[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: malloc performance
From:       Andrea Aime <aaime () libero ! it>
Date:       2002-02-14 18:25:45
[Download RAW message or body]

>  Hello,
>
>  see http://sources.redhat.com/ml/libc-alpha/2002-02/msg00107.html for
> details. In short, we're using malloc() very extensively, and a noticeable
> part of the execution time is spent handling dynamic allocations. 

I don't know anything about QT internals, but I've noticed that glib
uses special allocators to reduce the number of malloc calls on dynamic
data structures, that is, instead of allocating one linked list node at a 
time, the library allocs 128 of them in one shot, thus reducing the need for 
malloc calls...  is QT doing the same or is this an area for improvement?
I see the impressive performace gain you report on a huge HTML page load...
the responsible for such a high number of allocations is the DOM, I 
suppose (just a speculation), so, wouldn't a change in the DOM memory 
allocation strategy provide some performance increase in page loading?
Just my 2 cents... (and sorry for my bad english)
Best regards
Andrea Aime
[prev in list] [next in list] [prev in thread] [next in thread]