How to understand a patch works
Most likely the developers doing the kernel work won’t get the info at all. Lets take a look at the patch header:
author John Harrison <John.C.Harrison@Intel.com> 2023-02-15 17:11:01 -0800
committer Greg Kroah-Hartman <gregkh@linuxfoundation.org> 2023-03-10 09:40:14 +0100
commit 4eb6789f9177a5fdb90e1b7cdd4b069d1fb9ce45 (patch)
tree 7f0259b02632bf92193b3b5aa5c45d6f0fa3ebdd
parent 64bcaffa2d5c88ddfe12d6019ad58986bd0e314c (diff)
download linux-4eb6789f9177a5fdb90e1b7cdd4b069d1fb9ce45.tar.gz
drm/i915: Don't use BAR mappings for ring buffers with LLC
commit 85636167e3206c3fbd52254fc432991cc4e90194 upstream.
Direction from hardware is that ring buffers should never be mapped
via the BAR on systems with LLC. There are too many caching pitfalls
due to the way BAR accesses are routed. So it is safest to just not
use it.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Fixes: 9d80841ea4c9 ("drm/i915: Allow ringbuffers to be bound anywhere")
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org
Cc: <stable@vger.kernel.org> # v4.9+
We know that there was an author by Intel John Harrison doing the patch. If we look at the last CC we see that patch should be added to all kernels starting with 4.9 kernel. So most likely you might have similar issues with higher and lower kernel series than your used 5.15 series if the same patch gets applied there also. You can check that here @stasadev: kernel/git/stable/stable-queue.git - Linux kernel stable patch queue. More or less higher than 6.2.2 might also not work on your end: 6.2.3 « releases - kernel/git/stable/stable-queue.git - Linux kernel stable patch queue. So test that by checking which kernels we provide.
So check other kernel series if they also don’t boot as you still have 5.15.98 as your fallback one.
If it is true, simply check if other kernel series are affected too by this.
How to report properly
The first thing you can do is to add a confirm report similar as you have found already here [i915]drm:add_taint_for_CI [i915]] CI tainted:0x9 by intel_gt_init+0xae/0x2d0 [i915] (#8284) · Issues · drm / i915 / kernel · GitLab by adding your own details of hardware and maybe the dmesg log of the 5.15.99 and 5.15.98 of your own system additionally to the already reported one.
Next step is simply to summarize the issue in an email and sent that to upstream. Which emails? Well you have all what you need in the patch itself. Simply send them to the email addresses exactly as given. Also include the stable@vger.kernel.org and intel-gfx@lists.freedesktop.org mailing lists, as some developers still work on email.
What I think personally
Well, most likely it also can be a missing patch within the 5.15 kernel series. For example I see this in all the other kernel series when the patch is applied, except 5.15, even it is tacked to be included: drm-i915-don-t-use-stolen-memory-for-ring-buffers-with-llc.patch « 6.2.3 « releases - kernel/git/stable/stable-queue.git - Linux kernel stable patch queue So test as recommended other kernel series and see if they work for you.
What Manjaro can do
We might revert the patch as of now and follow up with upstream.