gbeauche
ea42af0568
Consume the packet even if we could not allocate a message block for it in
...
the Ethernet interrupt. The BeOS version does that, likewise for MOL. Otherwise,
we end up into an infinite loop reporting the memory allocation failure.
I think this is now the expected behavior as we wouldn't have stats
(num_rx_no_mem) for it if we couldn't get out of the EtherIRQ. ;-) Besides,
the packet will be resent for reliable networks.
2005-06-30 06:38:09 +00:00
gbeauche
7ac6ad2f46
Clean-ups: comments, ticks per sec output, don't restore r13 twice.
2005-06-28 16:50:30 +00:00
gbeauche
38d03d02e9
The alternate stack trick never worked as you can't modify an active stack.
...
i.e. it returned EPERM and ran into stack corruption to eventually crash the
emulator. This is noticeable in !hw_mac_cursor_accl mode (e.g. fullscreen DGA).
In order to the sigalstack() to be effective, we must kludge the kernel to
think it's running on another stack. In practise, we provide another stack
for the SIGUSR2 handler. sigusr2_handler_init() fulfills that purpose.
I hope this fixes remaining issues forever. At some point, I had multiple
*_init() handlers in case this is necessary.
2005-06-28 16:47:32 +00:00
gbeauche
a8ba24aecf
Use fast spinlocks only for small enough atomic operations. Otherwise, you
...
run into some performance problems in e.g. video graphics experience because
of busywaits in the current spin_lock() implementation.
2005-06-25 11:40:29 +00:00
gbeauche
198925c5ae
clean-ups from previous experiment with sched_yield(), this one caused a
...
slow-down too.
2005-06-25 11:36:35 +00:00
gbeauche
52138d2014
We HAVE_PTHREADS even if we use our own pthreads implementation, this also
...
induces availability of locking primitives. I will merge the !HAVE_PTHREADS
case (a la Basilisk II) for EMULATED_PPC when I get back to home.
2005-06-25 11:06:24 +00:00
gbeauche
b92ce82a32
Rewrite SheepThreads locks & semaphores. They now look much better and avoid
...
busy waits for acquiring spin locks.
2005-06-25 11:00:30 +00:00
gbeauche
e991e59fc1
fix test_and_set, aka fixes X async replies when clipboard active
2005-06-25 10:58:47 +00:00
gbeauche
2de44f1910
Fix for MacOS X assembler, why don't Apple use GNU as?
2005-06-25 06:33:39 +00:00
gbeauche
d0da2a0dda
- Restore TOC on EMUL_OP_EXEC_RETURN from Execute68k
...
- Restore r13 on EMUL_RETURN, though I never saw r13 clobbered
- Increase stack frame allocated to EmulOp processing on par with BeOS
- Factor GetResource wrappers with a macro + restore r13 too
2005-06-24 22:58:55 +00:00
gbeauche
d8a33b0952
Direct Color visual is OK for fbdev dga.
2005-06-24 22:31:28 +00:00
gbeauche
26ee263b6f
Don't fake the TVECT value on non-BeOS native systems. This is important
...
for systems that use a global r2 as the TLS register, e.g. Linux/ppc with
newer glibc. Also remove the syscall junk which were simply workarounds
for this bug. Remove a duplicate r2 restoration in EmulOp.
BTW, it's possible to get SheepShaver running on Linux/ppc systems with
NPTL rather than SheepThreads.
2005-06-23 16:23:31 +00:00
gbeauche
b9b90fce11
fix last commit
2005-06-23 11:56:43 +00:00
gbeauche
2deca51ca4
Add code to gather some stats on register usage.
2005-06-23 11:37:01 +00:00
gbeauche
c33f09b2d8
Fix crash in FBDev DGA mode when it failed to initialize (here, that was
...
because fb visual was Direct Color instead of True Color). We used to
initialize the_buffer* even if !open_display. Report that gracefully with
the dedicated STR_OPEN_WINDOW_ERR.
2005-06-22 16:40:24 +00:00
gbeauche
c41f403f56
Rework sheepshaver_cpu object allocation and get rid of POSIX'ish functions.
