(no commit message)
[ikiwiki.git] / docs / developer / gsoc2010 / index.mdwn
CommitLineData
619785d7 1[[!meta title="Google Summer of Code 2010"]]
2
fd9153a6 3[[!toc levels=0]]
619785d7 4
6decb607 5DragonFly BSD is planning to participate (pending acceptance) in the Google Summer of Code program for 2010.
619785d7 6
7
eec4180d 8Have a look at our SoC pages from [[2008|docs/developer/GoogleSoC2008/]] and [[2009|docs/developer/gsoc2009]] to get an overview about prior year's projects. The [Projects Page](/docs/developer/ProjectsPage/) is also a potential source of ideas.
619785d7 9
10
eec4180d 11For more details on Google's Summer of Code: [Google's SoC page](http://socghop.appspot.com/)
619785d7 12
619785d7 13
7b6108eb 14#### Project ideas
619785d7 15
7b6108eb 16##### VFS Quota System
c43da85e 17* Create a new kernel subsystem to manage quota's in a filesystem agnostic manner by interfacing with the kernel VFS layer.
18* Create filesystem-agnostic quota support tools for userland that obtain information in the same manner as eg: du(1) instead of parsing the filesystem internals directly as the existing quota tools do (see quotacheck(8), repquota(8), edquota(8), ...).
19* The quota file storage can be modeled after the existing UFS code that does the same, but should use the more general bytes, files and/or directories metrics instead of the somewhat UFS-specific blocks and inodes.
d87e1458 20
7b6108eb 21Meta information:
22
c43da85e 23* Difficulty: Moderate
24* Contact point: Samuel J. Greear <sjg@thesjg.com>
7b6108eb 25
26---
27
5d1a98e7 28##### HAMMER Data dedup
29* Add a data de-duplication mechanism to HAMMER.
30
31* Potential data matches using CRCs during pruning runs,
32 verify duplicate data, collapse the B-Tree reference,
33 and account for the additional ref in the allocator.
34
35Meta information:
36
37* Difficulty: Moderate
38* Contact point: dillon
39
40---
41
42##### Implement i386 32-bit ABI for x86_64 64-bit kernel
43* Add a 32-bit syscall table which translates 32-bit
44 system calls to 64-bit
45
46* Add support for 32 bit compatibility mode operation
47 and ELF binary detection.
48
49Meta information:
50
51* Difficulty: Moderate
52* Contact point: dillon
53
54---
55
56##### Implement ARC algorithm for vnode free list
57* Vnode recycling is LRU and can't efficiently handle data sets which
58 exceed the maxvnode limit.
59
60Meta information:
61
62* Difficulty: Modest
63* Contact point: dillon
64
65---
66
67##### Implement swapoff
68* We have swapon to add swap space, we need a swapoff to
69 remove it.
70
71Meta information:
72
73* Difficulty: Modest
74* Contact point: dillon
75
76---
77
78##### Implement SHA-2 password encryption
79* Implement new default encryption for master.passwd
80
81Meta information:
82
83* Difficulty: Fairly easy
84* Contact point: dillon
85
86---
87
7b6108eb 88##### Graphics Kernel Memory Manager Support ( GEM )
95aebbf4 89* Support dealing with graphics NUMA in kernel space for modern graphics hardware
90* http://en.wikipedia.org/wiki/Graphics_Execution_Manager
91
7b6108eb 92Meta information:
93
94* Difficulty: ?
95* Contact point: ?
96
97---
98
99##### Make DragonFly NUMA-aware
f2c1c563 100
101* Parse related ACPI tables
102* NUMA-aware memory allocation
7fdbf5f2
C
103* References:
104[ACPI SLIT parser](http://mail-index.netbsd.org/tech-kern/2009/11/23/msg006518.html)
105[ACPI SRAT parser](http://mail-index.netbsd.org/tech-kern/2009/11/23/msg006517.html)
106[NetBSD NUMA diff](http://www.netbsd.org/~cegger/numa2.diff)
107[NetBSD NUMA x86 diff](http://www.netbsd.org/~cegger/numa_x86.diff)
f2c1c563 108
7b6108eb 109Meta information:
110
111* Difficulty: ?
112* Contact point: ?
