Commit graph

521 commits

Author SHA1 Message Date
MerryMage
3cca3bbd0b jit_state: Split off CPSR.GE 2020-04-22 20:26:40 +01:00
MerryMage
6fde29f5d8 emit_x64: Remove unnecessary ABI overhead in ReadMemory, WriteMemory 2020-04-22 20:26:40 +01:00
MerryMage
6adc554b53 jit_state: Hide cpsr implementation 2020-04-22 20:26:40 +01:00
MerryMage
eb80aae9c0 block_of_code: Move MXCSR switching out of dispatch loop
Also clarify MXCSR entry/exit terminology
2020-04-22 20:26:40 +01:00
MerryMage
a4e85ad565 emit_x64: Make RSB a stack 2020-04-22 20:26:40 +01:00
MerryMage
2a818f9d8e Merge branch 'timing'
We do this to improve timing information before entering a supervior
function. We also do this to try and stay within JITted code as much
as possible, by updating the cycles we have remaining.
2020-04-22 20:26:37 +01:00
MerryMage
ea4c3292d5 BlockOfCode: Detect space remaining
We also clear the code cache when we run out of space.

This closes #111.
2020-04-22 20:26:12 +01:00
MerryMage
256749910f Add AddTicks and GetTicksRemaining callbacks 2020-04-22 20:26:12 +01:00
MerryMage
80c56aa89d Remove unnecessary use of boost::make_optional
Closes #119.
2020-04-22 20:26:12 +01:00
MerryMage
639f7cfd2d reg_alloc: Add IsLastUse optimization for UseScratch 2020-04-22 20:26:12 +01:00
MerryMage
6b122751fe reg_alloc: Remove reliance on IR::Inst::DecrementRemainingUses 2020-04-22 20:26:12 +01:00
MerryMage
30049ca928 emit_x86: Standardize time of DefineValue call 2020-04-22 20:26:12 +01:00
MerryMage
12eaf496fd emit_x64: Perform mask creation for packed instructions in SSE 2020-04-22 20:26:07 +01:00
MerryMage
305e4baa29 emit_x64: Eliminate conversion of GE flags
* We do this so that we can simplify PackedSelect.
* We also try to minimise xmm-gpr/gpr-xmm transfers in PackedSelect.
2020-04-22 20:26:07 +01:00
MerryMage
d1e0a29cd9 Implement IR instruction PackedSelect, reimplement SEL 2020-04-22 20:26:07 +01:00
MerryMage
18f11972c6 emit_x64: Remove SSSE3 implementation of PackedHalvingAddU8
It is much slower than the SSE2 implementation, so there's no point keeping it around.
2020-04-22 20:26:07 +01:00
MerryMage
c4b40909f7 emit_x64: Improve code emission of FPCompare{32,64}
Replace if-chain with table lookup
2020-04-22 20:26:07 +01:00
MerryMage
814e378249 VCMP and VCMPE were the other way around
- This was due to a misunderstanding of what the E in VCMPE means.
- The E refers to an exception being raised when a QNaN is encountered.
- Added unit tests for VCMP{E}
2020-04-22 20:26:07 +01:00
MerryMage
08f638d447 emit_x64: pmaxuw and pminuw require SSE 4.1
This commit is intended to close citra-emu/citra#3137.

pmaxuw and pminuw were used to perform unsigned comparisons; we emulate
these using a signed comparison by offsetting the inputs by 0x8000 for
CPUs that do not support SSE 4.1.
2020-04-22 20:26:07 +01:00
Mat M
77fe2aeeaa emit_x64: Amend doxygen parameters for InvalidateCacheRange() (#116) 2020-04-22 20:22:01 +01:00
MerryMage
19dcdde90b block_of_code: Add vzeroupper instructions where AVX-SSE transitions may occur 2020-04-22 20:22:01 +01:00
MerryMage
60d9392b5c block_of_code: BlockOfCode should provide cpu info 2020-04-22 20:22:01 +01:00
MerryMage
148c01e08f interface_x64: Remove is_executing assert from HaltExecution
In multithreaded code this can be triggered due to a race.
2017-10-14 23:35:01 +01:00
MerryMage
f6cf265bc5 block_of_code: BlockOfCode::ABI_* should be const 2017-09-29 01:35:24 +01:00
MerryMage
29471be317 Standardize location of storage-class specifiers: Place at beginning of declarations
Justification: C99 specifies that doing otherwise is an obsolescent feature.
2017-09-29 01:23:45 +01:00
MerryMage
b992e5f8ec Ranged cache invalidation
* Fix clearing code block on a partial invalidation
* Remove unnecessary use of boost::variant
* Code cleanup
2017-09-11 00:11:05 +01:00
MerryMage
568b52d4ba externals: Update Xbyak to v5.51
Xbyak now supports multi-byte nops
2017-08-17 21:34:54 +01:00
MerryMage
1613846ab0 reg_alloc: Handle XMM registers in LoadImmediate 2017-08-16 23:11:05 +01:00
Yuri Kunde Schlesner
38eb7e0314 emit_x64: Use alternative Xbyak names for and, or, xor
Also enabled XBYAK_NO_OP_NAMES, allowing us to stop using
-fno-operator-names.
2017-06-12 07:57:46 +01:00
MerryMage
2c9dcfa2db backend_x64: Rename UnwindHandler to ExceptionHandler 2017-04-20 14:08:56 +01:00
MerryMage
0d47f50f57 block_of_code: Implement farcode 2017-04-19 18:58:36 +01:00
MerryMage
9ac890c62d reg_alloc: Fix for LLVM's interpretation of the System V ABI
This aspect of the System V ABI is under-defined. LLVM choses a
different interpretation from GCC and ICC.

