| Chips4Makers[m] | Las lkcl For tape-out the Verilog has been patched manually to fix difficulties with litex generated code. I assume that is the problem you are seeing. | 09:38 |
|---|---|---|
| ghostmansd | Hi all, it seems I'm out for at least a week. My daughter got sick, so I'll be concentrating on parental deeds. I might be available occasionally, but it's not a likely scenario. | 14:26 |
| lkcl | totally understand, ghostmansd | 14:34 |
| ghostmansd | Thanks! | 14:35 |
| ghostmansd | That said, if there are questions which need my reaction, my both e-mails still work :-) | 14:35 |
| ghostmansd | It's just that the answer perhaps won't be immediate | 14:36 |
| programmerjake | hope she feels better! | 15:08 |
| ghostmansd | programmerjake: thank you for the best wishes! :-) | 15:41 |
| Las[m] | <Chips4Makers[m]> "Las lkcl For tape-out the..." <- So after running `make ls180_4ksram_verilog_build` in the `soc` repository to "build the litex libresoc SoC with 4k SRAMs", the output of that command is manually edited? | 17:46 |
| Las[m] | Chips4Makers: ^ | 17:46 |
| Chips4Makers[m] | yes; details to be provided by lkcl as he did last version. | 18:55 |
| Las[m] | That's quite unfortunate | 19:18 |
| Las[m] | Are there any plans to fix this? | 19:18 |
| Las[m] | Or is this unavoidable? | 19:18 |
| lkcl | ah yeah, i made one change - the name of the module ls1804k is changed to module ls180 | 19:47 |
| lkcl | it's a trivial thing that can be done with sed | 19:48 |
| lkcl | i just didn't get round to ever doing it | 19:48 |
| lkcl | Las[m], one thing, you do need to use the correct branches. | 19:51 |
| lkcl | src/coriolis$ git branch | 19:51 |
| lkcl | * devel | 19:51 |
| lkcl | this is part of the automated dev-env-setup script | 19:52 |
| Las[m] | yeah I'm already using the devel branch of Coriolis | 19:52 |
| Las[m] | When I get to the point where I can get `make ls180_4ksram_verilog_build` to work, I'll see what `sed`s I'll have to do to get it working with Coriolis and such | 19:53 |
| lkcl | they're completely independent | 19:54 |
| lkcl | for the very reason that Jean-Paul Chaput requested that we commit the auto-generated verilog | 19:54 |
| lkcl | which, normally, i do not approve of | 19:54 |
| * lkcl running ./build_full_4k_sdram_recon.sh to make sure it's working | 19:55 | |
| lkcl | ok that's now doing "Building RoutingPads" so it should be good | 19:57 |
| Las[m] | lkcl: Have you tested `build_full_4ksram.sh` from `experiments9` lately? | 20:00 |
| lkcl | there's basically no commits - at all - since the tape-out | 20:23 |
| lkcl | and it's been long enough i can't exactly remember which one of the build scripts i ran | 20:23 |
| lkcl | so whatever worked then *will* work now | 20:23 |
| lkcl | (another advantage of Jean-Paul's advice) | 20:24 |
| lkcl | my fingers however are telling me to run the build_full_4k_sdram_recon.sh script | 20:25 |
| lkcl | (coriolis2)lkcl@fizzy:~/soclayout/experiments9$ ls -altr *.sh | 20:26 |
| lkcl | -rwxr-xr-x 1 lkcl lkcl 333 Nov 27 2020 build.sh | 20:26 |
| lkcl | -rwxr-xr-x 1 lkcl lkcl 806 Apr 9 12:03 build_full.sh | 20:26 |
| lkcl | -rwxr-xr-x 1 lkcl lkcl 1204 Apr 24 13:54 mksym.sh | 20:26 |
| lkcl | -rwxr-xr-x 1 lkcl lkcl 972 Apr 28 10:13 build_full_4ksram.sh | 20:26 |
| lkcl | -rwxr-xr-x 1 lkcl lkcl 1335 Jun 15 12:37 build_full_4ksram_recon.sh | 20:26 |
| lkcl | yep that looks right. it was the last one i edited | 20:26 |
| Las[m] | Hmm, thanks, I'll try that! | 21:09 |
| Las[m] | Unfortunately still get `[ERROR] CoreToChip.buildChip(): Pad "p_sys_pllclk" refer net "sys_pllclk" which do not exist in core "ls180".` | 21:54 |
| Las[m] | lkcl: Where is `ls180` defined? | 21:54 |
| Las[m] | ah nevermind, I figured it ou | 21:55 |
| Las[m] | t | 21:55 |
| Las[m] | I can see it's defined | 21:56 |
| kylel | I'm ready to push shit_rot caller | 22:00 |
| kylel | make that shift lol | 22:00 |
| kylel | quite a typo | 22:01 |
| kylel | Anyways need perms | 22:04 |
| avl | Hi all! I have set up and ready to hack, I guess. I have downloaded packages and now I need a way to verify, that everything as expected. Also I read https://bugs.libre-soc.org/show_bug.cgi?id=550 and I wonder what is the current status on that? Which repo to hack and what sources migt be useful? | 23:48 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!