yanking the mouse around if they keys haven't been read yet to combat the
same situation that causes the keyboard to return DIERR_NOTACQUIRED in
KeyRead(): The window is sort of in focus and sort of not. User.dll
considers it to be focused and it's drawn as such, but another focused
window is on top of it, and DirectInput doesn't see it as focused.
- Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it
handles DIERR_INPUTLOST errors. This can happen if our window had the
focus stolen away from it before we tried to acquire the keyboard in
DI_Init2(). Strangely, MouseRead_DI() already did this.
- When a stack overflow occurs, report.txt now only includes the first and
last 16KB of the stack to make it more manageable.
- Limited StreamEditBinary() to the first 64KB of the file to keep it from
taking too long on large dumps.
- And now I know why gathering crash information in the same process that
crashed can be bad: Stack overflows. You get one spare page to play with
when the stack overflows. MiniDumpWriteDump() needs more than that and
causes an access violation when it runs out of leftover stack, silently
terminating the application. Windows XP x64 offers SetThreadStackGuarantee()
to increase this, but that isn't available on anything older, including
32-bit XP. To get around this, a new thread is created to write the mini
dump when the stack overflows.
- Changed A_Burnination() to be closer to Strife's.
- Fixed: When playing back demos, DoAddBot() can be called without an
associated call to SpawnBot(). So if the bot can't spawn, botnum can
go negative, which will cause problems later in DCajunMaster::Main()
when it sees that wanted_botnum (0) is higher than botnum (-1).
- Fixed: Stopping demo recording in multiplayer games should not abruptly
drop the recorder out of the game without notifying the other players.
In fact, there's no reason why it should drop them out of multiplayer at
all.
- Fixed: Earthquakes were unreliable in multiplayer games because
P_PredictPlayer() did not preserve the player's xviewshift.
- Fixed: PlayerIsGone() needs to stop any scripts that belong to the player
who left, in addition to executing disconnect scripts.
- Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo
in case the player left but somebody still has a reference to their actor.
- Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior
and set Buffer to NULL.
- Improved feedback for network game initialization with the console ticker.
- Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source
directory. They are identical, so keeping two copies of them is bad.
- Fixed: (At least with Creative's driver's,) EAX settings are global and not
per-application. So if you play a multiplayer ZDoom game on one computer
(or even another EAX-using application), ZDoom needs to restore the
environment when it regains focus.
- Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently,
PacketGet can receive ECONNRESET from nodes that aren't in the game. It
should be safe to just ignore these packets.
- Fixed: PlayerIsGone() should set the gone player's camera to NULL in case
the player who left was player 0. This is because if a remaining player
receives a "recoverable" error, they will become player 0. Once that happens,
they game will try to update sounds through their camera and crash in
FMODSoundRenderer::UpdateListener() because the zones array is now NULL.
G_NewInit() should also clear all the player structures.
SVN r233 (trunk)
contained unused sidedefs.
- Fixed: Color control sequences were written to the log file. Since any
entered console command contains such a sequence it was quite noticable.
SVN r208 (trunk)
memcpy to copy the player structures.
- Fixed compilation with MinGW again and removed most of the new warnings.
And following is the log that I forgot to paste in for the previous commit:
- Changed the memory management for FString. Instead of using a garbage
collected heap, it now uses normal heap calls and reference counting to
implement lazy copying. You may now use bitwise operators to move
(but not copy!) FStrings around in memory. This means that the
CopyForTArray template function is gone, since TArrays can now freely
move their contents around without bothering with their specifics.
There is one important caveat, however. It is not acceptable to blindly 0
an FString's contents. This necessitated the creation of a proper
constructor for player_s so that it can be reset without using memset. I
did a quick scan of all memsets in the source and didn't see anything else
with a similar problem, but it's possible I missed something.
- Fixed: Build tiles were never deallocated.
- Fixed: Using Build's palette.dat only got half the palette right.
SVN r117 (trunk)
functions added to the exit chain with atterm so that they can be called
in a deterministic order and not whatever order the linker decides to put
them in.
- Fixed: DCajunMaster did not free its getspawned.
- Fixed: P_FreeLevelData() did not free ACS scripts.
- Fixed: Level snapshots were not freed at exit.
- Fixed: The save/load menu list was not freed at exit.
- Fixed: FCompressedMemFile needs a destructor to free the m_ImplodedBuffer.
- Fixed: G_DoLoadGame() did not free the engine string.
- Fixed: M_ReadSaveStrings() did not free the engine string.
- Fixed: Processing DEM_SAVEGAME did not free the pathname string.
- Added a check for truncated flats to FFlatTexture::MakeTexture() because
Heretic's F_SKY1 is only four bytes long.
- Added a dump of the offending state to the "Cannot find state..." diagnostic.
- Fixed: FCompressedFile did not initialize m_Mode in its default constructor.
- Fixed: Heretic and Hexen status bars did not initialize ArtiRefresh.
- Fixed: PNGHandle destructor should use delete[] to free TextChunks.
SVN r111 (trunk)
merge a lot of these static destructor-only structs into regular
functions added to the exit chain with atterm so that they can be called
in a deterministic order and not whatever order the linker decides to put
them in. (Interestingly, the amount of memory used when repeatedly
executing the same map command at the console varies up and down, but it
now stays relatively stable rather than increasing unbounded.)
