[Reland][Runtimes] Merge 'compile_commands.json' files from runtimes build (#116303)
[llvm-project.git] / flang / lib / Evaluate / intrinsics-library.cpp
blobc1b270f518c0e0a6a22d7468477194c34f6fc730
1 //===-- lib/Evaluate/intrinsics-library.cpp -------------------------------===//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
9 // This file defines host runtime functions that can be used for folding
10 // intrinsic functions.
11 // The default host runtime folders are built with <cmath> and
12 // <complex> functions that are guaranteed to exist from the C++ standard.
14 #include "flang/Evaluate/intrinsics-library.h"
15 #include "fold-implementation.h"
16 #include "host.h"
17 #include "flang/Common/erfc-scaled.h"
18 #include "flang/Common/idioms.h"
19 #include "flang/Common/static-multimap-view.h"
20 #include "flang/Evaluate/expression.h"
21 #include <cfloat>
22 #include <cmath>
23 #include <complex>
24 #include <functional>
25 #if HAS_QUADMATHLIB
26 #include "quadmath.h"
27 #endif
28 #include "flang/Common/float128.h"
29 #include "flang/Common/float80.h"
30 #include <type_traits>
32 namespace Fortran::evaluate {
34 // Define a vector like class that can hold an arbitrary number of
35 // Dynamic type and be built at compile time. This is like a
36 // std::vector<DynamicType>, but constexpr only.
37 template <typename... FortranType> struct TypeVectorStorage {
38 static constexpr DynamicType values[]{FortranType{}.GetType()...};
39 static constexpr const DynamicType *start{&values[0]};
40 static constexpr const DynamicType *end{start + sizeof...(FortranType)};
42 template <> struct TypeVectorStorage<> {
43 static constexpr const DynamicType *start{nullptr}, *end{nullptr};
45 struct TypeVector {
46 template <typename... FortranType> static constexpr TypeVector Create() {
47 using storage = TypeVectorStorage<FortranType...>;
48 return TypeVector{storage::start, storage::end, sizeof...(FortranType)};
50 constexpr size_t size() const { return size_; };
51 using const_iterator = const DynamicType *;
52 constexpr const_iterator begin() const { return startPtr; }
53 constexpr const_iterator end() const { return endPtr; }
54 const DynamicType &operator[](size_t i) const { return *(startPtr + i); }
56 const DynamicType *startPtr{nullptr};
57 const DynamicType *endPtr{nullptr};
58 const size_t size_;
60 inline bool operator==(
61 const TypeVector &lhs, const std::vector<DynamicType> &rhs) {
62 if (lhs.size() != rhs.size()) {
63 return false;
65 for (size_t i{0}; i < lhs.size(); ++i) {
66 if (lhs[i] != rhs[i]) {
67 return false;
70 return true;
73 // HostRuntimeFunction holds a pointer to a Folder function that can fold
74 // a Fortran scalar intrinsic using host runtime functions (e.g libm).
75 // The folder take care of all conversions between Fortran types and the related
76 // host types as well as setting and cleaning-up the floating point environment.
77 // HostRuntimeFunction are intended to be built at compile time (members are all
78 // constexpr constructible) so that they can be stored in a compile time static
79 // map.
80 struct HostRuntimeFunction {
81 using Folder = Expr<SomeType> (*)(
82 FoldingContext &, std::vector<Expr<SomeType>> &&);
83 using Key = std::string_view;
84 // Needed for implicit compare with keys.
85 constexpr operator Key() const { return key; }
86 // Name of the related Fortran intrinsic.
87 Key key;
88 // DynamicType of the Expr<SomeType> returns by folder.
89 DynamicType resultType;
90 // DynamicTypes expected for the Expr<SomeType> arguments of the folder.
91 // The folder will crash if provided arguments of different types.
92 TypeVector argumentTypes;
93 // Folder to be called to fold the intrinsic with host runtime. The provided
94 // Expr<SomeType> arguments must wrap scalar constants of the type described
95 // in argumentTypes, otherwise folder will crash. Any floating point issue
96 // raised while executing the host runtime will be reported in FoldingContext
97 // messages.
98 Folder folder;
101 // Translate a host function type signature (template arguments) into a
102 // constexpr data representation based on Fortran DynamicType that can be
103 // stored.
104 template <typename TR, typename... TA> using FuncPointer = TR (*)(TA...);
105 template <typename T> struct FuncTypeAnalyzer {};
106 template <typename HostTR, typename... HostTA>
107 struct FuncTypeAnalyzer<FuncPointer<HostTR, HostTA...>> {
108 static constexpr DynamicType result{host::FortranType<HostTR>{}.GetType()};
109 static constexpr TypeVector arguments{
110 TypeVector::Create<host::FortranType<HostTA>...>()};
113 // Define helpers to deal with host floating environment.
114 template <typename TR>
115 static void CheckFloatingPointIssues(
116 host::HostFloatingPointEnvironment &hostFPE, const Scalar<TR> &x) {
117 if constexpr (TR::category == TypeCategory::Complex ||
118 TR::category == TypeCategory::Real) {
119 if (x.IsNotANumber()) {
120 hostFPE.SetFlag(RealFlag::InvalidArgument);
121 } else if (x.IsInfinite()) {
122 hostFPE.SetFlag(RealFlag::Overflow);
126 // Software Subnormal Flushing helper.
127 // Only flush floating-points. Forward other scalars untouched.
128 // Software flushing is only performed if hardware flushing is not available
129 // because it may not result in the same behavior as hardware flushing.
130 // Some runtime implementations are "working around" subnormal flushing to
131 // return results that they deem better than returning the result they would
132 // with a null argument. An example is logf that should return -inf if arguments
133 // are flushed to zero, but some implementations return -1.03972076416015625e2_4
134 // for all subnormal values instead. It is impossible to reproduce this with the
135 // simple software flushing below.
