da66c68bf6
[CPU] Made xe::cpu::StackWalker destructor virtual since it is an abstract class.
2018-11-17 14:50:30 -08:00
aa14741e74
[x64] Fix unwind info bug; Cleanup win code cache
2018-11-17 14:40:25 -06:00
55fa9e6917
Merge branch 'master' into d3d12
2018-11-17 20:06:55 +03:00
69af4a21ac
[x64] Fill unwind reservation info bytes with 0xCC as well
2018-11-17 11:04:25 -06:00
1f33c221a2
Merge branch 'master' into d3d12
2018-11-10 11:47:13 +03:00
ab53cf2a82
[CPU] Fix some compressed XEXs decompressing past the image size
2018-11-05 19:47:18 +00:00
d55a2a586f
Merge branch 'master' into d3d12
2018-11-02 09:30:53 +03:00
bbb5c938ec
[CPU] Fix XexModule::FindSaveRest not finding functions properly
2018-11-01 15:50:56 +00:00
c7befee395
Merge branch 'master' into d3d12
2018-10-28 20:31:49 +03:00
7443b7e61f
[CPU] Rename ImportLibrary fields to follow naming conventions
2018-10-28 09:41:31 -05:00
6e74ba93d6
[CPU] Add support for delta patches using *_offset fields
2018-10-22 12:26:14 +01:00
d2fd109af3
[CPU] Add more asserts for XEXP descriptor (+ log a warning)
2018-10-20 05:08:14 +01:00
265903fe66
[CPU] Add XEXP support to XexModule, if XEXP is in same folder as XEX
...
This was a headache to work out, big thanks to the lack of documentation on .xexp files... a ton of guesswork was involved here but luckily it turned out well.
I did have to make some pretty major changes to the way XEX files are loaded though.
Previously it'd just load everything in one go: XEX headers -> decrypt/decompress data -> load imports/symbols -> set loader data table entries, etc...
Now it's changed to something like this:
- Load base XEX headers + decrypted/decompressed image data, return X_STATUS_PENDING
- In the LoadFromFile call used to load the XEX, search for XEXP patch file (only .xexp in same folder atm)
- If patch exists: load XEXP, decrypt headers/data, apply patch to base XEX, dispose of XEXP
- Finish XEX load via LoadXexContinue() (handles imports/symbols/loader data...)
This saves us from needing to reset the imports/function/symbol stuff after patching (since all the XEX code will be a lot different), but I'm not really sure if I went about it the best way.
2018-10-20 04:36:21 +01:00
0b7f7e1657
[CPU] Move XEX2 code into XexModule class, autodetect XEX key
...
Code is mainly just copy/pasted from kernel/util/xex2.cc, I've tried fixing it up to work better in a class, but there's probably some things I missed.
Also includes some minor improvements to the XEX loader, like being able to try both XEX keys (retail/devkit) automatically, and some fixes to how the base address is determined.
(Previously there was code that would get base address from optional header, code that'd get it from xex_security_info, code that'd use a stored base address value...
Now everything reads it from a single stored value instead, which is set either from the xex_security_info, or if it exists from the optional header instead.
Maybe this can help improve compatibility with any weird XEX's that don't have a base address optional header?)
Compressed XEX loader also has some extra checks to make sure the compressed data hash matches what's expected.
Might increase loading times by a fraction, but could save reports from people unknowingly using corrupt XEXs.
(still no checks for non-compressed data though, maybe need to compare data with xex_security_info->ImageHash?)
2018-10-20 04:18:18 +01:00
add23849bf
[CPU] Implement lmw and stmw instructions
2018-10-14 13:57:45 -04:00
6e36101b42
[D3D12] Experimental write watch implementation for shared memory
2018-09-24 23:18:16 +03:00
db625892ea
[D3D12] Shared memory typo fix and improvements
2018-08-01 01:09:51 +03:00
4f7edff19d
[D3D12] SHM: Watches prototype, some uploading
2018-07-26 22:52:26 +03:00
dd19701dc4
[CPU] Unify constant operand checks in pack/unpack
2018-06-18 13:55:46 +03:00
c89cc6a229
[CPU] vupkd3d: NaN on negative overflow and tests
2018-06-11 21:34:13 +03:00
0378f532ab
[CPU] Fix W in 2101010 vupkd3d and revert untested saturation
2018-06-11 17:22:15 +03:00
06fcdef968
[CPU] Bring back saturation in D3DCOLOR vpkd3d128
2018-05-31 16:17:27 +03:00
d61aff4389
[CPU] v(u)pkd3d128: Support UINT_2101010 and don't saturate D3DCOLOR
...
