hybrid malloc() ?

[john moser <bluefoxicy at linux dot net>]

Hi.  Please CC me all replies.

I've been trying to get an idea accross about a hybrid malloc() allocator
worked off the current allocator (heap/mmap).  The idea is that the new
allocator would be much more efficient than the old in many cases.  A
simple outline of these cases is below.

 - In cases where an area of unused, allocated memory touches or crosses
   a page boundary, that area can be used for any allocation, including
   those larger than itself.
 - In cases where an area of used memory needs to be expanded, i.e. via
   realloc(), and crosses a page boundary, that area can be partially
   expanded in place, lessening the copying overhead.
 - In cases where an allocated page exists and is spanned entirely by
   an unused area of memory, that page can be freed to the system.

It has a setback as well.

 - In the normal operation of a program, many physical pages will be
   mapped to two or more areas of virtual memory.

The idea is on the redhat and sources bugzilla, and I blogged about it.
Nobody's replied to any of it, so I'm assuming the idea has not been
seen.  Relavent links follow.

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=123965
http://bluefoxicy.blogspot.com/2004_05_01_bluefoxicy_archive.html#108507154064758877
http://sources.redhat.com/bugzilla/show_bug.cgi?id=167



_____________________________________________________________
Linux.Net -->Open Source to everyone
Powered by Linare Corporation
http://www.linare.com/