OpenBSD Journal

Call for testing: Kernel memory allocator diff

Contributed by ray on from the crash-test-dummies dept.

Theo de Raadt (deraadt@) has requested testing for a diff,

There is this diff that has been floating around in developer circles that has already uncovered two bugs.

The kernel has two main memory allocators; a kernel malloc and a pool allocator. The kernel malloc has a variety of corruption checking components, but the pool allocator has traditionally been weak in this department.

It is not yet clear if we want to commit this since it is a bit expensive. That said, this is in snapshots. But it sure would be nice if people who are interested run this themselves, especially if they are running into kernel bugs.

Thanks.

Please help Theo shake out those nasty kernel bugs!

(Comments are closed)


Comments
  1. By Anonymous Coward (85.19.213.88) on

    Is there anything in particular that we should do to properly test this diff, other than applying it to current and running the patched kernel for a while?

    Comments
    1. By Tobias Weingartner (68.148.4.19) on

      Generally just that. If you exercise your kernel a bit more than usual,
      that is a good thing IMHO. Corruption sometimes only shows up after an
      extended up time. Other times it shows up right away.

  2. By Steven (142.179.226.243) steven_schneider@telus.net on

    If one is running snapshots then do you still have to apply the diff if you want to help test? From theo's post I have the impression it's already compiled into the snapshots.

    Please feel free to correct me if I'm mistaken.

    Comments
    1. By phessler (phessler) on http://theapt.org

      > If one is running snapshots then do you still have to apply the diff if you want to help test? From theo's post I have the impression it's already compiled into the snapshots.
      >
      > Please feel free to correct me if I'm mistaken.


      Yes, it is in snapshots. Please run with it both in your normal duties, and try to abuse it (as you defined abuse :) ).

  3. By dcoppa (195.35.99.31) on

    I'm not using snapshots, I've applied the diff to a -current src tree... Btw, no regression so far, it seems all good to me (as usual with openbsd ;) )

    Comments
    1. By Insan Praja SW (202.90.194.171) on

      > I'm not using snapshots, I've applied the diff to a -current src tree... Btw, no regression so far, it seems all good to me (as usual with openbsd ;) )
      >
      >

      I use november 11th 2008 source-tree, compile it on my routers.. it's fine so far.. I've noticed faster reboot on my machine and more windows on the systat.. nice

Latest Articles

Credits

Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]