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/)
16 ##### VFS Quota System
17 1. Kernel and quota support in the VFS layer
18 1. Filesystem-agnostic quota support tools for userland
27 ##### HAMMER Data dedup
28 * Add a data de-duplication mechanism to HAMMER.
30 * Potential data matches using CRCs during pruning runs,
31 verify duplicate data, collapse the B-Tree reference,
32 and account for the additional ref in the allocator.
36 * Difficulty: Moderate
37 * Contact point: dillon
41 ##### Implement i386 32-bit ABI for x86_64 64-bit kernel
42 * Add a 32-bit syscall table which translates 32-bit
43 system calls to 64-bit
45 * Add support for 32 bit compatibility mode operation
46 and ELF binary detection.
50 * Difficulty: Moderate
51 * Contact point: dillon
55 ##### Implement ARC algorithm for vnode free list
56 * Vnode recycling is LRU and can't efficiently handle data sets which
57 exceed the maxvnode limit.
62 * Contact point: dillon
66 ##### Implement swapoff
67 * We have swapon to add swap space, we need a swapoff to
73 * Contact point: dillon
77 ##### Implement SHA-2 password encryption
78 * Implement new default encryption for master.passwd
82 * Difficulty: Fairly easy
83 * Contact point: dillon
87 ##### Ultra Fast Boot & Shutdown Speed
88 * Be competitive with GNU/Linux, OSX & MS Windows refinements in this area.
97 ##### Graphics Kernel Memory Manager Support ( GEM )
98 * Support dealing with graphics NUMA in kernel space for modern graphics hardware
99 * http://en.wikipedia.org/wiki/Graphics_Execution_Manager
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)
125 ##### Security/Hardening improvements
126 * Encrypted swap/filesystems (From NetBSD or OpenBSD?)
127 * Extended toolchain hardening
128 * NX/XD support in kernel (at least for 64 bit kernels, DF doesn’t support PAE IIRC)
129 * More use of randomization (for example in PIDs)
130 * Port OpenBSD’s most recent malloc implementation as an option
139 ##### Data Integrity Framework Implementation
140 * Something akin to what was done for Linux: http://oss.oracle.com/projects/data-integrity/
149 ##### Volume Management based on NetBSD's port of LVM2
151 NetBSD reimplemented Linux's device mapper (currently only implementing
152 the linear, zero and error targets; Linux has support for a variety of
153 targets, including crypt, stripe, snap, multipath) as dm(4). Device mapper
154 provides the functionality on which to implement volume management; NetBSD
155 has imported LVM2 (which is GPL), but it is possible to create different
156 tools for volume management (e.g. IBM's EVMS was also built on top of device
159 The goal of this project is to port both the kernel code, dm(4), and the
160 LVM2 userspace libraries and tools from NetBSD. If time remains, the
161 student should also implement a proof of concept "stripe" target or, for the
162 more ambitious, a "crypt" target.
164 A possible roadmap for this project would be
166 1. Port the dm(4) code
168 This code uses proplib instead of binary ioctls for communicating with
169 userspace. Either port proplib, or convert the code to use ioctls.
171 1. Port the userspace tools
173 Integrate the tools in our source tree using a separate vendor branch, as
174 is normally done for contrib software (see development(7)). Make any
175 DragonFlyBSD-specific changes necessary.
177 1. (Optional) Implement either a "stripe" target or a crypt target.
179 The stripe target must be designed with robustness and extensibility in
180 mind, though it is not required to go all the way. It should be flexible
181 enough to allow for different RAID level implementations (at least 0, 1
182 and 5). Additionally, it should be possible to keep an internal (i.e. part
183 of the volume) log to speed up resyncing and parity checking. Implementing
184 those features would be ideal, but is not required.
186 The crypt target must allow for different ciphers and cipher parameters and
187 should make use of our in-kernel crypto infrastructure. It is probably
188 necessary to do the encryption asynchronously which will require extending
189 the current infrastructure.
194 * Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>
198 ##### Make DragonflyBSD Tickless
199 * 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.
200 This 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.
201 * http://hub.opensolaris.org/bin/view/Project+tickless/lbolt
210 ##### Make the DragonflyBSD Dispatcher Power-aware
211 * 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.
212 * 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.
213 * http://hub.opensolaris.org/bin/view/Project+tesla/CPUPM
222 ##### Port DragonFly to ARM platform
226 * Difficulty: Extreme hard
231 ##### Port valgrind to DragonFlyBSD
233 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.
235 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.
240 * Contact point: Aggelos Economopoulos <aoiko@cc.ece.ntua.gr>
244 ##### Adapt pkgsrc to create a package system with dependency independence.
245 * 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.
254 ##### Implement virtio drivers on DragonFly to speed up DragonFly as a KVM guest
255 * 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.
264 ##### Port FUSE or PUFFS from FreeBSD/NetBSD
266 * http://www.netbsd.org/docs/puffs/
267 * This would make many userspace filesystems available to DragonFly, e.g. sshfs to mention only one.
272 * Contact point: Michael Neumann <mneumann@ntecs.de>
276 ##### Make vkernels checkpointable
279 * Teach the checkpt syscall how to checkpoint multiple vmspaces.
280 * Add code to the vkernel which gets triggered on SIGCKPT to dump/load e.g. the current state of network drivers.
281 * This would allow us to save and restore or even migrate a complete DragonFly operating system running on the vkernel platform.
282 This could be especially handy on laptops (if we'd get X11 operating in vkernels).
283 * See also: http://www.dragonflybsd.org/docs/developer/CheckpointFeatures/
288 * Contact point: Michael Neumann <mneumann@ntecs.de>
292 ##### HAMMER compression
294 * Compress blocks as they get written to disk.
295 * Only file data (rec_type == DATA) should be compressed, not meta-data.
296 * the CRC should be that of the uncompressed data.
297 * ideally you'd need to associate the uncompressed data with the buffer cache buffer somehow, so that decompression is only performed once.
298 * compression could be turned on a per-file or per-pfs basis.
299 * gzip compression would be just fine at first.
304 * Contact point: Michael Neumann <mneumann@ntecs.de>