From: Hou Tao <houtao@huaweicloud.com>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Tejun Heo <tj@kernel.org>,
rcu@vger.kernel.org, Network Development <netdev@vger.kernel.org>,
bpf <bpf@vger.kernel.org>, Kernel Team <kernel-team@fb.com>,
Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andrii@kernel.org>,
David Vernet <void@manifault.com>,
"Paul E. McKenney" <paulmck@kernel.org>
Subject: Re: [PATCH v4 bpf-next 12/14] bpf: Introduce bpf_mem_free_rcu() similar to kfree_rcu().
Date: Fri, 7 Jul 2023 12:05:18 +0800 [thread overview]
Message-ID: <205ac9e9-ef8c-2b39-8d76-a937d6fc72d5@huaweicloud.com> (raw)
In-Reply-To: <CAADnVQKea47Q1WPtmVrHEZijb=Ms8QzufVj8eds5HmNXGxSRug@mail.gmail.com>
Hi,
On 7/7/2023 10:10 AM, Alexei Starovoitov wrote:
> On Thu, Jul 6, 2023 at 6:45 PM Hou Tao <houtao@huaweicloud.com> wrote:
>>
>>
>> On 7/6/2023 11:34 AM, Alexei Starovoitov wrote:
>>> From: Alexei Starovoitov <ast@kernel.org>
>>>
>>> Introduce bpf_mem_[cache_]free_rcu() similar to kfree_rcu().
>>> Unlike bpf_mem_[cache_]free() that links objects for immediate reuse into
>>> per-cpu free list the _rcu() flavor waits for RCU grace period and then moves
>>> objects into free_by_rcu_ttrace list where they are waiting for RCU
>>> task trace grace period to be freed into slab.
>>>
>>> The life cycle of objects:
>>> alloc: dequeue free_llist
>>> free: enqeueu free_llist
>>> free_rcu: enqueue free_by_rcu -> waiting_for_gp
>>> free_llist above high watermark -> free_by_rcu_ttrace
>>> after RCU GP waiting_for_gp -> free_by_rcu_ttrace
>>> free_by_rcu_ttrace -> waiting_for_gp_ttrace -> slab
>>>
>>> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
>> Acked-by: Hou Tao <houtao1@huawei.com>
> Thank you very much for code reviews and feedback.
You are welcome. I also learn a lot from this great patch set.
>
> btw I still believe that ABA is a non issue and prefer to keep the code as-is,
> but for the sake of experiment I've converted it to spin_lock
> (see attached patch which I think uglifies the code)
> and performance across bench htab-mem and map_perf_test
> seems to be about the same.
> Which was a bit surprising to me.
> Could you please benchmark it on your system?
Will do that later. It seems if there is no cross-CPU allocation and
free, the only possible contention is between __free_rcu() on CPU x and
alloc_bulk()/free_bulk() on a different CPU.
next prev parent reply other threads:[~2023-07-07 4:05 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-06 3:34 [PATCH v4 bpf-next 00/14] bpf: Introduce bpf_mem_cache_free_rcu() Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 01/14] bpf: Rename few bpf_mem_alloc fields Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 02/14] bpf: Simplify code of destroy_mem_alloc() with kmemdup() Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 03/14] bpf: Let free_all() return the number of freed elements Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 04/14] bpf: Refactor alloc_bulk() Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 05/14] bpf: Factor out inc/dec of active flag into helpers Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 06/14] bpf: Further refactor alloc_bulk() Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 07/14] bpf: Change bpf_mem_cache draining process Alexei Starovoitov
2023-07-06 12:55 ` Hou Tao
2023-07-06 3:34 ` [PATCH v4 bpf-next 08/14] bpf: Add a hint to allocated objects Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 09/14] bpf: Allow reuse from waiting_for_gp_ttrace list Alexei Starovoitov
2023-07-07 2:07 ` Hou Tao
2023-07-07 2:12 ` Alexei Starovoitov
2023-07-07 3:38 ` Hou Tao
2023-07-07 4:16 ` Alexei Starovoitov
2023-07-07 4:37 ` Hou Tao
2023-07-07 16:11 ` Alexei Starovoitov
2023-07-07 17:47 ` Paul E. McKenney
2023-07-07 22:22 ` Joel Fernandes
2023-07-08 7:03 ` Hou Tao
2023-07-10 4:45 ` Paul E. McKenney
2023-07-06 3:34 ` [PATCH v4 bpf-next 10/14] rcu: Export rcu_request_urgent_qs_task() Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 11/14] selftests/bpf: Improve test coverage of bpf_mem_alloc Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 12/14] bpf: Introduce bpf_mem_free_rcu() similar to kfree_rcu() Alexei Starovoitov
2023-07-07 1:45 ` Hou Tao
2023-07-07 2:10 ` Alexei Starovoitov
2023-07-07 4:05 ` Hou Tao [this message]
2023-07-08 7:00 ` Hou Tao
2023-07-06 3:34 ` [PATCH v4 bpf-next 13/14] bpf: Convert bpf_cpumask to bpf_mem_cache_free_rcu Alexei Starovoitov
2023-07-06 3:34 ` [PATCH v4 bpf-next 14/14] bpf: Add object leak check Alexei Starovoitov
2023-07-12 21:50 ` [PATCH v4 bpf-next 00/14] bpf: Introduce bpf_mem_cache_free_rcu() patchwork-bot+netdevbpf
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=205ac9e9-ef8c-2b39-8d76-a937d6fc72d5@huaweicloud.com \
--to=houtao@huaweicloud.com \
--cc=alexei.starovoitov@gmail.com \
--cc=andrii@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=kernel-team@fb.com \
--cc=netdev@vger.kernel.org \
--cc=paulmck@kernel.org \
--cc=rcu@vger.kernel.org \
--cc=tj@kernel.org \
--cc=void@manifault.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;
as well as URLs for NNTP newsgroup(s).