Hi Sasha,

On Thu, Dec 20, 2018 at 07:26:15PM +0000, Sasha Levin wrote:

Hi,

[This is an automated email]

This commit has been processed because it contains a "Fixes:" tag,
fixing commit: 432c6bacbd0c MIPS: Use per-mm page to execute branch delay slot instructions.

The bot has tested the following trees: v4.19.10, v4.14.89, v4.9.146, 

Neat! I like the idea of this automation :)

v4.19.10: Build OK!
v4.14.89: Build OK!
v4.9.146: Failed to apply! Possible dependencies:
    05ce77249d50 ("userfaultfd: non-cooperative: add madvise() event for MADV_DONTNEED request")
    163e11bc4f6e ("userfaultfd: hugetlbfs: UFFD_FEATURE_MISSING_HUGETLBFS")
    67dece7d4c58 ("x86/vdso: Set vDSO pointer only after success")
    72f87654c696 ("userfaultfd: non-cooperative: add mremap() event")
    893e26e61d04 ("userfaultfd: non-cooperative: Add fork() event")
    897ab3e0c49e ("userfaultfd: non-cooperative: add event for memory unmaps")
    9cd75c3cd4c3 ("userfaultfd: non-cooperative: add ability to report non-PF events from uffd descriptor")
    d811914d8757 ("userfaultfd: non-cooperative: rename *EVENT_MADVDONTNEED to *EVENT_REMOVE")

This list includes the correct soft dependency - commit 897ab3e0c49e
("userfaultfd: non-cooperative: add event for memory unmaps") which
added an extra argument to mmap_region().

How should we proceed with this patch?

The backport to v4.9 should simply drop the last argument (NULL) in the
call to mmap_region().

Is there some way I can indicate this sort of thing in future patches so
that the automation can spot that I already know it won't apply cleanly
to a particular range of kernel versions? Or even better, that I could
indicate what needs to be changed when backporting to those kernel
versions?

Thanks,
    Paul