On 04/10/2019 22:24, Justin Hibbits wrote: > On Fri, 4 Oct 2019 15:06:52 -0400 > Dennis Clarke <dclarke_at_blastwave.org> wrote: > >> On 10/4/19 10:05 AM, Andriy Gapon wrote: >>> >>> Does anyone use ZFS with a 32-bit kernel, that is also not i386 ? >>> If you do, could you please let me know? Along with uname -rmp >>> output. Thank you! >>> >> >> I don't know if that has even been attempted by anyone. The ZIL and >> ZFS log comonents require substantial amounts of memory and I am not >> aware of anyone with arm devices that have 8GB+ of memory. I have had >> FreeBSD current on RISC-V running fairly well with ZFS however that >> was a purely rv64imafdc architecture. >> >> I will watch this thread with curiosity. >> >> > > I did try using ZFS on 32-bit powerpc (8GB RAM), and even got a bugfix > pushed into the ZFS/Illumos repo for it, but it was too unstable to be > usable. I'd love to try again later though. The instability was not surprising, IMO. ZFS depends on 64-bit atomic operations. For platforms that do not provide them there is an attempt to emulate them, but it is not quite consistent. In particular, reads are just plain reads on all platforms, so torn values can be seen on all 32-bit platforms. But it's easier to fix those of them that have 64-bit operations. I am thinking about proposing to remove ZFS support from platforms that do not provide 64-bit atomics for the kernel. I think that those are 32-bit powerpc-s and some flavors of 32-bit mips, but I am quite confused about the latter. There are so many of them: mips.mips, mipsel, mipshf, mipselhf, mipsn32. -- Andriy GaponReceived on Sat Oct 05 2019 - 05:58:04 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:22 UTC