| *** tristan has joined #buildstream | 05:33 | |
| *** ChanServ sets mode: +o tristan | 05:33 | |
| *** tristan has quit IRC | 05:40 | |
| *** tristan has joined #buildstream | 05:53 | |
| *** ChanServ sets mode: +o tristan | 05:53 | |
| *** tiagogomes has joined #buildstream | 09:53 | |
| *** tiagogomes has quit IRC | 10:00 | |
| *** tiagogomes has joined #buildstream | 10:01 | |
| *** ssam2 has joined #buildstream | 10:05 | |
| *** paulsher3ood has quit IRC | 12:46 | |
| *** tristan has quit IRC | 12:59 | |
| *** ssam2 has quit IRC | 13:19 | |
| *** leeming has quit IRC | 13:20 | |
| *** ssam2 has joined #buildstream | 13:20 | |
| *** leeming has joined #buildstream | 13:20 | |
| *** tiagogomes_ has joined #buildstream | 13:20 | |
| *** tiagogomes has quit IRC | 13:21 | |
| *** tristan has joined #buildstream | 13:25 | |
| *** ChanServ sets mode: +o tristan | 13:25 | |
| *** tiagogomes_ has quit IRC | 13:36 | |
| *** leeming has quit IRC | 13:37 | |
| *** tiagogomes has joined #buildstream | 13:40 | |
| * ironfoot tries to bulid build-essential.bst | 15:38 | |
| ironfoot | so far so good. I wondered why it had to fetch the gnome-sdk for this | 15:39 |
|---|---|---|
| ironfoot | but it's included, I guess because it's also a build-essential for some things | 15:39 |
| ironfoot | tristan: i sent this MR for your bs-tests https://gitlab.com/tristanvb/buildstream-tests/merge_requests/1 | 15:39 |
| ironfoot | Hm.. so a build fails (in this case because bwrap is not installed) and the process continues untily I cancel it with ^C | 16:07 |
| ironfoot | I guess it tries to build everything it can? | 16:07 |
| ironfoot | (it was maybe still fetching gcc's git repo) | 16:08 |
| *** ghishadow_ has joined #buildstream | 16:23 | |
| tristan | ironfoot, hmmm, I might forget that lack of bwrap is not auto detected at setup.py time if it's not written down :-/ | 16:44 |
| * tristan notes the 2am-ness of the hour | 16:45 | |
| ironfoot | build-stream bed | 16:45 |
| tristan | ironfoot, gnome-sdk is required (but anything else would do), it's just what we use instead of host tools for build-essential (because I already had an import of that for the master branch) | 16:46 |
| tristan | and iirc there is an option to --keep-going or not, i forget which is the default | 16:46 |
| tristan | but it will be prefetching sources and such in advance | 16:47 |
| tristan | so any ongoing tasks will wait to complete if there is a failure regardless (but I want an interactive mode where I put everything to sleep and ask the user what to do, in the case we're connected to a terminal) | 16:47 |
| ironfoot | makes sense | 16:48 |
| ironfoot | It just looked like it was doing nothing, but I guess it was still fetching other sources | 16:48 |
| tristan | yeah, a ticker or sorts is also in the queue (show what ongoing tasks are active) | 16:49 |
| tristan | to be honest, I was having too much fun with the UI | 16:49 |
| tristan | so I told myself, stop sugar coating it and finish this demo ! | 16:49 |
| ironfoot | yeah | 16:49 |
| ironfoot | It's tricky to think about a nice UI showing progress of parallel jobs | 16:50 |
| * tristan is happy there are requests to improve the UI, cause it's not difficult and plain fun :) | 16:50 | |
| tristan | ironfoot, I'm thinking, something in between yocto-like UI and current UI, only in the case we're connected to a terminal | 16:50 |
| tristan | when not connected to terminal, just line-by-line output like we have (but improved) | 16:51 |
| tristan | so you would have maybe one or two lines showing "what's currently processing" (maybe with 00:00:00 ticking the amount of time spent ?)... but *also* have the log lines trailing | 16:52 |
| tristan | when something starts or completes or fails | 16:52 |
| tristan | anyway, that's probably what I'll do unless someone has a better idea :) | 16:53 |
| ironfoot | I'll have more ideas after using it some time | 16:58 |
| tristan | nod, churn at this stage is right | 17:03 |
| tristan | ironfoot, since you did try it; try `build-stream track gedit.bst` in the master branch too :) | 17:07 |
| tristan | (I dont think it'll do anything with build-essential) | 17:07 |
| ironfoot | will do as soon as this current build fails | 17:09 |
| ironfoot | :P | 17:09 |
| ironfoot | I'm trying to see if this is an actual bug or not: My docker container running BS exited after building the first component | 17:10 |
| ironfoot | with a "kernel:[544680.988154] unregister_netdevice: waiting for lo to become free. Usage count = 1" | 17:10 |
| tristan | ironfoot, odd, default is 4 fetchers, so there will be up to 4 simultaneous source downloads at most | 17:13 |
| tristan | not that it's ever caused any issue for me | 17:13 |
| tristan | will it fail though ? | 17:14 |
| ironfoot | it has just finished building gcc, and it didn't | 17:14 |
| tristan | if you have bwrap now, I think it shouldnt fail | 17:14 |
| ironfoot | note, this failure was after building binutils | 17:15 |
| tristan | ok but... what failure, was there a failure or not ? | 17:15 |
| * tristan confused | 17:15 | |
| tristan | the kernel message ? | 17:15 |
| ironfoot | docker crashing | 17:15 |
| tristan | oh wait ok the docker crashed I see | 17:15 |
| tristan | hmmm | 17:15 |
| ironfoot | don't worry, I'll keep an eye on this | 17:16 |
| tristan | ok I'm not sure why it would, may be a docker problem, and a possible workaround might be `--fetchers 1` | 17:16 |
| tristan | (although, doesnt sound like a desirable workaround heh) | 17:17 |
| ironfoot | nope | 17:17 |
| ironfoot | i really don't know if that kernel error is caused because of the number of fetchers | 17:20 |
| ironfoot | hm | 17:34 |
| ironfoot | I believe it's fetching linux.git twice :) | 17:34 |
| ironfoot | at the same time | 17:34 |
| *** ssam2_ has joined #buildstream | 18:00 | |
| *** ssam2 has quit IRC | 18:01 | |
| tristan | ironfoot, you're probably right... curiously, is *that* causing some problem ? | 18:05 |
| tristan | (that works well with the ostree source and is actually desirable in that case) | 18:05 |
| *** ghishadow_ has quit IRC | 18:33 | |
| *** ssam2_ has quit IRC | 18:43 | |
| ironfoot | I don't think is causing any problems. I'll think about this and open an issue | 19:12 |
| *** tristan has quit IRC | 21:32 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!