Re: Re: [parisc-linux] STI framebuffer and PINE

Re: Re: [parisc-linux] STI framebuffer and PINE

[Stephan Trajkoff]

I don't know if my STI can run more colours. I have no documentation=20
for my server
./build slx and ./build lnx is commands for compiling pine
before compiling PINE i did apt-get dist-upgrade
but again PINE package can't compile, only pico and pilot works, not=20
pine=20
PINE in hppa dist is only avalailable as source, not binary
my clients want pine, not mutt



-------- Original Message --------

=3D=3D> From: Ralf Hildebrandt <Ralf.Hildebrandt@charite.de>
=3D=3D> Date: Sun, 14 Jul 2002 10:12:41  0200

On Sun, Jul 14, 2002 at 09:52:30AM  0300, Stephan Trajkoff wrote:

> STI GSC/PCI graphics driver version 0.9 > STI word mode ROM at=20
f0060000, hpa at f4000000 > STI id 2b4ded6d-40a00499, conforms to
spec=20
rev. 8.04 > STI device: HPA208LC1280 > Console: switching to colour=20
frame buffer device 160x64 > fb0: stifb 1280x1024-8 frame buffer=20
device, id: 2b4ded6d, mmio: > 0xf4100000 > I have this shit, can=20
anyone help me to run XFREE in 15,16,24 or 32 > bit color? > my HP is=20
819/K210

Are you sure the framebuffer can run at more colors? I don't find
this=20
on http://docs.hp.com/hpux/hw/ !

> Another problem how I can run PINE? > I have downloaded with apt-get=20
pine 4.44

apt-get install pine

but pine is junk anyway. Try mutt.

> ./build lnx and ./build slx don't work

What is this?

> there is an error in PINE

Then report it using

reportbug pine

> before compiling pine, I did apt-get dist-upgrade

You compiled pine? I though you used apt-get?!

-- Ralf Hildebrandt (Im Auftrag des Referat V A)=20=20=20
Ralf.Hildebrandt@charite.de Charite Campus Virchow-Klinikum=20=20=20=20=20=
=20=20=20=20=20=20
      Tel.   49 (0)30-450 570-155 Referat V A - Kommunikationsnetze -=20
            Fax.   49 (0)30-450 570-916 If JavaScript is walking
alone=20
late at night through a bad part of town with a pocket full of $20=20
bills, ActiveX is dropping your trousers in the middle of the yard of=20
a maximum-security prison, bending over, and yelling 'Come and get
it,=20
boys!'=20

_______________________________________________ parisc-linux mailing=20
list parisc-linux@lists.parisc-linux.org=20
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux=20

-------------------------------------------------------------------
-->> Add WEB ACCESS to your POP email accounts with
-->> FREE Perl CGI scripts! Download today! http://www.adjeweb.com
-------------------------------------------------------------------

[parisc-linux] Re: Re: STI framebuffer and PINE

[Martin Schulze]

Stephan Trajkoff wrote:
> PINE in hppa dist is only avalailable as source, not binary

This is due to the license of PINE.  It cannot be distributed as
binary or binary built from modified source.  Hence, only the
plain source can be distributed.

> my clients want pine, not mutt

Even though it doesn't help you in the first place, but since
Mutt comes with a Pine.rc in the examples directory, you may
be able to educate one client or another to use the free Mutt
with these settings.  It may reduce the pressure on you a little
bit.

Regards,

	Joey

-- 
If you come from outside of Finland, you live in wrong country.
	-- motd of irc.funet.fi

Re: [parisc-linux] STI framebuffer and PINE

[Ralf Hildebrandt]

Am 14.07.2002 um 12:17:46 +0300 schrieb Stephan Trajkoff folgendes:
> I don't know if my STI can run more colours. I have no documentation 
> for my server

Well, check the pages I mentioned.

> ./build slx and ./build lnx is commands for compiling pine


> before compiling PINE i did apt-get dist-upgrade

Fine. Don't forget to
apt-get update
first

> but again PINE package can't compile, only pico and pilot works, not 
> pine 

Did you get a PINE source package (is there one?) or the original source?

> my clients want pine, not mutt

Their fault.

-- 
Ralf Hildebrandt (Im Auftrag des Referat V A)   Ralf.Hildebrandt@charite.de
Charite Campus Virchow-Klinikum                 Tel.  +49 (0)30-450 570-155
Referat V A - Kommunikationsnetze -             Fax.  +49 (0)30-450 570-916
A complex system that works is invariably found to have evolved from a
simple system that worked.                                -- John Gall 

Re: [parisc-linux] STI framebuffer and PINE

[Ralf Hildebrandt]

Am 14.07.2002 um 11:50:17 +0200 schrieb Ralf Hildebrandt folgendes:

> > ./build slx and ./build lnx is commands for compiling pine

WHat is the error you're getting?
 
