dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

[Borislav Petkov]

Hi all,

this fires in my randbuilds here:

vmlinux.o: warning: objtool: adis16400_write_raw() falls through to next function adis16400_show_serial_number()
ld: vmlinux.o: in function `dw_mipi_dsi_stm_remove':
dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'
ld: vmlinux.o: in function `dw_mipi_dsi_clk_register':
dw_mipi_dsi-stm.c:(.text+0x8db9f5): undefined reference to `clk_hw_register'
ld: vmlinux.o: in function `lvds_remove':
lvds.c:(.text+0x8dc605): undefined reference to `clk_hw_unregister'
make[2]: *** [scripts/Makefile.vmlinux:34: vmlinux] Error 1
make[1]: *** [/home/amd/bpetkov/kernel/linux/Makefile:1156: vmlinux] Error 2
make: *** [Makefile:224: __sub-make] Error 2

is there a fix somewhere?

People love to do

	depends on ... COMPILE_TEST

but then if no one takes care of it in time:

https://lore.kernel.org/oe-kbuild-all/202407212000.rpDH64jP-lkp@intel.com

that COMPILE_TEST thing is forcing me to simply blacklist it and is not really
helping.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

[Raphael Gallais-Pou]

On 9/5/24 10:19, Borislav Petkov wrote:
> Hi all,
>
> this fires in my randbuilds here:
>
> vmlinux.o: warning: objtool: adis16400_write_raw() falls through to next function adis16400_show_serial_number()
> ld: vmlinux.o: in function `dw_mipi_dsi_stm_remove':
> dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'
> ld: vmlinux.o: in function `dw_mipi_dsi_clk_register':
> dw_mipi_dsi-stm.c:(.text+0x8db9f5): undefined reference to `clk_hw_register'
> ld: vmlinux.o: in function `lvds_remove':
> lvds.c:(.text+0x8dc605): undefined reference to `clk_hw_unregister'
> make[2]: *** [scripts/Makefile.vmlinux:34: vmlinux] Error 1
> make[1]: *** [/home/amd/bpetkov/kernel/linux/Makefile:1156: vmlinux] Error 2
> make: *** [Makefile:224: __sub-make] Error 2
>
> is there a fix somewhere?

Hi Borislav,


Arnd Bergmann sent a patch regarding this issue on x86 that I merged several
weeks ago.

https://lore.kernel.org/lkml/20240719075454.3595358-1-arnd@kernel.org/
https://lore.kernel.org/all/c3d0757a-07c0-4f83-9f06-c3ad205aa1e9@foss.st.com/

>
> People love to do
>
> 	depends on ... COMPILE_TEST
>
> but then if no one takes care of it in time:
>
> https://lore.kernel.org/oe-kbuild-all/202407212000.rpDH64jP-lkp@intel.com

Now, I'm not familiar with PowerPC architecture and toolchains, but I think this
patch should fix your problem.
Do you have the above fixup in your tree ? If not please try it.

Thanks,

Regards,
Raphaël

>
> that COMPILE_TEST thing is forcing me to simply blacklist it and is not really
> helping.
>
> Thx.
>

Re: dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

[Borislav Petkov]

On Mon, Sep 09, 2024 at 08:57:57AM +0200, Raphael Gallais-Pou wrote:
> Arnd Bergmann sent a patch regarding this issue on x86 that I merged several
> weeks ago.
> 
> https://lore.kernel.org/lkml/20240719075454.3595358-1-arnd@kernel.org/
> https://lore.kernel.org/all/c3d0757a-07c0-4f83-9f06-c3ad205aa1e9@foss.st.com/

Thanks.

> Now, I'm not familiar with PowerPC architecture and toolchains, but I think
> this patch should fix your problem.  Do you have the above fixup in your
> tree ? If not please try it.

Not PowerPC - it is an x86 build. I do random config builds as part of build
testing and when a driver depends on COMPILE_TEST, it does gets built in some
.configs. So some may fail, as in this case.

So I think you should send that patch to Linus now so that such randconfig
builds do not fail anymore.

