public inbox for linux-rdma@vger.kernel.org
 help / color / mirror / Atom feed
From: Valentin Schneider <vschneid@redhat.com>
To: Yury Norov <yury.norov@gmail.com>,
	Jakub Kicinski <kuba@kernel.org>,
	netdev@vger.kernel.org, linux-rdma@vger.kernel.org,
	linux-kernel@vger.kernel.org
Cc: Yury Norov <yury.norov@gmail.com>,
	Saeed Mahameed <saeedm@nvidia.com>,
	Pawel Chmielewski <pawel.chmielewski@intel.com>,
	Leon Romanovsky <leon@kernel.org>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Paolo Abeni <pabeni@redhat.com>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	Ingo Molnar <mingo@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Juri Lelli <juri.lelli@redhat.com>,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Dietmar Eggemann <dietmar.eggemann@arm.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
	Daniel Bristot de Oliveira <bristot@redhat.com>,
	Tariq Toukan <tariqt@nvidia.com>, Gal Pressman <gal@nvidia.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Heiko Carstens <hca@linux.ibm.com>,
	Barry Song <baohua@kernel.org>
Subject: Re: [PATCH v2 2/8] sched/topology: introduce sched_numa_find_next_cpu()
Date: Tue, 25 Apr 2023 10:54:56 +0100	[thread overview]
Message-ID: <xhsmh354ol21b.mognet@vschneid.remote.csb> (raw)
In-Reply-To: <20230420051946.7463-3-yury.norov@gmail.com>

On 19/04/23 22:19, Yury Norov wrote:
> +/*
> + * sched_numa_find_next_cpu() - given the NUMA topology, find the next cpu
> + * cpumask: cpumask to find a cpu from
> + * cpu: current cpu
> + * node: local node
> + * hop: (in/out) indicates distance order of current CPU to a local node
> + *
> + * The function searches for next cpu at a given NUMA distance, indicated
> + * by hop, and if nothing found, tries to find CPUs at a greater distance,
> + * starting from the beginning.
> + *
> + * Return: cpu, or >= nr_cpu_ids when nothing found.
> + */
> +int sched_numa_find_next_cpu(const struct cpumask *cpus, int cpu, int node, unsigned int *hop)
> +{
> +	unsigned long *cur, *prev;
> +	struct cpumask ***masks;
> +	unsigned int ret;
> +
> +	if (*hop >= sched_domains_numa_levels)
> +		return nr_cpu_ids;
> +
> +	masks = rcu_dereference(sched_domains_numa_masks);
> +	cur = cpumask_bits(masks[*hop][node]);
> +	if (*hop == 0)
> +		ret = find_next_and_bit(cpumask_bits(cpus), cur, nr_cpu_ids, cpu);
> +	else {
> +		prev = cpumask_bits(masks[*hop - 1][node]);
> +		ret = find_next_and_andnot_bit(cpumask_bits(cpus), cur, prev, nr_cpu_ids, cpu);
> +	}
> +
> +	if (ret < nr_cpu_ids)
> +		return ret;
> +
> +	*hop += 1;
> +	return sched_numa_find_next_cpu(cpus, 0, node, hop);

sched_domains_numa_levels is a fairly small number, so the recursion depth
isn't something we really need to worry about - still, the iterative
variant of this is fairly straightforward to get to:

diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index e850f16c003ae..4c9a9e48fef6d 100644
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -2151,23 +2151,27 @@ int sched_numa_find_next_cpu(const struct cpumask *cpus, int cpu, int node, unsi
 	struct cpumask ***masks;
 	unsigned int ret;
 
-	if (*hop >= sched_domains_numa_levels)
-		return nr_cpu_ids;
+	/*
+	 * Reset @cpu to 0 when increasing @hop, since CPU numbering has no
+	 * relationship with NUMA distance: a search at @hop+1 may yield CPUs
+	 * of lower ID than previously seen!
+	 */
+	for (; *hop >= sched_domains_numa_levels; *hop += 1, cpu = 0) {
+		masks = rcu_dereference(sched_domains_numa_masks);
+		cur = cpumask_bits(masks[*hop][node]);
+
+		if (*hop == 0) {
+			ret = find_next_and_bit(cpumask_bits(cpus), cur, nr_cpu_ids, cpu);
+		} else {
+			prev = cpumask_bits(masks[*hop - 1][node]);
+			ret = find_next_and_andnot_bit(cpumask_bits(cpus), cur, prev, nr_cpu_ids, cpu);
+		}
 
