Commit graph

1596 commits

Author SHA1 Message Date
Lioncash
28a8b4d210 A64: Handle half-precision floating point in scalar FMOV
This is simply performing a scalar value transfer between registers
without conversions, so this is trivial to handle as-is.
2020-04-22 20:58:12 +01:00
Lioncash
d7ac5a664f A64: Handle half-precision floating point in FCVTL
Like FCVTN, now that we have half-precision floating point conversion
functions available, we can go ahead and use those to eliminate the
interpreter fallback.
2020-04-22 20:58:12 +01:00
Lioncash
fe84ecb780 A64: Handle half-precision floating point in scalar FABS
Now that we have the half-precision variant of the opcode added, we can
simply handle the instruction instead of treating it as undefined.
2020-04-22 20:58:12 +01:00
Lioncash
fac9224d5e A64: Handle half-precision floating point in FCVTN
Now that we have IR instructions for performing conversions with
half-precision floating point, we can also handle half-precision values
within FCVTN.
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
16de99d3e3 A64: Enable FCVT floating-point conversions for half-precision
With this, we no longer have to fall back to the interpreter in any of
the FCVT floating-point conversion instructions.
2020-04-22 20:58:12 +01:00
Lioncash
10abc77fad A64: Handle half-precision floating point in scalar FNEG
With the half-precision variant of the FPNeg opcode added, we can
utilize it here to emulate the half-precision variant of FNEG.
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
dff5da1063 common/fp/unpacked: Amend behavior of FPUnpackCV
This is supposed to call FPUnpackBase instead of FPUnpack. This would
result in alternate half-precision representations being misinterpreted
when it comes to dealing with NaNs.
2020-04-22 20:58:12 +01:00
Merry
f01afc5ae6 Merge pull request #456 from lioncash/mov
A64: Enable FMOV (general) for half-precision floating point
2020-04-22 20:58:12 +01:00
Lioncash
03bc2334fe common/fp/op/FPConvert: Amend off-by one in double NaN case in FPConvertNaN
Avoids potentially clobbering the intended sign bit value during
conversions to double-precision values. The other conversion types are
already properly handled, so those don't need to be addressed.
2020-04-22 20:58:12 +01:00
Lioncash
c57b146fb2 common/fp/op/FPConvert: Add half-precision instantiations to FPConvert 2020-04-22 20:58:12 +01:00
Merry
c1ce94872d Merge pull request #455 from lioncash/sqrdmulh-scalar
A64: Implement SQRDMULH and SQDMULL's scalar indexed variants
2020-04-22 20:58:11 +01:00
Lioncash
25a7256ee1 A64: Enable FMOV (general) for half-precision floating point
This just transfers values between vector registers and general-purpose
registers with no conversions performed, so this is trivial to add
support for half-precision to.
2020-04-22 20:58:11 +01:00
Lioncash
97dd3d0596 A64: Implement SQRDMULH's scalar indexed element variant 2020-04-22 20:58:11 +01:00
Lioncash
49b51e34f1 simd_vector_x_indexed_element: Deduplicate index and Vm operand construction 2020-04-22 20:58:11 +01:00
Lioncash
692aba91b6 A64: Implement SQDMULL{2}'s scalar indexed element variant 2020-04-22 20:58:11 +01:00
Lioncash
c043b831d5 A64: Implement SQDMULL{2}'s by-element variant 2020-04-22 20:58:11 +01:00
Lioncash
72af5a3dff simd_scalar_x_indexed_element: Factor out index and Vm argument construction
This will be useful in the implementations of SQRDMULH and SQDMULL{2} as
well.
