#emc-devel | Logs for 2008-04-17

[03:16:38] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[04:47:08] <SWPLinux> so, here's an interesting board to offload EMC2 onto: http://www.micromint.com/products/electrum200.html
[04:47:57] <SWPLinux> may not be fast enough for the trajectory planner, but should handle everything after that OK (though I don't know if that CPU has floating point. PID could be "interesting" if it doesn't)
[21:02:59] <SWPLinux> strange. halmeter doesn't seem to deal with s32 correctly
[21:03:06] <SWPLinux> the numbers are printed as u32
[21:05:15] <SWPLinux> oh. it's a 64-bit artifact. bummer
[21:06:13] <alex_joni> hrmm
[21:07:45] <SWPLinux> unsigned long and signed long aren't the right size on a 64-bit system, so the 32-bit negative number is printed as 2^32-ish
[21:10:53] <alex_joni> is that true only for halmeter?
[21:11:12] <alex_joni> or does it show up during halcmd show foo too?
[21:11:26] <SWPLinux> one sec
[21:12:07] <SWPLinux> halcmd is fine
[21:12:22] <alex_joni> * alex_joni points at halmeter
[21:12:57] <SWPLinux> sure, I looked at the source, and it uses {un,}signed long casts
[21:13:06] <SWPLinux> haven't looked at halcmd yet
[21:25:12] <CIA-30> EMC: 03swpadnos 07TRUNK * 10emc2/src/hal/utils/meter.c: Make halmeter display negative s32 values correctly on 64-bit systems
[22:55:18] <cradek> I've had a contribution of a mux8 component. what do you all think about whether or not this belongs in trunk?
[22:58:26] <SWPLinux> I think I'd rather generalize the mux component so you could do something like loadrt mux sizes=1,2,3,4
[22:58:36] <SWPLinux> (maybe have the size be the number of input bits - don't know about that)
[22:58:51] <cradek> interesting idea
[22:59:15] <SWPLinux> that can be done with pretty much all of the smaller components: and2, or2, sum2 ...
[22:59:19] <cradek> (that would break configs)
[22:59:22] <SWPLinux> yes
[22:59:40] <SWPLinux> unless default is sizes=2 (or 1 for number of bits)
[22:59:54] <SWPLinux> actually, it would be called "mux", but mux2 would be deprecated
[23:00:05] <cradek> I think logic is a superset of all those others, only a bit harder to use for basic stuff
[23:00:17] <SWPLinux> logic of course works for only boolean things
[23:00:23] <cradek> sure
[23:00:39] <cradek> oh, sum2, I only read and/or
[23:00:43] <SWPLinux> heh
[23:01:23] <SWPLinux> I guess sum and mult are the only mathematical ones there
[23:03:01] <SWPLinux> I'm not opposed to adding it, but I prefer the more general solution (not that I've written it yet :) )
[23:04:38] <cradek> you go to the cvs with the component you have
[23:05:23] <cradek> the best things to come out of this presidency are sound bites that we misquote/laugh about forever
[23:05:37] <cradek> how can we ever forget "the internets"
[23:06:24] <SWPLinux> fool me once, shame on you. fool me twice - uh, well - you can't fool me again
[23:06:47] <cradek> I think it's "can't get fooled again" - yeah that's another good one
[23:07:28] <SWPLinux> well, there's the Who song "Won't Get Fooled Again"
[23:07:43] <SWPLinux> but the saying says "... fool me twice, shame on you"
[23:07:46] <cradek> uh, what were they thinking?: http://www.samsoncables.com/catalog/PRODUCTS/hp33s_closeup.jpg
[23:07:54] <SWPLinux> too bad bonehead couldn't remember
[23:08:13] <SWPLinux> it's like a klingon calculator
[23:08:28] <SWPLinux> or cylon
[23:08:44] <SWPLinux> err, shame on me
[23:08:47] <SWPLinux> duh
[23:08:55] <SWPLinux> maybe I should brew that cup of coffee now
[23:09:02] <cradek> the internets say: "There's an old saying in Tennessee -- I know it's in Texas, probably in Tennessee -- that says, fool me once, shame on -- shame on you. Fool me -- you can't get fooled again."
[23:09:33] <SWPLinux> yep
[23:09:36] <SWPLinux> that's the one
[23:09:38] <SWPLinux> (of many)