Commit Graph

361 Commits

Author SHA1 Message Date
Flyinghead 1333d6f776 Remove TARGET_BOUNDED_EXECUTION. Build warnings and clean up 2019-07-28 17:37:04 +02:00
Flyinghead 9932d9bf43 OSX: fix bios files path (now in /data/) 2019-07-24 18:24:58 +02:00
Flyinghead 326bc580ec Merge branch 'fh/unified' into fh/wince-dynarec 2019-07-13 12:50:19 +02:00
Flyinghead bbc11a30aa Naomi rename vars and clean up. Logging changes. 2019-07-13 12:44:10 +02:00
Flyinghead 4f8e18215b Fixed savestates: restore compatibility with lr and master
Add new maple devices from lr
Clean up shil.cpp
2019-07-12 17:20:43 +02:00
Flyinghead 58cb50f587 fix reset hang on naomi 2019-07-10 17:25:11 +02:00
Flyinghead f3a91de1af Unify dreamcast, naomi and atomiswave builds WIP 2019-07-09 23:52:19 +02:00
Flyinghead 10e6b296c5 fix cue issue. better logging 2019-07-08 00:03:44 +02:00
flyinghead b540b437ee win32: unprotect ram regions individually in blockmanager
fixes abort when starting game in windows
2019-07-04 09:36:22 +02:00
Flyinghead 35cd1fcf14 sh4: use new logging 2019-07-01 15:22:04 +02:00
Flyinghead feb1b79353 sh4/dyna: use new logging 2019-07-01 12:17:51 +02:00
Flyinghead c60e2d29e8 modem: add termination function to stop pico thread
use new logging
log: set max verbosity to warning for release builds
2019-07-01 10:28:31 +02:00
Flyinghead c27975fb37 ssa: don't propagate const over interpreter fallback, sync_sr and sync_fpscr
rec-x64: support immediate args for xtract
Set write rtt to vram for Super Speed Racing (a.k.a Flag to Flag)
2019-06-24 18:56:09 +02:00
Flyinghead 49b49c5207 don't write-protect BIOS and IP.BIN blocks
Fixes Grandia II crash at start on arm.
Fix bug in bm_GetBlock()
2019-06-21 16:18:34 +02:00
Flyinghead 70ea7c4d76 ssa: use canonical implementations to compute const ops 2019-06-20 19:27:21 +02:00
Flyinghead 6ed098a929 wince: enable code block protection 2019-06-19 22:55:47 +02:00
Flyinghead 318852e261 dynarec: use mprotect to protect code areas
All ram pages containing code are write protected. In that case, no need
for block checks. Memory reads in the same block(s) can also be executed
at compile time and the results propagated as constants.
When a write is detected in a protected area, the corresponding blocks
are discarded and recompiled using traditional (slow) block checks.

