Christophe Leroy [off-list ref] writes:

On 02/08/2019 12:34 PM, Michael Ellerman wrote:

quoted

In commit 7820856a4fcd ("powerpc/mm/book3e/64: Remove unsupported
64Kpage size from 64bit booke") we dropped the 64K page size support
from the 64-bit nohash (Book3E) code.

But we didn't update the dependencies of the PPC_64K_PAGES option,
meaning a randconfig can still trigger this code and cause a build
breakage, eg:
   arch/powerpc/include/asm/nohash/64/pgtable.h:14:2: error: #error "Page size not supported"
   arch/powerpc/include/asm/nohash/mmu-book3e.h:275:2: error: #error Unsupported page size

So remove PPC_BOOK3E_64 from the dependencies. This also means we
don't need to worry about PPC_FSL_BOOK3E, because that was just trying
to prevent the PPC_BOOK3E_64=y && PPC_FSL_BOOK3E=y case.

Does it means some cleanup could be done, for instance:

arch/powerpc/include/asm/nohash/64/pgalloc.h:#ifndef CONFIG_PPC_64K_PAGES
arch/powerpc/include/asm/nohash/64/pgalloc.h:#endif /* 
CONFIG_PPC_64K_PAGES */
arch/powerpc/include/asm/nohash/64/pgtable.h:#ifdef CONFIG_PPC_64K_PAGES
arch/powerpc/include/asm/nohash/64/slice.h:#ifdef CONFIG_PPC_64K_PAGES
arch/powerpc/include/asm/nohash/64/slice.h:#else /* CONFIG_PPC_64K_PAGES */
arch/powerpc/include/asm/nohash/64/slice.h:#endif /* 
!CONFIG_PPC_64K_PAGES */
arch/powerpc/include/asm/nohash/pte-book3e.h:#ifdef CONFIG_PPC_64K_PAGES

arch/powerpc/mm/tlb_low_64e.S:#ifdef CONFIG_PPC_64K_PAGES
arch/powerpc/mm/tlb_low_64e.S:#ifndef CONFIG_PPC_64K_PAGES
arch/powerpc/mm/tlb_low_64e.S:#endif /* CONFIG_PPC_64K_PAGES */
arch/powerpc/mm/tlb_low_64e.S:#ifdef CONFIG_PPC_64K_PAGES
arch/powerpc/mm/tlb_low_64e.S:#ifdef CONFIG_PPC_64K_PAGES
arch/powerpc/mm/tlb_low_64e.S:#ifndef CONFIG_PPC_64K_PAGES
arch/powerpc/mm/tlb_low_64e.S:#endif /* CONFIG_PPC_64K_PAGES */
arch/powerpc/mm/tlb_low_64e.S:#ifndef CONFIG_PPC_64K_PAGES
arch/powerpc/mm/tlb_low_64e.S:#endif /* CONFIG_PPC_64K_PAGES */
arch/powerpc/mm/tlb_low_64e.S:#ifndef CONFIG_PPC_64K_PAGES
arch/powerpc/mm/tlb_low_64e.S:#ifdef CONFIG_PPC_64K_PAGES

Probably.

Some of the FSL chips do support 64K pages at least according to some
datasheets. I don't know what would be required to get it working, or if
it even works in practice.

So it would be nice to get 64K working on those chips, but probably no
one has time or motivation to do it. In which case yeah all that code
should be removed.

cheers