Mailing Lists: Apple Mailing Lists
Image of Mac OS face in stamp
 
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Performance of the Darwin 6.6 Libc malloc()

> I see so maybe you want your allocation rate to be labeled something
> like GB /s rather ?

This is what the graph units label should indeed read (and what it
actually did read in a previous incarnation).. The graphs will be updated
to reflect the proper units of GB/second for the y-axis; thank you for
noticing this oversight.

> The time used to actually use that kind of memory (f.e. memcpy( dst,
> src, 100000)) dwarfs the time that the allocation spends. So it is
> uninteresting.

The test is meant to be a comprehensive analysis of malloc() performance.
Questions and factors concerning data or memory usage were not within the
scope of the analysis.

The primary reason for halting the testing at 64 MB size allocations was
only due to the resolution of the timing mechanisms and memory available
for allocation. The tests completed too quickly to be comparably
meaningful across implementations. We found the test and results rather
interesting regardless; the purpose of the test is to observe and analyse
the performance of malloc() (only) as it compares to other implementations
regardless of usage.

Thank you to all who have contacted me with feedback, corrections, and
comments. Keep it coming. :) Those who have sent me questions and
concerns will hopefully find them addressed in the final paper.

Cheers!
Sean
_______________________________________________
darwin-development mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-development
Do not post admin requests to the list. They will be ignored.


Visit the Apple Store online or at retail locations.
1-800-MY-APPLE

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2007 Apple Inc. All rights reserved.