Re: [特别重要] CEWG Open Project 参与指南
aged...@gmail.com
Osier
Wu Zhangjin <wuzha...@gmail.com>编写：
Q. How is this different from other "open project" systems, like Google's "Summer of Code"?
A. Other systems often require that the submitter of the proposal be the one volunteering to do the work. With CE WG's Open Project Proposal, anyone who can think of a good idea can submit it, (possibly with hinting about someone who might be a good candidate to perform the work).
Of course you can also submit something you are interested in working on yourself, in the hopes that the CE Workgroup will fund the work.
Q. How much money is the CE WG willing to spend on these projects?
A. The exact amount shall remain a mystery. Historically our budget has been around $100,000, and we usually focus on projects in the $10K to $20k range. However, we have exceptions both above and below this amount. The exact amount of funding will depend a great deal on the projects that are proposed.
Q. Are you daft? Why would you let other people suggest ways to spend the CE WG's money?
A. Hey, it's open source. Good ideas can come from anywhere. Note that the CE WG still controls what projects get selected, and hence where the money will be spent. That's one of the privileges of CEWG membership. However, CEWG will publish the list of projects that it will fund, as well as the projects that were not funded. It is hoped that other entities may contribute money or resources to push some of the technologies or features forward, independently or in conjunction with CEWG's contribution.
Q. Where will the proposals be published?
A. Right here on the eLinux wiki. See Project Proposals for 2013
Q. How will CEWG decide on contractors for these projects?
A. After collecting the proposals, CEWG will solicit bids for contractors for these proposals. In some cases, CEWG may proactively contact individuals or groups they think would be good candidates for doing certain work (such as previous CEWG contractors, or people already working on certain projects, who are available.
After talking to various contractor candidates, the CEWG Architecture Group will select the contractors for the projects, contracts will be drawn up, and the work started.
OK，如果看完了基本的内容其实你应该已经知道了。从这个项目中我们可以拿到什么，下面我举例2011年我跟Shydox申请的这个项目为例。
--------------------------------------------------------------------------------------
* 项目提案(Proposal)：http://elinux.org/Work_on_Tiny_Linux_Kernel
1. Title
Work on Tiny Linux Kernel
2. Description
The main work I plan to perform:
* Improve Section garbage collection patches
** Ensure the other features(e.g. Ftrace, Gcov, Perf...) survive with it
** Enhance it for specific architectures (e.g. add -ffunction-sections and
-fdata-sections for specific arch; move the functions and data in assembly
to their own sections...)
* Forward-porting Linux-tiny to latest kernels
* Make kernel more configurable
** ptrace system call
** alram, getitimer, setitimer system call
** "kexec on crash"
** random number generators: /dev/{random, urandom}
** make duplicated features configurable: e.g. emulated software FPU and
hardware FPU support
** kernel and module parameters support
** remove sysfs support of specific drivers for !CONFIG_SYSFS
** convert big static array to configurable variable: e.g. modedb,
ata_device_blacklist
** NR_IRQS, COMMAND_LINE_SIZE, MAX_NR_CON_DRIVER, MAX_NR_CONSOLES ...
3. Schedule
* Improve Section garbage collection patches (4 weeks)
This mainly need to upstream the left patches(at least 4), add arch specific
support (X86, MIPS, ARM and PowerPC first) and add compatibility for the
other kernel features(Ftrace, Gcov, Perf first). Then, at least 10 sub items
should be covered.
On average, consider 8 hours per item, 2 weeks(8 hours per day, 5 days per
week) for primary developement and testing, 2 weeks for upstream may be
required.
* Forward-porting Linux-tiny to latest kernels (4 weeks)
Need to port over than 41 patches from 2.6.32 (have already ported most of
them to 2.6.36) to the latest 2.6.40? If 2 hour per patch, then, totally 80
hours or 2 weeks. Most of them are not suitable for upstream, another 2 weeks
may be required for the upstream candidates(around 10 patches or one quarter).
Since most of them are not suitable for upstream, they will be maintained in
one public source code repository, currently, they are maintained here [1],
later, they will be placed at [2].
* Make kernel more configurable (8 weeks)
This part currently include about 10 items, each one may need about 2 days of
developement and testing, another 2 days for upstream (If applicable),
totally, 40 days or 8 weeks.
4. Indication for qualifications
All of the above proposed items should be available at [2] under GPL v2,
including their source code, commit log and related testing result. All of
them should pass build and boot test on at least one board of these 4 archs:
X86, ARM, MIPS and PowerPC.
Most of the section garbage collection patches, more kernel configuration
patches and some of the Linux-tiny patches should be pushed into mainline (If
applicable), the others should be maintained at [2].
The project at [2] will be maintained all the time, the other ideas mentioned
or not mentioned in the proposal [3] will be implemented/added there. A
corresponding mainling list (the form of such a community) will be created
for this project for persistent maintaining in the future.
5. Cost
Consider 10$ per hour, total is 10$* (16 weeks * 5 days * 8 hours) = 6400$.
6. Contact
Email+Gtalk+MSN: wuzha...@gmail.com
7. Links
[1] http://dev.lemote.com/cgit/linux-loongson-community.git/log/?h=tiny36
[2] http://code.google.com/p/tinylinux/
[3] http://elinux.org/Work_on_Tiny_Linux_Kernel
WEN Pingbo
BTW, 在我离校之前，还能榆中本部来回跑，可以带着榆中的成员一起干，顺便来个现场监督:)
> 您收到此邮件是因为您订阅了 Google 网上论坛的“oss team”论坛。
> 要退订此论坛并停止接收此论坛的电子邮件，请发送电子邮件到 ossteam+u...@googlegroups.com。
> 要向此网上论坛发帖，请发送电子邮件至 oss...@googlegroups.com。
> 通过以下网址访问此论坛：http://groups.google.com/group/ossteam。
> 要查看更多选项，请访问 https://groups.google.com/groups/opt_out。