Bug 696255

Summary: [SGI 6.2 FEAT] NVidia Tesla GPU's
Product: Red Hat Enterprise Linux 6 Reporter: George Beshers <gbeshers>
Component: xorg-x11-driversAssignee: Adam Jackson <ajax>
Status: CLOSED WORKSFORME QA Contact: Desktop QE <desktop-qa-list>
Severity: high Docs Contact:
Priority: unspecified    
Version: 6.2CC: dwa, gbeshers, martinez, rja, tee, tpelka, travis
Target Milestone: rcKeywords: FutureFeature, OtherQA
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Enhancement
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-06-29 18:42:47 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 652290    

Description George Beshers 2011-04-13 17:30:30 UTC
Description of problem:
1. Feature Overview:
   a) Name of feature:
      NVidia Tesla GPU's

   b) Feature Description:
      The kernel vgaarb is not recognizing the Tesla GPU's:

      vgaarb: this pci device is not a vga device
      vgaarb: this pci device is not a vga device
      vgaarb: this pci device is not a vga device
      vgaarb: this pci device is not a vga device
      vgaarb: this pci device is not a vga device
      vgaarb: this pci device is not a vga device
      vgaarb: this pci device is not a vga device

      I don't know if this is affecting anything, as X faults with:

      (EE) Peppercon AG Multidevice: failed to initialize for relative axes.
      (EE) Peppercon AG Multidevice: failed to initialize for relative axes.



2. Feature Details:
   a) Architectures:
      64-bit Intel EM64T

   b) Bugzilla Dependencies:
      none

   c) Drivers or hardware dependencies:
      Tesla GPU

   d) Upstream acceptance information:
      In progress.

   e) External links:
      to be supplied

   f) Severity (H,M,L):
      High (required for Hardware Enablement)

   g) Feature Needed by:
      Needed for UV2 (Early to Mid 2012).

3. Business Justification:
   a) Why is this feature needed?
      Increasing interest in GPU based HPC

   b) What hardware does this enable?
      Tesla GPU's

   c) Business impact?
      Existing customers wanting to switch to RHEL.

   d) Other business drivers:

4. Primary contact at Red Hat, email, phone (chat)
   First_Name Last_Name
   xxx_xxxx
   Phone Number

5. Primary contact at Partner, email, phone (chat)
      Lori Gilbertson, loriann, 651-683-3433, N/A  
                                               
   Partner technical contact, email, phone, chat
      George Beshers, gbeshers/gbeshers, 508-212-6362, gbeshers



Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:

Comment 2 Adam Jackson 2011-04-14 14:40:43 UTC
>    b) Feature Description:
>       The kernel vgaarb is not recognizing the Tesla GPU's:
> 
>       vgaarb: this pci device is not a vga device
>       vgaarb: this pci device is not a vga device
>       vgaarb: this pci device is not a vga device
>       vgaarb: this pci device is not a vga device
>       vgaarb: this pci device is not a vga device
>       vgaarb: this pci device is not a vga device
>       vgaarb: this pci device is not a vga device

Please attach lspci -vnn from this machine, we should be able to quash that message but I'd like to be sure what kind of case provokes it.

>       I don't know if this is affecting anything, as X faults with:
> 
>       (EE) Peppercon AG Multidevice: failed to initialize for relative axes.
>       (EE) Peppercon AG Multidevice: failed to initialize for relative axes.

It's unlikely, though not impossible, that this is the actual cause of the fault.  Please also attach the Xorg.0.log from the failure case.

Comment 3 gbeshers 2011-04-14 16:49:33 UTC
Mike,

Can you see this BZ?  If so, can you respond to his request?

George

Comment 4 Mike Travis 2011-04-18 18:20:52 UTC
    Technical note added. If any revisions are required, please edit the "Technical Notes" field
    accordingly. All revisions will be proofread by the Engineering Content Services team.
    
    New Contents:
Mike,

Can you see this BZ?  If so, can you respond to his request?

Hi,

Yes, I can and will gather this up as soon as practical.

Thanks,
Mike

Comment 5 Tomas Pelka 2011-05-31 06:58:17 UTC
Hi Mike,

could you please add info requested by Adam?

Since we have no Tesla GPU's in-house, will SGI be able to check HW enablement? May I add OtherQA keyword?

Comment 6 Mike Travis 2011-06-08 20:17:16 UTC
I'm running linux 2.6.32-71.el6.x86_64 and have not been able to reproduce the problem.  I'll keep checking to see how the problem disappeared.

Comment 7 Tomas Pelka 2011-06-08 20:41:19 UTC
That's very good to hear Mike,

keep inform us please, thanks.

Comment 8 Marizol Martinez 2011-06-13 17:20:13 UTC
SGI -- Please provide an update. The Partner Code Submission Deadline for RHEL 6.2 is this Thu, 16-June.

Comment 10 Adam Jackson 2011-06-29 18:25:37 UTC
If this is still an issue, please attach the information requested in comment #2.

Comment 11 Marizol Martinez 2011-06-29 18:31:41 UTC
Deleting Technical Notes. Clearly a mistake.

Comment 12 Marizol Martinez 2011-06-29 18:31:41 UTC
Deleted Technical Notes Contents.

Old Contents:
Mike,

Can you see this BZ?  If so, can you respond to his request?

Hi,

Yes, I can and will gather this up as soon as practical.

Thanks,
Mike

Comment 13 gbeshers 2011-06-29 18:42:47 UTC
Closing this as it has not been seen in recent testing.

George