FPUtil/x86_64/FEnv.h

//===-- x86_64 floating point env manipulation functions --------*- C++ -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//

#ifndef LLVM_LIBC_UTILS_FPUTIL_X86_64_FENV_H
#define LLVM_LIBC_UTILS_FPUTIL_X86_64_FENV_H

#include <fenv.h>
#include <stdint.h>

namespace __llvm_libc {
namespace fputil {

namespace internal {

// Normally, one should be able to define FE_* macros to the exact rounding mode
// encodings. However, since we want LLVM libc to be compiled against headers
// from other libcs, we cannot assume that FE_* macros are always defined in
// such a manner. So, we will define enums corresponding to the x86_64 bit
// encodings. The implementations can map from FE_* to the corresponding enum
// values.

// The rounding control values in the x87 control register and the MXCSR
// register have the same 2-bit enoding but have different bit positions.
// See below for the bit positions.
struct RoundingControlValue {
  static constexpr uint16_t ToNearest = 0x0;
  static constexpr uint16_t Downward = 0x1;
  static constexpr uint16_t Upward = 0x2;
  static constexpr uint16_t TowardZero = 0x3;
};

static constexpr uint16_t X87RoundingControlBitPosition = 10;
static constexpr uint16_t MXCSRRoundingControlBitPosition = 13;

// The exception flags in the x87 status register and the MXCSR have the same
// encoding as well as the same bit positions.
struct ExceptionFlags {
  static constexpr uint16_t Invalid = 0x1;
  static constexpr uint16_t Denormal = 0x2; // This flag is not used
  static constexpr uint16_t DivByZero = 0x4;
  static constexpr uint16_t Overflow = 0x8;
  static constexpr uint16_t Underflow = 0x10;
  static constexpr uint16_t Inexact = 0x20;
};

// The exception control bits occupy six bits, one bit for each exception.
// In the x87 control word, they occupy the first 6 bits. In the MXCSR
// register, they occupy bits 7 to 12.
static constexpr uint16_t X87ExceptionControlBitPosition = 0;
static constexpr uint16_t MXCSRExceptionContolBitPoistion = 7;

// Exception flags are individual bits in the corresponding registers.
// So, we just OR the bit values to get the full set of exceptions.
static inline uint16_t getStatusValueForExcept(int excepts) {
  // We will make use of the fact that exception control bits are single
  // bit flags in the control registers.
  return (excepts & FE_INVALID ? ExceptionFlags::Invalid : 0) |
         (excepts & FE_DIVBYZERO ? ExceptionFlags::DivByZero : 0) |
         (excepts & FE_OVERFLOW ? ExceptionFlags::Overflow : 0) |
         (excepts & FE_UNDERFLOW ? ExceptionFlags::Underflow : 0) |
         (excepts & FE_INEXACT ? ExceptionFlags::Inexact : 0);
}

static inline int exceptionStatusToMacro(uint16_t status) {
  return (status & ExceptionFlags::Invalid ? FE_INVALID : 0) |
         (status & ExceptionFlags::DivByZero ? FE_DIVBYZERO : 0) |
         (status & ExceptionFlags::Overflow ? FE_OVERFLOW : 0) |
         (status & ExceptionFlags::Underflow ? FE_UNDERFLOW : 0) |
         (status & ExceptionFlags::Inexact ? FE_INEXACT : 0);
}

struct X87State {
  uint16_t ControlWord;
  uint16_t Unused1;
  uint16_t StatusWord;
  uint16_t Unused2;
  // TODO: Elaborate the remaining 20 bytes as required.
  uint32_t _[5];
};

static inline uint16_t getX87ControlWord() {
  uint16_t w;
  __asm__ __volatile__("fnstcw %0" : "=m"(w)::);
  return w;
}

static inline void writeX87ControlWord(uint16_t w) {
  __asm__ __volatile__("fldcw %0" : : "m"(w) :);
}

