| /* |
| * Licensed to the Apache Software Foundation (ASF) under one or more |
| * contributor license agreements. See the NOTICE file distributed with |
| * this work for additional information regarding copyright ownership. |
| * The ASF licenses this file to You under the Apache License, Version 2.0 |
| * (the "License"); you may not use this file except in compliance with |
| * the License. You may obtain a copy of the License at |
| * |
| * http://www.apache.org/licenses/LICENSE-2.0 |
| * |
| * Unless required by applicable law or agreed to in writing, software |
| * distributed under the License is distributed on an "AS IS" BASIS, |
| * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| * See the License for the specific language governing permissions and |
| * limitations under the License. |
| */ |
| |
| #if !defined(cbigint_h) |
| #define cbigint_h |
| |
| #include "JNIHelp.h" |
| |
| #include <sys/types.h> |
| #include <sys/param.h> |
| #include <stdint.h> |
| |
| /* IEEE floats consist of: sign bit, exponent field, significand field |
| single: 31 = sign bit, 30..23 = exponent (8 bits), 22..0 = significand (23 bits) |
| double: 63 = sign bit, 62..52 = exponent (11 bits), 51..0 = significand (52 bits) |
| inf == (all exponent bits set) and (all mantissa bits clear) |
| nan == (all exponent bits set) and (at least one mantissa bit set) |
| finite == (at least one exponent bit clear) |
| zero == (all exponent bits clear) and (all mantissa bits clear) |
| denormal == (all exponent bits clear) and (at least one mantissa bit set) |
| positive == sign bit clear |
| negative == sign bit set |
| */ |
| #if __BYTE_ORDER == __LITTLE_ENDIAN |
| #define DOUBLE_LO_OFFSET 0 |
| #define DOUBLE_HI_OFFSET 1 |
| #define LONG_LO_OFFSET 0 |
| #define LONG_HI_OFFSET 1 |
| #else |
| #define DOUBLE_LO_OFFSET 1 |
| #define DOUBLE_HI_OFFSET 0 |
| #define LONG_LO_OFFSET 1 |
| #define LONG_HI_OFFSET 0 |
| #endif |
| |
| #define DOUBLE_EXPONENT_MASK_HI 0x7FF00000 |
| #define DOUBLE_MANTISSA_MASK_HI 0x000FFFFF |
| |
| union U64U32DBL { |
| uint64_t u64val; |
| uint32_t u32val[2]; |
| int32_t i32val[2]; |
| double dval; |
| }; |
| |
| #define DOUBLE_TO_LONGBITS(dbl) (*(reinterpret_cast<uint64_t*>(&dbl))) |
| #define FLOAT_TO_INTBITS(flt) (*(reinterpret_cast<uint32_t*>(&flt))) |
| #define INTBITS_TO_FLOAT(bits) (*(reinterpret_cast<float*>(&bits))) |
| |
| /* Replace P_FLOAT_HI and P_FLOAT_LOW */ |
| /* These macros are used to access the high and low 32-bit parts of a double (64-bit) value. */ |
| #define LOW_U32_FROM_DBL_PTR(dblptr) ((reinterpret_cast<U64U32DBL*>(dblptr))->u32val[DOUBLE_LO_OFFSET]) |
| #define HIGH_U32_FROM_DBL_PTR(dblptr) ((reinterpret_cast<U64U32DBL*>(dblptr))->u32val[DOUBLE_HI_OFFSET]) |
| #define LOW_I32_FROM_DBL_PTR(dblptr) ((reinterpret_cast<U64U32DBL*>(dblptr))->i32val[DOUBLE_LO_OFFSET]) |
