| *** rsc <rsc!~robert@linuxnetz.de> has joined #libre-soc | 00:31 | |
| *** rsc <rsc!~robert@fedora/rsc> has joined #libre-soc | 00:31 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 00:42 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 00:42 | |
| *** lxo <lxo!~lxo@gateway/tor-sasl/lxo> has quit IRC | 03:47 | |
| *** lxo <lxo!~lxo@gateway/tor-sasl/lxo> has joined #libre-soc | 03:47 | |
| *** Ritish <Ritish!~Ritish@60.243.29.170> has joined #libre-soc | 04:16 | |
| bagiyal[m] | 😁 | 04:19 |
|---|---|---|
| Ritish | while I thought I actually finished installing dev-scripts, It seems I only installed a part of it | 06:36 |
| *** Gayatri <Gayatri!~Gayatri@60.243.29.170> has joined #libre-soc | 06:52 | |
| *** Gayatri <Gayatri!~Gayatri@60.243.29.170> has quit IRC | 07:00 | |
| *** Ritish <Ritish!~Ritish@60.243.29.170> has quit IRC | 07:00 | |
| *** Ritish <Ritish!~Ritish@27.5.247.63> has joined #libre-soc | 07:01 | |
| *** Gayatri <Gayatri!~Gayatri@27.5.247.63> has joined #libre-soc | 07:02 | |
| Gayatri | Good Morning Team | 07:03 |
| Gayatri | How can i check my all setup has been done for create SOC build | 07:04 |
| Gayatri | what will be the 1st step to do for creating SOC. Can i run dev-env-setup.git file at vim | 07:07 |
| *** Gayatri18 <Gayatri18!~Gayatri@27.5.247.63> has joined #libre-soc | 07:20 | |
| *** Gayatri82 <Gayatri82!~Gayatri@27.5.247.63> has joined #libre-soc | 07:21 | |
| *** Gayatri <Gayatri!~Gayatri@27.5.247.63> has quit IRC | 07:23 | |
| *** Gayatri82 <Gayatri82!~Gayatri@27.5.247.63> has quit IRC | 07:26 | |
| *** Gayatri <Gayatri!~Gayatri@27.5.247.63> has joined #libre-soc | 07:27 | |
| Gayatri | i completed running dev-env-setup. then what can i do | 07:33 |
| *** Gayatri18 <Gayatri18!~Gayatri@27.5.247.63> has quit IRC | 07:35 | |
| lkcl | Gayatri, i advised Ritish a couple of days ago. you can see in the IRC logs | 08:59 |
| Ritish | https://libre-soc.org/irclog/%23libre-soc.2022-12-24.log.html | 08:59 |
| Ritish | here | 08:59 |
| Ritish | lkcl, I tried running make test in the soc directory | 09:00 |
| Ritish | It shows a fatal error, "mpfr file not found" | 09:00 |
| Ritish | > mpfr.h | 09:00 |
| lkcl | i never run that. | 09:01 |
| Ritish | I see | 09:01 |
| lkcl | and it means you're missing libmpfr | 09:01 |
| Ritish | Hmm | 09:01 |
| Gayatri | Thanks luke. i will try to find mpfr.h | 09:01 |
| Gayatri | ok | 09:02 |
| lkcl | which in debian means you're missing ... | 09:02 |
| lkcl | grep mpfr.h /var/lib/dpkg/info/*.list | 09:02 |
| lkcl | surprise | 09:02 |
| lkcl | libmpfr-dev | 09:02 |
| Ritish | eek, just using ubuntu for now | 09:02 |
| lkcl | which in turn means that you didn't properly run... | 09:03 |
| lkcl | $ grep libmpfr-dev ../dev-env-setup/* | 09:03 |
| Gayatri | can we run in ubuntu | 09:03 |
| lkcl | ../dev-env-setup/binutils-gdb-install:apt-get install -y build-essential libgmp-dev libmpfr-dev libmpc-dev \ | 09:03 |
| lkcl | ../dev-env-setup/coriolis-install: libmpfr-dev libgmp-dev libmotif-dev libreadline-dev \ | 09:03 |
| lkcl | or | 09:03 |
| lkcl | ../dev-env-setup/install-hdl-apt-reqs: libmpfr-dev \ | 09:03 |
| Ritish | isgotchu | 09:03 |
| Ritish | gotchu* | 09:03 |
| lkcl | there is flat-out absolutely no chance in hell that any of us are going to do "OS" support | 09:03 |
| Gayatri | ok | 09:04 |
| Ritish | yessir | 09:04 |
| lkcl | if you want to run on a non-standard OS other than the *very specifically chosen* reproducible build OS, you are entirely on your own | 09:04 |
| lkcl | there are very specific reasons for that | 09:04 |
| lkcl | *even one* library different | 09:04 |
| lkcl | can result in a variation | 09:04 |
| Ritish | I understand | 09:04 |
| lkcl | which, when you are about to drop USD 16 million dollars on Mask Charges | 09:04 |
| lkcl | one engineer goes, | 09:05 |
| lkcl | "oh i have a different GDS-II file" | 09:05 |
| lkcl | and another goes | 09:05 |
| lkcl | "oh mine is different too" | 09:05 |
| Ritish | lol xD | 09:05 |
| lkcl | and after *weeks* of investigation, far after the missed deadline where the Foundry will no longer talk to you because you wasted their money (they only make money from selling wafers) | 09:06 |
| lkcl | it turns out that all the engineers picked random OSes instead of listening to experience. | 09:06 |
| lkcl | follow the advice! use the chroot environment scripts exactly! | 09:07 |
| Ritish | ayee | 09:07 |
