68kavr - considered competed

So that happened.
I consider 68kavr project competed.
That means that it loads OS from SD card
Today I've published my SD card image with the source of BIOS/LDRBIOS, as usuially, here , I'll check two last chapters and copy them there (to be able to discuss them and just they not to be lost one day) in few days.

Advanced band selection for OpenWRT/luci/Huawei modems

I have a wonderful OpenWRT on a nice TPLink 842 router, and I want to add band selection for LTE and selective UMTS - 900 and 2100 Mhz.
I have E3276 (not in NCM mode!) and several 3G (E1552,E1762) modems. I had to add the folowing:
/lib/netifd/proto/3g.sh:

                       elif echo "$cardinfo" | grep -q "Sierra Wireless"; then
                                SIERRA=1
                       elif echo "$cardinfo" | grep -q "MF823"; then
                                CODE="6" 
                                case "$service" in
                                        umts_only) CODE="5";;
                                        umts900_only) CODE="5";;
                                        umts2100_only) CODE="5";;
                                        gprs_only) CODE="4";;
                                        gprs900_only) CODE="4";;
                                        gprs1800_only) CODE="4";;
                                        lte_only) CODE="7";;
                                        lte_b7_only) CODE="7";;
                                        lte_b20_only) CODE="7";;

                                esac
                                export MODE="AT+ZPREFMODE=${CODE}"
                        elif echo "$cardinfo" | grep -qi "E327"; then
# E327 means both E3276 and E3272 at least; correct for your's 4G model
# 00000100 is EGSM900, 00000200 is primary GSM900
                                BAND="3FFFFFFF"
                                LTEBAND="7FFFFFFFFFFFFFFF"
                                case "$service" in
                                        umts_only) CODE="02";;
                                        umts900_only) CODE="02"
                                            BAND="0002000000000000";;
                                        umts2100_only) CODE="02"
                                            BAND="00400000";;
                                        gprs_only) CODE="01";;
                                        gprs900_only) CODE="01"
                                            BAND="00000100";;
                                        gprs1800_only) CODE="01"
                                            BAND="00000080";;
                                        lte_only) CODE="03";;
                                        lte_b7_only) CODE="03"
                                            LTEBAND="40";;
                                        lte_b20_only) CODE="03"
                                            LTEBAND="80000";;
                                        *) CODE="00";;
                                esac
                                export MODE="AT^SYSCFGEX=\"${CODE}\",${BAND},1,4,${LTEBAND},\"\",\"\""
                        elif echo "$cardinfo" | grep -qi huawei; then
                                BAND="3FFFFFFF"
                                case "$service" in
                                        lte_only) CODE="14,2";;
                                        lte_b7_only) CODE="14,2";;
                                        lte_b20_only) CODE="14,2";;
                                        umts_only) CODE="14,2";;
                                        umts900_only) CODE="14,2"
                                            BAND="0002000000000000";;
                                        umts2100_only) CODE="14,2"
                                            BAND="00400000";;
                                        gprs_only) CODE="13,1";;
                                        gprs900_only) CODE="13,1"
                                            BAND="00000100";;
                                        gprs1800_only) CODE="13,1"
                                            BAND="00000080";;
                                        *) CODE="2,2";;
                                esac
                                export MODE="AT^SYSCFG=${CODE},${BAND},2,4"
                        fi

and in /usr/lib/lua/luci/model/cbi/admin_network/proto_3g.lua

service = section:taboption("general", Value, "service", translate("Service Type"))
service:value("", translate("-- Please choose --"))
service:value("umts", "UMTS/GPRS")
service:value("umts_only", translate("UMTS only"))
service:value("umts900_only", translate("UMTS 900 only"))
service:value("umts2100_only", translate("UMTS 2100 only"))
service:value("gprs_only", translate("GPRS only"))
service:value("gprs900_only", translate("GPRS 900 only"))
service:value("gprs1800_only", translate("GPRS 1800 only"))
service:value("lte_only", translate("LTE only"))
service:value("lte_b7_only", translate("LTE Band 7 only"))
service:value("lte_b20_only", translate("LTE Band 20 only"))
service:value("evdo", "CDMA/EV-DO")

