*** tpb has joined #yosys | 00:00 | |
*** emeb has left #yosys | 00:12 | |
*** gsi_ has joined #yosys | 02:35 | |
*** gsi__ has quit IRC | 02:38 | |
*** _whitelogger has quit IRC | 03:31 | |
*** _whitelogger has joined #yosys | 03:33 | |
*** _whitelogger has quit IRC | 03:40 | |
*** _whitelogger has joined #yosys | 03:42 | |
*** _whitelogger has quit IRC | 03:52 | |
*** _whitelogger has joined #yosys | 03:54 | |
*** lutsabound has quit IRC | 03:58 | |
*** pie__ has joined #yosys | 04:43 | |
*** pie___ has quit IRC | 04:47 | |
*** chaseemory has quit IRC | 04:58 | |
*** cr1901_modern has quit IRC | 05:28 | |
*** cr1901_modern has joined #yosys | 05:43 | |
*** rohitksingh has joined #yosys | 05:43 | |
*** _whitelogger has quit IRC | 06:01 | |
*** _whitelogger has joined #yosys | 06:03 | |
*** leviathanch has joined #yosys | 06:08 | |
*** rohitksingh has quit IRC | 06:14 | |
*** _whitelogger has quit IRC | 06:40 | |
*** _whitelogger has joined #yosys | 06:42 | |
*** emeb_mac has quit IRC | 07:36 | |
*** leviathanch has quit IRC | 07:55 | |
*** seldridge has quit IRC | 08:32 | |
*** rohitksingh has joined #yosys | 08:43 | |
*** m4ssi has joined #yosys | 08:51 | |
*** rohitksingh has quit IRC | 09:18 | |
*** keesj has joined #yosys | 09:37 | |
*** m4ssi has quit IRC | 09:42 | |
*** rohitksingh has joined #yosys | 09:54 | |
*** leviathanch has joined #yosys | 10:01 | |
*** rohitksingh has quit IRC | 10:05 | |
*** rohitksingh has joined #yosys | 10:38 | |
*** citypw has quit IRC | 10:39 | |
*** ZipCPU has joined #yosys | 10:45 | |
*** citypw has joined #yosys | 10:52 | |
*** rohitksingh has quit IRC | 11:13 | |
*** jevinskie has joined #yosys | 11:23 | |
*** tmeissner has joined #yosys | 12:08 | |
*** dpiegdon has joined #yosys | 14:18 | |
dpiegdon | hi there! | 14:18 |
---|---|---|
*** m4ssi has joined #yosys | 14:19 | |
dpiegdon | i have been playing around with ringoscillators, and found a working example from clifford, at http://svn.clifford.at/handicraft/2015/ringosc . | 14:25 |
tpb | Title: handicraft - Revision 2994: /2015/ringosc (at svn.clifford.at) | 14:25 |
dpiegdon | I was able to get this running with arachnepnr easily, | 14:25 |
dpiegdon | but on nextpnr, which is timing driven, I always get the error: ERROR: timing analysis failed due to presence of combinatorial loops, incomplete specification of timing ports, etc. | 14:25 |
dpiegdon | which makes sense, I guess. but is there a way to tell nextpnr to ignore timing here? or alltogether? | 14:26 |
dpiegdon | (I have tried --no-tmdriv) | 14:26 |
*** lutsabound has joined #yosys | 14:39 | |
daveshah | dpiegdon: use --force | 14:43 |
dpiegdon | daveshah: ok, then the error is reduced to a warning, but no output file is generated. | 14:48 |
daveshah | What is your full command line? | 14:49 |
dpiegdon | nextpnr-ice40 --no-tmdriv --force --hx1k --package tq144 --pcf icestick.pcf --json $< | 14:50 |
dpiegdon | i have uploaded my example to https://github.com/dpiegdon/ringoscillator | 14:50 |
tpb | Title: GitHub - dpiegdon/ringoscillator: FPGA ringoscillator example. Originally taken from http://svn.clifford.at/handicraft/2015/ringosc/ (at github.com) | 14:50 |
daveshah | You need --asc output.asc to create any output | 14:52 |
dpiegdon | damn - copy and paste error | 14:56 |
dpiegdon | thanks! | 14:56 |
dpiegdon | yes, that works. ringing at 322MHz :) | 14:58 |
*** brasilino has joined #yosys | 15:20 | |
MoeIcenowy | by the way is it a way to evalulate the frequency of such a ring oscillator? | 15:35 |
MoeIcenowy | can icetime do it? | 15:35 |
sorear | with the icetime database it is possible to produce a conservative lower limit for the frequency (upper bound on delay of the closed-loop path) | 15:37 |
sorear | ring oscillators are very sensitive to things like temperature (and are frequently used as sensors) | 15:37 |
sorear | a more sophisticated timing analysis (for hold times) would also give you a lower bound on the delay → upper limit for the frequency over the operating voltage/temperature range | 15:38 |
*** m4ssi has quit IRC | 15:41 | |
*** fsasm has joined #yosys | 15:54 | |
dpiegdon | but icetime does not produce valid results here. neither for nextpnr nor for arachnepnr. it *did* produce some valid results for an earlier example. i am going to look into this a bit. | 15:55 |
dpiegdon | so for the original version from clifford, icetime shows the (~correct) value of 8.92 MHz. but only when using arachnepnr, not for nextpnr. also when reducing the delay line, the value is rougly the lower bound, as described by sorear. | 15:59 |
dpiegdon | but when i strip the example to only the ring oscillator (remove freq counter and control), icetime is clueless and shows 0MHz. | 16:00 |
dpiegdon | the oscillator *does* work, though | 16:00 |
dpiegdon | so i assume icetime does somehow uses the freq-counter logic loop to count the generated frequency. | 16:04 |
*** emeb_mac has joined #yosys | 16:12 | |
*** seldridge has joined #yosys | 16:39 | |
*** harryinashed has joined #yosys | 17:11 | |
*** harryinashed has quit IRC | 17:19 | |
*** harryinashed_ has joined #yosys | 17:22 | |
*** fsasm_ has joined #yosys | 17:22 | |
*** harryinashed_ has quit IRC | 17:23 | |
*** harryina_ has joined #yosys | 17:23 | |
*** fsasm has quit IRC | 17:24 | |
*** harryina_ has quit IRC | 17:24 | |
*** harryina_ has joined #yosys | 17:25 | |
*** harryina_ has left #yosys | 17:26 | |
*** harryina_ has joined #yosys | 17:26 | |
*** harryina_ has quit IRC | 17:27 | |
*** seldridge has quit IRC | 17:43 | |
*** tmeissner has quit IRC | 17:51 | |
*** brasilino has quit IRC | 18:04 | |
*** leviathanch has quit IRC | 18:43 | |
*** emeb_mac has quit IRC | 19:08 | |
*** emeb has joined #yosys | 19:24 | |
*** dpiegdon has quit IRC | 20:21 | |
*** dpiegdon has joined #yosys | 20:25 | |
*** dpiegdon has quit IRC | 20:44 | |
*** cr1901_modern has quit IRC | 21:23 | |
*** chaseemory has joined #yosys | 21:55 | |
*** fsasm_ has quit IRC | 22:02 | |
*** s_frit has quit IRC | 22:45 | |
*** cr1901_modern has joined #yosys | 22:45 | |
*** s_frit has joined #yosys | 22:45 |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!