113
114---
115
7b6108eb 116##### Volume Management based on NetBSD's port of LVM2
117
118NetBSD reimplemented Linux's device mapper (currently only implementing
8611186f
A
119the linear, zero and error targets; Linux has support for a variety of
120targets, including crypt, stripe, snap, multipath) as dm(4). Device mapper
121provides the functionality on which to implement volume management; NetBSD
122has imported LVM2 (which is GPL), but it is possible to create different
123tools for volume management (e.g. IBM's EVMS was also built on top of device
124mapper).
125
7b6108eb 126The goal of this project is to port both the kernel code, dm(4), and the
8611186f
A
127LVM2 userspace libraries and tools from NetBSD. If time remains, the
128student should also implement a proof of concept "stripe" target or, for the
129more ambitious, a "crypt" target.
130
7b6108eb 131A possible roadmap for this project would be
8611186f 132
7b6108eb 1331. Port the dm(4) code
8611186f 134
7b6108eb 135 This code uses proplib instead of binary ioctls for communicating with
8611186f
A
136userspace. Either port proplib, or convert the code to use ioctls.
137
7b6108eb 1381. Port the userspace tools
8611186f 139
7b6108eb 140 Integrate the tools in our source tree using a separate vendor branch, as
8611186f
A
141is normally done for contrib software (see development(7)). Make any
142DragonFlyBSD-specific changes necessary.
143
7b6108eb 1441. (Optional) Implement either a "stripe" target or a crypt target.
8611186f 145
7b6108eb 146 The stripe target must be designed with robustness and extensibility in
8611186f
A
147mind, though it is not required to go all the way. It should be flexible
148enough to allow for different RAID level implementations (at least 0, 1
149and 5). Additionally, it should be possible to keep an internal (i.e. part
150of the volume) log to speed up resyncing and parity checking. Implementing
151those features would be ideal, but is not required.
152
7b6108eb 153 The crypt target must allow for different ciphers and cipher parameters and
8611186f
A
154should make use of our in-kernel crypto infrastructure. It is probably
155necessary to do the encryption asynchronously which will require extending
156the current infrastructure.
157
7b6108eb 158Meta information:
159
160* Difficulty: Medium
161* Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>
162
163---
164
165##### Make DragonflyBSD Tickless
661bad5e 166* By default, the clock cyclic fires at 100 Hz, regardless of whether or not any timeouts/callouts are scheduled to fire/expire. This is suboptimal from a power efficiency standpoint, as at least one of the system's CPUs never become quiescent/idle enough to be brought into a low power state.
167This work involves re-implementing the services presently provided by clock() in a tickless (or event based) fashion, eliminating the need for the system to "wake up", only to realize that there's nothing to do on an otherwise idle system.
168* http://hub.opensolaris.org/bin/view/Project+tickless/lbolt
169
7b6108eb 170Meta information:
171
172* Difficulty: ?
173* Contact point: ?
174
175---
176
177##### Make the DragonflyBSD Dispatcher Power-aware
6eefd8d1 178* CPU power management as it it implemented today is relatively isolated from the rest of the system. As such, it is forced to periodically poll to measure the utilization of the system's CPU resources.
179* This project extends the kernel's existing topology aware scheduling facility to bring "power domain" awareness to the dispatcher. With this awareness in place, the dispatcher can implement coalescence dispatching policy to consolidate utilization onto a smaller subset of CPU domains, freeing up other domains to be power managed. In addition to being domain aware, the dispatcher will also tend to prefer to utilize domains already running at higher power/performance states...this will increase the duration and extent to which domains can remain quiescent, improving the kernel's ability to take advantage of features like deep C-states. Because the dispatcher will track power domain utilization along the way, it can drive active domain state changes in an event driven fashion, eliminating the need for the CPUPM subsystem to poll.
180* http://hub.opensolaris.org/bin/view/Project+tesla/CPUPM
9ce6d844 181
7b6108eb 182Meta information:
183
184* Difficulty: ?
185* Contact point: ?
186
187---
6e6a7613 188
7b6108eb 189##### Port DragonFly to ARM platform
ddd662fb 190
7b6108eb 191Meta information:
908e52ab 192
93720600 193* Difficulty: Extreme hard
7b6108eb 194* Contact point: ?
