linux-kernel - Re: [PATCH/RFC] llist: add llist_[add|del_first]

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>] [day] [month] [year] [list]

Message-ID: <87h8sbsmoq.fsf@yhuang-dev.intel.com>
Date:   Thu, 28 Dec 2017 08:47:33 +0800
From:   "Huang\, Ying" <ying.huang@...el.com>
To:     Vitaly Wool <vitalywool@...il.com>
Cc:     LKML <linux-kernel@...r.kernel.org>, <Oleksiy.Avramchenko@...y.com>
Subject: Re: [PATCH/RFC] llist: add llist_[add|del_first]_exclusive

Vitaly Wool <vitalywool@...il.com> writes:

> 2017-12-22 14:57 GMT+01:00 Huang, Ying <ying.huang@...el.com>:
>
>> Vitaly Wool <vitalywool@...il.com> writes:
>>
>> > 2017-12-20 1:57 GMT+01:00 Huang, Ying <ying.huang@...el.com>:
>> >
>> > <snip>
>> >
>> >>
>> >> > Could you please elaborate how this would be implemented "on top"?
>> >>
>> >> struct llist_node *my_del_first_exclusive(struct llist_head *head)
>> >> {
>> >>         struct llist_node *node = llist_del_first(head);
>> >>
>> >>         if (node)
>> >>                 node->next = LLIST_NODE_UNLISTED;
>> >> }
>> >>
>> >> bool my_add_exclusive(struct llist_node *node, struct llist_head *head)
>> >> {
>> >>         if (node->next != LLIST_NODE_UNLIST)
>> >>                 return false;
>> >>         if (cmpxchg(&node->next, LLIST_NODE_UNLIST, NULL) !=
>> >>             LLIST_NODE_UNLIST)
>> >>                 return false;
>> >>         llist_add(node, head);
>> >>         return true;
>> >> }
>> >>
>> >
>> > That would work, thanks. I'll update the patch.
>>
>> Is there any other users except your code?  If no, I think it's better
>> to keep it with its only user instead part of llist library at least for
>> now.  Because I don't know whether the usage model is popular.
>>
>>
>>
> I'm going to come up with a patch to binder that uses these specific
> extensions, so I expect there'll be some users for this.

I think it is better to keep this in binder unless there are other
users.

Best Regards,
Huang, Ying

> Thanks,
>    Vitaly