Wireshark-dev: Re: [Wireshark-dev] r48218: Remove the emem slab feature

From: Jakub Zawadzki <darkjames-ws@xxxxxxxxxxxx>
Date: Sun, 10 Mar 2013 17:22:19 +0100
On Sun, Mar 10, 2013 at 09:19:25AM -0400, Evan Huus wrote:
> On Sun, Mar 10, 2013 at 8:46 AM, Jakub Zawadzki
> <darkjames-ws@xxxxxxxxxxxx> wrote:
> >
> > I cheated a little and both tshark are based on r47905, but with patches:
> >   ws_slab-tshark was build with patch: http://pastebin.com/raw.php?i=rMexZBsh
> >   g_slices-tshark was build with patch: http://pastebin.com/raw.php?i=YdA50LKL
> 
> Your g_slice patch uses g_slice_alloc0, which zeros the returned
> memory like g_malloc0 does. A fairer comparison would be with
> g_slice_alloc(), since the emem slab doesn't make any guarantees about
> zeroing memory.

Right, again g_slices, this time only 5:
             18m59.489s
             18m48.376s
             18m43.400s
             18m55.353s
             19m0.579s

       avg:  18m53.439s
              +/- 10s

time difference to ws_slab: 1m58s (~10%)

> I'm honestly not sure why g_slice_alloc0 was used when
> debug_use_slices was set, since presumably in a debugging case we'd
> want to blow up if possible on uninitialized memory.
> 
> > but I'd rather wait for some other results/conclusions before I repeat test with r48218 and r48217
> >
> > [cut]
> >
> > avg:            16m55.380s      19m29.426s
> >                  +/- 25s         +/- 20s
> >
> >                        1m17.023s
> >                      (~8% +/- 4%)

I'm not sure how I calculated this one (divided by two? :>), it should be 2m34.046s (~14%) :P