Fix broken mesh polygons in the Call of Duty series. Also remove saturation of D3DCOLOR when packing because values near 3 may be accepted, while unpacking D3DCOLOR gives values near 1.
2018-05-31 14:42:11 +03:00
452c58149a
[CPU] Some extra srw tests.
2018-05-22 05:56:20 -05:00
198cc9b9b9
[CPU] Print a message and crash if the CPU does not support AVX.
2018-05-03 22:12:36 -05:00
14abe1a407
Fixup 360 support
2018-04-03 19:02:49 -05:00
157c7780af
[JIT] Add V128 ADD/SUB ( fixes #1095 )
2018-03-03 18:24:57 -06:00
1c1fdb4ccd
[JIT] Properly mask NaNs in vctsxs
2018-03-03 14:06:45 -06:00
894bbf9e26
[JIT] Fix incorrect definition for constant VectorConvertX2X
2018-03-02 19:03:48 -06:00
3a8f8f2ecb
[JIT] Fix OPCODE_SELECT V128 with certain combinations of constants
2018-03-02 18:56:28 -06:00
2af0a3c2f4
[JIT] Mask extraction constants in Value::Extract
2018-03-02 18:55:52 -06:00
49866e970b
[JIT] VectorConvertX2X unsigned support
2018-03-02 16:34:57 -06:00
fd0d2185ef
[JIT] Full support for vcfux
2018-03-02 15:11:29 -06:00
b0b920c040
Linting.
2018-03-02 12:14:09 -06:00
6c97dbaf81
[JIT] Simplify VECTOR_ADD/SUB i32 with saturate
2018-03-02 11:48:19 -06:00
131fcd1320
[JIT] Kill a couple instructions in VECTOR_CONVERT_F2I
2018-03-02 11:41:04 -06:00
04d20e33a5
[JIT] Add another test case to vctuxs
2018-03-02 00:05:43 -06:00
0577b6c9be
[JIT] Full vctuxs support
2018-03-01 23:49:44 -06:00
4766a5ec24
[JIT] Add FIXME on constant Vector F2I (saturation unsupported)
2018-03-01 21:15:54 -06:00
6fd75cea91
[CPU] Handle constant multiply in fmadd/fmsub in constant propagation pass
2018-03-01 20:04:56 -06:00
db53559611
Merge branch 'master'
2018-02-16 16:34:17 -06:00
43f9cebb0c
[JIT] Rewrite CNTLZ and remove one branch
2018-02-16 12:51:39 -06:00
ec501a32c6
[JIT] Apply the same ZeroExtend/Truncate to rlwimix and rlwnmx
2018-02-14 22:05:54 -06:00
e54c24e150
[JIT] New opcodes: OPCODE_LOAD_OFFSET and OPCODE_STORE_OFFSET
...
These take full advantage of x86 addressing, and eliminate extra add operations.
2018-02-14 16:26:49 -06:00
1de598e4ce
[JIT] Comment the offsets in PPCContext
2018-02-14 16:22:14 -06:00
7818cdab60
[CPU] Check for RawModule memory allocation success
2018-02-14 16:21:16 -06:00
02b5a07bc9
[JIT] rlwinmx: Use Truncate/ZeroExtend instead of And 0xFFFFFFFF
2018-02-14 13:50:57 -06:00
1d0b290c3d
[JIT] Remove all calls to ReloadMembase / ReloadContext (rcx/rdx free now)
2018-02-13 13:33:39 -06:00
e2bbae3896
[JIT] Don't bother using a temp for constant addresses < 0x80000000
2018-02-13 13:09:40 -06:00
a1677d34e1
[CPU] MMIOHandler IsRangeWatched now returns true if the entire range is watched.