Most other compilers assume the callee is responsible zero-ing the
upper bits of the register if necessary. LLVM assumes the caller
has zero-extended the register.

This is a quick fix for this problem until zext-tracking is
implemented.
2017-04-08 22:12:37 +01:00
MerryMage
a5bb81a97c backend_x64: Remove dispatch loop in Jit::Run 2017-04-08 10:04:53 +01:00
MerryMage
1b37420459 backend_x64: Simplify dispatcher 2017-04-08 09:35:45 +01:00
MerryMage
4c5de3905b emit_x64: Correct mutation of immutable in FPThreeOp{32,64}
operand (args[1]) was erroneously declared as non-scratch.
operand's value could be modified if FTZ was enabled.
2017-04-01 09:57:14 +01:00
MerryMage
05e97058c3 parallel: Add and Subtract with Exchange improvements
* Remove asx argument from PackedHalvingSubAdd{U16,S16} IR instruction
* Implement Packed{Halving,}{AddSub,SubAdd}{U16,S16} IR instructions
* Implement SASX, SSAX, UASX, USAX
2017-03-24 15:56:24 +00:00
Lynn
fd068ed6b8 Ranged cache invalidation 2017-03-20 11:58:25 +00:00
MerryMage
d9c69ad997 constant_pool: Implement a constant pool 2017-03-19 13:08:04 +00:00
Lioncash
d85137ed65 interface_x64: Amend LLVM disassembly code
This would previously attempt to perform pointer arithmetic on void pointers,
which would cause compilation errors.
2017-03-07 18:32:04 +00:00
MerryMage
92a01b0cd8 Prefer ASSERT to DEBUG_ASSERT 2017-02-26 23:30:40 +00:00
MerryMage
135346eb2e reg_alloc: Move implementations out of header 2017-02-26 23:30:39 +00:00
MerryMage
184db36caf reg_alloc: Call DecrementRemainingUses in only one place 2017-02-26 23:30:38 +00:00
MerryMage
51fc9fec05 reg_alloc: Reorganize 2017-02-26 23:30:37 +00:00
MerryMage
cf93ab3d31 reg_alloc: Remove old register allocator interface 2017-02-26 23:12:26 +00:00
MerryMage
08a467bf9a emit_x64: Port to new register allocator interface 2017-02-26 23:12:25 +00:00
MerryMage
f883bad2cc reg_alloc: New register allocation interface 2017-02-26 21:37:35 +00:00
MerryMage
13ac0c234e reg_alloc: Differentiate between ReadLock and WriteLock 2017-02-26 21:37:34 +00:00
MerryMage
6c3df057fa reg_alloc: Remove unused functions 2017-02-26 21:37:33 +00:00
MerryMage
1ee4c07f14 reg_alloc: Reimplement ScratchHostLocReg 2017-02-26 21:37:32 +00:00
MerryMage
640faab8a7 reg_alloc: UseHostLoc is no longer necessary 2017-02-26 21:37:30 +00:00
MerryMage
9518bbe06e reg_alloc: Reimplement UseScratchHostLocReg 2017-02-26 21:37:29 +00:00
MerryMage
e1d8238c50 reg_alloc: Stub UseOpArg 2017-02-26 21:37:27 +00:00
MerryMage
2b078152e7 reg_alloc: Reimplement UseHostLocReg 2017-02-26 21:37:26 +00:00
MerryMage
aefe550428 reg_alloc: Remove the Def concept from register allocator internals 2017-02-26 21:37:25 +00:00
MerryMage
65cccf070e reg_alloc: Properly encapsulate HostLocInfo 2017-02-26 21:37:24 +00:00
MerryMage
469bb6253f backend_x64: Factor EmitExclusiveWriteMemory64 into ExclusiveWrite 2017-02-26 15:34:26 +00:00
MerryMage
d7ab1f9c64 backend_x64: Fix ABI violation in ReadMemory and WriteMemory
Caller-save registers were not saved before call instruction.

