netdev - Re: [PATCH bpf-next 07/12] bpf: Add a hint to allocated objects.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Sat, 24 Jun 2023 11:54:04 +0800
From: Hou Tao <houtao@...weicloud.com>
To: Alexei Starovoitov <alexei.starovoitov@...il.com>
Cc: Daniel Borkmann <daniel@...earbox.net>,
 Andrii Nakryiko <andrii@...nel.org>, David Vernet <void@...ifault.com>,
 "Paul E. McKenney" <paulmck@...nel.org>, Tejun Heo <tj@...nel.org>,
 rcu@...r.kernel.org, Network Development <netdev@...r.kernel.org>,
 bpf <bpf@...r.kernel.org>, Kernel Team <kernel-team@...com>
Subject: Re: [PATCH bpf-next 07/12] bpf: Add a hint to allocated objects.

Hi,

On 6/24/2023 11:42 AM, Alexei Starovoitov wrote:
> On Fri, Jun 23, 2023 at 8:28 PM Hou Tao <houtao@...weicloud.com> wrote:
>>>                */
>>> -             obj = __llist_del_first(&c->free_by_rcu_ttrace);
>>> +             obj = llist_del_first(&c->free_by_rcu_ttrace);
>> According to the comments in llist.h, when there are concurrent
>> llist_del_first() and llist_del_all() operations, locking is needed.
> Good question.
> 1. When only one cpu is doing llist_del_first() locking is not needed.
>  This is the case here. Only this cpu is doing llist_del_first() from this 'c'.
> 2. The comments doesn't mention it, but llist_del_first() is ok on
> multiple cpus if ABA problem is addressed by other means.
Haven't checked the implementation details of lockless list. Will do
that later. "by other means" do you mean RCU ? because the reuse will be
possible only after one RCU GP.
> PS
> please trim your replies.
Sorry for the inconvenience. Will do next time.