> > but again PINE package can't compile, only pico and pilot works, not 
> > pine 

What happens?

-- 
Ralf Hildebrandt (Im Auftrag des Referat V A)   Ralf.Hildebrandt@charite.de
Charite Campus Virchow-Klinikum                 Tel.  +49 (0)30-450 570-155
Referat V A - Kommunikationsnetze -             Fax.  +49 (0)30-450 570-916
Your file was so big.
It might be very useful.
But now it is gone.

Re: Re: [parisc-linux] STI framebuffer and PINE

[Matthew Wilcox]

On Sun, Jul 14, 2002 at 12:17:46PM +0300, Stephan Trajkoff wrote:
> my clients want pine, not mutt

i have to say i think you're a complete lunatic for wanting to support
`clients' on a K class running PA/Linux right now.

-- 
Revolutions do not require corporate support.

Re: [parisc-linux] STI framebuffer and PINE

[Stephan Trajkoff]

My K class server works perfect with 240 users for mail.
this server has apache, squid, pine,qmail
4 cpu with 120mhz, 512mb ramm, 18GB HDD, what more???
my linux in my ISP is with P3 800Mhz, 256mb ram, 20GB HDD wtih apache, 
mysql, radius, routing, shapers.
Why did you call me lunatic???
Do you think that other servers in Bulgaria are P4 1.xGhz or palomino 
1.xGhz, the average is 500-1000Mhz celerons or p3
The old server was IBM7248-120, powerpc 133mhz, 32mb ram, 1.2GB SCSI HDD 
and hosted 240 users
It is bulgaria man, it is the reality:)
Come here and see motor scooter 50cc runing at 80km/h:)))

Re: [parisc-linux] STI framebuffer and PINE

[Ralf Hildebrandt]

On Mon, Jul 15, 2002 at 04:27:05PM +0300, Stephan Trajkoff wrote:

> My K class server works perfect with 240 users for mail.
> this server has apache, squid, pine,qmail

Quite nice.

> 4 cpu with 120mhz, 512mb ramm, 18GB HDD, what more???

* squid doesn't actually use more than one processor (state machine!)
* qmail is I/O bound.

Only a new Apache2 would benefit from a multi processor box.
If you replace qmail with postfix, you might speed things up a bit.

> my linux in my ISP is with P3 800Mhz, 256mb ram, 20GB HDD wtih apache, 
> mysql, radius, routing, shapers.
> Why did you call me lunatic???

Because Linux isn't really as stable on HPPA as on say Alpha or x386!

> Come here and see motor scooter 50cc runing at 80km/h:)))

I bet it's against the law :)

-- 
Ralf Hildebrandt (Im Auftrag des Referat V A)   Ralf.Hildebrandt@charite.de
Charite Campus Virchow-Klinikum                 Tel.  +49 (0)30-450 570-155
Referat V A - Kommunikationsnetze -             Fax.  +49 (0)30-450 570-916
Why you can't find your system administrators:
Went to Stop&Go to get ANOTHER case of insert favorite caffinated beverage here 

Re: [parisc-linux] STI framebuffer and PINE

[Ralf Hildebrandt]

Am 15.07.2002 um 15:37:59 +0100 schrieb Alan Cox folgendes:

> Someone has to go first. I ran www.linux.org.uk on the first ever Mac68K
> Linux box and that helped no end in the stabilisation process.
> 
> [ObOnTopic] What is the state of the 2.4 pa-risc code in terms of rate
> of change now. Would it be a good time to use -ac as a real sync up ?

Whoa! What an honor!
I was already wondering how to get all that pa5x stuff back into the
mainline 2.4.x kernel.

-- 
Ralf Hildebrandt (Im Auftrag des Referat V A)   Ralf.Hildebrandt@charite.de
Charite Campus Virchow-Klinikum                 Tel.  +49 (0)30-450 570-155
Referat V A - Kommunikationsnetze -             Fax.  +49 (0)30-450 570-916
"If it is blue and winding, it is biology. If it stinks, it is chemistry. If
it doesnŽt work, it is physics. If it is unintelligible, it is mathematics.
If it is all of above, it must has been made by M$."      -- Holger Dittmann

Re: Re: [parisc-linux] STI framebuffer and PINE

[Matthew Wilcox]

On Mon, Jul 15, 2002 at 03:37:59PM +0100, Alan Cox wrote:
> On Mon, 2002-07-15 at 14:20, Matthew Wilcox wrote:
> > i have to say i think you're a complete lunatic for wanting to support
> > `clients' on a K class running PA/Linux right now.
> 
> Someone has to go first. I ran www.linux.org.uk on the first ever Mac68K
> Linux box and that helped no end in the stabilisation process.

