lp:~roadmr/checkbox/1291940-amd-merge-from-legacy

Branch merges

Propose for merging

No branches dependent on this one.

Merged into lp:checkbox at revision 2935

Zygmunt Krynicki (community): Approve on 2014-04-22

Diff: 12235 lines (+12165/-3)

5 files modified

checkbox-support/checkbox_support/lib/pci.py (+3/-1)
checkbox-support/checkbox_support/parsers/tests/test_udevadm.py (+50/-1)
checkbox-support/checkbox_support/parsers/tests/udevadm_data/DELL_INSPIRON_3048.txt (+6656/-0)
checkbox-support/checkbox_support/parsers/tests/udevadm_data/DELL_INSPIRON_7737_NVIDIA.txt (+5446/-0)
checkbox-support/checkbox_support/parsers/udevadm.py (+10/-1)

api

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

2937. By Daniel Manrique on 2014-04-18

checkbox-support:parsers: Added data and tests for Dell Inspiron 7737.

This change had not been not merged from checkbox-legacy for some
reason. A commit previously backported from checkbox-legacy implements
the required heuristics for video devices with PCI subclass "2", like
newer NVidia GPUs (LP: #1270139). The data file and test are included
here for completeness and parity with checkbox-legacy.

2936. By Daniel Manrique on 2014-04-18

checkbox-support:parsers:udevadm: Updated heuristic for AMD VIDEO devices.

Fixes this bug: (LP: #1291940)

Also updated test expectations for other systems affected by this
update (we were previously missing a GPU).

Rationale and logic are discussed in the code's comments.

2935. By Daniel Manrique on 2014-04-18

checkbox-support:parsers:tests:udevadm: Added test with expectations for Dell Inspiron 3048.

Specifically we want to verify that its AMD GPU is correctly categorized as VIDEO.

It also adds the data file for this system, the AMD GPU has a
subclass_id of 0x80, which we hadn't handled before (i.e. we ignored
it as no actual GPUs used it).

2934. By Daniel Manrique on 2014-04-18

checkbox-support:lib:pci: Corrected CLASS_DISPLAY_OTHER value and added constant for a vendor-specific exception.

AMD devices use the OTHER display subclass in some valid GPUs so we need the
vendor ID as well to "quirk" their devices.

2933. By Zygmunt Krynicki on 2014-04-16

"automatic merge by tarmac [r=roadmr][bug=][author=zkrynicki]"

2932. By Zygmunt Krynicki on 2014-04-16

"automatic merge by tarmac [r=roadmr][bug=1308612][author=zkrynicki]"

2931. By Zygmunt Krynicki on 2014-04-16

"automatic merge by tarmac [r=zkrynicki][bug=1307494][author=zkrynicki]"

2930. By Ashley Lai on 2014-04-16

"automatic merge by tarmac [r=zkrynicki][bug=1298438][author=alai]"

2929. By Daniel Manrique on 2014-04-16

"use partial_id instead of name for jobs
 [r=zkrynicki][bug=1305288][author=roadmr]"

2928. By Daniel Manrique on 2014-04-16

"cpuinfo_resource: avoid printing empty keys (rather, keys with empty values).
 [r=zkrynicki][bug=1306255][author=roadmr]"