2005-06-22 16:38:15 +00:00
gbeauche
c4a7b10650
op_invoke*() symbols are expected to be in demangled form, i.e. "C" linkage
2005-06-22 16:35:59 +00:00
gbeauche
f887fb7bac
Enable build of the JIT with GCC 2.9x compilers (e.g. "2.96" from MDK 8.1).
2005-06-22 15:37:25 +00:00
gbeauche
ae52321fe7
HOST_FLOAT_WORDS_BIG_ENDIAN is a tristate and, when defined, can still be
...
0, so little endian format. i.e. don't check that with a simple #ifdef...
2005-06-22 15:33:22 +00:00
gbeauche
32b5b4f0bd
Integrate C++ demangler from libiberty so that to handle G++ 2.9X generated
...
symbols.
2005-06-22 15:30:54 +00:00
gbeauche
f226f793dd
Extend internal math library from GNU libc to accomodate older systems
...
with glibc 2.2.X or simply no C99 capable C library. Fix vrfiz instruction
to really truncate on float values.
2005-06-22 12:32:32 +00:00
gbeauche
28a167b3bd
Avoid <limits> to determine vector register numeric_limits<>. This fixes
...
build with ancient compilers like gcc "2.96".
2005-06-22 12:27:01 +00:00
gbeauche
3aa78832fc
pointer-to-member-functions fixlets, NULL may be (void *)0 on ancient
...
compilers which is not suitable here.
2005-06-22 12:25:43 +00:00
gbeauche
c686f5451c
Include <sys/mman.h> for fbdev dga without VOSF acceleration.
2005-06-22 12:24:36 +00:00
gbeauche
e616062d6d
Apple GCC uses __ppc__ as the arch token
2005-06-14 06:35:00 +00:00
gbeauche
46f9e6d2ea
Support JIT on Mach/ppc platforms. Mach/i386 (Darwin/x86) is to follow.
2005-06-14 06:32:52 +00:00
gbeauche
ef30f8e882
remove duplicate
2005-06-11 16:47:00 +00:00
gbeauche
e423a07632
Recognize POWER5 CPUs. ;-)
2005-06-08 12:12:50 +00:00
gbeauche
89643dbce0
Enable VidMode extension by default though some Xservers may still be
...
broken. Probably run-time check for the right VidMode version? My tests
were done with X.org 6.8.2 (VidMode 2.2)
2005-05-12 11:21:10 +00:00
gbeauche
0bce152c20
- Sync with latest B2 video_vosf.h updates.
...
- Enable VidMode extension with FBDev DGA graphics.
- Factor out FBDev/XF86 DGA code.
- Fix pointer grab in fbdev DGA mode, thus fixing scrolling screens in
lower VidModes.
- Only select VidModes that match the requested resolutions, exactly.
- Fix VideoQuitFullScreen() in non FBDev mode.
2005-05-12 11:20:00 +00:00
gbeauche
1853c38130
fix inline asm for gcc4
2005-04-24 17:22:49 +00:00
gbeauche
8db2a3ef62
implement lvsl/lvsr instructions
2005-04-15 17:03:49 +00:00
gbeauche
ac6b7d760e
What a silly thing I wrote! Really nuke the generated file, not the original
...
source. IOW, don't commit stuff when you are ill.
2005-04-07 22:35:47 +00:00
gbeauche
8f80b5ddba
remove generated ppc_asm.S on make clean
2005-04-03 15:35:20 +00:00
gbeauche
8e512ca7fe
Better lock and zero VOSF data, especially for DGA mode. Also make sure
...
that we really update the whole screen when sub-8bpp modes are used
2005-04-02 09:54:16 +00:00
gbeauche
b46eb9ae45
Only support True Color frame buffers. Make it possible to run-time switch
...
depth in FBDev DGA fullscreen mode.
