Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.apple2 > #24787 > unrolled thread

Marina IP updated with DNS resolution, other improvements

Back to article view | Back to comp.sys.apple2

Contents

  Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-20 18:54 +0000
    Marina IP updated with DNS resolution, other improvements Antoine Vignau <antoine.vignau@laposte.net> - 2015-05-21 00:51 -0700
      Re: Marina IP updated with DNS resolution, other improvements Steve Nickolas <usotsuki@buric.co> - 2015-05-21 10:42 +0200
        Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-21 18:55 +0000
      Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-21 18:54 +0000
        Re: Marina IP updated with DNS resolution, other improvements "Bill Garber" <willy46pa@comcast.net> - 2015-05-21 15:10 -0400
          Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-21 19:27 +0000
            Re: Marina IP updated with DNS resolution, other improvements Antoine Vignau <antoine.vignau@laposte.net> - 2015-05-21 12:44 -0700
              Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-21 20:00 +0000
                Re: Marina IP updated with DNS resolution, other improvements Jeff Blakeney <CUTjeffrey_blakeney@yahoo.ca> - 2015-05-22 15:56 -0400
                Re: Marina IP updated with DNS resolution, other improvements pitz <pitz.wong@gmail.com> - 2015-05-22 16:31 -0700
              Re: Marina IP updated with DNS resolution, other improvements "Bill Garber" <willy46pa@comcast.net> - 2015-05-21 18:37 -0400
                Re: Marina IP updated with DNS resolution, other improvements Smohn Jith <Smohn@smohn.com> - 2015-05-21 19:22 -0500
                  Re: Marina IP updated with DNS resolution, other improvements David Schmidt <schmidtd@my-deja.com> - 2015-05-21 21:43 -0400
                Re: Marina IP updated with DNS resolution, other improvements Warren Ernst <wernst@gmail.com> - 2015-05-29 11:46 -0700
            Re: Marina IP updated with DNS resolution, other improvements David Schmidt <schmidtd@my-deja.com> - 2015-05-21 16:13 -0400
            Re: Marina IP updated with DNS resolution, other improvements Michael J. Mahon <mjmahon@aol.com> - 2015-05-22 03:18 -0500
              Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-22 21:28 +0000
                Re: Marina IP updated with DNS resolution, other improvements Michael J. Mahon <mjmahon@aol.com> - 2015-05-22 18:38 -0500
                  Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-25 20:05 +0000
                    Re: Marina IP updated with DNS resolution, other improvements ol.sc@web.de (Oliver Schmidt) - 2015-05-26 18:38 +0000
                      Re: Marina IP updated with DNS resolution, other improvements Steve Nickolas <usotsuki@buric.co> - 2015-05-27 02:56 +0200
                Re: Marina IP updated with DNS resolution, other improvements David Schmidt <schmidtd@my-deja.com> - 2015-05-23 00:25 -0400
                  Re: Marina IP updated with DNS resolution, other improvements David Schmenk <dschmenk@gmail.com> - 2015-05-23 10:01 -0700
                    Re: Marina IP updated with DNS resolution, other improvements ol.sc@web.de (Oliver Schmidt) - 2015-05-24 19:55 +0000
                      Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-25 20:10 +0000
                        Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-05-25 20:13 +0000
                          Re: Marina IP updated with DNS resolution, other improvements Michael J. Mahon <mjmahon@aol.com> - 2015-05-25 16:01 -0500
        Re: Marina IP updated with DNS resolution, other improvements wssimms@gmail.com - 2015-05-21 22:18 -0700
          Re: Marina IP updated with DNS resolution, other improvements Antoine Vignau <antoine.vignau@laposte.net> - 2015-05-21 22:57 -0700
    Re: Marina IP updated with DNS resolution, other improvements D Finnigan <dog_cow@macgui.com> - 2015-06-01 18:22 +0000

Page 1 of 2  [1] 2  Next page →

#24787 — Marina IP updated with DNS resolution, other improvements

Today I released the newest edition of the Marina IP stack for Apple II
computers. This edition adds DNS name resolution, as well as several other
bug fixes and improvements. Domain name resolution means that users can use
human-readable names such as www.yahoo.com instead of hard-to-remember IP
addresses like 98.138.253.109 to access resources on the Internet.

