Re: NCHNAMLEN vnode cache limitation removal

[David Holland <dholland-tech%netbsd.org@localhost>]

On Wed, Sep 11, 2019 at 06:49:05AM +0300, Jason Thorpe wrote:
 > > On Sep 11, 2019, at 2:23 AM, Christos Zoulas <christos%zoulas.com@localhost> wrote:
 > > 
 > > Comments?
 > 
 > This looks good, and I think it would be fine for it to go in now.
 > However, I think we should probably instrument how many duplicate
 > names may end up in the name cache over the course of "normal"
 > operation (insert "standard" workloads here).  It may be worth
 > implementing a generic string cache to de-dup those names (which
 > has the secondary advantage of making the name cache entry itself
 > fixed-size again).

I'm confused; nothing in there should lead to duplicate entries...

(also it would be helpful to hear from someone who has an informed
opinion about how rearranging the structure is likely to affect cache
behavior, since my understanding is that namecache lookups are both
critical and sensitive.)


-- 
David A. Holland
dholland%netbsd.org@localhost