Daan

Thanks -- I got it running locally; but it doesn't quire reproduce. I set it to 100 iterations using the _latest_ `dev3` branch. The memory usage is much lower though...

I got it running now -- it still uses much less memory but I can see it grow now, from about 240MiB to 480MiB after 100 iterations. One thing I...

Thanks! It is good to see that v3 does much better as it was essentially redesigned to deal with threadpools better. Just to clarify, the _threads_ on Linux stay constant...

Hi @pitrou : can you repro reliably? I think the actual difference between 2.1.7 and 2.1.9 is not that large so maybe you can pinpoint the commit that causes the...

I missed the earlier comment :-(. But this is fixed now in commit 1c51484 -- however, it was really an artifact of another bug which was fixed in 3e32b4c ....

DIfficult to say what is wrong: - make sure you did follow all instructions -- is `mimalloc-redirect.dll` in the same directory as the executable? - run with `MIMALLOC_VERBOSE=1` in the...

Thanks. I hope it is working now for you?

Thanks for the PR and putting attenting to the `MADV_FREE_REUSABLE` flags! However, this needs a bit more thought I think to understand the implications well (and probably it is something...

Thanks @Jura-Z for the reply. I am surprised that `MADV_DONTNEED` doesn't do the right thing and you see the difference with `MADV_FREE_REUSABLE` -- I will work on using that flag...

I hesitantly added initial support for these flags; it is active in a non-debug build or if `MIMALLOC_PURGE_DECOMMITS=0`. Can you try if the latest `dev3` (or `dev`/`dev2`) improves matters for...