objc console new design
Andrew Jeffery
andrew at aj.id.au
Thu Apr 16 13:01:30 AEST 2020
On Wed, 15 Apr 2020, at 04:37, Vijay Khemka wrote:
>
>
> On 4/9/20, 9:47 PM, "Andrew Jeffery" <andrew at aj.id.au> wrote:
>
>
>
> On Tue, 7 Apr 2020, at 05:11, Vijay Khemka wrote:
> >
> >
> > On 4/5/20, 5:10 AM, "Andrew Jeffery" <andrew at aj.id.au> wrote:
> >
> >
> >
> > On Sat, 4 Apr 2020, at 08:09, Vijay Khemka wrote:
> > >
> > > Hi Andrew,
> > >
> > > How is server launched as per new design code. With latest
> update,
> > > console server is not launching in our platform. I saw
> config option
> > > “OBMC_CONSOLE_HOST_TTY ?= "ttyVUART0" “ has been removed
> from bb file
> > > and also there is no triggering point in obmc-console-ssh
> socket and
> > > service like earlier
> Wants=obmc-console@{OBMC_CONSOLE_HOST_TTY}.service
> > >
> <mailto:Wants=obmc-console@%7bOBMC_CONSOLE_HOST_TTY%7d.service>.
> > >
> > >
> > > I see udev rules added in latest patch for launching server
> which needs
> > > config definition of udev, do we need this for launching,
> please advise.
> >
> > Yes, obmc-console now ships a udev rules file that launches
> > obmc-console-server
> > via an `ENV{SYSTEMD_WANTS}` directive:
> >
> > If this is mandatory then why do we need to enable UDEV config as
> it
> > should be enabled by default.
> >
>
> It's not mandatory, just I don't know the requirements of any systems that
> aren't using udev. The switch exists in obmc-console so you can implement
> support for systems without udev if that's desired. Maybe adding the switch
> was premature?
>
> As per new design, udev rule is the only one who can start server .
That's not true, you could implement the system as Kumar's original patch[1]
if your system doesn't use udev. But I'm not aware of systems that aren't
using udev.
Andrew
[1] https://gerrit.openbmc-project.xyz/c/openbmc/meta-facebook/+/31091/6/recipes-phosphor/console/obmc-console_%2525.bbappend#19
More information about the openbmc
mailing list