136 template <typename T>
137 static constexpr inline const Scalar<T> FlushSubnormals(Scalar<T> &&x) {
138 if constexpr (T::category == TypeCategory::Real ||
139 T::category == TypeCategory::Complex) {
140 return x.FlushSubnormalToZero();
142 return x;
145 // This is the kernel called by all HostRuntimeFunction folders, it convert the
146 // Fortran Expr<SomeType> to the host runtime function argument types, calls
147 // the runtime function, and wrap back the result into an Expr<SomeType>.
148 // It deals with host floating point environment set-up and clean-up.
149 template <typename FuncType, typename TR, typename... TA, size_t... I>
150 static Expr<SomeType> ApplyHostFunctionHelper(FuncType func,
151 FoldingContext &context, std::vector<Expr<SomeType>> &&args,
152 std::index_sequence<I...>) {
153 host::HostFloatingPointEnvironment hostFPE;
154 hostFPE.SetUpHostFloatingPointEnvironment(context);
155 host::HostType<TR> hostResult{};
156 Scalar<TR> result{};
157 std::tuple<Scalar<TA>...> scalarArgs{
158 GetScalarConstantValue<TA>(args[I]).value()...};
159 if (context.targetCharacteristics().areSubnormalsFlushedToZero() &&
160 !hostFPE.hasSubnormalFlushingHardwareControl()) {
161 hostResult = func(host::CastFortranToHost<TA>(
162 FlushSubnormals<TA>(std::move(std::get<I>(scalarArgs))))...);
163 result = FlushSubnormals<TR>(host::CastHostToFortran<TR>(hostResult));
164 } else {
165 hostResult = func(host::CastFortranToHost<TA>(std::get<I>(scalarArgs))...);
166 result = host::CastHostToFortran<TR>(hostResult);
168 if (!hostFPE.hardwareFlagsAreReliable()) {
169 CheckFloatingPointIssues<TR>(hostFPE, result);
171 hostFPE.CheckAndRestoreFloatingPointEnvironment(context);
172 return AsGenericExpr(Constant<TR>(std::move(result)));
174 template <typename HostTR, typename... HostTA>
175 Expr<SomeType> ApplyHostFunction(FuncPointer<HostTR, HostTA...> func,
176 FoldingContext &context, std::vector<Expr<SomeType>> &&args) {
177 return ApplyHostFunctionHelper<decltype(func), host::FortranType<HostTR>,
178 host::FortranType<HostTA>...>(
179 func, context, std::move(args), std::index_sequence_for<HostTA...>{});
182 // FolderFactory builds a HostRuntimeFunction for the host runtime function
183 // passed as a template argument.
184 // Its static member function "fold" is the resulting folder. It captures the
185 // host runtime function pointer and pass it to the host runtime function folder
186 // kernel.
187 template <typename HostFuncType, HostFuncType func> class FolderFactory {
188 public:
189 static constexpr HostRuntimeFunction Create(const std::string_view &name) {
190 return HostRuntimeFunction{name, FuncTypeAnalyzer<HostFuncType>::result,
191 FuncTypeAnalyzer<HostFuncType>::arguments, &Fold};
194 private:
195 static Expr<SomeType> Fold(
196 FoldingContext &context, std::vector<Expr<SomeType>> &&args) {
197 return ApplyHostFunction(func, context, std::move(args));
201 // Define host runtime libraries that can be used for folding and
202 // fill their description if they are available.
203 enum class LibraryVersion {
204 Libm,
205 LibmExtensions,
206 PgmathFast,
207 PgmathRelaxed,
208 PgmathPrecise
210 template <typename HostT, LibraryVersion> struct HostRuntimeLibrary {
211 // When specialized, this class holds a static constexpr table containing
212 // all the HostRuntimeLibrary for functions of library LibraryVersion
213 // that returns a value of type HostT.
216 using HostRuntimeMap = common::StaticMultimapView<HostRuntimeFunction>;
218 // Map numerical intrinsic to <cmath>/<complex> functions
219 // (Note: ABS() is folded in fold-real.cpp.)
220 template <typename HostT>
221 struct HostRuntimeLibrary<HostT, LibraryVersion::Libm> {
222 using F = FuncPointer<HostT, HostT>;
223 using F2 = FuncPointer<HostT, HostT, HostT>;
224 static constexpr HostRuntimeFunction table[]{
225 FolderFactory<F, F{std::acos}>::Create("acos"),
226 FolderFactory<F, F{std::acosh}>::Create("acosh"),
227 FolderFactory<F, F{std::asin}>::Create("asin"),
228 FolderFactory<F, F{std::asinh}>::Create("asinh"),
229 FolderFactory<F, F{std::atan}>::Create("atan"),
230 FolderFactory<F2, F2{std::atan2}>::Create("atan2"),
231 FolderFactory<F, F{std::atanh}>::Create("atanh"),
232 FolderFactory<F, F{std::cos}>::Create("cos"),
233 FolderFactory<F, F{std::cosh}>::Create("cosh"),
234 FolderFactory<F, F{std::erf}>::Create("erf"),
235 FolderFactory<F, F{std::erfc}>::Create("erfc"),
236 FolderFactory<F, F{common::ErfcScaled}>::Create("erfc_scaled"),
237 FolderFactory<F, F{std::exp}>::Create("exp"),
238 FolderFactory<F, F{std::tgamma}>::Create("gamma"),
239 FolderFactory<F, F{std::log}>::Create("log"),
240 FolderFactory<F, F{std::log10}>::Create("log10"),
241 FolderFactory<F, F{std::lgamma}>::Create("log_gamma"),
242 FolderFactory<F2, F2{std::pow}>::Create("pow"),
243 FolderFactory<F, F{std::sin}>::Create("sin"),
244 FolderFactory<F, F{std::sinh}>::Create("sinh"),
245 FolderFactory<F, F{std::tan}>::Create("tan"),
246 FolderFactory<F, F{std::tanh}>::Create("tanh"),
248 // Note: cmath does not have modulo and erfc_scaled equivalent
250 // Note regarding lack of bessel function support:
251 // C++17 defined standard Bessel math functions std::cyl_bessel_j
252 // and std::cyl_neumann that can be used for Fortran j and y
253 // bessel functions. However, they are not yet implemented in
254 // clang libc++ (ok in GNU libstdc++). C maths functions j0...
