Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1212294

Summary: [lshw] reports some USB host controllers as type "bus" instead of "USB"
Product: [Retired] Beaker Reporter: Dan Callaghan <dcallagh>
Component: inventoryAssignee: matt jia <mjia>
Status: CLOSED CURRENTRELEASE QA Contact: Dan Callaghan <dcallagh>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 20CC: aigao, dcallagh, ebaak, mjia
Target Milestone: 21.0Keywords: Patch
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2015-08-26 06:17:38 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 541294    

Description Dan Callaghan 2015-04-16 06:24:51 UTC
For example, from a physical x86 system:

631xESB/632xESB/3100 Chipset EHCI USB2 Controller 8086:268C
type should be: USB

Calling it "USB" seems generally more useful but it's not clear which is the most accurate. It may not matter.

Comment 1 matt jia 2015-05-22 04:52:25 UTC
It would be good to have an example system that has (In reply to Dan Callaghan from comment #0)
> For example, from a physical x86 system:
> 
> 631xESB/632xESB/3100 Chipset EHCI USB2 Controller 8086:268C
> type should be: USB
> 
> Calling it "USB" seems generally more useful but it's not clear which is the
> most accurate. It may not matter.

It would be good to have an example system to make it easy to debug the problem.

Comment 4 matt jia 2015-05-25 00:15:10 UTC
(In reply to Dan Callaghan from comment #0)
> For example, from a physical x86 system:
> 
> 631xESB/632xESB/3100 Chipset EHCI USB2 Controller 8086:268C
> type should be: USB
> 
> Calling it "USB" seems generally more useful but it's not clear which is the
> most accurate. It may not matter.

Both smolt and lshw are using device class as 'Type'. smolt is using[1] to match the code. lshw is using 'bus'[2] as a base class serial for devices like USB, SCSI, Firewire. I think we could use both class and description attributes for the device type to make it more clear. For this example, the device type would be bus(USB controller).


[1] http://www.acm.uiuc.edu/sigops/roll_your_own/7.c.1.html
[2] http://www.ezix.org/project/wiki/HardwareLiSter

Comment 5 matt jia 2015-05-25 00:38:08 UTC
(In reply to matt jia from comment #4)
> Both smolt and lshw are using device class as 'Type'. smolt is using[1] to
> match the code. lshw is using 'bus'[2] as a base class serial for devices
> like USB, SCSI, Firewire. I think we could use both class and description
> attributes for the device type to make it more clear. For this example, the
> device type would be bus(USB controller).

We could also use bootstrap tooltip on an info icon. So when users move the cursor over the icon, a right tooltip with the text of the description attribute is shown on the web UI.

Comment 6 Amit Saha 2015-05-25 04:33:27 UTC
(In reply to matt jia from comment #4)
> (In reply to Dan Callaghan from comment #0)
> > For example, from a physical x86 system:
> > 
> > 631xESB/632xESB/3100 Chipset EHCI USB2 Controller 8086:268C
> > type should be: USB
> > 
> > Calling it "USB" seems generally more useful but it's not clear which is the
> > most accurate. It may not matter.
> 
> Both smolt and lshw are using device class as 'Type'. smolt is using[1] to
> match the code. lshw is using 'bus'[2] as a base class serial for devices
> like USB, SCSI, Firewire. I think we could use both class and description
> attributes for the device type to make it more clear. For this example, the
> device type would be bus(USB controller).
> 
> 
> [1] http://www.acm.uiuc.edu/sigops/roll_your_own/7.c.1.html
> [2] http://www.ezix.org/project/wiki/HardwareLiSter

Do you think we can do that in beaker-system-scan itself or do we need to patch lshw?

Comment 7 matt jia 2015-05-25 04:39:10 UTC
(In reply to Amit Saha from comment #6)
> > 
> > Both smolt and lshw are using device class as 'Type'. smolt is using[1] to
> > match the code. lshw is using 'bus'[2] as a base class serial for devices
> > like USB, SCSI, Firewire. I think we could use both class and description
> > attributes for the device type to make it more clear. For this example, the
> > device type would be bus(USB controller).
> > 
> > 
> > [1] http://www.acm.uiuc.edu/sigops/roll_your_own/7.c.1.html
> > [2] http://www.ezix.org/project/wiki/HardwareLiSter
> 
> Do you think we can do that in beaker-system-scan itself or do we need to
> patch lshw?

If this approach sounds ok, we can do it in beaker-system-scan.

Comment 8 Amit Saha 2015-05-25 05:46:56 UTC
(In reply to matt jia from comment #7)
> (In reply to Amit Saha from comment #6)
> > > 
> > > Both smolt and lshw are using device class as 'Type'. smolt is using[1] to
> > > match the code. lshw is using 'bus'[2] as a base class serial for devices
> > > like USB, SCSI, Firewire. I think we could use both class and description
> > > attributes for the device type to make it more clear. For this example, the
> > > device type would be bus(USB controller).
> > > 
> > > 
> > > [1] http://www.acm.uiuc.edu/sigops/roll_your_own/7.c.1.html
> > > [2] http://www.ezix.org/project/wiki/HardwareLiSter
> > 
> > Do you think we can do that in beaker-system-scan itself or do we need to
> > patch lshw?
> 
> If this approach sounds ok, we can do it in beaker-system-scan.

Can you please run a comparison task with this change so that we can see how the output will look like?

Comment 9 matt jia 2015-05-25 23:32:55 UTC
Another idea would be mapping the current existing smolt classes to lshw classes.

Comment 10 matt jia 2015-05-26 01:31:29 UTC
(In reply to matt jia from comment #9)
> Another idea would be mapping the current existing smolt classes to lshw
> classes.

Sorry, wrong direction, it should map lshw classes to smolt classes.

Comment 11 matt jia 2015-06-04 04:24:29 UTC
On Gerrit:

   http://gerrit.beaker-project.org/4236

Comment 20 Dan Callaghan 2015-08-26 06:17:38 UTC
Beaker 21.0 has been released.