- Fixed: The list of resolutions in the video modes menu was not freed
at exit.
- Fixed: mus_playing.name was not freed at exit.
- Fixed: SN_StopAllSequences() should be called at the start of
P_FreeLevelData(), not just before the call to P_SetupLevel() in
G_DoLoadLevel(), so it can run even at exit. And C_FullConsole() can
call P_FreeLevelData() to free more memory too.
- Fixed: StatusBar was not freed at exit.
- Fixed: spritesorter was not freed at exit.
- Fixed: Bad things happened if FString's data pool was destroyed before
all C_RemoveTabCommand() calls were made.
- Added an overload for FArchive << FString.
- Fixed: The players' log text was not freed at exit.
- Fixed: Bot information was not freed at exit.
- Fixed: doomcom was not freed at exit. But since it's always created,
there's no reason why it needs to be allocated from the heap. My guess
is that in the DOS days, the external packet driver was responsible for
allocating doomcom and passed its location with the -net parameter.
- Fixed: FBlockNodes were not freed at exit.
- Fixed: Openings were not freed at exit.
- Fixed: Drawsegs were not freed at exit.
- Fixed: Vissprites were not freed at exit.
- Fixed: Console command history was not freed at exit.
- Fixed: Visplanes were not freed at exit.
- Fixed: Call P_FreeLevelData() at exit.
- Fixed: Channel, SoundCurve, and PlayList in s_sound.cpp were not freed at
exit.
- Fixed: Sound sequences were not freed at exit.
- Fixed: DSeqNode::Serialize() did not resize the m_SequenceChoices array
when loading.
SVN r106 (trunk)
- The CRT no longer detects any memory leaks when I run to the IWAD picker and quit.
- Fixed: The memory used to hold the path to zdoom.wad/.pk3 was not freed if
the IWAD picker was cancelled.
- Fixed: Some implementations of cvar->GetGenericRep (CVAR_String) returned a
statically allocated string and others returned a dynamically allocated string.
To be consistant, they should all be static.
- Fixed: DObject also has three static TArrays that should not be explicitly
initialized: Objects, FreeIndices, and ToDestroy.
- Added a new do-nothing constructor for TArray that can be used for BSS objects
that are manipulated before startup. Specifically, this was added because
C_AddTabCommand() is called before main, but the TabCommands constructor is
called after the array already has over 100 entries in it, orphaning everything
that was already inserted. And since the code is much nicer-looking now, I didn't
want to revert to the old non-TArray version.
This could also have been used to fix FName, but I consider the current
implementation to be better than the old one, so I'm leaving it as-is.
- Fixed: Scanned IWAD paths were not freed if you exited without selecting one.
- Fixed: Dynamically allocated cvars were not freed on exit.
- Fixed: FConfigFile's destructor did not free space used for Values.
SVN r82 (trunk)
actually work.
- Turned the list of TabCommands into a TArray because I saw lots of console
commands in the memory leak report at exit. Then I realized those were actually
key bindings, so I changed the Bindings and DoubleBindings arrays into FString
arrays.
- Fixed: FStringCVar was missing a destructor.
- Added TArray::Insert().
- Fixed: TArray::Delete() used memmove().
- Renamed Malloc(), Realloc(), and Calloc() to M_Malloc(), M_Realloc(), and
M_Calloc() so that the debug versions can be defined as macros.
- Enabled the CRT's memory leak detection in WinMain().
- Moved contents of PO_DeInit() into P_FreeLevelData().
- Removed "PolyBlockMap = NULL;" from P_SetupLevel(), because the P_FreeLevelData()
call it makes next does the exact same thing, but also freeing it if needed.
- Fixed: Unneeded memcpy in UnpackUserCmd() when ucmd and basis are the same
SVN r75 (trunk)
its tiles are loaded from the same directory.
- RFF files now load their entire directories into the lumplist.
- Added char * and const char * type coversions for FString, so FStrings can be
freely passed to functions expecting C strings. (Except varargs functions,
which still require manually fetching the C string out of it.)
- Renamed the name class to FName.
- Renamed the string class to FString to emphasize that it is not std::string.
SVN r74 (trunk)
- Fixed the issues with .96x's railgun code and added it to the current
version.
- Fixed: Setting of the friendly Minotaur's angle was inconsistent and
could cause it to move backwards in a féw situation.
- Fixed: The minotaur did checks for type by checking for the MF_FRIENDLY
flag, not by checking for the actor class. That made it impossible to
spawn friendly 'normal' minotaurs.
- Moved a few virtual methods which are only applicable to the friendly
minotaur to AMinotaurFriend.
- Fixed: The friendly minotaur checked the master's health instead of
the target's when deciding whether to switch targets. Replaced with
MF4_NOTARGETSWITCH.
- Fixed: maybedrawnow must not draw the console unless the background
texture has been set up.
SVN r66 (trunk)
STlib_drawNum(), and FDoomStatusBarTexture::DrawToBar() should add
the textures left offset to the x coordinate before drawing.
These fix Twice Risen's status bar.
- Changed: VPrintf now uses string.VFormat(), instead of vsprintf().
SVN r53 (trunk)
Randy Heit
Christoph Oelckers