2018-02-09 19:23:24 -06:00
190108dab6
[CPU] Add some more docs to MMIO handler functions
2018-02-09 18:26:12 -06:00
7da973d9e6
[CPU] Handle cases in access watches where base addresses overlap
2018-02-09 17:21:35 -06:00
9e79babde0
Formatting.
2018-02-07 18:55:26 -06:00
26212bffb0
[x64 seq] Reuse the source variable rather than re-specifying the destination (to make things less confusing)
2018-01-24 20:57:11 -06:00
9395b73217
[Travis] Enable full Xenia build on linux
2017-12-20 16:27:01 -06:00
aaf281351d
Format all code with new clang-format
2017-12-14 20:44:43 -06:00
02e37be4e5
Merge branch 'linux' of git://github.com/dougvj/xenia into linux
...
# Conflicts:
# .travis.yml
2017-12-14 19:20:02 -06:00
b6e7e66fbe
Fix travis LINT
2017-09-20 23:29:39 +02:00
8d261ca73d
x64 JIT: Simplify splats
2017-08-08 00:12:38 -05:00
48a2435bde
Add header dependencies that prevented linux build
...
There were several files with dependancies on vector and cmath that
caused compiler errors under linux. In addition, there was a printf
formatting error that threw a compiler error as well
2017-07-11 11:47:24 -06:00
4b607a68ba
PPC: Fix typo in instruction disassembly for mtspr.
2017-06-11 14:40:59 -05:00
3117d737ed
PPC: Fixed padding for FormatDS instruction decode data.
2017-06-11 14:28:07 -05:00
e8f55d8ff2
PPC: Support FPSCR[RN] (rounding mode)
2017-05-20 12:15:19 -05:00
7a3af9b2c4
PPC: Fix vsl[b/h/w] and vsr[b/h/w] with variable shift amounts.
2017-05-15 11:08:09 -05:00
6ea6b8b51d
x64: Change binary strings to actual binary numbers.
2017-05-14 09:28:58 -05:00
e2434e50a0
Updated to latest Xbyak.
2017-05-14 02:03:01 -05:00
16cff62d36
Base: Remove memory::DeallocationType::kDecommitRelease (an invalid combination)
2017-05-14 00:59:52 -05:00
66dbd86989
PPC: Support NaN in fcmp[o/u]x.
2017-05-14 00:24:35 -05:00
eff9ba31d2
Fixed PPCContext::cr() returning incorrectly shifted state.
2017-05-14 00:02:47 -05:00
6e2bf0b4b1
Native test runner: Support FPSCR
2017-05-13 23:12:53 -05:00
6990d21a8d
Update test framework documentation
2017-05-12 18:34:01 -05:00
d3ed53c43e
PPC: Support v[u]pkd3d128 VPACK_NORMSHORT4
2017-05-12 17:55:52 -05:00
82efbd7bc5
HIR: Add opcode SET_ROUNDING_MODE (taking FPSCR as an argument)
2017-05-12 14:05:27 -05:00
dbece71945
native-tests: Add altivec save/restore support (only up to 32 registers)
2017-05-12 12:00:35 -05:00
7e9d63e33d
Gracefully handle test suites failing to load without exiting.
2017-05-11 15:59:34 -05:00
747f7d65d8
PPC-nativetests: Zero out the context between runs
2017-05-11 15:59:34 -05:00
737b78ad49
Last stretch to get PPC native tests working
2017-05-11 15:09:43 -05:00
7eceb9db09
PPC nativetests require gflags!
2017-05-11 11:41:34 -05:00
cf838439b7
Whoops! Fix formatting.
2017-05-11 11:38:50 -05:00
ccd6d4b199
(WIP) Add a native PowerPC testing frontend
2017-05-08 22:21:43 -05:00
cff80bd782
fctiw[z]x: Sign-extend the result (verified by HW)
2017-04-26 12:28:35 -05:00
d2605bce18
Fixing Altivec lvrx and stvrx
2017-03-31 11:00:51 -07:00
e699be0118
Logging: Add a flag to specify the minimum log level
...
--log_level = (0=error, 1=warning, 2=info, 3=debug)
2017-03-24 16:28:41 -05:00
f012845cb6
JIT: Propagate mul/div by constant one
2017-03-16 18:49:55 -05:00
5ed086f03f
x64 Code Cache: Notify Intel VTune of generated functions.
