netdev - RE: [PATCH net-next] devlink: Improve the port attributes description

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

Message-ID:
 <CY8PR12MB7195F16541A67545B4315533DC0F2@CY8PR12MB7195.namprd12.prod.outlook.com>
Date: Sat, 28 Dec 2024 03:33:03 +0000
From: Parav Pandit <parav@...dia.com>
To: Jakub Kicinski <kuba@...nel.org>
CC: "netdev@...r.kernel.org" <netdev@...r.kernel.org>, "jiri@...nulli.us"
	<jiri@...nulli.us>, "davem@...emloft.net" <davem@...emloft.net>,
	"edumazet@...gle.com" <edumazet@...gle.com>, "pabeni@...hat.com"
	<pabeni@...hat.com>, "horms@...nel.org" <horms@...nel.org>, Shay Drori
	<shayd@...dia.com>, Mark Bloch <mbloch@...dia.com>, Jiri Pirko
	<jiri@...dia.com>
Subject: RE: [PATCH net-next] devlink: Improve the port attributes description


> From: Jakub Kicinski <kuba@...nel.org>
> Sent: Friday, December 27, 2024 11:49 PM
> 
> On Tue, 24 Dec 2024 03:40:34 +0000 Parav Pandit wrote:
> > > On Thu, 19 Dec 2024 17:01:58 +0200 Parav Pandit wrote:
> > > > Improve the description of devlink port attributes PF, VF and SF
> > > > numbers.
> > >
> > > Please provide more context. It's not obvious why you remove PF from
> > > descriptions but not VF or SF.
> >
> > 'PF number' was vague and source of confusion. Some started thinking that
> it is some kind of index like how VF number is an index.
> > So 'PF number' is rewritten to bring the clarity that it's the function number
> of the PCI device which is very will described in the PCI spec.
> 
> Just to make sure I understand - you're trying to emphasize that the PF
> number is just an arbitrary ID of the PCIe PF within the chip, not necessarily
> related to any BDF numbering sequence?
Right. The proposed kdoc update clearly defines the pf as the 'F' of the BDF
that matches the PCIe spec, reflecting the existing implementation of
multiple drivers.

Also, the patch is clarifying the VF number (that is slightly deviated from
the pci spec).

> 
> If that's the case I think the motivation makes sense. But IMHO the execution
> is not ideal, I offer the fact we're having this exchange as a proof of the point
> not getting across :(
> 
> May be better to explain this in a couple of sentences somewhere (actually I
> get the feeling we already have such an explanation but I can't find it.
> Perhaps it was just talked about on the list) and then just point to that longer
> explanation in the attr kdocs?
What part of the 'longer explanation' would you like to have in the attr kdocs?
Do you mean to tell that it is the 'F' of the BDF?
'PCI function number' as wrote in v1 is a string that one can grep very easily
in the PCI spec; hence I didn't write it as 'F' of the BDF..

Can you please check v1 commit log - is the motivation explained enough?

> 
> > For VF number, the description is added describing it's an index starting
> from 0 (unlike pci spec where vf number starts from 1).
> > SF number is user supplied number so nothing to remove there.
> 
> nit: -EOUTLOOK.. please wrap the lines in your replies at 80 chars.
Ack.