Re: HFS Patch 16 ported to Parted 1.6.19

[Sven Luther]

On Sat, Dec 04, 2004 at 10:05:10PM +1100, Andrew Clausen wrote:
> On Sat, Dec 04, 2004 at 11:45:59AM +0100, Sven Luther wrote:
> > Ok. So you would give up looking at the development branch ?
> 
> I guess so.  Well, I could continue my current approach of "take
> a look at parted every 2 weeks".  This isn't very good for helping
> people get their patches applied.

I understand.

> > It really depends on how much development the stable branch is taking.
> 
> I imagine the stable branch will continue to take about the same
> amount of development as it has in the past 3 months or so.
> 
> > > Perhaps the biggest amount of work in looking after a development
> > > branch is applying the changes from the stable branch to the development
> > > one.  Anyone interested?
> > 
> > Well, what would be the task ? Doing the development, or just handling
> > the different patches and so provided by folk.
> 
> The latter only.

Basically you are looking for a new maintainer, i think. I proposed myself
last year, but failed miserably, so i will be carefull in volunteering my time
right now, ...

> > Should we not have a first round of discussion on the future of parted
> > before we start a devel branch, and would this led to parted 2.0 ?
> > Or do you want only a devel branch to early test things like the hfs+
> > patches ? 
> 
> I'm mostly talking about the latter.  I would like to make it
> easier for people who want to contribute to Parted to get the help
> they need... in particular testing.

Why not have a stable and devel branch, with the devel branch being widely
open, and people could just commit to it, and we setup an automated daily
build, with a blame mail to anyone who breaks the build, CCed to the mailing
list or something such.

> > BTW, on a separate note : 
> > 
> >   2004-11-21  Andrew Clausen <address@hidden>
> >   * libparted/linux.c (linux_new): get rid of pointless exception for
> >     exotic block devices.
> > 
> > Has a negative effect of now not making the difference between a
> > normal block device and a CD rom device or something such, which makes
> > partitioners out there offer CD drives for partitioning. I think this
> > is in the same category as the lvm2 patch, not sure though.
> 
> Sounds like it.  Know any ways to tell CDs apart from everything else?

Not really, but the kernel assuredly knows how to make the difference. I heard
about another way to fix this consisting in checking if the media is
read-only, and not modifying partition tables in this case ...

> > And secondly, about this EDD module. I have two questions : 1) it will only
> > work on x86, right ?
> 
> Yes.  It is only relevant on x86, though.

What about disks prepared on non-x86, and then moved ? 

> > And 2) in how far is it different from the int13 stuff
> > XFree86 uses to run bioses to initialize graphic cards that weren't primary
> > ones at startup time ? 
> 
> I don't know.  I suspect the mechanism for running the BIOS code is
> similar.

Still, the X code doesn't run in a kernel module as far as i know. I may be
wrong though.

Friendly,

Sven Luther