Refer to issue #98.
2017-02-26 15:34:25 +00:00
MerryMage
3768174783 ir_opt: Constant propagation pass works better with a DCE just before it 2017-02-26 15:28:35 +00:00
MerryMage
bbeea72eba ir_opt: Remove redundant shift instructions 2017-02-26 15:28:14 +00:00
MerryMage
517fe0f18e emit_x64: WriteMemory* microinstructions do not define a value 2017-02-25 11:54:47 +00:00
MerryMage
1ff60bc69f reg_alloc: Move OpArg into own header 2017-02-21 23:38:36 +00:00
MerryMage
058f7b5de6 emit_x64: Make EmitTerminal type-safe
Avoid the use of boost::variant::which, which tends to produce code which
is not verifiable at compile-time.
2017-02-16 19:40:51 +00:00
MerryMage
2af39dfaa8 emit_x64: Make reg_alloc a local variable
reg_alloc contains state that is only valid on a per-block basis, so there
is no reason for it to be a member variable.
2017-02-04 09:29:35 +00:00
MerryMage
a0e9417912 ir_opt: Initial constant propagation pass implementation 2017-01-30 21:49:46 +00:00
MerryMage
2447f2f360 callbacks: Factorize memory callbacks into inner structure 2017-01-30 21:42:51 +00:00
MerryMage
9ecdd32b84 coprocessor: Implement fast-path for Coproc{Send,Get}{OneWord,TwoWords}
Allow coprocessor interface to provide pointers instead of a callback.
This allows for a fastpath when all that is required is to read or write a
value and no other action needs to be taken.
2017-01-08 14:56:06 +00:00
MerryMage
48693eb6ff Implement coprocessor-related microinstructions
* CoprocInternalOperation
* CoprocSendOneWord
* CoprocSendTwoWords
* CoprocGetOneWord
* CoprocGetTwoWords
* CoprocLoadWords
* CoprocStoreWords
2017-01-08 14:56:06 +00:00
MerryMage
1efd3a764d IR: Remove unused microinstructions NegateLowWord and NegateHighWord 2017-01-05 20:16:39 +00:00
FernandoS27
d5610eb26c Implement UHASX, UHSAX, SHASX and SHSAX (#75) 2016-12-28 21:32:22 +00:00
MerryMage
c7e5216473 emit_x64: EraseInstruction now also invalidates the instruction
There is now no longer a need to call DecrementRemainingUses on the parent
instruction.
2016-12-22 18:43:11 +00:00
Fernando Sahmkow
677f62dd6f Implement SHSUB8 and SHSUB16 (#74)
* Implement IR operations PackedHalvingSubS8 and PackedHalvingSubS16
2016-12-22 12:02:24 +00:00
MerryMage
36082087de callbacks: Read code using MemoryReadCode callback 2016-12-21 21:39:14 +00:00
MerryMage
6a269a6ebd IR: Add microinstructions UnsignedSaturation and SignedSaturation 2016-12-21 19:51:25 +00:00
MerryMage
02b2ab7581 emit_x64: Pass tmp to ExtractMostSignificantBitFromPackedBytes in EmitPackedAddU8 2016-12-20 22:07:51 +00:00
MerryMage
097f6a83da emit_x64: Document ExtractAndDuplicateMostSignificantBitFromPackedWords 2016-12-20 22:06:14 +00:00
MerryMage
03f168094d emit_x64: Document ExtractMostSignificantBitFromPackedBytes 2016-12-20 22:05:51 +00:00
FernandoS27
8919265d2c Implement SADD8, SADD16, SSUB8, SSUB16, USUB16 2016-12-20 21:52:38 +00:00
FernandoS27
3f6ecfe245 Implemented USAD8 and USADA8 2016-12-20 21:52:38 +00:00
MerryMage
b1d3e7aae9 emit_x64: Refactor patching code
* Only have a single std::unordered_map for patching information
* Factor patch emitters into own functions
* Implement EmitX64::Unpatch
2016-12-20 14:06:55 +00:00
MerryMage
74a95ea51e block_of_code: Rename alloc to AllocateFromCodeSpace 2016-12-18 23:43:48 +00:00
MerryMage
96e46ba6b5 Implement QADD, QSUB, QDADD, QDSUB 2016-12-15 22:34:29 +00:00
MerryMage
276873bf70 Wrap #pragma warning with #ifdef _MSC_VER .. #endif 2016-12-15 21:36:02 +00:00
MerryMage
0e8b626d87 CMakeLists: Globally disable MSVC warning C4592
C4592: Symbol will be dynamically initialized (implementation limitation)
2016-12-15 21:09:55 +00:00
MerryMage
91e851a991 CMakeLists: Enable /W4 on MSVC 2016-12-15 20:52:23 +00:00
MerryMage
63caed7b09 emit_x64: Remove argument names of unused arguments 2016-12-15 20:52:22 +00:00
MerryMage
5bea2e1680 block_of_code: Support stack unwinding on Windows 2016-12-12 07:49:18 +00:00
MerryMage
4962d92b79 block_of_code: Do not regenerate prelude when clearing cache 2016-12-12 07:49:18 +00:00
MerryMage
179a3388f9 block_of_code: Provide an alloc function to allocate space in the code block 2016-12-12 07:49:18 +00:00
Lioncash
f467589346 emit_x64: Remove unnecessary casts 2016-12-05 20:30:19 +00:00
Lioncash
a37631c010 emit_x64: Use reinterpret_cast for pointer casts 2016-12-05 20:30:19 +00:00
Lioncash
fafa845f64 emit_x64: Make GetBasicBlock() const qualified 2016-12-05 12:46:22 +00:00
Lioncash
6a16edc0fb emit_x64: Move implementations into the cpp file
Prevents needing to rebuild everything including the emitter if any
details ever change.
2016-12-05 12:46:22 +00:00
Lioncash
282029f60a emit_x64: Forward declare BlockOfCode 2016-12-05 12:46:22 +00:00
Lioncash
6898b74c78 emit_x64: Get rid of indirect includes 2016-12-05 12:46:22 +00:00
MerryMage
54d051977f emit_x64: Use movdqa instead of movaps in EmitPackedSubU8
While movaps and movdqa are behaviourly equivalent, using movaps may incur
a domain crossing penalty on some microarchitectures. This is because
movaps is an instruction in the floating-point domain while the following
instructions are in the integer domain.
2016-12-05 01:00:51 +00:00
MerryMage
52e1445f43 Implement USUB8 2016-12-05 00:29:15 +00:00
MerryMage
5c1aab1666 Implement CLZ
Includes tests
2016-12-04 22:56:33 +00:00
MerryMage
1a1646d962 Implement UADD8 2016-12-04 20:52:33 +00:00
MerryMage
7cad6949e7 IR: Implement new pseudo-operation GetGEFromOp 2016-12-04 20:52:06 +00:00
MerryMage
25f21b5371 emit_x64: Inline nzcv computation into EmitFPCompare32 and EmitFPCompare64 2016-12-04 11:43:31 +00:00
MerryMage
cede5e442a emit_x64: Use xorps/xorpd when argument to TransferToFP32/TransferToFP64 is an immediate zero 2016-12-03 11:41:10 +00:00
MerryMage
e166965f3e Implement VCMP 2016-12-03 11:41:09 +00:00
MerryMage
f2fe376fc6 Support 64-bit immediates 2016-12-03 11:29:50 +00:00
Mat M
95f34c683c reg_alloc: Remove unnecessary breaks after returns (#54) 2016-12-02 19:14:44 +00:00
Mat M
de1f831d79 microinstruction: Make use_count private (#53)
Makes the operation a part of the direct interface.
2016-11-30 21:51:06 +00:00
MerryMage
3621a925b2 reg_alloc: Register allocator related constraints belong with the rest of the register allocator
HostLocToReg64 contained two DEBUG_ASSERTs invloving constraints that
really belonged to the register allocator.