2017-03-16 17:45:47 -05:00
c4b728b121
MMIO Handler: Don't allow more than one watch to occupy the same region (fire old watches if a new one overlaps)
2017-03-11 20:55:01 -06:00
257fbfc408
ELF Loader: Fix linux build
2017-03-07 12:33:18 -06:00
cbab45c4b7
Improve ELF loading
2017-03-06 17:59:19 -06:00
098d23c3d4
JIT - fctidz: Properly saturate
2017-03-06 01:06:32 -06:00
d8ed251ad1
Tests: Handle integer floating point numbers
2017-03-06 01:04:59 -06:00
31766eaf8d
JIT: oops
2017-03-05 18:06:39 -06:00
d5e7bc8dca
JIT - vctsxs: Properly saturate signed integers
2017-03-05 17:45:17 -06:00
ea89a5d179
JIT - fctixx: Properly handle numbers > INT_MAX
2017-03-05 15:04:26 -06:00
ad5748f3e7
Handle cases where the optional import header isn't present :P
2017-02-25 20:28:45 -06:00
c08d4e862e
PPC Testing: Don't discover . and ..
2017-02-18 00:43:45 -06:00
591e581a70
Merge pull request #663 from resetnow/linux-build-fixes
...
Linux + clang build fixes
2017-02-11 20:32:33 -06:00
d43e2c7ff8
xenia-cpu-ppc-tests is now building on linux
2017-02-10 23:54:10 -06:00
11ae05155d
Fix casting between XThread and Thread causing pointer misalignment
2017-02-10 23:35:35 -06:00
1334656c6a
xenia-cpu-ppc-tests: Adjust link order for clang
2017-02-07 18:15:30 -06:00
f0ba717142
Fix a bunch of missing cmath includes
...
...as well as missing std:: namespace specifiers
2017-02-07 20:26:08 +03:00
4c55039c22
Remove xenia-cpu dependency on xenia-kernel
2017-02-06 21:57:18 -06:00
b66f10f2b8
Testing travis xenia powerpc testing
2017-02-06 00:40:06 -06:00
8947a7626e
Linting fixes :)
2017-02-05 22:53:30 -06:00
803f0be049
Linux header fixes
2017-02-05 22:45:01 -06:00
d730784efb
Added XBDM stubs.
2017-01-09 04:51:55 -06:00
691c3d1021
EXTRACT_I32: Support constant src1
2016-11-11 16:50:42 -06:00
a148b965f1
KernelState should handle module launching
2016-10-24 11:01:10 -05:00
52c75c8dbc
x64 backend: Pass guest return address as first argument to guest functions.
2016-08-26 13:46:21 -05:00
7683069f6c
Set context/membase registers to rsi/rdi respectively.
...
Use offsetof for the backend thunks.
2016-08-26 11:00:42 -05:00
5f4416ee2f
x64 backend: Save nonvolatile XMM registers on host -> guest transitions
...
Define the context and membase registers in the x64 emitter.
2016-08-22 14:55:16 -05:00
e3fdb08ad7
x64 JIT: Allocate constant data outside of the guest heap.
2016-08-18 18:26:55 -05:00
22794902f3
CNTLZ: Invert the result using xor rather than subtraction
2016-08-01 13:25:44 -05:00
a72f1d949f
Constant propagation for vector SGT/UGE/UGT
...
Constant rounding (and fixed vec128 rounding only doing x)
2016-07-25 11:57:14 -05:00
8427acfada
Use vpminud to saturate rather than bitwise ops (shaves 6 instructions per 16 in 32 saturated pack)
2016-07-23 19:19:40 -05:00
2d55b12cc9
Fix a race condition in x64 code cache involving RtlGrowFunctionTable
2016-07-19 13:59:40 -05:00
18ff97e6b6
OPCODE_IS_NAN
2016-07-11 20:44:12 -05:00
3d1d4dea47
OPCODE_DOT_PRODUCT_4 constant propagation
2016-06-28 19:39:22 -05:00
f2ad6b8cb8
OPCODE_RECIP
2016-06-26 13:03:42 -05:00
8965fda402
RSqrt for VEC128_TYPE.
2016-06-20 13:31:11 -05:00
53e37c3167
Constant propogation for DOT_PRODUCT_3.
