| *** tpb has joined #yosys | 00:00 | |
| *** mwk has quit IRC | 00:04 | |
| *** dh73 has quit IRC | 00:07 | |
| *** mwk has joined #yosys | 00:33 | |
| *** Cerpin has quit IRC | 00:50 | |
| *** citypw has joined #yosys | 01:49 | |
| *** emeb has left #yosys | 01:57 | |
| *** awordnot has quit IRC | 02:03 | |
| *** N2TOH has quit IRC | 02:14 | |
| *** N2TOH has joined #yosys | 02:14 | |
| *** Cerpin has joined #yosys | 02:15 | |
| *** cr1901_modern has quit IRC | 02:18 | |
| *** attie has joined #yosys | 02:52 | |
| *** attie has quit IRC | 02:57 | |
| *** awordnot has joined #yosys | 03:16 | |
| *** rohitksingh has joined #yosys | 03:27 | |
| *** cr1901_modern has joined #yosys | 03:47 | |
| *** dh73 has joined #yosys | 04:18 | |
| *** dh73 has quit IRC | 05:14 | |
| *** rohitksingh has quit IRC | 05:19 | |
| *** heijligen has quit IRC | 06:05 | |
| *** piegames1 has quit IRC | 06:05 | |
| *** X-Scale has quit IRC | 06:16 | |
| *** [X-Scale] has joined #yosys | 06:17 | |
| *** heijligen has joined #yosys | 06:28 | |
| *** futarisIRCcloud has joined #yosys | 07:16 | |
| *** fevv8[m] has quit IRC | 07:59 | |
| *** ric96 has quit IRC | 07:59 | |
| *** tannewt has quit IRC | 07:59 | |
| *** dxld has quit IRC | 08:00 | |
| *** Wolf480pl has quit IRC | 08:00 | |
| *** tux3 has quit IRC | 08:00 | |
| *** Kokjo has quit IRC | 08:00 | |
| *** ZirconiumX has quit IRC | 08:00 | |
| *** svenn has quit IRC | 08:00 | |
| *** pepijndevos[m] has quit IRC | 08:00 | |
| *** forrestv has quit IRC | 08:00 | |
| *** TD-Linux has quit IRC | 08:00 | |
| *** grummel has quit IRC | 08:00 | |
| *** Ultrasauce has quit IRC | 08:00 | |
| *** flokli has quit IRC | 08:00 | |
| *** ZipCPU has quit IRC | 08:00 | |
| *** jhol has quit IRC | 08:00 | |
| *** knielsen has quit IRC | 08:00 | |
| *** kristianpaul has quit IRC | 08:00 | |
| *** rjo has quit IRC | 08:02 | |
| *** nrossi has quit IRC | 08:03 | |
| *** jryans has quit IRC | 08:03 | |
| *** ZirconiumX has joined #yosys | 08:04 | |
| *** svenn has joined #yosys | 08:04 | |
| *** flokli has joined #yosys | 08:04 | |
| *** forrestv has joined #yosys | 08:04 | |
| *** TD-Linux has joined #yosys | 08:04 | |
| *** ZipCPU has joined #yosys | 08:04 | |
| *** grummel has joined #yosys | 08:04 | |
| *** Ultrasauce has joined #yosys | 08:04 | |
| *** jhol has joined #yosys | 08:04 | |
| *** knielsen has joined #yosys | 08:04 | |
| *** kristianpaul has joined #yosys | 08:04 | |
| *** pie_[bnc] has joined #yosys | 08:05 | |
| *** az0re has joined #yosys | 08:12 | |
| *** ric96 has joined #yosys | 08:46 | |
| *** tannewt has joined #yosys | 08:46 | |
| *** dxld has joined #yosys | 08:46 | |
| *** Wolf480pl has joined #yosys | 08:46 | |
| *** tux3 has joined #yosys | 08:46 | |
| *** Kokjo has joined #yosys | 08:46 | |
| *** vidbina_ has joined #yosys | 08:57 | |
| *** _whitelogger has quit IRC | 09:07 | |
| *** _whitelogger has joined #yosys | 09:09 | |
| *** dys has joined #yosys | 10:09 | |
| *** indy has quit IRC | 10:51 | |
| *** attie has joined #yosys | 10:54 | |
| *** attie has quit IRC | 10:58 | |
| *** alexhw has quit IRC | 11:03 | |
| *** fsasm has joined #yosys | 11:04 | |
| *** piegames1 has joined #yosys | 11:18 | |
| *** vidbina_ has quit IRC | 11:56 | |
| *** fevv8[m] has joined #yosys | 12:41 | |
| *** rjo has joined #yosys | 12:41 | |
| *** nrossi has joined #yosys | 12:41 | |
| *** ematrix has joined #yosys | 12:41 | |
| *** promach3 has joined #yosys | 12:41 | |
| *** pepijndevos[m] has joined #yosys | 12:41 | |
| *** jryans has joined #yosys | 12:41 | |
| *** indy has joined #yosys | 12:48 | |
| *** develonepi3 has quit IRC | 13:55 | |
| *** captain_morgan20 has quit IRC | 14:18 | |
| *** captain_morgan20 has joined #yosys | 14:21 | |
| *** oldtopman has quit IRC | 14:34 | |
| *** captain_morgan20 has quit IRC | 14:39 | |
| *** _whitelogger has quit IRC | 14:55 | |
| *** _whitelogger has joined #yosys | 14:57 | |
| *** vidbina_ has joined #yosys | 15:19 | |
| *** emeb has joined #yosys | 15:34 | |
| *** alexhw has joined #yosys | 15:43 | |
| *** rohitksingh has joined #yosys | 15:43 | |
| *** citypw has quit IRC | 15:43 | |
| *** dh73 has joined #yosys | 16:02 | |
| *** rohitksingh has quit IRC | 16:51 | |
| *** vidbina_ has quit IRC | 16:56 | |
| *** vidbina_ has joined #yosys | 17:00 | |
| *** rohitksingh has joined #yosys | 17:10 | |
| *** dys has quit IRC | 17:48 | |
| *** bluesceada has quit IRC | 17:55 | |
| *** bluesceada has joined #yosys | 18:01 | |
| *** rohitksingh has quit IRC | 18:08 | |
| *** jwhitmore has joined #yosys | 18:18 | |
| *** attie has joined #yosys | 18:55 | |
| *** [X-Scale] is now known as X-Scale | 18:57 | |
| *** jwhitmore has quit IRC | 18:59 | |
| *** attie has quit IRC | 19:00 | |
| *** rohitksingh has joined #yosys | 19:13 | |
| *** blackbit has left #yosys | 19:20 | |
| *** rohitksingh has quit IRC | 19:24 | |
| *** dys has joined #yosys | 19:35 | |
| *** rohitksingh has joined #yosys | 19:37 | |
| *** cr1901_modern has quit IRC | 19:52 | |
| *** az0re has quit IRC | 19:53 | |
| *** fsasm has quit IRC | 19:57 | |
| *** cr1901_modern has joined #yosys | 19:57 | |
| *** emily is now known as emilazy | 20:06 | |
| *** ematrix is now known as emily | 20:06 | |
| *** az0re has joined #yosys | 20:13 | |
| *** dys has quit IRC | 20:14 | |
| *** rohitksingh has quit IRC | 20:21 | |
| *** emily is now known as ematrix | 20:32 | |
| *** emilazy is now known as emily | 20:33 | |
| *** emily is now known as emilazy | 20:33 | |
| *** ematrix is now known as emily | 20:33 | |
| *** emeb has left #yosys | 21:00 | |
| *** awygle has joined #yosys | 21:05 | |
| awygle | Can I communicate to yosys/abc/nextpnr that the calculation I've described must complete within X cycles of a Y MHz clock, please optimize it for area? | 21:05 |
|---|---|---|
| tnt | within X cycles ? nextpnr doesn't have multi-cycle constraint in the first place. | 21:06 |
| daveshah | Yeah, nextpnr doesn't support this yet | 21:07 |
| awygle | that's more or less what i expected | 21:07 |
| awygle | thanks tho | 21:07 |
| awygle | i was thinking maybe abc's "retiming" stuff could be used for this somehow | 21:07 |
| daveshah | You could probably give abc a delay constraint of X/Y | 21:07 |
| sorear | what tools *do* have that kind of multi-cycle constraint? | 21:07 |
| daveshah | But abc's delay constraints don't tend to yield vast area improvements | 21:07 |
| daveshah | At least some of the vendor onez | 21:08 |
| tnt | awygle: well it's not the same thing as retiming. | 21:08 |
| daveshah | *ones | 21:08 |
| awygle | another channel recommended trying vivado hls | 21:08 |
| tnt | that's _yet_ another different thing than multi-cycle constraints ... | 21:08 |
| awygle | tbf, i didn't say "multi cycle constraints", you did | 21:09 |
