Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210401183216.443C4443@viggo.jf.intel.com> <20210401183219.DC1928FA@viggo.jf.intel.com>
In-Reply-To: <20210401183219.DC1928FA@viggo.jf.intel.com>
From:   Wei Xu <weixugc@google.com>
Date:   Fri, 9 Apr 2021 20:07:08 -0700
Message-ID: <CAAPL-u8Jk-i-9-iSnU7_nb-k2ZMqdRk5c88d-M6Bi1rfv4kSLQ@mail.gmail.com>
Subject: Re: [PATCH 02/10] mm/numa: automatically generate node migration order
To:     Dave Hansen <dave.hansen@linux.intel.com>
Cc:     Linux MM <linux-mm@kvack.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Yang Shi <shy828301@gmail.com>,
        David Rientjes <rientjes@google.com>,
        Huang Ying <ying.huang@intel.com>,
        Dan Williams <dan.j.williams@intel.com>,
        David Hildenbrand <david@redhat.com>,
        Oscar Salvador <osalvador@suse.de>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Apr 1, 2021 at 11:35 AM Dave Hansen <dave.hansen@linux.intel.com> wrote:
> +/*
> + * node_demotion[] example:
> + *
> + * Consider a system with two sockets.  Each socket has
> + * three classes of memory attached: fast, medium and slow.
> + * Each memory class is placed in its own NUMA node.  The
> + * CPUs are placed in the node with the "fast" memory.  The
> + * 6 NUMA nodes (0-5) might be split among the sockets like
> + * this:
> + *
> + *     Socket A: 0, 1, 2
> + *     Socket B: 3, 4, 5
> + *
> + * When Node 0 fills up, its memory should be migrated to
> + * Node 1.  When Node 1 fills up, it should be migrated to
> + * Node 2.  The migration path start on the nodes with the
> + * processors (since allocations default to this node) and
> + * fast memory, progress through medium and end with the
> + * slow memory:
> + *
> + *     0 -> 1 -> 2 -> stop
> + *     3 -> 4 -> 5 -> stop
> + *
> + * This is represented in the node_demotion[] like this:
> + *
> + *     {  1, // Node 0 migrates to 1
> + *        2, // Node 1 migrates to 2
> + *       -1, // Node 2 does not migrate
> + *        4, // Node 3 migrates to 4
> + *        5, // Node 4 migrates to 5
> + *       -1} // Node 5 does not migrate
> + */

In this example, if we want to support multiple nodes as the demotion
target of a source node, we can group these nodes into three tiers
(classes):

fast class:
0 -> {1, 4}  // 1 is the preferred
3 -> {4, 1}  // 4 is the preferred

medium class:
1 -> {2, 5}  // 2 is the preferred
4 -> {5, 2}  // 5 is the preferred

slow class:
2 -> stop
5 -> stop

This can guarantee there are no cycles, either.  Does it sound sensible?

> +again:
> +       this_pass = next_pass;
> +       next_pass = NODE_MASK_NONE;
> +       /*
> +        * To avoid cycles in the migration "graph", ensure
> +        * that migration sources are not future targets by
> +        * setting them in 'used_targets'.  Do this only
> +        * once per pass so that multiple source nodes can
> +        * share a target node.
> +        *
> +        * 'used_targets' will become unavailable in future
> +        * passes.  This limits some opportunities for
> +        * multiple source nodes to share a destination.
> +        */
> +       nodes_or(used_targets, used_targets, this_pass);
> +       for_each_node_mask(node, this_pass) {
> +               int target_node = establish_migrate_target(node, &used_targets);
> +
> +               if (target_node == NUMA_NO_NODE)
> +                       continue;
> +
> +               /* Visit targets from this pass in the next pass: */
> +               node_set(target_node, next_pass);
> +       }
> +       /* Is another pass necessary? */
> +       if (!nodes_empty(next_pass))
> +               goto again;

This goto seems like exactly a "do {} while" loop.  Any particular reason not to
use "do {} while" here?