kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
BZK[m] has joined #openscad
J231725 has quit [Quit: Client closed]
J231725 has joined #openscad
Bocaneri has joined #openscad
Bocaneri is now known as Guest2961
Sauvin has quit [Ping timeout: 255 seconds]
Guest2961 is now known as Sauvin
guso78 has joined #openscad
SamantazFox has quit [Quit: Bye]
SamantazFox has joined #openscad
rawgreaze has quit [Ping timeout: 255 seconds]
rawgreaze has joined #openscad
guso7858 has joined #openscad
guso78 has quit [Ping timeout: 245 seconds]
SamantazFox has quit [Remote host closed the connection]
kow__ has joined #openscad
SamantazFox has joined #openscad
drkow has quit [Ping timeout: 248 seconds]
castaway has joined #openscad
teepee has quit [Ping timeout: 240 seconds]
teepee has joined #openscad
SamantazFox has quit [Remote host closed the connection]
castaway has quit [Ping timeout: 246 seconds]
castaway has joined #openscad
SamantazFox has joined #openscad
guso7858 has quit [Quit: Client closed]
L29Ah has joined #openscad
J231725 has quit [Quit: Client closed]
J231725 has joined #openscad
guso78 has joined #openscad
J231725 has quit [Quit: Client closed]
J231725 has joined #openscad
castaway has quit [Ping timeout: 246 seconds]
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
kintel has joined #openscad
<kintel>
Does anyone know a good way of detecting and reporting when the compiler gets OOM killed on Github CI?
<guso78>
i think it should be clearly be visible in the server logs if a process was killed due to OOM. question is if this info is available to the user.
<kintel>
Yeah, I can probably grep for magic strings in the logs, was hoping to have that surfaced in a more robust way.
<guso78>
i am wondering why its so silent here for almost one day. presumably people meet on the booth instead '=D
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
<guso78>
are there any free RDP windows clients available in the internet where i can test openscad installation and these machines have a "dispose" button so i can repeatedly test/improve the installation ?
pa has quit [Ping timeout: 240 seconds]
pah has joined #openscad
<InPhase>
kintel: "echo $?" will be 137 after OOM hits a process. It would be 139 for something like a segfault, so you can distinguish these.
<kintel>
InPhase Thx. do you know if e.g. cmake calls make, make calls gcc and gcc is killed, will the 137 code bubble up to cmake?
<kintel>
(for simplicity, this is a Linux runner)
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<InPhase>
kintel: Uhm, no, it seems to eat those.
<InPhase>
kintel: Hopefully you'll check the log. This will highlight it red, if we're fortunate enough that github picks up the ansi coloring: cmake --build . 2>&1 | grep --color=always -E ' Killed$|'
<InPhase>
kintel: Or the whole line like: cmake --build . 2>&1 | grep --color=always -E '.+ Killed$|'
<InPhase>
But I guess, who knows if that extra step messes anything else up.
Virindi has quit [Ping timeout: 248 seconds]
guso78 has quit [Quit: Client closed]
castaway has joined #openscad
ToAruShiroiNeko has quit [Ping timeout: 276 seconds]
Virindi has joined #openscad
To_Aru_Shiroi_Ne has joined #openscad
<teepee>
maybe that would want to be ... | tee | grep ... so there's some output on stdout, not 100% sure about github but some CI have a lower kill limit if there's no output anymore
<InPhase>
teepee: That subtle trailing | before the closing quote makes sure it all stays on stdout.
<InPhase>
teepee: i.e., highlight if it matches the part to the left, "or" just print everything else.
<teepee>
ah, interesting feature
<InPhase>
But what it does do, is eat stderr by bumping it into stdout. Not sure if that messes up any CI features watching for stderr content.
<teepee>
I would expect that to be fine, as no error output is really wanted. but I suppose we'll know only for sure after really running it
<lf94>
guso78[m]: how's it going
J231725 has quit [Quit: Client closed]
J231725 has joined #openscad
<guso78[m]>
Lf94 i am thinking about the Details of a New transforming Operation ...
<lf94>
nice
<lf94>
I'm porting over noise stuff to do terrain (inspired by J23)
<lf94>
from iq's site
<guso78[m]>
I knew you are refering to the Terrain.
<guso78[m]>
Maybe you want to add to my Landscape library 🤣
Bocaneri has joined #openscad
Bocaneri is now known as Guest8107
peeps[work] has joined #openscad
Sauvin has quit [Ping timeout: 276 seconds]
pah is now known as pa
<J231725>
perlin noise would be nice ( so i don't need to write it myself )
J231725 is now known as J23
guso78 has joined #openscad
teepee_ has joined #openscad
teepee has quit [Ping timeout: 240 seconds]
teepee_ is now known as teepee
Virindi has quit [Ping timeout: 260 seconds]
Virindi has joined #openscad
<J23>
can someone explain why pow( expression, 2 ) is working fine but pow( expression, 1.8) causes a program freeze ?
Guest6514 has joined #openscad
Guest6514 has quit [Client Quit]
<guso78>
you refer to openscad pow ?
<InPhase>
J23: Is expression negative?
<InPhase>
J23: You might want an abs in there.
<J23>
it is a sin (.... ) so yes it can be negative
<J23>
and if use abs i will loose the wave form
<InPhase>
J23: Most likely it's not a freeze, but a massive multiple in runtime if this is called frequently, as there is huge overhead on the nan processing. Likewise for any undefs. Especially if they propagate into geometry.
<InPhase>
:) That's awesome. Please save that for the calendar.
<J23>
at least it is written without my lib .. (but now is a new module in my lib )
<InPhase>
Maybe a very modest tweak to make a perfect animation loop.
<J23>
yeah looping just need to be with multiple of the frequencies to fit .. but there are 6 or so .. but yeah can be done ( and for a long enough loop still looks nice)
<InPhase>
peeps[work] had a really nice water animation some time ago based on... was it bicubic interpolation?
<InPhase>
But it did not have a beach ball!
<J23>
oh that is cool bicubic is bit like perlin noise
<InPhase>
And yeah, that one was more of a standing wave effect.
<J23>
can't wrap my head around .. bicubic give you the 3rd dimension -- wouldn't tricubic give the 4th dimension as output?
<InPhase>
peeps[work]: Also, that scad file was the original mental inspiration for the snowballs I put into the calendar. You wouldn't notice it because there probably aren't two sequential characters in common, but it mapped there in my head. :)
<peeps[work]>
J23: 4th dimension of time is needed for animation
<peeps[work]>
the nice thing is that it is also periodic in x,y, and t