segfault sounds strange, have you changed any essential part of your machine after building?
My first attempt would be to rebuild and reinstall.
Some feedback to performance and memory usage would be nice.
I managed to fix the mimalloc package, so there’s no need to manually create the symlink anymore.
Hi @Guzzisti,
Ok I uninstalled and rebuilt Art from Aur, and after another crash (?) it all works now.
I looked at the Ram’s power consumption with both versions of Art (the Aur version and the self-compiled version with build-art). I did exactly the same manipulations with both softwares (going from picture to picture by zooming in at 100% and coming back to a full view) on 5 pictures). It seems that the Aur version (above) is more greedy in Ram, the performances being about the same.
strange things happen on your machine…i have no clue why “after another crash” it should work and before it did not work… 
As far as i can see build-art uses tcmalloc by default. Did you install the AUR “release” version or the “git” version? Only the latter uses TCMALLOC currently.
Do both version use the same settings (especially “Performace” tab)?
A higher ram usage is not necessarily a bad thing, it just shouldn’t skyrocket and start to use swap.
I used the git version and indeed there are the same settings in the performance tab.
Maybe the crash of the first startup is due to the fact that they are the same cache folders used for the successive versions ?
Otherwise both versions work fine, I just found it amazing that the Ram usage is different.
But with 32GB of Ram I have a lot of margin left.
I may have a weird machine, but I swear I’m not changing anything !
Every time I update the Aur repository, Art crashes with the same error as above… I have to uninstall and reinstall the software for it to start…
here is what happens when Aur version 1.2.79 crashes
I didn’t do a real benchmarking, but I didn’t see any significant difference between TCMALLOC and MIMALLOC in terms of performance nor in memory usage.
HELP ME please!!!
The latest version of Art doesn’t even open anymore whether it’s the Aur version or the compiled version.
I’ve uninstalled everything, reinstalled everything, it doesn’t matter. The old spotremoval branches versions work… what’s going on ? how can I get you a bug report ? I’m on Manjaro, with a 2011 pro mac. Thanks for your help.
I’ve never encountered such a behaviour, so i can only place a rough guess: maybe something got messed up when using multiple version with the same config/cache folder.
Have you tried to clean to ART-related folders in ~/.cache (and may ~/.config as well)?
On W10, I made a build with mimalloc. I verified mimalloc is linked and it seems ok.
In my normal use of ART I don’t really feel a performance difference. So my question is how to benchmark? do I need to stress ART opening a lot of editor windows? What operation to perform? What I have to measure?..
by deleting the Art folder from .cache and .config, the Aur version restarts. but not the compiled version. The problem isn’t totally solved, but… The software often crashes while loading photo files…
Thanks @guzzisti
I think I should repeat: if it ain’t broke, don’t fix it. I normally use tcmalloc for all my programs, so I recommend to stick to it if possible. I’ve added mimalloc because I recently found out about it and I’m somewhat curious, but I don’t expect that it makes a significant difference (at least on Linux, on windows it might but I have no experience there)
@jllailes I really don’t know what’s going on with your system, those crashes are strange…no idea, sorry 
For information, after a lot of compilation tests, it seems that the problem for me comes from the build-art script. Indeed, any compilation with this script generates a binary that doesn’t run anymore.
Whereas a compilation with the manual method of rawtherapee (in rawpedia) works. Art works perfectly…
Seems like to transition to GCC 10.1 on Arch/Manjaro broke something. ART builds fine but segfaults on startup tough.
I’m investigating this.
On W10 with GCC 10.1,it doesn’t segfault but the images get pink.
seems to be linked to -ftree-vectorize, see pink color cast when using input profile when rawtherapee is compiled with gcc-10 and -O3 · Issue #5749 · Beep6581/RawTherapee · GitHub
I’m doing some build tests right now on my zen-based system, seems like it depends on a combination of optimization params.
Nevertheless i can say if ART compiles & runs successfully with GCC10.1, than the “pink image bug” as there as well. @Daniel_Catalina already created a bug report for ART this one:
@guzzisti what happens if you compile with -fno-tree-loop-vectorize? Is that enough to make it go away?
I’m currently concentrating on the segfault on startup which might be related to tcmalloc as far as my current test data shows. I’ll try -fno-tree-loop-vectorize as well.