mbox series

[RFC,0/4] mm/arm64: re-enable HVO

Message ID 20240806022114.3320543-1-yuzhao@google.com (mailing list archive)
Headers show
Series mm/arm64: re-enable HVO | expand

Message

Yu Zhao Aug. 6, 2024, 2:21 a.m. UTC 
This series presents one of the previously discussed approaches to
re-enable HugeTLB Vmemmap Optimization (HVO) on arm64. HVO was
disabled by commit 060a2c92d1b6 ("arm64: mm: hugetlb: Disable
HUGETLB_PAGE_OPTIMIZE_VMEMMAP") due to the following reason:

  This is deemed UNPREDICTABLE by the Arm architecture without a
  break-before-make sequence (make the PTE invalid, TLBI, write the
  new valid PTE). However, such sequence is not possible since the
  vmemmap may be concurrently accessed by the kernel.

Other approaches that have been discussed include:
  A. Handle kernel PF while doing BBM [1],
  B. Use stop_machine() while doing BBM [2], and,
  C. Enable FEAT_BBM level 2 and keep the memory contents at the old
     and new output addresses unchanged to avoid BBM (D8.16.1-2) [3].

A quick comparison between this approach (D) and the above approaches:
  --+------------------------------+-----------------------------+
    |              Pro             |             Con             |
  --+------------------------------+-----------------------------+
  A | Low latency, h/w independent | Predictability concerns [4] |
  B | Predictable, h/w independent | High latency                |
  C | Predictable, low latency     | H/w dependent, complex      |
  D | Predictable, h/w independent | Medium latency              |
  --+------------------------------+-----------------------------+

[1] https://lore.kernel.org/20240113094436.2506396-1-sunnanyong@huawei.com/
[2] https://lore.kernel.org/ZbKjHHeEdFYY1xR5@arm.com/
[3] https://lore.kernel.org/Zo68DP6siXfb6ZBR@arm.com/
[4] https://lore.kernel.org/20240326125409.GA9552@willie-the-truck/

Nanyong Sun (2):
  mm: HVO: introduce helper function to update and flush pgtable
  arm64: mm: Re-enable OPTIMIZE_HUGETLB_VMEMMAP

Yu Zhao (2):
  arm64: use IPIs to pause/resume remote CPUs
  arm64: pause remote CPUs to update vmemmap

 arch/arm64/Kconfig               |   1 +
 arch/arm64/include/asm/pgalloc.h |  55 ++++++++++++++++
 arch/arm64/include/asm/smp.h     |   3 +
 arch/arm64/kernel/smp.c          | 110 +++++++++++++++++++++++++++++++
 mm/hugetlb_vmemmap.c             |  69 +++++++++++++++----
 5 files changed, 226 insertions(+), 12 deletions(-)


base-commit: de9c2c66ad8e787abec7c9d7eff4f8c3cdd28aed