netdev - Re: [Patch net-next 2/3] net_sched: plug in qdisc ops change_tx_queue

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]

Message-ID: <b2f31002-39b6-7a99-09ef-a9717cba7a04@gmail.com>
Date:   Tue, 23 Jan 2018 10:17:03 -0800
From:   John Fastabend <john.fastabend@...il.com>
To:     Cong Wang <xiyou.wangcong@...il.com>
Cc:     Linux Kernel Network Developers <netdev@...r.kernel.org>
Subject: Re: [Patch net-next 2/3] net_sched: plug in qdisc ops
 change_tx_queue_len

On 01/23/2018 10:07 AM, Cong Wang wrote:
> On Mon, Jan 22, 2018 at 2:44 PM, Cong Wang <xiyou.wangcong@...il.com> wrote:
>>
>> Good point, but netdev_for_each_tx_queue() doesn't handle errors
>> in current code base, it is not trivial to make it returning int to reflect
>> errors, since we have to rollback a partial failure too.
> 
> I think for now we can just check for errors but not rollback,
> at least attach_default_qdiscs() does not rollback either.
> And I will add a comment saying we need to do it in future.
> 

Sorry I delayed responding was thinking about some solutions. Unrolling
is problematic because we may fail again on the unroll :( so I guess
for this to work we would need a two phase operation with a prep and
commit phase.

So agree for now checking for errors seems reasonable and is at least
better than without the patches. And as you say another patch set can
cleanup the bigger issue of rollback.

Thanks,
.John