Re: How to install official AMDGPU linux driver on Debian 11?

[Markos]

Em 21-10-2021 21:19, Linux-Fan escreveu:

Markos writes:


Em 17-10-2021 19:47, piorunz escreveu:

On 17/10/2021 22:27, Markos wrote:

Hi,

Please, could someone suggest a tutorial (for a basic user) on how to
install the driver for the graphics card for a laptop Lenovo IdeaPad
S145 with AMD Ryzen™ 5 3500U and AMD Radeon RX Vega 8 running 
Debian 11

(Bullseye).

I found a more complete tutorial just for Stretch and Buster:

https://wiki.debian.org/AMDGPUDriverOnStretchAndBuster2

What are the possible risks of problems using these AMD drivers?


[...]


No reply so far.

So, it seems that no one is interested in this question. :-(

Or none managed to do this installation, yet.


[...]

Before your initial post, there was already some discussion about a 
very similar case in the following thread:

https://lists.debian.org/debian-user/2021/10/msg00700.html

Summary: Just following AMDs instructions may lead to compile errors
(see https://lists.debian.org/debian-user/2021/10/msg00738.html)
whereas it worked for my GPU and downloaded driver:
(see https://lists.debian.org/debian-user/2021/10/msg00738.html)

I am interested in the questions of yours, but unfortunately cannot 
provide much of an assistance beyond what I already wrote in the other 
thread.


HTH
Linux-Fan

öö





Hi Linux-Fan,

Thank you for your comments.

Reading the messages I realized that this is a too big challenge for my 
level of knowledge.


And in the tutorial at 
https://linuxconfig.org/install-opencl-for-the-amdgpu-open-source-drivers-on-debian-and-ubuntu


I found an answer to my question about the risks of problems using these 
AMD drivers:


"WARNING: This may interfere with your existing drivers, resulting in 
poor performance and instability. Proceed with caution."


I even posted the same question on the AMD forum, but I haven't had an 
answer so far.


https://community.amd.com/t5/drivers-software/how-to-install-official-amdgpu-linux-driver-on-debian-11/td-p/493903

Thank you for your attention.

Best Regards,
Markos

Re: How to install official AMDGPU linux driver on Debian 11?

[piorunz]

On 22/10/2021 01:19, Linux-Fan wrote:


Before your initial post, there was already some discussion about a very
similar case in the following thread:
https://lists.debian.org/debian-user/2021/10/msg00700.html

Summary: Just following AMDs instructions may lead to compile errors
(see https://lists.debian.org/debian-user/2021/10/msg00738.html)
whereas it worked for my GPU and downloaded driver:
(see https://lists.debian.org/debian-user/2021/10/msg00738.html)

I am interested in the questions of yours, but unfortunately cannot
provide much of an assistance beyond what I already wrote in the other
thread.



Yes the other thread which I opened is just a few threads down. OpenCL
driver for modern Radeon 6800 XT which I have has 0% success rate
installing in Debian 11. I exhausted all options I found. You have
non-Radeon AMD card as I remember, probably driver is made by other team
at AMD. No idea. I invite anyone to have a try. Debian is disadvantaged
in this regard comparing to Ubuntu, which is supported by AMD. O_o

--
With kindest regards, Piotr.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

Re: How to install official AMDGPU linux driver on Debian 11?

[Linux-Fan]

Markos writes:


Em 17-10-2021 19:47, piorunz escreveu:

On 17/10/2021 22:27, Markos wrote:

Hi,

Please, could someone suggest a tutorial (for a basic user) on how to
install the driver for the graphics card for a laptop Lenovo IdeaPad
S145 with AMD Ryzen™ 5 3500U and AMD Radeon RX Vega 8 running Debian 11
(Bullseye).

I found a more complete tutorial just for Stretch and Buster:

https://wiki.debian.org/AMDGPUDriverOnStretchAndBuster2

What are the possible risks of problems using these AMD drivers?


[...]


No reply so far.

So, it seems that no one is interested in this question. :-(

Or none managed to do this installation, yet.


[...]

Before your initial post, there was already some discussion about a very  
similar case in the following thread:

https://lists.debian.org/debian-user/2021/10/msg00700.html

Summary: Just following AMDs instructions may lead to compile errors
(see https://lists.debian.org/debian-user/2021/10/msg00738.html)
whereas it worked for my GPU and downloaded driver:
(see https://lists.debian.org/debian-user/2021/10/msg00738.html)

I am interested in the questions of yours, but unfortunately cannot provide  
much of an assistance beyond what I already wrote in the other thread.


HTH
Linux-Fan

öö





pgpbWc1g5EV6Y.pgp
Description: PGP signature

Re: How to install official AMDGPU linux driver on Debian 11?

[piorunz]

On 21/10/2021 06:33, Felix Miata wrote:

Markos composed on 2021-10-21 00:29 (UTC-0300):


So, it seems that no one is interested in this question. :-(



Or none managed to do this installation, yet.


Could it be that readers here are content with the FOSS driver?


More likely readers here don't use/need OpenCL. FOSS driver doesn't
support that at all.

--
With kindest regards, Piotr.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

Re: How to install official AMDGPU linux driver on Debian 11?

[Felix Miata]

Markos composed on 2021-10-21 00:29 (UTC-0300):

> So, it seems that no one is interested in this question. :-(

> Or none managed to do this installation, yet. 
> 


Could it be that readers here are content with the FOSS driver?
-- 
Evolution as taught in public schools is, like religion,
based on faith, not based on science.

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata

Re: How to install official AMDGPU linux driver on Debian 11?

[Markos]

Em 17-10-2021 19:47, piorunz escreveu:

On 17/10/2021 22:27, Markos wrote:

Hi,

Please, could someone suggest a tutorial (for a basic user) on how to
install the driver for the graphics card for a laptop Lenovo IdeaPad
S145 with AMD Ryzen™ 5 3500U and AMD Radeon RX Vega 8 running Debian 11
(Bullseye).

I found a more complete tutorial just for Stretch and Buster:

https://wiki.debian.org/AMDGPUDriverOnStretchAndBuster2

What are the possible risks of problems using these AMD drivers?

Does using these AMD drivers make much difference in performance?

Thank you,

Markos



I'd love to know that too. I have Radeon 6900XT card and I have no idea
how to install proprietary OpenCL (which is part of proprietary driver)
on it. Other distros like Manjaro have this working (despite no official
AMD support like in Debian), but on Debian I haven't figured it out yet.

--
With kindest regards, Piotr.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄



Hi Piotr,

No reply so far.

So, it seems that no one is interested in this question. :-(

Or none managed to do this installation, yet.

Best Regards,
Markos

Re: How to install official AMDGPU linux driver on Debian 11?

[piorunz]

On 17/10/2021 22:27, Markos wrote:

Hi,

Please, could someone suggest a tutorial (for a basic user) on how to
install the driver for the graphics card for a laptop Lenovo IdeaPad
S145 with AMD Ryzen™ 5 3500U and AMD Radeon RX Vega 8 running Debian 11
(Bullseye).

I found a more complete tutorial just for Stretch and Buster:

https://wiki.debian.org/AMDGPUDriverOnStretchAndBuster2

What are the possible risks of problems using these AMD drivers?

Does using these AMD drivers make much difference in performance?

Thank you,

Markos



I'd love to know that too. I have Radeon 6900XT card and I have no idea
how to install proprietary OpenCL (which is part of proprietary driver)
on it. Other distros like Manjaro have this working (despite no official
AMD support like in Debian), but on Debian I haven't figured it out yet.

--
With kindest regards, Piotr.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

How to install official AMDGPU linux driver on Debian 11?

[Markos]

Hi,

Please, could someone suggest a tutorial (for a basic user) on how to 
install the driver for the graphics card for a laptop Lenovo IdeaPad 
S145 with AMD Ryzen™ 5 3500U and AMD Radeon RX Vega 8 running Debian 11 
(Bullseye).


I found a more complete tutorial just for Stretch and Buster:

https://wiki.debian.org/AMDGPUDriverOnStretchAndBuster2

What are the possible risks of problems using these AMD drivers?

Does using these AMD drivers make much difference in performance?