Backported the blkmap code finding change from upstream.
Use smart pointers for block management to avoid reference issues.
Added WriteAfterWrite ssa pass
Fixed crash in ssa ConstProp pass when op list is modified
2019-06-19 11:01:33 +02:00
Flyinghead 8d8e694e23 swap setpeq args in ssa if first is imm 2019-06-18 13:25:43 +02:00
Flyinghead 7120512205 wince: use exceptions in dyna/decoder.cpp and catch them in driver.cpp 2019-06-13 18:36:08 +02:00
Flyinghead e0949fb9aa Fix div matching: support 64-bit div and correct handling of negative
Some games do 64-bit / 32-bit division (Pro Pinball Trilogy)
DIV0s/1 use and output 1's complement numbers
The final reminder fixup wasn't correct for negative dividend
2019-06-13 18:27:21 +02:00
Flyinghead c49694599c rec-arm: use ssa regalloc
add `subc a,a` expression to simplify
2019-06-12 17:31:37 +02:00
Flyinghead e537d92bcc ssa: don't simplify FPU ops because of Inf of NaN 2019-06-12 08:52:42 +02:00
Flyinghead ff745ee4fe fix xtrct canonical implementation 2019-06-10 19:17:24 +02:00
Flyinghead 3dd16e80d2 arm64 and x64 recs use ssa regalloc 2019-06-10 13:57:10 +02:00
Flyinghead 623d70d710 ssa register allocator and more ssa stuff 2019-06-09 19:41:42 +02:00
Flyinghead bd30752b86 Use auto& to avoid copy when iterating. Inline mmu_Read/WriteMemNoEx 2019-06-08 13:04:35 +02:00
Flyinghead 0890150dda dynarec: ssa missing changes 2019-06-04 21:27:45 +02:00
Flyinghead 692556ca5d dynarec: ssa initial commit 2019-06-04 21:20:34 +02:00
Flyinghead 4540acc22e mmu: optimize mmu_full_SQ() 2019-05-28 19:41:21 +02:00
Flyinghead 8cb86de026 Make SMCCheck=Fast work on arm64
The issue is that flushing the dynarec cache makes rewriting fail for
the currently executing block. So this avoids flushing the cache too
often but the problem remains.
2019-05-28 19:40:30 +02:00
Flyinghead 51cf6cf271 regalloc: handle mem ops differently in wince mode
Flush all current writeback spans before the op
Flush instead of killing spans in case the mem access throws an
exception
Some optimizations
2019-05-28 19:37:24 +02:00
Flyinghead c6e92028f2 regalloc: cleanup. no functional change 2019-05-27 22:44:51 +02:00
Flyinghead 5b2cef3fdd dynarec: add negc and xtrct op codes
native implementation of negc and xtrct for x64 and arm64
rec-arm64: pass exception pc in w27 instead of sh4 context
inline mmu_intruction_translation() and clean up
2019-05-26 13:30:05 +02:00
Flyinghead 2c91ca96fc rec-arm64: fix transitions to/from mmu
exit mainloop() before regenerating it
make sure dynarec cache is cleared between runs
when deserializing, clear dynarec cache after setting mmu state
2019-05-25 18:03:18 +02:00
Flyinghead e7e387b443 wince: enable/disable vmem32 based on mmu state 2019-05-24 18:19:33 +02:00
Flyinghead e3d95f2258 wince: unify _vmem and vmem32. Use 4GB virtual space on 64-bit arch
On 64-bit architectures, _vmem first tries to allocate 4GB then falls
back to 512 MB.
The same virtual space is now used by _vmem and vmem32 (mmu)
2019-05-23 11:40:33 +02:00
Flyinghead 5b7d07315b wince: clean up unused stuff 2019-05-22 15:15:53 +02:00
Flyinghead c2c0215e1b Merge remote-tracking branch 'origin/master' into fh/wince-dynarec 2019-05-22 14:13:00 +02:00
Flyinghead fb76efb08d Standard functions for virtual mem alloc/protect/map/... 2019-05-22 11:41:12 +02:00
flyinghead 66cbc0acf6 wince: dynarec boost 1.2 -> 1.5 2019-05-15 15:26:08 +02:00
Flyinghead 2434d0183e wince: better sh4 instr cycle counting. Don't hash blocks if mmu on
Use actual instr latency cycles, including fp instr
Decrease the WinCE boost ratio to 1.2
Don't hash blocks when mmu on
2019-05-14 12:38:56 +02:00
David Guillen Fandos 65e2218792 Fix ARM64 rewrites with NO_RWX_PAGES
Tested on nvidia jetson and Android, so far works great.
2019-05-13 00:23:18 +02:00
David Guillen Fandos 3b760f9869 Add FEAT_NO_RWX_PAGES and implement it for x64 CPU (Win and Linux)
Tested: Both with and without the feature, works only for x64 CPUs for
now, but supported in both windows and linux (see vmem implementation
for it, using mem-mapped files).
2019-05-12 22:02:57 +02:00
David Guillen Fandos 1a2b1c8672 Remove traces of memory manager from Jit driver.
Still stuff to be removed in the Naomi file handler and the ARM
AREC, which need to get ported to the new vmem interface.
2019-05-12 19:47:11 +02:00
David Guillen Fandos b74db6ef53 Moving vmem to separate files with a proper interface.
Works so far for Linux and Android, need to do some testing on Windows.
2019-05-11 22:09:52 +02:00
David Guillen Fandos be1ecbaa8b Fix TARGET_NO_NVMEM and deprecate TARGET_NO_EXCEPTIONS
Linked them both toghether since you can't really define one and
not the other (plus Linux honors one windows the other in some
cases).
More refactoring on this area to follow.
2019-05-10 18:57:28 +02:00
Flyinghead 810b8a59da wince: 32-bit virtual mem space
use fast mem read/write for x64 and arm64 dynarecs
2019-04-29 18:23:00 +02:00
Flyinghead 6b78e1f879 fix dynarec regression 2019-04-28 20:41:39 +02:00
Flyinghead 553f6054ce NO_MMU build fixes 2019-04-19 11:56:11 +02:00
Flyinghead 9920880987 dynarec: reserve code cache space for frequently changing blocks
detect frequent SMC check failures and use a specific code cache area
for these blocks.
flush the temp area when full but keep the main code cache area
2019-04-19 11:45:05 +02:00
Flyinghead 76348b13ce wince: use cpu ratio of 2 in dynarec 2019-04-19 09:58:25 +02:00
Flyinghead d82e5e2017 don't use mmu for dma 2019-04-18 22:48:37 +02:00
Flyinghead 0f34d776f6 wince: delete sr.FD hack, restore FD block check in x64 rec
use longjmp for interpreter fallback exceptions
2019-04-18 14:15:01 +02:00
Flyinghead d68da6bdd4 wince: let fp disable exception run before compiling a block 2019-04-18 13:55:10 +02:00
Flyinghead a25918839b save fpu regs when sr.FD is set, and restore them when unset 2019-04-15 19:02:10 +02:00
Flyinghead 5d6e96463c x64 and win32 build fixes 2019-04-15 18:31:15 +02:00
Flyinghead dece3fc13e wince: use setjmp/longjmp instead of try/catch for dynarecs