The Marina IP stack is the newest networking implementation for the Apple II
series, designed to have better RFC-compliance than existing
implementations. Written in 6502 assembly language, it is compatible with
all Apple II models that accept the Uthernet card in slot 3.

For more information, and links to download and documentation, visit the
official Marina web site:
http://marina.a2hq.com/


Here is the change log:

- Added checks on destination address for incoming datagrams 
- Datagrams not addressed to our IP, or to broadcast, or DHCP replies, are
dropped. 
- Broadcast flag is set for incoming broadcast datagrams 
- IP datagrams with head longer than 20 bytes are checksummed and accepted,
but options are not processed 
- Fixed checksum routine to work with IP headers longer than 20 bytes 
- Fixed the check for UDP datagrams sent with no checksum (offset was off by
one) 
- Added NETBCAST for storing network broadcast address, e.g. 169.254.255 or
192.168.2.255 
- Incoming check for broadcasts now looks for "all ones" or the network
broadcast address
- Updated IPSEND to use the new check on NETBCAST 
- Wrote IPTOHEX subroutine to convert dotted decimal addresses to 4 hex
digits
- ARPTEST and UDPSENDTEST now take a user-supplied IP address 
- Bug fix: datagrams with Link-Local destination addresses were being
forwarded to router when a routable address was configured 
- intitial DNS resolver implementation that queries for A records 
- IPSEND now uses a retry counter for ARP requests. It will not fail if the
MAC address was not in the ARP table 
- IPSEND forces TTL to 255 when destination is a Link-Local address 
- Increased size of ARP table from 4 to 5 
- display of incoming IP datagrams shows source address 
- renamed UTHERDEMO file to MARINA 


-- 
]DF$
The Marina IP stack for Apple II--
http://marina.a2hq.com/

[toc] | [next] | [standalone]

#24791

That looks great, David.
Why slot 3? Will it run with the card in other slots?
av

[toc] | [prev] | [next] | [standalone]

#24792

On Thu, 21 May 2015, Antoine Vignau wrote:

> That looks great, David.
> Why slot 3? Will it run with the card in other slots?
> av
>

Well, AppleWin puts it in 3, so it might be that...

-uso.

[toc] | [prev] | [next] | [standalone]

#24795

Steve Nickolas wrote:
> On Thu, 21 May 2015, Antoine Vignau wrote:
> 
>> That looks great, David.
>> Why slot 3? Will it run with the card in other slots?
>> av
>>
> 
> Well, AppleWin puts it in 3, so it might be that...

I don't use that emulator, and in fact I have never programmed or tested
Marina in any emulator. I picked slot 3 because Uthernet has no ROM, and
that's where we've been putting that card by convention.

[toc] | [prev] | [next] | [standalone]

#24794

Antoine Vignau wrote:
> That looks great, David.
> Why slot 3? Will it run with the card in other slots?
> av
>

I have the slot I/O addresses hard-coded to $C0Bx. I don't like
self-modifying code, which is what several other Uthernet drivers use to get
slot independence.

-- 
]DF$
The Marina IP stack for Apple II--
http://marina.a2hq.com/

[toc] | [prev] | [next] | [standalone]

#24796

"D Finnigan" <dog_cow@macgui.com> wrote in message 
news:dog_cow-1432234539@macgui.com... 
> Antoine Vignau wrote: 
>> That looks great, David. 
>> Why slot 3? Will it run with the card in other slots? 
>> av 
> 
> I have the slot I/O addresses hard-coded to $C0Bx. I don't 
> like self-modifying code, which is what several other Uthernet 
> drivers use to get slot independence. 

Most, if not all, of the software that I've run using the Uthernet, 
have config programs where the user selects the Uthernet's 
slot location. 

Bill Garber - I love my 
C1 D0 D0 CC C5 A0 C9 C9 C7 D3 
http://www.sepa-electronics.com 

[toc] | [prev] | [next] | [standalone]

#24797

Bill Garber wrote:
> "D Finnigan" <dog_cow@macgui.com> wrote in message 
> news:dog_cow-1432234539@macgui.com... 
>> Antoine Vignau wrote: 
>>> That looks great, David. 
>>> Why slot 3? Will it run with the card in other slots? 
>>> av 
>> 
>> I have the slot I/O addresses hard-coded to $C0Bx. I don't 
>> like self-modifying code, which is what several other Uthernet 
>> drivers use to get slot independence. 
> 
> Most, if not all, of the software that I've run using the Uthernet, 
> have config programs where the user selects the Uthernet's 
> slot location. 