The register allocator prevents allocation of RSP and R15 because those
are reserved for the stack pointer and the state pointer respectively.
2016-11-30 19:42:41 +00:00
MerryMage
5f11b4f50e HostLoc: R15 is a GPR 2016-11-30 18:38:03 +00:00
Sebastian Valle
14eb70d7e4 VFP: Fixed the VCVT behavior when converting from unsigned 32-bit values. (#51)
Use a 64-bit register to hold the values so that we don't end up interpreting them as signed values.
2016-11-27 23:25:50 +00:00
Merry
0ff8c375af Implement UHSUB8 and UHSUB16 (#48) 2016-11-26 18:27:21 +00:00
Merry
cb17f9a3ed Implement SHADD8 and SHADD16 (#47) 2016-11-26 18:12:29 +00:00
MerryMage
c0c1bb1094 Implemented UHADD16 2016-11-26 11:28:20 +00:00
Sebastian Valle
4d44474ad4 Implemented the ARM UHADD8 instruction. (#45)
The x64 implementation uses the SSSE3 instruction PSHUFB.
A non-SSE fallback is provided in case the CPU doesn't support it.
2016-11-25 20:32:22 +00:00
MerryMage
b6f7b8babd ir: Implement GetGEFlags, SetGEFlags 2016-11-23 19:44:27 +00:00
Mat M
6a2174ebfa Add missing explicit specifiers (#27) 2016-09-07 12:08:48 +01:00
Mat M
6e0f27a500 types: Add helpers for determining single and doubleword extension registers (#26) 2016-09-07 12:08:35 +01:00
Mat M
5bc9ce544f arm_types: Move into arm folder (#25) 2016-09-06 00:52:33 +01:00
MerryMage
1f61a3d7bc jitstate: Rename fields s/guest_FPSCR/FPSCR/ 2016-09-05 14:42:21 +01:00
Mat M
6d53bb6d7e arm_types: Split out LocationDescriptor (#20)
This isn't really an ARM-specific type, since it's used to indicate a
Block location.
2016-09-05 11:54:09 +01:00
Mat M
858796a029 Eliminate variable shadowing warnings with MSVC (#17) 2016-09-04 11:30:57 +01:00
Mat M
7f9a0c3c38 Remove unnecessary explicit includes (#16) 2016-09-03 21:48:03 +01:00
Mat M
26db11cd71 reg_alloc: Use a strongly-typed enum for representing OpArg type (#15) 2016-09-03 18:30:03 +01:00
Mat M
05b189bc26 arm_types: Specialize std::hash for LocationDescriptor (#14)
Same thing, but with the benefit of working with anything that uses
std::hash by default.
2016-09-03 12:48:47 +01:00
Mat M
8c4df46580 FPSCR: Make value constructor explicit (#13)
Maintains consistency between the PSR helper.
2016-09-03 12:48:31 +01:00
Mat M
6ec651498d arm: Add PSR helper type (#3) 2016-09-02 17:34:33 +01:00
Mat M
1e781d911a reg_alloc: const correctness (#8) 2016-09-02 17:30:01 +01:00
MerryMage
711b3e29d3 interface: Allow ClearCache to be called at any time 2016-09-02 10:59:33 +01:00
Mat M
fb6d838bd9 dynarmic: Remove poison_memory ClearCache parameter (#1)
Unused since the switch to Xbyak
2016-09-01 09:47:09 +01:00
MerryMage
4321e829d1 callbacks: Add user_arg argument to InterpreterFallback 2016-09-01 02:00:08 +01:00
MerryMage
3b5c43b427 Optimization: Read page-table directly for memory access 2016-09-01 00:58:02 +01:00
MerryMage
57169ec093 abi: Implement ABI_PushCallerSaveRegistersAndAdjustStack and ABI_PopCallerSaveRegistersAndAdjustStack 2016-09-01 00:57:22 +01:00
MerryMage
702e181b35 backend_x64/abi: Reversing XMM list leads to incorrect ordering 2016-08-31 23:06:49 +01:00
MerryMage
b10c438e8e jitstate: Remove code argument from ResetRSB 2016-08-31 21:57:33 +01:00
Lioncash
ea6a4e82b5 block_of_code: Make CallFunction accept function pointers only 2016-08-31 21:51:44 +01:00
Lioncash
37d64f0c86 hostloc: Simplify static_assert 2016-08-28 22:10:23 +01:00
Lioncash
f2bf795876 intrusive_list: Interface changes
- Remove the root pointer from iterators.