WinCE fast mmu implementation
WIP arm64 dynarec
2019-04-15 18:02:34 +02:00
david miller cc9d5ec55b CMake/Master working, tested on windows with Clang && MSC (x86,x64) 2019-04-12 16:59:39 -04:00
David Miller 3d1b82854e
Merge branch 'master' into fh/win32-winresize 2019-04-12 13:03:30 -04:00
Flyinghead 1b04ef4cb1 Get rid of RTC in settings 2019-04-08 22:09:22 +02:00
flyinghead 7f0489ff28 visual studio compatibility
no modem support
no zip or 7z support so no naomi for now
hacked a .asm file as vs doesn't support inline assembly -> code dup
2019-03-30 19:26:05 +01:00
Flyinghead 505c5b6c4d dynarec: FullCheck was being ignored and FastCheck used instead
Fix crash on arm32
Fix infinite loop on x64
2019-03-30 10:06:19 +01:00
Stefanos Kornilios Mitsis Poiitidis af6993a819 dynarec: Refactor smc-option a bit 2019-03-30 07:16:44 +01:00
Flyinghead 6438a402df dynarec: add option to control smc code checks: none, fast, full 2019-03-29 19:23:37 +01:00
Flyinghead c809c6c56f Merge branch 'fh/mymaster' into fh/master-merge 2019-03-25 16:47:47 +01:00
Flyinghead ba00da2420 dynarec: don't throw exceptions if NO_MMU 2019-03-25 13:53:49 +01:00
Flyinghead cb6acab40f missing file from previous commit 2019-03-25 11:56:41 +01:00
Flyinghead ef43883fb5 dynarec: WinCE support WIP
Only for the x64 dynarec atm
Bugs remaining
2019-03-25 11:53:13 +01:00
Flyinghead 8564f497d4 sh4 interp: add IsFloatingPoint() to op desc. Add delay slot 0 hack
Add UsesFPU flag to floating point ops. Use flag instead of specific
test cases in op handlers.
Adjust thrown exception in delay slot (slot illegal exception and slot
fpu disable)
Re-add delay slot 0 hack (Looney Tunes Space Race)
2019-03-20 13:45:35 +01:00
Flyinghead d0f65b19d3 mmu: templatize mmu_full_lookup 2019-03-20 10:58:59 +01:00
Flyinghead a21eedc88a implement fpu disable exception and other interp and mmu fixes
implement sh4 fpu disable exception
implement assistance/PTEA MMU registers
fix some sh4 ops with side effect in interpreter
account for delay slot op cycles
avoid any side effect when using wince tracer
extract SH4_TIMESLICE to single header file (still not used by arm and
x86 recs)
2019-03-19 21:35:55 +01:00
Flyinghead 9e2938dff4 mmu: state change wasn't detected. WinCE syscalls tracing 2019-03-17 23:46:39 +01:00
Flyinghead 73d50486d5 mmu: dynamic switching with read/write mem function pointers 2019-03-17 22:59:18 +01:00
Flyinghead e5b18cd8a0 wince: implement pref SQ to YUV converter and 32bit vram
implement incremental YUV conversion with pref SQ
implement access to 32bit VRAM with pref SQ
init YUV converter when TA_YUV_TEX_CTRL is set
set FIFO available space to 256 through SB_TFREM reg
fake FIFO status through SB_FFST reg
2019-03-14 21:35:33 +01:00
Flyinghead 89c2fd54a9 fix function name typo 2019-03-13 20:04:14 +01:00
Flyinghead f4568ace3c backport disassembler missing piece from nulldc 2019-03-13 17:48:21 +01:00
Flyinghead e6d67baf02 sh4 sched: avoid error when jitter is high and delta becomes negative 2019-03-13 17:21:52 +01:00
Flyinghead 6305df9dab ignore fldi0/fldi1 in double precision instead of dying 2019-03-13 17:19:41 +01:00
Flyinghead 6d6492ddc1 mmu: flush tables when MMUCR.TI is written 2019-03-13 17:17:08 +01:00
Flyinghead 0d0fd212ff Implement Ch2 DMA to 32-bit VRAM
Fixes Giana's Return
2019-03-08 13:23:51 +01:00
Flyinghead 492e771272 Content browser (WIP)
Get rid of the renderer thread. It is now the main/UI thread on all
platforms. The emulator runs in a separate thread.
Content browser displayed at startup.
2019-02-25 17:52:53 +01:00
Flyinghead 35bb81b195 fix some printf format strings 2019-02-16 14:16:50 +01:00
flyinghead 3cdd39170d win32: call os_DoEvents on the emu/main thread and other fixes 2019-02-07 19:20:10 +01:00
Flyinghead 91cfd4b2f7 Reserve and allocate maximum RAM/VRAM/ARAM in all cases
Reserve enough virtual memory space for DC and Naomi
Allocate dynarec entry point tables for max possible ram
Free mem and release vmem on exit
2019-01-24 09:48:58 +01:00
Flyinghead 0cce6cc5a5 Clean up and comments. No functional change 2019-01-24 09:40:14 +01:00
Flyinghead cd4e4cbdc9 x64 dynarec: check if extension is supported by cpu. seh on win32
Check if FMA/AVX/SSE3 is supported before using it
fully naked main loop in win32 with proper seh directives
win32: more xmm regs to allocate and no need to save them when calling
out
2019-01-18 17:02:50 +01:00
flyinghead ed3f866835 win32 build fix 2019-01-16 17:42:36 +01:00
Flyinghead f852480b88 OSX: build fix 2019-01-16 14:44:40 +01:00
Flyinghead cb8e81d473 arm64: direct memory access and jit rewrite
generates direct vmem read & write accesses
trap sigsegv and rewrite using slow path
add w29 to allocatable registers
get rid of literals and use pc-rel branching
minor optimizations and cleanup
2019-01-16 13:04:16 +01:00
Flyinghead c2a048e8d8 arm64: use explode_spans to allocate regs for V2 and F64 params 2019-01-15 08:47:07 +01:00
Flyinghead e241613b8f x64 dynarec: implemented swaplb, fipr, ftrv, frswap and other fixes
native implementations for swaplb, fipr, ftrv and frswap
use explode_spans to map 2V and F64 to registers
save xmm registers when calling subroutine
2019-01-14 21:15:36 +01:00
Flyinghead b465f744ba dynarec: flush fpu regs before FTRV
fixes issue with tokyo xtreme racer (x64)
might need to flush before other ops
2019-01-13 23:21:58 +01:00
Flyinghead fc05727538 dynarecs clean-up
move GetRegPtr and ngen_FailedToFindBlock to sh4/dyna
2019-01-11 23:52:20 +01:00
Flyinghead 0f026552c9 fix comments 2019-01-11 15:54:03 +01:00
Flyinghead a9a2aad8f6 arm64: use register spans allocation. Implement some opcodes natively 2019-01-09 16:35:23 +01:00
Flyinghead 67a4eb8f1f arm64 dynarec using vixl 2019-01-07 21:50:46 +01:00
Flyinghead 10d9761a65 minor fix 2018-12-12 13:49:53 +01:00
Flyinghead 5763da184c Fix bug in WriteMemBlock_nommu_ptr when size is not word-aligned
Bump max opaque polygon to 8192 (alpilot)
Minor lr backport and clean up

