From: Jakub Kicinski <jakub.kicinski@netronome.com>
To: Martin KaFai Lau <kafai@fb.com>
Cc: Okash Khawaja <osk@fb.com>,
Daniel Borkmann <daniel@iogearbox.net>,
"Alexei Starovoitov" <ast@kernel.org>, Yonghong Song <yhs@fb.com>,
Quentin Monnet <quentin.monnet@netronome.com>,
"David S. Miller" <davem@davemloft.net>, <netdev@vger.kernel.org>,
<kernel-team@fb.com>, <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH bpf-next 2/3] bpf: btf: add btf json print functionality
Date: Thu, 21 Jun 2018 17:25:23 -0700 [thread overview]
Message-ID: <20180621172523.6cd00ed1@cakuba.netronome.com> (raw)
In-Reply-To: <20180621235746.dfq6kdtkogftw3ws@kafai-mbp.dhcp.thefacebook.com>
On Thu, 21 Jun 2018 16:58:15 -0700, Martin KaFai Lau wrote:
> On Thu, Jun 21, 2018 at 04:07:19PM -0700, Jakub Kicinski wrote:
> > On Thu, 21 Jun 2018 15:51:17 -0700, Martin KaFai Lau wrote:
> > > On Thu, Jun 21, 2018 at 02:59:35PM -0700, Jakub Kicinski wrote:
> > > > On Wed, 20 Jun 2018 13:30:53 -0700, Okash Khawaja wrote:
> > > > > $ sudo bpftool map dump -p id 14
> > > > > [{
> > > > > "key": 0
> > > > > },{
> > > > > "value": {
> > > > > "m": 1,
> > > > > "n": 2,
> > > > > "o": "c",
> > > > > "p": [15,16,17,18,15,16,17,18
> > > > > ],
> > > > > "q": [[25,26,27,28,25,26,27,28
> > > > > ],[35,36,37,38,35,36,37,38
> > > > > ],[45,46,47,48,45,46,47,48
> > > > > ],[55,56,57,58,55,56,57,58
> > > > > ]
> > > > > ],
> > > > > "r": 1,
> > > > > "s": 0x7ffff6f70568,
> > > > > "t": {
> > > > > "x": 5,
> > > > > "y": 10
> > > > > },
> > > > > "u": 100,
> > > > > "v": 20,
> > > > > "w1": 0x7,
> > > > > "w2": 0x3
> > > > > }
> > > > > }
> > > > > ]
> > > >
> > > > I don't think this format is okay, JSON output is an API you shouldn't
> > > > break. You can change the non-JSON output whatever way you like, but
> > > > JSON must remain backwards compatible.
> > > >
> > > > The dump today has object per entry, e.g.:
> > > >
> > > > {
> > > > "key":["0x00","0x00","0x00","0x00",
> > > > ],
> > > > "value": ["0x02","0x00","0x00","0x00","0x00","0x00","0x00","0x00"
> > > > ]
> > > > }
> > > >
> > > > This format must remain, you may only augment it with new fields. E.g.:
> > > >
> > > > {
> > > > "key":["0x00","0x00","0x00","0x00",
> > > > ],
> > > > "key_struct":{
> > > > "index":0
> > > > },
> > > > "value": ["0x02","0x00","0x00","0x00","0x00","0x00","0x00","0x00"
> > > > ],
> > > > "value_struct":{
> > > > "src_ip":2,
> > > > "dst_ip:0
> > > > }
> > > > }
> > > I am not sure how useful to have both "key|value" and "(key|value)_struct"
> > > while most people would prefer "key_struct"/"value_struct" if it is
> > > available.
> >
> > Agreed, it's not that useful, especially with the string-hex debacle :(
> > It's just about the backwards compat.
> >
> > > How about introducing a new option, like "-b", to print the
> > > map with BTF (if available) such that it won't break the existing
> > > one (-j or -p) while the "-b" output can keep using the "key"
> > > and "value".
> > >
> > > The existing json can be kept as is.
> >
> > That was my knee jerk reaction too, but on reflection it doesn't sound
> > that great. We expect people with new-enough bpftool to use btf, so it
> > should be available in the default output, without hiding it behind a
> > switch. We could add a switch to hide the old output, but that doesn't
> > give us back the names... What about Key and Value or k and v? Or
> > key_fields and value_fields?
> I thought the current default output is "plain" ;)
> Having said that, yes, the btf is currently printed in json.
