[PATCH 0/3] mm, pmem: Force unmap pmem on surprise remove

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Summary:

A dax_dev can be unbound from its driver at any time. Unbind can not
fail. The driver-core will always trigger ->remove() and the result from
->remove() is ignored. After ->remove() the driver-core proceeds to tear
down context. The filesystem-dax implementation can leave pfns mapped
after ->remove() if it is triggered while the filesystem is mounted.
Security and data-integrity is forfeit if the dax_dev is repurposed for
another security domain (new filesystem or change device modes), or if
the dax_dev is physically replaced. CXL is a hotplug bus that makes
dax_dev physical replace a real world prospect. 

All dax_dev pfns must be unmapped at remove. Detect the "remove while
mounted" case and trigger memory_failure() over the entire dax_dev
range.

Details:

The get_user_pages_fast() path expects all synchronization to be handled
by the pattern of checking for pte presence, taking a page reference,
and then validating that the pte was stable over that event. The
gup-fast path for devmap / DAX pages additionally attempts to take/hold
a live reference against the hosting pgmap over the page pin. The
rational for the pgmap reference is to synchronize against a dax-device
unbind / ->remove() event, but that is unnecessary if pte invalidation
is guaranteed in the ->remove() path.

Global dax-device pte invalidation *does* happen when the device is in
raw "device-dax" mode where there is a single shared inode to unmap at
remove, but the filesystem-dax path has a large number of actively
mapped inodes unknown to the driver at ->remove() time. So, that unmap
does not happen today for filesystem-dax. However, as Jason points out,
that unmap / invalidation *needs* to happen not only to cleanup
get_user_pages_fast() semantics, but in a future (see CXL) where dax_dev
->remove() is correlated with actual physical removal / replacement the
implications of allowing a physical pfn to be exchanged without tearing
down old mappings are severe (security and data-integrity).

What is not in this patch set is coordination with the dax_kmem driver
to trigger memory_failure() when the dax_dev is onlined as "System
RAM". The remove_memory() API was built with the assumption that
platform firmware negotiates all removal requests and the OS has a
chance to say "no". This is why dax_kmem today simply leaks
request_region() to burn that physical address space for any other
usage until the next reboot on a manual unbind event if the memory can't
be offlined. However a future to make sure that remove_memory() succeeds
after memory_failure() of the same range seems a better semantic than
permanently burning physical address space.

The topic of remove_memory() failures gets to the question of what
happens to active page references when the inopportune ->remove() event
happens. For transient pins the ->remove() event will wait for for all
pins to be dropped before allowing ->remove() to complete. Since
fileystem-dax forbids longterm pins all those pins are transient.
Device-dax, on the other hand, does allow longterm pins which means that
->remove() will hang unless / until the longterm pin is dropped.
Hopefully an unmap_mapping_range() event is sufficient to get the pin
dropped, but I suspect device-dax might need to trigger memory_failure()
as well to get the longterm pin holder to wake up and get out of the
way (TBD).

Lest we repeat the "longterm-pin-revoke" debate, which highlighted that
RDMA devices do not respond well to having context torn down, keep in
mind that this proposal is to do a best effort recovery of an event that
should not happen (surprise removal) under nominal operation.

---

Dan Williams (3):
      mm/memory-failure: Prepare for mass memory_failure()
      mm, dax, pmem: Introduce dev_pagemap_failure()
      mm/devmap: Remove pgmap accounting in the get_user_pages_fast() path


 drivers/dax/super.c      |   15 +++++++++++++++
 drivers/nvdimm/pmem.c    |   10 +++++++++-
 drivers/nvdimm/pmem.h    |    1 +
 include/linux/dax.h      |    5 +++++
 include/linux/memremap.h |    5 +++++
 include/linux/mm.h       |    3 +++
 mm/gup.c                 |   38 ++++++++++++++++----------------------
 mm/memory-failure.c      |   36 +++++++++++++++++++++++-------------
 mm/memremap.c            |   11 +++++++++++
 9 files changed, 88 insertions(+), 36 deletions(-)



[Index of Archives]     [Linux Ext4 Filesystem]     [Union Filesystem]     [Filesystem Testing]     [Ceph Users]     [Ecryptfs]     [AutoFS]     [Kernel Newbies]     [Share Photos]     [Security]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux Cachefs]     [Reiser Filesystem]     [Linux RAID]     [Samba]     [Device Mapper]     [CEPH Development]

  Powered by Linux