Certainly, and I'm aware of people using PA/Linux in production environments
today.  But the K class hardware is not well supported.  Lan and Console
are the only two devices which really work; the SCSI cards don't use the
normal NCR/Symbios chips but a proprietary chip instead.

Hmm.. anyone want to do a `testimonials' section on the website which
lists peoples (successful ;-) experiences with using PA/Linux in
production environments?

> [ObOnTopic] What is the state of the 2.4 pa-risc code in terms of rate
> of change now. Would it be a good time to use -ac as a real sync up ?

rate of change is fairly low; i attempted to do a merge with marcelo
during early 2.4.19 dev, but someone sent some conflicting patches in
first, i tripped over my own feet trying to use bitkeeper and ran out
of motivation to fix conflicts myself.

at OLS we decided to concentrate on 2.5 RSN.  it's more important to
work on that than keep trying to push bits into 2.4.

-- 
Revolutions do not require corporate support.

Re: Re: [parisc-linux] STI framebuffer and PINE

[Matthew Wilcox]

On Mon, Jul 15, 2002 at 03:57:48PM +0100, Alan Cox wrote:
> Is there somewhere I can yank a definitive stable 2.4 tree and I'll take
> a look at the 2.4 stuff. I've had a bit of practice at the merging game
> 8)

http://ftp.parisc-linux.org/cvs/linux-2.4.18-pa55.tar.gz (29MB)

-- 
Revolutions do not require corporate support.

Re: Re: [parisc-linux] STI framebuffer and PINE

[Alan Cox]

On Mon, 2002-07-15 at 14:20, Matthew Wilcox wrote:
> On Sun, Jul 14, 2002 at 12:17:46PM +0300, Stephan Trajkoff wrote:
> > my clients want pine, not mutt
> 
> i have to say i think you're a complete lunatic for wanting to support
> `clients' on a K class running PA/Linux right now.

Someone has to go first. I ran www.linux.org.uk on the first ever Mac68K
Linux box and that helped no end in the stabilisation process.

[ObOnTopic] What is the state of the 2.4 pa-risc code in terms of rate
of change now. Would it be a good time to use -ac as a real sync up ?

Re: Re: [parisc-linux] STI framebuffer and PINE

[Alan Cox]

On Mon, 2002-07-15 at 14:39, Matthew Wilcox wrote:
> > [ObOnTopic] What is the state of the 2.4 pa-risc code in terms of rate
> > of change now. Would it be a good time to use -ac as a real sync up ?
> 
> rate of change is fairly low; i attempted to do a merge with marcelo
> during early 2.4.19 dev, but someone sent some conflicting patches in
> first, i tripped over my own feet trying to use bitkeeper and ran out
> of motivation to fix conflicts myself.

People who do merges dont use bitkeeper

> at OLS we decided to concentrate on 2.5 RSN.  it's more important to
> work on that than keep trying to push bits into 2.4.

Is there somewhere I can yank a definitive stable 2.4 tree and I'll take
a look at the 2.4 stuff. I've had a bit of practice at the merging game
8)

Re: Re: [parisc-linux] STI framebuffer and PINE

[Alan Cox]

On Mon, 2002-07-15 at 14:55, Matthew Wilcox wrote:
> On Mon, Jul 15, 2002 at 03:57:48PM +0100, Alan Cox wrote:
> > Is there somewhere I can yank a definitive stable 2.4 tree and I'll take
> > a look at the 2.4 stuff. I've had a bit of practice at the merging game
> > 8)
> 
> http://ftp.parisc-linux.org/cvs/linux-2.4.18-pa55.tar.gz (29MB)

Munch...

I'll let folks know how it goes

Re: [parisc-linux] STI framebuffer and PINE

[Grant Grundler]

Matthew Wilcox wrote:
> But the K class hardware is not well supported.  Lan and Console
> are the only two devices which really work; the SCSI cards don't use the
> normal NCR/Symbios chips but a proprietary chip instead.

Not entirely correct. K-class has both HSC (GSC follow-on) and HP-PB "slots".
IIRC, the built-in SCSI is HSC based. But you are right in that all the HP-PB
crud isn't supported.

> Hmm.. anyone want to do a `testimonials' section on the website which
> lists peoples (successful ;-) experiences with using PA/Linux in
> production environments?

