mirrors/dynarmic
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
2260 commits 1 branch 0 tags 34 MiB
Commit graph

237 commits

Author SHA1 Message Date
MerryMage
5ec8e48593 A32: Implement ASIMD VMUL (floating-point) 
* Also add fpcr_controlled arguments to FPVectorMul IR instruction
* Merge ASIMD floating-point instruction implementations
 2020-06-20 14:20:28 +01:00
MerryMage
bb4f3aa407 A32: Implement ASIMD VMAX, VMIN (floating-point) 2020-06-20 03:21:07 +01:00
MerryMage
656419286c ir: Add fpcr_controlled argument to FPVector{Equal,Greater,GreaterEqual} 2020-06-20 00:50:40 +01:00
MerryMage
d3664b03fe ir_emitter: Default fpcr_controlled arguments to true 2020-06-19 22:51:23 +01:00
MerryMage
f3845cea9a A32: Implement ASIMD VQSUB instruction 2020-05-30 18:19:17 +01:00
MerryMage
4e90754873 IR: Implement VectorSaturated{Signed,Unsigned}{Add,Sub} 2020-05-30 15:55:32 +01:00
MerryMage
d0b45f6150 A32: Implement ARMv8 VST{1-4} (multiple) 2020-05-17 17:01:39 +01:00
MerryMage
a8a712c801 Relicense to 0BSD 2020-04-23 15:45:57 +01:00
MerryMage
4573511fe3 constant_propagation_pass: Prepare for IR matchers 2020-04-22 21:07:09 +01:00
MerryMage
f59b9fb020 IR: Add ReplicateBit microinstruction 2020-04-22 21:07:09 +01:00
MerryMage
09d3c77d74 IR: Add masked shift IR instructions 
Also use these in the A64 frontend to avoid the need to mask the shift amount.
 2020-04-22 21:06:17 +01:00
Lioncash
af3614553b A64/impl: Move AccType and MemOp enum into general IR emitter header 
These will be used by both frontends in the future, so this performs the
migratory changes separate from the changes that will make use of them.
 2020-04-22 21:04:23 +01:00
Lioncash
c6f99235e1 frontend/ir/ir_emitter: Remove unnecessary logical shift overloads 
These aren't necessary anymore, now that the U32U64 overload already
exists.
 2020-04-22 21:02:46 +01:00
Merry
09ee64ea98 Merge pull request #482 from lioncash/fixedfp 
A64: Handle half-precision variants of FP->Fixed instructions
 2020-04-22 21:02:45 +01:00
Lioncash
604f39f00a frontend/ir_emitter: Add half-precision->fixed-point opcodes 2020-04-22 21:01:45 +01:00
Merry
45864133f5 Merge pull request #478 from lioncash/stepfused 
A64: Handle half-precision variants of FRECPE and FRECPS
 2020-04-22 21:01:44 +01:00
Lioncash
824c551ba2 frontend/ir_emitter: Add half-precision opcode variant of FPRSqrtStepFused 2020-04-22 21:01:44 +01:00
Lioncash
5dba99b4f4 frontend/ir_emitter: Add half-precision opcode variant for FPRSqrtEstimate 2020-04-22 21:01:44 +01:00
Lioncash
2184d24e8f frontend/ir_emitter: Add half-precision opcode for FPRecipEstimate 2020-04-22 21:01:44 +01:00
Lioncash
6da0411111 frontend/ir_emitter: Add half-precision opcode for FPRecipStepFused 2020-04-22 21:01:44 +01:00
Lioncash
ad0c698f89 frontend/ir_emitter: Add half-precision variant of FPRoundInt 2020-04-22 21:01:44 +01:00
Merry
cb9a1b18b6 Merge pull request #475 from lioncash/muladd 
A64: Enable half-precision variants of floating-point multiply-add instructions
 2020-04-22 21:01:44 +01:00
Lioncash
a4cadf1cd9 frontend/ir_emitter: Add opcodes for signed saturated left shifts with unsigned saturation 2020-04-22 21:01:44 +01:00
Lioncash
bd82513199 frontend/ir_emitter: Add half-precision opcode for FPMulAdd 2020-04-22 21:01:44 +01:00
Merry
01bb1cdd88 Merge pull request #458 from lioncash/float-op 
A64: Handle half-precision floating point in FABS, FNEG, and scalar FMOV
 2020-04-22 20:58:12 +01:00
Lioncash
8309ec7a9f frontend/ir_emitter: Add half-precision variant of FPAbs 2020-04-22 20:58:12 +01:00
Lioncash
e4c259d69f frontend/ir_emitter: Add half->{single, double} and {double, single}->half conversion opcodes 2020-04-22 20:58:12 +01:00
Lioncash
c97efcb978 frontend/ir_emitter: Add half-precision variant of FPNeg 2020-04-22 20:58:12 +01:00
Lioncash
bd892ec4ef frontend/ir/ir_emitter: Amend FPRecipExponent to handle half-precision floating point 2020-04-22 20:58:11 +01:00
Merry
fb039e232c Merge pull request #442 from lioncash/fcvtxn 
A64: Implement scalar and vector variants of FCVTXN
 2020-04-22 20:58:11 +01:00
Lioncash
5cf1478620 frontend/ir: Add opcodes for vector square roots 2020-04-22 20:58:10 +01:00
Lioncash
7c81a58ed3 frontend/ir/ir_emitter: Alter parameters of FPDoubleToSingle() and FPSingleToDouble() to pass along desired rounding mode 
This will be necessary to special-case the non-IEEE Von Neumann rounding
to odd rounding mode.
 2020-04-22 20:58:10 +01:00
Lioncash
9cf3c25811 frontend/ir/ir_emitter: Add opcodes for floating point reciprocal exponents 2020-04-22 20:58:10 +01:00
MerryMage
fa8925c4df IR: Implement FPVectorMulX 2020-04-22 20:57:37 +01:00
MerryMage
f0920c0ded Fix VShift terminology 
An arithmetic shift is by definition a signed shift, and a logical shift is by definition an unsigned shift.