| #define HIGH_I32_FROM_DBL_PTR(dblptr) ((reinterpret_cast<U64U32DBL*>(dblptr))->i32val[DOUBLE_HI_OFFSET]) |
| #define LOW_U32_FROM_DBL(dbl) LOW_U32_FROM_DBL_PTR(&(dbl)) |
| #define HIGH_U32_FROM_DBL(dbl) HIGH_U32_FROM_DBL_PTR(&(dbl)) |
| #define LOW_U32_FROM_LONG64_PTR(long64ptr) ((reinterpret_cast<U64U32DBL*>(long64ptr))->u32val[LONG_LO_OFFSET]) |
| #define HIGH_U32_FROM_LONG64_PTR(long64ptr) ((reinterpret_cast<U64U32DBL*>(long64ptr))->u32val[LONG_HI_OFFSET]) |
| #define LOW_I32_FROM_LONG64_PTR(long64ptr) ((reinterpret_cast<U64U32DBL*>(long64ptr))->i32val[LONG_LO_OFFSET]) |
| #define HIGH_I32_FROM_LONG64_PTR(long64ptr) ((reinterpret_cast<U64U32DBL*>(long64ptr))->i32val[LONG_HI_OFFSET]) |
| #define LOW_U32_FROM_LONG64(long64) LOW_U32_FROM_LONG64_PTR(&(long64)) |
| #define HIGH_U32_FROM_LONG64(long64) HIGH_U32_FROM_LONG64_PTR(&(long64)) |
| #define LOW_I32_FROM_LONG64(long64) LOW_I32_FROM_LONG64_PTR(&(long64)) |
| #define HIGH_I32_FROM_LONG64(long64) HIGH_I32_FROM_LONG64_PTR(&(long64)) |
| #define IS_DENORMAL_DBL_PTR(dblptr) (((HIGH_U32_FROM_DBL_PTR(dblptr) & DOUBLE_EXPONENT_MASK_HI) == 0) && ((HIGH_U32_FROM_DBL_PTR(dblptr) & DOUBLE_MANTISSA_MASK_HI) != 0 || (LOW_U32_FROM_DBL_PTR(dblptr) != 0))) |
| #define IS_DENORMAL_DBL(dbl) IS_DENORMAL_DBL_PTR(&(dbl)) |
| |
| #define LOW_U32_FROM_VAR(u64) LOW_U32_FROM_LONG64(u64) |
| #define LOW_U32_FROM_PTR(u64ptr) LOW_U32_FROM_LONG64_PTR(u64ptr) |
| #define HIGH_U32_FROM_VAR(u64) HIGH_U32_FROM_LONG64(u64) |
| #define HIGH_U32_FROM_PTR(u64ptr) HIGH_U32_FROM_LONG64_PTR(u64ptr) |
| |
| void multiplyHighPrecision(uint64_t* arg1, int32_t length1, uint64_t* arg2, int32_t length2, |
| uint64_t* result, int32_t length); |
| uint32_t simpleAppendDecimalDigitHighPrecision(uint64_t* arg1, int32_t length, uint64_t digit); |
| jdouble toDoubleHighPrecision(uint64_t* arg, int32_t length); |
| uint64_t doubleMantissa(jdouble z); |
| int32_t compareHighPrecision(uint64_t* arg1, int32_t length1, uint64_t* arg2, int32_t length2); |
| int32_t highestSetBitHighPrecision(uint64_t* arg, int32_t length); |
| void subtractHighPrecision(uint64_t* arg1, int32_t length1, uint64_t* arg2, int32_t length2); |
| int32_t doubleExponent(jdouble z); |
| int32_t addHighPrecision(uint64_t* arg1, int32_t length1, uint64_t* arg2, int32_t length2); |
| int32_t lowestSetBit(uint64_t* y); |
| int32_t timesTenToTheEHighPrecision(uint64_t* result, int32_t length, jint e); |
| int32_t highestSetBit(uint64_t* y); |
| int32_t lowestSetBitHighPrecision(uint64_t* arg, int32_t length); |
| void simpleShiftLeftHighPrecision(uint64_t* arg1, int32_t length, int32_t arg2); |
| uint32_t floatMantissa(jfloat z); |
| int32_t simpleAddHighPrecision(uint64_t* arg1, int32_t length, uint64_t arg2); |
| int32_t floatExponent(jfloat z); |
| |
| #endif /* cbigint_h */ |