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: [RFC] Faster memset.


On 04/10/2013 02:25 AM, OndÅej BÃlka wrote:
>>> On previous test my implementations gains mostly when current
>>> implementation computed jump is not in cache and this benchmark
>>> underestimates this factor.
>>
>> So a win for one and a loss for the other.
>>
>> How much of a win and how much of a loss?
>>
> When I did profiling it supports theory that cache cost dominates and my 
> implementation is faster. Result is here. 
> 
> http://kam.mff.cuni.cz/~ondra/benchmark_string/memset_profile/result.html

Nice results.

> Results are slower when random test, I am not sure why.

I think we need to answer that before we can move forward.

I don't have any interesting ideas though.

Cheers,
Carlos.


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