To see how gcom sets the mode, add in /etc/gcom/setmode.gcom:

:start
 print "Trying to set mode ",$env("MODE")," (at /etc/gcom)\n"

And you can monitor by logread -f and see something like that:

Sat May  3 16:06:53 2014 daemon.notice netifd: Interface '3g' is setting up now
Sat May  3 16:06:55 2014 daemon.notice netifd: 3g (10250): Trying to set mode AT^SYSCFGEX="03",3FFFFFFF,1,4,7FFFFFFFFFFFFFFF,"","" (at /etc/gcom)
Sat May  3 16:06:57 2014 daemon.notice pppd[10273]: pppd 2.4.5 started by root, uid 0
Sat May  3 16:06:58 2014 local2.info chat[10275]: abort on (BUSY)

UPD 20140627:

• Also works with E3272 [Megafon-100?] (change grep -qi "E3276"; to grep -qi "E327"; to support both modems)
• Has a bug in 4G/LTE band selection. Instead of 1st ($CODE) , the 5th (7FFFFFFFFFFFFFFF) parameter should be changed. If I ever test this on real hardware, I'll update the post.

UPD 20150502: Fixed 4G band selection. UPD 20150801: Some code for ZTE 823D added, it must be in AT-command mode and it cannot select bands, just switches from GSM to UMTS and LTE. That's the only ZTE modem I own for now, so I cannot say if any other model will do the same stuff, check it.

68k project: step 7 - CP/M 68, booting from Srecords

Ok, lets start from code. Code for ChVII
  The firmware is mostly [or exactly?] the same, but there are two new directories - BIOS ,which contains BIOS (sort of ugly bios), and Disk, which contains pretty empty disk image and diskdefs for cpmtools. (local)
The disk image itself contains A: - P: drives, cpmtools with this diskdefs are able to write to A: . Write there a whole CP/M distribution. The disk image comes from Searle Grant's Z80 SBC, (or even my mentioned SD interface for it) as all the DPB stuff.
  For the bios reassembly use assemble script, you'll get bios.s. Next, load it and CPM15000.SR via cut-and-paste to SBC console.
  Personally I do it with kermit. It's much better for our use then saying minicom.
  Here is my kermit initialization (I hope I forgot nothing:)

set line /dev/ttyS0
set car off
set speed 38400
set flow rts
set transmit timeout 10

(and set input echo on, but I forgot what it is for and if it is really needed)    So from SBC prompt say
* boot
, it will initialise the card and load something ($2000 bytes?) to somewhere ($1000?). But currently we do not care.
  Next, load bios.s and CPM15000.SR by

 C-Kermit> transmit CPM15000.SR

and

 C-Kermit> transmit bios.s

Next, run it via go15000 and get A:.
  The next thing is to create CPMLDR and CPM.SYS. Sure, we will create them on the board. But I refuse to learn CP/M ed!
So, the next thing I going to do is to get some text editor. And that will be separate chapter or two.


Some things to mention:

• First, get .68K files from .REL . That's done by series of RELOCX.SUB scripts
• if AS68 says can't open 0:AS68SYMB.DAT, ensure you have AS68INIT from DISK3 and run AS68 -I AS68INIT for every TPA change
• http://forums.debian.net/viewtopic.php?f=16&t=112244 A very good article how to recalculate DPBs into cpmtools diskdefs.
  It does not help me much because it is long and clever; I always guess some things (and probably wrong)

A working CP/M 68K emulator found

it is fresh and newly-written, but it works
Also, it contains a working version of emacs! That solves my need for a text editor for CP/M 68k (but not a reason to finish my attempts to port dealin and ted editors)

I have a bootable CP/M 68!

