e8dd27f490
Tweaks to the tweaks.
...
SDB needs a rewrite.
2013-01-31 22:00:43 -08:00
837a414249
Minor tweaks to the SDB to get parsing of real games a bit further.
...
Unfortunately it looks like issue #10 will have to be addressed before
anything can progress :(
2013-01-31 21:40:07 -08:00
7f846afdfc
Adding support for disc images.
...
With this, games can now be loaded! Of course, prep fails.
2013-01-31 20:11:13 -08:00
f78fdba9c3
Wiring up xex loading via the new virtual filesystem.
...
File resolution is hacked, but works well enough for testing.
2013-01-31 16:52:50 -08:00
59189f12ab
Skeleton for the virtual filesystem.
2013-01-31 13:27:00 -08:00
c77bcbf879
Better module handling.
2013-01-31 11:27:05 -08:00
9cfc01940e
Thread IDs in critical sections.
2013-01-31 03:01:57 -08:00
56912aa2bd
Rough CriticalSection work.
2013-01-31 02:50:37 -08:00
fdbd30acb7
Partially implemented string functions.
2013-01-31 00:19:48 -08:00
49af0dbc85
Real modules and threads (mostly).
...
Need events/waiting to get proper launch thread behavior.
2013-01-30 22:44:32 -08:00
2ecacedaa6
Cleaning up LLVM exports and adding an access violation trap.
2013-01-30 14:52:41 -08:00
23378443e8
Kernel function work.
2013-01-30 14:26:28 -08:00
42009cda88
Fixes to get things compiling on OSX again.
2013-01-30 11:03:19 -08:00
1585730ec1
Running on Windows!
...
Minus gflags parsing right now.
2013-01-30 02:48:34 -08:00
94d55b5056
GYP updates for Windows.
...
This will break OSX for a moment.
2013-01-30 02:19:38 -08:00
9e05d44723
Implementing (maybe working) Win32 memory.cc.
2013-01-30 01:45:55 -08:00
ae4d8ad40e
Code fixes to get things somewhat compiling on Windows.
2013-01-30 01:35:08 -08:00
f117f870fb
Cleaning up build warnings.
2013-01-29 20:27:24 -08:00
80d74dbe03
Implementing rlwimi.
2013-01-29 18:35:51 -08:00
9c86c29140
Setting up locals in codegen based on disassembly results.
2013-01-29 17:47:34 -08:00
d3054839b7
Now tracking register accesses for each function.
2013-01-29 13:07:59 -08:00
d1333db404
Disassembly for implemented control/memory instructions.
2013-01-29 01:55:03 -08:00
8a560e4cbe
Adding disassembler bits for alu.
2013-01-29 00:41:39 -08:00
775c97bf53
Beginning work on disassembler functions.
...
Ideally, this would be automated, but it's easier to just do it manually.
2013-01-28 21:37:03 -08:00
7b62fa96bd
Progress on ghetto memory allocator.
2013-01-28 21:36:03 -08:00
0fde6135a0
Fixing XO instruction decoding.
2013-01-28 15:31:46 -08:00
9b0cb94e43
Fixing kernel exported variables.
2013-01-28 14:51:23 -08:00
9ad6bb917f
RtlImageXexHeaderField stub.
2013-01-28 14:25:16 -08:00
e93908dd72
Skeleton code for more kernel calls.
2013-01-28 13:28:22 -08:00
6c4af5aa70
Kernel calls and variables now working and tracing better.
2013-01-28 12:36:39 -08:00
5c2060af72
Now running up to the first required kernel call.
...
If memory address validation is turned off it runs a lot further.
2013-01-28 03:03:37 -08:00
46d5a0b51d
Executing a bunch of instructions!
...
Very hacky module startup code, but can now start XEXs! Time to start
implementing kernel stuff.
2013-01-27 22:14:24 -08:00
92a3e19cd9
Fixing to allow multiple executions/process.
