Re: linux-next: build failure after merge of the drm-misc tree

[Mark Brown]

On Wed, Sep 23, 2026 at 05:57:24PM -0400, Alex Deucher wrote:

> Your reports are valuable and we always take them seriously even if I
> don't reply as quickly as I should.  For the issues at hand they
> should all be fixed with the following patches:

> https://patchwork.freedesktop.org/patch/755725/?series=174828&rev=1
> https://patchwork.freedesktop.org/patch/755741/?series=174834&rev=1
> https://patchwork.freedesktop.org/patch/755769/?series=172614&rev=2

> I'll push a new -next branch with those fixes later today.

I'm still seeing at least:

/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn30/display_mode_vba_30.c: In function ‘UseMinimumDCFCLK’:
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn30/display_mode_vba_30.c:6666:1: error: the frame size of 2064 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
 6666 | }
      | ^
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn31/display_mode_vba_31.c: In function ‘UseMinimumDCFCLK’:
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn31/display_mode_vba_31.c:7460:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
 7460 | }
      | ^
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/display_mode_vba_util_32.c: In function ‘dml32_UseMinimumDCFCLK’:
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/display_mode_vba_util_32.c:3278:1: error: the frame size of 2176 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
 3278 | }
      | ^
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn314/display_mode_vba_314.c: In function ‘UseMinimumDCFCLK’:
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn314/display_mode_vba_314.c:7546:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
 7546 | }
      | ^
cc1: all warnings being treated as errors
make[7]: *** [/home/broonie/git/next/sfr/next/scripts/Makefile.build:290: drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/display_mode_vba_util_32.o] Error 1
make[7]: *** Waiting for unfinished jobs....
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml2_0/display_mode_core.c: In function ‘dml_prefetch_check’:
/home/broonie/git/next/sfr/next/drivers/gpu/drm/amd/amdgpu/../display/dc/dml2_0/display_mode_core.c:6821:1: error: the frame size of 2160 bytes is larger than 2056 bytes [-Werror=frame-larger-than=]
 6821 | }
      | ^

after merging amdgpu today, note that this is in the -next merge and not
just the amdgpu tree in isolation.  I am currently holding drm and
drm-misc at next-20260921, there is a conflict with the old version of
them though that was just comments but they'll be in the history at the
point I'm building so possibly whatever git did with the merge is
causing issues.

Any fixes will also need to propagate into drm and drm-misc.

I'm also still carrying a build fix marking the display KUnit tests as
BROKEN due to the issues with the conflicting changes with drm-misc
which several people have noticed, for example there's this from Arnd
last week:

   https://lore.kernel.org/r/20260915201006.3526945-1-arnd@xxxxxxxxxx

None of the mails I saw about that or the issues resulting from my
mismerge got any response either.

Attachment: signature.asc
Description: PGP signature