Unfortunately, I cannot test it right now because I've removed the triggering
randconfigs but when I hit it again, I'll make sure to apply yours.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

[Raphael Gallais-Pou]

On 9/9/24 10:45, Borislav Petkov wrote:
> On Mon, Sep 09, 2024 at 08:57:57AM +0200, Raphael Gallais-Pou wrote:
>> Arnd Bergmann sent a patch regarding this issue on x86 that I merged several
>> weeks ago.
>>
>> https://lore.kernel.org/lkml/20240719075454.3595358-1-arnd@kernel.org/
>> https://lore.kernel.org/all/c3d0757a-07c0-4f83-9f06-c3ad205aa1e9@foss.st.com/
> Thanks.
>
>> Now, I'm not familiar with PowerPC architecture and toolchains, but I think
>> this patch should fix your problem.  Do you have the above fixup in your
>> tree ? If not please try it.
> Not PowerPC - it is an x86 build. I do random config builds as part of build
> testing and when a driver depends on COMPILE_TEST, it does gets built in some
> .configs. So some may fail, as in this case.

Unless I am mistaken, the link you provided refers to a PowerPC linker error:

[...]

compiler: powerpc-linux-gcc (GCC) 14.1.0

[...]
   powerpc-linux-ld: drivers/gpu/drm/stm/dw_mipi_dsi-stm.o: in function `dw_mipi_dsi_stm_remove':
   dw_mipi_dsi-stm.c:(.text+0x664): undefined reference to `of_clk_del_provider'
>> powerpc-linux-ld: dw_mipi_dsi-stm.c:(.text+0x66c): undefined reference to `clk_hw_unregister'    powerpc-linux-ld: drivers/gpu/drm/stm/dw_mipi_dsi-stm.o: in function `dw_mipi_dsi_stm_probe':
   dw_mipi_dsi-stm.c:(.text+0x98c): undefined reference to `clk_hw_register'

>
> So I think you should send that patch to Linus now so that such randconfig
> builds do not fail anymore.

What do you mean by 'sending it to Linus' ? If you meant to do a pull request,
then no. This patch is already in the drm-misc tree, which means it will keep
its usual pace of merging with the rest of the drm-misc tree.

For more information about drm-misc tree:
https://drm.pages.freedesktop.org/maintainer-tools/repositories/drm-misc.html#merge-timeline

>
> Unfortunately, I cannot test it right now because I've removed the triggering
> randconfigs but when I hit it again, I'll make sure to apply yours.
Thank you for this. Do not hesitate to reach if you encounter any problems with
this patch.

Regards,
Raphaël

>
> Thx.
>

Re: dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

[Borislav Petkov]

On Tue, Sep 10, 2024 at 12:48:49PM +0200, Raphael Gallais-Pou wrote:
> Unless I am mistaken, the link you provided refers to a PowerPC linker error:

Ah, the kernel test robot is doing other architectures now too, sorry about
that. In any case, I am triggering it on x86 too.

> What do you mean by 'sending it to Linus' ? If you meant to do a pull request,
> then no. This patch is already in the drm-misc tree, which means it will keep
> its usual pace of merging with the rest of the drm-misc tree.

That's ok - I'll simply disable CONFIG_DRM_STM from building in my randconfigs
for the time being.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 2155 bytes --]

On Tue, Sep 10, 2024 at 12:48:49PM GMT, Raphael Gallais-Pou wrote:
> 
> On 9/9/24 10:45, Borislav Petkov wrote:
> > On Mon, Sep 09, 2024 at 08:57:57AM +0200, Raphael Gallais-Pou wrote:
> >> Arnd Bergmann sent a patch regarding this issue on x86 that I merged several
> >> weeks ago.
> >>
> >> https://lore.kernel.org/lkml/20240719075454.3595358-1-arnd@kernel.org/
> >> https://lore.kernel.org/all/c3d0757a-07c0-4f83-9f06-c3ad205aa1e9@foss.st.com/
> > Thanks.
> >
> >> Now, I'm not familiar with PowerPC architecture and toolchains, but I think
> >> this patch should fix your problem.  Do you have the above fixup in your
> >> tree ? If not please try it.
> > Not PowerPC - it is an x86 build. I do random config builds as part of build
> > testing and when a driver depends on COMPILE_TEST, it does gets built in some
> > .configs. So some may fail, as in this case.
> 
> Unless I am mistaken, the link you provided refers to a PowerPC linker error:
> 
> [...]
> 
> compiler: powerpc-linux-gcc (GCC) 14.1.0
> 
> [...]
>    powerpc-linux-ld: drivers/gpu/drm/stm/dw_mipi_dsi-stm.o: in function `dw_mipi_dsi_stm_remove':
>    dw_mipi_dsi-stm.c:(.text+0x664): undefined reference to `of_clk_del_provider'
> >> powerpc-linux-ld: dw_mipi_dsi-stm.c:(.text+0x66c): undefined reference to `clk_hw_unregister'    powerpc-linux-ld: drivers/gpu/drm/stm/dw_mipi_dsi-stm.o: in function `dw_mipi_dsi_stm_probe':
>    dw_mipi_dsi-stm.c:(.text+0x98c): undefined reference to `clk_hw_register'
> 
> >
> > So I think you should send that patch to Linus now so that such randconfig
> > builds do not fail anymore.
> 
> What do you mean by 'sending it to Linus' ? If you meant to do a pull request,
> then no. This patch is already in the drm-misc tree, which means it will keep
> its usual pace of merging with the rest of the drm-misc tree.
> 
> For more information about drm-misc tree:
> https://drm.pages.freedesktop.org/maintainer-tools/repositories/drm-misc.html#merge-timeline