2013-01-27 19:33:57 -08:00
12d9c3d15e
First instruction executed in the test runner!
...
Fixes #7 .
2013-01-27 17:49:32 -08:00
bba99d4a22
Refactoring to allow for raw binary loading.
2013-01-27 12:48:10 -08:00
d9b08f569f
Adding a bunch of instructions.
2013-01-26 21:51:31 -08:00
733fe85d37
Cleaning up function registers and setting up CR.
2013-01-26 01:25:31 -08:00
9b3d96a8ae
Cleaning up instruction decoding names to match the docs better.
2013-01-25 23:45:13 -08:00
ffdc1e5d4b
Hacking on some instructions.
2013-01-25 23:32:37 -08:00
ebda245773
Better handling of invalid/unimplemented instructions.
2013-01-25 00:51:56 -08:00
dcb958de54
Adding better register tracking through locals and fixing branches.
...
LLVM does an amazing job of optimizing this. There are many opportunities
to make it better, too, by preventing spilling where not required or only
spilling/filling things when needed.
2013-01-25 00:32:42 -08:00
47481fecf7
Adding tracing methods.
2013-01-24 00:10:24 -08:00
385a4ee23b
Cleaning up function creation.
2013-01-23 22:38:59 -08:00
a10a05b6f2
Misc output cleanup and a test run script.
2013-01-23 21:59:26 -08:00
860a0739ec
Adding gflags to handle command line flags.
...
It sucks, but is the best there is without using boost.
2013-01-23 21:31:23 -08:00
91f9e8b7bb
Generating a lot of code!
...
Still a few missing instructions/variants and other issues, but a big and
valid LLVM module is being generated.
2013-01-22 00:22:27 -08:00
791f14182c
Fixing a bunch of valgrind errors and reworking globals.
2013-01-21 14:31:59 -08:00
b29dd22850
Adding instruction emitting functions.
...
All empty, all ready to go!
2013-01-21 11:55:42 -08:00
95a8be078b
Rearranging code a bit to keep things cleaner.
2013-01-21 10:58:52 -08:00
b91d550ef1
Adding instruction generation loop.
...
Not yet doing anything, but close.
2013-01-21 01:11:57 -08:00
25af5f410a
Cleaning up the names in imports/functions.
2013-01-21 00:13:00 -08:00
68cc34bb79
Fixing up basic block targets and such.
2013-01-20 17:27:49 -08:00
852536ae0a
Basic function analysis.
...
Finds basic blocks and estimates proper function bounds.
Seems legit for compiled code.
2013-01-20 16:46:08 -08:00
d4b0bf73c1
Prepping flow for basic block building.
2013-01-20 09:48:39 -08:00
ca2908db32
Converting everything to C++ cause I'm a masochist.
2013-01-20 01:13:59 -08:00
8a5dcbc1dd
Hacky state for instruction gen.
2013-01-19 11:04:22 -08:00
ea5154af0f
Codegen experimentation.
2013-01-19 05:26:01 -08:00
c18e94c5be
Fleshing out the symbol database.
...
Now detecting a lot of functions via method hints and generating stub
functions. A few holes (in the test xex), but enough to move forward with
codegen.
2013-01-17 23:18:09 -08:00
6c5432eb45
Symbol database skeleton.
2013-01-14 00:02:24 -08:00
bfec194533
Adding instruction table and basic PPC state.
...
Can decode a single instruction.
2013-01-13 21:25:28 -08:00
099e37490a
Work on CPU, codegen skeleton, and thunk.
2013-01-13 15:48:18 -08:00
2f4bc598e5
CPU skeleton.
2013-01-13 00:34:08 -08:00
bcd575f9b0
Massive dump of xenia-info required code.
...
This is a working xenia-info for xex files (no gdfs files yet).
2013-01-12 23:25:41 -08:00
7e4a9220d7
Initial project skeleton.
...
This includes a working gyp-based build of an executable that uses LLVM.
2013-01-11 01:23:08 -08:00
Ben Vanik