This is unnecessary, since the only way to get a valid iterator is
either from a node itself (it transiently becomes an iterator via the
underlying interface), or through the iterator interface for the list.
This should also result in better code generation, as each increment or
decrement of an iterator is now branchless.

- Remove iterator_to

This is actually a pretty dangerous function, since it would immediately
create an iterator into the list using the given item, even if it's not
actually part of the list. This was only left around due to lack of
type handling around constructors.

- Add other overloads for erase() and remove()

Now handles iterators, pointers, and references.
2016-08-28 20:56:40 +01:00
MerryMage
7912a79fa5 emit_x64: align before emitting blocks 2016-08-27 11:04:43 +01:00
MerryMage
41c8dabf0b block_of_code: nop should probably default to a size of 1 2016-08-27 10:57:48 +01:00
MerryMage
dca3b2f079 Implement VMRS and VMSR 2016-08-26 22:47:54 +01:00
MerryMage
814348371e emit_x64: EmitX64::Emit: block.Location() returns by value 2016-08-26 19:43:29 +01:00
MerryMage
4f6ea715b2 emit_x64: EmitX64::Emit doesn't need descriptor argument 2016-08-26 19:14:25 +01:00
Lioncash
32c24d2cb3 Use 'false' instead of '0' in asserts 2016-08-26 18:52:08 +01:00
MerryMage
ba31f43672 reg_alloc: UseDefOpArgXmm: default value for argument desired_location should be any_xmm, not any_gpr 2016-08-26 18:50:08 +01:00
MerryMage
7fedf04e79 reg_alloc: Deduplicate constants in RegAlloc::HostCall that were already defined by abi.h 2016-08-26 18:43:50 +01:00
MerryMage
59a8e14d1c reg_alloc: Correct OpArg::setBit for Reg 2016-08-26 15:23:38 +01:00
MerryMage
065c53ebfc emit_x64: Make ZeroIfNaN64 branchless 2016-08-26 15:23:08 +01:00
MerryMage
9901ed0f51 block_of_code: Optimize nops 2016-08-26 13:46:19 +01:00
Lioncash
0102951bdd Convert formatting over to fmtlib 2016-08-26 13:13:19 +01:00
MerryMage
ed3a686d1d Implement public header files 2016-08-26 00:44:50 +01:00
MerryMage
656d4f7252 emit_x64: inhibit_emission is obsolete
Not used anymore; unused ever since intrusive lists were introduced.
2016-08-25 23:24:16 +01:00
MerryMage
4322c0907c microinstruction: Rename FindUseWithOpcode to GetAssociatedPseudoOperation, encapsulate associated variables 2016-08-25 21:08:47 +01:00
MerryMage
922d1fd198 Merge branch 'xbyak' 2016-08-25 16:54:48 +01:00
MerryMage
d04b9eaa81 backend_x64/block_of_code: Reset labels when ClearCache() is called 2016-08-25 16:18:18 +01:00
MerryMage
e32812cd00 Port x64 backend to xbyak 2016-08-25 16:18:17 +01:00
Lioncash
0e12fb6a56 basic_block: Move all variables behind a public interface 2016-08-25 16:14:37 +01:00
MerryMage
13908c5a58 reg_alloc: Insert braces around DEBUG_ASSERT
DEBUG_ASSERT becomes an empty statement in release-mode; an if statement
with an empty statement produces a compiler warning.
2016-08-25 13:09:18 +01:00
MerryMage
dc26afbd7e translate_arm: Translate more than one conditional instruction in a block 2016-08-25 13:05:33 +01:00
Lioncash
1395baefa9 interface: Return register files by const reference
Prevents unnecessary copies where they aren't particularly required.
2016-08-25 12:51:41 +01:00
MerryMage
22cca5ff72 emit_x64: Actually advance RSB pointer 2016-08-24 23:19:47 +01:00
Lioncash
eba3a06d80 frontend: Introduce FPSCR register helper class
Encapsulates all of the FPSCR state.
2016-08-24 20:51:14 +01:00
MerryMage
b5a86889cd Implement VCVT 2016-08-23 22:20:04 +01:00
MerryMage
c8b2f63c93 get_set_elimination_pass: Eliminate unnecessary gets/sets of extended registers 2016-08-23 15:57:20 +01:00
MerryMage
8d1b9f32ca Standardize indentation of switch statments 2016-08-23 12:19:27 +01:00
MerryMage
72250b119f backend_x64/block_of_code: Add more floating point constants
* MFloatPositiveZero32
* MFloatPositiveZero64
* MFloatMinS32
* MFloatMaxS32
* MFloatMinU32
* MFloatMaxU32
2016-08-22 15:54:19 +01:00
Lioncash
1abe881921 basic_block: Add proxy member functions for the instruction list
Currently basic block kind of acts like a 'dumb struct' which makes things
a little more verbose to write (as opposed to keeping it all in one place,
I guess). It's also a little wonky conceptually, considering a block is
composed of instructions (i.e. 'contains' them).

So providing accessors that make it act more like a container can make working
with algorithms a little nicer. It also makes the API a little more
defined.

