1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
#define NVGPU_ECC_ENABLED_SM_SHM 9
/* TEX ECC is enabled */
#define NVGPU_ECC_ENABLED_TEX 10
/* L2 ECC is enabled */
#define NVGPU_ECC_ENABLED_LTC 11
/* SM L1 DATA ECC is enabled */
#define NVGPU_ECC_ENABLED_SM_L1_DATA 12
/* SM L1 TAG ECC is enabled */
#define NVGPU_ECC_ENABLED_SM_L1_TAG 13
/* SM CBU ECC is enabled */
#define NVGPU_ECC_ENABLED_SM_CBU 14
/* SM ICAHE ECC is enabled */
#define NVGPU_ECC_ENABLED_SM_ICACHE 15
/*
* MM flags.
*/
#define NVGPU_MM_UNIFY_ADDRESS_SPACES 16
/* false if vidmem aperture actually points to sysmem */
#define NVGPU_MM_HONORS_APERTURE 17
/* unified or split memory with separate vidmem? */
#define NVGPU_MM_UNIFIED_MEMORY 18
/* User-space managed address spaces support */
#define NVGPU_SUPPORT_USERSPACE_MANAGED_AS 20
/* IO coherence support is available */
#define NVGPU_SUPPORT_IO_COHERENCE 21
/* MAP_BUFFER_EX with partial mappings */
#define NVGPU_SUPPORT_PARTIAL_MAPPINGS 22
/* MAP_BUFFER_EX with sparse allocations */
#define NVGPU_SUPPORT_SPARSE_ALLOCS 23
/* Direct PTE kind control is supported (map_buffer_ex) */
#define NVGPU_SUPPORT_MAP_DIRECT_KIND_CTRL 24
/* Support batch mapping */
#define NVGPU_SUPPORT_MAP_BUFFER_BATCH 25
/* Use coherent aperture for sysmem. */
#define NVGPU_USE_COHERENT_SYSMEM 26
/* Use physical scatter tables instead of IOMMU */
#define NVGPU_MM_USE_PHYSICAL_SG 27
/* WAR for gm20b chips. */
#define NVGPU_MM_FORCE_128K_PMU_VM 28
/*
* Host flags
*/
#define NVGPU_HAS_SYNCPOINTS 30
/* sync fence FDs are available in, e.g., submit_gpfifo */
#define NVGPU_SUPPORT_SYNC_FENCE_FDS 31
/* NVGPU_DBG_GPU_IOCTL_CYCLE_STATS is available */
#define NVGPU_SUPPORT_CYCLE_STATS 32
/* NVGPU_DBG_GPU_IOCTL_CYCLE_STATS_SNAPSHOT is available */
/* mpi-internal.h - Internal to the Multi Precision Integers
* Copyright (C) 1994, 1996 Free Software Foundation, Inc.
* Copyright (C) 1998, 2000 Free Software Foundation, Inc.
*
* This file is part of GnuPG.
*
* GnuPG is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* GnuPG is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
*
* Note: This code is heavily based on the GNU MP Library.
* Actually it's the same code with only minor changes in the
* way the data is stored; this is to support the abstraction
* of an optional secure memory allocation which may be used
* to avoid revealing of sensitive data due to paging etc.
* The GNU MP Library itself is published under the LGPL;
* however I decided to publish this code under the plain GPL.
*/
#ifndef G10_MPI_INTERNAL_H
#define G10_MPI_INTERNAL_H
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/mpi.h>
#include <linux/errno.h>
#define log_debug printk
#define log_bug printk
#define assert(x) \
do { \
if (!x) \
log_bug("failed assertion\n"); \
} while (0);
/* If KARATSUBA_THRESHOLD is not already defined, define it to a
* value which is good on most machines. */
/* tested 4, 16, 32 and 64, where 16 gave the best performance when
* checking a 768 and a 1024 bit ElGamal signature.
