New Patches Aim To Boost Linux 9p Performance By ~10x

[phoronix]

Phoronix: New Patches Aim To Boost Linux 9p Performance By ~10x

A new set of patches posted for the Plan 9 (9p) resource sharing protocol code inside the Linux kernel can deliver roughly 10x better performance for file transfers...

https://www.phoronix.com/news/Linux-9p-10x-Performance

[Amano]

I guess that those are the same optimisations that went into the new QEMU version. That promised a 6x to 12x speedup as well: https://www.phoronix.com/news/QEMU-7.2-Released I guess that those patches don't add up (QEMU obviously using the code internally and thus independently from the Linux kernel).

[jarekZ]

v9fs is a Unix implementation of the Plan 9 9p remote filesystem protocol.

For remote file server:

mount -t 9p 10.10.1.2 /mnt/9

For Plan 9 From User Space applications (http://swtch.com/plan9)

mount -t 9p `namespace`/acme /mnt/9 -o trans=unix,uname=$USER

For server running on QEMU host with virtio transport:

mount -t 9p -o trans=virtio <mount_tag> /mnt/9

TIL. Is this actually useful?

[ermo]

TIL. Is this actually useful?

It was certainly designed to be. YMMV obviously.

[iustinp]

It was certainly designed to be. YMMV obviously.

Well, what we need now is someone to test sharing via samba/+cifs(guest) vs. 9p. Curious myself…

[ryao]

TIL. Is this actually useful?

I have used it for doing ZFS development. I understand that ARM uses it for development too.

It is nice to be able to run the guest’s root filesystem in the host to do builds that are immediately available in the guest. It has no need to do mount/unmount.

[tracker1]

Hopefully this lands in WSL soon and improves the IO performance there as well.

[reza]

Hopefully this lands in WSL soon and improves the IO performance there as well.

I was actually about to say the same thing. I posted this in WSL github but so far no answer. Looking at the patch and the official Linux repo, I don't see the patch applied yet. Any input?