Ideally, the list would be only available through a function, but
currently, the pool allocator is exposed, which seems somewhat odd,
considering the block itself should manage its overall allocations
(with placement new, and regular new), rather than putting that
sanitizing directly on the IR emitter (it should just care about emission,
not block state). However, recontaining that can be followed up with,
as it's very trivial to do.
2016-08-22 13:44:56 +01:00
Lioncash
86f803da04 reg_alloc: Use Inst's HasUses() function where applicable 2016-08-20 21:26:09 +01:00
Lioncash
36a0ad5bc2 reg_alloc: const correctness for ValueLocation() 2016-08-19 19:33:57 +01:00
MerryMage
2d6a86e43c Remove <cassert> 2016-08-19 01:53:24 +01:00
MerryMage
269160ef0d emit_x64: Clear RSB-related caches when ClearCache() is called 2016-08-18 18:18:44 +01:00
MerryMage
1a3f3ac435 emit_x64: Correct behaviour of PackedOperation for immediate argument
x64 MOVD_xmm does not support an immediate oparg
2016-08-18 18:17:17 +01:00
MerryMage
b2de47954b EmitX64: Emit correct cycle count on cond failure 2016-08-18 18:16:18 +01:00
Lioncash
841098a0bc ir: separate components out a little more 2016-08-17 20:46:21 +01:00
Lioncash
cbd99e4367 jitstate: Use std::array's fill() in ResetRSB
Performs the equivalent behavior
2016-08-17 10:10:43 +01:00
Lioncash
74ee92ee38 jitstate: const correctness
GetReturnFromRunCodeAddress is const qualified, so this can accept a const
pointer. This also allows for making the constructor accept a const
pointer as well.
2016-08-17 10:10:43 +01:00
Lioncash
439619c827 reg_alloc: Make GetRegLoc return by const reference
Considering a HostLocInfo instance houses a std::vector, every time this
function is called can cause a potential heap allocation.

This can be somewhat unnecessary because this function is only used to query
for information we already have.

Considering this is used by several other internal query functions such as
IsRegisterOccupied, IsRegisterAllocated, and IsLastUse, this will result
in better codegen (returning an address is just 3 instructions excluding
the ret instruction for returning, meanwhile heap alloc can be 60+).