Thank you,

Markos

Re: USB "null modem" cables and related Linux driver questions

[rhkramer]

On Monday, June 18, 2018 08:44:20 AM Michael Stone wrote:
> He really doesn't want a sensible solution, just let it go.

+1

Re: USB "null modem" cables and related Linux driver questions

[Richard Owlett]

On 06/18/2018 07:17 AM, David wrote:

On 1 June 2018 at 00:21, Richard Owlett  wrote:


I have two computers with USB ports.
I wish them to communicate as simply as mid-20th-century computers did.


What is the make and model number of each computer?


No longer a relevant question.
I have purchased a cable based on the Prolific PL-25A1 chipset.
Current kernels can make it look like an Ethernet connection.
Once I finish some configuration homework {primary motivation has 
shifted to education} I'll be able to have a client-server setup.

Re: USB "null modem" cables and related Linux driver questions

[Michael Stone]

On Mon, Jun 18, 2018 at 10:17:21PM +1000, David wrote:

On 1 June 2018 at 00:21, Richard Owlett  wrote:


I have two computers with USB ports.
I wish them to communicate as simply as mid-20th-century computers did.


What is the make and model number of each computer?


He really doesn't want a sensible solution, just let it go.

Re: USB "null modem" cables and related Linux driver questions

[David]

On 1 June 2018 at 00:21, Richard Owlett  wrote:
>
> I have two computers with USB ports.
> I wish them to communicate as simply as mid-20th-century computers did.

What is the make and model number of each computer?

Re: USB "null modem" cables and related Linux driver questions

[Michael Stone]

On Thu, Jun 07, 2018 at 08:07:15AM +0100, Tixy wrote:

On Wed, 2018-06-06 at 22:26 -0600, Joe Pfeiffer wrote:

Richard Owlett  writes:

> I have two computers with USB ports.
> I wish them to communicate as simply as mid-20th-century computers
> did.
> Then we used RS232-C with a null modem &/or  appropriate software
> software at both ends.
>
> The underlying problem is that both ends egotistically expect to be
> *MASTER*.
>
> The hardware problem is solvable
> [e.g. http://www.ftdichip.com/Products/Cables/USBtoUSB.htm].

Given how FTDI does things, I'd be really surprised if this didn't
meet
your desires.  Have you actually tried it?  Are you sure there is no
driver in the kernel?


There is a driver in Linux because, from what the datasheet says, that
cable is two of FTDIs standard USB-to-serial chips wired together. I.e.
it's equivalent to getting 2 USB to serial cables and connecting them
with a null modem cable.


Yes, that's right. I have no idea why someone would want to do that, but 
that's exactly what it is.


There are also usb transfer cables which are basically a proprietary 
FIFO buffer with a usb master on both sides, which allow higher speed 
networking.  (Which at least seems more useful than going through a 
RS-232 conversion, but much more limited than an actual network. In some 
cases where you want more than gigabit speeds and will never have more 
than two computers it might be useful. I'm not sure how many of them can 
connect machines with different OSs, either. This is the "laplink" style 
connection.)


If you want a true, standardized, network session over your peripheral 
connection, then get firewire. (This is what not having a bus master 
allows.) It'll be old hardware, because it turns out that nobody 
actually wanted to pay more so that every device could be a master, and 
so firewire died. But once upon a time, this was one of its big 
advantages over usb.


Mike Stone

Re: USB "null modem" cables and related Linux driver questions

[Curt]

On 2018-06-07, Joe Pfeiffer  wrote:
>
> This depends on what you mean by "universal".  It was intended to be a
> protocol for computers to use to communicate with peripherals;
> "universal" in this context was restricted to peripherals.

And to the planet earth rather than all the way to the furthest reaches
of the cosmos and back as I doubt whether the folks living out by
Kepler-16 have much use for them.

Re: USB "null modem" cables and related Linux driver questions

[Tixy]

On Wed, 2018-06-06 at 22:26 -0600, Joe Pfeiffer wrote:
> Richard Owlett  writes:
> 
> > I have two computers with USB ports.
> > I wish them to communicate as simply as mid-20th-century computers
> > did.
> > Then we used RS232-C with a null modem &/or  appropriate software
> > software at both ends.
> > 
> > The underlying problem is that both ends egotistically expect to be
> > *MASTER*.
> > 
> > The hardware problem is solvable
> > [e.g. http://www.ftdichip.com/Products/Cables/USBtoUSB.htm].
> 
> Given how FTDI does things, I'd be really surprised if this didn't
> meet
> your desires.  Have you actually tried it?  Are you sure there is no
> driver in the kernel?

There is a driver in Linux because, from what the datasheet says, that
cable is two of FTDIs standard USB-to-serial chips wired together. I.e.
it's equivalent to getting 2 USB to serial cables and connecting them
with a null modem cable.

-- 
Tixy

Re: USB "null modem" cables and related Linux driver questions

[Joe Pfeiffer]

Richard Owlett  writes:

> On 06/01/2018 08:21 AM, to...@tuxteam.de wrote:
>> -BEGIN PGP SIGNED MESSAGE-
>> Hash: SHA1
>>
>> On Fri, Jun 01, 2018 at 08:23:42AM -0400, Stefan Monnier wrote:
 The one choice you have is that one of both sides takes a step
 back and plays "gadget" [...]
>>
>>> The gadget API is the programming API offered by the kernel for the OTG
>>> ports: no OTG => no gadget!
>>>
 [OTG]
>>
>>> More importantly, the USB ports which support OTG are driven by
>>> different hardware.
>>
>> Ah, so the hardware has to play along...
>>
>>> Right, you need both your hardware's USB port to support OTG and you
>>> need your kernel to have a driver that supports this hardware.
>>> AFAIK the driver is usually available.
>>
>> Did I say I was handwaving?
>>
>> Thanks for the clarifications!
>>
>
> It also suggests that I frequently grasp some of the implications of
> what I read. Thank you.
> As an illustration of my mindset:
> If they had really intended USB to be *UNIVERSAL* serial bus, then it
> should have been OTG from the get go.
> P.S. I know of thousands of reasons they did not.
>  Vast majority preceded by $ ;/

This depends on what you mean by "universal".  It was intended to be a
protocol for computers to use to communicate with peripherals;
"universal" in this context was restricted to peripherals.

Re: USB "null modem" cables and related Linux driver questions

[Joe Pfeiffer]

Richard Owlett  writes:

> I have two computers with USB ports.
> I wish them to communicate as simply as mid-20th-century computers did.
> Then we used RS232-C with a null modem &/or  appropriate software
> software at both ends.
>
> The underlying problem is that both ends egotistically expect to be
> *MASTER*.
>
> The hardware problem is solvable
> [e.g. http://www.ftdichip.com/Products/Cables/USBtoUSB.htm].

Given how FTDI does things, I'd be really surprised if this didn't meet
your desires.  Have you actually tried it?  Are you sure there is no
driver in the kernel?

As for your actual needs...  if you want two computers to communicate,
you really ought to be using some sort of networking hardware.  You've
got a computer with a USB port, no serial port, and no ethernet or wifi?

Re: USB "null modem" cables and related Linux driver questions

[David Wright]

On Fri 01 Jun 2018 at 09:08:53 (-0500), Richard Owlett wrote:
> On 06/01/2018 08:21 AM, to...@tuxteam.de wrote:
> >On Fri, Jun 01, 2018 at 08:23:42AM -0400, Stefan Monnier wrote:
> >>>The one choice you have is that one of both sides takes a step
> >>>back and plays "gadget" [...]
> >
> >>The gadget API is the programming API offered by the kernel for the OTG
> >>ports: no OTG => no gadget!
> >>
> >>>[OTG]
> >
> >>More importantly, the USB ports which support OTG are driven by
> >>different hardware.
> >
> >Ah, so the hardware has to play along...
> >
> >>Right, you need both your hardware's USB port to support OTG and you
> >>need your kernel to have a driver that supports this hardware.
> >>AFAIK the driver is usually available.
> >
> >Did I say I was handwaving?
> >
> >Thanks for the clarifications!
> >
> 
> It also suggests that I frequently grasp some of the implications of
> what I read. Thank you.
> As an illustration of my mindset:
> If they had really intended USB to be *UNIVERSAL* serial bus, then
> it should have been OTG from the get go.
> P.S. I know of thousands of reasons they did not.
>  Vast majority preceded by $ ;/

I think you're misunderstanding the use of the word universal.
USB was designed to be a universal way of connecting "any"
peripherals to a PC (sensu lato) which acts as a unique,
controlling host for them.