Fix corruption in doa2[m] and alpilot
Fix missing sound in Jambo Safari
2018-12-12 12:40:04 +01:00
Flyinghead fb92d8d8c5 less log 2018-11-09 13:23:47 +01:00
Flyinghead a3f898b7d0 Naomi: implement undocumented SH4 registers to make version h bios happy 2018-11-06 10:54:13 +01:00
Flyinghead 47be33d388 Free dynarec code blocks on exit 2018-10-29 16:10:39 +01:00
Flyinghead 8a56710841 New save state format 2018-10-29 15:11:34 +01:00
Flyinghead 22b18d97a0 Don't flush the dynarec cache when the sh4 instruction cache is flushed
Fixes Shikigami No Shiro II slowness
2018-10-28 01:29:44 +02:00
Flyinghead fb84df6665 Better logging 2018-10-20 19:38:21 +02:00
Flyinghead 4b38b9b788 less log 2018-10-16 15:35:28 +02:00
Flyinghead 82e0fc7f60 Don't crash on invalid ram write size. Report and ignore 2018-10-11 21:15:00 +02:00
Flyinghead f419786584 Fix sh4 FTRC op to use correct positive cutoff value.
Fix for interpreter, common dynarec and x86 rec.

Fixes wrong car color in Tokyo Xtreme Racer car selection screen.
2018-10-02 12:38:49 +02:00
Ender's Games ee3a474ee4 Revert "Interrupt Hack setting w/ less stigma"
This reverts commits e5c0f0e, eebf3fc, 14fc7d9
2018-09-26 09:06:07 -04:00
flyinghead c135ab0e90 Win32: fix msvc build. removed tick thread 2018-09-25 14:09:07 +02:00
flyinghead 21f47c03ec Fix dynarec x64 crash with mingw64.
Get rid of CDI warning pop up dialog
2018-09-25 12:27:37 +02:00
Flyinghead de147549c3 Save states implementation 2018-09-20 19:48:46 +02:00
Flyinghead 7ce4fccb37 Merge remote-tracking branch 'origin/master' into fh/mymaster 2018-09-20 17:28:41 +02:00
Sven daae7c8e68 add save states 2018-09-02 09:49:23 -04:00
Ender's Games 14fc7d910a Convert Interrupt Hack to name w/ less stigma 2018-08-26 23:13:25 -04:00
Ender's Games d8226c7b5d Fix: "Functions should be declared at file scope" 2018-08-22 21:14:42 -04:00
Ender's Games e5c0f0ee71 https://code.google.com/p/nulldc/source/detail?r=108
Originally ported from nullDC to libretro in commits:
2fa562db1b46c52b663b3dd4bb33a64907357458
f8eb58ac16a9e5adf662b99be5d00729264808e0
Modified for use w/ reicast per-game configuration
2018-08-22 21:14:02 -04:00
Ender's Games eb11d19687 Include header guards to prevent multiple inclusions 2018-08-21 09:28:54 -04:00
Ender's Games 8d9d40dffc Core: Changes provided by Android NDK compiler 2018-08-19 01:54:15 -04:00
Flyinghead 2fb9927688 Less console spamming 2018-08-17 18:30:54 +02:00
Ender's Games a3f585ea1c Port the dynarec safe flag from nullDC (See #84) 2018-08-16 20:00:10 -04:00
Flyinghead 2d3fd59e04 Revert f13b366e8d57c15a6a97cc0721d68ddb5268385f: the fixNaN function is
completely bogus and the correct one doesn't have any effect.
2018-07-14 09:13:56 +02:00
Flyinghead f13b366e8d Set the value for NaN according to the SH4 specs
The SH4 sets the signaling bit to 0 for qNaN: 7fbfffff instead of the
usual 7fffffff. Same games seem to rely on this.
Fixes Fur Fighters freeze and missing geometry in game.
2018-07-13 18:57:51 +02:00
Flyinghead ea35eeb728 Fix FTRC op in both interpreter and dynarec with respect to Inf and NaN
The -ffast-math gcc option implies the -ffinite-math-only option, which
produces wrong results with Inf and NaN. Use integer math to detect the
sign of float numbers in FTRC to avoid these issues.
Also the upper cut off value for conversion was apparently wrong.
Also fixed the x86 dynarec but not tested.
Fixes wrong car color in Tokyo Xtreme Racer car selection screen.
2018-07-13 12:02:32 +02:00
Flyinghead 648988e622 don't log div32 matching and some GDRom ops 2018-07-10 14:36:28 +02:00
Flyinghead 0df91770d2 Increase dynarec code cache size to 10 MB
Fixes frequent code cache invalidation due to lack of space, which kills
performance (Extreme Sports)
2018-07-06 17:19:37 +02:00
Flyinghead 315205caa9 Add setting to disable div32 matching (Pro Pinball Trilogy)
div32 matching doesn't handle division by zero and edge cases, which
causes crashes with some games.
Setting enabled by default for Pro Pinball Trilogy.
2018-07-06 09:49:39 +02:00
Nicolas HOUDELOT 5c343a219c fix typo 2018-03-05 01:57:00 +01:00
Stefanos Kornilios Mitsis Poiitidis 7c5e49a6d2 dyna: Fix f2i canonical + x86 to saturate 2016-05-14 04:15:16 +03:00
twinaphex b606593774 ftrc fix - revert back to nulldc defaults 2016-05-13 14:44:53 +02:00
Stefanos Kornilios Mitsis Poiitidis d85e4d8dff sh4/mmu: Don't use exceptions when disabled 2015-09-29 03:17:26 +02:00
Stefanos Kornilios Mitsis Poiitidis d99c803a0c sh4/mmu: Typo fixes, less log spam 2015-09-29 03:11:28 +02:00
Stefanos Kornilios Mitsis Poiitidis 4a060b5635 sh4/mmu: C++ exception based sh4 exception support. Won't run wince yet.
- Rewrite mem ops to only modify regs after exception path
- Throw & catch logic for interpreter that raises the exception
- Re-enabled some commented mmu code
2015-09-29 03:11:28 +02:00
Jan Holthuis e9beb86069 Merge pull request #798 from reicast/holzhaus/linux-use-xdg-config-home
Introduce separate config/data dirs (user & system wide)
2015-09-02 16:14:50 +02:00
Jan Holthuis 4267d51f90 stdclass: Make path getter function names more verbose 2015-09-02 15:49:00 +02:00
Jan Holthuis b6d0cddcaa stdclass: Add support for separate config/data dirs and system wide dirs
This adds support for separate config and data dirs.

On Linux, these will be compliant XDG Basedir Specification, i.e.
XDG_CONFIG_HOME and XDG_CONFIG_DIRS (or XDG_DATA_HOME and XDG_DATA_DIRS
respectively). On all other platforms, there currently just set to the
homedir path (so no previous behaviour has been changed).

If reicast wants to read and write a data file, it just calls
get_data_path("/samplefile.txt"). If it does not need to write to
that file, it just uses get_data_path("/samplefile.txt", false). That
way, we can also use system-wide dirs (like /usr/share/reicast on
linux), that the user usually doesn't have write access to.

The same applies for config file, where you use get_config_path(args)
respectively.
2015-09-02 15:48:53 +02:00
Stefanos Kornilios Mitsis Poiitidis dcd77326cc sh4/mmu: Import the old mmu implementation from nullDC.
Reicast doesn't support exceptions yet, so this isn't of much use now,
and is intended mostly as documentation. nullDC used some call stack
hooking magic to handle exceptions, which was never generic and clean
enough to be worth the effort to port to Reicast.
2015-08-27 17:05:06 +02:00
Stefanos Kornilios Mitsis Poiitidis f83130b84b Merge pull request #766 from reicast/fix/ios-and-osx
Fix ios and osx (projectfiles, nvmem)
2015-08-19 23:55:20 +02:00
TwistedUmbrella a4028154f9 Fix to prevent EXC_BAD_ACCESS on iPhone
This is inelegant and should only be temporary but resolves the issue
of CodeCache and ICache “getting lost” during initialization.
2015-08-19 15:33:18 -04:00
Stefanos Kornilios Mitsis Poiitidis 4f62b995b6 reios/naomi: First baby steps to support booting naomi roms
- Setups state, copies binary
- Binary locks up w/ a reboot loop

Naomi roms have a 512-byte header, executable length seems to be
at 368 or 3C0. The rom is copied from [0, len) to 0x0c020000.The
bios then hands over control at 0x0c021000
2015-08-19 01:29:46 +02:00
Jan Holthuis 5ab3d7b59b core/hw/sh4/dyna/shil.cpp: Fix '&&' within '||' warning
Here's the original compiler warning:
../../core/hw/sh4/dyna/shil.cpp:700:24: warning: '&&' within '||'
      [-Wlogical-op-parentheses]
  ...if (op->rd.is_reg() && op->rd._reg==reg_sr_T ||  op->op==shop_ifb)
         ~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~ ~~
../../core/hw/sh4/dyna/shil.cpp:700:24: note: place parentheses around the '&&'
      expression to silence this warning
  ...if (op->rd.is_reg() && op->rd._reg==reg_sr_T ||  op->op==shop_ifb)
                         ^
         (                                       )
../../core/hw/sh4/dyna/shil.cpp:843:25: warning: '&&' within '||'
      [-Wlogical-op-parentheses]
                        if (op->rs1.is_reg() && op->rs1._reg==reg_sr_T
                            ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
../../core/hw/sh4/dyna/shil.cpp:843:25: note: place parentheses around the '&&'
      expression to silence this warning
                        if (op->rs1.is_reg() && op->rs1._reg==reg_sr_T
                                             ^
                            (                                         )
../../core/hw/sh4/dyna/shil.cpp:844:25: warning: '&&' within '||'
      [-Wlogical-op-parentheses]
                                || op->rs2.is_reg() &&
op->rs2._reg==reg_sr_T
                                ~~ ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
../../core/hw/sh4/dyna/shil.cpp:844:25: note: place parentheses around the '&&'
      expression to silence this warning
                                || op->rs2.is_reg() && op->rs2._reg==reg_sr_T
                                                    ^
                                   (                                         )
../../core/hw/sh4/dyna/shil.cpp:845:25: warning: '&&' within '||'
      [-Wlogical-op-parentheses]
                                || op->rs3.is_reg() && op->rs3._reg==reg_sr_T
                                ~~ ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
../../core/hw/sh4/dyna/shil.cpp:845:25: note: place parentheses around the '&&'
      expression to silence this warning
                                || op->rs3.is_reg() && op->rs3._reg==reg_sr_T
                                                    ^
                                   (                                         )
2015-08-15 18:15:50 +02:00
Jan Holthuis 1d0ef81a43 core/hw/sh5/dyna/decoder.cpp: place parentheses around && expression
This compiler warning has been fixed:
../../core/hw/sh4/dyna/decoder.cpp:1181:66: warning: '&&' within '||'
      [-Wlogical-op-parentheses]
  ...|| blk->BlockType==BET_Cond_1 && blk->BranchBlock<=blk->addr)
     ~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../../core/hw/sh4/dyna/decoder.cpp:1181:66: note: place parentheses around the
      '&&' expression to silence this warning
  ...|| blk->BlockType==BET_Cond_1 && blk->BranchBlock<=blk->addr)
                                   ^
        (                                                        )
2015-08-15 18:15:50 +02:00
Stefanos Kornilios Mitsis Poiitidis 67ecd6d9f9 vmem: Automatic fallback to slowpath if alloc fails
This consolidates some of the work done for TARGET_NO_NVMEM and
feat/no-direct-memmap. If nvmem is disabled at compile time or alloc
fails _nvmem_enabled() will return false. Various other fixes
and cleanups all around.
2015-08-12 03:09:44 +02:00
Stefanos Kornilios Mitsis Poiitidis ca83428fa5 naomi: Fix linux builds 2015-08-11 19:13:12 +02:00
Stefanos Kornilios Mitsis Poiitidis 7d0d2ba572 naomi: Initial support for TARGET_NAOMI, windows only
- Import naomi code from nullDC, modify and cleanup
- Only unprotected dimm-board support, custom lst files
- Still a compile option
- Boots naomi bios and some games, no input yet
2015-08-11 19:13:11 +02:00
Stefanos Kornilios Mitsis Poiitidis 9bc0a8ff0f shrec/bm: BM lookup table size follows RAM_SIZE 2015-08-11 19:13:11 +02:00
Stefanos Kornilios Mitsis Poiitidis 365accfde2 windows: fix TARGET_NO_NVMEM 2015-08-11 19:12:20 +02:00
Stefanos Kornilios Mitsis Poiitidis c3c2c68f21 Merge pull request #729 from reicast/wip/softrend
Basic and buggy software renderer
2015-08-11 17:43:47 +02:00
Stefanos Kornilios Mitsis Poiitidis c60fca4973 sh4: Cleanup SetFloatStatusReg a bit 2015-08-11 07:57:23 -04:00
Stefanos Kornilios Mitsis Poiitidis 26cfcd79e9 OSX: Fiddling to get the rec-cpp/noexcept path running 2015-08-11 07:57:22 -04:00
Stefanos Kornilios Mitsis Poiitidis e3845465f0 win64/jit: Generate unwind table for the jit code cache
This required moving the SH4_TCB outside the exe, as new tables that belong on the executable are ignored.
This isn't perfect, but there's a large area to scan for available address space so it shouldn't be a problem
2015-08-11 00:26:02 -04:00
TwistedUmbrella bf75c95ae8 Update project for #755 and fix warnings 2015-08-17 07:59:39 -04:00
Stefanos Kornilios Mitsis Poiitidis 0343feffa6 Merge pull request #714 from reicast/wip/nacl-and-emscripten
NaCL & Emscripten targets

- CPU_GENERIC
- TARGET_NO_THREADS
- TARGET_NO_NIXPROF
- TARGET_NO_EXCEPTIONS
- TARGET_NO_NVMEM
- TARGET_BOUNDED_EXECUTION
- TARGET_NO_COREIO_HTTP
2015-08-03 11:28:29 +02:00
Stefanos Kornilios Mitsis Poiitidis 61394d7a44 shil: Disable external assembly implementations
Using external function pointers is broken after the rec-cpp structural changes,
and relatively low priority so disabled for now.

Fixes builds on arm, non-android (see #720)
2015-08-03 10:41:34 +02:00
Stefanos Kornilios Mitsis Poiitidis ac6bdddae3 Merge pull request #694 from randomstuff/cleanup
Minor Cleanup
2015-08-03 00:09:11 +02:00
Gabriel Corona 8c68f4a06b Fix a 'permissive' warning 2015-07-31 20:37:39 +02:00
Stefanos Kornilios Mitsis Poiitidis d8ccc9d2e7 emscripten: More rec-cpp progress
It almost builds now, except emscripten randomly gives up and crashes. No one likes my code these days...
2015-07-29 06:45:02 +02:00
Stefanos Kornilios Mitsis Poiitidis 17a4af378c nacl: Post-merge cleanups 2015-07-29 04:58:41 +02:00
~skmp 2d9fbdda40 nacl: Fix sqw_nommu to work w/ non-linear nvmem allocation 2015-07-29 04:27:32 +02:00
Stefanos Kornilios Mitsis Poiitidis b9bbb5c0c8 nacl: Work towards rec-cpp 2015-07-29 04:26:50 +02:00
~skmp 376a36bf57 Emscripten: Hacky support, can boot bios
- Add TARGET_BOUNDED_EXECUTION, TARGET_EMSCRIPTEN
- Add emscripten makefile

Of course, there's no jit at this point
2015-07-29 04:08:08 +02:00
Stefanos Kornilios Mitsis Poiitidis a82f9704d6 shrec: Limit max blocksize to 511 shops 2015-07-29 02:29:16 +02:00
~skmp 5b67e3f090 rec-cpp: SH4_TCB doesn't need to be in .text if not executable 2015-07-29 02:28:39 +02:00
Stefanos Kornilios Mitsis Poiitidis fb55e7f6f8 sh4/canonical: Provide impls for sync_sr and sync_fpscr
This shouldn't be a requirement as directly using shil_cf_ext should work,
but for some reason if I don't msvc optimizes the functions out.
2015-07-25 20:36:06 +02:00
Stefanos Kornilios Mitsis Poiitidis cbda836a96 she/core: Less stupid register swap functions 2015-07-25 20:33:53 +02:00
Stefanos Kornilios Mitsis Poiitidis b1f7015a98 rec/cpp: Mostly fully direct specialized dispatchers
- Nest shil implementations to shilop_##name::form::impl for easier template matching
- Add more direct handlers
- Add prints if a direct handler isn't found
- Remove multiple indirect handlers, leave them for fallback
2015-07-25 14:36:04 +02:00
Stefanos Kornilios Mitsis Poiitidis 39e369411c shil/canonical: Fix decls, comment 2015-07-25 14:23:45 +02:00
Stefanos Kornilios Mitsis Poiitidis 9cf9bacd13 build: Refactor HOST_NO_REC everywhere 2015-07-25 13:19:26 +02:00
Stefanos Kornilios Mitsis Poiitidis 67a8ceabdd shil/param: add imm_value() 2015-07-23 02:08:25 +02:00
Stefanos Kornilios Mitsis Poiitidis e50cfe3010 x64 tinyjit: linux support 2015-07-14 03:35:34 +02:00
Stefanos Kornilios Mitsis Poiitidis 3852cf9ac8 Minimal x64 jit: Desync'd, but runs 2015-07-13 23:56:42 +02:00
Stefanos Kornilios Mitsis Poiitidis b385575d6f Get x64 building 2015-07-13 21:13:51 +02:00
Jan Holthuis fb50c7570c sh4_opcodes: Fix printf arguments
This partly resolves #660 (needs to applied to linux-64 branch, too).
2015-06-13 16:55:12 +02:00
Stefanos Kornilios Mitsis Poiitidis 9c2a4dc5a4 iOS fixes 2015-05-16 01:04:30 -07:00
Stefanos Kornilios Mitsis Poiitidis 65ca1382d8 Merge remote-tracking branch 'remotes/origin/skmp/ios-wip' into feat/segfault-context-consolidation
Android/windows builds fine again. Android not tested yet, iOS/OSX not compiled yet
2015-05-16 07:12:19 +02:00
Stefanos Kornilios Mitsis Poiitidis 8c469e4b5c *nix: move context massaging to linux/context.cpp, ALIGN -> DECL_ALIGN
Yay some warnings gone from not conflicting with ALIGN anymore
2015-05-16 03:06:04 +02:00
Stefanos Kornilios Mitsis Poiitidis 5b609a6be4 linux: Sh4 dynarec works 2015-05-08 18:59:20 +02:00
Stefanos Kornilios Mitsis Poiitidis 5ef9d661a6 Merge remote-tracking branch 'origin/master' into skmp/linux86 2015-05-08 17:29:14 +02:00
Stefanos Kornilios Mitsis Poiitidis 5d3c71bee7 Merge master 2015-05-08 17:29:06 +02:00
Stefanos Kornilios Mitsis Poiitidis 3fe8f0eb94 tmu: Correctly schedule next interrupt in underflow, correctly differentiate between large values and underflows
- 32 bit code was "fudgy" on this, so simplified via 64bit counters for edge detection
- Maths to count cycles were wrong
- Let's not do 32 bit wrapped maths after beers without any kind of pen-and-paper design

