[Gtkradiant] [Bug 53] mouseclick + drag in 2d view causes stack overflow
Mon, 14 May 2001 15:56:40 -0500
What |Old Value |New Value
------- Additional Comments From firstname.lastname@example.org 2001-05-14 15:56 -------
I was able to reproduce the problem. But I can't isolate the crash criterion.
We get a stack overflow message, but no sign of recursion in any of the
functions involved around the crash.
1) There is no telling that this bug is related to 2D view slowdowns. Afaik
those slowdowns are not due to running out of virtual memory.
2) This crash doesn't occur in normal use. Unless you are a bad parkinson case
you are not going to shake a brush for 1 minute during a drag.
might be related to 2D view slowdowns, reported as a seperate bug:
closing as wontfix until further notice
[TTimo] ok ok
[TTimo] back to the stack overflow bug
[TTimo] you there Spog?
[TTimo] 1) does the stack crash happen randomly when you edit a brush, that is
when you interact normally with it without dragging it like a mad cow
[TTimo] 2) what do you mean by 2D view slowdown. Is it a general slowdown
because of system load (with swaping and memory usage) or does it happen after
a while using the editor
[TTimo] I never experienced the slowdowns, if they happen when running out of
mem, I think it's more than 15 mins usage, I got to use it for 2-3 hours to
*** W2k (~email@example.com) has joined #gtkradiant
<SmallPileofGibs> i've never had the stack crash during normal editing,
but i've had the slowdown
<SmallPileofGibs> the slowdown is a big lag between clicking+dragging and
the 2d view being updated
*** W2k is now known as W2k[tribes2]
[TTimo] and is it tied to swapping and OS running low on system resources?
<SmallPileofGibs> i've had 2 people complain if slowdown, using builds
<SmallPileofGibs> i don't know
<SmallPileofGibs> but i suspect it is
[TTimo] then I'll assume it's not related
<SmallPileofGibs> i would
[TTimo] and still those two problems are not related
<SmallPileofGibs> assume they're not related
<SmallPileofGibs> it's just my suspicion =)
*** neo279 is now known as neo[cs]
[TTimo] does it happen systematically after some time of usage for you?
<SmallPileofGibs> i haven't used it enough constantly to say
[TTimo] what annoys me is that the memory usage as shown by win2k is raising,
and yet BC doesn't see any leaking
[TTimo] which makes me think it can be just the OS screwing up, showing cached
[TTimo] not keeping an actual track of the heap allocations
<SmallPileofGibs> that would be strange