Re: New classes of possible memory optimizations

[Jeff Morriss <jeff.morriss.ws () gmail com>]

Evan Huus wrote:
> It is now possible with wmem to free individual allocated chunks (this
> does not affect the bulk-free that occurs after each packet or file).
> I don't expect it to be used much directly, however if there are large
> buffers that are needed only briefly, freeing them immediately
> afterwards could do good things for Wireshark's memory usage. If an
> exception is thrown and the free call is never hit, they will still be
> cleaned up after the packet or file is closed as usual.
>
> Also note that wmem supports realloc as well, so things like emem's
> strbuf now have a much simpler and more efficient way to grow (or
> shrink) when needed.

This thread:

https://www.wireshark.org/lists/wireshark-dev/201208/msg00179.html

shows a good use case for it.  (AFAIK the SCTP dissector is still using 
double memory when doing reassembly.)

___________________________________________________________________________
Sent via:    Wireshark-dev mailing list <wireshark-dev () wireshark org>
Archives:    http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
            mailto:wireshark-dev-request () wireshark org?subject=unsubscribe