| Time |
Nick |
Message |
| 00:59 |
|
Fenrir24 joined #minetest-dev |
| 01:05 |
|
proller joined #minetest-dev |
| 01:21 |
|
ShadowBot joined #minetest-dev |
| 01:32 |
|
Noisytoot joined #minetest-dev |
| 01:43 |
|
Fenrir24 joined #minetest-dev |
| 01:43 |
MTDiscord |
<bastrabun> sfan5: As long as gathering those performance data doesn't impact the performance of the server, feel free to request any kind of readout from YL. |
| 01:55 |
|
srifqi joined #minetest-dev |
| 01:56 |
srifqi |
hello. merging #14573 in around 4 minutes |
| 01:56 |
ShadowBot |
https://github.com/minetest/minetest/issues/14573 -- Do not make inventory_items_animations setting static thread_local by srifqi |
| 02:00 |
srifqi |
done |
| 02:12 |
|
thelounge656 joined #minetest-dev |
| 03:24 |
|
pmp-p joined #minetest-dev |
| 03:32 |
|
pmp-p joined #minetest-dev |
| 04:00 |
|
MTDiscord joined #minetest-dev |
| 05:03 |
BuckarooBanzai |
sfan5: i have the prometheus/engine metrics for a few of my servers here if that (somehow?) helps: https://monitoring.minetest.ch/d/BRwIg8xMz/engine-metrics?orgId=1&refresh=5s |
| 07:37 |
sfan5 |
@landarvargan @bastrabun BuckarooBanzai: C++ profiling does not impact performance, all you need is a build with debug symbols (RelWithDebInfo), see https://github.com/minetest/minetest/blob/master/doc/developing/misc.md#profiling-minetest-on-linux |
| 07:37 |
sfan5 |
I'm sure there's some way to attach perf to an existing process even |
| 08:22 |
|
Warr1024 joined #minetest-dev |
| 08:22 |
|
YuGiOhJCJ joined #minetest-dev |
| 08:41 |
|
Desour joined #minetest-dev |
| 08:49 |
Desour |
RelWithDebInfo *can* actually impact performance. for example, the mapblock mesh making seems to be considerably slower, which causes nodes to ghost around for some time after you dig them |
| 08:50 |
Desour |
and you don't necessarily need RelWithDebInfo for profiling, Release works mostly fine, for me at least in most cases |
| 08:51 |
rubenwardy |
specifically, sampler based profiling doesn't affect performance as much (ie: perf). Definitely don't try insturamentation based profiling (ie: callgrind) |
| 08:57 |
Desour |
btw. we already once had a PR for moving on_step execution to lua: #5502 |
| 08:57 |
ShadowBot |
https://github.com/minetest/minetest/issues/5502 -- Move object on_step execution to lua by HybridDog |
| 08:58 |
Desour |
but it'll be a bit harder nowadays that we have moveresult |
| 09:17 |
|
Lupercus joined #minetest-dev |
| 09:38 |
|
Lupercus joined #minetest-dev |
| 10:17 |
|
proller joined #minetest-dev |
| 10:34 |
sfan5 |
are you sure it's RelWithDebInfo doing that? |
| 10:34 |
sfan5 |
debug info should not affect codegen, but it could be that Release is using e.g. -O3 and RelWithDebInfo is using -O2 |
| 11:07 |
MTDiscord |
<josiah_wi> You could manually set the flags. Don't you need to set -profiling or something anyway? |
| 11:08 |
MTDiscord |
<josiah_wi> Maybe not for the perf tool. |
| 11:20 |
|
proller joined #minetest-dev |
| 11:31 |
|
Lupercus joined #minetest-dev |
| 11:38 |
|
turtleman joined #minetest-dev |
| 13:03 |
sfan5 |
pushing some additions to doc/developing/misc.md in 10m |
| 13:21 |
|
Fenrir24 joined #minetest-dev |
| 13:23 |
sfan5 |
@landarvargan @bastrabun BuckarooBanzai: see the new section in the document I linked earlier :) |
| 13:41 |
MTDiscord |
<wsor4035> @bastrabun @buckaroobanzai.0 |
| 13:41 |
MTDiscord |
<bastrabun> Thanks, will take a look. Josiah already advises me on those things. I'd like to send you a sample of our testserver, so you can see whether I did everything right. Would be pretty disappointing if I gave you The Grand perf.data File and it had nothing of value in it, because I made a mistake - which is not without precedence 😛 |
| 15:15 |
|
imi joined #minetest-dev |
| 15:28 |
|
proller joined #minetest-dev |
| 16:34 |
sfan5 |
sure that'd be nice |
| 17:10 |
|
Lupercus joined #minetest-dev |
| 17:20 |
|
proller joined #minetest-dev |
| 17:33 |
|
Lupercus joined #minetest-dev |
| 17:59 |
|
proller joined #minetest-dev |
| 18:26 |
|
Eragon joined #minetest-dev |
| 18:35 |
|
Lupercus joined #minetest-dev |
| 18:36 |
|
alias joined #minetest-dev |
| 19:48 |
sfan5 |
https://en.cppreference.com/w/cpp/coroutine/generator hmm this'd be nice to use |
| 20:20 |
sfan5 |
reviews needed, PRs are small -> #14568, #14566 |
| 20:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/14568 -- Add minetest.get_gen_notify to mapgen env by sfan5 |
| 20:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/14566 -- Check for negative width in `InvRef:set_width()` by OgelGames |
| 20:21 |
sfan5 |
also merging #14581 in 10m |
| 20:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/14581 -- Fix crash when player leaves before breath removal job by LoneWolfHT |
| 21:20 |
MTDiscord |
<josiah_wi> Is the one-approval policy experiment still active? |
| 21:23 |
[MTMatrix] |
<Zughy> yeah |
| 21:54 |
|
turtleman joined #minetest-dev |
| 22:07 |
|
[MTMatrix] joined #minetest-dev |
| 22:14 |
|
proller joined #minetest-dev |
| 22:32 |
|
panwolfram joined #minetest-dev |
| 22:39 |
|
proller joined #minetest-dev |
| 22:40 |
|
[MTMatrix] joined #minetest-dev |
| 22:45 |
|
[MTMatrix] joined #minetest-dev |
| 23:01 |
|
Fenrir24 joined #minetest-dev |
| 23:01 |
|
[MTMatrix] joined #minetest-dev |
| 23:03 |
|
Fenrir24 left #minetest-dev |
| 23:28 |
|
Fenrir24 joined #minetest-dev |
| 23:28 |
|
Fenrir24 left #minetest-dev |