It wasn't designed to duplicate networking hardware, communicating
between multiple hosts. Nor was OTG. OTG was designed to allow, for
example, what's normally a peripheral to be disconnected from the
host, be connected to a peripheral and effectively become a host
controlling it. Eg, a camera could act as a peripheral while
uploading photos to a PC, then act as a host when connected to an
inkjet to print them.

Cheers,
David.

Re: USB "null modem" cables and related Linux driver questions

[Michael Stone]

On Fri, Jun 01, 2018 at 02:16:47PM -0500, Richard Owlett wrote:
USB-Ethernet dongles would be a lot more useful in the long term 
than USB-Serial dongles.


*WHO* said anything about a "USB-Serial dongle"?
I want a USB-USB object. Subtle, but important, distinction.


Well, you started out talking about RS232 and null modem cables, so 
don't be surprised if people are having trouble figuring out what you're 
asking for. Then you asked for better ideas. Then you got mad about all 
the ideas and started laughing hysterically. I have no idea what you 
actually want at this point.


RS-232 had a really good 40 year run, but for general communications 
it was mostly obsolete nearly 20 years ago. You mentioned file 
transfers-


So what?
I get grief for not mentioning immediate goals.
I casually mention a potential use and get *UNIVERSAL ADVICE*:
   Don't do that. !


So you can't articulate what you're trying to do at all? That will 
certainly reduce the odds of getting helpful advice.


It seems like you want to connect two computers via USB "just because". 
Good luck with that.


Mike Stone

Re: USB "null modem" cables and related Linux driver questions

[Andy Smith]

Richard,

On Fri, Jun 01, 2018 at 02:16:47PM -0500, Richard Owlett wrote:
> On 06/01/2018 09:01 AM, Michael Stone wrote:
> >I can't think of any applications where that's going to be better
> >over a pair of USB-Serial dongles than a pair of USB-Ethernet
> >dongles.
> 
> Please read my stated goals.
> Not my PREsumed goals.
> *ROFL*

Laugh all you want at people trying to help you, but time and time
again you end up posting volumes of text here without finding a
solution because you suck at concisely describing what you want to
achieve. Everyone else fails to understand you, every time, so is
everyone just stupid, or is it that the common factor here is you?

Your first email said you wanted your machines to "communicate".
That's as specific as you got.

Dan then tried to narrow it down to one specific form of
communication by spending a lot of their time listing out everything
they could think of that you might have meant, and you replied,
"Your list pretty much covers it." Useless and frustrating.

Then Stefan tried:

> > What kind of "communicate" do you need there?

Your response?

> Essentially any :/

I don't know what kind of response you expect to that uselessly
vague comment. By doing that you force people to spend a lot of
their time trying to cover every base, and then you complain at them
when they try. It really seems like you just want to complain when
people try their best to help you, even against your best efforts.

I expect to just get a barrel of whinging back for this, but I hate
to write a totally useless email myself, so…

You mentioned you want to transfer files.

If both machines have Ethernet interfaces then (as you've already
been advised) the simplest, most reliable and most performant way
will be direct Ethernet connection, or connection via a switch. If
file transfer is your goal, no need to bother with USB.

If somehow these machines don't both have Ethernet and USB really is
the only way, then again as you've already been advised, USB
Ethernet is the way to go.

We are now leaving the realms of simplicity and performance for more
subjective lands of doing things "just because" or for the retro
computing experience. You can certainly use a pair of USB serial adaptors
and then run a null modem cable between them, then run PPP over
that. Absolutely no idea why you would want to do this rather than
run Ethernet over USB. Instead of PPP you could use more arcane
forms of serial data transfer like zmodem or kermit.

If it's just a console you want, i.e. the console output and login
prompt from one machine to appear in a serial terminal program on
the other, then you can do that over Ethernet too! See "netconsole".

For a serial console when you don't have Ethernet (or want it
separate from the Ethernet infrastructure) then again it's USB
serial dongles on both ends, a null modem cable in between, then use
a serial terminal emulator like minicom or screen on the end you
want to be the client. There are configuration details involved for
the server end to get it to send console output to this USB-serial,
and put a login prompt on it.

Ask which parts you need more info on. Please be specific. Please
don't make this harder than it needs to be!

Cheers,
Andy

-- 
https://bitfolk.com/ -- No-nonsense VPS hosting

Re: USB "null modem" cables and related Linux driver questions

[Richard Owlett]

On 06/01/2018 09:01 AM, Michael Stone wrote:

On Fri, Jun 01, 2018 at 04:56:32AM -0500, Richard Owlett wrote:

On 05/31/2018 06:58 PM, David Wright wrote:

(thanks for your link) gives an idea of the price, and in this case I
can see some justification for it because they describe the
electronics hidden inside the plugs (we hope).

But considering that two NICs cost less than that cable, I'd need a
pressing reason to purchase the cable instead.


There is a perceived elegance aspect.
There is also a practical aspect the only known working and 

8>> conveniently physically accessible ports are USB.


USB-Ethernet dongles would be a lot more useful in the long term than 
USB-Serial dongles.


*WHO* said anything about a "USB-Serial dongle"?
I want a USB-USB object. Subtle, but important, distinction.



RS-232 had a really good 40 year run, but for 
general communications it was mostly obsolete nearly 20 years ago. You 
mentioned file transfers-


So what?
I get grief for not mentioning immediate goals.
I casually mention a potential use and get *UNIVERSAL ADVICE*:
Don't do that. !


-I can't think of any applications where
that's going to be better over a pair of USB-Serial dongles than a pair 
of USB-Ethernet dongles.


Please read my stated goals.
Not my PREsumed goals.
*ROFL*



Mike Stone

Re: USB "null modem" cables and related Linux driver questions

[Gene Heskett]

On Friday 01 June 2018 06:20:59 Richard Owlett wrote:

> On 05/31/2018 10:07 PM, Stefan Monnier wrote:
> >> I have two computers with USB ports.
> >> I wish them to communicate as simply as mid-20th-century computers
> >> did.
> >
> > What kind of "communicate" do you need there?
>
> Essentially any ;/
> In fact one of the thought experiments I was pursuing was how to do
> file copying/sharing over RS232 - even I would not actually attempt to
> implement.

Other than speed of the copy while the error correction verify's it 
sector by sector, why not. Both rsync and rzsz are quite capable of 
sending a file halfway around the planet with the last 17 miles on a 
barbed wire fence. And getting identical crc's in the final check.

For a circuit that was actually that dirty, I think I'd choose rzsz as 
its default packet size is 256 bytes.  If the crc of that packet fails, 
it requests a resend until it gets it right. So does rsync, but rsync's 
default packet is 64k, demanding a far cleaner path. rzsz unforch has 
several cousins 3x removed, so there is less than 100% compatibility. 
And the linux version is one of the worse compatibility violators.