static inline uint16_t getX87StatusWord() {
  uint16_t w;
  __asm__ __volatile__("fnstsw %0" : "=m"(w)::);
  return w;
}

static inline void clearX87Exceptions() {
  __asm__ __volatile__("fnclex" : : :);
}

static inline uint32_t getMXCSR() {
  uint32_t w;
  __asm__ __volatile__("stmxcsr %0" : "=m"(w)::);
  return w;
}

static inline void writeMXCSR(uint32_t w) {
  __asm__ __volatile__("ldmxcsr %0" : : "m"(w) :);
}

static inline void getX87State(X87State &s) {
  __asm__ __volatile__("fnstenv %0" : "=m"(s));
}

static inline void writeX87State(const X87State &s) {
  __asm__ __volatile__("fldenv %0" : : "m"(s) :);
}

static inline void fwait() { __asm__ __volatile__("fwait"); }

} // namespace internal

static inline int enableExcept(int excepts) {
  // In the x87 control word and in MXCSR, an exception is blocked
  // if the corresponding bit is set. That is the reason for all the
  // bit-flip operations below as we need to turn the bits to zero
  // to enable them.

  uint16_t bitMask = internal::getStatusValueForExcept(excepts);

  uint16_t x87CW = internal::getX87ControlWord();
  uint16_t oldExcepts = ~x87CW & 0x3F; // Save previously enabled exceptions.
  x87CW &= ~bitMask;
  internal::writeX87ControlWord(x87CW);

  // Enabling SSE exceptions via MXCSR is a nice thing to do but
  // might not be of much use practically as SSE exceptions and the x87
  // exceptions are independent of each other.
  uint32_t mxcsr = internal::getMXCSR();
  mxcsr &= ~(bitMask << internal::MXCSRExceptionContolBitPoistion);
  internal::writeMXCSR(mxcsr);

  // Since the x87 exceptions and SSE exceptions are independent of each,
  // it doesn't make much sence to report both in the return value. Most
  // often, the standard floating point functions deal with FPU operations
  // so we will retrun only the old x87 exceptions.
  return internal::exceptionStatusToMacro(oldExcepts);
}

static inline int disableExcept(int excepts) {
  // In the x87 control word and in MXCSR, an exception is blocked
  // if the corresponding bit is set.

  uint16_t bitMask = internal::getStatusValueForExcept(excepts);

  uint16_t x87CW = internal::getX87ControlWord();
  uint16_t oldExcepts = ~x87CW & 0x3F; // Save previously enabled exceptions.
  x87CW |= bitMask;
  internal::writeX87ControlWord(x87CW);

  // Just like in enableExcept, it is not clear if disabling SSE exceptions
  // is required. But, we will still do it only as a "nice thing to do".
  uint32_t mxcsr = internal::getMXCSR();
  mxcsr |= (bitMask << internal::MXCSRExceptionContolBitPoistion);
  internal::writeMXCSR(mxcsr);

  return internal::exceptionStatusToMacro(oldExcepts);
}

static inline int clearExcept(int excepts) {
  // An instruction to write to x87 status word ins't available. So, we
  // just clear all of the x87 exceptions.
  // TODO: One can potentially use fegetenv/fesetenv to clear only the
  // listed exceptions in the x87 status word. We can do this if it is
  // really required.
  internal::clearX87Exceptions();

  uint32_t mxcsr = internal::getMXCSR();
  mxcsr &= ~internal::getStatusValueForExcept(excepts);
  internal::writeMXCSR(mxcsr);
  return 0;
}

static inline int testExcept(int excepts) {
  uint16_t statusValue = internal::getStatusValueForExcept(excepts);
  // Check both x87 status word and MXCSR.
  return internal::exceptionStatusToMacro(
      (statusValue & internal::getX87StatusWord()) |
      (statusValue & internal::getMXCSR()));
}

static inline int raiseExcept(int excepts) {
  uint16_t statusValue = internal::getStatusValueForExcept(excepts);

  // We set the status flag for exception one at a time and call the
  // fwait instruction to actually get the processor to raise the
  // exception by calling the exception handler. This scheme is per
  // the description in in "8.6 X87 FPU EXCEPTION SYNCHRONIZATION"
  // of the "Intel 64 and IA-32 Architectures Software Developer's
  // Manual, Vol 1".