2005-03-28 16:19:28 +00:00
gbeauche
f9e3fd28a1
better logics in sigstack allocation on native platforms
2005-03-28 09:50:58 +00:00
gbeauche
88b493bf38
fix make modules for Linux NetDriver
2005-03-28 09:06:46 +00:00
gbeauche
0f4440f99f
Handle "screen fbdev/WIDTH/HEIGHT" to force use of FBDev DGA instead of XF86
...
DGA mode. In other words, root user can now use FBDev DGA though it's always
recommended to run SheepShaver with a normal user. Increase "ramsize" bounds
in the GUI and fully migrate to new "screen" modes. Remove "windowmodes" and
"screenmodes" defaults.
2005-03-28 09:05:28 +00:00
gbeauche
5810c6a764
Use "screen TYPE/WIDTH/HEIGHT" prefs item by default on Unix systems. The
...
former options (windowmodes/screenmodes) are migrated if the user uses the
GUI. Also make a note about EXPERIMENTAL state for jit68k.
2005-03-27 22:32:46 +00:00
gbeauche
abd2c5406f
Add "screen" prefs item similar to Basilisk II. Don't ignore SEGV by default
...
as this could hide real bugs.
2005-03-27 22:08:33 +00:00
gbeauche
08e30a894e
disable 68k DR emulator for now (not stable enough yet)
2005-03-27 22:06:52 +00:00
gbeauche
d27832cb1c
- Implement fullscreen DGA via fbdev access under Linux. Besides, r/w access
...
to /dev/mem is required on Linux to use XF86 DGA mode. Otherwise, there is
now a fallback to fbdev.
- Forward port some features from Basilisk II (set_window_name,
disable_mouse_accel).
- Don't SIGSTOP the emulation thread on suspend since that would completely
stop the process on Linux. Use a frame buffer lock instead (as B2 does)
2005-03-27 19:05:18 +00:00
gbeauche
0bc6418ea6
Implement screen win/WIDTH/HEIGHT prefs item that overrides any other value
...
for windowmodes and screenmodes. Necessary for arbitrary full screen mode
sizes.
2005-03-27 14:53:04 +00:00
gbeauche
fbe378f780
Add custom video modes for full screen in unsual resolutions (e.g. 1400x1050).
...
X11 and SDL infrastructures have yet to be implemented
2005-03-27 13:44:45 +00:00
gbeauche
ca7a87465b
Disable high-res timings as it could still hang. The advantage is that we
...
now can use special mutexes to debug deadlocks
2005-03-27 13:41:46 +00:00
gbeauche
d9a8a5c10f
cross-compilation fixes
2005-03-24 23:39:19 +00:00
gbeauche
35a29f57df
rearrange for win32 cross-compilation
2005-03-24 23:38:10 +00:00
gbeauche
2d6df7d54d
Respin snapshot with functional Windows port + extra SDL keysym translations
...
in case keycodes were not used (e.g. includes key-'3')
2005-03-20 23:58:43 +00:00
gbeauche
86b049678f
Windows specific configure script and Makefile
2005-03-20 23:45:17 +00:00
gbeauche
0db0d48bf0
Avoid the use of floating-point when loading/storing from/to memory. This
...
could have caused some rounding thus alterations to integer registers on
context switches when lfd/stfd instructions were used. e.g. cygwin compilers
defaulted to i686 code generation and exhibed this behaviour, you could also
see this behavior with -march=i586 -mtune=pentiumpro. GCC is perfectly right
to do those optimizations.
2005-03-20 23:07:11 +00:00
gbeauche
1fb076bc7b
Check that we are running a Windows NT kernel >= 4.0 and drivers are
...
installed correctly (namely cdenable.sys)
2005-03-19 19:09:44 +00:00
gbeauche
9383e5bd50
Enable build with gtk2 and default to that UI if found. Otherwise, it's
...
still possible to select GTK1 as --with-gtk=gtk1 or --with-gtk=gtk.
2005-03-19 09:59:30 +00:00
gbeauche
b29b5e4e23
let it build against gtk2
2005-03-19 09:35:01 +00:00
gbeauche
8071d90849
workaround weird bug lying somewhere in cygwin generated micro-ops for
...