> > The "way back machine" to simulate a "null modem" serial cable
> > exists, as you've seen, but it's rarely the best solution for
> > nowadays's needs,
>
> "Best" is not an invariant absolute.
>
> > since nowadays connecting two computers is something completely
> > normal, supported by a deluge of tools, but they all expect a
> > "network" connection rather than a serial cable.
> >
> > In most cases those two computers also have ethernet or wifi "ports"
> > so you can connect them via such a network (which usually offers
> > faster transmission than a serial cable, lets you seamlessly
> > multiplex several connections, and lets you use the many tools
> > working over the network to connect computers).
>
> An explicit requirement is a wired, NOT WiFi, connection.
> I that seriously. My internet access is a WiFi hotspot with its WiFi
> capability disabled.
>
> > In some cases one of the two computers's USB port is an "OTG" port,
> > meaning that it can act either as "master" or not, in which case you
> > can just use a regular USB cable (and usually you then configure the
> > OTG side to pretend it's a network card, so it ends up looking to
> > the software like you've connected the two machines via an ethernet
> > cable. That's what I use between my BananaPi "router" and my office
> > desktop).
>
> I saw it and it meets most (all?) my requirements except my reading
> suggested:
>1. obsolete
>2. available only for Windows/Mac
>3. no way to determine if any of machines were equipped
>
> > If none of that are options, you can resort to using an "ethernet
> > dongle" on both sides and an ethernet cable between the two.
>
> That's a 1 versus 3 items required per connection.
>
> > All of those things will typically work "out of the box" on a
> > vanilla Linux kernel (the usbnet drivers have been incorporated
> > years ago).
> >
> > Oh, and in case those computers are somewhat old, they may also come
> > with Firewire ports, and those (contrary to USB) don't have the
> > "slave/master" distinction so you can connect your computers this
> > way with a plain normal Firewire cable (and make it appear to the
> > software, again, as some kind of ethernet-like connection).
> >
> >
> >  Stefan



-- 
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

Re: USB "null modem" cables and related Linux driver questions

[Joe]

On Fri, 1 Jun 2018 10:01:57 -0400
Michael Stone  wrote:

> On Fri, Jun 01, 2018 at 04:56:32AM -0500, Richard Owlett wrote:
> >On 05/31/2018 06:58 PM, David Wright wrote:  
> >>(thanks for your link) gives an idea of the price, and in this case
> >>I can see some justification for it because they describe the
> >>electronics hidden inside the plugs (we hope).
> >>
> >>But considering that two NICs cost less than that cable, I'd need a
> >>pressing reason to purchase the cable instead.  
> >
> >There is a perceived elegance aspect.
> >There is also a practical aspect the only known working and 
> >conveniently physically accessible ports are USB.  
> 
> USB-Ethernet dongles would be a lot more useful in the long term than 
> USB-Serial dongles. RS-232 had a really good 40 year run, but for 
> general communications it was mostly obsolete nearly 20 years ago.
> You mentioned file transfers--I can't think of any applications where
> that's going to be better over a pair of USB-Serial dongles than a
> pair of USB-Ethernet dongles.
> 
>
Radio? This gadget:

https://www.lensadaptor.com/mtf-effect-control-unit-3-kit

 uses straight point-to-point 8-bit serial over Zigbee type RF
tranceivers. Bluetooth is basically 8-bit serial over radio, though
with a very short range.

-- 
Joe
https://en.wikipedia.org/wiki/Zigbee

Re: USB "null modem" cables and related Linux driver questions

[David Wright]

On Fri 01 Jun 2018 at 05:26:01 (-0500), Richard Owlett wrote:
> On 06/01/2018 01:27 AM, deloptes wrote:
> >Richard Owlett wrote:
> >
> >>I have two computers with USB ports.
> >>I wish them to communicate as simply as mid-20th-century computers did.
> >>Then we used RS232-C with a null modem &/or  appropriate software
> >>software at both ends.
> >>
> >
> >J., why not take a crossover cable - all pcs have now ethernet port
> >
> >https://en.wikipedia.org/wiki/Ethernet_crossover_cable
> >
> >
> 
> Some combinations require crossover cable, some don't.
> I investigated that and wen so fa as to purchase a {still shrink
> wrapped} 8-port switch as a solution. Decided clutter not worth it.

Just forget anyone said "crossover cable¹". You'd have to be making a
direct connection between two museum pieces to require one, and as
you have a switch, you have absolutely no need to ever do that.
Coonect both to the switch instead.

Accessibility? Using the NIC (usually at the back), it gets inserted
permanently and then forgotten, rather than needing to be inserted
on each occasion you want connectivity. Tidier too.

Bear in mind you're using the hardware as intended: no risk of
blowing up your USB port. (Ever noticed how often flaky USB ports
are mentioned here?)

You might even find you can use wakeonlan on the desktop PCs
to save some power.

¹ If you *have* obtained one, just use it like an ordinary cable nowadays.

Cheers,
David.

Re: USB "null modem" cables and related Linux driver questions

[Richard Owlett]

On 06/01/2018 08:21 AM, to...@tuxteam.de wrote:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Fri, Jun 01, 2018 at 08:23:42AM -0400, Stefan Monnier wrote:

The one choice you have is that one of both sides takes a step
back and plays "gadget" [...]



The gadget API is the programming API offered by the kernel for the OTG
ports: no OTG => no gadget!


[OTG]



More importantly, the USB ports which support OTG are driven by
different hardware.


Ah, so the hardware has to play along...


Right, you need both your hardware's USB port to support OTG and you
need your kernel to have a driver that supports this hardware.
AFAIK the driver is usually available.


Did I say I was handwaving?

Thanks for the clarifications!



It also suggests that I frequently grasp some of the implications of 
what I read. Thank you.

As an illustration of my mindset:
If they had really intended USB to be *UNIVERSAL* serial bus, then it 
should have been OTG from the get go.

P.S. I know of thousands of reasons they did not.
 Vast majority preceded by $ ;/

Re: USB "null modem" cables and related Linux driver questions

[Michael Stone]

On Fri, Jun 01, 2018 at 04:56:32AM -0500, Richard Owlett wrote:

On 05/31/2018 06:58 PM, David Wright wrote:

(thanks for your link) gives an idea of the price, and in this case I
can see some justification for it because they describe the
electronics hidden inside the plugs (we hope).

But considering that two NICs cost less than that cable, I'd need a
pressing reason to purchase the cable instead.


There is a perceived elegance aspect.
There is also a practical aspect the only known working and 
conveniently physically accessible ports are USB.


USB-Ethernet dongles would be a lot more useful in the long term than 
USB-Serial dongles. RS-232 had a really good 40 year run, but for 
general communications it was mostly obsolete nearly 20 years ago. You 
mentioned file transfers--I can't think of any applications where
that's going to be better over a pair of USB-Serial dongles than a pair 
of USB-Ethernet dongles.


Mike Stone

Re: USB "null modem" cables and related Linux driver questions

[tomas]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Fri, Jun 01, 2018 at 08:23:42AM -0400, Stefan Monnier wrote:
> > The one choice you have is that one of both sides takes a step
> > back and plays "gadget" [...]

> The gadget API is the programming API offered by the kernel for the OTG
> ports: no OTG => no gadget!
> 
> > [OTG]

> More importantly, the USB ports which support OTG are driven by
> different hardware.

Ah, so the hardware has to play along...

> Right, you need both your hardware's USB port to support OTG and you
> need your kernel to have a driver that supports this hardware.
> AFAIK the driver is usually available.

Did I say I was handwaving?

Thanks for the clarifications!

Cheers
- -- tomás
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlsRSHYACgkQBcgs9XrR2kYP9ACfb7mVtI1Qu/CjDeTIB5uoyOPv
jNYAni1/EbprNJ8Gu6DuHMfyWoO+tCGP
=dhmw
-END PGP SIGNATURE-

Re: USB "null modem" cables and related Linux driver questions

[Stefan Monnier]

> The one choice you have is that one of both sides takes a step
> back and plays "gadget" (the jargon term, somewhat unfortunate
> as search engine fodder). There seems to be something out there
> for that, e.g. [2].

The gadget API is the programming API offered by the kernel for the OTG
ports: no OTG => no gadget!

> The other choice seems to be USB On The Go (aka "OTG") [3].
> You seem to need a special cable for that.

More importantly, the USB ports which support OTG are driven by
different hardware.

> There seem to be Linux drivers to let the USB "stack" play along
> with OTG [4], but I have no experience whatsoever with this

Right, you need both your hardware's USB port to support OTG and you
need your kernel to have a driver that supports this hardware.
AFAIK the driver is usually available.


Stefan

Re: USB "null modem" cables and related Linux driver questions

[Stefan Monnier]

>> In some cases one of the two computers's USB port is an "OTG" port,
>> meaning that it can act either as "master" or not, in which case you can
>> just use a regular USB cable (and usually you then configure the OTG
>> side to pretend it's a network card, so it ends up looking to the
>> software like you've connected the two machines via an ethernet cable.
>> That's what I use between my BananaPi "router" and my office desktop).
>
> I saw it and it meets most (all?) my requirements except my reading
> suggested:
>   1. obsolete
>   2. available only for Windows/Mac
>   3. no way to determine if any of machines were equipped

Not obsolete at all.
Availability has nothing to do with the OS (I never use either of macOS
or Windows).

