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] benchtests/Makefile: Run the string benchmarks four times by default.


On Wed, Sep 4, 2013 at 11:52 AM, OndÅej BÃlka <neleai@seznam.cz> wrote:
> On Wed, Sep 04, 2013 at 05:20:23PM +0100, Will Newton wrote:
>> On 4 September 2013 17:17, OndÅej BÃlka <neleai@seznam.cz> wrote:
>> > On Wed, Sep 04, 2013 at 03:48:24PM +0100, Will Newton wrote:
>> >>
>> >> The string benchmarks can be affected by physical page placement, so
>> >> running them multiple times is required to account for this. Also
>> >> backup the results of the previous run like is done for the other
>> >> benchmarks.
>> >>
>> > You do not need to do this. We should instead randomize addresses used
>> > which handles this problem.

What is the objective of randomizing the addresses? What are we trying
to accomplish by different page placement?

>> That seems like it would be considerably more complicated. Do you have
>> a reason why your approach is better?
>>
> It does not matter if it is complicated, it is required to get
> reasonable results.

Your proposed solution (in a later email) is in-fact not
overly-complicated, but for the record, there is a very good reason to
avoid complexity in a project like glibc, which is already extremely
complex.  Some poor soul will need to maintain all of the complicated
solutions long into the future.

Ryan


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