- Rename VectorLogicalVShiftS* -> VectorArithmeticVShift*
- Rename VectorLogicalVShiftU* -> VectorLogicalVShift*
 2020-04-22 20:55:50 +01:00
Lioncash
d426dfe942 ir: Add opcodes for unsigned saturating left shifts 2020-04-22 20:55:06 +01:00
MerryMage
02150bc0b7 IR: Add fbits argument to FPVectorFrom{Signed,Unsigned}Fixed 2020-04-22 20:55:06 +01:00
MerryMage
90193b0e3d IR: Add fbits argument to FixedToFP-related opcodes 2020-04-22 20:55:06 +01:00
Lioncash
b14eaaec46 ir: Add opcodes for left signed saturated shifts 2020-04-22 20:55:06 +01:00
MerryMage
3e447614c6 IR: Add VectorSignedSaturatedDoublingMultiplyLong 2020-04-22 20:55:06 +01:00
MerryMage
06b31448aa emit_x64_vector: Changes to VectorSignedSaturatedDoublingMultiply 
* Return both the upper and lower parts of the multiply if required
* SSE2 does not support the pmuldq instruction, do sign correction to an unsigned result instead
* Improve port utilisation where possible (punpck instructions were a bottleneck)
 2020-04-22 20:55:06 +01:00
MerryMage
08c0e017a5 IR: Implement Vector{Signed,Unsigned}Multiply{16,32} 2020-04-22 20:55:06 +01:00
Lioncash
e739624296 ir: Add opcodes for vector CLZ operations 
We can optimize these cases further for with the use of a fair bit of
shuffling via pshufb and the use of masks, but given the uncommon use of
this instruction, I wouldn't consider it to be beneficial in terms of
amount of code to be worth it over a simple manageable naive solution
like this.

If we ever do hit a case where vectorized CLZ happens to be a
bottleneck, then we can revisit this. At least with AVX-512CD, this can
be done with a single instruction for the 32-bit word case.
 2020-04-22 20:55:05 +01:00
Lioncash
d4a76aaa04 ir: Add opcodes form unsigned saturated accumulations of signed values 2020-04-22 20:55:05 +01:00
Lioncash
6f911a26da ir: Add opcodes for signed saturated accumulations of unsigned values 2020-04-22 20:55:05 +01:00
Lioncash
b6e74fd17d ir: Add opcodes for performing unsigned reciprocal square root estimates 2020-04-22 20:55:05 +01:00
Lioncash
af83360f89 ir: Add opcodes for unsigned reciprocal estimate 2020-04-22 20:55:05 +01:00
Lioncash
fca7eddb9e A64: Add opcodes for signed saturating negations 2020-04-22 20:53:46 +01:00
Lioncash
7ebfd0f31c ir: Add opcodes for scalar signed saturated doubling multiplies 2020-04-22 20:53:46 +01:00
Lioncash
a0231e5546 ir: Add opcodes for signed saturated doubling multiplies 2020-04-22 20:53:46 +01:00