255 // are not C standard but a GNU extension so they are not used
256 // to avoid introducing incompatibilities.
257 // Use libpgmath to get bessel function folding support.
258 // TODO: Add Bessel functions when possible.
259 static constexpr HostRuntimeMap map{table};
260 static_assert(map.Verify(), "map must be sorted");
263 // Helpers to map complex std::pow whose resolution in F2{std::pow} is
264 // ambiguous as of clang++ 20.
265 template <typename HostT>
266 static std::complex<HostT> StdPowF2(
267 const std::complex<HostT> &x, const std::complex<HostT> &y) {
268 return std::pow(x, y);
270 template <typename HostT>
271 static std::complex<HostT> StdPowF2A(
272 const HostT &x, const std::complex<HostT> &y) {
273 return std::pow(x, y);
275 template <typename HostT>
276 static std::complex<HostT> StdPowF2B(
277 const std::complex<HostT> &x, const HostT &y) {
278 return std::pow(x, y);
281 #ifdef _AIX
282 #ifdef __clang_major__
283 #pragma clang diagnostic ignored "-Wc99-extensions"
284 #endif
286 extern "C" {
287 float _Complex cacosf(float _Complex);
288 double _Complex cacos(double _Complex);
289 float _Complex csqrtf(float _Complex);
290 double _Complex csqrt(double _Complex);
293 enum CRI { Real, Imag };
294 template <typename TR, typename TA> static TR &reIm(TA &x, CRI n) {
295 return reinterpret_cast<TR(&)[2]>(x)[n];
297 template <typename TR, typename T> static TR CppToC(const std::complex<T> &x) {
298 TR r;
299 reIm<T, TR>(r, CRI::Real) = x.real();
300 reIm<T, TR>(r, CRI::Imag) = x.imag();
301 return r;
303 template <typename T, typename TA> static std::complex<T> CToCpp(const TA &x) {
304 TA &z{const_cast<TA &>(x)};
305 return std::complex<T>(reIm<T, TA>(z, CRI::Real), reIm<T, TA>(z, CRI::Imag));
307 #endif
309 template <typename HostT>
310 static std::complex<HostT> CSqrt(const std::complex<HostT> &x) {
311 std::complex<HostT> res;
312 #ifdef _AIX
313 // On AIX, the implementation of csqrt[f] and std::sqrt is different,
314 // use csqrt[f] in folding.
315 if constexpr (std::is_same_v<HostT, float>) {
316 float _Complex r{csqrtf(CppToC<float _Complex, float>(x))};
317 res = CToCpp<float, float _Complex>(r);
318 } else if constexpr (std::is_same_v<HostT, double>) {
319 double _Complex r{csqrt(CppToC<double _Complex, double>(x))};
320 res = CToCpp<double, double _Complex>(r);
321 } else {
322 DIE("bad complex component type");
324 #else
325 res = std::sqrt(x);
326 #endif
327 return res;
330 template <typename HostT>
331 static std::complex<HostT> CAcos(const std::complex<HostT> &x) {
332 std::complex<HostT> res;
333 #ifdef _AIX
334 // On AIX, the implementation of cacos[f] and std::acos is different,
335 // use cacos[f] in folding.
336 if constexpr (std::is_same_v<HostT, float>) {
337 float _Complex r{cacosf(CppToC<float _Complex, float>(x))};
338 res = CToCpp<float, float _Complex>(r);
339 } else if constexpr (std::is_same_v<HostT, double>) {
340 double _Complex r{cacos(CppToC<double _Complex, double>(x))};
341 res = CToCpp<double, double _Complex>(r);
342 } else {
343 DIE("bad complex component type");
345 #else
346 res = std::acos(x);
347 #endif
348 return res;
351 template <typename HostT>
352 struct HostRuntimeLibrary<std::complex<HostT>, LibraryVersion::Libm> {
353 using F = FuncPointer<std::complex<HostT>, const std::complex<HostT> &>;
354 using F2 = FuncPointer<std::complex<HostT>, const std::complex<HostT> &,
355 const std::complex<HostT> &>;
356 using F2A = FuncPointer<std::complex<HostT>, const HostT &,
357 const std::complex<HostT> &>;
358 using F2B = FuncPointer<std::complex<HostT>, const std::complex<HostT> &,
359 const HostT &>;
360 static constexpr HostRuntimeFunction table[]{
361 FolderFactory<F, F{CAcos}>::Create("acos"),
362 FolderFactory<F, F{std::acosh}>::Create("acosh"),
363 FolderFactory<F, F{std::asin}>::Create("asin"),
364 FolderFactory<F, F{std::asinh}>::Create("asinh"),
365 FolderFactory<F, F{std::atan}>::Create("atan"),
366 FolderFactory<F, F{std::atanh}>::Create("atanh"),
367 FolderFactory<F, F{std::cos}>::Create("cos"),
368 FolderFactory<F, F{std::cosh}>::Create("cosh"),
369 FolderFactory<F, F{std::exp}>::Create("exp"),
370 FolderFactory<F, F{std::log}>::Create("log"),
371 FolderFactory<F2, F2{StdPowF2}>::Create("pow"),
372 FolderFactory<F2A, F2A{StdPowF2A}>::Create("pow"),
373 FolderFactory<F2B, F2B{StdPowF2B}>::Create("pow"),
374 FolderFactory<F, F{std::sin}>::Create("sin"),
375 FolderFactory<F, F{std::sinh}>::Create("sinh"),
376 FolderFactory<F, F{CSqrt}>::Create("sqrt"),
377 FolderFactory<F, F{std::tan}>::Create("tan"),
378 FolderFactory<F, F{std::tanh}>::Create("tanh"),
380 static constexpr HostRuntimeMap map{table};
381 static_assert(map.Verify(), "map must be sorted");
383 // Note regarding cmath:
384 // - cmath does not have modulo and erfc_scaled equivalent
385 // - C++17 defined standard Bessel math functions std::cyl_bessel_j
386 // and std::cyl_neumann that can be used for Fortran j and y
387 // bessel functions. However, they are not yet implemented in
388 // clang libc++ (ok in GNU libstdc++). Instead, the Posix libm
389 // extensions are used when available below.
