Re: [Emc-developers] patch for parallel port pin assignment

[Joseph H. B. <as...@co...>]

jmk...@at... wrote:

> Do you already have it in "patch" format - ie. a list of source
> files that get modified, and the mods?  The reason I ask is 

Yes.  I amn working off a local CVS copy and will submit cvs diff
output if that's what you want.

> most of the developers have local CVS working copies.  We make
> changes to our local copies, compile and test, then do a CVS
> commit.  To import your patch, somebody will need to apply it
> to their CVS working copy and commit it.  That somebody could
> be you if we make you a developer.

That would be fine with me.  I do software development for a living
so have some idea what's going on here.

> Are your changes documented?  Ideally you also have patches
> to update the LyX documentation.  Failing that, make sure
> the code is well commented.  It is especially important to
> describe any added ini file items in detail.

I haven't touched the LyX files, but can if that's what's needed.

> Can you briefly describe how your changes work?

Currently there are a few places where there are hard-coded assumptions
about which axis and which bits (step/direction) connect to which
parallel port bits.

The change I made was to compute pin assignment tables at module load time
based on information in the ini file, extracted by the run script
and passed in as a module parameter.  I believe it should improve
performance somewhat because it changes the computation of the
port bit numbers to simple table lookups.

> How have you tested them?

Basic testing only -- it runs my Taig mill via a xylotex card, which
has the step/direction bits the opposite of emc's convention.

> Sorry for all the questions - it's not often that a new person 
> pops in with something already finished!

Let me know how you want to proceed.  Typical for open-source
projects is: someone reviews my patches and I make changes until
they are satisfied, then they commit them for me.
-- 
Joe Buehler