linux-kernel - [PATCH v4 0/2] HMM tests and minor fixes

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:   Mon, 4 Nov 2019 14:21:39 -0800
From:   Ralph Campbell <rcampbell@...dia.com>
To:     Jerome Glisse <jglisse@...hat.com>,
        John Hubbard <jhubbard@...dia.com>,
        Christoph Hellwig <hch@....de>,
        Jason Gunthorpe <jgg@...lanox.com>,
        "Shuah Khan" <shuah@...nel.org>
CC:     <linux-rdma@...r.kernel.org>, <linux-mm@...ck.org>,
        <linux-kernel@...r.kernel.org>, <linux-kselftest@...r.kernel.org>,
        "Ralph Campbell" <rcampbell@...dia.com>
Subject: [PATCH v4 0/2] HMM tests and minor fixes

These changes are based on Jason's rdma/hmm branch (5.4.0-rc5).
Patch 1 was previously posted here [1] but was dropped from that orginal
series. Hopefully, the tests will reduce concerns about edge conditions.
I'm sure more tests could be usefully added but I thought this was a good
starting point.

Changes since v3:
patch 1:
Unchanged except rebased on Jason's latest hmm (bbe3329e354d3ab5dc18).
patch 2:
Is now part of Jason's tree.
patch 3 (now 2):
Major changes to incorporate Jason's review feedback.
  * drivers/char/hmm_dmirror.c driver moved to lib/test_hmm.c
  * XArray used instead of "page tables".
  * platform device driver removed.
  * remove redundant copyright.

Changes since v2:
patch 1:
Removed hmm_range_needs_fault() and just use hmm_range_need_fault().
Updated the change log to include that it fixes a bug where
hmm_range_fault() incorrectly returned an error when no fault is requested.
patch 2:
Removed the confusing change log wording about DMA.
Changed hmm_range_fault() to return the PFN of the zero page like any other
page.
patch 3:
Adjusted the test code to match the new zero page behavior.

Changes since v1:
Rebased to Jason's rdma/hmm branch (5.4.0-rc1).
Cleaned up locking for the test driver's page tables.
Incorporated Christoph Hellwig's comments.

[1] https://lore.kernel.org/linux-mm/20190726005650.2566-6-rcampbell@nvidia.com/

Ralph Campbell (2):
  mm/hmm: make full use of walk_page_range()
  mm/hmm/test: add self tests for HMM

 MAINTAINERS                            |    3 +
 include/uapi/linux/test_hmm.h          |   59 ++
 lib/Kconfig.debug                      |   11 +
 lib/Makefile                           |    1 +
 lib/test_hmm.c                         | 1306 ++++++++++++++++++++++++
 mm/hmm.c                               |  121 ++-
 tools/testing/selftests/vm/.gitignore  |    1 +
 tools/testing/selftests/vm/Makefile    |    3 +
 tools/testing/selftests/vm/config      |    2 +
 tools/testing/selftests/vm/hmm-tests.c | 1295 +++++++++++++++++++++++
 tools/testing/selftests/vm/run_vmtests |   16 +
 tools/testing/selftests/vm/test_hmm.sh |   97 ++
 12 files changed, 2852 insertions(+), 63 deletions(-)
 create mode 100644 include/uapi/linux/test_hmm.h
 create mode 100644 lib/test_hmm.c
 create mode 100644 tools/testing/selftests/vm/hmm-tests.c
 create mode 100755 tools/testing/selftests/vm/test_hmm.sh

-- 
2.20.1