But yes, OTG rare (read: non-existing) on laptops/desktops.
They're very common on "embedded systems" which are expected to be
(occasionally) plugged into a computer.  E.g. many phones's USB port
is OTG.

>> If none of that are options, you can resort to using an "ethernet
>> dongle" on both sides and an ethernet cable between the two.
> That's a 1 versus 3 items required per connection.

I just gave you alternatives.  You'll let you decide what's better for
your planned use cases.


Stefan

Re: USB "null modem" cables and related Linux driver questions

[Richard Owlett]

On 06/01/2018 01:27 AM, deloptes wrote:

Richard Owlett wrote:


I have two computers with USB ports.
I wish them to communicate as simply as mid-20th-century computers did.
Then we used RS232-C with a null modem &/or  appropriate software
software at both ends.



J., why not take a crossover cable - all pcs have now ethernet port

https://en.wikipedia.org/wiki/Ethernet_crossover_cable




Some combinations require crossover cable, some don't.
I investigated that and wen so fa as to purchase a {still shrink 
wrapped} 8-port switch as a solution. Decided clutter not worth it.

Re: USB "null modem" cables and related Linux driver questions

[Richard Owlett]

On 05/31/2018 10:07 PM, Stefan Monnier wrote:

I have two computers with USB ports.
I wish them to communicate as simply as mid-20th-century computers did.


What kind of "communicate" do you need there?


Essentially any ;/
In fact one of the thought experiments I was pursuing was how to do file 
copying/sharing over RS232 - even I would not actually attempt to implement.




The "way back machine" to simulate a "null modem" serial cable exists,
as you've seen, but it's rarely the best solution for nowadays's needs,


"Best" is not an invariant absolute.


since nowadays connecting two computers is something completely normal,
supported by a deluge of tools, but they all expect a "network"
connection rather than a serial cable.

In most cases those two computers also have ethernet or wifi "ports" so
you can connect them via such a network (which usually offers faster
transmission than a serial cable, lets you seamlessly multiplex several
connections, and lets you use the many tools working over the network to
connect computers).


An explicit requirement is a wired, NOT WiFi, connection.
I that seriously. My internet access is a WiFi hotspot with its WiFi 
capability disabled.




In some cases one of the two computers's USB port is an "OTG" port,
meaning that it can act either as "master" or not, in which case you can
just use a regular USB cable (and usually you then configure the OTG
side to pretend it's a network card, so it ends up looking to the
software like you've connected the two machines via an ethernet cable.
That's what I use between my BananaPi "router" and my office desktop).


I saw it and it meets most (all?) my requirements except my reading 
suggested:

  1. obsolete
  2. available only for Windows/Mac
  3. no way to determine if any of machines were equipped



If none of that are options, you can resort to using an "ethernet
dongle" on both sides and an ethernet cable between the two.


That's a 1 versus 3 items required per connection.



All of those things will typically work "out of the box" on a vanilla
Linux kernel (the usbnet drivers have been incorporated years ago).

Oh, and in case those computers are somewhat old, they may also come
with Firewire ports, and those (contrary to USB) don't have the
"slave/master" distinction so you can connect your computers this way
with a plain normal Firewire cable (and make it appear to the software,
again, as some kind of ethernet-like connection).


 Stefan

Re: USB "null modem" cables and related Linux driver questions

[tomas]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Fri, Jun 01, 2018 at 08:27:13AM +0200, deloptes wrote:
> Richard Owlett wrote:
> 
> > I have two computers with USB ports.
> > I wish them to communicate as simply as mid-20th-century computers did.
> > Then we used RS232-C with a null modem &/or  appropriate software
> > software at both ends.
> > 
> 
> J., why not take a crossover cable - all pcs have now ethernet port

FWIW these days, most devices are capable of doing the crossover thing
automatically [1], so you can just use a straight cable. From personal
experience, I haven't needed a crossover cable the last ten years (and
I do muck around with oldish hardware).

Cheers

[1] https://en.wikipedia.org/wiki/Medium_Dependent_Interface#Auto_MDI-X

- -- tomás
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlsRAmAACgkQBcgs9XrR2kbpPwCfYH3wQicOhfiQrgAvcDUPytCZ
29YAn0sYNXgIXIFfMDq3tt5wCJUKM8DK
=tkoi
-END PGP SIGNATURE-

Re: USB "null modem" cables and related Linux driver questions

[deloptes]

Richard Owlett wrote:

> I have two computers with USB ports.
> I wish them to communicate as simply as mid-20th-century computers did.
> Then we used RS232-C with a null modem &/or  appropriate software
> software at both ends.
> 

J., why not take a crossover cable - all pcs have now ethernet port

https://en.wikipedia.org/wiki/Ethernet_crossover_cable

Re: USB "null modem" cables and related Linux driver questions

[Stefan Monnier]

> If none of that are options, you can resort to using an "ethernet
> dongle" on both sides and an ethernet cable between the two.

[ If one of the two computers has a free ethernet port, you can of
  course also such a dongle on the other computer.  ]

BTW, those ethernet dongles can be found pretty cheaply (like $10 or
less) and they can be handy in many cases to add an ethernet port (or
several ports, even).  IOW, it's a good tool to have in your box.


Stefan

Re: USB "null modem" cables and related Linux driver questions

[Stefan Monnier]

> I have two computers with USB ports.
> I wish them to communicate as simply as mid-20th-century computers did.

What kind of "communicate" do you need there?

The "way back machine" to simulate a "null modem" serial cable exists,
as you've seen, but it's rarely the best solution for nowadays's needs,
since nowadays connecting two computers is something completely normal,
supported by a deluge of tools, but they all expect a "network"
connection rather than a serial cable.

In most cases those two computers also have ethernet or wifi "ports" so
you can connect them via such a network (which usually offers faster
transmission than a serial cable, lets you seamlessly multiplex several
connections, and lets you use the many tools working over the network to
connect computers).

In some cases one of the two computers's USB port is an "OTG" port,
meaning that it can act either as "master" or not, in which case you can
just use a regular USB cable (and usually you then configure the OTG
side to pretend it's a network card, so it ends up looking to the
software like you've connected the two machines via an ethernet cable.
That's what I use between my BananaPi "router" and my office desktop).

If none of that are options, you can resort to using an "ethernet
dongle" on both sides and an ethernet cable between the two.

All of those things will typically work "out of the box" on a vanilla
Linux kernel (the usbnet drivers have been incorporated years ago).

Oh, and in case those computers are somewhat old, they may also come
with Firewire ports, and those (contrary to USB) don't have the
"slave/master" distinction so you can connect your computers this way
with a plain normal Firewire cable (and make it appear to the software,
again, as some kind of ethernet-like connection).


Stefan

Re: USB "null modem" cables and related Linux driver questions

[Dan Ritter]

On Thu, May 31, 2018 at 09:21:27AM -0500, Richard Owlett wrote:
> I have two computers with USB ports.
> I wish them to communicate as simply as mid-20th-century computers did.
> Then we used RS232-C with a null modem &/or  appropriate software software
> at both ends.
> 
> The underlying problem is that both ends egotistically expect to be
> *MASTER*.
> 
> The hardware problem is solvable [e.g.
> http://www.ftdichip.com/Products/Cables/USBtoUSB.htm].
> 
> The software is another case :<
> The best Linux specific link I've found is dated September 2005
> [http://www.linux-usb.org/usbnet/]
> 
> I've done DuckDuckGo searches for permutations/combinations of
>   usb, "peer to peer", lan, and bridge.
> Most is Windows/Mac centric.
> There were hints that needed drivers may exist in current Linux core.
> 
> Pointers to good references and/or better search terms?

Do you want:

- One end is a terminal, the other end offers login?   "agetty
  serial" and "minicom"

- TCP/IP over RS-232 serial?   Keyword is "PPP".

- TCP/IP over ethernet attached via USB? "usbnet" and treat them
  like ethernet adapters.

- Connect an actual hardware terminal like a VT320 to your Linux
  box? "agetty serial" again.

- Something else? Describe it.

-dsr-

Re: USB "null modem" cables and related Linux driver questions

[Reco]

Hi.

