netdev - Re: [net 09/15] net/mlx5e: Forbid devlink reload if IPSec rules are offloaded

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: Wed, 22 Nov 2023 22:12:15 -0800
From: Saeed Mahameed <saeed@...nel.org>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Leon Romanovsky <leon@...nel.org>, Jiri Pirko <jiri@...nulli.us>,
	"David S. Miller" <davem@...emloft.net>,
	Paolo Abeni <pabeni@...hat.com>, Eric Dumazet <edumazet@...gle.com>,
	Saeed Mahameed <saeedm@...dia.com>, netdev@...r.kernel.org,
	Tariq Toukan <tariqt@...dia.com>, Jianbo Liu <jianbol@...dia.com>
Subject: Re: [net 09/15] net/mlx5e: Forbid devlink reload if IPSec rules are
 offloaded

On 22 Nov 19:53, Jakub Kicinski wrote:
>On Wed, 22 Nov 2023 13:28:32 +0200 Leon Romanovsky wrote:
>> Unfortunately not, we (mlx5) were forced by employer of one of
>> the netdev maintainers to keep uplink netdev in devlink reload
>> while we are in eswitch.
>
>The way you phrased this makes it sound like employers of netdev
>maintainers get to exert power over this community.
>

I think Leon is just misinformed, the mlx5 netdev behavior Leon is
talking about was already removed and has nothing to do with eswitch,
and even that was never required by any employer or maintainer,
sorry for the confusion .. 

>This is an unacceptable insinuation.
>
>DEVLINK_RELOAD_LIMIT_NO_RESET should not cause link loss, sure.
>Even if Meta required that you implemented that (which it does
>not, AFAIK) - it's just an upstream API.
>

We only support this limit for FW_ACTIVATE_ACTION, and has no issue
in this flow.

Leon's issue is with internal mlx5 uplink implementation where on eswitch
mode changes we don't unregister the netdev which causes eswitch resource
leaks with ipsec rules, since we move eswitch to legacy mode on devlink
reload then the same issue happens on relaod, hence he needs to block it
in this patch, and we are already discussing a new design to fix devlink
reload in net-next.

This is Just a bug and has nothing to do with any requirements from anyone.

Thanks.