[Coco] PBJ PC PAK - Part II - Interfacing (source for driver)

[Dave R in Illinois]

After looking through most the code, it appears most of those include files
could be skipped, due to basic's direct talking with hardware.

 

Now correct me if I am wrong but I would need the following.

 

Hardware int dev

Connect dev

HW read and write regr

Dev protocol

Disconnect dev

Exit dev

 

Doesn't seem like a lot now that I think about it. Any suggestions or
anything overlooked?

 

Dave

 

 

>

  _____  

>From: Dave R in Illinois
>Sent: Thursday, January 31, 2008 4:19 PM
>Yes your assumption is correct. It is an IDE interface, that acts similar
to an SCSI, using a parallel port.


>Dave



  _____  



On Thursday 31 January 2008, Gene Heskett wrote:


>>Sorry but some odd reason, the code didn't show up, and every time I try
the 

>>message is now ignored. 

>> 

>>So here is a link to it. :-) 

>> 

>> 

>> 

>>http://lxr.oss.org.cn/source/drivers/block/paride/epat.c 

>> 

>My guess is that you will have to collect all those include files and 
>regenerate them to say the same thing in os9 style before you can get past 
>line 30 with the microware compiler plus all the extra's we have done for
it 
>over the years. Not impossible, but it will be time consuming to run all 
>those defines down and figure out what its doing. I suspect the resultant 
>driver may also be several kilobytes larger than a barebones effort in 
>assembly, mainly due to bells and whistles that may not be needed for a
given 
>application under os9/nitros9. 

>I also note that this version has huge bunches of EXPORT_SYMBOL_* stuff in
it 
>that the current 2.6.24 kernel version doesn't have, so changing to that 
>version may have a helpfull effect too. According to the 2.6.24 version, 
>Grant Guenther wrote it back in 1998 and its only been touched once since. 

>If this is the correct driver, its an IDE interface? For ls-120 drives and 
>such...