Re: GSoC 2017 Project "Make Anita support multiple virtual machine systems"

[christos%astron.com@localhost (Christos Zoulas)]

In article <CADd-wOYHU18CAFzs-yEb_WEL75VT1v281-3i0UxmhEr+M1GYAA%mail.gmail.com@localhost>,
Utkarsh Anand  <uanand009%gmail.com@localhost> wrote:
>-=-=-=-=-=-
>-=-=-=-=-=-
>
>Sir/Ma'am,
>I am Utkarsh Anand, a second year undergraduate student studying at Indian
>Institute of Technology, Kanpur. I have been using linux since the year
>2009. I am familiar with many programming languages (C, C++, Python,
>Javascript etc. to name a few). I would like to apply for "Make Anita
>support multiple virtual machine systems" as GSoC 2017 project. I have
>already gone through the entire code of anita 1.40. I had a few questions
>related to the project:
>1. It looks like it already had support for xen, although, the page:
>http://www.gson.org/netbsd/anita/ says that it has 'experimental support'.
>So, are there any issues that need to be resolved? Or any features that
>need to be added? I'd like to know about it.

Yes, I have exchanged some emails with gson@ and he brought me up to speed.
I tried to update the page, but I could not. I hope to resolve this later.

Xen has been added, but still we would like to cover other architectures
that are not handled by qemu; tme and simh come to mind.

There is also real work in getting many of the architectures working with
anita. For a port to be anita-friendly it must be using sysinst over a
serial console without any cheat (such as passing the kernel on the
command line); it should boot directly to sysinst from the install media.

>2. Line nos. 125-132 (the function definition for make_dense_image) define
>a function for creating a file of given file size. Was there any advantage
>(as in speed etc.) for choosing to write your own function instead of
>something like
><code>
>ret =
>os.spawnvp(os.P_WAIT,'dd',['dd','if=/dev/zero','of=/home/utkarsh/new.img','bs=1MiB','count=10'])
>if not ret:
>    print("image created!")
></code>
>We could get the block size, count and filename from the user, just like
>the original function and the user name from the environment variable USER.

Yes, I agree I think using dd is better.

>3. If the xen support is complete, the only thing left to do is virtualbox
>support, which should be similar to qemu and xen. Am I right?

Read above [simh/tme?]

>4. What other modules would you like to add?

I am not sure, let's work that out later. Let's do the other work first.

>Lastly, I would like to assure that I am ready to put in any amount of
>effort required for the project.

Great!

Best,

christos