Re: v2 [PATCH 4/4] nsswitch: use new internal API

[DJ Delorie <dj@redhat.com>]

"Carlos O'Donell" <carlos@redhat.com> writes:
> The system works perfectly fine, but in the tests I ran the /etc/nsswitch.conf
> doesn't reload.

Heh, yeah, the tests I'm writing also pointed this out.  The complex
lookups seem to work, but the simple ones were cached.

> It doesn't reload because of code like this:

Yup.

> (1) Test covering changing data after reload for host interfaces.
>
> * New container test case that has 2 NSS services for hosts providing
>   different data.

Note that the two services we've used in the past for this do *not* have
host support (just passwd and group) so this will take some time for me
to add support for.

> * Start the test with one set of data and call:
>   * gethostbyname
>   * gethostbyname_r

The non_r functions just call the _r functions, so do we really need to
test both?

> Yes, it means we will stat /etc/nsswitch.conf for every API call, but that
> is in the noise compared to the network costs of going to DNS/LDAP or
> parsing files. I don't think the added stat cost will make much of a difference
> compared to reading and parsing lines from files.

We still have some optimization opportunities for the future, though.
Florian a I decided to keep as much of the internal API the same as
possible for the first pass, so for example, we still look up names by
string compares instead of enums, and still have unneeded layers of
complexity that could be optimized away if they were written
differently.