391 #if _POSIX_C_SOURCE >= 200112L || _XOPEN_SOURCE >= 600
392 /// Define libm extensions
393 /// Bessel functions are defined in POSIX.1-2001.
395 // Remove float bessel functions for AIX and Darwin as they are not supported
396 #if !defined(_AIX) && !defined(__APPLE__)
397 template <> struct HostRuntimeLibrary<float, LibraryVersion::LibmExtensions> {
398 using F = FuncPointer<float, float>;
399 using FN = FuncPointer<float, int, float>;
400 static constexpr HostRuntimeFunction table[]{
401 FolderFactory<F, F{::j0f}>::Create("bessel_j0"),
402 FolderFactory<F, F{::j1f}>::Create("bessel_j1"),
403 FolderFactory<FN, FN{::jnf}>::Create("bessel_jn"),
404 FolderFactory<F, F{::y0f}>::Create("bessel_y0"),
405 FolderFactory<F, F{::y1f}>::Create("bessel_y1"),
406 FolderFactory<FN, FN{::ynf}>::Create("bessel_yn"),
408 static constexpr HostRuntimeMap map{table};
409 static_assert(map.Verify(), "map must be sorted");
411 #endif
413 #if HAS_QUADMATHLIB
414 template <> struct HostRuntimeLibrary<__float128, LibraryVersion::Libm> {
415 using F = FuncPointer<__float128, __float128>;
416 using F2 = FuncPointer<__float128, __float128, __float128>;
417 using FN = FuncPointer<__float128, int, __float128>;
418 static constexpr HostRuntimeFunction table[]{
419 FolderFactory<F, F{::acosq}>::Create("acos"),
420 FolderFactory<F, F{::acoshq}>::Create("acosh"),
421 FolderFactory<F, F{::asinq}>::Create("asin"),
422 FolderFactory<F, F{::asinhq}>::Create("asinh"),
423 FolderFactory<F, F{::atanq}>::Create("atan"),
424 FolderFactory<F2, F2{::atan2q}>::Create("atan2"),
425 FolderFactory<F, F{::atanhq}>::Create("atanh"),
426 FolderFactory<F, F{::j0q}>::Create("bessel_j0"),
427 FolderFactory<F, F{::j1q}>::Create("bessel_j1"),
428 FolderFactory<FN, FN{::jnq}>::Create("bessel_jn"),
429 FolderFactory<F, F{::y0q}>::Create("bessel_y0"),
430 FolderFactory<F, F{::y1q}>::Create("bessel_y1"),
431 FolderFactory<FN, FN{::ynq}>::Create("bessel_yn"),
432 FolderFactory<F, F{::cosq}>::Create("cos"),
433 FolderFactory<F, F{::coshq}>::Create("cosh"),
434 FolderFactory<F, F{::erfq}>::Create("erf"),
435 FolderFactory<F, F{::erfcq}>::Create("erfc"),
436 FolderFactory<F, F{::expq}>::Create("exp"),
437 FolderFactory<F, F{::tgammaq}>::Create("gamma"),
438 FolderFactory<F, F{::logq}>::Create("log"),
439 FolderFactory<F, F{::log10q}>::Create("log10"),
440 FolderFactory<F, F{::lgammaq}>::Create("log_gamma"),
441 FolderFactory<F2, F2{::powq}>::Create("pow"),
442 FolderFactory<F, F{::sinq}>::Create("sin"),
443 FolderFactory<F, F{::sinhq}>::Create("sinh"),
444 FolderFactory<F, F{::tanq}>::Create("tan"),
445 FolderFactory<F, F{::tanhq}>::Create("tanh"),
447 static constexpr HostRuntimeMap map{table};
448 static_assert(map.Verify(), "map must be sorted");
450 template <> struct HostRuntimeLibrary<__complex128, LibraryVersion::Libm> {
451 using F = FuncPointer<__complex128, __complex128>;
452 using F2 = FuncPointer<__complex128, __complex128, __complex128>;
453 static constexpr HostRuntimeFunction table[]{
454 FolderFactory<F, F{::cacosq}>::Create("acos"),
455 FolderFactory<F, F{::cacoshq}>::Create("acosh"),
456 FolderFactory<F, F{::casinq}>::Create("asin"),
457 FolderFactory<F, F{::casinhq}>::Create("asinh"),
458 FolderFactory<F, F{::catanq}>::Create("atan"),
459 FolderFactory<F, F{::catanhq}>::Create("atanh"),
460 FolderFactory<F, F{::ccosq}>::Create("cos"),
461 FolderFactory<F, F{::ccoshq}>::Create("cosh"),
462 FolderFactory<F, F{::cexpq}>::Create("exp"),
463 FolderFactory<F, F{::clogq}>::Create("log"),
464 FolderFactory<F2, F2{::cpowq}>::Create("pow"),
465 FolderFactory<F, F{::csinq}>::Create("sin"),
466 FolderFactory<F, F{::csinhq}>::Create("sinh"),
467 FolderFactory<F, F{::csqrtq}>::Create("sqrt"),
468 FolderFactory<F, F{::ctanq}>::Create("tan"),
469 FolderFactory<F, F{::ctanhq}>::Create("tanh"),
471 static constexpr HostRuntimeMap map{table};
472 static_assert(map.Verify(), "map must be sorted");
474 #endif
476 template <> struct HostRuntimeLibrary<double, LibraryVersion::LibmExtensions> {
477 using F = FuncPointer<double, double>;
478 using FN = FuncPointer<double, int, double>;