| tnt | retiming is moving pipeline register around to try and equalize the paths between FFs. | 21:09 |
| sorear | do you have throughput requirements? | 21:09 |
| awygle | you can see the relation though, yes? i want to add or remove pipeline registers to try to minimize total area. that the one might apply to the other is not a crazy thought. | 21:10 |
| tnt | multi-cycle constraint is allowing the comb logic to take more time between FFs than the normal clock period because you guarantee externaly, somehow, that the input is going to be state for N cycles on you will only clock enable the output after N cycles. | 21:10 |
| tnt | HLS will potentially generate completely different logic and data path and everything depending on a set of parameters. | 21:11 |
| awygle | sorear: kind of? i have a calculation, which must happen within (say) 16 clock cycles. after 16 clock cycles i should be able to launch the calculation again. that's all i really want. | 21:11 |
| daveshah | abc's retiming is pretty crap tbh | 21:11 |
| daveshah | There was talk of doing retiming in Yosys itself but that seems to have stalled, there isn't even a timing framework yet afaics | 21:11 |
| tnt | also in fpga, pipeline registers can often be free since they're already after each lut wether you use them or not. | 21:12 |
| daveshah | Yep | 21:13 |
| daveshah | I think doing post placement retiming in nextpnr could be interesting for larger devices | 21:13 |
| daveshah | Particularly once you get into Xilinx "3D" die crossing | 21:13 |
| daveshah | But that's a digression | 21:13 |
| awygle | I'll play with the hls stuff, see how that goes. Thanks all. | 21:16 |
| tnt | So realistically area really won't change drastically. If you have 16 cycles the only thing that would change it a lot is if you can do resource re-use / sharing without paying too much for the muxes needed for that, but no tool is going to do that for you, that's your job as the designer. | 21:16 |
| awygle | a tool absolutely should be able to tell me that, it's a completely deterministically knowable thing. | 21:17 |
| tnt | Feel free to write it. | 21:19 |
| *** cr1901_modern has quit IRC | 21:19 | |
| *** vidbina_ has quit IRC | 21:19 | |
| *** alexhw has quit IRC | 21:19 | |
| *** ebb has quit IRC | 21:19 | |
| *** srk has quit IRC | 21:19 | |
| *** simeonm has quit IRC | 21:19 | |
| *** adjtm has quit IRC | 21:19 | |
| *** cyrozap has quit IRC | 21:19 | |
| *** gorbak25 has quit IRC | 21:19 | |
| *** FL4SHK has quit IRC | 21:19 | |
| *** az0re has quit IRC | 21:20 | |
| *** adjtm has joined #yosys | 21:23 | |
| *** gorbak25 has joined #yosys | 21:27 | |
| *** dys has joined #yosys | 21:43 | |
| *** cr1901_modern has joined #yosys | 21:47 | |
| *** alexhw has joined #yosys | 22:03 | |
| *** ebb has joined #yosys | 22:03 | |
| *** srk has joined #yosys | 22:03 | |
| *** simeonm has joined #yosys | 22:03 | |
| *** cyrozap has joined #yosys | 22:03 | |
| *** FL4SHK has joined #yosys | 22:03 | |
| *** rohitksingh has joined #yosys | 22:17 | |
| *** rohitksingh has quit IRC | 22:54 | |
| *** rohitksingh has joined #yosys | 23:01 | |
| *** s_frit_ has joined #yosys | 23:26 | |
| *** s_frit has quit IRC | 23:27 | |
| *** rohitksingh has quit IRC | 23:45 | |
| *** dh73 has quit IRC | 23:47 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!