* (wk 22.12.97) */
#ifndef KARATSUBA_THRESHOLD
#define KARATSUBA_THRESHOLD 16
#endif
/* The code can't handle KARATSUBA_THRESHOLD smaller than 2. */
#if KARATSUBA_THRESHOLD < 2
#undef KARATSUBA_THRESHOLD
#define KARATSUBA_THRESHOLD 2
#endif
typedef mpi_limb_t *mpi_ptr_t; /* pointer to a limb */
typedef int mpi_size_t; /* (must be a signed type) */
#define ABS(x) (x >= 0 ? x : -x)
#define MIN(l, o) ((l) < (o) ? (l) : (o))
#define MAX(h, i) ((h) > (i) ? (h) : (i))
static inline int RESIZE_IF_NEEDED(MPI a, unsigned b)
{
if (a->alloced < b)
return mpi_resize(a, b);
return 0;
}
/* Copy N limbs from S to D. */
#define MPN_COPY(d, s, n) \
do { \
mpi_size_t _i; \
for (_i = 0; _i < (n); _i++) \
(d)[_i] = (s)[_i]; \
} while (0)
#define MPN_COPY_INCR(d, s, n) \
do { \
mpi_size_t _i; \
for (_i = 0; _i < (n); _i++) \
(d)[_i] = (d)[_i]; \
} while (0)
#define MPN_COPY_DECR(d, s, n) \
do { \
mpi_size_t _i; \
for (_i = (n)-1; _i >= 0; _i--) \
(d)[_i] = (s)[_i]; \
} while (0)
/* Zero N limbs at D */
#define MPN_ZERO(d, n) \
do { \
int _i; \
for (_i = 0; _i < (n); _i++) \
(d)[_i] = 0; \
} while (0)
#define MPN_NORMALIZE(d, n) \
do { \
while ((n) > 0) { \
if ((d)[(n)-1]) \
break; \
(n)--; \
} \
} while (0)
#define MPN_NORMALIZE_NOT_ZERO(d, n) \
do { \
for (;;) { \
if ((d)[(n)-1]) \
break; \
(n)--; \
} \
} while (0)
#define MPN_MUL_N_RECURSE(prodp, up, vp, size, tspace) \
do { \
if ((size) < KARATSUBA_THRESHOLD) \
mul_n_basecase(prodp, up, vp, size); \
else \
mul_n(prodp, up, vp, size, tspace); \
} while (0);
/* Divide the two-limb number in (NH,,NL) by D, with DI being the largest
* limb not larger than (2**(2*BITS_PER_MP_LIMB))/D - (2**BITS_PER_MP_LIMB).
* If this would yield overflow, DI should be the largest possible number
* (i.e., only ones). For correct operation, the most significant bit of D
* has to be set. Put the quotient in Q and the remainder in R.
*/
#define UDIV_QRNND_PREINV(q, r, nh, nl, d, di) \
do { \
mpi_limb_t _q, _ql, _r; \
mpi_limb_t _xh, _xl; \
umul_ppmm(_q, _ql, (nh), (di)); \
_q += (nh); /* DI is 2**BITS_PER_MPI_LIMB too small */ \
umul_ppmm(_xh, _xl, _q, (d)); \
sub_ddmmss(_xh, _r, (nh), (nl), _xh, _xl); \
if (_xh) { \
sub_ddmmss(_xh, _r, _xh, _r, 0, (d)); \
_q++; \
if (_xh) { \
sub_ddmmss(_xh, _r, _xh, _r, 0, (d)); \
_q++; \
} \
} \
if (_r >= (d)) { \
_r -= (d); \
_q++; \
} \
(r) = _r; \
(q) = _q; \
} while (0)
/*-- mpiutil.c --*/
mpi_ptr_t mpi_alloc_limb_space(unsigned nlimbs);
void mpi_free_limb_space(mpi_ptr_t a);
void mpi_assign_limb_space(MPI a, mpi_ptr_t ap, unsigned nlimbs);
/*-- mpi-bit.c --*/
void mpi_rshift_limbs(MPI a, unsigned int count);
int mpi_lshift_limbs(MPI a, unsigned int count);
/*-- mpihelp-add.c --*/
mpi_limb_t mpihelp_add_1(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr,
mpi_size_t s1_size, mpi_limb_t s2_limb);
mpi_limb_t mpihelp_add_n(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr,
mpi_ptr_t s2_ptr, mpi_size_t size);
