Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-18 Thread Bob Camp
Hi

The nice thing about PCI is that you can tie it tightly to the core time wise.

Not so easy with USB and all of it's layers.

Bob


On Dec 18, 2009, at 4:27 PM, b...@lysator.liu.se wrote:

> Hi,
> 
> I also have one or a few home projects that could use this.
> 
> Does it have to be a PCI-card? USB FPGA boards are so much cheaper. Say
> that you read your computer clock and toggle an output pin and timestamp
> this pin in your FPGA as well. Then it should still be possible to
> timestamp external PPS-pulses relative to the computer clock.
> 
> Are there other needs to be on the PCI bus?
> 
> --
> 
>   Björn
> 
> 
>> Hi
>> 
>> That's actually a petty tempting offer since this is a home project.
>> 
>> Off to investigate off the shelf PCI cards with Altera Cyclone III's on
>> them. I *think* I have all the IP for Altera.
>> 
>> After a quick look I need to add the qualifier - cards that cost less than
>> 15 45xx boards...
>> 
>> 
>> Bob
>> 
>> -Original Message-
>> From: p...@critter.freebsd.dk [mailto:p...@critter.freebsd.dk] On Behalf Of
>> Poul-Henning Kamp
>> Sent: Wednesday, December 16, 2009 8:00 AM
>> To: Bob Camp
>> Cc: soekris-tech@lists.soekris.com
>> Subject: Re: [Soekris] net5501 freebsd 8.0 compile options
>> 
>> In message , Bob Camp writes:
>>> Hi
>>> 
>>> I had a nasty suspicion that was the case.
>>> 
>>> So now it's either get another 15 45xx boards or figure out another way
>>> to do it.
>> 
>> If you know anybody just moderately VHDL capable, getting a low-profile
>> PCI card with an FPGA, a counter and a bunch of latches shouldn't
>> be too hard.
>> 
>> I did this 10 years ago for the Xilings XC6200 "Hot-1", but that does
>> not exist anymore :-(
>> 
>> If you make it open source I'll happily write the kernel driver for it.
>> 
>> --
>> Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
>> p...@freebsd.org | TCP/IP since RFC 956
>> FreeBSD committer   | BSD since 4.3-tahoe
>> Never attribute to malice what can adequately be explained by
>> incompetence.
>> 
>> 
>> 
>> ___
>> Soekris-tech mailing list
>> Soekris-tech@lists.soekris.com
>> http://lists.soekris.com/mailman/listinfo/soekris-tech
>> 
> 
> 

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-16 Thread Michael Sierchio
Poul-Henning Kamp wrote:

> If you make it open source I'll happily write the kernel driver for it.

BSD license, or nasty GPL?

(or, "What? Have you stopped drinking BEER?")

Skål!

- M
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-16 Thread Bob Camp
Hi

That's actually a petty tempting offer since this is a home project. 

Off to investigate off the shelf PCI cards with Altera Cyclone III's on
them. I *think* I have all the IP for Altera. 

After a quick look I need to add the qualifier - cards that cost less than
15 45xx boards...


Bob

-Original Message-
From: p...@critter.freebsd.dk [mailto:p...@critter.freebsd.dk] On Behalf Of
Poul-Henning Kamp
Sent: Wednesday, December 16, 2009 8:00 AM
To: Bob Camp
Cc: soekris-tech@lists.soekris.com
Subject: Re: [Soekris] net5501 freebsd 8.0 compile options 

In message , Bob Camp writes:
>Hi
>
>I had a nasty suspicion that was the case. 
>
>So now it's either get another 15 45xx boards or figure out another way 
>to do it.

If you know anybody just moderately VHDL capable, getting a low-profile
PCI card with an FPGA, a counter and a bunch of latches shouldn't
be too hard.

I did this 10 years ago for the Xilings XC6200 "Hot-1", but that does
not exist anymore :-(

If you make it open source I'll happily write the kernel driver for it.

-- 
Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer   | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.



___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-16 Thread Poul-Henning Kamp
In message , Bob Camp writes: >Hi
>
>I had a nasty suspicion that was the case. 
>
>So now it's either get another 15 45xx boards or figure out another way 
>to do it.

If you know anybody just moderately VHDL capable, getting a low-profile
PCI card with an FPGA, a counter and a bunch of latches shouldn't
be too hard.

I did this 10 years ago for the Xilings XC6200 "Hot-1", but that does
not exist anymore :-(

If you make it open source I'll happily write the kernel driver for it.

-- 
Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer   | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-16 Thread Bob Camp
Hi

I had a nasty suspicion that was the case. 

So now it's either get another 15 45xx boards or figure out another way to do 
it.

Bob


On Dec 16, 2009, at 3:04 AM, Poul-Henning Kamp wrote:

> In message <011c0fceacbb4e59a0945d7c10a92...@vectron.com>, "Bob Camp" writes:
>> Hi
>> 
>> While we're on the topic.
>> 
>> How many PPS's can a 48xx stamp at more or less the same time?
> 
> One.
> 
> 
> -- 
> Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
> p...@freebsd.org | TCP/IP since RFC 956
> FreeBSD committer   | BSD since 4.3-tahoe
> Never attribute to malice what can adequately be explained by incompetence.
> 

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-16 Thread Poul-Henning Kamp
In message <011c0fceacbb4e59a0945d7c10a92...@vectron.com>, "Bob Camp" writes:
>Hi
>
>While we're on the topic.
>
>How many PPS's can a 48xx stamp at more or less the same time?

