On Monday 07 November 2016 12:43:19 Gene Heskett wrote:

> On Monday 07 November 2016 10:02:02 Peter C. Wallace wrote:
> > On Mon, 7 Nov 2016, Gene Heskett wrote:
> > > Date: Mon, 7 Nov 2016 04:26:21 -0500
> > > From: Gene Heskett <[email protected]>
> > > Reply-To: "Enhanced Machine Controller (EMC)"
> > >     <[email protected]>
> > > To: [email protected]
> > > Subject: [Emc-users] unbricking the pi
> > >
> > > Greetings all pi players;
> > >
> > > Is there a key to be held down during boot that will give me the
> > > effect of adding single to the kernel command line, effective
> > > bypassing the login so I can stand a snowballs chance of running
> > > raspi-config and undoing the dumbassed mistake I made?
> > >
> > > Then is there anything that can be done to make the wireless
> > > keyboards work?  Wally didn't have any more of the logitech k-360,
> > > so I bought an unknown "darkweb" that looks like a clone of the
> > > k-360, but has a smaller numbpad with a touchpad under it, about
> > > 2" wide by 1.35" high thats worthless as a full width stroke only
> > > moves the mouse about an inch on the monitor screen.  And that is
> > > still subject to the cold molasses motion.
> > >
> > > Second unrelated question, I see in the x86 /lib/modules tree,
> > > 3 "*spi.ko" modules.  Can one of them do this spi over an epp
> > > parport? Any documentation there may be, has never made it to
> > > /usr/share/doc
> > >
> > >
> > > Or, if Peter can tell me how to reprogram this 7i90HD back to
> > > accessing it via epp when it is now set for spi access? I am sold
> > > on the plethora of I/O it can do and would like to be able to use
> > > it on this lathe from an x86 box.
> >
> > There are 2 independent jumper selectable configs on the 7I90HD (see
> > manual) I would always leave one config in EPP  mode. If you wish to
> > reprogram the SPI config, you power cycle the 7I90HD with the jumper
> > in the EPP position move the jumper back to the SPI position and
> > then program whatever bit file you wish via EPP (never running
> > mesaflash with the jumper in the EPP position)
>
> I'll check the book, I hope that is what I did...  If that is what I
> did, I should find that jumper in the spi position NOW, right?

All jumpers are in the "up" position, and were when I flashed it.

But its now moot as I have unbricked it by editing the config.txt file in 
the boot partition, which restored the textual login.  No network to 
rather than putz with dhcp, I made it static, works a treat.

Watching the supply, its 5.07 volts and never flickers a digit.  And with 
x out of the picture, the keyboard is usable. ssh -Y raspi, and running 
linuxcnc from the milling machine, its still slow enough to pour a fresh 
cuppa while getting its screen drawn, but after that, ran the logo file 
fairly well, even at full screen. That monitor was actually a 22" 
Samsung HiDef that has a failed tuner, so is being driven at 
1920xwhatever over a vga connection from a dual core Dell 748 box.

Now I'm fiddling with the files in the linuxcnc/config/lathe directory, 
first with an eye toward making the vfd drive work, because none of the 
spindle stuff has been net'ed yet, but also to convert it from a 3 axis 
mill to a 2 axis lathe. I should probably compare those .ini files with 
the same files on TLM.

If that orange-pi can do the x stuff, I may be good to go. log into the 
orange pi from its keyboard, and ssh -Y into the raspi *might* do it. 
Needlessly complex IMO. Ought to be able to script this stuff.

Who knows, maybe the orange-pi will show up today, it was in noo joisey 
friday night.

> Thanks Peter, I may do that, but with the plethora of spi bit files,
> we users would sure love to see a human readable description of what
> each one does as opposed to putting it in then booting up lcnc to see
> what falls out, which on the pi is not in dmesg, so you have to run
> linuxcnc -l >7i90_cap.txt, and print that to get a record of what it
> has available in that bitfile.  The pdf only has 5 or so examples, but
> there are 30 or more *spi*.bit files to choose from.

Make that 15 or so after the daughter card driver versions are 
eliminated.

One of the problems with the slow startup might be that the default pi 
installed of uspace-linuxcnc, does not make an hm2 directory 
in /lib/firmware, so as an experiment to see if I can get rid of a 
warning about "hm2_7i90.0" that is output in the lower right screen pane 
on running it. So, since perms to write it directly are hard to get, I 
used mc as me to put all those *spi*.bit files in 
a /home/gene/linuxcnc/7i90 directory & softlinked that to /lib/firmware 
on the pi.

Progress I believe.  Better ideas appreciated.

> Whats in it now will work, 4 of everything,

I'll take that back, only 2 pwm's but nothing is net'ed in the hal file, 
yet.

> occupy's all of p1, so I 
> can build down from the top of p3 for my gpio uses. Thats IF I can use
> the pi's.  So far its 1 step fwd, 2 back. With 4 cores, it still can't
> read the keyboard on a full time basis, throwing away that which it
> missed. Frustrating is not an adequate description...
>
> Perhaps you can tell me which of the 3 kernel *-spi.ko files will turn
> the parport on an x86 mobo into an spi port?  That might be an
> exploreable tunnel to poke about in, or would the epp version be
> faster?
>
Thanks Peter, everybody.
[...]
Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>

------------------------------------------------------------------------------
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to