2016-06-20 13:30:19 -05:00
450ca87120
Added comparison operator for --break_on_instruction.
2016-06-19 21:01:18 -05:00
b5bef8c54d
Undoing the previous commit in favor of finding a better solution (xmm1/xmm2 can't safely be used here).
2016-06-19 02:33:52 -05:00
d04a6cc983
Added support for when both src1 and src2 are constants in EmitCommutativeBinaryXmmOp and EmitAssociativeBinaryXmmOp.
2016-06-19 02:10:40 -05:00
d5865740e2
Add a few more vector tests
2016-06-12 21:45:51 -05:00
0ef16b10e2
Constant Propagation for OPCODE_VECTOR_ADD
2016-06-12 21:41:02 -05:00
841f6fe239
Merge branch 'vulkan'
2016-06-11 19:23:23 -05:00
32011a841e
Formatting.
2016-06-05 17:48:06 -05:00
ff5529714c
clang-format to Google Standards
...
Added formatting to code
2016-05-29 13:22:14 +09:30
f2ca689ea0
Expanded vec128 and VectorSub
...
Added double precision (f64) values to the vec128 structure
Added cases for all formats of data for vector subtraction in the
Value::VectorSub function
NOTE: Unsure what the saturate function is for however maintained
original functionality
2016-05-17 05:34:09 +09:30
0e3c113375
Physical write watches -> access watches (read and/or write watching)
2016-03-17 21:55:16 -05:00
06ba273492
Warn and be ok with symbol services failing.
2016-02-21 14:26:16 -08:00
05107d2d3e
Skeleton xenia::gpu::vulkan implementation, enough to start trace viewer.
2016-02-18 16:43:41 -08:00
5d9a4b6830
Handle all other cases in switch statements
2016-01-20 11:00:37 -05:00
4c0f9e6342
Add missing includes
2016-01-20 10:36:01 -05:00
6777ce6668
Reconcile debugger and save state stuff into a single implementation.
...
Fixes #497 and fixes #496 .
Still rough edges, but at least less duplication.
2016-01-18 11:48:21 -08:00
9473f20c7f
Fixing travis lint diff.
2016-01-13 22:55:28 -08:00
2e826f9321
Supporting constants in Haswell codepaths for mul_add/sub.
2016-01-13 22:29:39 -08:00
0771eea0b3
Change comments to comply with local regulations.
2016-01-13 19:41:15 -06:00
bd6bf16bd1
New flag: disable_global_lock - Disables global lock usage in guest code.
2016-01-13 19:40:00 -06:00
14c5e48120
Enable speculative PACK_TYPE_FLOAT16_3
2016-01-13 01:48:18 -06:00
36372d1eae
Enforce xex memory protections
2016-01-13 01:47:06 -06:00
f0802d75fa
Use atomic compare exchange in stdcx/stwcx (still under the global lock)
2016-01-13 01:45:31 -06:00
51dde62f09
x64 backend vector optimizations
2016-01-13 01:43:23 -06:00
7b6f1388d6
OPCODE_ATOMIC_COMPARE_EXCHANGE
2016-01-13 01:41:58 -06:00
3ccb2a978d
Propagate vector constants.
2016-01-13 01:40:44 -06:00
6d2f083b09
Up the maximum function count to 50,000
2016-01-13 01:30:15 -06:00
86b706d87c
Merge pull request #509 from x1nixmzeng/xex-resources
...
Rebrand the Xenia window with module metadata
2016-01-10 08:55:47 -08:00
3eb602c93a
Assert library name index is in range of the string table
...
As suggested by @DrChat
2016-01-09 17:55:57 +00:00
6530b9dc49
Added callback when new module is launched
...
This avoids having to guess the display window title format
Also manually fixed the linting errors picked out by travis which do not
get picked up using xb format locally
2016-01-09 17:36:46 +00:00
06f259c87d
Mask the name index instead
...
The maximum number of import libraries (32) is already asserted
2016-01-09 02:43:29 +00:00
7e4fdf7669
Fixed rare crash when accessing library import by name
...
The library name index may only be 8-bits.
