| *** tpb has joined #yosys | 00:00 | |
| *** emeb_mac has joined #yosys | 00:00 | |
| *** promach__ has joined #yosys | 00:25 | |
| *** quigonjinn has quit IRC | 00:25 | |
| *** promach__ is now known as promach2 | 00:30 | |
| *** emeb has quit IRC | 02:30 | |
| *** cemerick_ has joined #yosys | 02:40 | |
| *** xrexeon has quit IRC | 02:43 | |
| *** jkiv has quit IRC | 04:25 | |
| *** leviathan has joined #yosys | 04:31 | |
| *** cemerick_ has quit IRC | 04:36 | |
| *** leviathan has quit IRC | 04:45 | |
| *** leviathan has joined #yosys | 04:46 | |
| *** X-Scale has quit IRC | 05:51 | |
| *** cemerick_ has joined #yosys | 05:55 | |
| *** leviathan has quit IRC | 06:02 | |
| *** leviathan has joined #yosys | 06:05 | |
| *** cemerick_ has quit IRC | 06:28 | |
| *** dys has joined #yosys | 06:32 | |
| *** GuzTech has joined #yosys | 06:58 | |
| *** dys has quit IRC | 06:58 | |
| *** jpo has quit IRC | 07:02 | |
| *** emeb_mac has quit IRC | 07:06 | |
| *** dys has joined #yosys | 07:15 | |
| *** jwhitmore has joined #yosys | 07:22 | |
| *** jwhitmore has quit IRC | 07:26 | |
| *** fouric has quit IRC | 08:08 | |
| *** fouric has joined #yosys | 08:09 | |
| *** quigonjinn has joined #yosys | 08:16 | |
| *** leviathan has quit IRC | 08:26 | |
| *** proteusguy has quit IRC | 08:37 | |
| *** proteusguy has joined #yosys | 08:40 | |
| *** dmin7 has joined #yosys | 08:43 | |
| *** leviathan has joined #yosys | 09:04 | |
| *** proteus-guy has quit IRC | 09:23 | |
| *** proteus-guy has joined #yosys | 09:24 | |
| *** proteusguy has quit IRC | 09:41 | |
| *** fsasm has joined #yosys | 09:53 | |
| *** mazzoo has quit IRC | 10:20 | |
| *** mazzoo has joined #yosys | 10:22 | |
| *** promach2 has quit IRC | 10:24 | |
| *** proteusguy has joined #yosys | 10:31 | |
| *** promach2 has joined #yosys | 10:37 | |
| *** xrexeon has joined #yosys | 10:38 | |
| *** promach2 has quit IRC | 10:48 | |
| *** promach__ has joined #yosys | 10:49 | |
| *** promach__ is now known as promach2 | 10:49 | |
| *** promach2 has quit IRC | 10:54 | |
| *** fsasm has quit IRC | 10:56 | |
| *** mjoldfield has quit IRC | 11:11 | |
| *** mjoldfield has joined #yosys | 11:42 | |
| *** jwhitmore has joined #yosys | 11:55 | |
| *** jwhitmore has quit IRC | 12:00 | |
| *** dys has quit IRC | 12:14 | |
| *** dmin7 has quit IRC | 12:39 | |
| *** eduardo__ has joined #yosys | 12:43 | |
| *** eduardo_ has quit IRC | 12:46 | |
| *** dys has joined #yosys | 12:54 | |
| *** ssvb has quit IRC | 14:00 | |
| *** emeb has joined #yosys | 14:01 | |
| *** fsasm has joined #yosys | 14:14 | |
| mattvenn | quit | 14:21 |
|---|---|---|
| *** mattvenn has quit IRC | 14:21 | |
| *** jayaura has quit IRC | 14:43 | |
| *** cfelton has quit IRC | 14:45 | |
| *** marex-cloud has quit IRC | 14:45 | |
| *** _florent_ has quit IRC | 14:46 | |
| *** cfelton has joined #yosys | 14:46 | |
| *** marex-cloud has joined #yosys | 14:46 | |
| *** mithro has quit IRC | 14:47 | |
| *** _florent_ has joined #yosys | 14:48 | |
| *** mithro has joined #yosys | 14:50 | |
| *** jayaura has joined #yosys | 14:50 | |
| *** jayaura is now known as Guest82714 | 14:50 | |
| *** fsasm has quit IRC | 14:57 | |
| emeb | Skimmed thru this: https://arxiv.org/abs/1805.03648 | 15:00 |
| tpb | Title: [1805.03648] Parallel Programming for FPGAs (at arxiv.org) | 15:00 |
| *** promach__ has joined #yosys | 15:00 | |
| *** promach__ is now known as promach2 | 15:00 | |
| emeb | Wondering how much HLS really improves overall design quality / productivity. | 15:00 |
| promach2 | For https://gist.github.com/promach/454cba2c790db7925e176eee61bb1fe5#file-test_uart-v-L192, why am I having this error "ERROR: Left operand of replicate expression is not constant at ../rtl/test_UART.v:192" ? | 15:01 |
| tpb | Title: test_UART.v ยท GitHub (at gist.github.com) | 15:01 |
| ZipCPU | promach2: Ahm ... because it isn't constant? The error is pretty self explanatory. | 15:11 |
| promach2 | ZipCPU: huh ? I am using genvar, why is it not constant ? | 15:16 |
| shapr | I had a shower thought yesterday about FPGA layout | 15:16 |
| shapr | but I'm not quite sure how to turn it into words | 15:16 |
| shapr | first thought was, place and route on a sphere would be much easier, especially since the edges connect | 15:17 |
| daveshah | promach2: because you are not using the genvar directly, but as an index to received_data_zeroes which is non-constant | 15:17 |
| shapr | that turned into thinking about n-dimensional surface layout, I suspect it'd be easy to do, but I'm not sure about bringing it back down from higher dimensions | 15:18 |
| ZipCPU | shapr: My thought was that the first step to P&R should be an aggregation step, words should be placed together for example. | 15:18 |
| ZipCPU | That would solve any carry chain issues. | 15:18 |
| shapr | ZipCPU: elucidate? | 15:18 |
| qu1j0t3 | there's a LOT of literature on this problem :) | 15:18 |
| ZipCPU | qu1j0t3: And I haven't read any of it! ;D | 15:18 |
| shapr | I was thinking about how SA gets stuck in local minima, I suspect doing layout in more dimensions and then reducing down to two may help with that | 15:19 |
| ZipCPU | shapr: Ooohh, yea, okay ... fascinating thought. | 15:20 |
| qu1j0t3 | there are some other algorithms you could look at -- cuckoo search etc | 15:20 |
| shapr | related to cuckoo hashing, probably? | 15:20 |
| qu1j0t3 | no | 15:20 |
| emeb | "reducing down to two" <- there's the rub. | 15:20 |
| shapr | emeb: yeah, agreed | 15:20 |
| ZipCPU | My thought was that: reg [15:0] A; always @(posedge i_clk) A <= A + 1; requires a series of logic elements that need to be placed near each other. The first step should therefore be a block placement algorithm. | 15:21 |
| shapr | first thought about that was 'sticking down' part of a spherical layout, then re-running SA in 2D | 15:21 |
| qu1j0t3 | the thing is "stuck in the wrong local minimum" is literally the issue, because the thing is NP-complete. | 15:21 |
| qu1j0t3 | you can only mitigate it | 15:21 |
| qu1j0t3 | & i think there are things you can do to help SA out? | 15:21 |
| * ZipCPU pulls at his beard | 15:21 | |
| ZipCPU | SA? | 15:22 |
| ZipCPU | Situational awareness? | 15:22 |
| qu1j0t3 | shapr: do you mean a torus? | 15:22 |
| qu1j0t3 | ZipCPU: simulated annealing | 15:22 |
| ZipCPU | Ahh ... thanks | 15:22 |
| shapr | torus would be fine too | 15:22 |
| shapr | just anything that makes the search easier at first | 15:22 |
| shapr | I also thought that a generic place-and-route might start with zero layout constraints on a sphere, just shrinking the sphere down to the perfect size to fit the layout | 15:23 |
| shapr | and then add in constraints | 15:23 |
| shapr | such as output pins and that kind of thing | 15:23 |
| *** emeb has quit IRC | 15:24 | |
| shapr | now I wish I had a BLIF parser in Haskell | 15:24 |