On Thu, May 31, 2018 at 09:21:27AM -0500, Richard Owlett wrote:
> I have two computers with USB ports.
> I wish them to communicate as simply as mid-20th-century computers did.
> Then we used RS232-C with a null modem &/or  appropriate software software
> at both ends.
> 
> The underlying problem is that both ends egotistically expect to be
> *MASTER*.

I did the thing in the not-so-distant past.

It took two USB-RS232 converters (FDTI chipsets should work out of the
box), and a conventional RS232 null modem cable. Communication was
limited to 112500 bps, but I needed it for the serial console anyway.

As for the software part, agetty and minicom were all that was needed.

Reco

USB "null modem" cables and related Linux driver questions

[Richard Owlett]

I have two computers with USB ports.
I wish them to communicate as simply as mid-20th-century computers did.
Then we used RS232-C with a null modem &/or  appropriate software 
software at both ends.


The underlying problem is that both ends egotistically expect to be 
*MASTER*.


The hardware problem is solvable [e.g. 
http://www.ftdichip.com/Products/Cables/USBtoUSB.htm].


The software is another case :<
The best Linux specific link I've found is dated September 2005 
[http://www.linux-usb.org/usbnet/]


I've done DuckDuckGo searches for permutations/combinations of
  usb, "peer to peer", lan, and bridge.
Most is Windows/Mac centric.
There were hints that needed drivers may exist in current Linux core.

Pointers to good references and/or better search terms?
TIA

Re: Samsung CLX4195SN and the Samsung Unified Linux Driver on Debian

[Paul Cartwright]

On 01/14/2014 06:03 PM, Lisi Reisz wrote:
> Debian Wheezy, stock kernel, ULD tarball downloaded from the Samsung 
> website and compiled.
>
> I am getting the following error message:
> 
> The components listed below are necessary for proper Unified Linux 
> Driver operation. Click Cancel now, install these components from 
> your Linux distribution CD-ROM or from Linux vendor Web site and then 
> run Unified Linux Driver Installer again.
> You may click Install Anyway to continue installation, but the result 
> would be unpredictable.
> Please click Help for explanations.
>
> - SANE - "Scanner Access Now Easy" - scanner API
> 
I have a Samsung CLP-315W and... I got the same error message. I ended
up removing the smartpanel, because it won't work on a printer that is
NOT directly attached ( mine is wireless).  I used the
http://localhost:631/admin/ administration method.

-- 
Paul Cartwright
Registered Linux User #367800 and new counter #561587


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/52d66ec7.5060...@gmail.com

Samsung CLX4195SN and the Samsung Unified Linux Driver on Debian

[Lisi Reisz]

Debian Wheezy, stock kernel, ULD tarball downloaded from the Samsung 
website and compiled.

I am getting the following error message:

The components listed below are necessary for proper Unified Linux 
Driver operation. Click Cancel now, install these components from 
your Linux distribution CD-ROM or from Linux vendor Web site and then 
run Unified Linux Driver Installer again.
You may click Install Anyway to continue installation, but the result 
would be unpredictable.
Please click Help for explanations.

- SANE - "Scanner Access Now Easy" - scanner API


I am told that this may mean one of two things:
1) It is the wrong version of the driver package for the version of 
Debian that I am running.
or
2) the SANE files it needs are not in the location where it is looking 
for them, and need copying or moving so that the files are where the 
driver package is looking for them.  They recommend copying.

Does anyone know, or is anyone prepared to try to guess, where the 
driver may look?  The chap I was talking to seemed to be suggesting 
that the driver package would look in /etc/bin/, but I have 
no /etc/bin.  I could of course create one.  Do people think that 
that is what I ought to do?

I have googled.  I have found the error message, but so far not found 
a solution.  One hit suggested that it might need a 2.x series 
kernel.  I hope that that is not the solution!  Other than that I 
found nothing constructive.

Thanks,
Lisi


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201401142303.27265.lisi.re...@gmail.com

Re: Linux Driver For PostgreSQL .......

[Mitchell Laks]

On Thursday 23 June 2005 05:33 pm, Redefined Horizons wrote:
> With a Java client application you can use JDBC.
> With a Microsoft Application you can use ODBC.
>
> I would like to write a client application in GNOME witha GTK GUI.
> What driver would I use in this case?
>
> Thanks,
>
> Scott  Huey

i have done this. I

would use perl with its database interface via perl-dbi and dbd-pg for the
postgresql interface. and i would use gtk2-perl for the gui.

this is the easiest way to do it.

also then you get a web interface for free: you can use  your backend database 
interaction code to created a web interface with cgi.pm as well 

and your backend code would be mostly valid for mysql or oracle as well via 
the dbi perl interface.

mitchell

---


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Linux Driver For PostgreSQL .......

[Lee Braiden]

On Friday 24 Jun 2005 11:56, Gregory Seidman wrote:
> In addition, the Qt database API seems to be pretty popular. It looks like
> you're planning on a GNOME app, but you might try KDE instead.

Yes, especially in combination with KDevelop and QT-Designer, which will 
automatically generate forms etc. for you.

Of course, if it's a basic DB app, something like KNoda or Kexi might be more 
appropriate.

-- 
Lee.

Please do not CC replies directly to me.  I'll read them on the list.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Linux Driver For PostgreSQL .......

[Gregory Seidman]

On Thu, Jun 23, 2005 at 10:46:03PM +0100, Fred L Youhanaie wrote:
} Redefined Horizons wrote:
} >With a Java client application you can use JDBC.
} >With a Microsoft Application you can use ODBC.
} >
} >I would like to write a client application in GNOME witha GTK GUI.
} >What driver would I use in this case?
} 
} I haven't tried this myself, Is unixODBC from http://www.odbc.org/ of 
} any use?

It is, indeed. I have used it with great success with both PostgreSQL and
DB2. It takes a little bit of work to configure it, and you need to know
the ODBC API (the Microsoft documentation is the best available; they did
invent it, after all).

In addition, the Qt database API seems to be pretty popular. It looks like
you're planning on a GNOME app, but you might try KDE instead.

} HTH
} Cheers
} f.
--Greg


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Linux Driver For PostgreSQL .......

[Fred L Youhanaie]

Redefined Horizons wrote:

With a Java client application you can use JDBC.
With a Microsoft Application you can use ODBC.

I would like to write a client application in GNOME witha GTK GUI.
What driver would I use in this case?


I haven't tried this myself, Is unixODBC from http://www.odbc.org/ of 
any use?


HTH

Cheers
f.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Linux Driver For PostgreSQL .......

[Lee Braiden]

On Thursday 23 Jun 2005 22:33, Redefined Horizons wrote:
> With a Java client application you can use JDBC.
> With a Microsoft Application you can use ODBC.

These are database abstraction layers.

> I would like to write a client application in GNOME witha GTK GUI.
> What driver would I use in this case?

GNOME's DB abstraction layer is called GNOMEDB.  Your preferred programming 
language may also have various options available.  Python includes a DB 
abstraction layer called DB-API, for instance, while for C++ there are 
options like libsqlxx, and Perl has Perl-DBI.

-- 
Lee.

Please do not CC replies directly to me.  I'll read them on the list.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Linux Driver For PostgreSQL .......

[Redefined Horizons]

With a Java client application you can use JDBC.
With a Microsoft Application you can use ODBC.

I would like to write a client application in GNOME witha GTK GUI.
What driver would I use in this case?

Thanks,

Scott  Huey

ATI Mobility Radeon 9600 4X AGP / XFree86 4.3.0 / ATI Linux Driver: Colour Depth Problems

[Martin HEIN]

Hello,

I'm running on my HP nc6000 (with an external hp L1925 TFT-Monitor) Debian GNU/Linux 3.0r3, XFree86 4.3.0, the prorpietary ATI Radeon Linux driver, and as display manager GDM; XFree is configured to use 1280x1024 24bpp externally, and 1024x768 24bpp internally.

After booting the system GDM appears just as intended (and configured) to, but after approximately two seconds the systems reduces the color depth from 24bpp to 8bpp, which admittedly looks very crude (or artish, at least from a certain point of view ;-). When I login an run fgl_glgears, the color depth is restored to 24bpp.

Any ideas how to prevent the depth reduction and keep the original, configured depth?

