[haiku-development] Re: GSoc 2010: File system cache or Kernel as userland process
- From: "Axel Dörfler" <axeld@xxxxxxxxxxxxxxxx>
- To: haiku-development@xxxxxxxxxxxxx
- Date: Thu, 25 Mar 2010 14:44:52 +0100
Hi Nitish,
Nitish Sharma <sharmanitishdutt@xxxxxxxxx> wrote:
> I am a Computer science masters student and highly interested in
> making
> contribution to Haiku operating system. I have good experience in C
> and a
> brief stint with C++ and Java. I have experience in GNU/Linux kernel
> development and right now working on Android kernel development.
Welcome, and sorry for the delayed answer!
(I answer here rather than on the kernel list, since it might be
interesting here as well)
> While browsing through Gsoc 2010 Haiku ideas and open tickets, these
> two
> tickets interested me the most: http://dev.haiku-os.org/ticket/1070
> and
> http://dev.haiku-os.org/ticket/1102 . I believe it would be a good
> experience to work on a non-Linux kernel and Haiku is the perfect
> platform
> (& best maintained) for that.
> Please guide me that what would make a good GSoc project: File system
> cache
> or kernel as userland process. Any suggested reading?
There already is a file system cache; the task would only consists of
implementing proper precaching algorithms. I would not consider this a
good GSoC project, though, since it's quite involved, and one should
know a lot about kernel VM speed tradeoffs (ie. not the implementation
itself, but to get it work right).
The kernel as a userland process would be a lot of work, too, but I
think it would be easier to work your way into it, in case you are
interested.
> Moreover, considering my experience, if you think some other project
> is more
> demanding, for the current state of Haiku, and aligned to my skill
> set, then
> please do let me know.
If there is something else that would interest you (it doesn't have to
be part of the bug tracker or ideas page), then just shout.
Bye,
Axel.
Other related posts:
