This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Improve bench-strlen


Hi Siddhesh,
  
>On 21/12/18 8:26 PM, Wilco Dijkstra wrote:
>> The current bench-strlen compares against a slow byte-oriented strlen which
>> is not useful given it's too easy to beat.  Remove it and compare against the
>> generic C strlen version and memchr.
>> 
>> OK to commit?
>
> OK.  We should do this for all functions.

Absolutely. I already did a few more, but wanted to get one out for review to
get some feedback. I'm also planning to remove the "stupid_xxx" variants as
they don't make sense at all.

The other thing we need to do is increase iterations to get more accurate results
and sort out the odd buffer handling from page counts to bytes (with the default
increased so you don't have to request more in many cases). And then we can
look into improving the actual tests run, for example I already mentioned we run
too many combinations in various tests (particularly unaligned) for the results to
be meaningful.

Cheers,
Wilco

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]