Yes, and they're using self-modifying code to change the slot I/O
addresses.* I may do the same thing one day, or I may just make an assembly
option. Right now, the slot addresses are EQUates in my source files. If
someone really wants the Marina demo for another slot, he or she is welcome
to email me with the desired slot and I will reassemble for it.

* for example, ip65 as used in ADTPro:
http://adtpro.cvs.sourceforge.net/viewvc/adtpro/adtpro/client/src/ip65/drivers/cs8900a.s?revision=1.2&view=markup
lines 278-365

-- 
]DF$
The Marina IP stack for Apple II--
http://marina.a2hq.com/

[toc] | [prev] | [next] | [standalone]

#24798

This is why I keep asking hardware developers to follow Apple (old) rules to provide some sort of ROM to identify cards.

Where is the convention of an ethernet card to be on slot 3, please? For instance, my cards are in slot 1 by my own convention;-)

Antoine

[toc] | [prev] | [next] | [standalone]

#24799

Antoine Vignau wrote:
> This is why I keep asking hardware developers to follow Apple (old) rules
> to provide some sort of ROM to identify cards.
> 
> Where is the convention of an ethernet card to be on slot 3, please? For
> instance, my cards are in slot 1 by my own convention;-)

It's the convention that cards with no ROM go in slot 3 because in the Apple
IIe, the built-in 80-columns firmware usually overrides the ROM in slot 3.

But let's open the scope of this discussion: what slot are people using for
the Uthernet if not slot 3?

-- 
]DF$
The Marina IP stack for Apple II--
http://marina.a2hq.com/

[toc] | [prev] | [next] | [standalone]

#24822

On 21/05/2015 4:00 PM, D Finnigan wrote:
> Antoine Vignau wrote:
>> This is why I keep asking hardware developers to follow Apple (old) rules
>> to provide some sort of ROM to identify cards.
>>
>> Where is the convention of an ethernet card to be on slot 3, please? For
>> instance, my cards are in slot 1 by my own convention;-)
>
> It's the convention that cards with no ROM go in slot 3 because in the Apple
> IIe, the built-in 80-columns firmware usually overrides the ROM in slot 3.
>
> But let's open the scope of this discussion: what slot are people using for
> the Uthernet if not slot 3?
>

I have my Uthernet in slot 2.  This way I can swap between communicating 
via ethernet or serial by changing the IIgs control panel between 
"Modem" and "Your Card".

My ZipGS is in slot 3.

[toc] | [prev] | [next] | [standalone]

#24827

On Thursday, May 21, 2015 at 1:02:09 PM UTC-7, D Finnigan wrote:
> But let's open the scope of this discussion: what slot are people using for
> the Uthernet if not slot 3?
> 
On a IIgs, my accelerator has to sit in slot #3 (because of the short cables to the 65816 socket).  With ROM3, I have the Uthernet typically in slot 1, also switching between using AppleTalk and using Ethernet.

[toc] | [prev] | [next] | [standalone]

#24801

"Antoine Vignau" <antoine.vignau@laposte.net> wrote in message 
news:3bf442c8-ace8-40af-b71b-5115285b8549@googlegroups.com... 
> This is why I keep asking hardware developers to follow 
> Apple (old) rules to provide some sort of ROM to identify 
> cards. 
> 
> Where is the convention of an ethernet card to be on 
> slot 3, please? For instance, my cards are in slot 1 
> by my own convention;-) 
> 
> Antoine 

In a //e, I would put the Uthernet in Slot#3 because 
of the No-ROM rule, but in my ROM03 IIgs, Slot#4 is 
best because the mouse works still with the Uthernet 
in there. Also, an accellerator card goes in Slot#3 
in the IIgs... Usually. 

