[v5,0/6] virtio pmem driver

Message

Pankaj Gupta April 10, 2019, 4:08 a.m. UTC

This patch series has implementation for "virtio pmem". 
 "virtio pmem" is fake persistent memory(nvdimm) in guest 
 which allows to bypass the guest page cache. This also
 implements a VIRTIO based asynchronous flush mechanism.  
 
 Sharing guest kernel driver in this patchset with the 
 changes suggested in v4. Tested with Qemu side device 
 emulation [6] for virtio-pmem. 

 We have incorporated all the suggestions in V4. Documented 
 the impact of possible page cache side channel attacks with 
 suggested countermeasures.
 
 Details of project idea for 'virtio pmem' flushing interface 
 is shared [3] & [4].

 Implementation is divided into two parts:
 New virtio pmem guest driver and qemu code changes for new 
 virtio pmem paravirtualized device.

1. Guest virtio-pmem kernel driver
---------------------------------
   - Reads persistent memory range from paravirt device and 
     registers with 'nvdimm_bus'.  
   - 'nvdimm/pmem' driver uses this information to allocate 
     persistent memory region and setup filesystem operations 
     to the allocated memory. 
   - virtio pmem driver implements asynchronous flushing 
     interface to flush from guest to host.

2. Qemu virtio-pmem device
---------------------------------
   - Creates virtio pmem device and exposes a memory range to 
     KVM guest. 
   - At host side this is file backed memory which acts as 
     persistent memory. 
   - Qemu side flush uses aio thread pool API's and virtio 
     for asynchronous guest multi request handling. 

   David Hildenbrand CCed also posted a modified version[7] of 
   qemu virtio-pmem code based on updated Qemu memory device API. 

 Virtio-pmem security implications and countermeasures:
 -----------------------------------------------------

 In previous posting of kernel driver, there was discussion [9]
 on possible implications of page cache side channel attacks with 
 virtio pmem. After thorough analysis of details of known side 
 channel attacks, below are the suggestions:

 - Depends entirely on how host backing image file is mapped 
   into guest address space. 

 - virtio-pmem device emulation, by default shared mapping is used
   to map host backing file. It is recommended to use separate
   backing file at host side for every guest. This will prevent
   any possibility of executing common code from multiple guests
   and any chance of inferring guest local data based based on 
   execution time.

 - If backing file is required to be shared among multiple guests 
   it is recommended to don't support host page cache eviction 
   commands from the guest driver. This will avoid any possibility
   of inferring guest local data or host data from another guest. 

 - Proposed device specification [8] for virtio-pmem device with 
   details of possible security implications and suggested 
   countermeasures for device emulation.

 Virtio-pmem errors handling:
 ----------------------------------------
  Checked behaviour of virtio-pmem for below types of errors
  Need suggestions on expected behaviour for handling these errors?

  - Hardware Errors: Uncorrectable recoverable Errors: 
  a] virtio-pmem: 
    - As per current logic if error page belongs to Qemu process, 
      host MCE handler isolates(hwpoison) that page and send SIGBUS. 
      Qemu SIGBUS handler injects exception to KVM guest. 
    - KVM guest then isolates the page and send SIGBUS to guest 
      userspace process which has mapped the page. 
  
  b] Existing implementation for ACPI pmem driver: 
    - Handles such errors with MCE notifier and creates a list 
      of bad blocks. Read/direct access DAX operation return EIO 
      if accessed memory page fall in bad block list.
    - It also starts backgound scrubbing.  
    - Similar functionality can be reused in virtio-pmem with MCE 
      notifier but without scrubbing(no ACPI/ARS)? Need inputs to 
      confirm if this behaviour is ok or needs any change?

Changes from PATCH v4: [1]
- Factor out MAP_SYNC supported functionality to a common helper
				[Dave, Darrick, Jan]
- Comment, indentation and virtqueue_kick failure handle - Yuval Shaia

Changes from PATCH v3: [2]
- Use generic dax_synchronous() helper to check for DAXDEV_SYNC 
  flag - [Dan, Darrick, Jan]
- Add 'is_nvdimm_async' function
- Document page cache side channel attacks implications & 
  countermeasures - [Dave Chinner, Michael]

Changes from PATCH v2: 
- Disable MAP_SYNC for ext4 & XFS filesystems - [Dan] 
- Use name 'virtio pmem' in place of 'fake dax' 

Changes from PATCH v1: 
- 0-day build test for build dependency on libnvdimm 

 Changes suggested by - [Dan Williams]
- Split the driver into two parts virtio & pmem  
- Move queuing of async block request to block layer
- Add "sync" parameter in nvdimm_flush function
- Use indirect call for nvdimm_flush
- Don’t move declarations to common global header e.g nd.h
- nvdimm_flush() return 0 or -EIO if it fails
- Teach nsio_rw_bytes() that the flush can fail
- Rename nvdimm_flush() to generic_nvdimm_flush()
- Use 'nd_region->provider_data' for long dereferencing
- Remove virtio_pmem_freeze/restore functions
- Remove BSD license text with SPDX license text

- Add might_sleep() in virtio_pmem_flush - [Luiz]
- Make spin_lock_irqsave() narrow

Changes from RFC v3
- Rebase to latest upstream - Luiz
- Call ndregion->flush in place of nvdimm_flush- Luiz
- kmalloc return check - Luiz
- virtqueue full handling - Stefan
- Don't map entire virtio_pmem_req to device - Stefan
- request leak, correct sizeof req- Stefan
- Move declaration to virtio_pmem.c

Changes from RFC v2:
- Add flush function in the nd_region in place of switching
  on a flag - Dan & Stefan
- Add flush completion function with proper locking and wait
  for host side flush completion - Stefan & Dan
- Keep userspace API in uapi header file - Stefan, MST
- Use LE fields & New device id - MST
- Indentation & spacing suggestions - MST & Eric
- Remove extra header files & add licensing - Stefan

Changes from RFC v1:
- Reuse existing 'pmem' code for registering persistent 
  memory and other operations instead of creating an entirely 
  new block driver.
- Use VIRTIO driver to register memory information with 
  nvdimm_bus and create region_type accordingly. 
- Call VIRTIO flush from existing pmem driver.

Pankaj Gupta (6):
   libnvdimm: nd_region flush callback support
   virtio-pmem: Add virtio-pmem guest driver
   libnvdimm: add nd_region buffered dax_dev flag
   dax: check synchronous mapping is supported
   ext4: disable map_sync for virtio pmem
   xfs: disable map_sync for virtio pmem

[1] https://lkml.org/lkml/2019/4/3/394
[2] https://lkml.org/lkml/2019/1/9/471
[3] https://www.spinics.net/lists/kvm/msg149761.html
[4] https://www.spinics.net/lists/kvm/msg153095.html  
[5] https://lkml.org/lkml/2018/8/31/413
[6] https://marc.info/?l=linux-kernel&m=153572228719237&w=2 
[7] https://marc.info/?l=qemu-devel&m=153555721901824&w=2
[8] https://lists.oasis-open.org/archives/virtio-dev/201903/msg00083.html
[9] https://lkml.org/lkml/2019/1/9/1191

 drivers/acpi/nfit/core.c         |    4 -
 drivers/dax/bus.c                |    2 
 drivers/dax/super.c              |   13 +++-
 drivers/md/dm.c                  |    2 
 drivers/nvdimm/claim.c           |    6 +
 drivers/nvdimm/nd.h              |    1 
 drivers/nvdimm/pmem.c            |   17 +++--
 drivers/nvdimm/region_devs.c     |   45 +++++++++++++-
 drivers/nvdimm/virtio_pmem.c     |   88 +++++++++++++++++++++++++++
 drivers/virtio/Kconfig           |   10 +++
 drivers/virtio/Makefile          |    1 
 drivers/virtio/pmem.c            |  124 +++++++++++++++++++++++++++++++++++++++
 fs/ext4/file.c                   |   11 +--
 fs/xfs/xfs_file.c                |   10 +--
 include/linux/dax.h              |   32 +++++++++-
 include/linux/libnvdimm.h        |    9 ++
 include/linux/virtio_pmem.h      |   60 ++++++++++++++++++
 include/uapi/linux/virtio_ids.h  |    1 
 include/uapi/linux/virtio_pmem.h |   10 +++
 19 files changed, 419 insertions(+), 27 deletions(-)

Comments

Arkadiusz Miśkiewicz April 10, 2019, 8:08 a.m. UTC | #1

On 10/04/2019 06:08, Pankaj Gupta wrote:
>  This patch series has implementation for "virtio pmem". 
>  "virtio pmem" is fake persistent memory(nvdimm) in guest 
>  which allows to bypass the guest page cache. This also

Will kernel pstore be able to use this persistent memory for storing
crash dumps?