| lkcl | do not deviate! | 09:07 |
| Ritish | yepp! | 09:07 |
| lkcl | apart from anything, it is not fair to expect other team members to "support" variations in OSes (instead of helping with much more important things like "bugs" or "development") | 09:08 |
| lkcl | once you have several *years* of experience with a different OS, then you will find naturally that you can work out different packages etc. | 09:08 |
| lkcl | but even there | 09:08 |
| lkcl | because of the "reproducible builds" requirement | 09:09 |
| lkcl | it's still not ok to use a non-standard OS | 09:09 |
| lkcl | so | 09:09 |
| Ritish | I very much agree to that, you shouldn't waste your time in something as trivial as that :/ | 09:09 |
| Ritish | I'll do what I can on ubuntu, since we need to submit the thingy before 31st | 09:10 |
| lkcl | even one package missed - or different - and the entire multi-level cascade fails. | 09:10 |
| lkcl | we still find issues that need pinning down with the devscripts, even now. | 09:10 |
| Ritish | If it's working, alright, if it's not, alright ig :) | 09:10 |
| Ritish | Hmmm | 09:10 |
| lkcl | reason: we had to use backports for qemu | 09:11 |
| lkcl | but | 09:11 |
| lkcl | of course, backports by definition *changes the damn packages* | 09:11 |
| Ritish | yikes | 09:11 |
| Ritish | qemu is? | 09:11 |
| Gayatri | ok luke | 09:11 |
| lkcl | one particularly irkesome one was libcurl3-gnutls | 09:12 |
| lkcl | which had to be pinned at exactly version 7.64 | 09:12 |
| Ritish | damn | 09:12 |
| lkcl | debian security team tried for a while to "sort" out (back-port) various security fixes | 09:13 |
| lkcl | which went horribly wrong (security fixes that introduced more bugs) | 09:13 |
| Ritish | oops | 09:13 |
| lkcl | sigh | 09:13 |
| lkcl | it stopped git-https working for a while | 09:13 |
| lkcl | https://git.libre-soc.org/?p=dev-env-setup.git;a=blob;f=mk-deb-chroot;h=a02e46ffea4bb2b1297ce3ea8ee9648c788e9fdf;hb=HEAD#l139 | 09:14 |
| lkcl | can you imagine trying to forward-port all this to a dozen OSes? | 09:14 |
| lkcl | "ubuntu don't work, why is that?" | 09:15 |
| lkcl | which version do you have? | 09:15 |
| Ritish | 20.04 | 09:15 |
| lkcl | 16 17 18 19 20 21 22 23 or 24? | 09:15 |
| Ritish | ;-; | 09:15 |
| lkcl | :) | 09:15 |
| Ritish | okay, I got that | 09:15 |
| Ritish | eeee | 09:15 |
| lkcl | we have had people run the chroot devscripts on archlinux | 09:16 |
| lkcl | arch has debootstrap | 09:16 |
| Ritish | ohh | 09:16 |
| lkcl | once you're in a chroot which has been specifically set up as a reproducible build chroot (i.e. exactly the same as everyone else) | 09:17 |
| lkcl | the *host* OS that you are using tends not to matter | 09:17 |
| lkcl | BUT | 09:17 |
| Ritish | but o.o | 09:17 |
| lkcl | if you use particularly non-standard OSes (archlinux) you end up with incompatibilities on things like x11 | 09:17 |
| *** Gayatri <Gayatri!~Gayatri@27.5.247.63> has quit IRC | 09:18 | |
| lkcl | so can't run GUIs unless you install special libraries in the *host* OS | 09:18 |
| Ritish | gotchu | 09:18 |
| lkcl | ubuntu is... close enough... to debian, such that you are less likely to run into such issues | 09:19 |
| lkcl | but you are still on your own | 09:20 |
| Ritish | aye aye captain | 09:20 |
| Ritish | lkcl, is it fine if I work on the university of oregon debian, remotely | 10:07 |
| Ritish | or would you think it's not possible to complete it in little time before 31st | 10:07 |
| *** Gayatri <Gayatri!~Gayatri@27.5.247.63> has joined #libre-soc | 10:32 | |
| *** Gayatri <Gayatri!~Gayatri@27.5.247.63> has quit IRC | 11:06 | |
| *** Ritish <Ritish!~Ritish@27.5.247.63> has quit IRC | 12:04 | |
| *** choozy <choozy!~choozy@75-63-174-82.ftth.glasoperator.nl> has joined #libre-soc | 13:03 | |
| *** Ritish <Ritish!~Ritish@202.131.158.114> has joined #libre-soc | 13:26 | |
| lkcl | Ritish, it's a great idea, they have some monster machines. | 14:24 |
| lkcl | but if you are thinking of doing that purely because "oregon machines use debian native", remember that's not strictly necessary | 14:26 |
| lkcl | the fact that you have ubuntu native means that the chroots have a very high probability of not having any problems. | 14:27 |
| lkcl | plus | 14:27 |
| lkcl | even on the oregon debian host, you will *still* have to run the chroot installation script (mk-deb-chroot) in order to create a stable debian/10 "repro" build environment | 14:28 |