2020-04-22 20:58:11 +01:00
Lioncash
224ff0afaa A64: Implement SQRDMULH's by-index vector variant 2020-04-22 20:58:11 +01:00
Lioncash
3a3542414b A64: Implement FRECPX's half-precision floating point variant 2020-04-22 20:58:11 +01:00
Lioncash
bd892ec4ef frontend/ir/ir_emitter: Amend FPRecipExponent to handle half-precision floating point 2020-04-22 20:58:11 +01:00
Lioncash
974fbf0677 frontend/ir/value: Add U16U32U64 type to represent floating point types 2020-04-22 20:58:11 +01:00
Lioncash
eb3e0d5908 common/fp/op/FPRecipExponent: Add half-precision floating point specialization 2020-04-22 20:58:11 +01:00
Lioncash
a829c93406 common/fp/unpacked: Correct edge-cases within FPUnpack for half-precision floating point
This corrects one case where floating-point exceptions could be set when
they're not supposed to be.

This also corrects a case where values were being treated as NaNs when
they weren't supposed to be.
2020-04-22 20:58:11 +01:00
Lioncash
7030b9af95 common/fp/process_nan: Add half-precision instantiations for NaN processing functions 2020-04-22 20:58:11 +01:00
Lioncash
14f55d7476 common/fp/unpacked: Add half-precision instantiation of FPRoundBase 2020-04-22 20:58:11 +01:00
Lioncash
7e814de445 common/fp/unpacked: Handle half-precision unpacking in FPUnpackBase 2020-04-22 20:58:11 +01:00
Lioncash
8f9fe8690a common/fp/unpacked: Adjust FPUnpack to operate like ARM pseudocode
This function is defined as always disabling the AHP bit in the fpcr
before performing any operations.

At the same time, rename the original FPUnpack function to FPUnpackBase
to match the pseudocode in the ARM reference manual.
2020-04-22 20:58:11 +01:00
Merry
37c4c39d62 Merge pull request #448 from lioncash/saturate
A64: Implement SQSHRN, SQSHRUN, and UQSHRN's scalar variants
2020-04-22 20:58:11 +01:00
Merry
f5d774bdbd Merge pull request #449 from lioncash/hp
common/fp/info: Add specialization of FPInfo for half-precision floating point
2020-04-22 20:58:11 +01:00
Lioncash
126c29a9e9 A64: Implement SQSHRN, SQSHRUN, and UQSHRN's scalar variants
These can just be implemented in terms of the vector variants for the
time being.
2020-04-22 20:58:11 +01:00
Lioncash
0b67b94b6c common/fp/info: Add specialization of FPInfo for half-precision floating point
Puts the necessary info struct in place for further use.
2020-04-22 20:58:11 +01:00
Lioncash
dd7433f9d3 A64: Amend prototypes of some SIMD scalar shift by immediate opcodes
These take a vector for a destination.
2020-04-22 20:58:11 +01:00
Lioncash
99c494bae9 common/fp/unpacked: Add FPRoundCV
Corresponds to the equivalent pseudocode within the ARMv8 reference
manual. This will be necessary for supporting half-precision
floating-point.

This also makes use of it within FPConvert
2020-04-22 20:58:11 +01:00
Merry
bbd5330ad2 Merge pull request #447 from lioncash/flag
A64: Implement CFINV, RMIF, AXFlag and XAFlag
2020-04-22 20:58:11 +01:00
Lioncash
490bebbd9a common/fp/unpacked: Add FPUnpackCV
Adds a template function that performs the same behavior as in the ARM
pseudocode, and utilizes it in FPConvert, which will be necessary for
half-float support.
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
6aed4036ef ir_opt/a64_get_set_elimination_pass: Add handling for NZCV raw get and set operations 2020-04-22 20:58:11 +01:00
Merry
4f937c1ee1 Merge pull request #446 from lioncash/sqshl
A64: Implement scalar variants of SQSHL (register) and UQSHL (register)
2020-04-22 20:58:11 +01:00
Lioncash
aa22db534b A64: Implement AXFlag and XAFlag 2020-04-22 20:58:11 +01:00
Merry
d74cccbc84 Merge pull request #445 from lioncash/sqrt
A64: Implement single and double-precision vector variant of FSQRT
2020-04-22 20:58:11 +01:00
Lioncash
20ffe568d0 A64: Implement RMIF 2020-04-22 20:58:11 +01:00
Merry
6d7e7c3269 Merge pull request #443 from lioncash/flag
A64: Rearrange flag format/manipulation instructions
2020-04-22 20:58:11 +01:00
Lioncash
51b526e453 A64: Implement CFINV 2020-04-22 20:58:11 +01:00
Merry
5d01f1b462 Merge pull request #441 from lioncash/constexpr
common/bit_util: Mark a few functions as constexpr
2020-04-22 20:58:11 +01:00
Lioncash
597a8be5d5 ir: Add A64-specific opcodes for getting and setting raw NZCV values
This will be necessary to implement the flag manipulation and flag
format instructions.
