Re: clean up and streamline probe_kernel_* and friends v4

On Tue, 26 May 2020 08:13:09 +0200 Christoph Hellwig [off-list ref] wrote:

On Mon, May 25, 2020 at 03:19:12PM -0700, Andrew Morton wrote:

quoted

hm.  Applying linux-next to this series generates a lot of rejects against
powerpc:

-rw-rw-r-- 1 akpm akpm  493 May 25 15:06 arch/powerpc/kernel/kgdb.c.rej
-rw-rw-r-- 1 akpm akpm 6461 May 25 15:06 arch/powerpc/kernel/trace/ftrace.c.rej
-rw-rw-r-- 1 akpm akpm  447 May 25 15:06 arch/powerpc/mm/fault.c.rej
-rw-rw-r-- 1 akpm akpm  623 May 25 15:06 arch/powerpc/perf/core-book3s.c.rej
-rw-rw-r-- 1 akpm akpm 1408 May 25 15:06 arch/riscv/kernel/patch.c.rej

the arch/powerpc/kernel/trace/ftrace.c ones aren't very trivial.

It's -rc7.  Perhaps we should park all this until 5.8-rc1?

As this is a pre-condition for the set_fs removal I'd really like to
get the actual changes in.  All these conflicts seem to be about the
last three cleanup patches just doing renaming, so can we just skip
those three for now?  Then we can do the rename right after 5.8-rc1
when we have the least chances for conflicts.

That seems to have worked.  "[PATCH 23/23] maccess: return -ERANGE when
copy_from_kernel_nofault_allowed fails" needed a bit of massaging to both
the patch and to the patch title.