| Time |
Nick |
Message |
| 00:16 |
paramat |
good progress #7192 Shara. another productive day ^_^ |
| 00:16 |
ShadowBot |
https://github.com/minetest/minetest/issues/7192 -- Biome API / cavegen: Add definable cave liquid for a biome by paramat |
| 00:17 |
Shara |
Looking forward to seeing this added :) |
| 00:17 |
Fixer |
remember that serverside movement PR? give it some love |
| 00:17 |
Shara |
Thanks for working on this one paramat |
| 00:17 |
Shara |
Fixer: I hope that will get some love as well |
| 00:56 |
|
nobodywasishere joined #minetest-dev |
| 01:00 |
|
MarisaG joined #minetest-dev |
| 01:02 |
|
MarisaG left #minetest-dev |
| 01:07 |
|
Tmanyo joined #minetest-dev |
| 02:02 |
|
reductum joined #minetest-dev |
| 02:35 |
|
troller joined #minetest-dev |
| 02:45 |
|
AndroBuilder joined #minetest-dev |
| 02:45 |
|
reductum joined #minetest-dev |
| 05:23 |
nerzhul |
nobody loves my commits :( xD |
| 05:42 |
nerzhul |
i push another set of performance-type-promotion-in-math-fn fixes |
| 05:42 |
|
Player-2 joined #minetest-dev |
| 05:43 |
nerzhul |
=> https://github.com/minetest/minetest/commit/8e0b80aa369df7d1142053f077df928106c6ce8c |
| 05:49 |
|
ssieb joined #minetest-dev |
| 06:40 |
nerzhul |
i push a little fix for a sunlight color bug + some issues reported by PVS studio |
| 07:48 |
nerzhul |
ty for the review Krock, i fixed your valid point |
| 08:00 |
|
Krock joined #minetest-dev |
| 08:26 |
nerzhul |
will merge this morning when travis pass and i look at it :p |
| 08:53 |
Krock |
will merge game#2089 in 5 minutes |
| 08:53 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/2089 -- Add food_* groups to default edibles by tenplus1 |
| 08:56 |
nerzhul |
merging #7195 |
| 08:56 |
ShadowBot |
https://github.com/minetest/minetest/issues/7195 -- Huge LBM lookup performance improvement on mapblock loading by nerzhul |
| 08:56 |
nerzhul |
done |
| 08:58 |
Krock |
merging.. |
| 09:06 |
|
troller joined #minetest-dev |
| 09:21 |
|
Fixer joined #minetest-dev |
| 10:14 |
|
Hijiri_ joined #minetest-dev |
| 10:15 |
|
ThomasMonroe_ joined #minetest-dev |
| 10:15 |
|
exio joined #minetest-dev |
| 10:16 |
|
Taoki joined #minetest-dev |
| 10:16 |
|
Darcidride joined #minetest-dev |
| 10:18 |
|
BakerPrime joined #minetest-dev |
| 10:28 |
|
ThomasMonroe_ joined #minetest-dev |
| 10:31 |
|
proller__ joined #minetest-dev |
| 11:04 |
|
proller__ joined #minetest-dev |
| 11:28 |
|
Krock joined #minetest-dev |
| 11:47 |
|
Wuzzy joined #minetest-dev |
| 12:33 |
|
longerstaff13-m joined #minetest-dev |
| 12:38 |
|
longerstaff13 joined #minetest-dev |
| 12:38 |
|
longerstaff13 joined #minetest-dev |
| 12:42 |
|
srifqi joined #minetest-dev |
| 12:51 |
|
srifqi left #minetest-dev |
| 12:54 |
|
srifqi joined #minetest-dev |
| 13:38 |
|
srifqi left #minetest-dev |
| 13:46 |
|
proller__ joined #minetest-dev |
| 14:13 |
|
antims joined #minetest-dev |
| 14:57 |
|
twoelk joined #minetest-dev |
| 15:19 |
|
paramat joined #minetest-dev |
| 15:40 |
|
proller__ joined #minetest-dev |
| 15:56 |
|
cx384 joined #minetest-dev |
| 15:58 |
|
clavi joined #minetest-dev |
| 16:01 |
|
Beton joined #minetest-dev |
| 16:06 |
|
TC01 joined #minetest-dev |
| 16:11 |
|
Gael-de-Sailly joined #minetest-dev |
| 16:15 |
|
CBugDCoder joined #minetest-dev |
| 16:50 |
Thomas-S |