| Ritish | Right | 14:30 |
| Ritish | I'll spend some time on ubuntu, if that doesn't go well, I'll try getting access to the oregon machines | 14:31 |
| lkcl | just run the instructions, exactly as they're written. no deviation, no experimentation, nothing. | 14:32 |
| lkcl | under this page is an example: | 14:33 |
| lkcl | https://libre-soc.org/HDL_workflow/devscripts/ | 14:33 |
| lkcl | notice how it says: | 14:33 |
| lkcl | $ sudo bash | 14:33 |
| lkcl | # ./mk-deb-chroot libresoc | 14:33 |
| lkcl | the expectation is that you *will* do *exactly that* - and nothing else | 14:33 |
| lkcl | thus, the expectation is that you *will* have created a debian chroot with that command | 14:36 |
| lkcl | the next command enters that chroot (schroot -c libresoc) | 14:36 |
| lkcl | and from there the expectation is that you will then run the installation commands *in* that chroot. | 14:37 |
| lkcl | remember, i said a few days ago: from experience over the past 2-3 years, everyone who has not properly followed the instructions has failed to get to a working point, usually wasting 2-3 weeks of their time. | 14:41 |
| lkcl | and by contrast everyone who has properly followed the instructions - explicitly and to the letter - has got up and running within 24-48 hours | 14:41 |
| Ritish | I gotchu | 14:46 |
| Ritish | but | 14:46 |
| Ritish | I assume running this is a necessary process as well : https://git.libre-soc.org/?p=dev-env-setup.git;a=blob;f=hdl-dev-repos;hb=HEAD | 14:47 |
| Ritish | because I forgot, when I did earlier | 14:47 |
| Ritish | I dont exactly remember running the chroot command, but I'll have to go through the logs. Even if I discovered a mistake rn, I wonder what I could do | 14:48 |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 15:34 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.166.93> has joined #libre-soc | 15:35 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.166.93> has quit IRC | 15:50 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.166.93> has joined #libre-soc | 15:51 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.166.93> has quit IRC | 15:55 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 16:00 | |
| lkcl | start again. in essence, the packages installed in ubuntu are radically different and it'd be a total waste of time to try to "correct" that. | 17:25 |
| lkcl | consider what you've done to be a write-off / learning-experience: follow the bleedin instructions to the letter! | 17:25 |
| lkcl | :) | 17:25 |
| lkcl | this is an extension of what i talked about on the 24th | 17:26 |
| lkcl | you haven't got time to make comparisons against "something that's fully automated" vs "something you did manually" when there's 250+ software package dependencies installed | 17:27 |
| lkcl | complete utter waste of your time | 17:27 |
| lkcl | by running the scripts *as instructed*, with zero thought and zero possibility of error (beyond annoying things like "interrupted internet access", sigh) you end up at an exact position that everyone else knows you will be at | 17:28 |
| lkcl | and thus *do not* have to go through days to weeks of investigation... over the internet | 17:28 |
| lkcl | that time spent investigating could have been spent instead doing actual development | 17:29 |
| Ritish | aye! will start again from 0 tomorrow >_< | 17:36 |
| Ritish | peace | 17:36 |
| *** Ritish <Ritish!~Ritish@202.131.158.114> has quit IRC | 17:37 | |
| lkcl | :) | 17:48 |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 18:03 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 18:05 | |
| openpowerbot | [mattermost] <gayatri> can we work at university of Oregon | 18:25 |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has quit IRC | 19:42 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.172.6> has joined #libre-soc | 19:42 | |
| lkcl | gayatri: of course. i replied to Ritish explaining it. https://libre-soc.org/irclog/%23libre-soc.2022-12-27.log.html#t2022-12-27T14:24:20 | 19:44 |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@176.59.172.6> has quit IRC | 19:53 | |
| *** ghostmansd[m] <ghostmansd[m]!~ghostmans@broadband-109-173-83-100.ip.moscow.rt.ru> has joined #libre-soc | 19:55 | |
| *** choozy <choozy!~choozy@75-63-174-82.ftth.glasoperator.nl> has quit IRC | 23:05 | |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!