All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
To: Mario.Limonciello@dell.com
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: DP Aux interfaces inquiry
Date: Wed, 25 Jan 2017 20:40:48 +0200	[thread overview]
Message-ID: <20170125184048.GY31595@intel.com> (raw)
In-Reply-To: <ba495b4d4eed48db878ad3c98c618348@ausx13mpc120.AMER.DELL.COM>

On Wed, Jan 25, 2017 at 04:43:58PM +0000, Mario.Limonciello@dell.com wrote:
> Thanks for your comments.  Some nested below.
> 
> > -----Original Message-----
> > From: Ville Syrjälä [mailto:ville.syrjala@linux.intel.com]
> > Sent: Wednesday, January 25, 2017 9:57 AM
> > To: Limonciello, Mario <Mario_Limonciello@Dell.com>
> > Cc: intel-gfx@lists.freedesktop.org
> > Subject: Re: [Intel-gfx] DP Aux interfaces inquiry
> > 
> > On Tue, Jan 24, 2017 at 07:51:28PM +0000, Mario.Limonciello@dell.com
> > wrote:
> > > Hi,
> > >
> > > Recently Synaptics collaborated with Dell on a plugin [1] for fwupd that
> > allows flashing Synaptics MST hubs using the DP aux interface to manipulate
> > the DPCD [2].
> > > Currently the plugin hardcodes the max number of DP aux devices to look for
> > to 3 (as that's what we've seen so far on HW), but we were wondering:
> > >
> > > 	1) If there is a way to query the number of devices that the kernel is
> > going to be creating?
> > > 	2) Are there any instances of more than 3 devices in the wild today that
> > anyone is aware of?
> > 
> > These depend on the board you're dealing with, and on the number of gpus you
> > have in the system (and whether they actually have drivers loaded for them).
> 
> OK, that's what I was suspecting.
> 
> > 
> > We should also perhaps expose the aux device node for MST devices. At which
> > point the number of aux nodes could change dynamically when you plug MST
> > devices in/out.
> > 
> 
> Hmm, for the devices themselves?  The way Synaptics handles cascaded MST devices
> today is a remote control mechanism.  They're able to turn on remote control
> for one MST hub, and it will forward control commands (and payloads) to the next
> cascaded hub.

I'm not really any kind of MST expert, so I don't really know what
people get up to with these things. But there are definitely remote DPCD
read/write messages in the spec, so I think we should be able to expose
the DPCD of any downstream device via a device node.

I took a quick stab at it:
git://github.com/vsyrjala/linux.git dp_mst_port_aux_dev

Doesn't quite seem to work though, so I probably made a mistake somewhere.

> 
> If/when you do this can you send a uevent up to userspace?  It would be good
> for fwupd to be able to listen to it and refresh devices based upon what happened
> from nodes coming and going.

Hotplug uevents should be fired off whenever a new display
appears/disappears.

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2017-01-25 18:40 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-01-24 19:51 DP Aux interfaces inquiry Mario.Limonciello
2017-01-25 15:56 ` Ville Syrjälä
2017-01-25 16:43   ` Mario.Limonciello
2017-01-25 17:13     ` Rafael Antognolli
2017-01-25 17:31       ` Ville Syrjälä
2017-01-25 18:40     ` Ville Syrjälä [this message]
2017-01-27 19:42       ` Mario.Limonciello

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=20170125184048.GY31595@intel.com \
    --to=ville.syrjala@linux.intel.com \
    --cc=Mario.Limonciello@dell.com \
    --cc=intel-gfx@lists.freedesktop.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 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.