RE: THANK YOU for the Linuxdays (fwd)

RE: THANK YOU for the Linuxdays (fwd)

[Frank Benke]

hello folks;

since you all seem to be interested in linux and are somehow related to
the comercial use of it, i strongly encourage you to sign the petition for
including the logical volume management in the linux kernel. linus
concidered lvm as NOT important and is not convinced that this include is
of common interest. this petition shall convince him otherwise.

the website is found at http://linux.msede.com/lvm .

if you have friends and colleagues, which might think positive about lvm,
please inform them and ask them to sign as well. logical volume management
is as important as filesystem-journaling and the entire lvm in its alpha
release runs quite well here at gis.

thanks for your attention

regards frank

Re: THANK YOU for the Linuxdays (fwd)

[Ross J. Reedstrom]

Frank - 
Have you spent much time reading the linux-kernel mailing list? Linus is
usually little swayed by 'popular opinion' as to what features go into
the kernel. The principle criteria are that any new feature not hurt
other features, and that the quality of the code is good. This petition
drive reminds me of ~3 years ago, when some people following the GGI
(General Graphics Interface) project wanted to get the early alpha
code adopted into the mainstream kernel. In my opinion (and that of
many kernel list regulars) the 'campaign' _slowed_ acceptance, rather
than helping. In fact, the basic underlying architecture (a generic
framebuffer interface) came into the mainstream kernel from the ports
to architectures(SPARC, I seem to recall) that required it, and was
eventually adapted to the i386 architecture, as well.

So, I'm not convinced that a petition is going to help at all, and
might hurt. If Linus Torvalds has rejected the lvm patches based on 'I
don't think many people need this' (which is _not_ a usual criteria),
it might help a little, to let the developers of the LVM patches point
to as interest. But they should have download logs indicating a lot of
people grabbing the patches, which says the same thing, doesn't it? I'm
glad the patches work well for you. Be happy features like this can be
developed by third parties. Help the LVM developers create clean code
that patches in easily, and it'll end up in the kernel.

Ross

Ob.PA-RISC: Sorry my first post here is off topic. Perhaps I can
salvage that.  I've just aquired what I believe to be a HP9000 730/66
a.k.a. 'King Cobra' workstation. I haven't much time to play with it
right now, but I _will_ be following this list, and trying to get your
work up and running on it. Hopefully, this isn't too different from other
7XX machines. Anyone know of any 'gotcha's ? (Oh, and I have zero HP/UX
experience (all Irix and Linux background).


On Wed, Sep 08, 1999 at 10:05:34AM +0200, Frank Benke wrote:
> hello folks;
> 
> since you all seem to be interested in linux and are somehow related to
> the comercial use of it, i strongly encourage you to sign the petition for
> including the logical volume management in the linux kernel. linus
> concidered lvm as NOT important and is not convinced that this include is
> of common interest. this petition shall convince him otherwise.
> 
> the website is found at http://linux.msede.com/lvm .
> 
> if you have friends and colleagues, which might think positive about lvm,
> please inform them and ask them to sign as well. logical volume management
> is as important as filesystem-journaling and the entire lvm in its alpha
> release runs quite well here at gis.
> 
> thanks for your attention
> 
> regards frank
> 
> 
> 
> ---------------------------------------------------------------------------
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.

-- 
Ross J. Reedstrom, Ph.D., <reedstrm@rice.edu> 
NSBRI Research Scientist/Programmer
Computer and Information Technology Institute
Rice University, 6100 S. Main St.,  Houston, TX 77005

Re: THANK YOU for the Linuxdays (fwd)

[Frank Benke]

okok ... i shouldnt have posted that. sorry. i dared because i understood
up to now the status of the current discussion was, that the only point of
critics from linus is the lack of need (and so far the performance
decrease is no fair tradeoff). linus should not be forced to include
bad/incomplete/buggy code, he should only see that there are people who
see this need and who want to make this tradeoff. and besides, we work on
improvements. 

and ross, as far as i understood the model philosophy, your 730 should
behave like 720 or 735.

regards frank

Re: THANK YOU for the Linuxdays (fwd)

[Matthew Wilcox]

On Wed, Sep 08, 1999 at 10:03:31AM -0500, Ross J. Reedstrom wrote:
> Ob.PA-RISC: Sorry my first post here is off topic. Perhaps I can
> salvage that.  I've just aquired what I believe to be a HP9000 730/66
> a.k.a. 'King Cobra' workstation. I haven't much time to play with it
> right now, but I _will_ be following this list, and trying to get your
> work up and running on it. Hopefully, this isn't too different from other
> 7XX machines. Anyone know of any 'gotcha's ? (Oh, and I have zero HP/UX
> experience (all Irix and Linux background).

Maybe this will be OK.  It's PA-RISC 1.1a, a PA7000 processor like the
705/710/720/750 rather than the PA7100LC as the 712/715/etc machines have.
We have docs on the 7100 and 7300 CPUs on the website, but no docs on
the 7000.  I suspect that getting documentation on this CPU will be
low-priority; even lower priority than the HIL keyboard driver :-)
So it may work, but don't be surprised if it doesn't.

-- 
Matthew Wilcox <willy@bofh.ai>
"Windows and MacOS are products, contrived by engineers in the service of
specific companies. Unix, by contrast, is not so much a product as it is a
painstakingly compiled oral history of the hacker subculture." - N Stephenson

Re: THANK YOU for the Linuxdays (fwd)

[Mike Hibler]

> Date: Wed, 8 Sep 1999 18:49:32 +0200
> From: Matthew Wilcox <Matthew.Wilcox@genedata.com>
> To: "Ross J. Reedstrom" <reedstrm@wallace.ece.rice.edu>
> Subject: Re: THANK YOU for the Linuxdays (fwd)
> 
> On Wed, Sep 08, 1999 at 10:03:31AM -0500, Ross J. Reedstrom wrote:
> > Ob.PA-RISC: Sorry my first post here is off topic. Perhaps I can
> > salvage that.  I've just aquired what I believe to be a HP9000 730/66
> > a.k.a. 'King Cobra' workstation. I haven't much time to play with it
> > right now, but I _will_ be following this list, and trying to get your
> > work up and running on it. Hopefully, this isn't too different from other
> > 7XX machines. Anyone know of any 'gotcha's ? (Oh, and I have zero HP/UX
> > experience (all Irix and Linux background).
> 
> Maybe this will be OK.  It's PA-RISC 1.1a, a PA7000 processor like the
> 705/710/720/750 rather than the PA7100LC as the 712/715/etc machines have.
> We have docs on the 7100 and 7300 CPUs on the website, but no docs on
> the 7000.  I suspect that getting documentation on this CPU will be
> low-priority; even lower priority than the HIL keyboard driver :-)
> So it may work, but don't be surprised if it doesn't.
> 

Those original snakes have the ASP bus, which is quite similar to the LASI.
You can look at the Mach code for differences.  A couple of other differences
that might bite you: the snakes don't have HW TLB miss handling support,
the block TLBs have some funky restrictions, and they have the NCR53C700
SCSI controller which is a little quirky compared to the 53C710 in the later
machines.  But none of these are show-stoppers.