Re: [PATCH v2 2/3] 9p: Introduce option for negative dentry cache retention time

[Christian Schoenebeck]

On Thursday, 12 February 2026 10:24:27 CET Remi Pommarel wrote:
> On Wed, Feb 11, 2026 at 04:58:02PM +0100, Christian Schoenebeck wrote:
> > On Wednesday, 21 January 2026 20:56:09 CET Remi Pommarel wrote:
[...]
> > Wouldn't it make sense to enable this option with some meaningful value
> > for
> > say cache=loose by default? 24 hours maybe?
> 
> That is an interesting question, I have seen pretty satisfying (at least
> for me) perf results on the different builds I ran, even with a 1 to 2
> seconds cache timeout, maybe this would be a good tradeoff for
> cache=loose being almost transparent in the eye of the user ? But maybe
> this is too specific to the build workflow (that hit the same negative
> dentries fast enough) ?

Always hard to pick magic numbers. But I would also say that 1s...2s is 
probably a use-case specific pick specifically for compiling sources.

When running 9p as rootfs you will also frequently run into libs querying the 
same non-existing configuration files and DLLs over and over again. So I would 
pick a higher value. Personally I would be fine with anything between few 
minutes ... 24h for cache=loose. For other cache modes this could be lower.

/Christian