This bug was not present in the previous implementation due to a bitmask
(0xFF) - see
ea99ba8e3b/src/xenia/kernel/util/xex2.cc (L272)
2016-01-09 01:19:05 +00:00
a95be25e2f
Merge branch 'linuxfix' of https://github.com/sephiroth99/xenia into sephiroth99-linuxfix
2015-12-31 14:39:23 -08:00
952d35911c
Folding build_tools back into the main repo for simplicity.
2015-12-30 16:53:14 -08:00
214532a3e8
Fiddling with FPSCR and such. Still not implemented, but wiring done.
2015-12-30 16:53:14 -08:00
c296d581cf
Readback CA in XER.
2015-12-30 16:53:12 -08:00
9451c0b1cf
Fix different types in std::min arguments
2015-12-30 17:42:18 -05:00
3f49aa3cef
Add default case for unhandled target_type values
2015-12-30 17:42:18 -05:00
da71bc26a2
Use correct value in unhandled case assert
2015-12-30 17:42:18 -05:00
cbc74c92f4
Fixing CR update for divwu/divdu/mulhwu.
2015-12-30 10:04:57 -08:00
69c7bac8db
Fixing branch address calculation.
2015-12-30 09:47:24 -08:00
3c332ef1f5
Adding/removing saturate checks where required.
2015-12-30 09:31:03 -08:00
74d71f7a4a
Fixing the rlw* instructions.
2015-12-30 09:02:29 -08:00
b09e32ec15
Adding tests for broken rlwnm and rliwnm. Not yet fixed.
2015-12-29 20:58:38 -08:00
321e2663ef
Fixing mullw.
2015-12-29 20:42:20 -08:00
d7eff7c9e3
Marking some memory instructions as kM.
2015-12-29 17:53:49 -08:00
cc37fb1358
Updating to the latest capstone. Fixes #440 .
2015-12-29 15:19:08 -08:00
9c694f07b3
Better trap message.
2015-12-29 13:40:53 -08:00
948aa2400d
Formatting all code. Too gnarly to rebase properly.
2015-12-29 13:13:34 -08:00
b9159ddee8
X64CodeCache: Add a couple of overflow debug asserts
2015-12-29 13:09:52 -08:00
383a173a18
REBASE: fixing xthread instruction decoding.
2015-12-29 13:09:50 -08:00
1442f50058
X64Backend: Take a reference of the pair rather than a copy (and formatting)
2015-12-29 13:09:40 -08:00
0f82e0d090
Make the processor automatically generate any ungenerated functions when installing a breakpoint.
2015-12-29 13:09:37 -08:00
ce2d6da2a1
Processor breakpoints: Change iterators to C++11 syntax where necessary
2015-12-29 13:09:35 -08:00
1ce6a7580b
Fix CaptureStackTrace with null out_stack_hash
2015-12-29 12:35:59 -08:00
d99008bfac
RawModule::SetAddressRange - Notify the backend about this range.
2015-12-29 12:35:58 -08:00
38a09cbcc2
ExecuteRaw: Pass in 0xBC... magic so Xenia will return to the guest caller.
2015-12-29 12:35:57 -08:00
6d98628962
Remove Function::Call extern handler shortcut (as we need old behavior for savestates)
2015-12-29 12:35:56 -08:00
67a45203e3
Uninstall the X64 backend exception callback on exit.
2015-12-29 12:35:56 -08:00
e206d13af7
Flag mfmsr/mtmsr/mtmsrd as context-synchronizing instructions
2015-12-29 12:35:04 -08:00
87cdce1440
Export kBlocking tag
2015-12-29 12:32:48 -08:00
db8eb83057
Track export data on Guest Functions
2015-12-29 12:32:47 -08:00
42c657c40a
Remove processor dependency on XThread for breakpoints
2015-12-29 12:32:47 -08:00
666f5543a8
Cleanup ThreadState and XThread
2015-12-29 12:32:46 -08:00
41d5b41523
Automatically install any CPU breakpoints in any newly-defined functions if necessary.
2015-12-29 12:32:01 -08:00
d09e3b7953
Fix incorrect backend breakpoints implementation
2015-12-29 12:32:00 -08:00
28468f8a18
cpu::Processor now tracks a pointer to the emulator.
2015-12-29 12:32:00 -08:00
7d59258839
Processor::ExecuteRaw
...