479 static constexpr HostRuntimeFunction table[]{
480 FolderFactory<F, F{::j0}>::Create("bessel_j0"),
481 FolderFactory<F, F{::j1}>::Create("bessel_j1"),
482 FolderFactory<FN, FN{::jn}>::Create("bessel_jn"),
483 FolderFactory<F, F{::y0}>::Create("bessel_y0"),
484 FolderFactory<F, F{::y1}>::Create("bessel_y1"),
485 FolderFactory<FN, FN{::yn}>::Create("bessel_yn"),
487 static constexpr HostRuntimeMap map{table};
488 static_assert(map.Verify(), "map must be sorted");
491 #if defined(__GLIBC__) && (HAS_FLOAT80 || HAS_LDBL128)
492 template <>
493 struct HostRuntimeLibrary<long double, LibraryVersion::LibmExtensions> {
494 using F = FuncPointer<long double, long double>;
495 using FN = FuncPointer<long double, int, long double>;
496 static constexpr HostRuntimeFunction table[]{
497 FolderFactory<F, F{::j0l}>::Create("bessel_j0"),
498 FolderFactory<F, F{::j1l}>::Create("bessel_j1"),
499 FolderFactory<FN, FN{::jnl}>::Create("bessel_jn"),
500 FolderFactory<F, F{::y0l}>::Create("bessel_y0"),
501 FolderFactory<F, F{::y1l}>::Create("bessel_y1"),
502 FolderFactory<FN, FN{::ynl}>::Create("bessel_yn"),
504 static constexpr HostRuntimeMap map{table};
505 static_assert(map.Verify(), "map must be sorted");
507 #endif // HAS_FLOAT80 || HAS_LDBL128
508 #endif //_POSIX_C_SOURCE >= 200112L || _XOPEN_SOURCE >= 600
510 /// Define pgmath description
511 #if LINK_WITH_LIBPGMATH
512 // Only use libpgmath for folding if it is available.
513 // First declare all libpgmaths functions
514 #define PGMATH_LINKING
515 #define PGMATH_DECLARE
516 #include "flang/Evaluate/pgmath.h.inc"
518 #define REAL_FOLDER(name, func) \
519 FolderFactory<decltype(&func), &func>::Create(#name)
520 template <> struct HostRuntimeLibrary<float, LibraryVersion::PgmathFast> {
521 static constexpr HostRuntimeFunction table[]{
522 #define PGMATH_FAST
523 #define PGMATH_USE_S(name, func) REAL_FOLDER(name, func),
524 #include "flang/Evaluate/pgmath.h.inc"
526 static constexpr HostRuntimeMap map{table};
527 static_assert(map.Verify(), "map must be sorted");
529 template <> struct HostRuntimeLibrary<double, LibraryVersion::PgmathFast> {
530 static constexpr HostRuntimeFunction table[]{
531 #define PGMATH_FAST
532 #define PGMATH_USE_D(name, func) REAL_FOLDER(name, func),
533 #include "flang/Evaluate/pgmath.h.inc"
535 static constexpr HostRuntimeMap map{table};
536 static_assert(map.Verify(), "map must be sorted");
538 template <> struct HostRuntimeLibrary<float, LibraryVersion::PgmathRelaxed> {
539 static constexpr HostRuntimeFunction table[]{
540 #define PGMATH_RELAXED
541 #define PGMATH_USE_S(name, func) REAL_FOLDER(name, func),
542 #include "flang/Evaluate/pgmath.h.inc"
544 static constexpr HostRuntimeMap map{table};
545 static_assert(map.Verify(), "map must be sorted");
547 template <> struct HostRuntimeLibrary<double, LibraryVersion::PgmathRelaxed> {
548 static constexpr HostRuntimeFunction table[]{
549 #define PGMATH_RELAXED
550 #define PGMATH_USE_D(name, func) REAL_FOLDER(name, func),
551 #include "flang/Evaluate/pgmath.h.inc"
553 static constexpr HostRuntimeMap map{table};
554 static_assert(map.Verify(), "map must be sorted");
556 template <> struct HostRuntimeLibrary<float, LibraryVersion::PgmathPrecise> {
557 static constexpr HostRuntimeFunction table[]{
558 #define PGMATH_PRECISE
559 #define PGMATH_USE_S(name, func) REAL_FOLDER(name, func),
560 #include "flang/Evaluate/pgmath.h.inc"
562 static constexpr HostRuntimeMap map{table};
563 static_assert(map.Verify(), "map must be sorted");
565 template <> struct HostRuntimeLibrary<double, LibraryVersion::PgmathPrecise> {
566 static constexpr HostRuntimeFunction table[]{
567 #define PGMATH_PRECISE
568 #define PGMATH_USE_D(name, func) REAL_FOLDER(name, func),
569 #include "flang/Evaluate/pgmath.h.inc"
571 static constexpr HostRuntimeMap map{table};
572 static_assert(map.Verify(), "map must be sorted");
575 // TODO: double _Complex/float _Complex have been removed from llvm flang
576 // pgmath.h.inc because they caused warnings, they need to be added back
577 // so that the complex pgmath versions can be used when requested.