Yes,I did it!
Looks like AS68 has some problems with conditional assembling, so I had to pass my assembler source via linux's cpp. As well it does not like symbols longer then 8 chars (and it's clear why), so using cpp was a good idea (and then transfer it on board).
Anyway, I've got an $18000 TPA. Good!

I've burned one more Atmega8A

And found that I have no any Atmega8As till tomorrow. Hopefully this online shop promise to deliver me some chips on next day after an order [in Moscow]; that's good!
I have some Atmega8L's (max 8 Mhz), and tried to make them work at least at 9600.
Ha! It works on 38400! My 68kavr flow control algorithm considered to be good.

68k project: step 6 - lets' start thinking about CP/M

First, some news of the project: The updated scheme HERE . Generally nothing new, only MAX232A is drawn
Here is the picture of my board taken 20140325.
First, I have some minor (but nice!) changes in my firmware. First is
GOaddressmonitor command , which goes to the specified address.
And the second... (have I ever told you I'm genius? REALLY? Oh, sorry) - I catch the BREAK from a serial line and inititate CPU Reset! (actually, even MCU reset).
So, in C-Kermit> (if you use C-Kermit as do I) 's terminal mode, I can press [CTRL]+[\] [B] , and get a reset!
More, I own an HP 9000 A500 server and it has as "Service Processor". And I'd like to have a full Service Processor console in future like the one on that server. Saying, it can turn on and off ATX power (if I have a spare pin and free memory on Atmega8)....
Sometimes in future
So, let's talk about CP/M iteself
This is my first attempt to port CP/M to any computer by the method explained in, our case, "System Guide". The prerequestings are:

• I should have System Guide (I do)
• I should have CP/M 68 in S-records (I do from http://www.cpm.z80.de/binary.html : I use this)
• The system must accept S-Records (my does)
• I must have an assembler which generates S-Records (I use A68k from TI 68K calculator project:

  
  68000 Assembler - version 2.71.F3s (Sep 21, 2004)
  Copyright 1985 by Brian R. Anderson
  AmigaDOS conversion copyright 1991 by Charlie Gibbs.
  Adapted for use with Fargo by David Ellsworth.
  Bugfixes and additions by Julien Muchembled, Paul Froissart and Kevin Kofler
  
  

  Warning: do not use asm68k

  
  asmx multi-assembler version 2.0b5
  Copyright 1998-2007 Bruce Tomlin
  
  

  buggy as shit. May be vasm is good - but I'm not ready to write linker scripts just to get SRecord from the source with static lables and given ORG)

So, on the CP/M distribution there is a file CPM15000.SR - that's an CP/M in S-Records (which is loaded from address $15000).
On the first disk of distribution set there is a very useful README.TXT which says that to create a BIOS I should figure out _init (== ORG of BIOS == $1B000) and _ccp ($150BC) symbols in CPM15000.MAP (and that differs from what is written in "System Guide"; there said you should patch Srecords file [for 1.0 and 1.1 CP/M 68k vers.]).
So lets go to "System Guide" directly to the Appendix B on page 59 and start type "Sample Bios" into text editor.
As you can see, BIOS install TRAP #3 and CP/M works with BIOS via this TRAP. I've tried to start with

• constat
• conin
• conout

, but it was not enough; setexc also a must for CP/M to show you it's A:>. The first thing it does, it installs an TRAP handlers for almost all the TRAPS! I will not retype it here, that will be attached later, when it will be at least read/write sectors. To start CP/M first load CP/M itself, then BIOS, then type go15000
After I wrote basic sector read/write access I was able to say DIR,TYPE and even REN. But could not run any program, that's because I've forgot getseg call at all. And then I was able to run DDT68000, but during read it compained "Cannot write sector" (and it was not writing sector at all!)
That was because I forgot flush call.
Funny - this code contains *No* CP/M specific code yet, and is just update of firmware ;-) Code for ChVI