mpi_limb_t mpihelp_add(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr, mpi_size_t s1_size,
mpi_ptr_t s2_ptr, mpi_size_t s2_size);
/*-- mpihelp-sub.c --*/
mpi_limb_t mpihelp_sub_1(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr,
mpi_size_t s1_size, mpi_limb_t s2_limb);
mpi_limb_t mpihelp_sub_n(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr,
mpi_ptr_t s2_ptr, mpi_size_t size);
mpi_limb_t mpihelp_sub(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr, mpi_size_t s1_size,
mpi_ptr_t s2_ptr, mpi_size_t s2_size);
/*-- mpihelp-cmp.c --*/
int mpihelp_cmp(mpi_ptr_t op1_ptr, mpi_ptr_t op2_ptr, mpi_size_t size);
/*-- mpihelp-mul.c --*/
struct karatsuba_ctx {
struct karatsuba_ctx *next;
mpi_ptr_t tspace;
mpi_size_t tspace_size;
mpi_ptr_t tp;
mpi_size_t tp_size;
};
void mpihelp_release_karatsuba_ctx(struct karatsuba_ctx *ctx);
mpi_limb_t mpihelp_addmul_1(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr,
mpi_size_t s1_size, mpi_limb_t s2_limb);
mpi_limb_t mpihelp_submul_1(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr,
mpi_size_t s1_size, mpi_limb_t s2_limb);
int mpihelp_mul_n(mpi_ptr_t prodp, mpi_ptr_t up, mpi_ptr_t vp, mpi_size_t size);
int mpihelp_mul(mpi_ptr_t prodp, mpi_ptr_t up, mpi_size_t usize,
mpi_ptr_t vp, mpi_size_t vsize, mpi_limb_t *_result);
void mpih_sqr_n_basecase(mpi_ptr_t prodp, mpi_ptr_t up, mpi_size_t size);
void mpih_sqr_n(mpi_ptr_t prodp, mpi_ptr_t up, mpi_size_t size,
mpi_ptr_t tspace);
int mpihelp_mul_karatsuba_case(mpi_ptr_t prodp,
mpi_ptr_t up, mpi_size_t usize,
mpi_ptr_t vp, mpi_size_t vsize,
struct karatsuba_ctx *ctx);
/*-- mpihelp-mul_1.c (or xxx/cpu/ *.S) --*/
mpi_limb_t mpihelp_mul_1(mpi_ptr_t res_ptr, mpi_ptr_t s1_ptr,
mpi_size_t s1_size, mpi_limb_t s2_limb);
/*-- mpihelp-div.c --*/
mpi_limb_t mpihelp_mod_1(mpi_ptr_t dividend_ptr, mpi_size_t dividend_size,
mpi_limb_t divisor_limb);
mpi_limb_t mpihelp_divrem(mpi_ptr_t qp, mpi_size_t qextra_limbs,
mpi_ptr_t np, mpi_size_t nsize,
mpi_ptr_t dp, mpi_size_t dsize);
mpi_limb_t mpihelp_divmod_1(mpi_ptr_t quot_ptr,
mpi_ptr_t dividend_ptr, mpi_size_t dividend_size,
mpi_limb_t divisor_limb);
/*-- mpihelp-shift.c --*/
mpi_limb_t mpihelp_lshift(mpi_ptr_t wp, mpi_ptr_t up, mpi_size_t usize,
unsigned cnt);
mpi_limb_t mpihelp_rshift(mpi_ptr_t wp, mpi_ptr_t up, mpi_size_t usize,
unsigned cnt);
/* Define stuff for longlong.h. */
#define W_TYPE_SIZE BITS_PER_MPI_LIMB
typedef mpi_limb_t UWtype;
typedef unsigned int UHWtype;
#if defined(__GNUC__)
typedef unsigned int UQItype __attribute__ ((mode(QI)));
typedef int SItype __attribute__ ((mode(SI)));
typedef unsigned int USItype __attribute__ ((mode(SI)));
typedef int DItype __attribute__ ((mode(DI)));
typedef unsigned int UDItype __attribute__ ((mode(DI)));
#else
typedef unsigned char UQItype;
typedef long SItype;
typedef unsigned long USItype;
#endif
#ifdef __GNUC__
#include "mpi-inline.h"
#endif
#endif /*G10_MPI_INTERNAL_H */
|