579 #endif /* LINK_WITH_LIBPGMATH */
581 // Helper to check if a HostRuntimeLibrary specialization exists
582 template <typename T, typename = void> struct IsAvailable : std::false_type {};
583 template <typename T>
584 struct IsAvailable<T, decltype((void)T::table, void())> : std::true_type {};
585 // Define helpers to find host runtime library map according to desired version
586 // and type.
587 template <typename HostT, LibraryVersion version>
588 static const HostRuntimeMap *GetHostRuntimeMapHelper(
589 [[maybe_unused]] DynamicType resultType) {
590 // A library must only be instantiated if LibraryVersion is
591 // available on the host and if HostT maps to a Fortran type.
592 // For instance, whenever long double and double are both 64-bits, double
593 // is mapped to Fortran 64bits real type, and long double will be left
594 // unmapped.
595 if constexpr (host::FortranTypeExists<HostT>()) {
596 using Lib = HostRuntimeLibrary<HostT, version>;
597 if constexpr (IsAvailable<Lib>::value) {
598 if (host::FortranType<HostT>{}.GetType() == resultType) {
599 return &Lib::map;
603 return nullptr;
605 template <LibraryVersion version>
606 static const HostRuntimeMap *GetHostRuntimeMapVersion(DynamicType resultType) {
607 if (resultType.category() == TypeCategory::Real) {
608 if (const auto *map{GetHostRuntimeMapHelper<float, version>(resultType)}) {
609 return map;
611 if (const auto *map{GetHostRuntimeMapHelper<double, version>(resultType)}) {
612 return map;
614 if (const auto *map{
615 GetHostRuntimeMapHelper<long double, version>(resultType)}) {
616 return map;
618 #if HAS_QUADMATHLIB
619 if (const auto *map{
620 GetHostRuntimeMapHelper<__float128, version>(resultType)}) {
621 return map;
623 #endif
625 if (resultType.category() == TypeCategory::Complex) {
626 if (const auto *map{GetHostRuntimeMapHelper<std::complex<float>, version>(
627 resultType)}) {
628 return map;
630 if (const auto *map{GetHostRuntimeMapHelper<std::complex<double>, version>(
631 resultType)}) {
632 return map;
634 if (const auto *map{
635 GetHostRuntimeMapHelper<std::complex<long double>, version>(
636 resultType)}) {
637 return map;
639 #if HAS_QUADMATHLIB
640 if (const auto *map{
641 GetHostRuntimeMapHelper<__complex128, version>(resultType)}) {
642 return map;
644 #endif
646 return nullptr;
648 static const HostRuntimeMap *GetHostRuntimeMap(
649 LibraryVersion version, DynamicType resultType) {
650 switch (version) {
651 case LibraryVersion::Libm:
652 return GetHostRuntimeMapVersion<LibraryVersion::Libm>(resultType);
653 case LibraryVersion::LibmExtensions:
654 return GetHostRuntimeMapVersion<LibraryVersion::LibmExtensions>(resultType);
655 case LibraryVersion::PgmathPrecise:
656 return GetHostRuntimeMapVersion<LibraryVersion::PgmathPrecise>(resultType);
657 case LibraryVersion::PgmathRelaxed:
658 return GetHostRuntimeMapVersion<LibraryVersion::PgmathRelaxed>(resultType);
659 case LibraryVersion::PgmathFast:
660 return GetHostRuntimeMapVersion<LibraryVersion::PgmathFast>(resultType);
662 return nullptr;
665 static const HostRuntimeFunction *SearchInHostRuntimeMap(
666 const HostRuntimeMap &map, const std::string &name, DynamicType resultType,
667 const std::vector<DynamicType> &argTypes) {
668 auto sameNameRange{map.equal_range(name)};
669 for (const auto *iter{sameNameRange.first}; iter != sameNameRange.second;
670 ++iter) {
671 if (iter->resultType == resultType && iter->argumentTypes == argTypes) {
672 return &*iter;
675 return nullptr;
678 // Search host runtime libraries for an exact type match.
679 static const HostRuntimeFunction *SearchHostRuntime(const std::string &name,
680 DynamicType resultType, const std::vector<DynamicType> &argTypes) {
681 // TODO: When command line options regarding targeted numerical library is
682 // available, this needs to be revisited to take it into account. So far,
683 // default to libpgmath if F18 is built with it.
684 #if LINK_WITH_LIBPGMATH
685 if (const auto *map{
686 GetHostRuntimeMap(LibraryVersion::PgmathPrecise, resultType)}) {
687 if (const auto *hostFunction{
688 SearchInHostRuntimeMap(*map, name, resultType, argTypes)}) {
689 return hostFunction;
692 // Default to libm if functions or types are not available in pgmath.
693 #endif
694 if (const auto *map{GetHostRuntimeMap(LibraryVersion::Libm, resultType)}) {
695 if (const auto *hostFunction{
696 SearchInHostRuntimeMap(*map, name, resultType, argTypes)}) {
697 return hostFunction;
700 if (const auto *map{
701 GetHostRuntimeMap(LibraryVersion::LibmExtensions, resultType)}) {
702 if (const auto *hostFunction{
703 SearchInHostRuntimeMap(*map, name, resultType, argTypes)}) {
704 return hostFunction;
707 return nullptr;
710 // Return a DynamicType that can hold all values of a given type.
711 // This is used to allow 16bit float to be folded with 32bits and
712 // x87 float to be folded with IEEE 128bits.
