Need your opinion regarding GR8NET capabilities

Page 11/55
4 | 5 | 6 | 7 | 8 | 9 | 10 | | 12 | 13 | 14 | 15 | 16

By hit9918

Prophet (2864)

hit9918's picture

22-11-2015, 19:26

"only one adapter will have to respond with CPL to the port 40"
I don't know, this sounds like not sticking to the sheme.

how to use switched ports in interrupts.
save the current selection and restore at end of interrupt.
"only one adapter responds" sounds like problems.

By Eugeny_Brychkov

Paragon (1080)

Eugeny_Brychkov's picture

22-11-2015, 19:33

hit9918 wrote:

"only one adapter responds" sounds like problems.

It is not a problem, it is how it should be. If there're 4 adapters, I need to design algorithm how to ensure that only one talks to CPU at a time with required information.

By hit9918

Prophet (2864)

hit9918's picture

22-11-2015, 19:50

Well and then it sounds like you got two adapters on one port object.
And then on port 41 sub-select adapter.

problem example:
out 0x40,object
out 0x41,sub-select non existant adapter

some interrupt thing happens.
it does IN 0x40 to save port selection.
it returns 0xFF because you said "non existant adapter does not respond".

at end of interrupt, it does restore
out 0x40,0 (the CPL value)

the main program continues with port select sheme messed up.

I wonder why you cannot sepatate the two topics "port selection" and "adapter".
when one port object has two adapters, not "one adapter" responds, but the port object responds.
maybe you wanted to fit the "sub-device exists" check into the port object sheme. It doesnt sound good.
isn't there another way to detect that there is no adapter. some register returning FF.

By Grauw

Ascended (8310)

Grauw's picture

22-11-2015, 21:04

Eugeny_Brychkov wrote:

1. only one adapter will have to respond with CPL to the port 40, as well as for other ports. It can be achieved the same way I did it now - by choosing "selected" adapter. However if application will "select" nonexistent adapter, none will reply to any port. The remedy is to enumerate all 3 adapters through I/O ports and find if there're any. This is not in contradiction or violation to the switched port definition, but a slight difference, and can be considered as "fault" situation because software is not allowed to choose nonexistent adapter.

It sounds good to me!

By the way, just an idea, throwing it out there, maybe useful maybe not: Have you considered to use 43H-48H for adapters 2-4? Yes the fourth adapter would conflict with Franky but I think it’s acceptable; typically one would only have one GR8NET anyway, and incompatibility with Franky if someone really uses four GR8NETs is no big deal I think, and the likelihood is low.

Eugeny_Brychkov wrote:

Could be there issue if some other application/system code write to port 40h in the middle of access to the GR8NET? For assembler routine it is rather simple, just use DI/EI, but it is a big question for BASIC. Will user be able to do the following:
OUT &H40,&H8E:A=INP(&H41)
and there will be no code which will change contents of port 40h between execution of these OUT and INP commands?

In theory yes, if there’s an ISR communicating with a device which uses switched I/O ports there is no guarantee it switches back to the original bank I think. However I don’t know of any at the moment, I think usually one needn’t worry about it. Hopefully any ISR would also restore the original switched I/O port. If one would want to be sure in a Basic tool made available to the public, it could execute a little asm program through a small DATA block.

Eugeny_Brychkov wrote:

Of course it seems it is a question of compliance of the other device to the standard (that's why there's CPL returned - code should be required to read current device ID, then process its own, and write CPL or original device ID back into port 40h and only then exit), but I have to be sure that I will not have failures of the adapter operation because of 3rd party non-compliance or failures. We already have examples like previously mentioned "Franky" device.

I think if you always switch back to bank 0 or 255 after accessing it, the odds of anything writing unintentionally to the GR8NET I/O ports are very low.

Not sure why Franky used those I/O ports, I thought initially it was because the SMS also uses those ports but that doesn’t seem to be the case, but it’s luckily the only exception I know of.

By Eugeny_Brychkov

Paragon (1080)

Eugeny_Brychkov's picture

23-11-2015, 13:49

After considering pros and cons of my current I/O port access technique versus switched port technique I decided to stick to current one, but change ports from 10-11 to 5e-5f. Please book these ports 5E and 5F for GR8NET in the port layout documentation. Thank you.

By zPasi

Champion (410)

zPasi's picture

24-11-2015, 12:48

Ok. Any idea what this GR8NET adapter will cost?

And, what about WiFi? Probably not internal / included, but as an add-on module, is it possible? I know there are relatively easy-to-use WiFi modules that can be connected to a microcontroller or processor. They just require a few GPIO-pins and software to run.

By Eugeny_Brychkov

Paragon (1080)

Eugeny_Brychkov's picture

24-11-2015, 21:12

Again, the one who wants the device please email me.

@zPasi: no WiFi at this time. You may add external module or router with wireless through twisted pair. I do not think there will be problems with such configuration.

By zPasi

Champion (410)

zPasi's picture

25-11-2015, 15:29

Eugeny_Brychkov wrote:

Again, the one who wants the device please email me.

I sent mail to info (at) gr8bit.ru. Is this the right address?

By edoz

Prophet (2168)

edoz's picture

25-11-2015, 15:34

Eugeny_Brychkov wrote:

Again, the one who wants the device please email me.

@zPasi: no WiFi at this time. You may add external module or router with wireless through twisted pair. I do not think there will be problems with such configuration.

@zPasi.. i use this wirless bridge.. my MSX is connected to the bridge by a cable the Bridge is connected by WIFI to my home Wifi router. This works very nice! (wap3000)

By edoz

Prophet (2168)

edoz's picture

25-11-2015, 15:41

BDW: the device looks so cool in the document!

Page 11/55
4 | 5 | 6 | 7 | 8 | 9 | 10 | | 12 | 13 | 14 | 15 | 16