[Pan-users] Re: freeze, high CPU getting new headers

[walt]

On Mon, 04 Aug 2008 18:32:34 +0000, Matthijs Benschop wrote:

> Op Mon, 04 Aug 2008 13:07:34 +0000, schreef walt:
> 
> 
>> Have you tried strace during the long pause?  If you start pan in gdb
>> and hit Ctrl-C during the busy time you can type 'bt' to see what part
>> of the code is running at that point.  You need debugging symbols for
>> that, though.  Charles uses the -g flag by default to build pan, but if
>> the offending code is in some other library you won't get as much info,
>> but at least you can see what library the offending code belongs to.
> 
> Done that, get ready for a long paste..
> 
> So it seems to stall at mainloop(); ..
> 
> Btw, I always get segfault when closing 'post article' window and when
> expanding the group list in the left pane...

I see from your earlier post that you're using a 64-bit machine, so that
obviously could make a big difference, dunno.  Duncan compiles his own
on a 64-bit machine, but he also builds the entire system that way.

<snippage> 
>     address@hidden) at ../../pan/data/data.h:442
> #18 0x00000000004efd89 in pan::DataImpl::MyTree::add_articles
> (this=0x2658000,
>     nodes_in=<value optimized out>) at my-tree.cc:534
> #19 0x00000000004f05a9 in pan::DataImpl::MyTree::apply_filter <====
> (this=0x2658000, address@hidden)
>     at my-tree.cc:235
> #20 0x00000000004f0d9a in pan::DataImpl::MyTree::add_articles
> (this=0x2658000, address@hidden)
>     at my-tree.cc:351...

Just occurred to me that I don't use any filters, so the 'apply_filter'
step in my case would be a null_op.  What about you (and David)?