Re: [myhdl-list] Distributed revision control for MyHDL

[Blubaugh, D. A. <dbl...@be...>]

Jan,

Hello.  This is David Blubaugh.  I have also worked on distributed
systems before.  I have worked on a sensor fusion system based on the
kalman filter, which was in a distributed network form.  I agree that
distributed systems are indeed what the future holds.  I was just
wondering if the distributed technique that you described can be carried
over to the generation of verilog source code?   


Thanks,


David Blubaugh




 

-----Original Message-----
From: myh...@li...
[mailto:myh...@li...] On Behalf Of Jan
Decaluwe
Sent: Thursday, May 29, 2008 2:27 PM
To: myh...@li...
Subject: [myhdl-list] Distributed revision control for MyHDL

Hello:

In the past I've called subversion a "winning solution" for revision
control and I have been using it for MyHDL.

Despite this, I never truly liked it. In particular, the way to do tags
and branches ("everything is a copy") seems attractive at first, but is
very unpractical, to the point of being unusable.

Recently I have understood that there are really 2 types of revision
control systems: centralized versus distributed. While subversion is a
winner among centralized systems, I have now become convinced that
distributed is the wave of the future.

I had experience with darcs (a distributed system) before, but I never
found that very convincing. What triggered my thinking were 2 things:
- a google video from Linus Torvalds about git
- a very positive experience with git in a real design project

With git, I found that after some days I was routinely doing things that
I had considered "advanced" with others systems, such as short-lived
feature branches.

Also, it is clear that distributed is much more powerful and versatile
in an open-source environment. Some issues inherent to centralized (e.g.
need for a connection to a server) simply go away.

After this, I decided I wanted something like this for MyHDL.
Besides git, only 1 other system is "tolerated" by Linus, and that is
mercurial. As this is a Python-based system (unlike git) and as many
high-profile projects are converting to it, you can guess what choice I
made ... I think git is probably more powerful at this point, but
mercurial is probably somewhat easier to learn and handle.

mercurial contains a convert tool that can convert subversion
repositories, which uses the subversion Python bindings. I had to do
quite some hacking (recompiling subversion with some libraries
disabled) before the conversion worked, but it finally did.
Then I followed a recipe to publish the repo on sourceforge.

You can see the result here:

    http://myhdl.sourceforge.net/hg/myhdl

You can use this to browse the repo, but also for other things - but
this I will explain in a subsequent post.

Jan

--
Jan Decaluwe - Resources bvba - http://www.jandecaluwe.com
Kaboutermansstraat 97, B-3000 Leuven, Belgium
     From Python to silicon:
     http://myhdl.jandecaluwe.com


------------------------------------------------------------------------
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
myhdl-list mailing list
myh...@li...
https://lists.sourceforge.net/lists/listinfo/myhdl-list

This e-mail transmission contains information that is confidential and may be 
privileged. It is intended only for the addressee(s) named above. If you receive 
this e-mail in error, please do not read, copy or disseminate it in any manner. 
If you are not the intended recipient, any disclosure, copying, distribution or 
use of the contents of this information is prohibited. Please reply to the 
message immediately by informing the sender that the message was misdirected. 
After replying, please erase it from your computer system. Your assistance in 
correcting this error is appreciated.