[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAEf4BzYMExNVP353xUmkD=M7_QKDG8Ukm0T7D9aCZG=-GToiaA@mail.gmail.com>
Date: Wed, 9 Nov 2022 20:05:49 -0800
From: Andrii Nakryiko <andrii.nakryiko@...il.com>
To: Xu Kuohai <xukuohai@...wei.com>
Cc: bpf@...r.kernel.org, linux-kernel@...r.kernel.org,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Andrii Nakryiko <andrii@...nel.org>,
Martin KaFai Lau <martin.lau@...ux.dev>,
Song Liu <song@...nel.org>, Yonghong Song <yhs@...com>,
John Fastabend <john.fastabend@...il.com>,
KP Singh <kpsingh@...nel.org>,
Stanislav Fomichev <sdf@...gle.com>,
Hao Luo <haoluo@...gle.com>, Jiri Olsa <jolsa@...nel.org>
Subject: Re: [PATCH bpf-next v3] bpf: Initialize same number of free nodes for
each pcpu_freelist
On Wed, Nov 9, 2022 at 7:33 PM Xu Kuohai <xukuohai@...wei.com> wrote:
>
> pcpu_freelist_populate() initializes nr_elems / num_possible_cpus() + 1
> free nodes for some CPUs, and then possibly one CPU with fewer nodes,
> followed by remaining cpus with 0 nodes. For example, when nr_elems == 256
> and num_possible_cpus() == 32, CPU 0~27 each gets 9 free nodes, CPU 28 gets
> 4 free nodes, CPU 29~31 get 0 free nodes, while in fact each CPU should get
> 8 nodes equally.
>
> This patch initializes nr_elems / num_possible_cpus() free nodes for each
> CPU firstly, then allocates the remaining free nodes by one for each CPU
> until no free nodes left.
>
> Signed-off-by: Xu Kuohai <xukuohai@...wei.com>
> Acked-by: Yonghong Song <yhs@...com>
> ---
> v3: Simplify code as suggested by Andrii
> v2: Update commit message and add Yonghong's ack
> ---
> kernel/bpf/percpu_freelist.c | 27 ++++++++++++++-------------
> 1 file changed, 14 insertions(+), 13 deletions(-)
>
> diff --git a/kernel/bpf/percpu_freelist.c b/kernel/bpf/percpu_freelist.c
> index b6e7f5c5b9ab..bd60070c079f 100644
> --- a/kernel/bpf/percpu_freelist.c
> +++ b/kernel/bpf/percpu_freelist.c
> @@ -100,22 +100,23 @@ void pcpu_freelist_populate(struct pcpu_freelist *s, void *buf, u32 elem_size,
> u32 nr_elems)
> {
> struct pcpu_freelist_head *head;
> - int i, cpu, pcpu_entries;
> + unsigned int cpu, cpu_idx, i, j, n, m;
>
> - pcpu_entries = nr_elems / num_possible_cpus() + 1;
> - i = 0;
> + n = nr_elems / num_possible_cpus();
> + m = nr_elems % num_possible_cpus();
> +
> + cpu_idx = 0;
>
> for_each_possible_cpu(cpu) {
> -again:
> - head = per_cpu_ptr(s->freelist, cpu);
> - /* No locking required as this is not visible yet. */
> - pcpu_freelist_push_node(head, buf);
> - i++;
> - buf += elem_size;
> - if (i == nr_elems)
> - break;
> - if (i % pcpu_entries)
> - goto again;
> + j = min(n + (cpu_idx < m ? 1 : 0), nr_elems);
why the min() here?
> + for (i = 0; i < j; i++) {
> + head = per_cpu_ptr(s->freelist, cpu);
> + /* No locking required as this is not visible yet. */
> + pcpu_freelist_push_node(head, buf);
> + buf += elem_size;
> + }
> + nr_elems -= j;
> + cpu_idx++;
> }
> }
>
> --
> 2.30.2
>
Powered by blists - more mailing lists