Page MenuHomeFreeBSD

Adjust vm.max_wired in order to avoid hitting EAGAIN artificially
AbandonedPublic

Authored by ngie on Jan 4 2016, 4:54 AM.
Tags
None
Referenced Files
Unknown Object (File)
Fri, Apr 26, 1:01 AM
Unknown Object (File)
Fri, Apr 26, 12:22 AM
Unknown Object (File)
Fri, Apr 26, 12:21 AM
Unknown Object (File)
Dec 20 2023, 2:35 AM
Unknown Object (File)
Sep 25 2023, 8:03 AM
Unknown Object (File)
Aug 3 2023, 2:36 AM
Unknown Object (File)
Jul 16 2023, 6:48 AM
Unknown Object (File)
May 25 2023, 6:57 AM
Subscribers

Details

Reviewers
cem
Summary

Adjust vm.max_wired in order to avoid hitting EAGAIN artificially

Set vm.max_wired to INT_MAX in :mlock_err, :mlock_mmap, and :mlock_nested to
avoid hitting EAGAIN artificially on the system when running the tests

Require root privileges in order to set the sysctl

Add allow_sysctl_side_effects to require.config as this test is now adjusting
sysctls that can affect the global system state

MFC after: 1 week
Obtained from: Isilon OneFS (^/onefs/head@r511419)
Sponsored by: EMC / Isilon Storage Division

Diff Detail

Repository
rS FreeBSD src repository - subversion
Lint
No Lint Coverage
Unit
No Test Coverage
Build Status
Buildable 1938
Build 1945: arc lint + arc unit

Event Timeline

ngie retitled this revision from to Adjust vm.max_wired in order to avoid hitting EAGAIN artificially.
ngie updated this object.
ngie edited the test plan for this revision. (Show Details)
ngie added a reviewer: cem.
ngie added subscribers: benno, markj.

Please take me off "submitted by" — I didn't write this patch. I suggested the use of vm.max_wired as a workaround but my patch looked a lot different.

contrib/netbsd-tests/lib/libc/sys/t_mlock.c
97–98

I'm not sure what purpose writing out a "vm.max_wired" file serves.

ngie edited edge metadata.
contrib/netbsd-tests/lib/libc/sys/t_mlock.c
97–98

I'm using a file as scratch to save/restore the previous state of the sysctl. Code that "does things" (runs a sysctl, creates a file, etc) shouldn't be done in ATF_TP_ADD_TCS (that's what the code we have in OneFS does today in Riptide).

This code is always executed, but only once (the old code did it more than once in the successful case, IIRC..).

ngie marked 2 inline comments as done.