netdev - Re: mlxsw and rtnl lock

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: <59774544-88d7-1f2a-82c6-28bb6c7ac747@gmail.com>
Date:   Tue, 29 Aug 2017 14:04:59 -0600
From:   David Ahern <dsahern@...il.com>
To:     Arkadi Sharshevsky <arkadis@...lanox.com>,
        Ido Schimmel <idosch@...sch.org>
Cc:     Jiri Pirko <jiri@...nulli.us>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        mlxsw@...lanox.com
Subject: Re: mlxsw and rtnl lock

On 8/29/17 12:10 AM, Arkadi Sharshevsky wrote:
> 
> 
> On 08/28/2017 09:00 PM, David Ahern wrote:
>> On 8/26/17 11:04 AM, Ido Schimmel wrote:
>>> Regarding the silent abort, that's intentional. You can look at the same
>>> code in v4.9 - when the chain was still blocking - and you'll see that
>>> we didn't propagate the error even then. This was discussed in the past
>>> and the conclusion was that user doesn't expect to operation to fail. If
>>> hardware resources are exceeded, we let the kernel take care of the
>>> forwarding instead.
>>>
>>
>> In addition to Roopa's comments... The silent abort is not a good user
>> experience. Right now it's add a network address or route, cross fingers
>> and hope it does not overflow some limit (nexthop, ecmp, neighbor,
>> prefix, etc) that triggers the offload abort.
>>
>> The mlxsw driver queries for some limits (e.g., max rifs) but I don't
>> see any query related to current usage, and there is no API to pass any
>> of that data to user space so user space has no programmatic way to
>> handle this. I realize you are aware of this limitation. The point is to
>> emphasize the need to resolve this.
>>
> 
> We actually thought about providing he user some tools to understand
> the ASIC's limitations by introducing the 'resource' object to devlink.
> 
> By linking dpipe tables to resources the user can understand which
> hardware processes share a common resource, furthermore this resources
> usage could be observed. By this more visibility can be obtained.
> 
> Its not a remedy for the silent abort, but, maybe a notification
> can be sent from devlink in case of abort that some resources is
> full.
> 
> This proposition was sent as RFC several weeks ago.
> 

Do you have patches (kernel and devlink) for the proposal?