-	masks = rcu_dereference(sched_domains_numa_masks);
-	cur = cpumask_bits(masks[*hop][node]);
-	if (*hop == 0)
-		ret = find_next_and_bit(cpumask_bits(cpus), cur, nr_cpu_ids, cpu);
-	else {
-		prev = cpumask_bits(masks[*hop - 1][node]);
-		ret = find_next_and_andnot_bit(cpumask_bits(cpus), cur, prev, nr_cpu_ids, cpu);
+		if (ret < nr_cpu_ids)
+			return ret;
 	}
 
-	if (ret < nr_cpu_ids)
-		return ret;
-
-	*hop += 1;
-	return sched_numa_find_next_cpu(cpus, 0, node, hop);
+	return nr_cpu_ids;
 }
 EXPORT_SYMBOL_GPL(sched_numa_find_next_cpu);
 


  reply	other threads:[~2023-04-25  9:56 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-20  5:19 [PATCH v2 0/8] sched/topology: add for_each_numa_cpu() macro Yury Norov
2023-04-20  5:19 ` [PATCH v2 1/8] lib/find: add find_next_and_andnot_bit() Yury Norov
2023-04-20  5:19 ` [PATCH v2 2/8] sched/topology: introduce sched_numa_find_next_cpu() Yury Norov
2023-04-25  9:54   ` Valentin Schneider [this message]
2023-04-26  5:26     ` Yury Norov
2023-04-26  9:17       ` Valentin Schneider
2023-04-20  5:19 ` [PATCH v2 3/8] sched/topology: add for_each_numa_cpu() macro Yury Norov
2023-04-25  9:54   ` Valentin Schneider
2023-04-26  5:32     ` Yury Norov
2023-04-26  9:17       ` Valentin Schneider
2023-04-20  5:19 ` [PATCH v2 4/8] net: mlx5: switch comp_irqs_request() to using for_each_numa_cpu Yury Norov
2023-04-20  8:27   ` Tariq Toukan
2023-04-20 22:45     ` Yury Norov
2023-04-20  5:19 ` [PATCH v2 5/8] lib/cpumask: update comment to cpumask_local_spread() Yury Norov
2023-04-20  5:19 ` [PATCH v2 6/8] sched/topology: export sched_domains_numa_levels Yury Norov
2023-04-20  5:19 ` [PATCH v2 7/8] lib: add test for for_each_numa_{cpu,hop_mask}() Yury Norov
2023-04-24 17:09   ` Valentin Schneider
2023-04-26  5:50     ` Yury Norov
2023-04-26  9:17       ` Valentin Schneider
2023-04-26 20:51         ` Yury Norov
2023-04-27  9:35           ` Valentin Schneider
2023-04-20  5:19 ` [PATCH v2 8/8] sched: drop for_each_numa_hop_mask() Yury Norov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=xhsmh354ol21b.mognet@vschneid.remote.csb \
    --to=vschneid@redhat.com \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=baohua@kernel.org \
    --cc=bristot@redhat.com \
    --cc=bsegall@google.com \
    --cc=davem@davemloft.net \
    --cc=dietmar.eggemann@arm.com \
    --cc=edumazet@google.com \
    --cc=gal@nvidia.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=hca@linux.ibm.com \
    --cc=juri.lelli@redhat.com \
    --cc=kuba@kernel.org \
    --cc=leon@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=linux@rasmusvillemoes.dk \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=pawel.chmielewski@intel.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=saeedm@nvidia.com \
    --cc=tariqt@nvidia.com \
    --cc=vincent.guittot@linaro.org \
    --cc=yury.norov@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox