| *** tpb has joined #symbiflow | 00:00 | |
| *** futarisIRCcloud has quit IRC | 02:01 | |
| *** citypw has joined #symbiflow | 02:31 | |
| *** citypw has quit IRC | 02:50 | |
| *** Bertl is now known as Bertl_zZ | 03:05 | |
| *** futarisIRCcloud has joined #symbiflow | 03:56 | |
| *** proteusguy has quit IRC | 04:29 | |
| *** proteusguy has joined #symbiflow | 05:47 | |
| *** futarisIRCcloud has quit IRC | 07:05 | |
| *** OmniMancer has joined #symbiflow | 07:07 | |
| *** Xark has quit IRC | 08:32 | |
| *** Xark has joined #symbiflow | 08:35 | |
| *** kraiskil has joined #symbiflow | 08:43 | |
| *** citypw has joined #symbiflow | 08:50 | |
| *** proteusguy has quit IRC | 08:58 | |
| *** kraiskil has quit IRC | 08:59 | |
| *** proteusguy has joined #symbiflow | 08:59 | |
| *** citypw has quit IRC | 09:28 | |
| *** Bertl_zZ is now known as Bertl | 09:30 | |
| *** Maya-sama has joined #symbiflow | 11:25 | |
| *** Miyu has quit IRC | 11:28 | |
| *** kraiskil has joined #symbiflow | 11:30 | |
| *** hackkitten has joined #symbiflow | 11:30 | |
| *** Maya-sama has quit IRC | 11:34 | |
| *** kraiskil has quit IRC | 11:52 | |
| *** kraiskil has joined #symbiflow | 12:06 | |
| *** proteusguy has quit IRC | 12:38 | |
| *** bjorkintosh has joined #symbiflow | 12:39 | |
| *** kraiskil has quit IRC | 13:53 | |
| *** proteusguy has joined #symbiflow | 14:00 | |
| *** kraiskil has joined #symbiflow | 14:07 | |
| *** lopsided98 has quit IRC | 14:32 | |
| *** lopsided98 has joined #symbiflow | 14:34 | |
| *** Bertl is now known as Bertl_oO | 15:06 | |
| mithro | Morning! | 15:59 |
|---|---|---|
| mithro | acomodi: Any chance you have looked at https://github.com/SymbiFlow/symbiflow-arch-defs/pull/779 ? | 16:00 |
| tpb | Title: xc7: Use the common_slice definition when LUTRAM used as a LUT. by mithro · Pull Request #779 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com) | 16:00 |
| *** kraiskil has quit IRC | 16:01 | |
| sf-slack2 | <mkurc> Morning ! | 16:03 |
| mithro | Is anyone awaiting reviews from me? | 16:06 |
| *** kraiskil has joined #symbiflow | 16:20 | |
| *** OmniMancer has quit IRC | 16:37 | |
| *** Bertl_oO is now known as Bertl | 16:50 | |
| sf-slack2 | <acomodi> mithro: not yet in detail, was traveling today. I will get to it first thing tomorrow so that the PR can be merged ASAP | 16:58 |
| mithro | Okay | 16:59 |
| mithro | tmichalak: I opened https://github.com/SymbiFlow/prjxray/issues/867 to guide us towards figuring out what needs to be done in prjxray | 17:00 |
| tpb | Title: Create a test design using LiteX, LiteEth and LiteDRAM and figure out what bits are still needed · Issue #867 · SymbiFlow/prjxray · GitHub (at github.com) | 17:00 |
| sf-slack2 | <mkurc> @mithro @litghost. Some time ago I did looking what bits we are missing in the 7-series. These are tiles that have pips and we do not have segbits for them: HCLK_IOI3, CMT_TOP_R_LOWER_B, CMT_TOP_R_LOWER_T, CMT_TOP_R_UPPER_B, CMT_TOP_R_UPPER_T, CMT_FIFO_R, CMT_TOP_L_LOWER_B, CMT_TOP_L_LOWER_T, CMT_TOP_L_UPPER_B, CMT_TOP_L_UPPER_T, CMT_FIFO_L | 17:05 |
| litghost | mkurc: Maybe file an issue with details? | 17:07 |
| mithro | mkurc: how did you do that? | 17:08 |
| sf-slack2 | <mkurc> I manually looked up `ppips_*.db` files to see which tiles that I am not familiar with have non-pseudo pips. Then I looked into corresponding `segbits_*.db`files and checked whether there are bits for those pips. | 17:45 |
| sf-slack2 | <mkurc> I focused mainly on the clock tree. | 17:46 |
| *** kraiskil has quit IRC | 18:10 | |
| sf-slack2 | <mkurc> Now when I think of that, I can write a script which does the same to double check my findings. | 18:11 |
| mithro | mkurc: But how did you figure out which bits are *needed* ? | 18:13 |
| sf-slack2 | <mkurc> @mithro Well, each non-pseudo pip should have at least some (I guess) | 18:14 |
| sf-slack2 | <mkurc> I identified which pips do not have bits | 18:15 |
| mithro | mkurc: But if a design never uses these pips, then we don't care all that much? | 18:15 |
| sf-slack2 | <mkurc> Oh, I didn't check which bits would be required for a particular design | 18:16 |
| hackerfoo | litghost: I sketched an idea about compressing and sampling the cost matrix for lookahead: | 18:16 |
| * hackerfoo posted a file: Notes122.pdf (109KB) < http://sandbox.hackerfoo.com:8008/_matrix/media/v1/download/sandbox.hackerfoo.com/NLIgEXUhdecwhprjwwukdsVA > | 18:17 | |
| hackerfoo | Let me know if that makes sense, or you'd like me to write up something more detailed. | 18:18 |
| hackerfoo | From my work on simulated annealing, most connections will be near the diagonal, but I didn't use a space filling curve, because I didn't care about compression. | 18:20 |
| hackerfoo | So this should focus the diagonal better. | 18:21 |
| litghost | That link does not appear to work? | 18:27 |
| hackerfoo | litghost: Drive link: https://drive.google.com/file/d/1DqPwElYrXBJZn5D8f2IErOFD7xU8m0BK/view?usp=sharing | 18:28 |
| tpb | Title: Notes122.pdf - Google Drive (at drive.google.com) | 18:28 |
| litghost | Ya, 1-10 GB for 1 million nodes is still a loser | 18:29 |
| litghost | 1 million nodes is a small FPGA | 18:29 |
| hackerfoo | Really? Why? | 18:29 |
| litghost | Because we need to scale to 10 - 100 million nodes | 18:29 |
| litghost | The current a7 graph is 1/5 of a 50k part | 18:29 |
| litghost | even in artix 7-series, there is s 300k part | 18:29 |
| hackerfoo | Oh. The width of the diagonal is likely much less than sqrt(N), but I don't know. | 18:30 |
| litghost | once you get to uS+, you have 1 M parts | 18:30 |
| litghost | So we are still working on toy sized graphs | 18:30 |
| litghost | We have +1-2 order of magnitude to go to reach the big parts | 18:30 |
| hackerfoo | So this is the number of pins? Can pins be grouped together? | 18:31 |
| litghost | Nodes, which can be either pin or interconnect | 18:31 |
| hackerfoo | I would guess the lower bound to be O(N log^2 N), which could scale to 100 million. But that's probably too optimistic. | 18:34 |
| hackerfoo | Maybe this coupled with some symmetry tricks could do it. At least it's a start. | 18:35 |
| litghost | To be clear, N in this case is number of nodes? | 18:35 |
| hackerfoo | Yeah. | 18:35 |
| hackerfoo | We could keep improving the compression without affecting the results, since its just a way to store the true data, although sampling all of that data would take forever, hence random sampling. | 18:38 |
| hackerfoo | There's probably good ways to accelerate that, though, since you shouldn't have to route every src -> dst individually. | 18:40 |
| *** kraiskil has joined #symbiflow | 18:56 | |
| *** kraiskil has quit IRC | 20:43 | |
| *** kraiskil has joined #symbiflow | 20:55 | |
| *** kraiskil has quit IRC | 21:58 | |
| *** Bertl is now known as Bertl_zZ | 23:03 | |
| *** bjorkintosh has quit IRC | 23:20 | |
| *** futarisIRCcloud has joined #symbiflow | 23:45 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!