ESIEE folks have that already in their "what's supported" HW listing.
Please see the ESIEE "puffin" site for where to submit them.

> > [ObOnTopic] What is the state of the 2.4 pa-risc code in terms of rate
> > of change now. Would it be a good time to use -ac as a real sync up ?
> 
> rate of change is fairly low; i attempted to do a merge with marcelo
> during early 2.4.19 dev, but someone sent some conflicting patches in
> first, i tripped over my own feet trying to use bitkeeper and ran out
> of motivation to fix conflicts myself.

I agree, hppa is quite stable on UP and most SMP systems.
BTW - no one ever said "Thank you" for doing that - kudos!

> at OLS we decided to concentrate on 2.5 RSN.  it's more important to
> work on that than keep trying to push bits into 2.4.

yes - but I'd still like to see our changes get into 2.4.
I just want to be doing development on 2.5.

Alan Cox wrote:
| Is there somewhere I can yank a definitive stable 2.4 tree and I'll take
| a look at the 2.4 stuff. I've had a bit of practice at the merging game
| 8)

Just a bit? ;^)

For a full source tree:
	ftp://ftp.parisc-linux.org/cvs/linux-latest.tar.gz

lrwxrwxrwx    1 bame     ftpadmin       24 Jul 13 18:35 linux-latest.tar.gz -> linux-2.4.18-pa55.tar.gz
-rw-r--r--    1 bame     ftpadmin 30508560 Jul 13 18:35 linux-2.4.18-pa55.tar.gz

hth,
grant

Re: [parisc-linux] 2.4.19 merge

[Grant Grundler]

[ sorry - change the subject line ]

Alan Cox wrote:
> [ObOnTopic] What is the state of the 2.4 pa-risc code in terms of rate
> of change now. Would it be a good time to use -ac as a real sync up ?

BTW, I've been testing Ivan K's PCI rewrite/changes on A500 for the past
couple of monthes but never committed those to CVS. I guess I didn't
want more churn until Ivan could tell me his bits were in 2.4.

Basically with his code, I needed few or no changes to drivers/pci.
I'd have to review the diff again.
Ivan last sent me a patch on 18.4.2002.
You want matching changes for arch/parisc as well?

thanks,
grant

Re: [parisc-linux] 2.4.19 merge

[Rafael E. Herrera]

Seems like the merge already begun:

http://linuxtoday.com/news_story.php3?ltsn=2002-07-15-015-26-NW-KN-DV

-- 
      Rafael

Re: [parisc-linux] 2.4.19 merge

[Alan Cox]

On Tue, 2002-07-16 at 00:03, Grant Grundler wrote:
> 
> [ sorry - change the subject line ]
> 
> Alan Cox wrote:
> > [ObOnTopic] What is the state of the 2.4 pa-risc code in terms of rate
> > of change now. Would it be a good time to use -ac as a real sync up ?
> 
> BTW, I've been testing Ivan K's PCI rewrite/changes on A500 for the past
> couple of monthes but never committed those to CVS. I guess I didn't
> want more churn until Ivan could tell me his bits were in 2.4.
> 
> Basically with his code, I needed few or no changes to drivers/pci.
> I'd have to review the diff again.
> Ivan last sent me a patch on 18.4.2002.
> You want matching changes for arch/parisc as well?

That would be interesting to see yes

Re: [parisc-linux] 2.4.19 merge

[Grant Grundler]

Alan Cox wrote:
> > Basically with his code, I needed few or no changes to drivers/pci.
> > I'd have to review the diff again.
> > Ivan last sent me a patch on 18.4.2002.
> > You want matching changes for arch/parisc as well?
> 
> That would be interesting to see yes

Original diff from Ivan (plus email comments) is
	ftp://ftp.parisc-linux.org/patches/pci-20020418-ink.diff

Diff against -pa55 (including my changes) is at:
	ftp://ftp.parisc-linux.org/patches/pci-20020418-pa55.diff

DEBUG should probably be disabled in drivers/pci/setup-bus.c.

grant

Re: [parisc-linux] STI framebuffer and PINE

[Matthieu Delahaye]

Grant Grundler wrote:

>Matthew Wilcox wrote:
>  
>
>>But the K class hardware is not well supported.  Lan and Console
>>are the only two devices which really work; the SCSI cards don't use the
>>normal NCR/Symbios chips but a proprietary chip instead.
>>    
>>
>
>Not entirely correct. K-class has both HSC (GSC follow-on) and HP-PB "slots".
>IIRC, the built-in SCSI is HSC based. But you are right in that all the HP-PB
>crud isn't supported.
>
>  
>
>>Hmm.. anyone want to do a `testimonials' section on the website which
>>lists peoples (successful ;-) experiences with using PA/Linux in
>>production environments?
>>    
>>
>
>ESIEE folks have that already in their "what's supported" HW listing.
>Please see the ESIEE "puffin" site for where to submit them.
>  
>
I don't think Matthew wath thinking about that. This HW list gives just 
an indication if this is working or not.
What I understand here is a list of something like:
I currently use a B132 in production environment since November, 2001. 
It provides a CVS server, apache with PHP and  a postgresql database. 
The kernel is pretty old (2.4.14-32), but the only shutdown (and end of 
services) I had to notice where caused by powerfailure.
Use of Php is quite recent.
Matthew, does these (real) example correspond to your suggestion?

>  
>
>[...]
>
>hth,
>grant
>_______________________________________________
>parisc-linux mailing list
>parisc-linux@lists.parisc-linux.org
>http://lists.parisc-linux.org/mailman/listinfo/parisc-linux
>
>
>  
>

Re: testimonials (was Re: [parisc-linux] STI framebuffer and PINE)

[Thibaut VARENE]

Le mardi 16 juillet 2002, =E0 09:29 , Matthieu Delahaye a =E9crit :

> Grant Grundler wrote:
>
>> Matthew Wilcox wrote:
>>
>>> Hmm.. anyone want to do a `testimonials' section on the website =
which
>>> lists peoples (successful ;-) experiences with using PA/Linux in
>>> production environments?
>>>
>>
>> ESIEE folks have that already in their "what's supported" HW listing.
>> Please see the ESIEE "puffin" site for where to submit them.
>>
> I don't think Matthew wath thinking about that. This HW list gives =
just=20
> an indication if this is working or not.
> What I understand here is a list of something like:
> I currently use a B132 in production environment since November, 2001.=20=

> It provides a CVS server, apache with PHP and  a postgresql database.=20=

> The kernel is pretty old (2.4.14-32), but the only shutdown (and end =
of=20
> services) I had to notice where caused by powerfailure.
> Use of Php is quite recent.
> Matthew, does these (real) example correspond to your suggestion?
>
Hi fellows!

I'm currently working on a big review of our supported hardware list.
A testimonial section is definitely something that can be considered !

Please check this mail i sent a month ago about this review:
http://lists.parisc-linux.org/pipermail/parisc-
linux/2002-June/016572.html

I'd enjoy any feedback about it, and of course any supplementary=20
testimonial
would be greatly welcomed.

Thank you for your support,


Thibaut VARENE
PA/Linux ESIEE Team
http://pateam.esiee.fr/

Re: [parisc-linux] 2.4.19 merge

[Alan Cox]

On Tue, 2002-07-16 at 00:30, Rafael E. Herrera wrote:
> Seems like the merge already begun:
> 
> http://linuxtoday.com/news_story.php3?ltsn=2002-07-15-015-26-NW-KN-DV

Thats the easy 90%. The stuff which doesn't overlap other platforms is
in there so I have a tree that I can point people back to if we get
breakage from later merges.

Some of it like the PCI changes is going to be hard.

Re: [parisc-linux] STI framebuffer and PINE

[Matthew Wilcox]

On Tue, Jul 16, 2002 at 09:29:12AM +0200, Matthieu Delahaye wrote:
> I don't think Matthew wath thinking about that. This HW list gives just 
> an indication if this is working or not.
> What I understand here is a list of something like:
> I currently use a B132 in production environment since November, 2001. 
> It provides a CVS server, apache with PHP and  a postgresql database. 
> The kernel is pretty old (2.4.14-32), but the only shutdown (and end of 
> services) I had to notice where caused by powerfailure.
> Use of Php is quite recent.
> Matthew, does these (real) example correspond to your suggestion?

Yep, that's perfect!

-- 
Revolutions do not require corporate support.

Re: [parisc-linux] 2.4.19 merge

[Randolph Chung]

> Thats the easy 90%. The stuff which doesn't overlap other platforms is
> in there so I have a tree that I can point people back to if we get
> breakage from later merges.

Alan, fwiw I'm playing with some arch-specific things that got missed in
the merge so that we can at least get a basic image that builds. Are 
you doing the same, or should we try to get something that will build
and feed that back to you? In any case I have some questions about some
pieces that I'll post here RSN.

> Some of it like the PCI changes is going to be hard.

What's the best way to proceed on this? I can try to merge Grant's
changes into the tree I'm working on and submit a working set (they
don't apply very cleanly directly on the rc1-ac5 tree)

randolph
--  
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/