[v4,0/5] mm: Fix pfn_to_online_page() with respect to ZONE_DEVICE

Message

Dan Williams Jan. 14, 2021, 12:43 a.m. UTC

Changes since v3 [1]:
- Switch to "if (IS_ENABLED(CONFIG_HAVE_ARCH_PFN_VALID) &&
  !pfn_valid(pfn))" (David)
- Finish collecting reviewed-bys across all patches in the series
- Drop the libnvdimm fixup, to be merged through nvdimm.git not -mm

[1]: http://lore.kernel.org/r/[e-mail]

---

Andrew,

All patches in this series have been reviewed and the kbuild-robot
reports a build-success over 172 configs. They pass an updated version
of the nvdimm unit tests to exercise corner cases of
pfn_to_online_page() and get_dev_pagemap() [2], and apply cleanly to
current -next.

Please apply, thanks.

[2]: http://lore.kernel.org/r/[e-mail]

---

Michal reminds that the discussion about how to ensure pfn-walkers do
not get confused by ZONE_DEVICE pages never resolved. A pfn-walker that
uses pfn_to_online_page() may inadvertently translate a pfn as online
and in the page allocator, when it is offline managed by a ZONE_DEVICE
mapping (details in Patch 3: ("mm: Teach pfn_to_online_page() about
ZONE_DEVICE section collisions")).

The 2 proposals under consideration are teach pfn_to_online_page() to be
precise in the presence of mixed-zone sections, or teach the memory-add
code to drop the System RAM associated with ZONE_DEVICE collisions. In
order to not regress memory capacity by a few 10s to 100s of MiB the
approach taken in this set is to add precision to pfn_to_online_page().

In the course of validating pfn_to_online_page() a couple other fixes
fell out:

1/ soft_offline_page() fails to drop the reference taken in the
   madvise(..., MADV_SOFT_OFFLINE) case.

2/ memory_failure() uses get_dev_pagemap() to lookup ZONE_DEVICE pages,
   however that mapping may contain data pages and metadata raw pfns.
   Introduce pgmap_pfn_valid() to delineate the 2 types and fail the
   handling of raw metadata pfns.

---

Dan Williams (5):
      mm: Move pfn_to_online_page() out of line
      mm: Teach pfn_to_online_page() to consider subsection validity
      mm: Teach pfn_to_online_page() about ZONE_DEVICE section collisions
      mm: Fix page reference leak in soft_offline_page()
      mm: Fix memory_failure() handling of dax-namespace metadata


 include/linux/memory_hotplug.h |   17 +---------
 include/linux/memremap.h       |    6 +++
 include/linux/mmzone.h         |   22 +++++++++----
 mm/memory-failure.c            |   26 +++++++++++++--
 mm/memory_hotplug.c            |   69 ++++++++++++++++++++++++++++++++++++++++
 mm/memremap.c                  |   15 +++++++++
 6 files changed, 128 insertions(+), 27 deletions(-)