NACK Re: [M][PULL][PATCH 00/19] drm/i915: Some fixes for Meteor Lake

Timo Aaltonen tjaalton at ubuntu.com
Mon Oct 23 14:01:50 UTC 2023


Stefan Bader kirjoitti 23.10.2023 klo 12.24:
> On 15.09.23 13:04, Timo Aaltonen wrote:
>> From: Timo Aaltonen <timo.aaltonen at canonical.com>
>>
>>
>> BugLink: https://bugs.launchpad.net/bugs/2032781
>> BugLink: https://bugs.launchpad.net/bugs/2029470
>>
>> These are all in 6.6~rc, but one patch needed some minor formatting
>> changes to avoid more backports. They have also been in oem-6.5 for
>> a while, and MTL is still protected here by requiring force_probe.
>>
>> Having that in mind, these should be safe to apply for Mantic.
> 
> This has been sitting around for quite a while which might tell you how 
> much people like huge patch bombs...
> I must admit I am very torn about what to do with this. On the positive 
> side this still applies. Patch #18 needs to be adjusted though to the 
> standard backport format we use:
> 
> (backported from commit <sha>)
> [tjaalton: minor adjustments]
> ...
> 
> Also you say you carry those in the oem-kernel for a while now without 
> issues (though that maybe does not cover older hw as long as they stil 
> are on an older oem kernel).
> But it refers to 2 bug report which both give not too much reason to 
> take this as stable updates. I would at least expect a summary of issue 
> which are addressed. So verification could be done against this. Also 
> testing likely should also cover a couple of other i915 GPU models. 
> Somehow I have not that much trust in any GPU driver developers to add 
> things for newer hardware completely without causing side effects on 
> older HW.
> Also those bug reports are against linux-oem-6.5 and not Mantic (yes 
> that would be quick to change).
> 
> So right now I think I am tending to ask you to re-submit ... I am just 
> avoiding the N-word in the subject in case there is a convincing 
> argument I cannot deny...
> 
> -Stefan

Well, it was meant to land before kernel freeze, but I didn't push it 
too actively. So lets just NACK this instead now.

-- 
t




More information about the kernel-team mailing list