pan-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Pan-users] [ANNOUNCE] Pan release 0.150


From: Duncan
Subject: Re: [Pan-users] [ANNOUNCE] Pan release 0.150
Date: Fri, 4 Mar 2022 06:51:11 -0000 (UTC)
User-agent: Pan/0.150 (Moucherotte; 4c6043e9c)

Duncan posted on Wed, 2 Mar 2022 10:31:00 -0000 (UTC) as excerpted:

> Working here (gentoo, building from git), but still crashy.  In
> particular I've found pan crashes quite reliably if I hover the mouse
> over the body pane, so apparently some pointer event triggering it. 
> I've not debugged beyond that.  If I steer clear of the body pane
> (literally, maneuver the mouse around it so as to avoid triggering that
> crash), pan's reasonably stable.

I thought to test a variant of that tonight.  Does a crash get triggered 
if I enter a group with now messages shown (view only unread, no unread to 
show) and play the mouse around the blank body pane?

Yes, it does, altho it /might/ take a bit longer and /might/ only trigger 
if I first click in the blank body pane, then cross between it and the 
groups pane.  (And no, hovering the divider between the two or grabbing 
and moving it to change the relative sizes of the two doesn't appear to 
trigger a crash.)

I tried several times, of course restarting pan afterward each time to try 
again and finally to stop and post this reply.  Just clicking, hovering or 
moving around the blank body pane didn't /appear/ /to/ trigger a crash, 
unless there was some delay between the trigger and the crash, but shortly 
after (not immediately) I clicked in the blank body pane, then moved over 
the groups pane, it would *sometimes* crash.  Not always, but consistently 
enough over several tries to reliably crash pan with say 20 seconds of 
repeating.

So it doesn't seem to /require/ having a post showing, tho I have the 
feeling that it's easier to trigger with a post showing, perhaps 
triggering with a post showing without switching panes as seems to be 
required for the blank body pane case.

Since you (DD) have confirmed *not* seeing it there, I suppose it could be 
a miss-compilation too, or over-optimization caused by my cxxflags.  I 
should try a generic-as-possible plain -O2, no --march=native --ftree-
vectorize, etc, and see if that eliminates the problem...

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman




reply via email to

[Prev in Thread] Current Thread [Next in Thread]