Re: [lmi] wx segfault iterating on GetChildren()

[Greg Chicares]

On 2006-3-16 22:44 UTC, Evgeniy Tarassov wrote:
>
> Vadim has asked me to check if the tuple
> mpatrol+mingw(gcc3.2.3)+wxWidget2.6.2
> allows to compile lmi.
> It does work with almost no problems/workarounds.

That sounds like very welcome news.

> 1. when trying to get mpatrol to link even to a hello world i got
> 'symbol undefined error' for SymGetModuleInfo routines defined in
> GbdHelp.lid(dll), and i did'nt managed to get it work. But after
> defining DYNLINK to DYNLINK_NONE in mpatrol/src/target.h file and
> recompiling mpatrol it did it.

My solution is
  http://cvs.savannah.gnu.org/viewcvs/lmi/lmi/mpatrol-mingw-GNUmakefile
because I understand make better than automake. I wrote that
file three years ago and have forgotten how it differs from
the original distribution, but I have used it successfully
as recently as last year.

> 2. fatal_error() function does not seem to work as intended. Any time
> it is being called it produces exception 'basic_ios::clear(iostate)
> caused exception' which is shown through wxSafeShowMessage and initial
> diagnostic message is lost.

I remember seeing that problem. My recollection is vague, but
I think it might have been caused by a defect in libstdc++.
Let me look into it and get back to you.

> Notes:
> 1. when using autotools with gcc3.2.3 i've noticed that -Wendif-labels
> isn't a legal switch (for gcc 3.2.3) and should be added
> conditionally.

OK, thanks, I'll work on fixing that.

> I will try to understand what happens with fatal_error().

Probably I should try to save you that trouble, because I've
already researched that problem; I just don't remember what
I found, so I ought to document it in the source code after
I find the explanation in my old notes.

> As my experience with mpatrol is equal to zero i wanted to ask you if
> you know any 'polite' way of configuring mpatrol instead of hacking
> into its headers. Thank you in advance!

I simply altered the makefile.

Also see 'mpatrol_patch_test.cpp', which should give a link to
the patch it tests: I guess that's line 37 of revision 1.2 .
The patch blocks diagnostics for zero-byte moves with memmove()
IIRC, and I patched it because wx did zero-byte moves, at least
in version...wx-2.39, I think.

I really look forward to getting mpatrol working as part of
my routine builds. It will take some work, and probably I've
spoiled some things by not routinely building it. My fear is
that the fatal_error() problem might not be easy to fix in
MinGW gcc-3.2.3, and I know I added that 'alert.hpp' facility
without testing it with mpatrol. But let me look into that.