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()

Am Donnerstag, 26.06.03 um 08:59 Uhr schrieb Christopher Sean Morrison:

Sparked by the discussions a couple months ago on the "faster malloc
implementation" thread, a comparative analysis of the Darwin 6.6 Libc
malloc implementation has been put together. The memory test was run on a
Ti notebook running Mac OS X and Yellow Dog Linux. The same test was also
run on a Xeon and IRIX server for comparison and reference. In short,
there are definately issues and cause for concern with the present Darwin
implementation; there were several surprises found, both good and bad.

Hopefully, things may be changing significantly with Darwin 7.0 given the
possible availability of another better-performing malloc implementation.
This hope is based on the SPEC benchmark reports released for the new G5
chips (that reportedly used a fast malloc implementation that will be
available in 10.3). Or perhaps this report will spark an increased
concern and demand for a better malloc implementation.

Either way, I'm providing the entire analysis, source code, log files, and
graphs for all to see and review. This analysis is intended to test a
very simple and specific domain of malloc performance behaviors. The
report provides all of the details.

http://intelligentcode.com/memoryTest/ [main]
http://homepage.mac.com/brlcad/memoryTest/ [mirror]



[ Oh man, if there is a better malloc in Darwin 7.0, that's inconvenient news for me. I hope its not faster than the one I am writing, otherwise that'd mean quite a lot of development time down the toilet :) ]

I looked at some of above results, but I don't know. maybe I am just too dumb to read them. f.e.

http://intelligentcode.com/memoryTest/Graphs/PNG/Test-Darwin-Linux- IRIX-65536-1048576-Real.pict.png

or

http://intelligentcode.com/memoryTest/Graphs/PNG/Test-Darwin-Linux- IRIX-1048576-67108864-Real.pict.png

are these graphs supposed to mean, that the more memory is allocated, the faster the allocation happens ? That'd be one funny malloc. Also how come the allocation unit size is something like between 100000 bytes and 7e+07, ain't that enormous allocation units ? That doesn't sound right to me. If its right, who cares how fast that is ? You usually allocate that big blocks once or twice during the lifetime of the program. :)

Please shed some light on this.

Ciao
Nat!
------------------------------------------------------
Some people drink deep from the fountains of life, and
some just gargle. -- DLR
_______________________________________________
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.