Bill Garber - I love my 
C1 D0 D0 CC C5 A0 C9 C9 C7 D3 
http://www.sepa-electronics.com 

[toc] | [prev] | [next] | [standalone]

#24802

On 5/21/2015 5:37 PM, Bill Garber wrote:
>
> "Antoine Vignau" <antoine.vignau@laposte.net> wrote in message
> news:3bf442c8-ace8-40af-b71b-5115285b8549@googlegroups.com...
>> This is why I keep asking hardware developers to follow Apple (old)
>> rules to provide some sort of ROM to identify cards.
>> Where is the convention of an ethernet card to be on slot 3, please?
>> For instance, my cards are in slot 1 by my own convention;-)
>> Antoine
>
> In a //e, I would put the Uthernet in Slot#3 because of the No-ROM rule,
> but in my ROM03 IIgs, Slot#4 is best because the mouse works still with
> the Uthernet in there. Also, an accellerator card goes in Slot#3 in the
> IIgs... Usually.
> Bill Garber - I love my C1 D0 D0 CC C5 A0 C9 C9 C7 D3
> http://www.sepa-electronics.com
>

Why wouldn't the Uthernet go in Slot 2 and replace the modem?

[toc] | [prev] | [next] | [standalone]

#24809

On 5/21/2015 8:22 PM, Smohn Jith wrote:
> On 5/21/2015 5:37 PM, Bill Garber wrote:
>>
>> "Antoine Vignau" <antoine.vignau@laposte.net> wrote in message
>> news:3bf442c8-ace8-40af-b71b-5115285b8549@googlegroups.com...
>>> This is why I keep asking hardware developers to follow Apple (old)
>>> rules to provide some sort of ROM to identify cards.
>>> Where is the convention of an ethernet card to be on slot 3, please?
>>> For instance, my cards are in slot 1 by my own convention;-)
>>> Antoine
>>
>> In a //e, I would put the Uthernet in Slot#3 because of the No-ROM rule,
>> but in my ROM03 IIgs, Slot#4 is best because the mouse works still with
>> the Uthernet in there. Also, an accellerator card goes in Slot#3 in the
>> IIgs... Usually.
>> Bill Garber - I love my C1 D0 D0 CC C5 A0 C9 C9 C7 D3
>> http://www.sepa-electronics.com
>>
>
> Why wouldn't the Uthernet go in Slot 2 and replace the modem?

Because your null modem cable plugs in there, of course. ;-)

[toc] | [prev] | [next] | [standalone]

#24894

On Thursday, May 21, 2015 at 3:37:07 PM UTC-7, Bill Garber wrote:
> "Antoine Vignau" <antoine.vignau@laposte.net> wrote in message 
> news:3bf442c8-ace8-40af-b71b-5115285b8549@googlegroups.com... 
> > This is why I keep asking hardware developers to follow 
> > Apple (old) rules to provide some sort of ROM to identify 
> > cards. 
> > 
> > Where is the convention of an ethernet card to be on 
> > slot 3, please? For instance, my cards are in slot 1 
> > by my own convention;-) 
> > 
> > Antoine 
> 
> In a //e, I would put the Uthernet in Slot#3 because 
> of the No-ROM rule, but in my ROM03 IIgs, Slot#4 is 
> best because the mouse works still with the Uthernet 
> in there. Also, an accellerator card goes in Slot#3 
> in the IIgs... Usually. 
> 
> Bill Garber - I love my 
> C1 D0 D0 CC C5 A0 C9 C9 C7 D3 
> http://www.sepa-electronics.com

Ditto for me: my uthernet card is in Slot 4 on my Rom03 GS.

-Warr

[toc] | [prev] | [next] | [standalone]

#24800

On 5/21/2015 3:27 PM, D Finnigan wrote:
> Bill Garber wrote:
>> "D Finnigan" <dog_cow@macgui.com> wrote in message
>> news:dog_cow-1432234539@macgui.com...
>>> Antoine Vignau wrote:
>>>> That looks great, David.
>>>> Why slot 3? Will it run with the card in other slots?
>>>> av
>>>
>>> I have the slot I/O addresses hard-coded to $C0Bx. I don't
>>> like self-modifying code, which is what several other Uthernet
>>> drivers use to get slot independence.
>>
>> Most, if not all, of the software that I've run using the Uthernet,
>> have config programs where the user selects the Uthernet's
>> slot location.
>
> Yes, and they're using self-modifying code to change the slot I/O
> addresses.

So load the slot offset in X, and use that as an index to the I/O 
address.  That's how onboard ROMs have to do it when they can't depend 
on the slot they're in.  Personally, I'm not allergic to self-abusing 
code.  I am, however, allergic to slot dependence and would use the 
offset technique.