nerzhul, paramat, rubenwardy, Krock: Please take a look at #7199. It's a bug report about serious problems introduced by #6587. Thanks in advance for your efforts! |
| 16:50 |
ShadowBot |
https://github.com/minetest/minetest/issues/7199 -- Bugs regarding objects |
| 16:50 |
ShadowBot |
https://github.com/minetest/minetest/issues/6587 -- Optimize entity-entity collision by numberZero |
| 16:51 |
Krock |
that's a nice error rainbow you've got there |
| 16:52 |
|
Fritigern joined #minetest-dev |
| 16:57 |
Krock |
Thomas-S, that stack overflow is caused by a infinite function call of display_api.update_entities, place_entities and the on_activate callback when core.add_entity returns nil |
| 16:58 |
Krock |
I wonder why it does that.. *checks* |
| 16:59 |
Thomas-S |
I'll take a look at the display_modpack code soon |
| 17:03 |
Krock |
Thomas-S, but why do you list the homedecor modpack to reproduce when the errors appear in display_api? |
| 17:03 |
Thomas-S |
The homedecor error is a separate error. |
| 17:05 |
Krock |
I see. two problems that might have the same source |
| 17:06 |
|
Taoki joined #minetest-dev |
| 17:09 |
Krock |
display_lib does not add any signs |
| 17:10 |
Thomas-S |
but display_modpack does |
| 17:17 |
nerzhul |
Krock if you find the thing and it's in the activeobjectmap, please fix it and add the corresponding unittest or fix it to ensure we won't have a such error anymore |
| 17:19 |
Krock |
oh my god. Minetest didn't crash but when leaving to the main menu again the RAM use jitters from 200 to 700 MiB |
| 17:20 |
nerzhul |
Krock interesting |
| 17:20 |
Krock |
https://i.imgur.com/cziJ2Vk.png |
| 17:21 |
nerzhul |
valgrind, if you just connect and disconnect do you have a such thing ? |
| 17:21 |
nerzhul |
euh... Krock not valgrind :D |
| 17:21 |
Krock |
nerzhul, first I have to find out why it spawns a few thousand entities |
| 17:23 |
nerzhul |
maybe the memleak is common for bot H ? :) |
| 17:23 |
nerzhul |
playing valgrind memcheck connect/mapblock load disconnect |
| 17:29 |
nerzhul |
Krock in regular MT + MT game (maybe mtgame not up to date) no problem for me |
| 17:29 |
nerzhul |
==4963== definitely lost: 0 bytes in 0 blocks |
| 17:29 |
nerzhul |
==4963== indirectly lost: 0 bytes in 0 blocks |
| 17:29 |
nerzhul |
==4963== possibly lost: 0 bytes in 0 blocks |
| 17:29 |
nerzhul |
we got many initialization error in our lua stack but no memleak in regular mapblock load & ABM |
| 17:32 |
Krock |
from spawning until on_activate everything is normal, but then the "name" field of the spawned entity is nil |
| 17:48 |
|
Foz joined #minetest-dev |
| 17:55 |
|
Foz joined #minetest-dev |
| 17:58 |
Krock |
oh my god. I need another mod which doesn't do such complicated cross-calls three times after each node placement |
| 17:59 |
nerzhul |
lol |
| 18:00 |
nerzhul |
Krock it seems ActiveObjects are never removed |
| 18:02 |
nerzhul |
by folding destructor of ActiveObject i don't see calls to destructor |
| 18:02 |
Krock |
but I can pick up dropped items by punching |
| 18:03 |
Krock |
get_objects_inside_radius cannot detect them after they were :remove'd |
| 18:05 |
|
ssieb joined #minetest-dev |
| 18:15 |
|
proller__ joined #minetest-dev |
| 18:33 |
nerzhul |
Krock strangely i don't find any cleanup on activeobject list |
| 18:33 |
nerzhul |
