Re: [PATCH 02/16] user_ns: use new hashtable implementation

[ebiederm@xmission.com (Eric W. Biederman)]

Sasha Levin <levinsasha928@gmail.com> writes:

> On 08/15/2012 01:52 AM, Eric W. Biederman wrote:
>> Sasha Levin <levinsasha928@gmail.com> writes:
>>> > Switch user_ns to use the new hashtable implementation. This reduces the amount of
>>> > generic unrelated code in user_ns.
>> Two concerns here.
>> 1) When adding a new entry you recompute the hash where previously that
>>    was not done.  I believe that will slow down adding of new entries.
>
> Since the hashtable doesn't expose the internal hashing to the user of the
> hashtable it would require adding a new interface to deal with in.
>
> I don't feel that a whole new interface to optimize out something which is very
> cheap (one multiplication + shift) in this case, I'd rather avoid having a new
> interface.
>
> Is it ok with you if I leave it as is in the next version of the
> patch?

No problem.  I was largely reviewing for potential gotchas, and things
that make me go hmm.

I don't think setuid is called enough for the hash table to be much of
a hotpath.  Although I haven't profiled it.

Eric