Update ZFS "todo" to say it's being done. Perhaps there is more than people

can help pjd with here, but they need to contact him first.  We need to
decide if this item should still be on the TODO list.

Note that jb's work on dtrace is a significant work-in-progress so to
contact him.  Does this need to be a todo item?

Note that the "use floating point in kernel rah rah rah" item requires
experimental validation -- the task isn't to hack code, but to decide if
the hacked code was a good idea.  Does this need to be a todo item?

Note that the DFBSD process checkpointing facility is both experimental and
incomplete.  Another "Don't just port this" sort of thing.  Do we want this
to be a todo item?

In the Linuxulator update task, note that Roman is doing this.  Do we need
this todo item anymore?

Add fine-grained locking to syscons task.  Some of the items in this idea
are silly.

Note that the PXE HTTP idea requires someone to implement TCP in pxeboot,
add networking experience on the requirements list.

Add gnn to the TCP/IP regression suite idea, as he's working on such a
thing.

Make it clear in the distributed audit daemon task that portable C coding
is required.
This commit is contained in:
Robert Watson 2007-02-16 11:23:34 +00:00
parent e844cade77
commit 5ba6d7a85b
Notes: svn2git 2020-12-08 03:00:23 +00:00
svn path=/www/; revision=29574

View file

@ -1,6 +1,6 @@
<!DOCTYPE HTML PUBLIC "-//FreeBSD//DTD HTML 4.01 Transitional-Based Extension//EN" [
<!ENTITY base CDATA "../..">
<!ENTITY date "$FreeBSD: www/en/projects/ideas/index.sgml,v 1.80 2007/02/12 20:50:30 joel Exp $">
<!ENTITY date "$FreeBSD: www/en/projects/ideas/index.sgml,v 1.81 2007/02/14 17:12:44 joel Exp $">
<!ENTITY title "FreeBSD list of projects and ideas for volunteers">
<!ENTITY % navinclude.developers "INCLUDE">
<!ENTITY % developers SYSTEM "../../developers.sgml"> %developers;
@ -286,14 +286,12 @@
<p>OpenSolaris&#153; gained support for a new file system called ZFS (Zettabyte
File System) as of build 27a. ZFS is a new approach to file
system design and data management, and includes several interesting features
such as transactional semantics, snapshots and good scalability. Porting ZFS
to FreeBSD is highly encouraged, but should be considered a difficult
task since the current implementation of ZFS is very
Solaris-specific. &a.pjd; has <a
href="http://lists.freebsd.org/pipermail/freebsd-current/2006-August/065306.html">started</a>
such as transactional semantics, snapshots and good scalability. &a.pjd;
has <a
href="http://lists.freebsd.org/pipermail/freebsd-current/2006-August/065306.html">created</a>
a largely functional
<a
href="http://perforce.freebsd.org/depotTreeBrowser.cgi?FSPC=//depot/user/pjd/zfs">porting</a>
ZFS to FreeBSD.</p>
href="http://perforce.freebsd.org/depotTreeBrowser.cgi?FSPC=//depot/user/pjd/zfs">work-in-progress port to FreeBSD</a>.</p>
<p><strong>Requirements</strong>:</p>
<ul>
<li>Strong C knowledge.</li>
@ -402,7 +400,8 @@
<p>DTrace is a dynamic tracing facility designed by Sun Microsystems and
released in Solaris 10. They have since released the major part of
Solaris under the banner of OpenSolaris and the Common Development and
Distribution License (CDDL) 1.0.</p>
Distribution License (CDDL) 1.0. &a.jb; has created an initial port and
should be contacted for information on what tasks remain to be done.</p>
<p><strong>Requirements</strong>:</p>
<ul>
<li>Ability to read and understand foreign C code.</li>
@ -521,7 +520,9 @@
improvement for at least those benchmarks).</p>
<p><a href="mailto:rookie@gufi.org">Attilio Rao</a> has almost <a