i found the regular but not the shutdown removal |
| 18:34 |
nerzhul |
hmm |
| 18:37 |
nerzhul |
it sounds okay, i don't know where is the mem leak |
| 18:39 |
nerzhul |
#7194 is now ready for a merge, paramat seems good. Krock ? |
| 18:39 |
ShadowBot |
https://github.com/minetest/minetest/issues/7194 -- [clang-tidy] Promote some performance-* as a coding error by nerzhul |
| 18:42 |
Krock |
*shrug* .. I guess |
| 18:45 |
nerzhul |
then i can merge it ? :) |
| 18:48 |
Krock |
quash&merge, yes. |
| 18:49 |
nerzhul |
ofc, done, thanks |
| 19:03 |
|
longerstaff13-m joined #minetest-dev |
| 19:06 |
nerzhul |
vallgrind doesn't like all our lua_pushstring(L, str.c_str()) :( |
| 19:08 |
sofar |
valgrind doesn't like programmers, I'm convinced |
| 19:10 |
nerzhul |
oh i got an interesting API idea for minetest core engine, reading mientest_game |
| 19:11 |
nerzhul |
registering particle spwaners definitions in core and returning an ID, permitting mods to add particle spawner by id directly instead of triggering a huge lua object deserialize on each spawn |
| 19:15 |
sofar |
how large is the serialized id? |
| 19:15 |
sofar |
but, yes, interesting |
| 19:16 |
nerzhul |
define and then spawn it just defining a position ? maybe can be nice, it doesn't cover all cases... but it's an iea |
| 19:17 |
nerzhul |
hmmm, maybe there is more ideas, but it's one :) |
| 19:21 |
sofar |
it makes sense for the spawners |
| 19:21 |
sofar |
not for each particle ofc |
| 19:24 |
nerzhul |
the only memleak i found (difficult due to the slow process) is on shader source, 1Kb memleak |
| 19:38 |
sofar |
that's only leaking in between server switches |
| 19:38 |
sofar |
right? |
| 19:38 |
nerzhul |
i only see that in my check |
| 19:38 |
nerzhul |
==8614== definitely lost: 2,304 bytes in 22 blocks |
| 19:38 |
nerzhul |
==8614== indirectly lost: 6,111 bytes in 156 blocks |
| 19:51 |
Krock |
nerzhul, I found an issue but that already existed pre-PR |
| 19:52 |
|
Tmanyo joined #minetest-dev |
| 19:52 |
|
paramat joined #minetest-dev |
| 19:53 |
Krock |
nvm. false alert :( |
| 20:06 |
|
YuGiOhJCJ joined #minetest-dev |
| 20:15 |
|
DI3HARD139 joined #minetest-dev |
| 20:16 |
Krock |
nerzhul, now I found the entity issue. the position is not updated on create |
| 20:16 |
Krock |
i.e. the server does not map its position correctly when it's only placed using add_entity |
| 20:18 |
Krock |
`refmap` |
| 20:30 |
nerzhul |
nice |
| 20:37 |
Krock |
and the collision box data is loaded AFTER that map is built |
| 20:38 |
Krock |
serverenvironment.cpp L1667 found that stupid bug |
| 20:39 |
Krock |
unittests pass but it doesn't work like this in our code |
| 20:40 |
nerzhul |
we don't cover everything :) |
| 21:05 |
|
numzero joined #minetest-dev |
| 21:32 |
|
longerstaff13-m joined #minetest-dev |
| 21:42 |
|
numzero1 joined #minetest-dev |
| 21:47 |
|
proller joined #minetest-dev |
| 21:50 |
|
paramat joined #minetest-dev |
| 22:36 |
paramat |
#7192 ready |
| 22:36 |
ShadowBot |
https://github.com/minetest/minetest/issues/7192 -- Biome API / cavegen: Add definable cave liquid for a biome by paramat |
| 23:17 |
|
twoelk left #minetest-dev |
| 23:31 |
|
longerstaff13-m joined #minetest-dev |
| 23:37 |
|
Cornelia joined #minetest-dev |
| 23:40 |
|
longerstaff13-m joined #minetest-dev |
| 23:42 |
|
proller joined #minetest-dev |
