linux-kernel - [PATCH resend v2 0/5] mm/madvise: introduce MADV_POPULATE

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 11 May 2021 10:15:29 +0200
From:   David Hildenbrand <david@...hat.com>
To:     linux-kernel@...r.kernel.org
Cc:     linux-mm@...ck.org, David Hildenbrand <david@...hat.com>
Subject: [PATCH resend v2 0/5] mm/madvise: introduce MADV_POPULATE_(READ|WRITE) to prefault page tables

This is a resend of v2 because linux-mm seems to have swallowed patch #2.

Excessive details on MADV_POPULATE_(READ|WRITE) can be found in patch #2.

v1 -> v2:
- "mm/madvise: introduce MADV_POPULATE_(READ|WRITE) to prefault page
   tables"
-- Renamed patch/series to match what's happening -- prefault page tables
-- Clarified MADV_POPULATE_READ semantics on file holes and that we might
   want fallocate().
-- Updated/clarified description
-- Dropped -EINVAL and -EBUSY checks
-- Added a comment regarding FOLL_TOUCH and why we don't care that
   pages will get set dirty when triggering write-faults for now.
-- Reran and extended performance measurements by more fallocate()
   combinations

RFCv2 -> v1
- "mm: fix variable name in declaration of populate_vma_page_range()"
-- Added
- "mm/madvise: introduce MADV_POPULATE_(READ|WRITE) to prefault ..."
-- Fix detection of memory holes when we have to re-lookup the VMA
-- Return -EHWPOISON to user space when we hit HW poisoned pages
-- Make variable names in definition and declaration consistent
- "MAINTAINERS: add tools/testing/selftests/vm/ to MEMORY MANAGEMENT"
-- Added
- "selftests/vm: add protection_keys_32 / protection_keys_64 to gitignore"
-- Added
- "selftests/vm: add test for MADV_POPULATE_(READ|WRITE)"
-- Added

RFC -> RFCv2:
- Fix re-locking (-> set "locked = 1;")
- Don't mimic MAP_POPULATE semantics:
--> Explicit READ/WRITE request instead of selecting it automatically,
    which makes it more generic and better suited for some use cases (e.g., we
    usually want to prefault shmem writable)
--> Require proper access permissions
- Introduce and use faultin_vma_page_range()
--> Properly handle HWPOISON pages (FOLL_HWPOISON)
--> Require proper access permissions (!FOLL_FORCE)
- Let faultin_vma_page_range() check for compatible mappings/permissions
- Extend patch description and add some performance numbers

David Hildenbrand (5):
  mm: make variable names for populate_vma_page_range() consistent
  mm/madvise: introduce MADV_POPULATE_(READ|WRITE) to prefault page
    tables
  MAINTAINERS: add tools/testing/selftests/vm/ to MEMORY MANAGEMENT
  selftests/vm: add protection_keys_32 / protection_keys_64 to gitignore
  selftests/vm: add test for MADV_POPULATE_(READ|WRITE)

 MAINTAINERS                                |   1 +
 arch/alpha/include/uapi/asm/mman.h         |   3 +
 arch/mips/include/uapi/asm/mman.h          |   3 +
 arch/parisc/include/uapi/asm/mman.h        |   3 +
 arch/xtensa/include/uapi/asm/mman.h        |   3 +
 include/uapi/asm-generic/mman-common.h     |   3 +
 mm/gup.c                                   |  58 ++++
 mm/internal.h                              |   5 +-
 mm/madvise.c                               |  66 ++++
 tools/testing/selftests/vm/.gitignore      |   3 +
 tools/testing/selftests/vm/Makefile        |   1 +
 tools/testing/selftests/vm/madv_populate.c | 342 +++++++++++++++++++++
 tools/testing/selftests/vm/run_vmtests.sh  |  16 +
 13 files changed, 506 insertions(+), 1 deletion(-)
 create mode 100644 tools/testing/selftests/vm/madv_populate.c


base-commit: bf05bf16c76bb44ab5156223e1e58e26dfe30a88
-- 
2.30.2