Analogously treat maskwallscan() and transmaskwallscan(), although I could
not get the respective accesses to be oob, too.
git-svn-id: https://svn.eduke32.com/eduke32@4757 1a8010ca-5511-0410-912e-c29ae57300e0
Notes:
- In the 'clean' target, the "core*" arg to 'rm' was removed as I find it too
unspecific. (Someone could have an unrelated file with that prefix lying
around.)
- The targets 'clean' and 'veryclean' now echo the commands they are about
to execute.
Also, in Makefile.common, resurrect sanitized Clang builds if DEBUGANYWAY=2
is passed (previously, that was triggered with RELEASE=0 DEBUGANYWAY=1).
git-svn-id: https://svn.eduke32.com/eduke32@4756 1a8010ca-5511-0410-912e-c29ae57300e0
- for the first point: 'struct glattribs sdlayer_gl_attributes[]' was defined
at file scope, but 'glmultisample > 0' is not a compile-time constant.
- for the second point: r4749 changed the meaning of some conditional
compilation (supposedly, unintentionally). I'm reverting it to pre-r4749
but keeping the rewritten style. It's concerned with mouse thumb buttons.
git-svn-id: https://svn.eduke32.com/eduke32@4755 1a8010ca-5511-0410-912e-c29ae57300e0
VM_OnEvent() has become VM_OnEvent(), VM_OnEventWithReturn(), VM_OnEventWithDist(), and VM_OnEventWithBoth() (the latter of which is only ever used once...). Of course, this required every call to VM_OnEvent() be changed.
memberlabel_t and vmstate_t have been changed to use the regular "int" type versus explicitly specifying int32_t as they did previously. The rationale for this change is simply that it looks cleaner, and I think we should move toward just using "int" in most cases where there's no particular reason to specify an explicit data type.
Also changes CON_KILLIT to just "return" instead of "continue". DONT_BUILD.
git-svn-id: https://svn.eduke32.com/eduke32@4745 1a8010ca-5511-0410-912e-c29ae57300e0
Otherwise, the out-of-memory handler is not called in case of failure.
git-svn-id: https://svn.eduke32.com/eduke32@4736 1a8010ca-5511-0410-912e-c29ae57300e0
We must not use Xrealloc(), since gamearrays are now allocated with
Xaligned_alloc().
NOTE: I did not check all potential problematic cases, only those to get
LNGA3 up and running.
git-svn-id: https://svn.eduke32.com/eduke32@4735 1a8010ca-5511-0410-912e-c29ae57300e0
Optimizations are left to the reader :P. I would prefer to look for ways of
giving the compilers hints about loops to unroll instead of having to read
the resulting mess of manual unrolling, though.
git-svn-id: https://svn.eduke32.com/eduke32@4728 1a8010ca-5511-0410-912e-c29ae57300e0
The practical rationale: Clang-sanitize catches this, so this is bad.
The real rationale: I *think* it is undefined behavior to even form such a
pointer in C99. However, I would be hard pressed to provide a nice formal
argument in terms of the Standard wording right now. It looks like
6.5.3.2#4 is to blame.
git-svn-id: https://svn.eduke32.com/eduke32@4727 1a8010ca-5511-0410-912e-c29ae57300e0
On Linux, it started manifesting itself since r4719 as crashes whose
backtraces pointed to malloc() and were not helpful. Valgrind finally
uncovered this.
git-svn-id: https://svn.eduke32.com/eduke32@4724 1a8010ca-5511-0410-912e-c29ae57300e0
- Needed to rip out enum GameEvent_t into own header file to prevent what
would be a circular #include otherwise
- Added some forgotten #ifdef LUNATIC
- updated structure and function declarations on the Lua FFI side
git-svn-id: https://svn.eduke32.com/eduke32@4710 1a8010ca-5511-0410-912e-c29ae57300e0
For example, in
CON_ERRPRINTF("invalid target sprite (%d) %d %d\n", iActor, vm.g_i, TrackerCast(vm.g_sp->picnum));
vm.g_i can be -1 and vm.g_sp can be NULL then. (Not anymore.)
DONT_BUILD.
git-svn-id: https://svn.eduke32.com/eduke32@4709 1a8010ca-5511-0410-912e-c29ae57300e0
