openbmc.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Konstantin Aladyshev <aladyshev22@gmail.com>
To: openbmc@lists.ozlabs.org, a.kartashev@yadro.com,
	 Deepak Kodihalli <deepak.kodihalli.83@gmail.com>
Subject: Re: Boot Source Override feature problems
Date: Wed, 23 Jun 2021 21:08:58 +0300	[thread overview]
Message-ID: <CACSj6VXYXiDhZo_=SkXZxMszMeN+fcAdPk_vuW4mqzdJKuWMFQ@mail.gmail.com> (raw)

On Wed, Jun 23, 2021 at 8:09 PM Andrei Kartashev <a.kartashev@yadro.com> wrote:
>
> On Wed, 2021-06-23 at 19:36 +0300, Konstantin Aladyshev wrote:
> > We need to have two `xyz.openbmc_project.Object.Enable` interfaces:
> > 1) one that stores information whether the BootSourceOverride feature
> > is enabled or not,
> > 2) another one that stores information whether the BootSourceOverride
> > feature is permanent or one_time.
> >
> > The current implementation has only 1), proposed design has both 1)
> > and 2):
> > ```
> >  /xyz/openbmc_project/control/host0/boot:
> >       - Interface: xyz.openbmc_project.Control.Boot.Source
> >       - Interface: xyz.openbmc_project.Control.Boot.Mode
> >       - Interface: xyz.openbmc_project.Control.Boot.Type
> >       - Interface: xyz.openbmc_project.Object.Enable              <--
> > -------  1)
> >  /xyz/openbmc_project/control/host0/boot/one_time:
> >       - Interface: xyz.openbmc_project.Object.Enable              <--
> > -------  2)
> > ```
> >
>
> Right, but your initial proposal was to use "Permanent" flag instead of
> second one:
>     - Interface: xyz.openbmc_project.Control.Boot.Permanent #
> true/false

To use something like  `xyz.openbmc_project.Control.Boot.Permanent` we
need to create another interface in the `phosphor-dbus-interfaces`
repository, which is something the OpenBMC community want to avoid
without necessities.
As the same functionality can be achieved with the
`xyz.openbmc_project.Object.Enable` interface under
`/xyz/openbmc_project/control/host0/boot` I've decided to take this
approach instead.

>
> BTW, can anyone explain me, why do we have all this as separate
> interfaces with only one properties?
> As for me, it should be one interface
> "xyz.openbmc_project.Control.Boot" with several properties.
>

I don't know.

> > Also earlier there were two sets of Boot.Source/Boot.Mode/Boot.Type
> > settings. But the second one isn't really necessary as the
> > BootSourceOverride feature doesn't fallback to permanent override
> > after one-time override. So we need to keep only one set of
> > Boot.Source/Boot.Mode/Boot.Type settings.
> >
>
> The fact that it was broken doesn't automatically mean that this need
> to be removed. May be it worth to fix this...
>

I didn't mean that something was broken here. I just meant that the
typical BootSourceOverride functionality in a BMC is not intended to
work this way.

> But here I tend to agree that
> "/xyz/openbmc_project/control/host0/boot/one_time" can be wasted. I
> just think it can be wasted completely. Then you can use GetAll on
> /xyz/openbmc_project/control/host0/boot to get all required data.
>

Currently "/xyz/openbmc_project/control/host0/boot/one_time" is used
to keep two separate `xyz.openbmc_project.Object.Enable` objects, so
we can't waste it.


> > Best regards,
> > Konstantin Aladyshev
>
> --
> Best regards,
> Andrei Kartashev
>
>

             reply	other threads:[~2021-06-23 18:09 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-23 18:08 Konstantin Aladyshev [this message]
  -- strict thread matches above, loose matches on Subject: below --
2021-06-23 16:36 Boot Source Override feature problems Konstantin Aladyshev
2021-05-26 20:50 Konstantin Aladyshev
2021-06-11 12:17 ` Deepak Kodihalli
2021-06-21 15:59   ` Konstantin Aladyshev
2021-06-22 16:49     ` Andrei Kartashev

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='CACSj6VXYXiDhZo_=SkXZxMszMeN+fcAdPk_vuW4mqzdJKuWMFQ@mail.gmail.com' \
    --to=aladyshev22@gmail.com \
    --cc=a.kartashev@yadro.com \
    --cc=deepak.kodihalli.83@gmail.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).