| doc |
|
|
| hpma512.s |
|
26033 |
| hppa20.s |
|
28203 |
| logtab.h |
0 1 2 3 |
1574 |
| montmulf.c |
1/(2^16) |
8190 |
| montmulf.h |
The functions that are to be called from outside of the .s file have the
following interfaces and array size requirements:
|
3097 |
| montmulf.il |
|
2226 |
| montmulf.s |
|
78139 |
| montmulfv8.il |
|
2226 |
| montmulfv8.s |
|
60099 |
| montmulfv9.il |
|
1845 |
| montmulfv9.s |
|
79580 |
| mp_comba.c |
TomsFastMath, a fast ISO C bignum library.
This project is meant to fill in where LibTomMath
falls short. That is speed ;-)
This project is public domain and free for all purposes.
Tom St Denis, tomstdenis@iahu.ca
|
79362 |
| mp_comba_amd64_masm.asm |
|
332877 |
| mp_comba_amd64_sun.s |
|
312089 |
| mp_gf2m-priv.h |
enable fast divide and mod operations on MP_DIGIT_BITS |
3527 |
| mp_gf2m.c |
Multiply two binary polynomials mp_digits a, b.
Result is a polynomial with degree < 2 * MP_DIGIT_BITS - 1.
Output in two mp_digits rh, rl.
|
16393 |
| mp_gf2m.h |
For modular arithmetic, the irreducible polynomial f(t) is represented
as an array of int[], where f(t) is of the form:
f(t) = t^p[0] + t^p[1] + ... + t^p[k]
where m = p[0] > p[1] > ... > p[k] = 0.
|
1120 |
| mpcpucache.c |
This file implements a single function: s_mpi_getProcessorLineSize();
s_mpi_getProcessorLineSize() returns the size in bytes of the cache line
if a cache exists, or zero if there is no cache. If more than one
cache line exists, it should return the smallest line size (which is
usually the L1 cache).
mp_modexp uses this information to make sure that private key information
isn't being leaked through the cache.
Currently the file returns good data for most modern x86 processors, and
reasonable data on 64-bit ppc processors. All other processors are assumed
to have a cache line size of 32 bytes.
|
23757 |
| mpcpucache_amd64.s |
|
10622 |
| mpcpucache_x86.s |
|
11189 |
| mpi-config.h |
For boolean options,
0 = no
1 = yes
Other options are documented individually.
|
1287 |
| mpi-priv.h |
If we aren't using a wired-in logarithm table, we need to include
the math library to get the log() function
|
9209 |
| mpi.c |
16-bit thumb or ARM v3 doesn't work inlined assember version |
121599 |
| mpi.h |
no error, all is well |
12225 |
| mpi_amd64.c |
MPI glue
|
725 |
| mpi_amd64_common.S |
|
9208 |
| mpi_amd64_masm.asm |
|
7781 |
| mpi_amd64_sun.s |
|
8796 |
| mpi_arm.c |
This inlined version is for 32-bit ARM platform only |
3926 |
| mpi_hp.c |
This file contains routines that perform vector multiplication. |
2462 |
| mpi_i86pc.s |
|
7029 |
| mpi_mips.s |
|
8834 |
| mpi_sparc.c |
Multiplication performance enhancements for sparc v8+vis CPUs. |
5733 |
| mpi_sse2.s |
|
7991 |
| mpi_x86.s |
|
12676 |
| mpi_x86_asm.c |
ebp - 36: caller's esi
ebp - 32: caller's edi
ebp - 28:
ebp - 24:
ebp - 20:
ebp - 16:
ebp - 12:
ebp - 8:
ebp - 4:
ebp + 0: caller's ebp
ebp + 4: return address
ebp + 8: a argument
ebp + 12: a_len argument
ebp + 16: b argument
ebp + 20: c argument
registers:
eax:
ebx: carry
ecx: a_len
edx:
esi: a ptr
edi: c ptr
|
12811 |
| mplogic.c |
{{{ Lookup table for population count |
10118 |
| mplogic.h |
The logical operations treat an mp_int as if it were a bit vector,
without regard to its sign (an mp_int is represented in a signed
magnitude format). Values are treated as if they had an infinite
string of zeros left of the most-significant bit.
|
1815 |
| mpmontg.c |
This file implements moduluar exponentiation using Montgomery's
method for modular reduction. This file implements the method
described as "Improvement 2" in the paper "A Cryptogrpahic Library for
the Motorola DSP56000" by Stephen R. Dusse' and Burton S. Kaliski Jr.
published in "Advances in Cryptology: Proceedings of EUROCRYPT '90"
"Lecture Notes in Computer Science" volume 473, 1991, pg 230-244,
published by Springer Verlag.
|
35044 |
| mpprime.c |
determines size of hard-wired prime table |
15056 |
| mpprime.h |
number of primes available |
1644 |
| mpv_sparc.c |
/
typedef int t_s32;
typedef unsigned int t_u32;
#if defined(__sparcv9)
typedef long t_s64;
typedef unsigned long t_u64;
#else
typedef long long t_s64;
typedef unsigned long long t_u64;
#endif
typedef double t_d64;
/************************************************************** |
5482 |
| mpv_sparcv8.s |
|
64933 |
| mpv_sparcv9.s |
|
67722 |
| mpvalpha.c |
This is empty for the loop in s_mpv_mul_d |
5462 |
| mulsqr.c |
make sure squaring code is included |
2103 |
| primes.c |
|
56352 |
| README |
This Source Code Form is subject to the terms of the Mozilla Public |
28010 |
| vis_32.il |
|
26216 |
| vis_64.il |
|
19882 |
| vis_proto.h |
Prototypes for the inline templates in vis.il
|
9353 |