2020-04-22 20:58:11 +01:00
Merry
743c52fdc5 Merge pull request #440 from lioncash/include
common/fp: Remove unnecessary includes
2020-04-22 20:58:11 +01:00
Lioncash
d3515279df A64: Implement the vector version of FCVTXN 2020-04-22 20:58:10 +01:00
Lioncash
17aea0b997 A64: Implement UQSHL (register)'s scalar variant
This can be implemented in terms of the vector variant.
2020-04-22 20:58:10 +01:00
Lioncash
c99d4b762e A64: Implement single and double-precision vector variant of FSQRT 2020-04-22 20:58:10 +01:00
Lioncash
54e0b487f3 A64: Rearrange flag format/manipulation instructions
Gives these instructions better categorical labeling.
2020-04-22 20:58:10 +01:00
Lioncash
88d1977cb9 common/bit_util: Make a few functions as constexpr
These four functions can be made constexpr with no issue.
2020-04-22 20:58:10 +01:00
Lioncash
f33e5939b7 common/fp: Remove unnecessary includes 2020-04-22 20:58:10 +01:00
Lioncash
302f56b36a A64: Fall back to interpreting for FCADD and FCMLA half-precision variants
Rather than straight-up treating them as undefined, we can fall back to an
interpreter in this case.
2020-04-22 20:58:10 +01:00
Lioncash
4339a8fff6 A64: Implement the scalar version of FCVTXN 2020-04-22 20:58:10 +01:00
Lioncash
35ddf68ad5 A64: Implement SQSHL (register)'s scalar variant
We can implement this in terms of the vector variant.
2020-04-22 20:58:10 +01:00
Lioncash
5cf1478620 frontend/ir: Add opcodes for vector square roots 2020-04-22 20:58:10 +01:00
Lioncash
36027ebef5 frontend/ir/microinstruction: Add missing cases for FPRecipExponent{32,64} for ReadsFromAndWritesToFPSRCumulativeExceptionBits()
This was intended to be added within #437, but was missed
2020-04-22 20:58:10 +01:00
Merry
40b081438a Merge pull request #439 from lioncash/fcmla
A64: Implement FCADD and FCMLA
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
Merry
d91192681a Merge pull request #438 from lioncash/fmulx
A64: Implement scalar double/single precision FMULX (by element)
2020-04-22 20:58:10 +01:00
Lioncash
ed29ef8cca A64: Implement FCMLA 2020-04-22 20:58:10 +01:00
Lioncash
95af9dafbe common/fp/op: Add FP conversion functions 2020-04-22 20:58:10 +01:00
Merry
9f11720a69 Merge pull request #437 from lioncash/frecpx
A64: Implement FRECPX (single, double precision)
2020-04-22 20:58:10 +01:00
Lioncash
bdcea0b0dc A64: Implement scalar double/single precision FMULX (by element) 2020-04-22 20:58:10 +01:00
Lioncash
5ce17574f9 A64: Implement FCADD 2020-04-22 20:58:10 +01:00
Merry
34d917f34e Merge pull request #436 from lioncash/no-alloc
A64: Implement LDNP/STNP
2020-04-22 20:58:10 +01:00
Lioncash
e44730ba6d A64: Implement FRECPX (single, double precision) 2020-04-22 20:58:10 +01:00
Lioncash
bfaeb08d3c A64: Implement LDNP/STNP
LDNP and STNP indicate that a memory access is non-temporal/streaming
(i.e. unlikely to be repeated), allowing data caching to not be
performed. However, given this is only a hint, we can treat these two
instructions as regular LDP and STP instructions for the time being.