| * shapr digs into https://github.com/satnam6502/lava | 15:25 | |
| promach2 | daveshah: thanks for the tips | 15:27 |
| shapr | qu1j0t3: what's the trade-off between a torus and sphere? | 15:28 |
| *** cemerick_ has joined #yosys | 15:30 | |
| qu1j0t3 | if you are working 2 dimensional ordinates you might have some trouble mapping to a sphere? | 15:31 |
| qu1j0t3 | there's also particle swarm optimisation. if i were doing this, i'd probably develop metrics and constraints i could use with pluggable algorithms, as there are quite a few possibilities | 15:31 |
| qu1j0t3 | but things like cuckoo search and PSO look promising to me if SA isn't cutting it. but you will definitely want to review literature so you get some benefit from the large investments already made | 15:33 |
| shapr | yes, I want a place and route experimental toolkit | 15:34 |
| *** emeb_mac has joined #yosys | 15:37 | |
| *** thoughtpolice has quit IRC | 15:37 | |
| *** elms has quit IRC | 15:37 | |
| *** furan-- has quit IRC | 15:37 | |
| *** lvrp16 has quit IRC | 15:37 | |
| *** thoughtpolice has joined #yosys | 15:40 | |
| *** danieljabailey has quit IRC | 15:41 | |
| *** furan-- has joined #yosys | 15:41 | |
| *** lvrp16 has joined #yosys | 15:42 | |
| *** danieljabailey has joined #yosys | 15:43 | |
| *** elms has joined #yosys | 15:43 | |
| shapr | awygle: were you able to find the source code for the GPU star+ place-and-route paper? | 16:04 |
| shapr | I found the primary author's github account, but don't see the source there | 16:04 |
| *** promach2 has quit IRC | 16:05 | |
| *** jkiv has joined #yosys | 16:17 | |
| *** X-Scale has joined #yosys | 16:36 | |
| shapr | is there a specification for the blif format? | 16:48 |
| *** emeb_mac has quit IRC | 16:51 | |
| awygle | shapr: all i know of is https://www.cse.iitb.ac.in/~supratik/courses/cs226/spr16/blif.pdf | 17:02 |
| awygle | and no, i didn't acquire the source. if you pinged that author and they could provide it that would be great | 17:03 |
| *** m_t has joined #yosys | 17:33 | |
| *** jwhitmore has joined #yosys | 17:45 | |
| *** leviathan has quit IRC | 17:54 | |
| *** jwhitmore has quit IRC | 18:18 | |
| *** dmin7 has joined #yosys | 18:30 | |
| *** jwhitmore has joined #yosys | 18:43 | |
| *** proteusguy has quit IRC | 18:51 | |
| *** cemerick_ has quit IRC | 18:52 | |
| *** dxld has quit IRC | 19:03 | |
| *** proteusguy has joined #yosys | 19:04 | |
| *** dxld has joined #yosys | 19:14 | |
| *** pie__ has joined #yosys | 19:56 | |
| *** pie_ has quit IRC | 20:00 | |
| *** jkiv has quit IRC | 20:08 | |
| *** jkiv has joined #yosys | 20:18 | |
| *** jkiv has quit IRC | 20:25 | |
| *** jkiv has joined #yosys | 20:54 | |
| *** jkiv has joined #yosys | 20:55 | |
| *** jkiv has quit IRC | 20:55 | |
| *** jkiv has joined #yosys | 20:56 | |
| *** jkiv has joined #yosys | 20:57 | |
| *** jkiv has quit IRC | 20:57 | |
| *** jkiv has joined #yosys | 20:58 | |
| *** jkiv has joined #yosys | 20:59 | |
| *** jkiv has quit IRC | 20:59 | |
| *** dxld has quit IRC | 21:09 | |
| *** dxld has joined #yosys | 21:09 | |
| *** m_t has quit IRC | 21:17 | |
| *** cr1901_modern has quit IRC | 21:33 | |
| *** emeb_mac has joined #yosys | 21:36 | |
| *** dys has quit IRC | 21:43 | |
| *** GuzTech has quit IRC | 21:59 | |
| *** jwhitmore has quit IRC | 22:51 | |
| *** dxld has quit IRC | 23:45 | |
| *** dxld has joined #yosys | 23:51 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!