On Mon, Jun 24, 2013 at 12:19 PM, Daniel Shahaf <danie...@apache.org> wrote:
> On Mon, Jun 24, 2013 at 10:15:31AM -0000, danie...@apache.org wrote: > > Author: danielsh > > Date: Mon Jun 24 10:15:31 2013 > > New Revision: 1495978 > > > > URL: http://svn.apache.org/r1495978 > > Log: > > Silence a compiler warning. > > > > * subversion/libsvn_fs_fs/tree.c > > (cache_lookup): Only declare FACTOR when it is used. > > > > Modified: > > subversion/trunk/subversion/libsvn_fs_fs/tree.c > > > > Modified: subversion/trunk/subversion/libsvn_fs_fs/tree.c > > URL: > http://svn.apache.org/viewvc/subversion/trunk/subversion/libsvn_fs_fs/tree.c?rev=1495978&r1=1495977&r2=1495978&view=diff > > > ============================================================================== > > --- subversion/trunk/subversion/libsvn_fs_fs/tree.c (original) > > +++ subversion/trunk/subversion/libsvn_fs_fs/tree.c Mon Jun 24 10:15:31 > 2013 > > @@ -342,8 +342,10 @@ cache_lookup( fs_fs_dag_cache_t *cache > > apr_size_t path_len = strlen(path); > > apr_uint32_t hash_value = (apr_uint32_t)revision; > > > > +#if SVN_UNALIGNED_ACCESS_IS_OK > > /* "randomizing" / distributing factor used in our hash function */ > > const apr_uint32_t factor = 0xd1f3da69; > > +#endif > > > > Is this the right fix? Right now, the !SVN_UNALIGNED_ACCESS_IS_OK hash > function is just (x[n] + 33^1 x[n-1] + ... + 33^(n-1) x[1] + 33^n x[0]). > Yep. That is good enough for the access patterns the we see here: tree walks. -- Stefan^2.
