Real Input and Output on the Pi emulation

[TimoHartong]

It will probably be the weekend before I set up the hardware needed to test the MCP23017 interface.

I have committed a software update which may resolve the issue, if you would like to try it.
<knip>

I have tested the code and no joy ;-(

[TimoHartong]

The issue is that init already goes not correctly:
timoh@raspberrypi:~ $ i2cdump -y 1 0x20
No size specified (using byte-data access)
0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
10: 00 00 bc bc bc 50 00 00 00 00 00 00 00 00 00 00 ..???P..........
20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
40: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
50: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
I would expect on 0x00 : 0x00 and on 0x01 : 0xff

[Bill B]

I have tested the code and no joy ;-(

OK. I will have to get some hardware set up and test properly.

Have you tried the alternate method using the kernel driver?

[TimoHartong]

No but I have done some hacking in your code ( added logging in the code ) and I see BOOLEAN hw_options (int *pargc, const char ***pargv, int *pi) in hardware.c is never called it also means that hw_read is never called so I2C and GPIO is never configured as it is should ;-(

[TimoHartong]

I have tested the code and no joy ;-(

OK. I will have to get some hardware set up and test properly.

Have you tried the alternate method using the kernel driver?

I have tested the code and no joy ;-(

OK. I will have to get some hardware set up and test properly.

Have you tried the alternate method using the kernel driver?

To get your current code working I have done the following things:
in hardware.c line comment out gio_output (8, pinPOT, 0xFF); ( around line 428 )
in gpio.c
Changed the line if ( ( pdev->fd > 0 ) && ( pdev->iData ) ) xio_put (pdev, pdev->iMask, pdev->iData); into
if ( pdev->fd > 0 )
xio_put (pdev, pdev->iMask, pdev->iData);
( around line 754 in function gio_put

in gpio.c in the function xio_put I have changed the regiters addresses from 0x14 into 0x12

in gpio.c in the function i2c_get changed the line msg[0].len = iLen + 1; into msg[0].len = sizeof(bAddr);

in memu.c
I ihave added the lines under
diag_message(DIAG_ALWAYS, "=== memu started at %s ===", buf);
hw_read ("hardware.cfg");
hw_init ();
around line 3625

[TimoHartong]

Hello I have made a page with software I have written to test the emulator and some photos of the print
https://timohartong.riscos.fr/memotech/

[Bill B]

Thanks, that will be most helpful when I get my test hardware setup.

[TimoHartong]

Your efforts are highly appreciated I hope to place schematics also today.

[TimoHartong]

Another update : when the unit starts up completely fresh the lines in hardware.c :
gio_input (8, pinPIN, 0xFF);
gio_output (8, pinPOT, 0xFF);
Do work so I have put them back. It is really a bit looking

[Bill B]

I have not yet got hardware set-up for testing, however I have been reviewing the code following your comments:

No but I have done some hacking in your code ( added logging in the code ) and I see BOOLEAN hw_options (int *pargc, const char ***pargv, int *pi) in hardware.c is never called it also means that hw_read is never called so I2C and GPIO is never configured as it is should ;-(

It is called, but not obvious from the source code. Note the following line in the CMakeLists.txt file, which is only enabled if HW_GPIO or HW_MCP23017 is defined:

Code: Select all

-DALT_OPTIONS=hw_options

As a result of that definition line 2559 of memu.c becomes a call to hw_options.

in gpio.c
Changed the line if ( ( pdev->fd > 0 ) && ( pdev->iData ) ) xio_put (pdev, pdev->iMask, pdev->iData); into
if ( pdev->fd > 0 )
xio_put (pdev, pdev->iMask, pdev->iData);
( around line 754 in function gio_put

A typo (copy / paste error) on my part. The line should be as follows, replacing iData by iMask in the condition:

Code: Select all

            if ( ( pdev->fd > 0 ) && ( pdev->iMask ) )	  xio_put (pdev, pdev->iMask, pdev->iData);

in gpio.c in the function xio_put I have changed the regiters addresses from 0x14 into 0x12

Address 0x12 is the GPIO register while 0x14 is the OLAT register. From the MCP23017 datasheet:

The GPIO register reflects the value on the port. Reading from this register reads the port. Writing to this register modifies the Output Latch (OLAT) register.

The OLAT register provides access to the output latches. A read from this register results in a read of the OLAT and not the port itself. A write to this register modifies the output latches that modifies the pins configured as outputs.

Since xio_put is reading and updating the output values, the OLAT register is the correct one to use. The update should not be affected by values on any input pins.

in gpio.c in the function i2c_get changed the line msg[0].len = iLen + 1; into msg[0].len = sizeof(bAddr);

Good spot. I agree.

in memu.c
I ihave added the lines under
diag_message(DIAG_ALWAYS, "=== memu started at %s ===", buf);
hw_read ("hardware.cfg");
hw_init ();
around line 3625

As above, this is done in hw_options which is called if and only if hardware interfacing is selected as a compilation option.

I have pushed an update which includes my version of these changes.