713 static DynamicType BiggerType(DynamicType type) {
714 if (type.category() == TypeCategory::Real ||
715 type.category() == TypeCategory::Complex) {
716 // 16 bits floats to IEEE 32 bits float
717 if (type.kind() == common::RealKindForPrecision(11) ||
718 type.kind() == common::RealKindForPrecision(8)) {
719 return {type.category(), common::RealKindForPrecision(24)};
721 // x87 float to IEEE 128 bits float
722 if (type.kind() == common::RealKindForPrecision(64)) {
723 return {type.category(), common::RealKindForPrecision(113)};
726 return type;
729 /// Structure to register intrinsic argument checks that must be performed.
730 using ArgumentVerifierFunc = bool (*)(
731 const std::vector<Expr<SomeType>> &, FoldingContext &);
732 struct ArgumentVerifier {
733 using Key = std::string_view;
734 // Needed for implicit compare with keys.
735 constexpr operator Key() const { return key; }
736 Key key;
737 ArgumentVerifierFunc verifier;
740 static constexpr int lastArg{-1};
741 static constexpr int firstArg{0};
743 static const Expr<SomeType> &GetArg(
744 int position, const std::vector<Expr<SomeType>> &args) {
745 if (position == lastArg) {
746 CHECK(!args.empty());
747 return args.back();
749 CHECK(position >= 0 && static_cast<std::size_t>(position) < args.size());
750 return args[position];
753 template <typename T>
754 static bool IsInRange(const Expr<T> &expr, int lb, int ub) {
755 if (auto scalar{GetScalarConstantValue<T>(expr)}) {
756 auto lbValue{Scalar<T>::FromInteger(value::Integer<8>{lb}).value};
757 auto ubValue{Scalar<T>::FromInteger(value::Integer<8>{ub}).value};
758 return Satisfies(RelationalOperator::LE, lbValue.Compare(*scalar)) &&
759 Satisfies(RelationalOperator::LE, scalar->Compare(ubValue));
761 return true;
764 /// Verify that the argument in an intrinsic call belongs to [lb, ub] if is
765 /// real.
766 template <int lb, int ub>
767 static bool VerifyInRangeIfReal(
768 const std::vector<Expr<SomeType>> &args, FoldingContext &context) {
769 if (const auto *someReal{
770 std::get_if<Expr<SomeReal>>(&GetArg(firstArg, args).u)}) {
771 bool isInRange{
772 std::visit([&](const auto &x) -> bool { return IsInRange(x, lb, ub); },
773 someReal->u)};
774 if (!isInRange) {
775 context.messages().Say(
776 "argument is out of range [%d., %d.]"_warn_en_US, lb, ub);
778 return isInRange;
780 return true;
783 template <int argPosition, const char *argName>
784 static bool VerifyStrictlyPositiveIfReal(
785 const std::vector<Expr<SomeType>> &args, FoldingContext &context) {
786 if (const auto *someReal =
787 std::get_if<Expr<SomeReal>>(&GetArg(argPosition, args).u)) {
788 const bool isStrictlyPositive{std::visit(
789 [&](const auto &x) -> bool {
790 using T = typename std::decay_t<decltype(x)>::Result;
791 auto scalar{GetScalarConstantValue<T>(x)};
792 return Satisfies(
793 RelationalOperator::LT, Scalar<T>{}.Compare(*scalar));
795 someReal->u)};
796 if (!isStrictlyPositive) {
797 context.messages().Say(
798 "argument '%s' must be strictly positive"_warn_en_US, argName);
800 return isStrictlyPositive;
802 return true;
805 /// Verify that an intrinsic call argument is not zero if it is real.
806 template <int argPosition, const char *argName>
807 static bool VerifyNotZeroIfReal(
808 const std::vector<Expr<SomeType>> &args, FoldingContext &context) {
809 if (const auto *someReal =
810 std::get_if<Expr<SomeReal>>(&GetArg(argPosition, args).u)) {
811 const bool isNotZero{std::visit(
812 [&](const auto &x) -> bool {
813 using T = typename std::decay_t<decltype(x)>::Result;
814 auto scalar{GetScalarConstantValue<T>(x)};
815 return !scalar || !scalar->IsZero();
817 someReal->u)};
818 if (!isNotZero) {
819 context.messages().Say(
820 "argument '%s' must be different from zero"_warn_en_US, argName);
822 return isNotZero;
824 return true;
827 /// Verify that the argument in an intrinsic call is not zero if is complex.
828 static bool VerifyNotZeroIfComplex(
829 const std::vector<Expr<SomeType>> &args, FoldingContext &context) {
830 if (const auto *someComplex =
831 std::get_if<Expr<SomeComplex>>(&GetArg(firstArg, args).u)) {
832 const bool isNotZero{std::visit(
833 [&](const auto &z) -> bool {
834 using T = typename std::decay_t<decltype(z)>::Result;
835 auto scalar{GetScalarConstantValue<T>(z)};
836 return !scalar || !scalar->IsZero();
838 someComplex->u)};
839 if (!isNotZero) {
840 context.messages().Say(
841 "complex argument must be different from zero"_warn_en_US);
843 return isNotZero;
845 return true;
848 // Verify that the argument in an intrinsic call is not zero and not a negative
849 // integer.
850 static bool VerifyGammaLikeArgument(
851 const std::vector<Expr<SomeType>> &args, FoldingContext &context) {
852 if (const auto *someReal =
853 std::get_if<Expr<SomeReal>>(&GetArg(firstArg, args).u)) {
854 const bool isValid{std::visit(
855 [&](const auto &x) -> bool {
856 using T = typename std::decay_t<decltype(x)>::Result;
857 auto scalar{GetScalarConstantValue<T>(x)};
858 if (scalar) {
859 return !scalar->IsZero() &&
860 !(scalar->IsNegative() &&
861 scalar->ToWholeNumber().value == scalar);
863 return true;
865 someReal->u)};
866 if (!isValid) {
867 context.messages().Say(
868 "argument must not be a negative integer or zero"_warn_en_US);
870 return isValid;
872 return true;
875 // Verify that two real arguments are not both zero.
