Red Hat Bugzilla – Bug 456293
[Intel 5.3 FEAT] Add final branding strings for EagleLake graphics (McCreary) to graphics driver
Last modified: 2015-05-08 09:59:41 EDT
1. Feature Name:
Final branding strings for EagleLake graphics (McCreary) are available. Please
add them to the graphics driver.
Architectures (mark all that apply)
x 32-bit x86
x 64-bit Intel EM64T
Priority (H,M,L): H
Target Release Date:
Drivers or hardware dependency:
3. Business Justification:
a. Why is this feature needed?
b. What hardware does this enable?
Eagle lake graphics
c. Forecast, volume or high end platform?
d. Any configuration info?
e. When needed?
f. Are there other dependencies (drivers).
5. Hardware to Red Hat?
I believe that Red Hat has the hardware now.
6. Partner management contact, email, phone, chat
7. Partner technical contact, email, phone, chat
John - where are the actual strings for this located?
Gordon Jin gave me the following information:
http://developer.intel.com/products/desktop/chipsets/ lists the branding
names for P45/P43/G45/G43.
I don't understand why Q45/Q43/G41 are not listed.
Note P45/P43 don't have integrated gfx.
Hardware is scheduled to be on-site (Westford, Raleigh and Brno) on 07-August.
Until we receive word from Intel that the final working code is all upstream,
I'm going to put this into Conditional NAK: PATCH.
(In reply to comment #11)
> Until we receive word from Intel that the final working code is all upstream,
> I'm going to put this into Conditional NAK: PATCH.
The final branding strings are G45/G43, Q45/Q43. The final code is all in Linus's upstream kernel and upstream projects.
- xf86-video-intel (master branch or 2.4.* release)
- mesa (master branch or 7.0.4 release)
- kernel (2.6.26)
Red Hat engineering is waiting for Intel (Keith P) to confirm that the changes upstream are final.
Michael Fu told me:
Confirmed with zhenyu. G43 use the same pciid as G45. So you can just apply the patch for G41 and we're ready to go.
If you need more info then that, please let me know. We believe that the code is upstream.
Correct comment #12, the final branding strings for EagleLake include G41, G45/G43, Q45/Q43. The patch for G41 has been submitted to upstream. Others are all in upstream code.
Intel branding in 5.3 is in line with upstream, afaict.
Created attachment 323743 [details]
patch for kenel/agp intel-agp.c to broadcast cantiga
How about we also update the agpgart information loged in dmesg?
GM45 is still described as "Intel Integrated Graphics Device" now. attached a small patch to kernel/driver/char/agp/intel-agp.c FYI. the codes already in upstream.
Comment #26 identified the missing branding code for GM45 in agpgart and the code is already in upstream now. Could you please consider adding the patch into RHEL5.3? Thanks.
does Geoff need to post something here or will RH handle it?
We are cloning this BZ so that the kernel AGP portion can be taken care of by RH. See https://bugzilla.redhat.com/show_bug.cgi?id=472438
Partners, this bug should be fixed in the latest RHEL 5.3 Snapshot. We believe that you have some interest in its correct functionality, so we're making a friendly request to send us some testing feedback.
If you have a chance to test it, please share with us your findings. If you have successfully VERIFIED the fix, please add PartnerVerified to the Bugzilla keywords, along with a description of the results. Thanks!
According to comment #12, here are some test results for RHEL5.3 SP4 to share with you:
1. Confirmed G45/G43/G41 branding info in 2D driver;
2. Confirmed G45/G43 branding info in kernel AGP;
3. G41 branding info patch is submited to upstream already but blocked there due to lack of hardware validation. So, G41 not in upstream, not in RHEL5.3. Plan to open a new bug to track G41 kernel support.
4. Mesa 3D isn't supported on eagleLake & Cantiga in RHEL5.3.
According to comment #36, this bug has been VERIFIED. Please set to to ASSIGNED if you find this bug is not correctly fixed. If you encounter any new issues, clone this bug (NEW) and describe them. Thanks for the feedback!
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.