FP load/store of doubles
2005-03-19 07:18:18 +00:00
gbeauche
7fcb230d67
the current ethernet code is not direct addressing clean, so enable it only
...
if real addressing mode is available (e.g. this excludes win32 platforms for
now)
2005-03-19 04:31:59 +00:00
gbeauche
c044b1d8eb
add windows specific sysdeps.h as they are stable enough now. lowered constant
...
offset to 0x02000000 so that we can cope with the new RAM_BASE value.
2005-03-18 00:03:32 +00:00
gbeauche
2990a3ddfc
add missing AboutWindow file for Windows, that have yet to be fully
...
implemented however
2005-03-18 00:01:37 +00:00
gbeauche
cf16210d3a
add other symlinks to B2/Windows files
2005-03-17 23:56:04 +00:00
gbeauche
0f90a78b5a
Always set RAM_BASE to 0x20000000 for now as there could be some weird
...
problems depending on the amount of memory requested. Also fix
initialization of the system dependent timers. Remove unixism.
2005-03-17 23:48:50 +00:00
gbeauche
068e2075fc
fix for native compilation under windows
2005-03-17 23:45:28 +00:00
gbeauche
536c81f641
fix when cross-compiling or using cygwin
2005-03-17 23:44:48 +00:00
gbeauche
287296b363
it makes more sense to dump the crash dump header to stderr
2005-03-17 23:40:01 +00:00
gbeauche
dd7b09ebe8
Use a mutex to protect wakeup_time instead of a semaphore, do we want some
...
sort of barrier synchronisation? Anyhow, the current implementation looks
reasonable enough now when using old LinuxThreads without TLS.
XXX: in the past, the sem_post in sigsuspend_handler would not release the
calling thread thus entering into a mutual waiting condition.
2005-03-16 00:35:51 +00:00
gbeauche
260614a873
fix initialization of timer thread that could be put in indefinite suspended
...
state from the start.
2005-03-15 20:46:50 +00:00
gbeauche
cd31ea2d80
mention improvements to the altivec code
2005-03-14 00:32:12 +00:00
gbeauche
cbf0c3debf
shorten vsldoi native code by one instruction, aka. avoid superfluous
...
shuffle of stale vector value
2005-03-14 00:30:44 +00:00
gbeauche
4918fe8d2d
default to 8 MB cache size on non-{alpha,powerpc} platforms
2005-03-13 23:41:42 +00:00
gbeauche
05a7453d54
MMX/SSE/SSE2 optimizations are now converted to full inline assembly code,
...
aka avoid use of (possibly broken) GCC intrinsics. Add some SSE2 optimizations.
Translate VSLDOI, MFVSCR, MTVSCR instructions. AltiVec Fractal Carbon now
shows more than 1 GFlops performance!
2005-03-13 12:49:30 +00:00
gbeauche
7cc49ca739
fix vsldoi implementation and testsuite
2005-03-13 11:38:35 +00:00
gbeauche
4d96076c62
fix SheepShaver wrappers in PowerPC CPU tester
2005-03-13 11:23:52 +00:00
gbeauche
fa6c3a521b
Try to document major changes between released snapshots
2005-03-09 00:12:27 +00:00
gbeauche
e7d8a54e21
Enable high precision timings on POSIX systems supporting clock_nanosleep().
...
Since pthread_suspend_np() is not available to Linux (but NetBSD 2.0), thread
suspend is implemented likewise to boehm-gc.
2005-03-05 19:07:35 +00:00
gbeauche
48213aa5eb
Revert to no nested native ppc interrupt processing, also filter out cases
...
where InterruptFlags is 0.
2005-03-05 18:33:30 +00:00
gbeauche
fb42e00b8d
Don't require an instruction skipper to fake SCSI globals
2005-03-05 15:44:03 +00:00
gbeauche
9019e71cfc
Preserve all necessary registers on interrupt, thus also permitting nested
...
interrupts to occur. SheepShaver locks should now be reduced.
