1 [[!meta title="Google Summer of Code 2010"]]
5 DragonFly BSD is planning to participate (pending acceptance) in the Google Summer of Code program for 2010.
8 Have 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.
11 For more details on Google's Summer of Code: [Google's SoC page](http://socghop.appspot.com/)
13 Note to prospective students: These project proposals are meant to be a first approximation; we're looking forward to your own suggestions (even for completely new directions) and will try to integrate your ideas to make the GSoC project more interesting to all parties. Even when a proposal is very specific about the goals that must be achieved and the path that should be taken, these are always negotiable. Keep in mind that we have tried to limit our proposals to those that (based on our past experience) are appropriate for the GSoC program.
17 * Prerequisites: knowledge that the student should have before starting the project. It may be possible to acquire the knowledge in the course of the project, but the estimated difficulty would increase substantially. On the bright side, you can expect to have a much deeper understanding of these fields (and gain some real-world experience) after you successfully complete the respective project.
18 * Difficulty: Estimated difficulty of the project, taking into account the complexity of the task and the time constraints of the GSoC program.
19 * Contact point: The person you should contact for any further information or clarifications.
23 ##### VFS Quota System
24 * Create a new kernel subsystem to manage quota's in a filesystem agnostic manner by interfacing with the kernel VFS layer.
25 * 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), ...).
26 * 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.
30 * Prerequisites: C, introductory filesystems internals
31 * Difficulty: Moderate
32 * Contact point: Samuel J. Greear <sjg@thesjg.com>
36 ##### HAMMER Data dedup
38 The HAMMER filesystem is very efficient in sharing data between its fine-grained
39 snapshots, but when you copy (or otherwise duplicate) a file or directory tree, the data
40 is no longer shared. This is suboptimal because then we make poor use of disk space
41 and the same data gets cached multiple times wasting precious RAM space.
43 The goal of this project is to add a data de-duplication mechanism to the HAMMER
44 filesystem. A reasonable approach would be to detect potential data matches using
45 CRCs during pruning runs. Then you could verify there is actual duplication of data
46 (i.e. the match is not a false positive), collapse the B-Tree reference and account for the
47 additional reference in the allocator.
51 * Prerequisites: C, modern filesystem internals
52 * Difficulty: Moderate
53 * Contact point: dillon
57 ##### Implement i386 32-bit ABI for x86_64 64-bit kernel
58 * Add a 32-bit syscall table which translates 32-bit
59 system calls to 64-bit
61 * Add support for 32 bit compatibility mode operation
62 and ELF binary detection.
67 * Difficulty: Moderate
68 * Contact point: dillon
72 ##### Implement ARC algorithm for vnode free list
73 * Vnode recycling is LRU and can't efficiently handle data sets which
74 exceed the maxvnode limit.
78 * Prerequisites: C, OS internals
80 * Contact point: dillon
84 ##### Implement swapoff
85 * We have swapon to add swap space, we need a swapoff to
90 * Prerequisites: C, elementary OS memory management
92 * Contact point: dillon
96 ##### Graphics Kernel Memory Manager Support ( GEM )
97 * Support dealing with graphics NUMA in kernel space for modern graphics hardware
98 * http://en.wikipedia.org/wiki/Graphics_Execution_Manager
102 * Prerequisites: C, knowledge of modern computer graphics system architecture
103 * Difficulty: Moderate
104 * Contact point: kernel@crater.dragonflybsd.org
108 ##### Make DragonFly NUMA-aware
110 * Parse related ACPI tables
111 * NUMA-aware memory allocation
113 [ACPI SLIT parser](http://mail-index.netbsd.org/tech-kern/2009/11/23/msg006518.html)
114 [ACPI SRAT parser](http://mail-index.netbsd.org/tech-kern/2009/11/23/msg006517.html)
115 [NetBSD NUMA diff](http://www.netbsd.org/~cegger/numa2.diff)
116 [NetBSD NUMA x86 diff](http://www.netbsd.org/~cegger/numa_x86.diff)
120 * Prerequisites: C, introductory computer architecture
122 * Contact point: kernel@crater.dragonflybsd.org
126 ##### Volume Management based on NetBSD's port of LVM2
128 NetBSD reimplemented Linux's device mapper (currently only implementing
129 the linear, zero and error targets; Linux has support for a variety of
130 targets, including crypt, stripe, snap, multipath) as dm(4). Device mapper
131 provides the functionality on which to implement volume management; NetBSD
132 has imported LVM2 (which is GPL), but it is possible to create different
133 tools for volume management (e.g. IBM's EVMS was also built on top of device
136 The goal of this project is to port both the kernel code, dm(4), and the
137 LVM2 userspace libraries and tools from NetBSD. If time remains, the
138 student should also implement a proof of concept "stripe" target or, for the
139 more ambitious, a "crypt" target.
141 A possible roadmap for this project would be
143 1. Port the dm(4) code
145 This code uses proplib instead of binary ioctls for communicating with
146 userspace. Either port proplib, or convert the code to use ioctls.
148 1. Port the userspace tools
150 Integrate the tools in our source tree using a separate vendor branch, as
151 is normally done for contrib software (see development(7)). Make any
152 DragonFlyBSD-specific changes necessary.
154 1. (Optional) Implement either a "stripe" target or a crypt target.
156 The stripe target must be designed with robustness and extensibility in
157 mind, though it is not required to go all the way. It should be flexible
158 enough to allow for different RAID level implementations (at least 0, 1
159 and 5). Additionally, it should be possible to keep an internal (i.e. part
160 of the volume) log to speed up resyncing and parity checking. Implementing
161 those features would be ideal, but is not required.
