[Development] Proposal for "container-oriented deterministic memory manager"

[Phil Bouchard]

On 01/05/2017 06:02 AM, Konstantin Tokarev wrote:
>
>
> 05.01.2017, 03:46, "Phil Bouchard" <philippeb8 at gmail.com>:
>> On 01/02/2017 04:50 PM, Phil Bouchard wrote:
>>>  On 12/29/2016 04:14 AM, Simon Hausmann wrote:
>>>>  Hi,
>>>
>>>  Sorry for the delay...
>>>
>>>  First I would like to point out this popular Javascript test runs 1.5
>>>  faster using Qt over WebKit:
>>>
>>>  - ~100 FPS on my laptop (x86 @ 2.40GHz) using Chrome (WebKit):
>>>  http://www.themaninblue.com/experiment/AnimationBenchmark/html/
>>>
>>>  - ~150 FPS on my laptop (x86 @ 2.40GHz) using Qt (without QtQuickCompiler):
>>>  http://finitetheory.com/personal/phil/JSBenchmark.tar.gz
>>>
>>>  - I am still waiting for the QtQuickCompiler request to be fulfilled but
>>>  anybody who has it already is welcome to try it out and please let us
>>>  know the results.
>>
>> For the record I was able to benchmark the QtQuickCompiler on a x86 @
>> 3.4 GHz and I get: ~250 FPS and without the QtQuickCompiler I get
>> something similar; which means this is pretty much the maximum speed
>> that test can get.
>
> AFAIU QtQuickCompiler has nothing to do with memory management, its main purpose is reduction of start up time and obfuscation of sources.

Ok I assumed that execution time would be affected because the code is 
compiled.

> Note that measuring FPS above screen refresh rate (usually 60) does not make a lot of sense, as underlying graphics systems may not preserve it (and if vsync is enabled, screen refresh rate is a hard upper bound).

You're right but I just wanted to have numbers.

Simon was mentioning that it would be preferable having root_ptr with an 
underlying mechanism to select memory pools based on the type allocated 
or the frequency of the allocation made.  I don't see how it could be 
further optimized.

Does anybody have better memory management benchmarks I could focus on?


Thanks a lot,
-Phil