[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAPhsuW6+kqJ+=+ssxdZ7+ZsCgiUC2rrLPZsWb6YdXEiN7ZhW9Q@mail.gmail.com>
Date: Mon, 14 Sep 2020 14:28:22 -0700
From: Song Liu <song@...nel.org>
To: Yonghong Song <yhs@...com>
Cc: bpf <bpf@...r.kernel.org>, Networking <netdev@...r.kernel.org>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Kernel Team <kernel-team@...com>,
Martin KaFai Lau <kafai@...com>
Subject: Re: [PATCH bpf-next] bpf: using rcu_read_lock for bpf_sk_storage_map iterator
On Mon, Sep 14, 2020 at 11:47 AM Yonghong Song <yhs@...com> wrote:
>
> Currently, we use bucket_lock when traversing bpf_sk_storage_map
> elements. Since bpf_iter programs cannot use bpf_sk_storage_get()
> and bpf_sk_storage_delete() helpers which may also grab bucket lock,
> we do not have a deadlock issue which exists for hashmap when
> using bucket_lock ([1]).
The paragraph above describes why we can use bucket_lock, which is more
or less irrelevant to this change. Also, I am not sure why we refer to [1] here.
>
> If a bucket contains a lot of sockets, during bpf_iter traversing
> a bucket, concurrent bpf_sk_storage_{get,delete}() may experience
> some undesirable delays. Using rcu_read_lock() is a reasonable
It will be great to include some performance comparison.
> compromise here. Although it may lose some precision, e.g.,
> access stale sockets, but it will not hurt performance of other
> bpf programs.
>
> [1] https://lore.kernel.org/bpf/20200902235341.2001534-1-yhs@fb.com
>
> Cc: Martin KaFai Lau <kafai@...com>
> Signed-off-by: Yonghong Song <yhs@...com>
Other than these,
Acked-by: Song Liu <songliubraving@...com>
[...]
Powered by blists - more mailing lists