Merry Xmas, ;-)
Martin

--
T-Mobile Austria GmbH
Ing.Martin HEIN
Information Technologies / Services
Knowledge Management & Process Automation / Senior Manager

Rennweg 97-99
A-1030 Vienna
Fon: +43-1-79585-6337
Fax: +43-1-79585-6584
Mobile: +43-676-82006337
E-Mail: [EMAIL PROTECTED]
Web: http://www.T-Mobile.at
--
NOTICE AND DISCLAIMER: 
This email (including attachments) is confidential.  If you have received this email in error please notify the sender immediately and delete this email from your system without copying or disseminating it or placing any reliance upon its contents.
--

PGP.sig
Description: This is a digitally signed message part

linux driver

[sridhar]

Hi

I am new to linux device drivers.

i have written a small driver with init_module and cleanup_module
compiled it using gcc -c option.
tried to load it using insmod but says that it is compiled with
kernel 2.4.20 but the current version is 2.4.20-8

how do i make it to compile it for this version.

Thanks in advance
sridhar.





___
No banners. No pop-ups. No kidding.
Make My Way your home on the Web - http://www.myway.com


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Argosy HD-530 USB2/FW Linux Driver.

[Felipe Zottola Diz]

The Argosy USB 2.0 / Firewire 2.5" Hard Disk Interface
is wholly usable in Linux.

I am running RedHat 9.0, and issuing 'uname -r' on my
linux box, it returns a kernel version 2.4.20-8... yeah!
I know... that's nothing out of this world!

To make things work nicely, there are a two files to
look at, which are:

The first one:

/etc/fstab

That probably already shows you the device you want to mount, but if not,
just read ahead. We'll return to this in a while, to make /etc/fstab look
how it should.

The other one:

/etc/sysconfig/hwconf

Take a special attention to sections describing your USB hardware.

To make all simpler here's my file section which applies to the case:

...

-
class: HD
bus: SCSI
detached: 0
device: sda
driver: ignore
desc: "Ibm-djsa -220"
host: 0
id: 0
channel: 0
lun: 0
generic: sg0
-

...

And guess what... the description field (desc:) shows the same label
Windows XP gives to your device (with the 2.5 HDD inside, of course).

So, try a text search inside your /etc/sysconfig/hwconf, using part
of the label. Don't forget using lower/upper case (ask your text editor
not to perform a case sensitive search).

When you have it, just use the label matching the "device" field, as in
my case "sda".

What we only need to know is which particular partition on your enclosed
disk you want to mount to. If you have only one (what it might be the case),
you should specify sda1 (or sda2, sda3, sdb1, sdc3, etc). Think as it were:
sd all together.

Now we are ready for the magic:

'mount -t vfat /dev/sda1 /mnt/other'

OOOpps!!

"vfat" & "other"???

vfat: is the filesystem type of your partition, or better said, you should
put your filesystem type instead of vfat, but as my 2.5" HDD was formatted
in FAT32 filesystem then I've used vfat, which that's how linux names FATxx
filesystems. Note that standard linux distributions do NOT manage NTFS
filesystems, unless you made some tweaks.

other: that should be the famous mount point (a folder created to use it as
a known place to map your files). So choose a nice name to map your files
and create a folder on a nice place. Look what's inside /mnt, which
typically
will be:

.
..
floppy
cdrom

So issue a 'mkdir mypreferredfoldername'

and what you'll have is:

.
..
floppy
cdrom
mypreferredfoldername

Finally, the command will be:

'mount -t vfat /dev/sda1 /mnt/mypreferredfoldername'

Then take a look to what's inside '/mnt/myblahblah...'

There you go! (Hopefully).

Now, add this line to your forgotten /etc/fstab file:

/dev/sda1/mnt/myblah  vfatnoauto,owner,users 0 0

Try to ident the text strings, just to make it look neat and clean.
It might help to figure out the details.

The next time what you only need is to attach your device and
mount using an abreviated form of mount:

mount /mnt/myblah

Sorry for being too boring, and hope this helped something!

Bye ;->

Felipe Zottola Diz.
Spain.



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

ATI Linux driver on kernel 2.6.0-testX, VIA KT400 Chipset

[Nicos Gollan]

I think this fits the thread, so I'll post it as a reply.

If you want to use the most recent ATI drivers on a 2.6.0-test7 kernel and/or 
with a motherboard using the VIA KT400 chipset, some changes are needed. 
You'll find a patch on the forums at www.rage3d.com (search for the Linux 
messageboard, it's a sticky thread). This should make it work.

-- 
Got Backup?


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Driver Download and Installation ATI Proprietary Linux Driver 3.2.8 Download--for those interested

[Roberto Sanchez]

Harshwardhan Nagaonkar wrote:
Rob Sims wrote:

On Thursday 09 October 2003 07:16 pm, Haralambos Geortgilakis wrote:


KDE works when I use the "radeon" driver, but I get no DRI.

Where do I go from here?  I didn't find a user forum for the ATI 
drivers, just a no-response bug reporting link.


Hmm... well I'm glad you were fortunate enough to get the kernel module 
working on a 2.6 kernel (aren't the ATI prop. modules designed for 2.2.x 
and specifically 2.4.x... atleast the Release Notes a.k.a Readme says 
that). Good backward compatibility in 2.6 it *seems*.

ATi recently (on October 8) released their new drivers for windows and
Linux.  The README for the LInux driver claims that they successfully
built it against a 2.6.0-test6 kernel.
On a sort of offtopic note, while I am generally against propritery
software, especially device drivers, I think it is great that they have
taken the initiative to support the 2.6 series as it just coming out.
-Roberto


pgp0.pgp
Description: PGP signature

Re: Driver Download and Installation ATI Proprietary Linux Driver 3.2.8 Download--for those interested

[Tom]

On Mon, Oct 13, 2003 at 03:37:06PM -0600, Harshwardhan Nagaonkar wrote:
> Tom wrote:
> >On Mon, Oct 13, 2003 at 02:48:15PM -0600, Harshwardhan Nagaonkar wrote:
> >
> >>Rob Sims wrote:
> >>
> >>>On Thursday 09 October 2003 07:16 pm, Haralambos Geortgilakis wrote:
> 
> >You can get the latest "unreleased" ATI drivers at:
> >http://www.schneider-digital.de/html/download_ati.html
> 
> Cool. However, ATI has the drivers version 3.2.8 on their site, while 
> Schneider has 3.0.2-x version. I usually tend to go with the 
> manufacturer on this sort of version number difference. Is there some 
> reason to why the version numbers are less on schneider-digital's site? 
> TIA for that.

Two points: one, this guys website is inconsistent.  Those are the 3.2.5 
drivers (check the readme) regardless of what the HTML says.

Second, couple months back, this guy had later drivers than ATI's site 
did (in particular support for 2.6).  Sounds like the official drivers 
passed him!

For most of the past year that site had later drivers than the official.  
Anyway, I found them with google: "linux and fglrx".  If you google 
"debian fglrx" you'll find a guy in Italy who's packaged them for 
Debian.  You can hack together his scripts to package these as well.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Driver Download and Installation ATI Proprietary Linux Driver 3.2.8 Download--for those interested

[Harshwardhan Nagaonkar]

Tom wrote:
On Mon, Oct 13, 2003 at 02:48:15PM -0600, Harshwardhan Nagaonkar wrote:

Rob Sims wrote:

On Thursday 09 October 2003 07:16 pm, Haralambos Geortgilakis wrote:

You can get the latest "unreleased" ATI drivers at:
http://www.schneider-digital.de/html/download_ati.html
Cool. However, ATI has the drivers version 3.2.8 on their site, while 
Schneider has 3.0.2-x version. I usually tend to go with the 
manufacturer on this sort of version number difference. Is there some 
reason to why the version numbers are less on schneider-digital's site? 
TIA for that.

Propriatary drivers for ATI don't bug me -- I mean, if the knowledge
of how to make killer drivers was "in the general community", superfast
graphics boards would be a dime a dozen.  It is qualitatively different
from netcards, hd controllers, even sound, cause it's so bleeding edge.
Oh, I am sorry if my mail seemed like I didn't like proprietary drivers 
or anything. I myself use the ATI proprietary drivers since they provide 
better 3d graphics speed.

