From: Christian Hopps <chopps@chopps•org>
To: Sabrina Dubroca <sd@queasysnail•net>
Cc: Christian Hopps <chopps@chopps•org>,
devel@linux-ipsec•org,
Steffen Klassert <steffen.klassert@secunet•com>,
netdev@vger•kernel.org, Christian Hopps <chopps@labn•net>
Subject: Re: [PATCH ipsec-next v5 06/17] xfrm: add mode_cbs module functionality
Date: Wed, 31 Jul 2024 14:41:46 -0400 [thread overview]
Message-ID: <m2ikwl8kjf.fsf@ja.int.chopps.org> (raw)
In-Reply-To: <m2mslx8kwx.fsf@ja.int.chopps.org>
Christian Hopps <chopps@chopps•org> writes:
> Sabrina Dubroca <sd@queasysnail•net> writes:
>
>> 2024-07-30, 17:29:06 -0400, Christian Hopps wrote:
>>>
>>> Sabrina Dubroca <sd@queasysnail•net> writes:
>>>
>>> > 2024-07-14, 16:22:34 -0400, Christian Hopps wrote:
>>> > > +struct xfrm_mode_cbs {
>>> >
>>> > It would be nice to add kdoc for the whole thing.
>>>
>>> Ok, I'll move the inline comments to a kdoc. FWIW, all the other structs in
>>> this header, including the main `xfrm_state` struct use the same inline
>>> comment documentation style I copied.
>>
>> Sure, but I don't think we should model new code on old habits.
>>
>>> > > diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
>>> > > index 7cee9c0a2cdc..6ff05604f973 100644
>>> > > --- a/net/xfrm/xfrm_input.c
>>> > > +++ b/net/xfrm/xfrm_input.c
>>> > > @@ -494,6 +497,10 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
>>> > >
>>> > > family = x->props.family;
>>> > >
>>> > > + /* An encap_type of -3 indicates reconstructed inner packet */
>>> >
>>> > And I think it's time to document all the encap_types above the
>>> > function (and in particular, how xfrm_inner_mode_input/encap_type=-3
>>> > pair together), and/or define some constants. Also, is -2 used
>>> > anywhere (I only see -1 and -3)? If not, then why -3?
>>>
>>> At the time this was added ISTR that there was some belief that -2
>>> was used perhaps in an upcoming patch, so I picked -3. I can't find
>>> a -2 use case though so I will switch to -2 instead.
>>>
>>> Re documentation: I think the inline comments where encap_type is
>>> used is sufficient documentation for the 2 negative values.
>>
>> I don't think it is. Inline comments are good to explain the internal
>> behavior, but that's more external behavior.
>
>>> There's
>>> a lot going on in this function and someone wishing to change (or
>>> understand) something is going to have to walk the code and use
>>> cases regardless of a bit of extra verbiage on the encap_value
>>> beyond what's already there. Fully documenting how xfrm_input works
>>> (in all it's use cases) seems beyond the scope of this patch to me.
>>
>> Sure, and that's really not what I'm asking for here. Something like
>> "encap_type=-3 makes xfrm_input jump right back to where it stopped
>> when xfrm_inner_mode_input returned -EINPROGRESS" is useful without
>> having to dive into the mess that is xfrm_input.
>
> If I'm not adding your suggested text into an inline comment where am I doing this?
>
> Bear in mind that encap_type can also have non-negative values, am I documenting
> all these cases too? It just seems like going down this path is asking for the
> entire function to be documented, perhaps I'm missing something though.
>
> Are other people going to be OK with a top of function comment that only documents the single (now) `-2` value for encap_type?
Actually, I will document the "Resume" negative value special cases in a top of function comment. If we want more generalized documentation for this function I think it should be in a different patch/project.
Thanks,
Chris.
> Thanks,
> Chris.
next prev parent reply other threads:[~2024-07-31 18:48 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-14 20:22 [PATCH ipsec-next v5 00/17] Add IP-TFS mode to xfrm Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 01/17] xfrm: config: add CONFIG_XFRM_IPTFS Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 02/17] include: uapi: add ip_tfs_*_hdr packet formats Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 03/17] include: uapi: add IPPROTO_AGGFRAG for AGGFRAG in ESP Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 04/17] xfrm: sysctl: allow configuration of global default values Christian Hopps
2024-07-18 13:13 ` Sabrina Dubroca
2024-07-18 13:15 ` Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 05/17] xfrm: netlink: add config (netlink) options Christian Hopps
2024-07-18 13:45 ` Sabrina Dubroca
2024-07-18 14:08 ` [devel-ipsec] " Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 06/17] xfrm: add mode_cbs module functionality Christian Hopps
2024-07-19 6:14 ` [devel-ipsec] " Florian Westphal
2024-07-20 0:06 ` Christian Hopps
2024-07-25 13:32 ` Sabrina Dubroca
2024-07-30 21:29 ` Christian Hopps
2024-07-31 17:10 ` Sabrina Dubroca
2024-07-31 18:32 ` Christian Hopps
2024-07-31 18:41 ` Christian Hopps [this message]
2024-07-14 20:22 ` [PATCH ipsec-next v5 07/17] xfrm: add generic iptfs defines and functionality Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 08/17] xfrm: iptfs: add new iptfs xfrm mode impl Christian Hopps
2024-07-15 12:39 ` Simon Horman
2024-07-18 5:56 ` [devel-ipsec] " Christian Hopps
2024-07-19 6:16 ` Florian Westphal
2024-07-20 0:30 ` Christian Hopps
2024-07-25 13:33 ` Sabrina Dubroca
2024-07-25 14:56 ` Christian Hopps
2024-07-31 16:29 ` Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 09/17] xfrm: iptfs: add user packet (tunnel ingress) handling Christian Hopps
2024-07-15 12:55 ` Simon Horman
2024-07-18 5:35 ` [devel-ipsec] " Christian Hopps
2024-07-18 6:32 ` Christian Hopps
2024-07-16 14:00 ` kernel test robot
2024-07-14 20:22 ` [PATCH ipsec-next v5 10/17] xfrm: iptfs: share page fragments of inner packets Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 11/17] xfrm: iptfs: add fragmenting of larger than MTU user packets Christian Hopps
2024-07-15 13:12 ` Simon Horman
2024-07-18 5:57 ` [devel-ipsec] " Christian Hopps
2024-07-18 6:56 ` Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 12/17] xfrm: iptfs: add basic receive packet (tunnel egress) handling Christian Hopps
2024-07-15 13:16 ` Simon Horman
2024-07-18 6:14 ` [devel-ipsec] " Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 13/17] xfrm: iptfs: handle received fragmented inner packets Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 14/17] xfrm: iptfs: add reusing received skb for the tunnel egress packet Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 15/17] xfrm: iptfs: add skb-fragment sharing code Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 16/17] xfrm: iptfs: handle reordering of received packets Christian Hopps
2024-07-14 20:22 ` [PATCH ipsec-next v5 17/17] xfrm: iptfs: add tracepoint functionality Christian Hopps
2024-07-15 11:53 ` kernel test robot
2024-07-17 5:21 ` kernel test robot
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=m2ikwl8kjf.fsf@ja.int.chopps.org \
--to=chopps@chopps$(echo .)org \
--cc=chopps@labn$(echo .)net \
--cc=devel@linux-ipsec$(echo .)org \
--cc=netdev@vger$(echo .)kernel.org \
--cc=sd@queasysnail$(echo .)net \
--cc=steffen.klassert@secunet$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox