that case, we do need add it when creating the window otherwise we
result int a nice BadMatch for X_CreateWindow request. FIXME?
- Fix typo and do enable VOSF in DGA mode, if necessary.
POSIX feature from _POSIX_THREAD_PROCESS_SHARED group. Besides, Ulrich Drepper
mentions "the kernel support for this option wasn't available until 2.5.7.
Future versions of the thread library will support this option."
<http://people.redhat.com/drepper/posix-option-groups.html>
BTW, this nukes a warning when build on FreeBSD 4.6
to programs by using a '|' followed by a command line as the modem or
printer port setting (instead of a device name like '/dev/ttyS0')
[Brian Johnson]
- the option "--config FILE" tells B2 to use a different config file
configure script would hang whereas standalone testing will pass all tests.
Any idea why??
- Unix/sigsegv.cpp (ix86_skip_instruction): Add decoder for mozbl instruction.
(SIGSEGV_REGISTER_FILE [FreeBSD/x86]): Note why we start at EDI offset
rather than plain sigcontext pointer. i.e. I don't know sigset_t size
beforehand and don't intend to.
- sigsegv.cpp (ix86_skip_instruction): Add decoder for movzwl instructions.
(main): oddly, FreeBSD doesn't seem to let a write to a page if it is
write-only. Aka. make the page readable too.
converted for the new scheme; not actually tested with a mult-monitor
setup yet but at least single-monitor display doesn't seem to be broken
(UAE banked addressing would definitely require some extensions to handle
multiple frame buffers)
- struct video_mode has an extra field that is free for use by platform-
specific code
type (first parameter) is caddr_t instead of void *. Explicitly cast
address to (caddr_t) type and C++ implicit pointer conversion rules
will do the rest. aka. caddr_t -> void * is OK unlike the opposite.
detect whether the compiler supports -fno-merge-constants or not
- Makefile.in: make sure to link gencpu with g++ since it contains objects
compiled from C++ code. Fix build problems with gcc 3.0.1
it breaks B2 in fpu/fpu_x86.cpp on the second call to fpu_init(), at the
fsave instruction. This problem appeared with the following compilers:
- "Official" : gcc-3.1 (mainline CVS)
- Mandrake : gcc-2.96 >= 0.59mdk, gcc-3.0.1 >= 0.1mdk
- Red Hat : gcc-2.96 >= 89, gcc-3.0 >= 1
possible to, for example, use MacOS-partitioned hard disks and removable
media under B2/Unix even if the OS doesn't understand Mac partition maps
by specifying the appropriate block device name as a Mac volume
- fixed typo in audio_dummy.cpp
- added minimally required UDP tunneling code to ether_dummy.cpp
- main_unix.cpp: if pthreads are not supported, we trigger the Ethernet
interrupt in the 60Hz ticker; this makes UDP tunneling work under
NetBSD/m68k (as the only form of networking)
that it can be vm_protect()'ed at will afterwards
- video_x.cpp: let driver_base free() the_buffer_copy and the_host_buffer,
if necessary
- video_vosf.h: reverted to use malloc() for mainBuffer.dirtyPages and
mainBuffer.pageInfo
- vm_alloc.cpp: small cleanups
- fixed typos
- fixed possible crash in driver_window dtor (image data would be freed twice,
once by XDestroyImage() and once in driver_base dtor)
- fixed compilation problems with banked memory
- fixed typos
to "true". This uses the BSD socket API, so it's fairly portable (currently
only imeplemented under Unix, though). This works by sending raw Ethernet
packets as UDP packets to a fixed port number ("udpport", default is 6066),
using IP broadcasts to simulate Ethernet broad- and multicasts. Currently
only tested with AppleTalk.
- ADBMouseMoved(), ADBMouseDown/Up() and ADBKeyDown/Up() trigger the ADB
interrupt
- ADB mutex is only used for mouse movement (the only input state where it
matters)
- adb.cpp: toggling relative mouse mode resets mouse_x/y
- PrimeTime(0) schedules a timer task with 0 delay time; this is still not
the correct implementation, but it makes MacSyndicate work...
- Unix: pthreads are preferred to POSIX.4 timers for 60Hz ticks because the
timers drift badly under Linux and the thread can compensate for drifting
well enough
- Unix: moved GetTicks_usec() and Delay_usec() to timer_unix.cpp
- video_x.cpp: X mouse acceleration is disabled in relative mouse mode because
MacOS does its own acceleration
- video_x.cpp: palette[].pixel and palette[].flags are always preset
- video_x.cpp: decoupled X event handling from 60Hz video refresh cycle by
using select() with a timeout on the X fd
of the internal structures used for the VOSF system
- use vm_acquire()/vm_release() for VOSF buffers and hope the_buffer is
allocated above RAM address space (temporary workaround for 64-bit
addressing systems)
- don't free() screen buffers in driver_base dtor
- don't free() memory mapped buffers in driver_base dtor
order. No host memory region used for Mac emulation (ScratchMem, RAM, ROM,
frame buffer) shall be allocated below the RAM space. Actually, MEMBaseDiff
should be set to the min(above-mentioned address spaces).
==> Temporary fix for 64-bit addressing systems (e.g. Linux/ia64)
of MacOS before 7.6:
- Apple mode IDs are now allocated contiguously from 0x80 (the video_*.cpp
module must call video_init_depth_list() after adding all modes)
- if the video driver didn't receive a GetVideoParameters call, it patches
ScrnBase and the main GDevice upon a video mode switch (otherwise MacOS
will continue to use the old frame buffer base)
- the rowBytes values in the video parameters slot resources are correct
for all bit depths
implemented under Unix
- adb.cpp uses mutexes for thread-safe mouse handling
- video_x.cpp: pressing Ctrl-F5 in windowed mode switches to a "grabbed"
relative mouse mode, useful for some games
- video_x.cpp: fixed some bugs relating to the hotkeys (key releases are no
longer treated as hotkeys)
linked list
- color depth switching updates slot ROM
- video_x.cpp always supports 1-bit window modes
- timer_create()/clock_gettime() are pulled from librt if present
However, vm_init() and vm_exit() are called in main_unix.cpp to ensure proper
initialization of the internal zero_fd descriptor, if needed. i.e. no
anonymous mapping for mmap()-based memory allocation.
resolutions are 512x384, 640x480, 800x600, 1024x768 and 1280x1024 (the prefs
editor has to be updated to reflect this). The resolution selected in the
prefs editor is used as the default, but it can be changed in the Monitors
control panel. So far only tested with direct addressing.
contains all supported depths, default mode is stored in XPRAM upon startup,
and added video_switch_to_mode() call (currently unimplemented in all drivers)
- added uniform virtual memory allocation
(supports mmap(), vm_allocate(), or fallbacks to malloc()/free())
- cleaned up memory allocation in main_unix.cpp
- ether_linux.cpp moved and renamed to ether_unix.cpp, now also works with
the tap driver under FreeBSD [Michael Alyn Miller]
- fpu_x86_asm.h: fixed problem in with newer GCC pre-processors
- now compiles under Linux 2.4 kernels
- Ethernet hardware address is virtualized on the Mac side
- automatically filters IP packets based on the IP address of the Mac side
- Cleaned up the process for determining the ranges of pages touched
that have to be blitted onto the screen (find_next_page_set() and
find_next_page_clear() functions)
Changes made to do_handle_screen_fault()
- An unhandled address is now dealt by the default SIGSEGV handler
- Print out the address of the instruction that caused the exception
that have to be blitted onto the screen (find_next_page_set() and
find_next_page_clear() functions)
- Cleaned up some comments
Changes from Brian J. Johnson
- Fixed mainBuffer.dirtyPages[] array overrun in VOSF code
- Fixed calculation of the frames-per-second value
but doesn't actually close the window to prevent Basilisk II from being
accidentally killed without a proper shutdown of MacOS
- replaced many #ifdef HAVE_PTHREADS constructs with LOCK/UNLOCK macros
- minor cleanups
