harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Xiao-Feng Li" <xiaofeng...@gmail.com>
Subject Re: Google Summer Of Code 2008, harmony-gc-1
Date Sat, 12 Apr 2008 01:08:29 GMT
Yes, agree with Alexei.

Yuri, please all put your message copy to GSoC your application
comment, so that other reviewers can get it as well.

GSoC is intended for good students for open source projects, rather
than good projects. The application proposal you write probably could
better reflect your skills, understandings, and passion on the
project.  :)


On Sat, Apr 12, 2008 at 8:26 AM, Alexei Fedotov
<alexei.fedotov@gmail.com> wrote:
> Hello Yuri,
>  Welcome to the dev@ list! I hope you will find our project interesting
>  regardless of GSoC. :-)
>  I was not sure I caught your description of milestones: they should
>  include some measurable indicators which would help your mentors to
>  decide if you successfully completed a part of your commitment. For
>  example, "I will complete GC visualization tool on 11 of July. The
>  tool will include the following features: a heap fragmentation
>  diagram, object size and lifetime histograms". I do not address here
>  if another GC visualization tool is needed: please, check
>  harmony-gc-3.
>  Thanks.
>  On Sat, Apr 12, 2008 at 2:36 AM, oberon <oberon.ua@gmail.com> wrote:
>  > Hello,
>  >
>  >   I have applied for Compressor Garbage Collector (harmony-gc-1). Well, I
>  >  have misunderstood what does abstract in google application means so I have
>  >  just copied the paper abstract. I have also promised to send you updates,
>  >  but I am sending those only now because of quite unpleasant incident in my
>  >  university :(.
>  >
>  >   First of all I will say that in previous semester I have had a course
>  >  about architecture of a PC. Well, I do not know how to translate it more
>  >  accurately, but it is about low level programming. Particularly we have
>  >  studied CPU interrupts, basic graphics system, architecture of file systems,
>  >  FPU, and other. Well, this was quite old-school course since we have not
>  >  studied anything advanced. For example we studied only CGA/EGA/VGA, FAT16
>  >  file system only (only partially FAT32, and absolutely no NTFS, ext2/3, or
>  >  anything more modern) and things like that. As my course work I had written
>  >  small DOS game which caught 9th interrupt to handle user input. To get high
>  >  credit for the course we have had to present report on some topic that
>  >  relate to the course but was not covered in it. I have picked a topic about
>  >  .NET garbage collections. I've googled a bit, found few articles on that and
>  >  made a report :). Well, this was my first acquaintance with internal
>  >  structure with garbage collectors. Very recently I have found this project
>  >  and have written short application for it. I am very excited that I have
>  >  found opportunity to write real-world garbage collector that could be
>  >  useful, and I really hope that you will rank me high enough to get this
>  >  position.
>  >
>  >   My plans are quite simple. First of all I will write very basic test
>  >  environment where all virtual memory services will be mocked-up. This will
>  >  include some visualization program so that I could ensure that I have
>  >  understood algo correctly and see how algorithm actually works. Secondly I
>  >  would study Harmony itself a bit more deeply. The interesting part of work
>  >  is about correctly using virtual memory mappings, traps and protections.
>  >  There is a reference to an article about virtual memory primitives for user
>  >  programs, which I will read. In parallel I will study how GC are currently
>  >  implemented in Harmony. I am not very sure if you expect me to short-tell
>  >  algorithm to check if I have understood it appropriately. Please say if so,
>  >  I will do my best. If you need other details clarified I will gladly present
>  >  them here.
>  >
>  >   As for milestones. I have spoken with our head of department and he said
>  >  that it is possible to use this work as course work. So if you will accept
>  >  me I will start really soon, since it has to be presented in mid-summer
>  >  (1st-16th of July). I will have few ACM competitions which would require my
>  >  attendance, but I believe I will be able to work on this project 10-20 hours
>  >  a week; sometimes more, sometimes less. I can also send you weekly progress
>  >  report on my work.
>  >
>  >  --
>  >  Best regards,
>  >  Yuriy Znovyak mailto:oberon.ua@gmail.com
>  >
>  --
>  With best regards,
>  Alexei


View raw message