Should fix #594, also see #645
2015-05-08 02:29:24 +02:00
Stefanos Kornilios Mitsis Poiitidis c25ce04b48 sh4_sched: Avoid scheduling to -1, because, magic values. 2015-05-07 22:10:00 +02:00
Stefanos Kornilios Mitsis Poiitidis 4b46de89bb sh4_sched: Handle disabled events correctly 2015-05-07 22:03:55 +02:00
Stefanos Kornilios Mitsis Poiitidis 20f7a41fe9 sched: Do not re-fire expired callbacks, use sh4_sched_remaing on sh4_sched_tick 2015-05-07 21:33:36 +02:00
Stefanos Kornilios Mitsis Poiitidis 3b1c38ad55 sh4/driver: don't log on div32 missmatch 2015-03-22 01:26:43 +01:00
Stefanos Kornilios Mitsis Poiitidis cca8558d08 Add Serial Console option (debug.SerialConsoleEnable) for KOS, etc 2015-02-25 20:56:58 +01:00
Stefanos Kornilios Mitsis Poiitidis 9ed6945826 sh4/cpu_iNimp: Nicer error output 2015-02-25 19:31:33 +01:00
Stefanos Kornilios Mitsis Poiitidis ae8ce5ad23 sh4/rdv_CompilePC: Fix boot time automatic cache flushes
The bios boots from 0xA... addresses, not 0x8... so the flushes were never triggered
2015-02-25 19:30:08 +01:00
Stefanos Kornilios Mitsis Poiitidis 7c64d57b35 Merge branch 'master' into skmp/bios-hle 2015-02-25 15:21:59 +01:00
Stefanos Kornilios Mitsis Poiitidis f2c07917c7 Partially working dyna for iOS. Very few games working atm.
This works, but is extremelly hacky. Must be started without attached debugger, lldb doesn't want to let go of EXC_BAD_ADDRESS, but reicast really depends on it getting delivered as SIGSEGV/SIGBUS. Also xcode has a really bad day upon seeing the jit code. Oh well.

There's some dynarec bug that causes color corruption on bios logo/boot triagles, TA crash on ikaruga and infinitive loop on crazy taxi. I'd guess some fp-memory-write thingy, abi, or smth. Too bad.

- Force code to compile in arm mode (arm jit -> thumb mem functions is complicated)
- SIGILL, SIGBUS. Works w/o Mach exceptions and EXC_BAD_ADDRESS
- Code buffers move to __TEXT, munmapped && memmapped to actually work
- Primitive input. Button + start, or left (works to get out of bios date screen)
- Fixup emitter for thumb2/interworking (didn't work though, reverted to arm cc)
- Block Manager: Disable mem saving / page fault alloc-on-demand logic
- Move cycle counter to r11, r9 is not clean on iOS. Remove r11 from reg alloc list
- Cache flushes for iOS
- log to log.txt
- load game.chd
2015-01-18 23:52:12 -08:00
Stefanos Kornilios Mitsis Poiitidis d055dbd7cb TMU: Also set the underflow flag. Fixes KOS. 2015-01-07 03:12:33 +02:00
Stefanos Kornilios Mitsis Poiitidis 6ad25b8f53 TMU: Improve interrupt handling 2015-01-06 21:50:00 +02:00
Stefanos Kornilios Mitsis Poiitidis 11e9ed167a Fix building for android 2015-01-06 16:07:18 +02:00
Stefanos Kornilios Mitsis Poiitidis 605fa1c40a TMU: Hello interrupts
Let's pretend that the TMU unit actually exists, it makes games like
Gaunlet Legends happy ~
2015-01-06 15:00:39 +02:00
Stefanos Kornilios Mitsis Poiitidis c5cc2f1e50 Document sh4_shced, fix tmu a bit 2015-01-06 14:10:48 +02:00
Stefanos Kornilios Mitsis Poiitidis 25eee26ce5 First steps towards a HLE bios implementation
This is very simplistic, but can already boot and run many games

Known working so far
- Dead Or Alive 2 LE
- Head Hunter
- Marvel VS Capcom 2

Not working
- (Almost) all others
2014-12-29 22:05:35 +01:00
Stefanos Kornilios Mitsis Poiitidis 56ebda10a9 ios interpreter works, bios runs w/ 3d visulals
THIS IS A MESS. Sorry @angelXwind :p. We'll have to clean up and merge for r8, but most of the nasty work is there.

What works:
- full init, memalloc, mprotects, etc
- boot
- gles2

What doesn't:
- Exceptions for page faults
  - Breaks texture cache
  - Breaks dynarec because of fastpath
  - segfaults don't work, needs special mach exc_server magic that still eludes me after 5 hours
- Dynarecs. RWX issues, and broken buffer mappings. Also some minnor linking stuff (most of it is taken care of)
- Audio. I just killed it off.
- All of the ui. Was getting in the way of me not knowing how to edit story boards...
- Input

Also, merged branch 'skmp/mac86-hackport' into skmp/ios-wip

Have fun~
2014-12-17 10:19:44 -08:00
Stefanos Kornilios Mitsis Poiitidis 86e2408ace wip to get ios compiling 2014-12-17 03:10:18 -08:00
Stefanos Kornilios Mitsis Poiitidis 34bc42b7a4 Add x64 target, fix Sh4RCB pointer sizing, u32->unat in vmem 2014-12-11 14:02:10 +01:00
Stefanos Kornilios Mitsis Poiitidis 2f171e3596 Support HOST_NO_REC for windows 2014-12-11 13:59:31 +01:00
Stefanos Kornilios Mitsis Poiitidis 63485e7a72 SetFloatStatusReg: support gcc for x86 2014-12-09 22:15:44 +01:00
MrPsyMan 7916cd63f3 Removed commit ed8c25ef7c as it causes compatibility problems. 2014-06-05 04:34:25 +03:00