On Sun, 2019-06-30 at 08:57 -0700, Matthew Wilcox wrote: > On Sun, Jun 30, 2019 at 09:52:29AM -0400, Trond Myklebust wrote: > > +/* FIXME: dynamically size this for the machine somehow? */ > > +#define NFSD_FILE_HASH_BITS 12 > > +#define NFSD_FILE_HASH_SIZE (1 << > > NFSD_FILE_HASH_BITS) > > +#define NFSD_LAUNDRETTE_DELAY (2 * HZ) > > Isn't this what rhashtable is for? Maybe. I'm less concerned than Jeff was over the size of the hash table. 4096 buckets should scale quite well up to the millions of entries. We're highly unlikely to have that many active files at the same time, and particularly not for NFSv3, which is the main critical user of this lookup table. -- Trond Myklebust Linux NFS client maintainer, Hammerspace trond.myklebust@xxxxxxxxxxxxxxx