Changed the breakpoint lock to a recursive mutex.
2015-12-29 12:28:57 -08:00
ac706f6923
Allow breakpoint addresses to be changed after creation
2015-12-29 12:28:56 -08:00
e383e2f101
Processor breakpoint support
2015-12-29 12:28:56 -08:00
214957da3a
Adding finer-grained instruction classification.
2015-12-29 11:02:00 -08:00
cc72de0f64
Moving disasm to new tables.
...
Also adding useful CPU docs.
2015-12-28 20:49:22 -08:00
cd938be678
Removing all uses of the old PPC tables besides disasm.
2015-12-28 10:08:38 -08:00
990d973c74
New ppc opcode decoder.
...
Seems to work and match up with the old decoder (which may be bad).
Old decoder remains until all functionality is replaced.
2015-12-27 22:53:05 -08:00
ca8d658ffe
Speeding up PPC tests significantly.
2015-12-27 12:03:30 -08:00
0e58208add
Dropping elemental-forms and swapping out with imgui.
...
Too much code for such little use. This should simplify porting.
2015-12-26 20:25:24 -08:00
15816327b4
Renaming xe::cpu::frontend to xe::cpu::ppc.
2015-12-14 21:17:55 -08:00
9273359cdd
Removing vestigial ContextInfo.
2015-12-14 20:57:14 -08:00
5e6c1f5a50
Removing some dead code.
2015-12-14 20:49:50 -08:00
3ebd53ba11
Fix crandc/crnand/crnor/crorc not properly taking the complement of the CR fields.
2015-12-13 18:32:57 -06:00
81d0376106
Fixing use before check for null.
2015-12-05 23:24:40 -08:00
bc1e7684ca
Pass rounding mode from i->flags for OPCODE_CONVERT
2015-12-02 14:03:15 -06:00
17d18f7154
Implement a few cases for Value::Cast/Convert
2015-12-02 14:01:28 -06:00
57a823ae39
Add (commented out) code to setup page protections for xex modules.
2015-12-01 19:53:17 -06:00
525d62d437
constant OPCODE_CONVERT and more cases for OPCODE_LOAD
2015-12-01 19:51:28 -06:00
873016f06f
Fix double-swap in constant load
2015-12-01 19:03:51 -06:00
82fe65475e
Fixing sqrt/rsqrt mixup.
...
Queue the 'how did this ever work?!'
2015-11-30 17:22:08 -08:00
51c3df1e2e
Making mtmsr & co synchronize the context.
2015-11-25 17:37:20 -08:00
dfd92757a9
Adding CONTEXT_BARRIER to force the PPC context to synchronize.
...
This is just an annotation right now, as it's not actually needed.
2015-11-25 14:48:36 -08:00
1956b8708e
Fixing include orders and whitelisting others.
...
Will be broken until clang SVN 253772 is available on travis/snapshots.
2015-11-21 11:34:46 -08:00
529227e4e6
MMIOHandler::InvalidateRange
2015-11-08 18:18:28 -06:00
4af6c41740
Moving profiling.h to base - UI could be separated for cleanliness.
2015-11-07 12:31:07 -08:00
aeb635e31d
vcmpbfp tests
2015-11-07 14:27:24 -06:00
abf37ed9be
Fix vcmpbfp returning that the value is out of bounds for values equal to the bounds.
2015-11-06 19:04:59 -06:00
35a9697af7
Merge pull request #446 from AlexAltea/master
...
Fixed mtcrf which was wrongly implemented as mtcr
2015-11-06 15:11:10 -08:00
9985187b23
Implement vpkuwum/vpkuhum
...
Fixed vpkuwus with signed numbers
Implement some constant sources for vec average/unpack
2015-11-06 15:18:48 -06:00
64b0cc9d5e
Enabled tests for vpkuhum/vpkuwum and added test with negative numbers for vpkuwus
2015-11-06 14:57:42 -06:00
5087ccd565
Tracking capstone next.
2015-11-05 18:43:15 -08:00
5525f6c0a2
Tracking master xbyak.
2015-11-05 17:17:52 -08:00
430d4c7182
Fix rare deadlock in the MMIO handler.
2015-11-05 18:13:15 -06:00
f3fe260a17
Fallback case if indirection table could not be allocated.