One.


-- 
Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer   | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-15 Thread Bob Camp
Hi

While we're on the topic.

How many PPS's can a 48xx stamp at more or less the same time?

I'm too lazy to dig into the data sheets

Bob

-Original Message-
From: p...@critter.freebsd.dk [mailto:p...@critter.freebsd.dk] On Behalf Of
Poul-Henning Kamp
Sent: Tuesday, December 15, 2009 12:29 PM
To: David E. Cross
Cc: Bob Camp; soekris-tech@lists.soekris.com
Subject: Re: [Soekris] net5501 freebsd 8.0 compile options 

In message <4b27c5ac.9070...@wsg.net>, "David E. Cross" writes:

>I thought it was in the CPU itself.  Seems a shame that AMD would remove 
>that :(

It is, but it is a different CPU.

The 45xx has the Elan CPU, the 48xx and 55xx has (different) Geodes.

-- 
Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer   | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.



___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-15 Thread Poul-Henning Kamp
In message <4b27c5ac.9070...@wsg.net>, "David E. Cross" writes:

>I thought it was in the CPU itself.  Seems a shame that AMD would remove 
>that :(

It is, but it is a different CPU.

The 45xx has the Elan CPU, the 48xx and 55xx has (different) Geodes.

-- 
Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer   | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-15 Thread Bob Camp
Hi

That was my reaction when I found out 

Bob

-Original Message-
From: David E. Cross [mailto:da...@wsg.net] 
Sent: Tuesday, December 15, 2009 12:28 PM
To: Poul-Henning Kamp
Cc: Bob Camp; soekris-tech@lists.soekris.com
Subject: Re: [Soekris] net5501 freebsd 8.0 compile options

Khaan! :)

(thanks)

-- 
David E. Cross

Poul-Henning Kamp wrote:
> In message <065a73544e7648008d44f09c127b4...@vectron.com>, "Bob Camp"
writes:
>   
>> Hi
>>
>> My understanding is that the precision time stuff is specific to the 45xx
>> boards. The "magic hardware" that does the time stamping is not present
on
>> the 55xx boards.
>> 
>
> You are absolutely correct.
>
>   



___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-15 Thread David E. Cross
Khaan! :)

(thanks)

-- 
David E. Cross

Poul-Henning Kamp wrote:
> In message <065a73544e7648008d44f09c127b4...@vectron.com>, "Bob Camp" writes:
>   
>> Hi
>>
>> My understanding is that the precision time stuff is specific to the 45xx
>> boards. The "magic hardware" that does the time stamping is not present on
>> the 55xx boards.
>> 
>
> You are absolutely correct.
>
>   

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-15 Thread David E. Cross
I thought it was in the CPU itself.  Seems a shame that AMD would remove 
that :(

I am waiting for PHK to chime in though :)

-- 
David E. Cross

Bob Camp wrote:
> Hi
>
> My understanding is that the precision time stuff is specific to the 45xx
> boards. The "magic hardware" that does the time stamping is not present on
> the 55xx boards.
>
> Been wrong before though 
>
> Bob
>
> -Original Message-
> From: soekris-tech-boun...@lists.soekris.com
> [mailto:soekris-tech-boun...@lists.soekris.com] On Behalf Of David E. Cross
> Sent: Tuesday, December 15, 2009 11:59 AM
> To: soekris-tech@lists.soekris.com
> Subject: [Soekris] net5501 freebsd 8.0 compile options
>
> Ok, I have everything MOSTLY going now, its just down to tweaks and 
> optimizations.
>
> Specifically kernel config at this point; the documentation is kind of 
> vague on:
> CPU_GEODE
> CPU_SOEKRIS
> CPU_ELAN
>
> Specifically looking  for a high precession timing application for a UT+ 
> GPS clock.
>
> I realize that the GEODE is not an ELAN, but does it have the same 
> time-stamping abilities?
> since that seems to be part of what ELAN turns on?
>
>   

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-15 Thread Poul-Henning Kamp
In message <065a73544e7648008d44f09c127b4...@vectron.com>, "Bob Camp" writes:
>Hi
>
>My understanding is that the precision time stuff is specific to the 45xx
>boards. The "magic hardware" that does the time stamping is not present on
>the 55xx boards.

You are absolutely correct.

-- 
Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer   | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech


Re: [Soekris] net5501 freebsd 8.0 compile options

2009-12-15 Thread Bob Camp
Hi

My understanding is that the precision time stuff is specific to the 45xx
boards. The "magic hardware" that does the time stamping is not present on
the 55xx boards.

Been wrong before though 

Bob

-Original Message-
From: soekris-tech-boun...@lists.soekris.com
[mailto:soekris-tech-boun...@lists.soekris.com] On Behalf Of David E. Cross
Sent: Tuesday, December 15, 2009 11:59 AM
To: soekris-tech@lists.soekris.com
Subject: [Soekris] net5501 freebsd 8.0 compile options

Ok, I have everything MOSTLY going now, its just down to tweaks and 
optimizations.

Specifically kernel config at this point; the documentation is kind of 
vague on:
CPU_GEODE
CPU_SOEKRIS
CPU_ELAN

Specifically looking  for a high precession timing application for a UT+ 
GPS clock.

I realize that the GEODE is not an ELAN, but does it have the same 
time-stamping abilities?
since that seems to be part of what ELAN turns on?

-- 
David E. Cross

___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech



___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech