My network connection has been down for almost a week so I could nit 
pursue the suggestions you made. Now that I have, I think it is worth 
going for the hardware solution as there are many places that the 
parport driver is impacted and it will take a lot of testing.
I will still do the i2c for parport but with some time to test properly 
later.

On 2014-05-18 15:59, Steve Stallings wrote:
> Marius,
>
> The hardware direction control issues may be simpler
> than you expect.
>
> The I2C bus does not use tri-state and direction
> control. Rather it uses "open collector" or "open
> drain" style outputs with a pull-up resistor. When
> it wants to read from an input, it writes a logic
> one to that output so that the driver is not pulling
> down.
>
> The resistor pulls up and if there is nothing else
> driving the signal, then the input from that same
> signal will read as one. If something else connected
> to that signal drives the signal to zero, then it
> will overcome the pull-up resistor and the input
> will read as a zero.
>
> The standard parallel printer port includes four
> signals (pins 1, 14, 16, and 17) that behave in this
> same manner and may be usable for direct connection
> to an I2C bus.
>
> Recommended reading:
>
> http://www.linux-cae.net/Projects/Parallel/Dage/parport.html
>
> Parallel ports that follow the original IBM-PC design
> recommendations will have a 4.7K ohm pull-up resistor.
> For many applications this will directly work with the
> I2C bus. In a few cases the resistor may be a different
> value and in a very few rare cases parallel ports have
> been built that use totem-pole drivers for these pins.
>
> If you are not comfortable with connecting your I2C bus
> directly to the parallel port, you can also use external
> buffers that are open-collector to achieve the same thing
> without needing to control the direction of the parallel
> port's data pins. It just requires one more pin from the
> parallel port because the data out and data in will need
> to be on different pins. See this application note:
>
> http://www.maximintegrated.com/app-notes/index.mvp/id/3230
>
> Do keep in mind that some of the signal paths in a parallel
> port interface have an extra signal inversion, so what is
> described as a one or zero at the parallel port connector
> may show up as the reverse when software reads the interface
> registers.
>
> Regards,
> Steve Stallings
> www.PMDX.com
>
>
>> -----Original Message-----
>> From: Marius Liebenberg [mailto:[email protected]]
>> Sent: Sunday, May 18, 2014 4:12 AM
>> To: [email protected]
>> Subject: Re: [Emc-developers] New function for parport driver required
>>
>> Partly as you say but it only puts the control lines into
>> input mode so
>> that the port has 8 out and 9 input lines. To use it as bidirectional
>> pins the approach is slightly different. So I have to add the
>> bi-directional pins as a config option and then they must be
>> handled is
>> input or output as required.
>> Currently the code only handles input or output mode. I
>> started with the
>> corrections already and might ask for some testing help later.
>>
>> One wire will work if we have bi-dir pins. Just need a driver
>> comp then.
>> Once I have the bi-dir stuff sorted I can continue with the
>> I2C driver.
>>
>>
>> On 2014-05-18 09:23, Andy Pugh wrote:
>>>> On 17 May 2014, at 22:55, Marius Liebenberg
>> <[email protected]> wrote:
>>>> I need an added feature for the parport driver namely to
>> be able to put
>>>> the control pins in bi-directional mode in order to drive
>> i2c devices.
>>> I think that this is partly implemented already as the "x" mode.
>>>
>>> It might also be a useful feature for bit-banging onewire.
>>>
>>>
>> --------------------------------------------------------------
>> ----------------
>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing
>> - For FREE
>>> Instantly run your Selenium tests across 300+ browser/OS combos.
>>> Get unparalleled scalability from the best Selenium testing
>> platform available
>>> Simple to use. Nothing to install. Get started now for free."
>>> http://p.sf.net/sfu/SauceLabs
>>> _______________________________________________
>>> Emc-developers mailing list
>>> [email protected]
>>> https://lists.sourceforge.net/lists/listinfo/emc-developers
>>>
>> -- 
>>
>> Regards /Groete
>>
>> Marius D. Liebenberg
>> +27 82 698 3251
>> +27 12 743 6064
>> QQ 1767394877
>>
>>
>> --------------------------------------------------------------
>> ----------------
>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>> Instantly run your Selenium tests across 300+ browser/OS combos.
>> Get unparalleled scalability from the best Selenium testing
>> platform available
>> Simple to use. Nothing to install. Get started now for free."
>> http://p.sf.net/sfu/SauceLabs
>> _______________________________________________
>> Emc-developers mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/emc-developers
>>
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Emc-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/emc-developers
>

-- 

Regards /Groete

Marius D. Liebenberg
+27 82 698 3251
+27 12 743 6064
QQ 1767394877


------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to