This also renames the function to have the same name as its non-const
counterpart, since overloading will just select the correct function
instead of putting that onus on the developer.
2016-08-17 10:08:08 +01:00
MerryMage
0ebb572e2d Optimization: Make RSB a ring buffer instead of a stack 2016-08-15 15:48:22 +01:00
MerryMage
7d7ac0af71 Optimization: Make SVC use RSB 2016-08-15 15:02:08 +01:00
MerryMage
6c45619aa1 Optimization: Implement terminal LinkBlockFast 2016-08-15 14:33:17 +01:00
MerryMage
624e84fa09 Optimization: Tweak RSB 2016-08-15 14:08:06 +01:00
MerryMage
070298b948 Optimization: bugfix! Return Stack Buffer location hash calculation was incorrect 2016-08-15 13:21:58 +01:00
MerryMage
e164ede4dc TranslateArm: Implement MRS, MSR (imm), MSR (reg) 2016-08-15 11:50:49 +01:00
MerryMage
8fc21f481a RegAlloc: Handle case when def is unused 2016-08-13 01:55:03 +01:00
MerryMage
d43d97b990 EmitX64/EmitPushRSB: Assert that patch location is of correct size 2016-08-13 00:52:31 +01:00
MerryMage
960d14d18e Optimization: Implement Return Stack Buffer 2016-08-13 00:10:23 +01:00
bunnei
8e68e6fdd9 TranslateArm: Implement QADD16/QSUB16/UQADD16/UQSUB16. 2016-08-12 19:00:44 +01:00
bunnei
4b09c0d032 TranslateArm: Implement QADD8 and UQADD8. 2016-08-12 19:00:44 +01:00
bunnei
127fbe99cb TranslateArm: Implement QSUB8. 2016-08-12 19:00:44 +01:00
bunnei
86fe29c6d2 TranslateArm: Implement UQSUB8. 2016-08-12 19:00:44 +01:00
MerryMage
1029fd27ce Update documentation (2016-08-12) 2016-08-12 18:17:31 +01:00
MerryMage
df39308e03 TranslateArm: Implement CLREX, LDREX, LDREXB, LDREXD, LDREXH, STREX, STREXB, STREXD, STREXH, SWP, SWPB 2016-08-09 22:57:20 +01:00
MerryMage
29d30bf931 Interface: Added Jit::Reset to reset CPU state 2016-08-09 22:45:54 +01:00
Tillmann Karras
5d26899ac9 Add simplified LogicalShiftRight64 IR opcode 2016-08-08 22:27:05 +01:00
Tillmann Karras
ccb2aa96a5 Add support for the APSR.Q flag 2016-08-08 22:27:04 +01:00
MerryMage
975f011fc0 BackendX64/RegAlloc: Do not allocate RSP for guest use 2016-08-08 16:01:07 +01:00
MerryMage
abd113f160 EmitX64: Renamed patch_jmp_locations to patch_jg_locations 2016-08-08 15:56:07 +01:00
MerryMage
52fa998e6b EmitX64: EmitTerminalLinkBlock: Fix behaviour when setting T and E flags 2016-08-07 22:47:43 +01:00
MerryMage
04c1a0d2de EmitX64: Switch MXCSR when switching to interpreter 2016-08-07 22:47:17 +01:00
MerryMage
a32063fa60 EmitX64: Implement block linking 2016-08-07 22:11:39 +01:00
MerryMage
328422b740 RegAlloc: HostCall flushes all XMM regsiters 2016-08-07 21:02:16 +01:00
MerryMage
a2c2db277b VFP: Implement VMOV (all variants) 2016-08-07 19:25:12 +01:00
MerryMage
aba705f6b9 BackendX64: Merge Routines into BlockOfCode 2016-08-07 18:08:48 +01:00
MerryMage
0f412247ed VFP: Implement VSQRT 2016-08-07 12:19:07 +01:00
MerryMage
3f1345a1a5 VFP: Implement VNMUL, VDIV 2016-08-07 10:56:12 +01:00
MerryMage
12e7f2c359 VFP: Implement VMUL 2016-08-07 10:21:14 +01:00
MerryMage
97b5fa173f VFP: Implement VSUB 2016-08-07 01:45:52 +01:00
MerryMage
ce6b5f8210 VFP: Implement VABS 2016-08-07 01:27:18 +01:00
MerryMage
f88b1b4c2e FPSCR: Save and restore MSCSR across supervisor call, fix MXCSR exception mask 2016-08-07 01:10:19 +01:00
Tillmann Karras
9264e2e04c Use XOR when loading a zero immediate 2016-08-06 21:17:11 +01:00
Tillmann Karras
846d07d7b5 Add Sub64 opcode 2016-08-06 21:17:11 +01:00
Tillmann Karras
b9f4f1ed0f Add carry support to MostSignificantWord 2016-08-06 21:17:11 +01:00
MerryMage
411e804b0d Interface: Forward declare Arm::LocationDescriptor 2016-08-06 20:11:35 +01:00
MerryMage
4b31ea25a7 VFP: Implement VADD.{F32,F64} 2016-08-06 20:03:15 +01:00
MerryMage
94d5738f62 BackendX64/Routines: Add floating-point constants 2016-08-06 20:01:47 +01:00
MerryMage
8754728a82 BackendX64/RegAlloc: Corrected code emitted by EmitMove for XMM->Spill case 2016-08-06 20:01:47 +01:00
MerryMage
8cc4fe8a10 BackendX64/RegAlloc: HostLocToX64 now handles XMM registers properly 2016-08-06 20:01:47 +01:00
MerryMage
856298577d EmitX64: Don't give MOVSX or MOVZX an immediate oparg 2016-08-06 01:03:39 +01:00
MerryMage
640ce48baa VFP: Implement {Get,Set}ExtendedRegister{32,64} 2016-08-05 19:06:10 +01:00
MerryMage
4c0a85f3b3 EmitX64: Correct EmitPack2x32To1x64 implementation 2016-08-05 18:43:24 +01:00
MerryMage
742eeb8913 BackendX64/RegAlloc: Correct debugging asserts and correct UseDef behaviour for spill locations 2016-08-05 18:43:22 +01:00
MerryMage
d80dcc5367 BackendX64/EmitX64: Eliminate unnecessary MOVs in Add64, Mul, Mul64, SignExtendWordToLong, ZeroExtendWordToLong, Pack2x32To1x64 2016-08-05 15:27:29 +01:00
MerryMage
2b025183a2 BackendX64/RegAlloc: Correct UseDefRegsiter behaviour for last use 2016-08-05 15:24:25 +01:00
MerryMage
b4aa01ccf4 Merge remote-tracking branch 'tilkax/master' 2016-08-05 14:14:06 +01:00
MerryMage
94e75ad32f BackendX64/EmitX64: Reduce number of MOVs by using reg_alloc.{RegisterAddDef,UseDefOpArg,UseOpArg} 2016-08-05 14:11:27 +01:00
MerryMage
92bd5f214b BackendX64/RegAlloc: Add RegisterAddDef, UseDefOpArg, UseOpArg 2016-08-05 14:10:39 +01:00