That's not entirely correct. This should have been merged in
drm-misc-fixes to begin with. I've cherry-picked the patch and pushed
it.

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 273 bytes --]

Re: dw_mipi_dsi-stm.c:(.text+0x8db9a3): undefined reference to `clk_hw_unregister'

[Raphael Gallais-Pou]

On 9/10/24 13:59, Maxime Ripard wrote:
> On Tue, Sep 10, 2024 at 12:48:49PM GMT, Raphael Gallais-Pou wrote:
>> On 9/9/24 10:45, Borislav Petkov wrote:
>>> On Mon, Sep 09, 2024 at 08:57:57AM +0200, Raphael Gallais-Pou wrote:
>>>> Arnd Bergmann sent a patch regarding this issue on x86 that I merged several
>>>> weeks ago.
>>>>
>>>> https://lore.kernel.org/lkml/20240719075454.3595358-1-arnd@kernel.org/
>>>> https://lore.kernel.org/all/c3d0757a-07c0-4f83-9f06-c3ad205aa1e9@foss.st.com/
>>> Thanks.
>>>
>>>> Now, I'm not familiar with PowerPC architecture and toolchains, but I think
>>>> this patch should fix your problem.  Do you have the above fixup in your
>>>> tree ? If not please try it.
>>> Not PowerPC - it is an x86 build. I do random config builds as part of build
>>> testing and when a driver depends on COMPILE_TEST, it does gets built in some
>>> .configs. So some may fail, as in this case.
>> Unless I am mistaken, the link you provided refers to a PowerPC linker error:
>>
>> [...]
>>
>> compiler: powerpc-linux-gcc (GCC) 14.1.0
>>
>> [...]
>>    powerpc-linux-ld: drivers/gpu/drm/stm/dw_mipi_dsi-stm.o: in function `dw_mipi_dsi_stm_remove':
>>    dw_mipi_dsi-stm.c:(.text+0x664): undefined reference to `of_clk_del_provider'
>>>> powerpc-linux-ld: dw_mipi_dsi-stm.c:(.text+0x66c): undefined reference to `clk_hw_unregister'    powerpc-linux-ld: drivers/gpu/drm/stm/dw_mipi_dsi-stm.o: in function `dw_mipi_dsi_stm_probe':
>>    dw_mipi_dsi-stm.c:(.text+0x98c): undefined reference to `clk_hw_register'
>>
>>> So I think you should send that patch to Linus now so that such randconfig
>>> builds do not fail anymore.
>> What do you mean by 'sending it to Linus' ? If you meant to do a pull request,
>> then no. This patch is already in the drm-misc tree, which means it will keep
>> its usual pace of merging with the rest of the drm-misc tree.
>>
>> For more information about drm-misc tree:
>> https://drm.pages.freedesktop.org/maintainer-tools/repositories/drm-misc.html#merge-timeline
> That's not entirely correct. This should have been merged in
> drm-misc-fixes to begin with. I've cherry-picked the patch and pushed
> it.

Hi Maxime,

Oops I stand corrected, I effectively should have thought about sending this to
drm-misc-fixes.
Thank you for doing this.

Regards,

Raphaël

>
> Maxime