  // FPU status word is read for each exception seperately as the
  // exception handler can potentially write to it (typically to clear
  // the corresponding exception flag). By reading it separately, we
  // ensure that the writes by the exception handler are maintained
  // when raising the next exception.

  if (statusValue & internal::ExceptionFlags::Invalid) {
    internal::X87State state;
    internal::getX87State(state);
    state.StatusWord |= internal::ExceptionFlags::Invalid;
    internal::writeX87State(state);
    internal::fwait();
  }
  if (statusValue & internal::ExceptionFlags::DivByZero) {
    internal::X87State state;
    internal::getX87State(state);
    state.StatusWord |= internal::ExceptionFlags::DivByZero;
    internal::writeX87State(state);
    internal::fwait();
  }
  if (statusValue & internal::ExceptionFlags::Overflow) {
    internal::X87State state;
    internal::getX87State(state);
    state.StatusWord |= internal::ExceptionFlags::Overflow;
    internal::writeX87State(state);
    internal::fwait();
  }
  if (statusValue & internal::ExceptionFlags::Underflow) {
    internal::X87State state;
    internal::getX87State(state);
    state.StatusWord |= internal::ExceptionFlags::Underflow;
    internal::writeX87State(state);
    internal::fwait();
  }
  if (statusValue & internal::ExceptionFlags::Inexact) {
    internal::X87State state;
    internal::getX87State(state);
    state.StatusWord |= internal::ExceptionFlags::Inexact;
    internal::writeX87State(state);
    internal::fwait();
  }

  // There is no special synchronization scheme available to
  // raise SEE exceptions. So, we will ignore that for now.
  // Just plain writing to the MXCSR register does not guarantee
  // the exception handler will be called.

  return 0;
}

static inline int getRound() {
  uint16_t bitValue =
      (internal::getMXCSR() >> internal::MXCSRRoundingControlBitPosition) & 0x3;
  switch (bitValue) {
  case internal::RoundingControlValue::ToNearest:
    return FE_TONEAREST;
  case internal::RoundingControlValue::Downward:
    return FE_DOWNWARD;
  case internal::RoundingControlValue::Upward:
    return FE_UPWARD;
  case internal::RoundingControlValue::TowardZero:
    return FE_TOWARDZERO;
  default:
    return -1; // Error value.
  }
}

static inline int setRound(int mode) {
  uint16_t bitValue;
  switch (mode) {
  case FE_TONEAREST:
    bitValue = internal::RoundingControlValue::ToNearest;
    break;
  case FE_DOWNWARD:
    bitValue = internal::RoundingControlValue::Downward;
    break;
  case FE_UPWARD:
    bitValue = internal::RoundingControlValue::Upward;
    break;
  case FE_TOWARDZERO:
    bitValue = internal::RoundingControlValue::TowardZero;
    break;
  default:
    return 1; // To indicate failure
  }

  uint16_t x87Value = bitValue << internal::X87RoundingControlBitPosition;
  uint16_t x87Control = internal::getX87ControlWord();
  x87Control =
      (x87Control & ~(0x3 << internal::X87RoundingControlBitPosition)) |
      x87Value;
  internal::writeX87ControlWord(x87Control);

  uint32_t mxcsrValue = bitValue << internal::MXCSRRoundingControlBitPosition;
  uint32_t mxcsrControl = internal::getMXCSR();
  mxcsrControl =
      (mxcsrControl & ~(0x3 << internal::MXCSRRoundingControlBitPosition)) |
      mxcsrValue;
  internal::writeMXCSR(mxcsrControl);

  return 0;
}

} // namespace fputil
} // namespace __llvm_libc

#endif // LLVM_LIBC_UTILS_FPUTIL_X86_64_FENV_H