Re: RFD: Limiting scope of ep_ memory

[Evan Huus <eapache () gmail com>]

On Mon, Oct 22, 2012 at 5:36 PM, Jakub Zawadzki
<darkjames-ws () darkjames pl> wrote:
> On Mon, Oct 22, 2012 at 12:43:49PM -0400, Evan Huus wrote:
> > Perhaps pinfo should have its own scope that is between ep and
> > se? This would be much easier given the changes suggested in [1].
> >
> > [1] https://www.wireshark.org/lists/wireshark-dev/201210/msg00178.html
>
> Evan, when do you plan to merge it?

Well, I've had absolutely zero feedback on it so far, so I'm not sure
if that's a good sign or a bad one...

I was planning on taking it slow, maybe merging it this weekend if
nobody raised any objections. I can merge it basically right now
though if you can use it - since it doesn't touch any existing code
it's not that disruptive. I just didn't want to merge it before the
general design had gotten a couple of acks on the list, since API
changes after it's in use are a pain.

> I think I need something simple: pi_*(packet_info *pinfo, ...) which just
> use glib/system allocator, and store all pointers in list (pinfo->gc_list).
> Later memory would be freed in epan_dissect_cleanup().
>
> But I can wait for wmem.

A simple glib+list allocator is already available in the wmem I linked
in the email, so I think it would be pretty trivial to have a
pinfo->pool, allocate everything in that pool and then just free the
pool in epan_dissect_cleanup().

Evan
___________________________________________________________________________
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