2005-03-05 15:25:10 +00:00
gbeauche
663facbf97
Fix native Linux/ppc with recent enough glibc that supports TLS; r2 is used
...
in that case. Tell me if I broke other arches, e.g. r13 is no longer saved
in Video and Ethernet stubs, though it seems to be OK.
Colateral feature: SheepShaver should now run on Linux/ppc64 with relevant
32-bit runtime. Native Linux/ppc64 support is harder as low mem globals are
32-bit in mind and e.g. the TLS register there is %r13, %r2 is the TOC
(PowerOpen/AIX ABI)
2005-02-27 21:52:06 +00:00
gbeauche
3f1eb78c6e
having both ppc_asm.S and ppc_asm.s in the same dir on MacOS X is a bad idea...
2005-02-21 22:29:49 +00:00
gbeauche
4aa9a22f41
native NetBSD/ppc support
2005-02-20 18:26:39 +00:00
gbeauche
c602bcb201
Move Linux/asm_linux.S to ppc_asm.S suitable for Linux, MacOS X and NetBSD
2005-02-20 18:25:45 +00:00
gbeauche
81204ca2d9
NetBSD/ppc paranoia checks
2005-02-20 18:25:05 +00:00
gbeauche
8500381648
Initial support for NetBSD/ppc in native mode (some crashes occur but I
...
could boot MacOS 9.0.4)
2005-02-20 18:08:50 +00:00
gbeauche
436522186e
Implement SheepThreads on NetBSD/ppc so that sigaltstack()s are really used
2005-02-20 18:06:40 +00:00
gbeauche
6d5303c868
Support FreeBSD 5.3:
...
- fix implementation of offsetof() with GCC >= 3.4 and C++ code
2005-02-13 18:18:59 +00:00
gbeauche
7df6429591
Support NetBSD 2.0 for x86:
...
- implement trunc() function, isless() and isgreater() macros
- execute genexec.pl through a real perl executable
- disable generation of debug info from dyngen u-ops
2005-02-13 15:32:11 +00:00
gbeauche
df0d5d2a41
Happy New Year 2005!
2005-01-30 21:48:22 +00:00
gbeauche
f401bb1494
really invalidate caches when ROM was patched in native mode
2005-01-30 21:25:24 +00:00
gbeauche
26314e240e
Add InitAll() which covers common initializations so that to avoid duplicate
...
code and possible bugs (e.g. on BeOS/PPC). Likewise for ExitAll().
2005-01-30 21:19:07 +00:00
gbeauche
f891b23e32
add PowerPC,G4 node
2005-01-30 21:13:23 +00:00
gbeauche
126e549397
add FP regs & state to preserved context on interrupt
2005-01-30 21:12:07 +00:00
gbeauche
9b17e04150
Fix Kernel Data for Gossamer ROMs: PVR, CPUClockSpeed et al.
2005-01-30 18:49:48 +00:00
gbeauche
a79ecf38ef
workaround direct access to FCBS from Apple Personal Diagnostics in MacOS 9
2005-01-30 17:39:44 +00:00
gbeauche
4316979264
fix NULL pmf initialization
2005-01-22 23:24:28 +00:00
gbeauche
5a2246b224
take care when only compiling in SDL audio support with X11 graphics
2005-01-19 20:26:01 +00:00
gbeauche
fb3f855d29
better handling of no-dyngen cases in makefiles
2005-01-10 22:50:54 +00:00
gbeauche
2f947f5667
Append disassembler to dyngen, if available
2004-12-21 23:27:11 +00:00
gbeauche
f5a9e6e110
add facility to debug basic blocks extracted from *-ops.o
2004-12-21 23:20:31 +00:00
gbeauche
f2d7689435
direct addressing fixes to video.cpp way to handle the VSL service owner
2004-12-19 23:05:34 +00:00
gbeauche
6b89caf030
little endian fixes to gamma handling code
2004-12-19 15:14:22 +00:00
gbeauche
6ecaec47d9
Don't overwrite our serial drivers (9.0)
2004-12-19 09:26:30 +00:00