MerryMage
ca40015145 IR: Add Breakpoint IR instruction (for debugging purposes, emits a host-breakpoint) 2016-08-05 14:07:27 +01:00
Tillmann Karras
72c503016c Fix Pack2x32To1x64
Not sure how to fix this properly.
2016-08-05 02:09:30 +01:00
Tillmann Karras
3fdc093d10 Add more IR opcodes for multiply instructions
Pack2x32To1x64, LeastSignificantWord, MostSignificantWord, IsZero64,
Add64, Mul, Mul64, SignExtendWordToLong, ZeroExtendWordToLong
2016-08-05 02:09:30 +01:00
Tillmann Karras
af27ef8d6c Optionally disassemble x86_64 code using LLVM 2016-08-05 02:08:41 +01:00
Tillmann Karras
2488926341 Add IR opcode RotateRightExtended
to rotate through the carry flag
2016-08-03 00:47:16 +01:00
Tillmann Karras
306e070ab5 Use opcodes.inc for emit_x64.h 2016-08-03 00:44:08 +01:00
MerryMage
1252bd653d RegAlloc: Define constructors for HostLocInfo to make MSVC happy 2016-08-03 00:25:42 +01:00
MerryMage
4414ec5bc8 RegAlloc: Allow allocation of XMM registers 2016-08-02 13:46:12 +01:00
MerryMage
864081d1a0 BackendX64: ArithmeticShiftRight: Fix incorrect immediate size for SAR 2016-08-02 12:00:11 +01:00
MerryMage
93af160c97 arm_types: Add FPSCR to Arm::LocationDescriptor and make Arm::LocationDescriptor have a FauxO-like interface 2016-08-02 11:54:02 +01:00
MerryMage
be87038ffd IROpt: Port get/set elimination pass to current IR 2016-08-02 11:51:05 +01:00
MerryMage
51448aa06d More Speed 2016-07-22 23:55:00 +01:00
MerryMage
5fbfc6c155 Implement some simple IR optimizations (get/set eliminiation and DCE) 2016-07-21 21:48:45 +01:00
MerryMage
90d317b868 Implement memory endianness. Implement Thumb SETEND instruction. 2016-07-20 15:34:17 +01:00
MerryMage
3f11a149d7 Implement Thumb Instructions: BLX (imm), BL (imm) 2016-07-18 22:18:58 +01:00
MerryMage
e0d6e28b67 Implement Thumb instructions: BX, BLX (reg), B (T1), B (T2) 2016-07-18 21:04:39 +01:00
MerryMage
8a310777a1 backend/EmitX64: Handle new_pc<1:0> == '10' case in BXWritePC 2016-07-18 20:01:48 +01:00
Subv
703a46ec99 Pass the current IR::Block by reference to the emitter.
This avoids calling the copy constructor more times than needed.
2016-07-18 11:27:33 -05:00
MerryMage
f7e3d7b8d2 Implement Thumb PUSH instruction 2016-07-18 15:11:16 +01:00
MerryMage
c18a3eeab4 Better MSVC support
* Avoiding use of templated variables.
* Now compling on MSVC with /WX (warnings as errors).
* Fixed all MSVC warnings.
* Fixed MSVC source_groups.
2016-07-18 10:38:22 +01:00
MerryMage
3720da4e19 Implement thumb16_{SXTH,SXTB,UXTH,UXTB,REV,REV16,REVSH} 2016-07-16 19:23:42 +01:00
MerryMage
07eaf100ba Reorganise src/frontend: Add subdirectories disassembler and translate 2016-07-14 14:39:43 +01:00
MerryMage
9b2aff166a Implement arm_SVC 2016-07-14 14:29:46 +01:00
MerryMage
7d7751c157 Allow IR blocks to require a cond for block entry.
* IR: Add cond, cond_failed.
* backend_x64/EmitX64: Implement EmitCondPrelude
2016-07-14 12:52:53 +01:00
MerryMage
4ab4ca58f9 backend_x64/EmitX64: Improve emitted code for non-carry ArithmeticShiftRight 2016-07-14 09:02:27 +01:00
MerryMage
08e848044d backend_x64: Inline Routines::GenReturnFromRunCode into emitted code 2016-07-12 16:46:27 +01:00
MerryMage
619b451902 clang support 2016-07-12 14:31:43 +01:00
MerryMage
8449deb0bc MSVC support 2016-07-12 13:28:09 +01:00
MerryMage
09420d190b IR: Implement IR microinstructions ALUWritePC and LoadWritePC 2016-07-12 10:58:14 +01:00
MerryMage
1410221b47 Implement thumb1_STR_reg, thumb1_STRH_reg, thumb1_STRB_reg 2016-07-11 23:11:05 +01:00
MerryMage
e7922e4fef Implement thumb1_LDR_literal, thumb1_LDR_imm_t1 2016-07-11 22:43:53 +01:00
MerryMage
cbcf61a9e6 backend_x64/RegAlloc: Provide convenience function HostCall to save registers necessary as per host ABI 2016-07-11 15:28:10 +01:00
MerryMage
d11df9067d Implement thumb1_BIC_reg 2016-07-10 10:44:45 +08:00
MerryMage
98a64a92b1 Implement thumb1_ORR_reg 2016-07-10 09:06:38 +08:00
MerryMage
8145b33882 Implemented thumb1_ROR_reg 2016-07-10 08:18:17 +08:00
MerryMage
aa72323823 Implement thumb1_CMP_imm 2016-07-08 21:32:01 +08:00
MerryMage
92142d5a22 Implement thumb1_SUB_reg 2016-07-08 18:49:30 +08:00
MerryMage
df0c324923 Implement thumb1_EOR_reg 2016-07-08 18:14:54 +08:00
MerryMage
8a0511d297 Implement thumb1_AND_reg 2016-07-08 17:44:53 +08:00
MerryMage
d0b48bfb59 Implement thumb1_ADD_reg_t1 and thumb1_ADD_reg_t2 2016-07-08 17:44:51 +08:00
MerryMage
e93fb0ba2b EmitX64: remove emit_fns map, use a switch statement instead 2016-07-08 15:28:56 +08:00
MerryMage
421ab344ad EmitX64::EmitTerminalInterpret: Restore RSP before CALL 2016-07-07 22:03:45 +08:00
MerryMage
e5f6450a24 Start implementing Thumb disassembler 2016-07-07 21:51:47 +08:00
MerryMage
f31b530703 Fuzz thumb instructions 2016-07-07 19:01:47 +08:00
MerryMage
5711e62419 Implement terminal instructions 2016-07-07 17:53:09 +08:00
MerryMage
14388ea690 Proper implementation of Arm::Translate 2016-07-04 21:37:50 +08:00
MerryMage
d743adf518 Reorganisation, Import Skyeye, This is a mess 2016-07-04 17:22:11 +08:00
MerryMage
65df15633d First Commit 2016-07-01 21:01:06 +08:00