All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jiri Pirko <jiri@resnulli.us>
To: Jakub Kicinski <kuba@kernel.org>
Cc: Ido Schimmel <idosch@idosch.org>,
	Ido Schimmel <idosch@nvidia.com>,
	netdev@vger.kernel.org, davem@davemloft.net, pabeni@redhat.com,
	jiri@nvidia.com, petrm@nvidia.com, dsahern@gmail.com,
	andrew@lunn.ch, mlxsw@nvidia.com
Subject: Re: [PATCH net-next 00/11] mlxsw: extend line card model by devices and info
Date: Fri, 27 May 2022 09:27:47 +0200	[thread overview]
Message-ID: <YpB9cwqcSAMslKLu@nanopsycho> (raw)
In-Reply-To: <20220526103539.60dcb7f0@kernel.org>

Thu, May 26, 2022 at 07:35:39PM CEST, kuba@kernel.org wrote:
>On Thu, 26 May 2022 13:45:49 +0200 Jiri Pirko wrote:
>> >Separate instance:
>> >
>> >	for (i = 0; i < sw->num_lcs; i++) {
>> >		devlink_register(&sw->lc_dl[i]);
>> >		devlink_line_card_link(&sw->lc[i], &sw->lc_dl[i]);
>> >	}
>> >
>> >then report that under the linecard
>> >
>> >	nla_nest_start(msg, DEVLINK_SUBORDINATE_INSTANCE);
>> >	devlink_nl_put_handle(msg, lc->devlink);
>> >	nla_nest_end(msg...)
>> >
>> >then user can update the linecard like any devlink instance, switch,
>> >NIC etc. It's better code reuse and I don't see any downside, TBH.  
>> 
>> Okay, I was thinking about this a litle bit more, and I would like to
>> explore extending the components path. Exposing the components in
>> "devlink dev info" and then using them in "devlink dev flash". LC could
>> be just one of multiple potential users of components. Will send RFC
>> soon.
>
>Feel free to send a mockup of the devlink user space outputs.
>The core code for devlink is just meaningless marshaling anyway.

Okay. So the output of devlink dev info would be extended by
"components" nest. This nest would carry array of components which
contain versions. The name of the component is openin each array member
nest:

$ devlink dev info
pci/0000:01:00.0:
  driver mlxsw_spectrum2
  versions:
      fixed:
        hw.revision A0
        fw.psid MT_0000000199
      running:
        fw.version 29.2010.2302
        fw 29.2010.2302
  components:
    lc1:
      versions:
        fixed:
          hw.revision 0
          fw.psid MT_0000000111
        running:
          fw 19.2010.1310
          ini.version 4
    lc2:
      versions:
        fixed:
          hw.revision 0
          fw.psid MT_0000000111
        running:
          fw 19.2010.1310
          ini.version 4
    someothercomponentname:
      versions:
        running:
	   fw: 888

Now on top of exsisting "devlink dev flash" cmd without component, user
may specify the component name from the array above:

$ devlink dev flash pci/0000:01:00.0 component lc1 file mellanox/fw-AGB-rel-19_2010_1312-022-EVB.mfa2

$ devlink dev flash pci/0000:01:00.0 component someothercomponentname file foo.bin

Note this is generic vehicle, line cards would benefit but it is usable
for multiple ASIC FW partitions for example.

Note that on "devlink dev flash" there is no change. This is implemented
currently. Only "devlink dev info" is extended to show the component
list.