href="http://lists.freebsd.org/pipermail/freebsd-hackers/2006-May/016702.html">completed</a>
this work, but he would like to receive more feedback.</p>
this work, but he would like to receive more feedback. In particular, a
thorough performance evaluation is required to determine whether these
improvements are purely theoretical or practical</p>
<p><strong>Requirements</strong>:</p>
<ul>
<li>Ability to read and understand foreign C code.</li>
@ -677,7 +678,8 @@
<p>Process checkpointing allows to migrate some processes to other machines or
to let some processes "survive" a reboot (subject to some constraints).
Interesting files in the DragonFly CVS repository are sys/sys/ckpt.h,
sys/checkpt/* and sys/kern/imgact_elf.c.</p>
sys/checkpt/* and sys/kern/imgact_elf.c; however, this implementation is
both highly experimental and incomplete.</p>
<p><strong>Requirements</strong>:</p>
<ul>
<li>Ability to read and understand foreign C code.</li>
@ -777,10 +779,10 @@
<h2>Syscons modularization</h2>
<p>Separate the syscons code into distinct parts for input, output,
console handling (switching, screen savers etc.) and terminal
emulation. Also implement vt100 and vt220 emulation to supplement
the existing SCO emulation. Add a gettytab(5) capability for
specifying the terminal emulation, and add entries to /etc/gettytab
for the alternative emulations.</p>
emulation. Introduce fine-grained locking. Also implement vt100 and
vt220 emulation to supplement the existing SCO emulation. Add a
gettytab(5) capability for specifying the terminal emulation, and add
entries to /etc/gettytab for the alternative emulations.</p>
<p>Optionally implement xterm emulation. The top line of the screen
should serve as a title bar, displaying the title set with the \e]0;
escape sequence as well as the vty number.</p>
@ -806,6 +808,9 @@
applications continue to work on FreeBSD. More information about the current
status can be obtained from the <a
href="http://wiki.freebsd.org/linux-kernel">wiki page</a>.</p>
<p><a href="http://wiki.freebsd.org/RomanDivacky">Roman Divacky</a> had
a 2006 Summer of Code Project on this topic, and has continued his work
following the end of the summer project.</p>
<p><strong>Requirements</strong>:</p>
<ul>
<li>Ability to read and understand foreign C code.</li>
@ -883,10 +888,13 @@
<p>Implementing HTTP support for pxeboot would allow us to boot a machine using
PXE and pull down a kernel from a web server rather than NFS. This will
allow us to install from DHCPD + Apache or even just DHCPD + a remote web
server.</p>
server. As PXE does not provide an integrated TCP stack, at least a
minimal TCP implementation would need to be present in the FreeBSD PXE
loader.</p>
<p><strong>Requirements</strong>:</p>
<ul>
<li>Good PXE knowledge.</li>
<li>Detailed knowledge of TCP/IP.</li>
</ul>
<hr>
@ -985,9 +993,10 @@
<a name="p-tcpipreg"></a>
<h2>TCP/IP regression test suite</h2>
<p><strong>Technical contact</strong>: <a
href="mailto:rwatson@FreeBSD.org">&a.rwatson;</a></p>
href="mailto:rwatson@FreeBSD.org">&a.rwatson;</a>, <a
href="mailto:gnn@FreeBSD.org">&a.gnn;</a></p>
<p>Design and implement a wire level regression test suite to exercise various
states in the TCP/IP protocol suite. Ideally with IPv4 and IPv6
states in the TCP/IP protocol suite. Ideally with both IPv4 and IPv6
support.</p>
<p><strong>Requirements</strong>:</p>
<ul>
@ -1180,6 +1189,7 @@
clear.</p>
<p><strong>Requirements</strong>:</p>
<ul>
<li>Strong (portable) C programming skills.</li>
<li>Knowledge of the audit subsystem.</li>
</ul>