inlined string tests are slower on i686

[Ulrich Drepper]

Greg McGary <greg@mcgary.org> writes:

> The non-inlined version is 4% faster
> 
> Does anyone have more comprehensive benchmarks that justify string
> inlines for ix86?

I expect something like this.  The increased code size is of course
visible.  But this is not the normal case.  In a normal program you'll
never find that many string functions in one place and then the jump
to a not so often used function is more expensive.

-- 
---------------.                          ,-.   1325 Chesapeake Terrace
Ulrich Drepper  \    ,-------------------'   \  Sunnyvale, CA 94089 USA
Red Hat          `--' drepper at redhat.com   `------------------------