From: Andrei Kartashev <a.kartashev@yadro.com>
To: "Garrett, Mike (HPE Server Firmware)" <mike.garrett@hpe.com>
Cc: OpenBMC Maillist <openbmc@lists.ozlabs.org>
Subject: Re: Leveraging and extending smbios-mdr
Date: Wed, 18 Aug 2021 09:33:44 +0300 [thread overview]
Message-ID: <0bf2da56744a03e1597d75e88e555ea4363448da.camel@yadro.com> (raw)
In-Reply-To: <DF4PR8401MB0634FB1810144BBDB9C42E158FFE9@DF4PR8401MB0634.NAMPRD84.PROD.OUTLOOK.COM>
You can retrigger read via dbus, using
xyz.openbmc_project.Smbios.MDR_V2->.AgentSynchronizeData call to
smbios-mdrv2.
See example here:
https://github.com/openbmc/intel-ipmi-oem/blob/master/src/smbiosmdrv2handler.cpp#L1240
On Tue, 2021-08-17 at 21:29 +0000, Garrett, Mike (HPE Server Firmware)
wrote:
> Actually I'm talking about https://github.com/openbmc/smbios-mdr
>
> It looks like this daemon tries to take in a data file at startup only
> (/var/lib/smbios/smbios/smbios2). I'm wondering if there's a provision
> in the design to refresh the data file if appears or is updated later.
> This file (at least in my experience) has to be created by an
> interaction with the BIOS during boot. So it would be very reasonable
> for the BMC to boot before the host and there to be no file available
> yet. Also, if something is updated on the host, a reboot may refresh
> this file. So I think there are a couple of reasons why it would be
> good for this repo to react to data file updates.
>
> If it doesn't exist and folks think it's a good idea, we can look at
> creating that capability, but I want to make sure it doesn't already
> exist.
>
> Mike
>
> > -----Original Message-----
> > From: Artem Senichev <artemsen@gmail.com>
> > Sent: Thursday, July 29, 2021 2:49 AM
> > To: Garrett, Mike (HPE Server Firmware) <mike.garrett@hpe.com>
> > Cc: OpenBMC Maillist <openbmc@lists.ozlabs.org>; Pedrana, Nick
> > <allan.nic.pedrana@hpe.com>
> > Subject: Re: Leveraging and extending smbios-mdr
> >
> > On Tue, Jul 27, 2021 at 03:47:04PM +0000, Garrett, Mike (HPE Server
> > Firmware) wrote:
> > > Hello Artem and all,
> > >
> > > We are integrating smbios-mdr into our build. Our SMBIOS info is
> > downloaded and built as a file in the filesystem using proprietary
> > means (our
> > CHIF - channel interface). Our hope is that we can simply point the
> > smbios-
> > mdr service at it and light up new information about the platform in
> > dbus,
> > Redfish, and the GUI.
> > >
> > > However, I can't find much in the way of documentation about
> > > smbios-
> > mdr.
> >
> > If we are talking about the MDR2 project (https://github.com/Intel-
> > BMC/mdrv2),
> > there is no documentation, but I don't think we really need it.
> > It is just a SMBIOS dump parser for some table types (CPU, DIMM).
> > The table format is fully documented in the SMBIOS specification:
> > https://www.dmtf.org/standards/smbios
> > !NpxR!xpRhPV3JWN6U7DS7nROiFT5h3CNwyuFM9FLOZhUK0yNczSR8w5MX7
> > iq1WAqwQuc$
> >
> > > I'm especially interested in how to extend the functionality past
> > > just CPUs
> > and DIMMs into things like OEM records supplied by the BIOS.
> >
> > What about phosphor-inventory-manager?
> >
> > > Seems like the possibilities are:
> > >
> > > 1. Smbios-mdr has a generic dbus API and we write an independent
> > > service to query it for this info and publish to dbus 2. We create
> > > a
> > > generic extension mechanism for smbios-mdr for vendors to add OEM
> > support and push to dbus directly from smbios-mdr (like w/
> > CPUs/DIMMs) 3.
> > We fork smbios-mdr and extend it (not preferred).
> > >
> > > I'd like to hear how best to leverage this recipe.
> >
> > --
> > Regards,
> > Artem Senichev
> > Software Engineer, YADRO.
--
Best regards,
Andrei Kartashev
prev parent reply other threads:[~2021-08-18 6:34 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-27 15:47 Leveraging and extending smbios-mdr Garrett, Mike (HPE Server Firmware)
2021-07-29 7:49 ` Artem Senichev
2021-08-17 21:29 ` Garrett, Mike (HPE Server Firmware)
2021-08-18 6:33 ` Andrei Kartashev [this message]
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=0bf2da56744a03e1597d75e88e555ea4363448da.camel@yadro.com \
--to=a.kartashev@yadro.com \
--cc=mike.garrett@hpe.com \
--cc=openbmc@lists.ozlabs.org \
/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).