0207739d76
Can't call functions through known-nil pointers, even if they don't touch local data - LLVM's optimizer might have done something stupid.
2017-11-30 01:07:03 +01:00
0ec1e5e3b2
Don't erase and rewrite the dispatcher when the cache is cleared. Fixes #9708
2017-05-26 15:48:03 +02:00
656645fb93
Improve a sad comment
2017-03-23 10:02:29 +01:00
5ce7e5e109
UWP ARM: Shrink the JIT hack
2017-03-23 10:02:29 +01:00
29ad3180a0
Add hack to make the JIT stable(r) on UWP ARM. Can't make it right without FlushInstructionCache which is not available. Sigh...
2017-03-23 10:02:29 +01:00
62dcb9c70c
Log if Comp_SysCall encounters bad syscall instructions
2017-03-10 00:13:06 +01:00
635b2ada43
Remove a function that didn't make a lot of sense.
2017-01-26 09:50:16 +01:00
8c3552de74
cmake: Detect features at compile time
...
Instead of relying on manually passed down flags from CMake,
we now have ppsspp_config.h file to create the platform defines for us.
This improves support for multiplatform builds (such as iOS).
2016-10-19 12:31:19 +02:00
22b6988424
Fix issue with iOS JIT block linking, see #8122
2016-08-29 17:32:46 +02:00
4e5f1e2797
Fix block linking with W^X mode
2016-08-28 18:54:14 +02:00
b264657d56
Implement W^X-compatible path. Works on Windows and Android, hopefully iOS (untested).
...
Disabled block linking in this mode, can re-enable with some more work later.
To enable W^X on other platforms than iOS, simply change PlatformIsWXExclusive.
2016-08-28 14:52:08 +02:00
f8659b8e1e
Move the IR interpreter out into its own file. Rename it in the UI. First CMake and Android fixes.
2016-05-08 22:06:13 +02:00
49a6a2f6cf
Minor refactor - let's not access a global when we don't need to
2016-05-06 21:22:36 +02:00
ffe4c266ef
Add CodeBlockCommon base class to remove further arch-specificity in JitBlockCache
...
Remove unused ArmThunk.
2016-05-01 11:40:00 +02:00
5aadce59a2
Move architecture-specific code out of JitBlockCache
2016-05-01 11:39:58 +02:00
eebe3e7bce
armjit: Fix block exit safety writes.
...
This ensures we don't link to invalid blocks, fixes BKPTs on relinking,
and fixes BKPTs when relinking a prelinked block.
Should help #8524 .
2016-01-30 20:44:57 -08:00
d628b9b57b
Minor fixes, mostly comments
2015-10-10 10:03:34 +02:00
b2b5f3424f
Fix for 32-bit ARM
2015-10-08 23:11:57 +02:00
c41baab747
Pregenerate code to handle rounding mode switches. This time, for all three cores.
2015-10-08 19:58:37 +02:00
bfed830f91
Remove the ability to disable rounding mode support. It's time.
2015-10-08 14:54:42 +02:00
6dd86cd843
Get rid of the ForceFlushToZero hidden config option
2015-10-08 14:54:41 +02:00
8fdceba7ca
Add timing for all the basics.
...
This way we can see overall stats for a frame.
2015-07-03 12:05:08 -07:00
70fa830ba5
Split out the ReplaceJalTo test logic.
...
This makes it so the IR, in the future, can work correctly for
replacements.
2015-04-12 13:35:10 -07:00
81dec36da8
Use an accessor to read the compilerPC.
...
In the IR it will be read from the block.
2015-04-11 01:14:37 -07:00
a897723e6a
Separate out jit reading nearby instructions.
...
This makes it easier to use an IR for these things, or remove them.
2015-04-11 00:53:24 -07:00
fbaffdceab
Remove some outdated comments, minor stuff
2015-04-06 18:13:36 +02:00
a12e448fb4
ARM64: Stub vertex decoder jit, implementing just enough for the cube.elf cube.
2015-04-06 18:13:18 +02:00
58b059ca14
Some casting cleanup, misc
2015-04-06 18:13:06 +02:00
5de011fe95
Again, sigh..
2014-12-13 21:53:28 +01:00
05a8e2e35d
Some work towards being able to build two JITs together
...
This will be useful for testing/debugging, but not there yet.
2014-12-13 21:13:54 +01:00
9dd6bb56bb
jit: Make available js_ and jo_ in regcaches.
2014-12-07 21:07:23 -08:00
50bb3e1e05
Minor fixes
2014-12-08 00:18:13 +01:00
51d55bd645
Namespacing cleanup (it's bad to do "using namespace" in a header)
2014-12-07 14:44:15 +01:00
4457dca4c9
Rename the ARM Jit class to ArmJit
2014-12-07 14:25:22 +01:00
7740caeade
Buildfix the arm emitter test in the unittest.
...
Also do some preparation for being able to have two JITs compiled at the same time
which may be useful in testing parts of the ARM jit on Windows.
2014-12-07 14:12:13 +01:00
d98bde8e50
Merge the RegCache changes from the old neon-vfpu branch
2014-12-06 12:26:58 +01:00
5bb9d32eaa
jit: Fix partial invalidation of larger blocks.
...
Fixes #7031 .
2014-10-27 19:04:19 -07:00
1064f580e4
armjit: Add proxy blocks for continuing.
2014-10-12 17:20:26 -07:00
01f9521dc5
jit: Invalidate blocks even if they end unevenly.
...
This allows blocks to start and end where ever they need, which should be
good for replacements and for continuing.
2014-10-12 17:13:04 -07:00
928e2adfc9
jit: Avoid applying/restoring the rounding mode.
...
If the game never sets it, we can skip around syscalls, interpreter,
replacements, etc.
2014-10-12 12:51:45 -07:00
8d0dca71fe
jit: Rename the rounding mode funcs to clarify.
...
They apply/restore the value, set/clear is confusing.
2014-10-12 11:35:20 -07:00
f84ebf6bff
sprintf->snprintf, fix some too short buffers
2014-09-14 00:14:11 +02:00
52b6f1095e
armjit: Fix rounding mode, allow non flush-to-zero.
...
Default: force flush to zero (for RunFast mode.) But now it's an ini
option so we can more easily compare armjit differences.
2014-09-11 07:58:51 -07:00
8daff0a25e
armjit: Fix some downcount issues with rounding.
...
When setting the rounding mode we need to be super careful about not
destroying flags or R0 if they are needed.
2014-08-30 16:43:13 -07:00
b7da82eebb
Merge pull request #6762 from unknownbrackets/fpu-rounding
...
Handle fpu rounding mode at least in jits
2014-08-23 10:43:22 +02:00
e9b5e6f277
armjit: Maintain rounding mode throughout jit.
2014-08-22 19:57:50 -07:00
1fcbb7bbd4
armjit: Respect the rounding mode for mul/etc.
2014-08-22 00:32:01 -07:00
97e93f48fd
Clean up LitPool code and re-enable flushing in AsmJit
2014-08-20 18:29:37 +10:00
245a2a3be0
Don't zero out downcount in replacements.
...
It doesn't write out js.downcountAmount in any of these cases, so zeroing
it is wrong.
2014-08-03 13:22:30 -07:00
d060a06fa6
Disable a bunch of function replacements.
...
These are just for speed, let's turn them off. Using a flag because:
* I think there's still some issue with savestates, not sure.
* We might swap this flag to a separate option.
2014-08-03 13:15:41 -07:00
Henrik Rydgård