Checkbox

monitor tests need to accept video output identifiers used by proprietary drivers

Bug #992727 reported by Brendan Donegan on 2012-05-01

6

This bug affects 1 person

Affects		Status	Importance	Assigned to	Milestone
	Checkbox	Fix Released	High	Sylvain Pineau	Checkbox 0.14.x "Before Q release"

Bug Description

The monitor tests currently use a requires field like:

requires: display.hdmi == 'supported'

to see if a particular test should run on a particular system. It seems that some of the proprietary drivers use some eccentric names for these video ports. For example on one ATI system the fglrx driver called the VGA port 'CRT1' and the HDMI port 'DFP1'. The requires field for those tests should include those identifiers, like:

requires: display.hdmi == 'supported' or display.dfp == 'supported'

Related branches

lp:~sylvain-pineau/checkbox/display_resource

Merged into lp:checkbox at revision 1719

Brendan Donegan (community): Approve on 2012-10-01

Sylvain Pineau: Needs Resubmitting on 2012-10-01

lp:~roadmr/ubuntu/quantal/checkbox/0.14.8

Merged into lp:ubuntu/quantal/checkbox at revision 53

Daniel Holbach (community): Approve on 2012-10-08

Micah Gersten (community): Needs Fixing on 2012-10-05

Ubuntu branches: Pending requested 2012-10-04

lp:ubuntu/quantal/checkbox

Revision history for this message

Daniel Manrique (roadmr) wrote on 2012-05-01:

#1

The test dependencies shouldn't need to deal with proprietary driver wonkiness. The display resource should instead map all possible identifiers to a single, agreed-upon name (and really, whose braindead idea was it that you can only plug a CRT to the 15-ping VGA connector? what does DFP1 stand for?) I'd say mapping them to industry-standard names like, oh I don't know, VGA or HDMI or displayport (I do see maybe thunderbolt appearing in the future) and keeping that concern out of the test definitions may be good.

Revision history for this message

Jeff Lane  (bladernr) wrote on 2012-05-01: Re: [Bug 992727] Re: monitor tests need to accept video output identifiers used by proprietary drivers

#2

On 05/01/2012 02:02 PM, Daniel Manrique wrote:
> The test dependencies shouldn't need to deal with proprietary driver
> wonkiness. The display resource should instead map all possible
> identifiers to a single, agreed-upon name (and really, whose braindead
> idea was it that you can only plug a CRT to the 15-ping VGA connector?
> what does DFP1 stand for?) I'd say mapping them to industry-standard
> names like, oh I don't know, VGA or HDMI or displayport (I do see maybe
> thunderbolt appearing in the future) and keeping that concern out of the
> test definitions may be good.

And a lot cleaner. +1 on having the display resource map these strings
to more understandable names...

--
Jeff Lane - Hardware Certification Engineer and Test Tools Developer
Ubuntu Ham: W4KDH
Freenode IRC: bladernr or bladernr_
gpg: 1024D/3A14B2DD 8C88 B076 0DD7 B404 1417 C466 4ABD 3635 3A14 B2DD

Revision history for this message

Brendan Donegan (brendan-donegan) wrote on 2012-05-02:

#3

That was of course the other option - it makes the display resource command a bit more complicated, maybe even requiring a Python script, but it will look cleaner in the test definitions for sure

Revision history for this message

Brendan Donegan (brendan-donegan) wrote on 2012-05-02:

#4

Oh, and regardless of the path taken to a solution, can we please gather information about the names used by different drivers and add it to this bug?

Revision history for this message

Brendan Donegan (brendan-donegan) wrote on 2012-05-15:

#5

Here's one used for Display Port

DisplayPort-0

Revision history for this message

Daniel Manrique (roadmr) wrote on 2012-09-07:

#6

OK, so work to do: create a display_resource script that maps all the wonky names used by drivers into the actual physical port names: vga, hdmi, dvi, displayport at least. Then as more descriptors start appearing we can just add them to the script's table.

Changed in checkbox:
status:	New → Triaged
importance:	Undecided → Low

Revision history for this message

Brendan Donegan (brendan-donegan) wrote on 2012-09-19:

#7

I don't agree with the Low importance - if the resource doesn't work properly then we may end up not testing ports that we need to. That is bad. I'm going to bump to High since I want to see this fixed before certification

Changed in checkbox:
importance:	Low → High
milestone:	none → 0.14.x

Brendan Donegan (brendan-donegan) on 2012-09-27

Changed in checkbox:
assignee:	nobody → Sylvain Pineau (sylvain-pineau)
status:	Triaged → In Progress

Zygmunt Krynicki (zyga) on 2012-10-01

Changed in checkbox:
status:	In Progress → Fix Committed

Brendan Donegan (brendan-donegan) on 2012-11-01

Changed in checkbox:
status:	Fix Committed → Fix Released

Report a bug

This report contains Public information

Everyone can see this information.

You are

Subscribing...

Edit bug mail

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.