  reply	other threads:[~2022-05-27  7:27 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-25  3:44 [PATCH net-next 00/11] mlxsw: extend line card model by devices and info Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 01/11] devlink: introduce line card devices support Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 02/11] devlink: introduce line card info get message Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 03/11] devlink: introduce line card device info infrastructure Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 04/11] mlxsw: reg: Extend MDDQ by device_info Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 05/11] mlxsw: core_linecards: Probe provisioned line cards for devices and attach them Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 06/11] selftests: mlxsw: Check devices on provisioned line card Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 07/11] mlxsw: core_linecards: Expose HW revision and INI version Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 08/11] selftests: mlxsw: Check line card info on provisioned line card Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 09/11] mlxsw: reg: Extend MDDQ device_info by FW version fields Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 10/11] mlxsw: core_linecards: Expose device FW version over device info Ido Schimmel
2022-04-25  3:44 ` [PATCH net-next 11/11] selftests: mlxsw: Check device info on activated line card Ido Schimmel
2022-04-25  9:50 ` [PATCH net-next 00/11] mlxsw: extend line card model by devices and info patchwork-bot+netdevbpf
2022-04-25 16:00 ` Jakub Kicinski
2022-04-25 19:39   ` Ido Schimmel
2022-04-25 19:52     ` Jakub Kicinski
2022-04-26  6:57       ` Jiri Pirko
2022-04-26 12:11         ` Andrew Lunn
2022-04-26 12:36           ` Jiri Pirko
2022-04-26 12:41         ` Jakub Kicinski
2022-04-26 14:00           ` Jiri Pirko
2022-04-26 14:51             ` Jakub Kicinski
2022-04-27  7:35               ` Jiri Pirko
2022-04-27 14:14                 ` Jakub Kicinski
2022-04-29 11:51                   ` Jiri Pirko
2022-04-29 18:45                     ` Jakub Kicinski
2022-04-29 19:29                       ` Jiri Pirko
2022-04-29 22:38                         ` Jakub Kicinski
2022-04-30  6:27                           ` Jiri Pirko
2022-05-02 14:39                             ` Jakub Kicinski
2022-05-23  9:42                               ` Jiri Pirko
2022-05-23 17:56                                 ` Jakub Kicinski
2022-05-24  6:46                                   ` Jiri Pirko
2022-05-24 14:31                                     ` Jiri Pirko
2022-05-24 18:00                                       ` Jakub Kicinski
2022-05-25  6:20                                         ` Jiri Pirko
2022-05-25 15:50                                           ` Jakub Kicinski
2022-05-26  9:05                                             ` Jiri Pirko
2022-05-26 10:47                                               ` Jiri Pirko
2022-05-26 11:45                                             ` Jiri Pirko
2022-05-26 17:35                                               ` Jakub Kicinski
2022-05-27  7:27                                                 ` Jiri Pirko [this message]
2022-05-28  0:10                                                   ` Jakub Kicinski
2022-05-28  9:09                                                     ` Jiri Pirko
2022-05-28 19:02                                                       ` Jakub Kicinski
2022-05-29  9:23                                                         ` Jiri Pirko
2022-05-30 19:54                                                           ` Jakub Kicinski
2022-05-31  7:11                                                             ` Jiri Pirko
2022-05-31 15:05                                                               ` Jakub Kicinski
2022-05-31 15:51                                                                 ` Jiri Pirko
2022-05-31 16:08                                                                   ` Jakub Kicinski
2022-05-31 19:34                                                                     ` Jiri Pirko
2022-05-31 22:41                                                                       ` Jakub Kicinski
2022-06-01  7:35                                                                         ` Jiri Pirko
2022-05-28 15:58                                       ` David Ahern
2022-05-29  9:24                                         ` Jiri Pirko
2022-05-31  2:11                                           ` David Ahern
2022-05-31  7:05                                             ` Jiri Pirko
2022-04-26 15:24             ` Andrew Lunn
2022-04-27  7:37               ` Jiri Pirko
2022-04-26  6:47     ` Jiri Pirko
2022-04-26 12:27       ` Andrew Lunn
2022-04-26 12:41         ` Jiri Pirko
2022-04-26 13:45           ` Andrew Lunn
2022-04-26 14:05             ` Jiri Pirko
2022-04-26 15:36               ` Andrew Lunn

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=YpB9cwqcSAMslKLu@nanopsycho \
    --to=jiri@resnulli.us \
    --cc=andrew@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=dsahern@gmail.com \
    --cc=idosch@idosch.org \
    --cc=idosch@nvidia.com \
    --cc=jiri@nvidia.com \
    --cc=kuba@kernel.org \
    --cc=mlxsw@nvidia.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=petrm@nvidia.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.