Forza

It's known that some nas vendors have a customised btrfs that isn't fully compatible. It's why I asked if the filesystem was created after debian was installed. It unfortunately seems...

@dehnhard if you use plain Debian tools, then it couldn't be anything to do with the SoC, unless the tools and kernel from Debian are custom Zyxel made? @kdave is...

I think we can rule out custom modifications by Zyxel, etc. Shame on me 🙄, but I dumped the `send.h` together with the he error code from the original post...

@lrcressy it's a very unfortunate thing that happened to you. When it comes to the `single` profile, there are several things that makes what you want very difficult. If you...

Hi @EldarAgalarov, Do you have some data to backup those claims? Also, different aechitectures may behave differently. Intel CPUs do have hw instructions for crc32c, which makes it very fast...

I have an ARM64 system hosted on a Neoverse-N1. ``` ❯ ./hash-speedtest CPU flags: 0x1 CPU features: Block size: 4096 Iterations: 100000 Implementation: builtin Units: CPU cycles NULL-NOP: cycles: 0,...

> those `hash-speedtest` is part of btrfs-progs sources. I thought it was meant as a better test of supported speeds?

Here are some tests using the various backends. Still, I am assuming they do not reflect actual Btrfs performance, even when comparing relative performance? It would be more useful if...

@DaLiV As you clearly found out, one of the worst case scenarios for current Btrfs design is an fsync after each small write. On ZFS, which is also a COW...

@DaLiV what is your goal with your posts? I don't see anything constructive with your remarks. Btrfs has many feature other Linux filesystems do not, which to me are more...