Richard Goedeken
e162e1ccf1
allow front-end to set the speed limiter on/off prior to starting a rom
2011-07-16 11:17:41 -07:00
Richard Goedeken
4286ee7b9b
updated FRONTEND_API_VERSION to 0x020001 for new m64p_command and m64p_core_param
2011-07-16 11:00:03 -07:00
Richard Goedeken
f310d37785
missing semicolon
2011-07-16 10:30:23 -07:00
Richard Goedeken
596aec45ab
added new core parameter: M64CORE_SPEED_LIMITER along with get/set functionality
2011-07-16 10:24:22 -07:00
Richard Goedeken
367b07176a
make the MainSpeedLimit variable a local instead of global, and add a modifier function
2011-07-16 09:54:03 -07:00
Richard Goedeken
47538f3bc6
added new M64CMD_CORE_STATE_SET core command, and main_speedset() function
2011-07-16 09:51:38 -07:00
Richard Goedeken
3ea632b500
Added speed limiter control variable in main.c
2011-07-16 07:42:16 -07:00
richard42
a2c2b2ef91
Win32 fixes
2011-07-11 23:05:14 -07:00
Sven Eckelmann
c50a3b5037
Remove extraneous parentheses around comparison which could hide GCC warnings
2011-07-10 22:43:45 +02:00
Sven Eckelmann
c792b2b39d
Remove comparison unsigned < 0
2011-07-10 22:24:19 +02:00
Sven Eckelmann
5efe246437
Don't require executable memory for cached interpreter
2011-07-10 11:34:10 +02:00
Sven Eckelmann
b1f9a9d154
Readd set_rounding in fpu instructions for interpreter modes
2011-07-10 11:06:50 +02:00
Sven Eckelmann
5fd974cac9
Use common fpu implementation for interpreter modes to fix missing round/trunc/ceil/floor
2011-07-10 10:25:27 +02:00
Sven Eckelmann
625ab60be8
Show error on failed malloc_exec in cached interpreter mode instead of segfaulting
2011-07-10 01:36:28 +02:00
Sven Eckelmann
b6ada95740
Show error on failed first malloc_exec instead of segfaulting
2011-07-09 23:45:13 +02:00
Sven Eckelmann
a1c8fc62b7
Allocate exec memory using mmap
...
mprotect has no defined behavior when it is used with memory regions not
allocated by mmap. Therefore, it is easier to use mmap directly with the
RWX flags instead of using mmap and then mprotect.
2011-07-09 23:02:35 +02:00
Sven Eckelmann
48d49d8135
Prevent data corruption with caller managed OSD text
...
OSD text like the volume information is update by the caller in main.c, but the
drawing function in osd.c maybe already deleted the corresponding data
structure. The workaround was to check if the pointer to data structure could
be found in the list. It could happen that another text was shown inbetween and
the libc reused the old memory region for another text message. The
osd_message_valid function would wrongly inform the caller that the OSD text
wasn't deleted and the following osd_update_message would modify the wrong
message.
We must forbid that the messages is deleted by both the caller and the osd
drawing mainloop. The mainloop restricts itself to removing it from the list
when the caller marked it as user_managed directly after it was allocated. The
caller has to remove it manually using osd_delete_message in that situation.
2011-07-09 22:08:58 +02:00
Sven Eckelmann
b0bc4ceb4d
Cleanup savestates_select_filename as suggested in #412
2011-07-09 20:08:11 +02:00
Richard Goedeken
58cb394345
fix warning in NO_ASM build
2011-07-07 19:26:17 -07:00
Sven Eckelmann
912cd431ff
Don't use _BIG_ENDIAN to check for endianness as it is always set on OpenBSD
2011-06-26 20:18:10 +02:00
Sven Eckelmann
141c5f2acc
Use rom.c as base definition place for isGoldeneyeRom
2011-06-26 19:42:10 +02:00
Sven Eckelmann
efaef036d2
Provide prototypes of exported functions for use inside the core
2011-06-26 01:04:53 +02:00
Sven Eckelmann
e20674dbca
Don't use plain 0 for pointers
2011-06-25 23:05:19 +02:00
Sven Eckelmann
56208dfc40
Remove carriage return from line ending
2011-06-25 23:03:55 +02:00
Sven Eckelmann
34ca3d7c49
Mark local only functions as static or fix their definition
2011-06-25 22:40:58 +02:00
Sven Eckelmann
5137e8e264
Don't define variable isGoldeneyeRom in each file
2011-06-25 21:47:50 +02:00
Sven Eckelmann
cf4a4027c2
Provide prototypes to assist plugin developers
2011-06-25 21:07:42 +02:00
Sven Eckelmann
52999b1e13
Remove unused variable tmp
2011-06-25 15:14:00 +02:00
Sven Eckelmann
8e68e71935
Prevent over-optimization of rjump related data
...
GCC tries to remove and/or rename local symbols when trying to optimize
the code. This is a common problem when using link time optimization or
other optimization with deep inspection of dataflow. Inline assembler
code has to be annotated when it tries to access data behind such
symbols to prevent over optimization and to correctly name these
symbols.
This also removes partially the requirement of having different
implementation of the rjump code for different combination of OS and
PIC/PIE targets.
2011-06-05 18:52:30 +02:00
Richard Goedeken
e3c987f3ec
updated core for new Mupen64plus 2.0 API versioning scheme
2011-05-16 21:47:02 -07:00
Richard Goedeken
a60d4a6cd8
bugfix issue 423: we should call video extension override _before_ checking for SDL state in VidExt_GL_GetProcAddress function
2011-04-13 10:30:09 -07:00
Richard Goedeken
f2d2a76cfd
bugfix: we need to call the Video Extension SwapBuffers function during pause, not the SDL one
2011-04-13 08:45:43 -07:00
Richard Goedeken
1d538f41e9
added documentation for new GLGetAttr function in the video extension. fixed the CoreOverrideVidext function
2011-04-13 08:21:41 -07:00
Richard Goedeken
d3b9892037
1. added new VidExt_GL_GetAttribute function. 2. Added verbose message when inserting time delay due to frame rate limiter
2011-04-09 21:17:40 -07:00
Sven Eckelmann
d254233f05
Port n64 internal clock to Windows/VC++
2011-04-05 09:48:21 +02:00
Sven Eckelmann
26d19e48fc
Fix swapping of n64 images
...
---
src/main/rom.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
2011-03-12 23:24:50 +01:00
Jason Benaim
47b3f89d2d
Add support for n64 internal clock
...
---
src/memory/pif.c | 45 +++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 45 insertions(+), 0 deletions(-)
2011-03-12 23:24:46 +01:00
richard42
ca0a434a41
changed warning to info message when core can not open config file for reading and must use defaults
2011-02-06 23:37:06 -05:00
Richard Goedeken
8656be8a96
bugfix: issue #410 - segfault in dma_pi_write()-->strlen() if /home/username/.local/share/mupen64plus/ owned by root
2011-01-15 17:11:38 -05:00
Richard Goedeken
c2d0d094e2
moved callback function pointer typedefs from callbacks.h into m64p_frontend.h, so that they may be used by front-end apps
2010-12-25 21:06:19 -05:00
Richard Goedeken
6e53b65a67
updated version and RELEASE info for 1.99.4 tag
2010-11-22 21:09:31 -05:00
Richard Goedeken
07a57d8e88
Added 2 missing error descriptions for CoreErrorMessage function
2010-10-21 13:54:36 -04:00
Richard Goedeken
d4df8a20cf
added some permissive type checking to ConfigGetParameter() function, and a new error type
2010-10-01 21:57:32 -04:00
Richard Goedeken
25383b54d9
verify MaxSize parameter in ConfigGetParameter()
2010-10-01 19:42:21 -04:00
Richard Goedeken
d0fd89cb07
avoid segfault in the video extension code if SDL initialization fails (because video plugin fails)
2010-09-26 23:48:53 -04:00
Richard Goedeken
109c4f2ec9
Added new CoreGetRomSettings() function for front-ends
2010-07-31 15:58:00 -04:00
Sven Eckelmann
b16e55fff9
Add dummy PATH_MAX on systems without it
...
Posix only defines that PATH_MAX may is defined through limits.h.
GNU/Hurd has removed this definition without violating the standards.
Mupen64Plus still relies on it and we work around that problem by
defining PATH_MAX as 4096 in all places it is needed. This is not 100%
correct as their are ways to generate longer paths, but mupen64plus must
be heavily refactored in some places to fix the actual problem.
---
2010-07-13 21:31:12 +02:00
Richard Goedeken
69962d0c01
From Narann: Optimize Goldeneye hack in order to avoid the strncmp in virtual_to_physical_address() function.
2010-05-18 19:30:31 -04:00
Sven Eckelmann
065458fd0c
synchronize declaration and definition of dyna_start
2010-04-06 21:54:32 +02:00
Sven Eckelmann
5637eac24b
allow to run dynarec in hardware DEP protected windows
2010-04-06 21:45:56 +02:00
