#linuxcnc-devel Logs

Oct 13 2017

08:36 AM skunkworks: pcw_home, https://groups.yahoo.com/neo/groups/mach1mach2cnc/conversations/messages/153486
08:38 AM skunkworks: (pokeys guy made a comment that linuxcnc requires a pci or such interface - no ethernet. I posted that indeed there is ethernet solutions for linuxcnc from mesa)
08:38 AM skunkworks: (Johnd from the mailing list was asking on the mach list if anyone made hardware that worked with both linuxcnc and mach)
08:39 AM skunkworks: I still think steve stallings could do it - he has the hard part done with the mach setup
08:44 AM skunkworks: (well - I assume making the motion part work in the external hardware is easier than doing the linuxcnc thing - but I could be wrong...)
09:42 AM cradek: beagles are an extra-hard-to-use kind of computer, and machinekit is an extra-hard-to-use-because-the-docs-are-bad fork of linuxcnc
10:24 AM skunkworks: gene gene gene...
10:52 AM pcw_home: I think Mesa shares some blame in making configurations difficult/complex
02:03 PM MrHindsight is now known as CaptHindsight
04:48 PM jepler: if you don't think you need a laser cutter, https://www.youtube.com/watch?v=B0h_WkJGZ28
04:59 PM cradek: that's awfully cute
05:00 PM andypugh: There’s probably an optimum wavelength for popcorn.
05:01 PM seb_kuzminsky: andypugh: my guess is about 12 cm
05:02 PM cradek: heh is that what a microwave does?
05:02 PM seb_kuzminsky: yeah :-)
05:03 PM cradek: what luck! intelligent design!
05:25 PM andypugh: In this caee. Yes.
05:26 PM andypugh: (or are you perhaps sugesting that corn kernels were designed by the FSM specifically to absorb microwaves?
05:35 PM andypugh: cradek: I modified my touchy so that the FO and SO buttons update when halui changes the overrides. Would you be OK with me pushing that change?
05:36 PM andypugh: (I wanred that because I actually habe 4 mpgs rather than the single one that touchy expects)
05:36 PM andypugh: (t, v)
05:48 PM pcw_mesa: First sserial update in a while = external threshold mode for 7I84
05:48 PM pcw_mesa: if OK will add to 7I76, 7I76E, 7I77, 7I70
05:52 PM andypugh: I have the dpll oddity next on my list.
05:52 PM pcw_mesa: oh the timer 2/3 issue?
05:53 PM andypugh: Yes
05:53 PM pcw_mesa: or 1/3 I forget
05:53 PM andypugh: It’s the even-numbered ones if you don’t set up the odd ones, as I understand it.
05:54 PM andypugh: 50/50 its
05:54 PM andypugh: in the firmware?
05:54 PM pcw_mesa: it was not immediately obvious from the code what was wrong but if you read back the registers its really apparent
05:55 PM andypugh: Indeed. But as far as I can see (cursory examination) the driver sends the right data.
05:55 PM andypugh: I will get to the bottom of it, but not until sunday at the very earliest.
05:55 PM pcw_mesa: the 2 16 bit timer registers are packed into 2 32 bit hardware registers and somehow the logic for updating these fails
05:56 PM pcw_mesa: the 4 16 bit I mean
05:57 PM andypugh: It’s possible the problem is in the “would be TRAM if we had ever got round to doing that bit” code.
05:58 PM andypugh: All the drivers assume that TRAM is working, but nothing has ever seemed to need it.
05:59 PM pcw_mesa: I forget what the code does but to avoid a hardware read I would assume there were 2 or 4 shadow registers that get updated when a timer is changed
05:59 PM pcw_mesa: and then this triggers the appropriate 32 bit hardware write
06:01 PM pcw_mesa: As it is now it can be made to work if you do things in a specific order