Wireshark-dev: Re: [Wireshark-dev] qtshark startup time ~1 minute just printing messages
Date Index · Thread Index · Other Months · All Mailing Lists
Date Prev · Date Next · Thread Prev · Thread Next
From: Evan Huus <eapache@xxxxxxxxx>
Date: Mon, 9 Sep 2013 21:08:59 -0400
It should be fixed in r51914 (see the commit message for more details). Joerg, please let me know if it now works for you (or if you are in fact seeing a different bug).

Evan


On Mon, Sep 9, 2013 at 8:58 PM, Evan Huus <eapache@xxxxxxxxx> wrote:
Got it, though I'm not sure the optimal fix - basically, the problem is that when a Paint takes more than info_update_freq_ milliseconds then every single event causes a Paint (because just the last Paint took enough time to trigger the timer). I will dig further


On Mon, Sep 9, 2013 at 8:47 PM, Evan Huus <eapache@xxxxxxxxx> wrote:
On Mon, Sep 9, 2013 at 6:48 PM, Gerald Combs <gerald@xxxxxxxxxxxxx> wrote:
On 9/9/13 3:18 PM, Joerg Mayer wrote:
> On Mon, Sep 09, 2013 at 05:38:27PM -0400, Evan Huus wrote:
>> This is actually quite reproducible (on my machine at least). If I maximize
>> the qtshark window immediately upon startup then the splash seems to do a
>> full-window redraw for every change, and takes a significantly
>> longer-than-usual amount of time to finish.
>
> Ah, that explains that last time I got annoyed by this it was more like
> 1,75 minutes. The window was larger.

What happens if you disable antialiasing in the overlay or set its
opacity to 1.0? I can't duplicate this (to any appreciable level, at
least) on any of my test machines.

No change - the splash is marginally faster in general, but it still slows down substantially when the window size is changed. I suspect it's some condition that is accidentally disabling the "Only update every splash_register_freq milliseconds" check, I will try and debug further.