linux-kernel - Re: [PATCH v2 0/3] 9p: Performance improvements for build workloads

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <4711141.LvFx2qVVIh@weasel>
Date: Wed, 04 Feb 2026 12:37:55 +0100
From: Christian Schoenebeck <linux_oss@...debyte.com>
To: v9fs@...ts.linux.dev, Remi Pommarel <repk@...plefau.lt>
Cc: linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
 Eric Van Hensbergen <ericvh@...nel.org>, Latchesar Ionkov <lucho@...kov.net>,
 Dominique Martinet <asmadeus@...ewreck.org>,
 Remi Pommarel <repk@...plefau.lt>
Subject: Re: [PATCH v2 0/3] 9p: Performance improvements for build workloads

On Wednesday, 21 January 2026 20:56:07 CET Remi Pommarel wrote:
> This patchset introduces several performance optimizations for the 9p
> filesystem when used with cache=loose option (exclusive or read only
> mounts). These improvements particularly target workloads with frequent
> lookups of non-existent paths and repeated symlink resolutions.
[...]
> Here is summary of the different hostapd/wpa_supplicant build times:
> 
>   - Baseline (no patch): 2m18.702s
>   - negative dentry caching (patches 1-2): 1m46.198s (23% improvement)
>   - Above + symlink caching (patches 1-3): 1m26.302s (an additional 18%
>     improvement, 37% in total)
> 
> With this ~37% performance gain, 9pfs with cache=loose can compete with
> virtiofs for (at least) this specific scenario. Although this benchmark
> is not the most typical, I do think that these caching optimizations
> could benefit a wide range of other workflows as well.

I did a wide range of tests. In broad average I'm also seeing ~40% improvement 
when compiling. Some individual sources even had 60% improvements and more. So 
there is quite a big variance.

I did not encounter misbehaviours in my tests, so feel free to add:

Tested-by: Christian Schoenebeck <linux_oss@...debyte.com>

I still need to make a proper review though.

/Christian