876 static bool VerifyAtan2LikeArguments(
877 const std::vector<Expr<SomeType>> &args, FoldingContext &context) {
878 if (const auto *someReal =
879 std::get_if<Expr<SomeReal>>(&GetArg(firstArg, args).u)) {
880 const bool isValid{std::visit(
881 [&](const auto &typedExpr) -> bool {
882 using T = typename std::decay_t<decltype(typedExpr)>::Result;
883 auto x{GetScalarConstantValue<T>(typedExpr)};
884 auto y{GetScalarConstantValue<T>(GetArg(lastArg, args))};
885 if (x && y) {
886 return !(x->IsZero() && y->IsZero());
888 return true;
890 someReal->u)};
891 if (!isValid) {
892 context.messages().Say(
893 "'x' and 'y' arguments must not be both zero"_warn_en_US);
895 return isValid;
897 return true;
900 template <ArgumentVerifierFunc... F>
901 static bool CombineVerifiers(
902 const std::vector<Expr<SomeType>> &args, FoldingContext &context) {
903 return (... && F(args, context));
906 /// Define argument names to be used error messages when the intrinsic have
907 /// several arguments.
908 static constexpr char xName[]{"x"};
909 static constexpr char pName[]{"p"};
911 /// Register argument verifiers for all intrinsics folded with runtime.
912 static constexpr ArgumentVerifier intrinsicArgumentVerifiers[]{
913 {"acos", VerifyInRangeIfReal<-1, 1>},
914 {"asin", VerifyInRangeIfReal<-1, 1>},
915 {"atan2", VerifyAtan2LikeArguments},
916 {"bessel_y0", VerifyStrictlyPositiveIfReal<firstArg, xName>},
917 {"bessel_y1", VerifyStrictlyPositiveIfReal<firstArg, xName>},
918 {"bessel_yn", VerifyStrictlyPositiveIfReal<lastArg, xName>},
919 {"gamma", VerifyGammaLikeArgument},
920 {"log",
921 CombineVerifiers<VerifyStrictlyPositiveIfReal<firstArg, xName>,
922 VerifyNotZeroIfComplex>},
923 {"log10", VerifyStrictlyPositiveIfReal<firstArg, xName>},
924 {"log_gamma", VerifyGammaLikeArgument},
925 {"mod", VerifyNotZeroIfReal<lastArg, pName>},
928 const ArgumentVerifierFunc *findVerifier(const std::string &intrinsicName) {
929 static constexpr Fortran::common::StaticMultimapView<ArgumentVerifier>
930 verifiers(intrinsicArgumentVerifiers);
931 static_assert(verifiers.Verify(), "map must be sorted");
932 auto range{verifiers.equal_range(intrinsicName)};
933 if (range.first != range.second) {
934 return &range.first->verifier;
936 return nullptr;
939 /// Ensure argument verifiers, if any, are run before calling the runtime
940 /// wrapper to fold an intrinsic.
941 static HostRuntimeWrapper AddArgumentVerifierIfAny(
942 const std::string &intrinsicName, const HostRuntimeFunction &hostFunction) {
943 if (const auto *verifier{findVerifier(intrinsicName)}) {
944 const HostRuntimeFunction *hostFunctionPtr = &hostFunction;
945 return [hostFunctionPtr, verifier](
946 FoldingContext &context, std::vector<Expr<SomeType>> &&args) {
947 const bool validArguments{(*verifier)(args, context)};
948 if (!validArguments) {
949 // Silence fp signal warnings since a more detailed warning about
950 // invalid arguments was already emitted.
951 parser::Messages localBuffer;
952 parser::ContextualMessages localMessages{&localBuffer};
953 FoldingContext localContext{context, localMessages};
954 return hostFunctionPtr->folder(localContext, std::move(args));
956 return hostFunctionPtr->folder(context, std::move(args));
959 return hostFunction.folder;
962 std::optional<HostRuntimeWrapper> GetHostRuntimeWrapper(const std::string &name,
963 DynamicType resultType, const std::vector<DynamicType> &argTypes) {
964 if (const auto *hostFunction{SearchHostRuntime(name, resultType, argTypes)}) {
965 return AddArgumentVerifierIfAny(name, *hostFunction);
967 // If no exact match, search with "bigger" types and insert type
968 // conversions around the folder.
969 std::vector<evaluate::DynamicType> biggerArgTypes;
970 evaluate::DynamicType biggerResultType{BiggerType(resultType)};
971 for (auto type : argTypes) {
972 biggerArgTypes.emplace_back(BiggerType(type));
974 if (const auto *hostFunction{
975 SearchHostRuntime(name, biggerResultType, biggerArgTypes)}) {
976 auto hostFolderWithChecks{AddArgumentVerifierIfAny(name, *hostFunction)};
977 return [hostFunction, resultType, hostFolderWithChecks](
978 FoldingContext &context, std::vector<Expr<SomeType>> &&args) {
979 auto nArgs{args.size()};
980 for (size_t i{0}; i < nArgs; ++i) {
981 args[i] = Fold(context,
982 ConvertToType(hostFunction->argumentTypes[i], std::move(args[i]))
983 .value());
985 return Fold(context,
986 ConvertToType(
987 resultType, hostFolderWithChecks(context, std::move(args)))
988 .value());
991 return std::nullopt;
993 } // namespace Fortran::evaluate