tomasz.sowa/ttmath
1
0
Fork 0
Code Issues Pull Requests Releases Activity

current chk version - too many changes on both sides for now ;-(

Browse Source 
git-svn-id: svn://ttmath.org/publicrep/ttmath/branches/chk@150 e52654a7-88a9-db11-a3e9-0013d4bc506e
This commit is contained in:
Christian Kaiser
2009-05-19 10:50:41 +00:00
parent 9b576ddbe2 5e5a106605
commit fdc292e91a
9 changed files with 12491 additions and 12487 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CHANGELOG
+20 -1
View File
@@ -1,4 +1,4 @@
Version 0.8.5 (2009.05.11):
Version 0.8.5 prerelease (2009.05.15):
* fixed: Big::Mod(x) didn't correctly return a carry
and the result was sometimes very big (even greater than x)
* fixed: global function Mod(x) didn't set an ErrorCode object
@@ -11,6 +11,25 @@ Version 0.8.5 (2009.05.11):
the same is to Cos() function
* changed: PrepareSin(x) is using Big::Mod() now when reducing 2PI period
should be a little accurate especially on a very big 'x'
* changed: uint Mul(const UInt<value_size> & ss2, uint algorithm = 100)
void MulBig(const UInt<value_size> & ss2, UInt<value_size*2> & result, uint algorithm = 100)
those methods by default use MulFastest() and MulFastestBig()
* changed: changed a little Mul2Big() to cooperate with Mul3Big()
* added: uint UInt::Mul3(const UInt<value_size> & ss2)
void UInt::Mul3Big(const UInt<value_size> & ss2, UInt<value_size*2> & result)
a new multiplication algorithm: Karatsuba multiplication,
on a vector UInt<100> with all items different from zero this algorithm is faster
about 3 times than Mul2Big(), and on a vector UInt<1000> with all items different from
zero this algorithm is faster more than 5 times than Mul2Big()
(measured on 32bit platform with GCC 4.3.3 with -O3 and -DTTMATH_RELEASE)
* added: uint MulFastest(const UInt<value_size> & ss2)
void MulFastestBig(const UInt<value_size> & ss2, UInt<value_size*2> & result)
those methods are trying to select the fastest multiplication algorithm
* added: uint AddVector(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result)
uint SubVector(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result)
three forms: asm x86, asm x86_64, no-asm
those methods are used by the Karatsuba multiplication algorithm
Version 0.8.4 (2009.05.08):
* fixed: UInt::DivInt() didn't check whether the divisor is zero
ttmath/ttmath.h
+2038 -2065
View File
File diff suppressed because it is too large Load Diff
ttmath/ttmathbig.h
+4047 -4056
View File
File diff suppressed because it is too large Load Diff
ttmath/ttmathparser.h
+2544 -2555
View File
File diff suppressed because it is too large Load Diff
ttmath/ttmathtypes.h
+3 -5
View File
@@ -64,7 +64,7 @@
*/
#define TTMATH_MAJOR_VER 0
#define TTMATH_MINOR_VER 8
#define TTMATH_REVISION_VER 5
#define TTMATH_REVISION_VER 4
#define TTMATH_PRERELEASE_VER 1
@@ -120,6 +120,7 @@ namespace ttmath
typedef unsigned int uint;
typedef signed int sint;
/*!
this type is twice bigger than uint
(64bit on a 32bit platforms)
@@ -128,11 +129,8 @@ namespace ttmath
but it is defined in C99 and in upcoming C++0x /3.9.1 (2)/ and many compilers support it
this type is used in UInt::MulTwoWords and UInt::DivTwoWords when macro TTMATH_NOASM is defined
but only on a 32bit platform
*/
#ifdef TTMATH_NOASM
typedef unsigned long long int ulint;
#endif
typedef unsigned long long int ulint;
/*!
the mask for the highest bit in the unsigned 32bit word (2^31)
ttmath/ttmathuint.h
+2939 -2934
View File
File diff suppressed because it is too large Load Diff
ttmath/ttmathuint_noasm.h
+885 -857
View File
File diff suppressed because it is too large Load Diff
ttmath/ttmathuint_x86_64.h
+11 -10
View File
@@ -158,7 +158,7 @@ namespace ttmath
table[1] = 30 + 2;
table[2] = 5;
of course if there was a carry from table[2] it would be returned
of course if there was a carry from table[3] it would be returned
*/
template<uint value_size>
uint UInt<value_size>::AddInt(uint value, uint index)
@@ -374,7 +374,7 @@ namespace ttmath
table[1] = 30 - 2;
table[2] = 5;
of course if there was a carry from table[2] it would be returned
of course if there was a carry from table[3] it would be returned
*/
template<uint value_size>
uint UInt<value_size>::SubInt(uint value, uint index)
@@ -695,7 +695,7 @@ namespace ttmath
/*
this method returns the number of the highest set bit in one 64-bit word
this method returns the number of the highest set bit in one 32-bit word
if the 'x' is zero this method returns '-1'
***this method is created only on a 64bit platform***
@@ -800,17 +800,18 @@ namespace ttmath
/*!
multiplication: result_high:result_low = a * b
result_high - higher word of the result
result_low - lower word of the result
multiplication: result2:result1 = a * b
result2 - higher word
result1 - lower word of the result
this methos never returns a carry
this method is used in the second version of the multiplication algorithms
***this method is created only on a 64bit platform***
it is an auxiliary method for version two of the multiplication algorithm
*/
template<uint value_size>
void UInt<value_size>::MulTwoWords(uint a, uint b, uint * result_high, uint * result_low)
void UInt<value_size>::MulTwoWords(uint a, uint b, uint * result2, uint * result1)
{
/*
we must use these temporary variables in order to inform the compilator
@@ -843,8 +844,8 @@ namespace ttmath
#endif
*result_low = result1_;
*result_high = result2_;
*result1 = result1_;
*result2 = result2_;
}
ttmath/ttmathuint_x86_amd64_msvc.asm
+4 -4
View File
@@ -114,9 +114,9 @@ loop1:
mov r9, 0 ; set to 0 -> cy still set!
dec rdx
jnz loop1
jc return_1 ; most of the times, there will be NO carry (I hope)
done:
jc return_1 ; most of the times, there will be NO carry (I hope)
xor rax, rax
ret
@@ -184,9 +184,9 @@ loop1:
mov r9, 1
dec rdx
jnz loop1
done:
jc return_1 ; most of the times, there will be NO carry (I hope)
done:
xor rax, rax
ret