Richard Goedeken
7e41f2022a
updated LICENSES file, removed some unneeded text files
2009-12-12 23:17:24 -05:00
Richard Goedeken
60e7eec02c
removed unneeded release.mk file
2009-12-12 08:15:33 -05:00
Richard Goedeken
01f34f7af5
update version number to 1.99.1 for beta 1 release
2009-12-12 01:17:33 -05:00
Richard Goedeken
4bd48d31d5
fixed install/uninstall options in makefile, removed broken rebuild option
2009-12-12 01:15:57 -05:00
Richard Goedeken
c32403c0cb
rename ui-console binary from mupen64plus-cli to just mupen64plus
2009-12-12 00:32:16 -05:00
Richard Goedeken
5253052870
cleanup and refactor makefile, remove pre.mk and configure script
2009-12-07 22:45:32 -05:00
Richard Goedeken
b5650c716c
update m64p_build.sh script in tarball to handle versioned core library, and copy the cheat ini file
2009-12-05 22:25:37 -05:00
Richard Goedeken
876b694b00
removed some unnecessary data files
2009-12-05 22:08:19 -05:00
Richard Goedeken
7172a026e8
updated sample regtest config file for versioned core library
2009-12-05 21:52:30 -05:00
Richard Goedeken
d8a8251eab
generate versioned shared library for Linux and Freebsd
2009-12-05 20:08:57 -05:00
Richard Goedeken
ed92f7c960
bugfix: is_valid_rom() always failed after input parameter was changed from unsigned to signed
2009-12-02 21:34:30 -05:00
Richard Goedeken
39f67809a6
added missing includes in main/lirc.c
2009-12-02 21:28:10 -05:00
Richard Goedeken
53b9e93b7f
added -Wall to CFLAGS and cleaned up the warnings for 64-bit normal and debugger builds
2009-12-01 22:46:47 -05:00
Richard Goedeken
5629f353b1
change core compare API so that callback function receives current r4300 opcode. This only works for pure interpreter and dynarec cores
2009-12-01 19:45:58 -05:00
Richard Goedeken
b6c03547c3
missed a return code from last commit
2009-11-30 23:09:30 -05:00
Richard Goedeken
a928f13da7
removed the old core comparison code and replaced with function calls through the new API. The front-end application will handle the communication between the 2 running cores
2009-11-30 22:01:43 -05:00
Richard Goedeken
ba5a20d0f4
implemented final debugger api functions
2009-11-30 16:19:53 -05:00
Richard Goedeken
06f5e8d0d2
implemented Memory debugger functions for 2.0 api
2009-11-29 23:12:19 -05:00
Richard Goedeken
5722075ec6
implemented General debugger functions for 2.0 api
2009-11-28 22:58:06 -05:00
Richard Goedeken
bc95d9e79a
clean-up and initial code for Debugger support through new API. DebugSetCallbacks() implemented
2009-11-28 14:06:14 -05:00
Richard Goedeken
93302a481a
bugfix in Makefile for DEBUGGER=1 build, updated help message
2009-11-28 14:01:50 -05:00
Richard Goedeken
309a8bde50
bugfix: frame advance feature should advance every frame, instead of every vertical interrupt (every field)
2009-11-27 17:38:56 -05:00
Richard Goedeken
e0bf18e234
added all of the core State Change callback functionality. api/callbacks.c is now feature complete
2009-11-27 16:43:40 -05:00
Richard Goedeken
64b6552e7c
bugfix: missed a couple of translation function calls in last commit
2009-11-27 15:53:52 -05:00
Richard Goedeken
89d19906ef
since the only language file that we ever had for the core was English, it seems that there's not that much of a demand for translation support in the core. So all of the translation code (and the empty language file) have been removed
2009-11-27 14:48:27 -05:00
Richard Goedeken
420fe90d52
completely re-wrote the cheat code handler: greatly simplified cheat.h/cheat.c to remove unneded functions including all file i/o, wrote cheat code handler according to new mupen64plus 2.0 api. Updated api/frontend.c to call new cheat handler. api/frontend.c is now feature-complete
2009-11-27 13:55:29 -05:00
Richard Goedeken
042b46a519
cleaned up cheat ini file a bit, mostly replacing references in the comments to the F9 key with GS for GameShark button.
2009-11-27 13:48:24 -05:00
Richard Goedeken
48caacd092
modified pre.mk for OSX compatibility patch from Auria and refactored for simplicity
2009-11-23 20:09:08 -05:00
Richard Goedeken
99d888b212
fixed m64p_update.sh script in tools/m64p_helper_scripts.tar.gz
2009-11-22 19:13:02 -05:00
Richard Goedeken
377e8238d5
added and refactored code to finish all cases of CoreDoCommand() in api/frontend.c
2009-11-22 17:09:35 -05:00
Richard Goedeken
8df96712f7
replaced all printf() statements in core with DebugMessage() calls, except those in compare_core.c which is going to be removed in the future
2009-11-21 20:47:52 -05:00
Richard Goedeken
e025556f4a
finally: removed ugly winlnxdefs.h!
2009-11-21 14:12:16 -05:00
Richard Goedeken
42c4c479bc
bugfix to allow diagonal hat movements for core joystick commands
2009-11-21 11:26:09 -05:00
Richard Goedeken
28b3aefdc9
modified SDL event loop joystick code so that gameshark button press is captured, and joystick commands that are level-triggered instead of edge-triggered (such as fast foward) can be accommodated
2009-11-21 10:13:09 -05:00
Richard Goedeken
5268db9437
fixed warnings in the 64-bit debug info build
2009-11-21 00:12:01 -05:00
Richard Goedeken
b69c66c558
fixed the outstanding SDL event issues by re-writing the code which handles the joystick-event-driven core commands. Now the axis-based commands use hysteresis and there is a single global event function for determining if the gameshark button is pressed
2009-11-20 23:54:38 -05:00
Richard Goedeken
8a74ba68a0
fixed technical bug in fread/fwrite calls in compare_core(): the size comes before the quantity
2009-11-19 23:53:08 -05:00
Richard Goedeken
05ff71f4c7
fixed all of the gcc warnings from upgraded compiler (4.3.4) in the 64-bit standard build
2009-11-19 23:34:38 -05:00
Richard Goedeken
c821c5cb53
Moved all of the SDL event-related stuff into a new source file eventloop.c. There are still 2 things to fix: the joystick axis-driven core commands are hosed without hysteresis, and the key_pressed and event_active functions should be removed. They are only used by gs_button_pressed in cheat.c and should be replaced with a better mechanism
2009-11-18 23:15:11 -05:00
Richard Goedeken
8852fb3e29
merged all of the patches from last night into local copy with 64-bit dynarec fixes
2009-11-15 01:13:21 -05:00
Richard Goedeken
6c78c3f2fb
bugfix in dyna_start(): we need to load address of reg[] into r15, not its value
2009-11-15 00:58:38 -05:00
Richard Goedeken
7ed11ec79e
modified all of the dynarec's generator functions to use the revised assembler functions for r15-relative addressing
2009-11-15 00:48:30 -05:00
Richard Goedeken
4d081ae72d
Modified all instructions using m8abs operands in the assembler
2009-11-14 22:17:39 -05:00
Richard Goedeken
50bcee4503
Modified all instructions using m16abs operands in the assembler
2009-11-14 21:52:21 -05:00
Richard Goedeken
cb8ad59515
Modified all instructions using m32abs operands in the assembler
2009-11-14 21:29:53 -05:00
Richard Goedeken
7ce4ed70f6
start working on 64-bit Dynarec. Added r15 register load to dyna_start() for the base of relative memory accesses. Modified all instructions using m64abs operands in the assembler
2009-11-14 19:30:33 -05:00
Sven Eckelmann
421c5b059c
Fix OSD crash after pause-stop-start-pause of emulator
...
The osd message handler for pause messages are managed outside the
message queue of the osd and will only be updated during
pauseContinueEmulation, but the message is bound to an OpenGL context
and thus must be also be deleted when the emulator is stopped and the
video plugin shuts down.
2009-11-14 01:01:38 -05:00
Sven Eckelmann
4923bcf6ad
Don't override FGR with 0xffffffff after mtc0
...
When somebody switches between 64-bit wide FGR to 32-bit FGR it will
rearrange all registers in memory. This is partly done by shifting the
different parts of a new register pair for a double together. It is
important that this is done with unsigned integers or otherwise sign
extension is done and the upper part of the register will always be
0xffffffff when the bit 31 of the lower part of the register is 1.
2009-11-14 00:58:37 -05:00
Sven Eckelmann
27b4796613
Don't shuffle FGR loaded from Project64
...
I couldn't receive any comments from zilmar about the way the FGR
register is stored inside the savegame of pj64. So I wrote some
savegames to help me to reverse engineer it. The result was that we do
it nearly perfect, but the PJ64 savegame doesn't shuffle the data back
to 64 bit wide FGR before saving when it is currently in 32 bit wide
FGR. So we shouldn't do it when saving to or loading from PJ64 savegame.
2009-11-14 00:56:43 -05:00
Sven Eckelmann
814fa8df74
Set video width and status during savestate load
...
Not setting status and width for video plugins could result in
corrupted output in some programs.
2009-11-14 00:54:22 -05:00
