Re: [staging:staging-linus 11/11] (.text+0x4e8): multiple definition of `mite_dma_disarm'; drivers/comedi/drivers/mite.o:(.text+0x1c4): first defined here

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Wed, Jan 08, 2025 at 10:17:22AM +0100, Lukas Bulwahn wrote:
> Hi Greg, hi Dave,
> 
> I have seen this kernel test robot report
> (https://lore.kernel.org/all/Z34YDLM4zmg4cMpU@rli9-mobl/), as well as
> the second report (see
> https://lore.kernel.org/all/202501081213.5vHUhT8P-lkp@intel.com/).
> 
> I looked at the code and I am certain that the commit I submitted is
> not the actual root cause. The change I made has enabled the driver to
> be properly included in the build and hence, this is the first time
> the driver is actually properly compiled, showing these linker errors
> and warnings.
> 
> I assume Dave Penkler is going to fix these issues as he submitted the
> driver and knows best how to address the issues before the next merge
> window opens. If you think I should take care, I would probably just
> submit a patch to mark this driver as broken.
> 
> In my humble opinion, reverting the patch I submitted is not a good
> idea, but the forward path here is fixing the build issues or marking
> it as broken, and with that, have it parked in a state where it is
> clear that it is currently not being built and cannot be used.

I've fixed one of these now:
	https://lore.kernel.org/r/2025010809-padding-survive-91b3@gregkh
the other one is odd so we can ignore that for now until I figure it
out as the variable is being used in the MODULE_DEVICE_TABLE() macro.

And yes, it's not your fault at all :)

thanks,

greg k-h