163 The crypt target must allow for different ciphers and cipher parameters and
164 should make use of our in-kernel crypto infrastructure. It is probably
165 necessary to do the encryption asynchronously which will require extending
166 the current infrastructure.
170 * Prerequisites: C, elementary OS internals
172 * Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>
176 ##### Port DragonFly to Xen platform
180 * Prerequisites: C, x86 assembly
182 * Contact point: kernel@crater.dragonflybsd.org
186 ##### Port valgrind to DragonFlyBSD
188 Valgrind 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.
190 The 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.
194 * Prerequisites: C, x86 assembly, low-level OS internals
196 * Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>
200 ##### Adapt pkgsrc to create a package system with dependency independence.
201 * 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.
207 * Contact point: kernel@crater.dragonflybsd.org
211 ##### Implement virtio drivers on DragonFly to speed up DragonFly as a KVM guest
213 As virtualization is coming more and more and KVM will be a strong player in that field,
214 we want DragonFly to have top-notch support for this virtualization platform. For this
215 purpose, we'd like to have a virtio-based implementation of a paravirtualized disk and
216 network driver. [virtio](http://www.ibm.com/developerworks/linux/library/l-virtio/index.html)
217 is an abstraction to a ring buffer that is shared between the host and the guest. On top of this
218 abstraction, one can build a variety of paravirtualized devices, as specified in
219 [virtio-spec](http://ozlabs.org/~rusty/virtio-spec/virtio-spec-0.8.6.pdf).
221 The goal of this project is to create a virtio-ring implementation and then to implement drivers
222 for the network and block devices described in the specification linked to above. This is a great
223 project for a student who wants to get experience writing (real-world, high-performance) device
224 drivers without having to deal with the quirks of real hardware.
228 * Prerequisites: C, elementary OS internals
230 * Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>, kernel@crater.dragonflybsd.org
234 ##### Port FUSE or PUFFS from FreeBSD/NetBSD
236 * http://www.netbsd.org/docs/puffs/
237 * This would make many userspace filesystems available to DragonFly, e.g. sshfs to mention only one.
241 * Prerequisites: C, elementary OS internals
243 * Contact point: Michael Neumann <mneumann@ntecs.de>
247 ##### Make vkernels checkpointable
250 * Teach the checkpt syscall how to checkpoint multiple vmspaces.
251 * Add code to the vkernel which gets triggered on SIGCKPT to dump/load e.g. the current state of network drivers.
252 * This would allow us to save and restore or even migrate a complete DragonFly operating system running on the vkernel platform.
253 This could be especially handy on laptops (if we'd get X11 operating in vkernels).
254 * See also: http://www.dragonflybsd.org/docs/developer/CheckpointFeatures/
258 * Prerequisites: C, OS internals
260 * Contact point: Michael Neumann <mneumann@ntecs.de>
264 ##### HAMMER compression
266 * Compress blocks as they get written to disk.
267 * Only file data (rec_type == DATA) should be compressed, not meta-data.
268 * the CRC should be that of the uncompressed data.
269 * ideally you'd need to associate the uncompressed data with the buffer cache buffer somehow, so that decompression is only performed once.
270 * compression could be turned on a per-file or per-pfs basis.
271 * gzip compression would be just fine at first.
275 * Prerequisites: C, filesystem internals
277 * Contact point: Michael Neumann <mneumann@ntecs.de>
283 * Port the whole usb4bsd stuff to DragonFly, as our own usb stack is too outdated.
285 * The usb4bsd branch of hselasky (?) has several userland wrappers and quite good abstraction to simplify the porting.
287 * (is polachok doing this or not?)
291 * Prerequisites: C, OS internals
292 * Difficulty: Moderate
293 * Contact point: kernel@crater.dragonflybsd.org
297 ##### Userland System V Shared Memory / Semaphore / Message Queue implementation
298 * Implement some or all of these subsystems in their entirety, or as completely as possible in userland using a daemon, mmap and the DragonFly umtx_sleep(2)/umtx_wakeup(2) or other userland facilities.
299 * Any security or other major hurdles to this approach that would likely have to be implemented in-kernel should be noted in the students application.
300 * Test and benchmark the new facilities with heavy SysV consumers such as PostgreSQL
301 * Identify performance tradeoffs made in the userland implementation versus the existing kernel implementation. If time permits identify and apply solutions to these tradeoffs so that the userland implementation performs on par with or better than the kernel implementation.
305 * Prerequisites: C, x86 assembly
306 * Difficulty: Moderate
307 * Contact point: Samuel J. Greear <sjg@thesjg.com>
312 ##### Update our interrupt routing and PCI code
313 * Update our interrupt routing to MSI (Message Signalled Interrupts)
314 * Update the PCI code to take advantage of MSI and also of newer power-saving features (see FreeBSD's work in this area)
318 * Prerequisites: C, OS internals
320 * Contact point: kernel@crater.dragonflybsd.org
324 ##### Proportional RSS
326 The Resident Stack Size displayed by top keeps track of the number of resident pages in
327 a certain process's adress space. It is very useful to locate memory hogs, but doesn't take
328 into account page sharing. For example, if N processes map library L and L's resident pages
329 are 1G, this 1G is added to the RSS of all N processes. A more useful number would be the
330 Proportional (or Effective) RSS, for which we divide the number of mapped shared pages by
331 the number of processes sharing each page. So in the previous example we would add 1GB/N
332 to each process that has L mapped.
334 The goal of this project is to hack the kernel to allow for effective calculation of the
335 Proportional RSS and modify top to use it in addition to the RSS (i.e. it should display it by
336 default and be able to sort based on it).
340 * Prerequisites: C, Elementary OS internals
342 * Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>