[toc] | [prev] | [next] | [standalone]

#24814

D Finnigan <dog_cow@macgui.com> wrote:
> Bill Garber wrote:
>> "D Finnigan" <dog_cow@macgui.com> wrote in message 
>> news:dog_cow-1432234539@macgui.com... 
>>> Antoine Vignau wrote: 
>>>> That looks great, David. 
>>>> Why slot 3? Will it run with the card in other slots? 
>>>> av 
>>> 
>>> I have the slot I/O addresses hard-coded to $C0Bx. I don't 
>>> like self-modifying code, which is what several other Uthernet 
>>> drivers use to get slot independence. 
>> 
>> Most, if not all, of the software that I've run using the Uthernet, 
>> have config programs where the user selects the Uthernet's 
>> slot location. 
> 
> Yes, and they're using self-modifying code to change the slot I/O
> addresses.* I may do the same thing one day, or I may just make an assembly
> option. Right now, the slot addresses are EQUates in my source files. If
> someone really wants the Marina demo for another slot, he or she is welcome
> to email me with the desired slot and I will reassemble for it.
> 
> * for example, ip65 as used in ADTPro:
> http://adtpro.cvs.sourceforge.net/viewvc/adtpro/adtpro/client/src/ip65/drivers/cs8900a.s?revision=1.2&view=markup
> lines 278-365


Anybody remember what index registers are for?

And configuration code is quite different from "self-modifying" code--which
I consider no problem for code not meant for ROM.

Making a constant into a variable is one of the most common program
modifications, since it prevents the proliferation of different "versions"
of the code. 
-- 
-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon

[toc] | [prev] | [next] | [standalone]

#24826

Michael J. Mahon wrote:
> 
> Anybody remember what index registers are for?

No need to poke fun at us, sir! Yes I have looked at the Disk II controller
and the Super Serial card firmware for slot independence.

I'll see what I can do. I'm already using X in some places.

-- 
]DF$
The Marina IP stack for Apple II--
http://marina.a2hq.com/

[toc] | [prev] | [next] | [standalone]

#24828

D Finnigan <dog_cow@macgui.com> wrote:
> Michael J. Mahon wrote:
>> 
>> Anybody remember what index registers are for?
> 
> No need to poke fun at us, sir! Yes I have looked at the Disk II controller
> and the Super Serial card firmware for slot independence.
> 
> I'll see what I can do. I'm already using X in some places.

Sorry I left off the smiley--I was chuckling as I wrote. ;-)

Unlike slot ROMs, which can just JSR to an RTS and look at the return
address on the stack to see which slot they're running in, I think you'll
either have to ask the user or search for the network card by its behavior.


It's always a question of whether to save the result or just redo it each
time it's run.

It's a real luxury having a choice of TCP/IP stacks. I appreciate your
efforts!
-- 
-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon

[toc] | [prev] | [next] | [standalone]

#24843

Michael J. Mahon <mjmahon@aol.com> wrote:
> D Finnigan <dog_cow@macgui.com> wrote:
>> Michael J. Mahon wrote:
>>> 
>>> Anybody remember what index registers are for?
>> 
>> No need to poke fun at us, sir! Yes I have looked at the Disk II controller
>> and the Super Serial card firmware for slot independence.
>> 
>> I'll see what I can do. I'm already using X in some places.
> 
> Sorry I left off the smiley--I was chuckling as I wrote. ;-)

Yes that's how I interpreted your message, and how I intended my response.
:-)

> 
> Unlike slot ROMs, which can just JSR to an RTS and look at the return
> address on the stack to see which slot they're running in, I think you'll
> either have to ask the user or search for the network card by its behavior.

The card's Ethernet controller has ID bytes to identify it. Problem is, who
knows what might get activated on other cards during your slot scan?

> 
> 
> It's always a question of whether to save the result or just redo it each
> time it's run.
> 
> It's a real luxury having a choice of TCP/IP stacks. I appreciate your
> efforts!

Yep, mostly just a learning exercise for me. I've learned *quite* a lot so
far. I'm currently planning the sockets API that will also underpin the
Applesoft ampersand extension.

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | comp.sys.apple2

csiph-web