[ECOS] NAND framework

[ECOS] NAND framework

[Andrew Lunn]

Hi Folks

Its time to address the issue of which NAND framework we will include
into eCos. This has been dragging on for a while, far too long, and is
hindering the adoption of NAND in eCos. So three of the most
independent eCos maintainers, John Dallaway, Gary Thomas and myself,
have come to an agreement about the way forward.

The issue is thorny because there is no clear technically superior
solution. Neither VU/Rutgers nor eCosCentric/Ross implementation is
clearly better than the other. So the decision is more philosophical
than technical.

We have chosen to include Rutger Hofman's implementation. 

Right from the very start Rutger did all the right things, in terms of
a developing a piece of software for an open source community
project. He posted his intention to write a NAND framework. He posted
his proposed APIs and accepted feedback, participated in discussion
and made changes which resulted from the discussion. He has posted
his code early and often and it has been used in at least two projects
we know of. As maintainers this is the style of development we must
support and encourage.

eCoscentric's solution is not without merit and eCosCentric have
consistently shown they produce high quality solutions for eCos. We
really hope that this decision does not discourage eCosCentric from
making further valuable contributions to eCos.

Hopefully the code will be committed to CVS in the next few days. As
always, everybody is welcome to contribute improvements in the form of
patches.

          Andrew

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] NAND framework

[Rutger Hofman]

Andrew Lunn wrote:
> Its time to address the issue of which NAND framework we will include
> into eCos.
[snip]
> We have chosen to include Rutger Hofman's implementation. 

I am glad this is now settled, and of course I am really happy that my 
contribution to the open-source community will be incorporated! I will 
contact Ross to see if we can come to some cross-fertilization.

> Hopefully the code will be committed to CVS in the next few days. As
> always, everybody is welcome to contribute improvements in the form of
> patches.

I would like a few days to polish up the code a bit. Whom should I 
contact when I am done?

Thanks,

Rutger Hofman
VU Amsterdam

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] NAND framework

[Andrew Lunn]

> I would like a few days to polish up the code a bit. Whom should I
> contact when I am done?

The ecos-patches list, even if all you do is say there is another
release on your website.

I took a very quick look last night at some files. Some copyright
headers need adding, others correcting. Also, it would be nice to
remove dead code inside comments, and maybe some of the debug
prints. I found a few in one one of the drivers which just said
"Here", etc. Also, please add ChangeLog files in each package, even if
it only says its the first release of the driver.

        Andrew

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] NAND framework

[Jonathan Larmour]

Regarding the NAND flash driver framework announcement:

   http://ecos.sourceware.org/ml/ecos-discuss/2009-08/msg00057.html

some issues need to be clarified.

After further in-depth consideration, we now realise that this decision 
was premature. There was inadequate discussion devoted to any technical 
basis for the decision and the rationale behind that, including the 
opportunity for corrections from the contributors, the full maintainer 
team, and the rest of the community.

Also we need to clarify that the full maintainer team should have been 
involved in the discussion. There was no intention by anyone to imply that 
any of the maintainers, whoever they are employed by, have not been acting 
in the best interests of eCos users. Any impression to the contrary given 
by the earlier mail was unintentional. It is a well-understood requirement 
of being an eCos maintainer that all must act independently from their 
employers. Maintainers do not act on their employer's behalf. Should there 
be a possibility of a conflict of interest, the maintainers will look at 
each case on its own merits.

All the maintainers would also like to apologise to Rutger for the 
premature announcement, and thank him for his patience, and in return we 
intend to finish a technical evaluation rapidly to end the uncertainty. 
This evaluation will be discussed on the public lists this time to allow 
everyone interested an opportunity to contribute.

Jifl, on behalf of the eCos maintainers

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] NAND framework

[Jonathan Larmour]

Jonathan Larmour wrote:
> we 
> intend to finish a technical evaluation rapidly to end the uncertainty. 
> This evaluation will be discussed on the public lists this time to allow 
> everyone interested an opportunity to contribute.

For clarity, I'm kicking this off on ecos-devel, rather than ecos-discuss.

Jifl
-- 
--["No sense being pessimistic, it wouldn't work anyway"]-- Opinions==mine

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss