linux-stable.git/tools/testing/selftests/vm, branch linux-4.9.y

kselftest/vm: fix tests build with old libc

2022-03-23T08:00:33+00:00

[ Upstream commit b773827e361952b3f53ac6fa4c4e39ccd632102e ]

The error message when I build vm tests on debian10 (GLIBC 2.28):

    userfaultfd.c: In function `userfaultfd_pagemap_test':
    userfaultfd.c:1393:37: error: `MADV_PAGEOUT' undeclared (first use
    in this function); did you mean `MADV_RANDOM'?
      if (madvise(area_dst, test_pgsize, MADV_PAGEOUT))
                                         ^~~~~~~~~~~~
                                         MADV_RANDOM

This patch includes these newer definitions from UAPI linux/mman.h, is
useful to fix tests build on systems without these definitions in glibc
sys/mman.h.

Link: https://lkml.kernel.org/r/20220227055330.43087-2-zhouchengming@bytedance.com
Signed-off-by: Chengming Zhou 
Reviewed-by: Shuah Khan 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Sasha Levin

kselftest: fix OOM in memory compaction test

2018-02-22T14:43:52+00:00

commit 4c1baad223906943b595a887305f2e8124821dad upstream.

Running the compaction_test sometimes results in out-of-memory
failures. When I debugged this, it turned out that the code to
reset the number of hugepages to the initial value is simply
broken since we write into an open sysctl file descriptor
multiple times without seeking back to the start.

Adding the lseek here fixes the problem.

Cc: stable@vger.kernel.org
Reported-by: Naresh Kamboju 
Link: https://bugs.linaro.org/show_bug.cgi?id=3145
Signed-off-by: Arnd Bergmann 
Signed-off-by: Shuah Khan 
Signed-off-by: Greg Kroah-Hartman

userfaultfd: selftest: vm: allow to build in vm/ directory

2017-12-20T09:07:18+00:00

[ Upstream commit 46aa6a302b53f543f8e8b8e1714dc5e449ad36a6 ]

linux/tools/testing/selftests/vm $ make

  gcc -Wall -I ../../../../usr/include     compaction_test.c -lrt -o /compaction_test
  /usr/lib/gcc/x86_64-pc-linux-gnu/4.9.4/../../../../x86_64-pc-linux-gnu/bin/ld: cannot open output file /compaction_test: Permission denied
  collect2: error: ld returned 1 exit status
  make: *** [../lib.mk:54: /compaction_test] Error 1

Since commit a8ba798bc8ec ("selftests: enable O and KBUILD_OUTPUT")
selftests/vm build fails if run from the "selftests/vm" directory, but
it works in the selftests/ directory.  It's quicker to be able to do a
local vm-only build after a tree wipe and this patch allows for it
again.

Link: http://lkml.kernel.org/r/20170302173738.18994-4-aarcange@redhat.com
Signed-off-by: Andrea Arcangeli 
Cc: Mike Rapoport 
Cc: "Dr. David Alan Gilbert" 
Cc: Mike Kravetz 
Cc: Pavel Emelyanov 
Cc: Hillf Danton 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

selftests: expanding more mlock selftest

2016-10-08T01:46:28+00:00

This patch will randomly perform mlock/mlock2 on a given memory region,
and verify the RLIMIT_MEMLOCK limitation works properly.

Suggested-by: David Rientjes 
Link: http://lkml.kernel.org/r/1473325970-11393-4-git-send-email-wei.guo.simon@gmail.com
Signed-off-by: Simon Guo 
Cc: Shuah Khan 
Cc: Vlastimil Babka 
Cc: "Kirill A. Shutemov" 
Cc: Michal Hocko 
Cc: Eric B Munson 
Cc: Simon Guo 
Cc: Mel Gorman 
Cc: Alexey Klimov 
Cc: Andrea Arcangeli 
Cc: Thierry Reding 
Cc: Mike Kravetz 
Cc: Geert Uytterhoeven 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

selftest: move seek_to_smaps_entry() out of mlock2-tests.c

2016-10-08T01:46:28+00:00

Function seek_to_smaps_entry() can be useful for other selftest
functionalities, so move it out to header file.

Link: http://lkml.kernel.org/r/1473325970-11393-3-git-send-email-wei.guo.simon@gmail.com
Signed-off-by: Simon Guo 
Cc: Shuah Khan 
Cc: Vlastimil Babka 
Cc: "Kirill A. Shutemov" 
Cc: Michal Hocko 
Cc: Eric B Munson 
Cc: Simon Guo 
Cc: Mel Gorman 
Cc: Alexey Klimov 
Cc: Andrea Arcangeli 
Cc: Thierry Reding 
Cc: Mike Kravetz 
Cc: Geert Uytterhoeven 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

selftests/vm: add test for mlock() when areas are intersected

2016-10-08T01:46:28+00:00

This patch adds mlock() test for multiple invocation on the same address
area, and verify it doesn't mess the rlimit mlock limitation.

Link: http://lkml.kernel.org/r/1472554781-9835-5-git-send-email-wei.guo.simon@gmail.com
Signed-off-by: Simon Guo 
Cc: Alexey Klimov 
Cc: Eric B Munson 
Cc: Geert Uytterhoeven 
Cc: "Kirill A. Shutemov" 
Cc: Mel Gorman 
Cc: Michal Hocko 
Cc: Shuah Khan 
Cc: Simon Guo 
Cc: Thierry Reding 
Cc: Vlastimil Babka 
Cc: David Rientjes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

selftest: split mlock2_ funcs into separate mlock2.h

2016-10-08T01:46:28+00:00

To prepare mlock2.h whose functionality will be reused.

Link: http://lkml.kernel.org/r/1472554781-9835-4-git-send-email-wei.guo.simon@gmail.com
Signed-off-by: Simon Guo 
Cc: Alexey Klimov 
Cc: Eric B Munson 
Cc: Geert Uytterhoeven 
Cc: "Kirill A. Shutemov" 
Cc: Mel Gorman 
Cc: Michal Hocko 
Cc: Shuah Khan 
Cc: Simon Guo 
Cc: Thierry Reding 
Cc: Vlastimil Babka 
Cc: David Rientjes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

selftests/vm: Don't mlockall MCL_CURRENT in on-fault-limit test

2016-06-27T19:19:24+00:00

The default MEMLOCK limit is not big enough to accomodate all the
current pages of the test program process, so the test fails
at this step.
By removing the MCL_CURRENT flag, we allow the mlockall
call to succeed. The mmap is twice the size of the current limit,
so it will still fail as expected.

Signed-off-by: Yannick Brosseau 
Signed-off-by: Shuah Khan

selftests/vm: write strlen length instead of sizeof to nr_hugepages

2016-06-27T19:18:26+00:00

When setting back the initial value to nr_hugepages, the
test was writing a length sizeof of the string and checking
that strlen was writen. Since those values are not the same,
use strlen in both place instead.

Also make the error messages more explicit to help in future
debugging.

Signed-off-by: Yannick Brosseau 
Signed-off-by: Shuah Khan

selftests/thuge-gen: Use platform specific MAP_HUGETLB value

2016-05-22T19:43:41+00:00

Do not hardcode MAP_HUGETLB to 0x40000, since quite some architectures
use a different value.

Tested with a parisc architecture 64bit kernel.

Signed-off-by: Helge Deller