Also, the point being with closed-source drivers is that even *if* 
somebody could tweak the drivers to get more speed, they are not able to 
 since they don't have the specifications to the card and/or the source 
to the original manufacturers drivers.

--
Harshwardhan Nagaonkar
Electrical Engineering Sysop
Brigham Young University, UT-84602
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Driver Download and Installation ATI Proprietary Linux Driver 3.2.8 Download--for those interested

[Tom]

On Mon, Oct 13, 2003 at 02:48:15PM -0600, Harshwardhan Nagaonkar wrote:
> Rob Sims wrote:
> >On Thursday 09 October 2003 07:16 pm, Haralambos Geortgilakis wrote:
[snip]
> Hmm... well I'm glad you were fortunate enough to get the kernel module 
> working on a 2.6 kernel (aren't the ATI prop. modules designed for 2.2.x 
> and specifically 2.4.x... atleast the Release Notes a.k.a Readme says 
> that). Good backward compatibility in 2.6 it *seems*.

You can get the latest "unreleased" ATI drivers at:
http://www.schneider-digital.de/html/download_ati.html

I have v3.2.5 for X 4.2.  It builds under kernel 2.6.

Propriatary drivers for ATI don't bug me -- I mean, if the knowledge
of how to make killer drivers was "in the general community", superfast
graphics boards would be a dime a dozen.  It is qualitatively different
from netcards, hd controllers, even sound, cause it's so bleeding edge.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Driver Download and Installation ATI Proprietary Linux Driver 3.2.8 Download--for those interested

[Harshwardhan Nagaonkar]

Rob Sims wrote:
On Thursday 09 October 2003 07:16 pm, Haralambos Geortgilakis wrote:

KDE works when I use the "radeon" driver, but I get no DRI.

Where do I go from here?  I didn't find a user forum for the ATI drivers, 
just a no-response bug reporting link.


Hmm... well I'm glad you were fortunate enough to get the kernel module 
working on a 2.6 kernel (aren't the ATI prop. modules designed for 2.2.x 
and specifically 2.4.x... atleast the Release Notes a.k.a Readme says 
that). Good backward compatibility in 2.6 it *seems*.

You mentioned that KDM refuses to start. Maybe the next step is to try 
starting a minimal windowmanager like 
blackbox/fluxbox/twm/take-your-pick and see what the XFree logs say. 
That might be a step towards better seeing if the problem is universal...

--
Harshwardhan Nagaonkar
Electrical Engineering Sysop
Brigham Young University, UT-84602
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Driver Download and Installation ATI Proprietary Linux Driver 3.2.8 Download--for those interested

[Rob Sims]

On Thursday 09 October 2003 07:16 pm, Haralambos Geortgilakis wrote:
> Hi yall,
> 
> subject says what wants said. Follow the yellow brick url too
> 
> http://www.ati.com/support/drivers/linux/radeon-linux.html?type=linux&prodType=graphic&prod=productsLINUXdriver&submit.x=18&submit.y=10

I tried this driver with a custom 2.6.0-test6 kernel.  The XF86config-4
file is that generated by fglrxconfig.  I selected single framebuffer mode 
to enable a dual headed single screen, 1600x1200x24.  The card is a 
Radeon 8500. I'm running unstable, x-window-system  4.2.1-12.1

X appears to try to start a few times, and I'm dropped back to a text 
console.  There are no errors in XFree86.0.log.  However, kdm.log 
contains:

[fglrx] API ERROR: could not register entrypoint for SecondaryColor3bEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3bvEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3dEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3dvEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3fEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3fvEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3iEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3ivEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3sEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3svEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3ubEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3ubvEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3uiEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3uivEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3usEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColor3usvEXT
[fglrx] API ERROR: could not register entrypoint for SecondaryColorPointerEXT

As a further check, I tried running xdm (normal for me is kdm), and it 
starts X and runs.  I can't get past xdm, guessing that the KDE start has 
the same problems as KDM.

A Google search on SecondaryColor3bEXT yielded four non-useful hits.

These "API ERROR" strings appear in /usr/X11R6/lib/modules/dri/fglrx_dri.so:
-rwxr-xr-x1 root root  5908219 Sep 21 20:44 fglrx_dri.so

KDE works when I use the "radeon" driver, but I get no DRI.

Where do I go from here?  I didn't find a user forum for the ATI drivers, 
just a no-response bug reporting link.
--
Rob


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Driver Download and Installation ATI Proprietary Linux Driver 3.2.8 Download--for those interested

[Haralambos Geortgilakis]

Hi yall,

subject says what wants said. Follow the yellow brick url too

http://www.ati.com/support/drivers/linux/radeon-linux.html?type=linux&prodType=graphic&prod=productsLINUXdriver&submit.x=18&submit.y=10

There are specific Debianized instructions too.

Kudos to rage3d for the link.

*BFN*

Greek Geek  :-)

MSDOS is not dead, it just smells that way. -- Henry Spencer

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Printer recommendations - Epson have wriiten Linux driver!

[Phillip Deackes]

On Fri, 9 Mar 2001 15:44:35 -0500
D-Man <[EMAIL PROTECTED]> wrote:

> On Thu, Mar 08, 2001 at 05:43:29PM -0900, Ethan Benson wrote:
> 
> I haven't seen what it lists, but the HP DeskJet 6xx my roommate
> bought worked.  

You just *have* to get an Epson 680 or higher. Epson Japan has produced a
Linux driver for it that gives the same print quality as you get from the
Windows drivers. It is fabulous!! Never before have I seen such good
quality from a colour printer on Linux! There is a gui tool for setting
paper & print quality etc. too. The driver works through your normal
/etc/printcap so just slots in nicely.

The web page for the driver is at:

http://www.epkowa.co.jp/english/pro_e/pips_e.html

There is a readme accessible from the download page which explains how to
set it all up. I downloaded the .rpm and installed that stright onto my
debian box - for some reason, alien wouldn't convert it, but if you do
'rpm -i --nodeps whatever.rpm' it will install cleanly.

-- 
Phillip Deackes
Using Progeny Debian Linux

Cisco 605 Linux Driver for DSL?

[Art Lemasters]

 Is there a Linux driver for the Cisco 605 "modem" for DSL
subscribers?  US West says that UNIX is not compatible with
it (although the 675 router/"modem" is, with a NIC installed).
Will the existing driver(s) for modems work with the 605?

Art Lemasters

BTW, it's good to be back on the Net, and hello to C.L.U.E. 
(Colorado Linux Users and Enthusiasts) again!

Re: Linux driver !

[Peter Iannarelli]

Hello Hubert:

If you need this driver to run X, starting at card number 227 in the
xf86config program are the Matrox cards.

Peter

-Original Message-
From: Hubert <[EMAIL PROTECTED]>
To: debian-user@lists.debian.org 
Date: Tuesday, September 15, 1998 6:09 AM
Subject: Linux driver !


>Dear colleague,
>
>I need linux driver for Matrox Millenium II.
>Where can I find that?
>
>
>Hubert
>
>
>--
>Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] <
/dev/null
>
>

Linux driver !

[Hubert]

Dear colleague,

I need linux driver for Matrox Millenium II.
Where can I find that?


Hubert

Linux Driver for WD7193,WD7197,WD7296 SCSI Controllers

[Andreas Dreyer]

Hi!I hava a Western Digital 
WD7193 SCSI Controller and it is not compatible to the WD7000 Driver.If 
I don't get a driver for WD7193,WD7197,WD7296 or compatible SCSI Controllers I 
willnever be able to use Linux because my harddrives and CDRom drives are 
conntected to it.And there is not driver available from Western Digital. 

 
Can anybody help me 
please?
 
Thank 
You,Andy

WD7193, WD7197, WD7296 Linux driver

[Andreas Dreyer]

Hi!
I need a driver for the Western 
Digital SCSI Controller WD7193 (WD7197,WD7296 schould work, too).
The WD7000 does not work (can't mount 
root fs) and without such a driver I will never be able to use Linux because my 
harddrives and cdroms are connectet to the WD7193 controller.
Please help me!
Thank You!
 
Andy Dreyer
e-mail:
[EMAIL PROTECTED]