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
Merry
dbf47db713 Merge pull request #434 from lioncash/format
A32/translate_arm: Formatting/tidying up
2020-04-22 20:58:10 +01:00
Lioncash
b168c2a9f9 common/fp/op: Add operations for floating-point reciprocal exponents 2020-04-22 20:58:10 +01:00
Lioncash
05a6ab691d translate_arm/coprocessor: Minor tidying up 2020-04-22 20:58:10 +01:00
Lioncash
1e32a09c03 translate_arm/vfp2: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
e209b31073 translate_arm/synchronization: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
9514e3602e translate_arm/status_register_access: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
c6aa1a708a translate_arm/saturated: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
a72813599a translate_arm/reversal: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
7be56e6b67 translate_arm/parallel: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
3c00a616d6 translate_arm/packing: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
c711188f46 translate_arm/multiply: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
c8dad40d81 translate_arm/misc: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
a7bf5ff77d translate_arm/load_store: Invert conditionals where applicable 2020-04-22 20:58:10 +01:00
Lioncash
2e180a7f14 backend/x64/a32_interface: Mark Context move constructor and move assignment as noexcept
Provides a more "correct" move constructor/assignment operator, since
these relevant functions shouldn't throw exceptions.

Has the benefit of playing nicely with std::move_if_noexcept and other
noexcept library facilities.
2020-04-22 20:58:09 +01:00
Lioncash
f4b19a7393 translate_arm/extension: Invert conditionals where applicable 2020-04-22 20:58:09 +01:00
Lioncash
deb9dd4acc block_of_code: Replace cast with [[maybe_unused]] in DoesCpuSupport() 2020-04-22 20:58:09 +01:00
Lioncash
c2de6ecfd0 translate_arm/exception_generating: Invert conditionals where applicable 2020-04-22 20:58:09 +01:00
Lioncash
d8a8d3b073 translate_arm/data_processing: Invert conditionals where applicable 2020-04-22 20:58:09 +01:00
Lioncash
df5c51ff47 translate_arm/branch: Invert conditionals where applicable
Allows unindenting code a bit.
2020-04-22 20:58:09 +01:00
Lioncash
3290a9fdc2 common: Remove address_range.h
The AddressRange structure isn't used anywhere within the codebase, so
this can be removed. Particularly because there's no real appeal/heavy
potential use of it in the future that isn't trivial to add back if
needed.
2020-04-22 20:57:38 +01:00
Lioncash
ee973f13c7 frontend/A32/ir_emitter: Mark PC() and AlignPC() as const-qualified member functions
These don't modify instance state, so they can be const-qualified member
functions.
2020-04-22 20:57:38 +01:00
Lioncash
3a2dd09122 frontend/A64/ir_emitter: Mark PC() and AlignPC() as const qualified member functions
These don't actually alter any instance state.
2020-04-22 20:57:38 +01:00
Lioncash
575ae852a9 constant_propagation_pass: Fold byte reversal opcodes where applicable
These are reasonably trivial to fold away when applicable. We just
perform the swap and replace the instruction with the constant value.
2020-04-22 20:57:37 +01:00
Merry
2c53f354ab Merge pull request #418 from lioncash/fold-op
constant_propagation_pass: Handle folding for Least/MostSignificant{Bit, Byte, Half, Word} opcodes
2020-04-22 20:57:37 +01:00
Merry
ad14a33672 Merge pull request #417 from lioncash/swap
common: Move byte swapping functions to bit_utils.h
2020-04-22 20:57:37 +01:00
Lioncash
d302d9bd0c constant_propagation_pass: Handle folding for Least/MostSignificant{Bit, Byte, Half, Word} opcodes
These are quite trivial to fold.
2020-04-22 20:57:37 +01:00
Lioncash
7139942976 common: Move byte swapping functions to bit_utils.h
These are quite general functions, so they can just be moved into common
instead of recreating a namespace here.
2020-04-22 20:57:37 +01:00
MerryMage
7c8fcaef26 emit_x64_vector_floating_point: AVX && DN implementation of EmitFPVectorMulX 2020-04-22 20:57:37 +01:00