linux-kernel - Re: [PATCH v3 3/4] can: skb:: move can_dropped_invalid_skb and can_skb_headroom

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: <22590a57-c7c6-39c6-06d5-11c6e4e1534b@hartkopp.net>
Date:   Tue, 17 May 2022 16:35:05 +0200
From:   Oliver Hartkopp <socketcan@...tkopp.net>
To:     Max Staudt <max@...as.org>
Cc:     Marc Kleine-Budde <mkl@...gutronix.de>,
        Vincent MAILHOL <mailhol.vincent@...adoo.fr>,
        linux-can@...r.kernel.org, linux-kernel@...r.kernel.org,
        netdev@...r.kernel.org
Subject: Re: [PATCH v3 3/4] can: skb:: move can_dropped_invalid_skb and
 can_skb_headroom_valid to skb.c



On 17.05.22 15:43, Max Staudt wrote:
> On Tue, 17 May 2022 15:35:03 +0200
> Oliver Hartkopp <socketcan@...tkopp.net> wrote:
> 
>> Oh, I didn't want to introduce two new kernel modules but to have
>> can_dev in different 'feature levels'.
> 
> Which I agree is a nice idea, as long as heisenbugs can be avoided :)
> 
> (as for the separate modules vs. feature levels of can-dev - sorry, my
> two paragraphs were each referring to a different idea. I mixed them
> into one single email...)
> 
> 
> Maybe the can-skb and rx-offload parts could be a *visible* sub-option
> of can-dev in Kconfig, which is normally optional, but immediately
> force-selected once a CAN HW driver is selected?

I think it should be even more simple.

When you enter the current Kconfig page of "CAN Device Drivers" every 
selection of vcan/vxcan/slcan should directly select CAN_DEV_SW.

The rest could stay the same, e.g. selecting CAN_DEV "Platform CAN 
drivers with Netlink support" which then enables CAN_CALC_BITTIMING and 
CAN_LEDS to be selectable. Which also makes sure the old .config files 
still apply.

And finally the selection of flexcan, ti_hecc and
mcp251xfd automatically selects CAN_DEV_RX_OFFLOAD.

Then only some more Makefile magic has to be done to build can-dev.ko 
accordingly.

Best regards,
Oliver



> 
> 
>> But e.g. the people that are running Linux instances in a cloud only
>> using vcan and vxcan would not need to carry the entire
>> infrastructure of CAN hardware support and rx-offload.
> 
> Out of curiosity, do you have an example use case for this vcan cloud
> setup? I can't dream one up...
> 
> 
> 
> Max