix it:
--- diff --git a/kernel/irq/affinity.c b/kernel/irq/affinity.c index d052947..eb8b689 100644 --- a/kernel/irq/affinity.c +++ b/kernel/irq/affinity.c @@ -98,13 +98,16 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) int ncpus, v, vecs_to_assign, vecs_per_node; /* Spread the vectors per node */ - vecs_per_node = (affv - curvec) / nodes; + vecs_per_node = (affv - (curvec - affd->pre_vectors)) / nodes; /* Get the cpus on this node which are in the mask */ cpumask_and(nmsk, cpu_online_mask, cpumask_of_node(n)); --
| |