908e52ab 195
7b6108eb 196---
6e6a7613 197
7b6108eb 198##### Port valgrind to DragonFlyBSD
199
200Valgrind is a very useful tool on a system like DragonFly that's under heavy development. Since valgrind is very target specific, a student doing the port will have to get acquainted with many low level details of the system libraries and the user<->kernel interface (system calls, signal delivery, threading...). This is a project that should appeal to aspiring systems programmers. Ideally, we would want the port to be usable with vkernel processes, thus enabling complex checking of the core kernel code.
201
202The goal of this project is to port valgrind to the DragonFlyBSD platform so that at least the memcheck tool runs sufficiently well to be useful. This is in itself a challenging task. If time remains, the student should try to get at least a trivial valgrind tool to work on a vkernel process.
203
204Meta information:
205
206* Difficulty: Hard
207* Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>
208
209---
210
211##### Adapt pkgsrc to create a package system with dependency independence.
f4e9f977 212* Create a set of tools that modifies how the pkgsrc packages are installed, allowing for the ability to upgrade individual packages, without stopping applications that depend on said packages from working. One method of achieving this is detailed at http://www.dragonflybsd.org/goals/#packages but other methods may be possible. PC-BSD have written a tool called PBI Builder which modifies FreeBSD ports for their dependency independence PBI system, this could be used as a starting point for the DragonFly BSD tools.
3f995320 213
7b6108eb 214Meta information:
215
216* Difficulty: ?
217* Contact point: ?
218
219---
220
221##### Implement virtio drivers on DragonFly to speed up DragonFly as a KVM guest
d184f2ee 222* As virtualization is coming more and more and KVM will be a strong player in that field, it might be a good idea to be the first BSD to have a virtio implementation that enables us to run at a better speed in comparison to the other BSDs and maybe close to Linux on this virtualization platform.
223
7b6108eb 224Meta information:
225
226* Difficulty: ?
227* Contact point: ?
228
229---
230
231##### Port FUSE or PUFFS from FreeBSD/NetBSD
9953fcc7 232
233* http://www.netbsd.org/docs/puffs/
234* This would make many userspace filesystems available to DragonFly, e.g. sshfs to mention only one.
235
7b6108eb 236Meta information:
237
93720600 238* Difficulty: Medium
239* Contact point: Michael Neumann <mneumann@ntecs.de>
7b6108eb 240
241---
242
243##### Make vkernels checkpointable
0358d02a 244
245* See checkpt(1).
246* Teach the checkpt syscall how to checkpoint multiple vmspaces.
247* Add code to the vkernel which gets triggered on SIGCKPT to dump/load e.g. the current state of network drivers.
248* This would allow us to save and restore or even migrate a complete DragonFly operating system running on the vkernel platform.
249This could be especially handy on laptops (if we'd get X11 operating in vkernels).
250* See also: http://www.dragonflybsd.org/docs/developer/CheckpointFeatures/
d184f2ee 251
7b6108eb 252Meta information:
253
93720600 254* Difficulty: Medium
255* Contact point: Michael Neumann <mneumann@ntecs.de>
7b6108eb 256
257---
258
259##### HAMMER compression
16e3abac 260
261* Compress blocks as they get written to disk.
262* Only file data (rec_type == DATA) should be compressed, not meta-data.
263* the CRC should be that of the uncompressed data.
264* ideally you'd need to associate the uncompressed data with the buffer cache buffer somehow, so that decompression is only performed once.
265* compression could be turned on a per-file or per-pfs basis.
93720600 266* gzip compression would be just fine at first.
d184f2ee 267
7b6108eb 268Meta information:
269
93720600 270* Difficulty: Medium
271* Contact point: Michael Neumann <mneumann@ntecs.de>
7b6108eb 272
273---
274
88d3450e 275
276##### Port usb4bsd
277* Port the whole usb4bsd stuff to DragonFly, as our own usb stack is too outdated.
278
279* The usb4bsd branch of hselasky (?) has several userland wrappers and quite good abstraction to simplify the porting.
280
281* (is polachok doing this or not?)
282
283Meta information:
284
285* Difficulty: Moderate
286* Contact point: -
287
288---
289
290
9ce6d844 291 (please add)