Re: [PATCH 2/3] nvme-pci: limit queue count to housekeeping cpus

[Christoph Hellwig <hch@lst.de>]

> diff --git a/block/blk-mq-cpumap.c b/block/blk-mq-cpumap.c
> index 9638b25fd521..43c039900ef6 100644
> --- a/block/blk-mq-cpumap.c
> +++ b/block/blk-mq-cpumap.c
> @@ -11,10 +11,23 @@
>  #include <linux/smp.h>
>  #include <linux/cpu.h>
>  #include <linux/group_cpus.h>
> +#include <linux/sched/isolation.h>
>  
>  #include "blk.h"
>  #include "blk-mq.h"
>  
> +unsigned int blk_mq_num_possible_queues(void)
> +{
> +	const struct cpumask *io_queue_mask;
> +
> +	io_queue_mask = housekeeping_cpumask(HK_TYPE_IO_QUEUE);
> +	if (!cpumask_empty(io_queue_mask))
> +		return cpumask_weight(io_queue_mask);
> +
> +	return num_possible_cpus();
> +}
> +EXPORT_SYMBOL_GPL(blk_mq_num_possible_queues);

This should be split into a separate patch.  And it could really use
a kerneldoc comment.

> -	return num_possible_cpus() + dev->nr_write_queues + dev->nr_poll_queues;
> +	return blk_mq_num_possible_queues() + dev->nr_write_queues + dev->nr_poll_queues;

Please avoid the overly long line here.

Otherwise this looks good to me.