2015-11-04 06:46:52 -06:00
e0b3d24a50
Merge pull request #445 from DrChat/kernel_trampolines
...
Generate trampoline functions for GetProcAddress on kernel modules
2015-10-24 17:16:33 -07:00
c6751c1e5f
Fixed mtcrf which was wrongly implemented as mtcr
2015-10-25 00:44:33 +01:00
729f762c5d
RawModule SetAddressRange/set_name
2015-10-24 14:02:08 -05:00
55e811450d
Lint
2015-10-22 20:22:34 -05:00
407d79cf3e
Fix MMIO handler race condition by rechecking page access protections under the global lock.
2015-10-22 20:18:08 -05:00
91dfb6a8a0
Merge pull request #436 from DrChat/misc_changes
...
Misc changes
2015-10-17 01:15:20 -07:00
9dfabba583
Fix compile error uncovered by Clang 3.8.
2015-09-25 23:50:05 -04:00
c33e95981b
Formatting.
2015-09-25 18:46:50 -05:00
bf39e33dc0
Raw modules need to commit an executable range too!
2015-09-25 18:35:33 -05:00
a9b557e8b9
Allow passing of >7 arguments through Processor::Execute
...
Set entry status to failed if LookupFunction failed
2015-09-25 18:24:12 -05:00
ee92f75a84
Updating to the latest clang-format. Make sure to update yourselves!
2015-09-22 08:26:19 -07:00
08ae855e07
It seems 'export' is a reserved keyword for clang.
2015-09-22 07:45:52 -07:00
542f5ba8e5
Move GetTempReg higher in the source file, before its first usage.
2015-09-22 07:39:26 -07:00
4c8634bc31
Fixing breakpoints.
2015-09-21 21:10:57 -07:00
5d033f9cb3
A new debugger.
...
Lots of bugs/rough edges/etc - issues will be filed.
Old-style debugging still works (just use --emit_source_annotations to get
the helpful movs back and --break_on_instruction will still fire).
2015-09-20 21:31:05 -07:00
a2bc0443f2
Quick-kill the MMIO handler if the fault address is above the mapping range.
2015-09-07 21:26:33 -05:00
e5fbf840d2
Shuffling kernel/.
2015-09-06 18:07:52 -07:00
790ce8aee1
Switch back to std:: muteces. mutices. mutexen.
2015-09-06 13:34:08 -07:00
9ae807e56e
Initialize builtins to 0.
2015-09-06 10:07:27 -07:00
3c96b6fa0a
DANGER DANGER. Switching to global critical region.
...
This changes almost all locks held by guest threads to use a single global
critical region. This emulates the behavior on the PPC of disabling
interrupts (by calls like KeRaiseIrqlToDpcLevel or masking interrupts),
and prevents deadlocks from occuring when threads are suspended or
otherwise blocked.
This has performance implications and a pass is needed to ensure the
locking is as granular as possible. It could also break everything
because it's fundamentally unsound. We'll see.
2015-09-06 09:30:54 -07:00
33270cd2a0
Making tracing use the fast logger path.
2015-09-05 21:12:10 -07:00
f5e374f9b5
Changing the way the global lock works. Some things are better, I think.
...
Regressions are likely.
2015-09-01 09:45:32 -07:00
5355183590
Don't double-break on debugbreak.
2015-09-01 09:45:32 -07:00
51e70be054
Don't assert on VRSAVE mfspr/mtspr (used by gcc)
2015-08-31 10:51:51 -05:00
a4ed79695d
WSACreateEvent/CloseEvent/SetEvent.
2015-08-30 21:31:52 -07:00
7233df2778
Switch default --ignore_undefined_externs to true.
2015-08-30 20:48:41 -07:00
2f9b37c142
Logging unimplemented instructions.
2015-08-30 20:46:06 -07:00
82a72e1974
Adding MemoryBarrier opcode.
2015-08-30 17:25:02 -07:00
65812438c4
Short-circuiting lvrx/stvrx. Should help bad accesses in many games.
...
Fixes #411 .
2015-08-30 16:38:01 -07:00
dfa5b90c36
Fixing test output.
2015-08-30 16:26:36 -07:00
NPCompete