>
> Ideally, the default json output should do what most people want:
> print btf and btf only (if it is available).
> but I don't see a way out without new option if we need to
> be backward compat :(
>
> Agree that showing the btf in the existing json output will be useful (e.g.
> to hint people that BTF is available). If btf is showing in old json,
> also agree that the names should be the same with the new json.
> key_fields and value_fields may hint it has >1 fields though.
> May be "formatted_key" and "formatted_value"?
SGTM! Or even maybe as a "formatted" object?:
{
"key":["0x00","0x00","0x00","0x00",
],
"value": ["0x02","0x00","0x00","0x00","0x00","0x00","0x00","0x00"
],
"formatted":{
"key":{
"index":0
},
"value":{
"src_ip":2,
"dst_ip:0
}
}
}
> > > > The name XYZ_struct may not be the best, perhaps you can come up with a
> > > > better one?
> > > >
> > > > Does that make sense? Am I missing what you're doing here?
> > > >
> > > > One process note - please make sure you run checkpatch.pl --strict on
> > > > bpftool patches before posting.
> > > >
> > > > Thanks for working on this!
next prev parent reply other threads:[~2018-06-22 0:25 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-20 20:30 [PATCH bpf-next 0/3] bpf: btf: json print btf info with bpftool map dump Okash Khawaja
2018-06-20 20:30 ` [PATCH bpf-next 1/3] bpf: btf: export btf types and name by offset from lib Okash Khawaja
2018-06-20 22:40 ` Song Liu
2018-06-20 22:48 ` Okash Khawaja
2018-06-20 23:24 ` Song Liu
2018-06-20 20:30 ` [PATCH bpf-next 2/3] bpf: btf: add btf json print functionality Okash Khawaja
2018-06-20 23:14 ` Song Liu
2018-06-21 10:31 ` Okash Khawaja
2018-06-21 10:42 ` Quentin Monnet
2018-06-22 10:24 ` Okash Khawaja
2018-06-22 10:39 ` Quentin Monnet
2018-06-22 18:44 ` Jakub Kicinski
2018-06-21 21:59 ` Jakub Kicinski
2018-06-21 22:51 ` Martin KaFai Lau
2018-06-21 23:07 ` Jakub Kicinski
2018-06-21 23:58 ` Martin KaFai Lau
2018-06-22 0:25 ` Jakub Kicinski [this message]
2018-06-22 1:20 ` Martin KaFai Lau
2018-06-22 11:17 ` Okash Khawaja
2018-06-22 18:43 ` Jakub Kicinski
2018-06-22 18:40 ` Jakub Kicinski
2018-06-22 20:58 ` Martin KaFai Lau
2018-06-22 21:27 ` Jakub Kicinski
2018-06-22 21:49 ` Jakub Kicinski
2018-06-22 23:19 ` Martin KaFai Lau
2018-06-22 23:40 ` Jakub Kicinski
2018-06-22 23:58 ` Martin KaFai Lau
2018-06-22 22:48 ` Okash Khawaja
2018-06-22 22:54 ` Martin KaFai Lau
2018-06-22 23:32 ` Jakub Kicinski
2018-06-23 0:26 ` Martin KaFai Lau
2018-06-26 16:48 ` Okash Khawaja
2018-06-26 20:31 ` Jakub Kicinski
2018-06-26 22:27 ` Martin KaFai Lau
2018-06-26 22:35 ` Jakub Kicinski
2018-06-27 10:34 ` Daniel Borkmann
2018-06-27 11:47 ` Okash Khawaja
2018-06-27 12:56 ` Daniel Borkmann
2018-07-01 10:31 ` Okash Khawaja
2018-07-02 17:19 ` Jakub Kicinski
2018-06-20 20:30 ` [PATCH bpf-next 3/3] bpf: btf: json print map dump with btf info Okash Khawaja
2018-06-20 23:22 ` Song Liu
2018-06-21 10:05 ` Okash Khawaja
2018-06-21 10:24 ` Quentin Monnet
2018-06-21 14:26 ` Okash Khawaja
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=20180621172523.6cd00ed1@cakuba.netronome.com \
--to=jakub.kicinski@netronome.com \
--cc=ast@kernel.org \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=kafai@fb.com \
--cc=kernel-team@fb.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=osk@fb.com \
--cc=quentin.monnet@netronome.com \
--cc=yhs@fb.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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;
as well as URLs for NNTP newsgroup(s).