Assembly string functions in i386 libc
Sean C. Farley
scf at FreeBSD.org
Thu Jul 12 22:18:18 UTC 2007
On Fri, 13 Jul 2007, Dag-Erling Smørgrav wrote:
> "Sean C. Farley" <scf at FreeBSD.org> writes:
>> On Thu, 12 Jul 2007, Dag-Erling Smørgrav wrote:
>>> The first rule of optimization is: don't do it.
>>> The second rule of optimization is: don't do it yet.
>>> The third rule of optimization is: don't optimize what you haven't
>>> measured.
>> I am a rule breaker at least for the first two. :) I tried to
>> follow the third rule.
>>
>>> Can you show us an actual application that spends a significant part
>>> of its run time in strlen()?
>> My test program that loops over strlen().
>
> So the answer is no, and you don't understand the third rule which you
> claim to follow.
I never claimed to succeed; I only tried. The two types of tests I can
think would be useful were execution of strlen() by itself and within a
common program. I had thought I had tested the first type of test. The
second one I did later with different versions of strlen() using diff.
What types of tests would have been better? I know about gprof for
finding where a specific program is spending its time, but I was
focusing on the strlen() call.
Also, having two different versions in the source tree is still puzzling
me.
Sean
--
scf at FreeBSD.org
More information about the freebsd-arch
mailing list