Re: Reduce memory consumption by re-reading data from file for reassembled packets.

[Anders Broman <a.broman () telia com>]

Anders Broman skrev 2010-11-01 21:26:
> Jeff Morriss skrev 2010-11-01 20:21:
> > Anders Broman wrote:
> > > - Store the file pointer in the topmost TVBUFF_REAL_DATA changing the
> > > type to something like
> > >     TVBUFF_REAL_DATA_FROM_FILE to differentiating it from tvb:s not
> > > constructed from file data.
> > >
> > > - When making sub tvb they'd have the type TVBUFF_SUBSET_FROM_FILE
> > I think the top-level would also need to store the offset into the file.
> >     SUBSET and COMPOSITE TVBs would not have to change at all, except if
> > maybe to have a bit that says their backing is persistent.
> Sorry, I didn't mean the file handle but rather the file offset.  The
> SUBSET tvbs would then need
> their own file offset. I haven't looked at the COMPOSITE type - a problem?
> > > - The reassembly routines should then be changed to not store the
> > > fragments but rather the file offset
> > >     and length when all fragments are available and needs to be presented
> > > that data is read from file and
> > >     stuffed into a "reassembled data tvb".  Possibly data should be kept
> > > around until the final packet in a reassembly sequence on the first pass
> > > for speed.
> > I had been thinking it would build up composite TVBs using the passed-in
> > TVBs.  Since the TVBs would now be "persistent" this would be okay (we
> > recently discussed composite TVBs and why they're not used much
> > now--basically it's because the underlying TVBs aren't persistent).
> >
> > > Did I miss something? Feasible?
> > I don't think so and I think yes.
> >
> > (I came back from Sharkfest thinking I'd build a prototype of this, but
> > you can see how much work I've done on it! :-()
> Should we go ahead and check in changes toward this goal in trunk or
> open up a test-branch
> or just privately try to build a prototype, collaboration might be
> quicker. The needed changes to tvbuff.[ch]
> shouldn't be that extensive - I hope.
> Anders
Hi,
Here's the first set of changes(untested) if some one wants to have a 
look. Comments?
Regards
Anders

Attachment: file_offset_in_tvb.diff
Description:

___________________________________________________________________________
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