| /netbsd-current/external/lgpl3/gmp/dist/mpn/x86/pentium/ |
| H A D | copyi.asm | 86 C ecx counter, limbs, negative
120 C ecx 0 to 7, representing respectively 7 to 0 limbs remaining
|
| H A D | com.asm | 81 C ecx counter, limbs, negative
124 C ecx 0 to 7, representing respectively 7 to 0 limbs remaining
|
| H A D | mod_34lsub1.asm | 102 C ecx counter, limbs
124 C ecx is -2, -1 or 0, representing 0, 1 or 2 more limbs, respectively
|
| H A D | sqr_basecase.asm | 35 C product at around 20x20 limbs.
153 C three limbs
492 C ebp scratch (fetched dst limbs)
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/x86/k6/ |
| H A D | mod_34lsub1.asm | 115 C eax counter, limbs
134 C ecx is -3, -2 or -1 representing 0, 1 or 2 more limbs, respectively
|
| H A D | sqr_basecase.asm | 35 C product (measured on the speed difference between 17 and 33 limbs,
64 dnl limbs, UNROLL_COUNT needs to be SQR_TOOM2_THRESHOLD-3.
84 C 9.0 and 9.2 c/tri-prod measured on the difference between 17 and 33 limbs.
178 C three limbs
297 C loop, but this was only a tiny speedup; at 35 limbs it took 24 cycles off
435 C edx VAR_COUNTER, limbs, negative
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/powerpc32/ |
| H A D | addmul_1.asm | 68 cmpwi cr0,r5,9 C more than 9 limbs?
69 bgt cr0,L(big) C branch if more than 9 limbs
|
| H A D | rshift.asm | 51 cmpwi cr0, r5, 30 C more than 30 limbs?
53 bgt L(BIG) C branch if more than 12 limbs
|
| H A D | submul_1.asm | 54 cmpwi cr0,r5,9 C more than 9 limbs?
55 bgt cr0,L(big) C branch if more than 9 limbs
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/x86/ |
| H A D | mod_34lsub1.asm | 107 C ecx counter, limbs
124 C ecx is -2, -1 or 0 representing 0, 1 or 2 more limbs, respectively
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/pa32/ |
| H A D | sub_n.asm | 50 sub %r20,%r19,%r28 C subtract first limbs ignoring cy
|
| H A D | add_n.asm | 50 add %r20,%r19,%r28 C add first limbs ignoring cy
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/arm64/ |
| H A D | copyi.asm | 69 C Copy last 0-3 limbs. Note that rp is aligned after loop, but not when we
|
| H A D | com.asm | 73 C Copy last 0-3 limbs. Note that rp is aligned after loop, but not when we
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/x86/p6/ |
| H A D | bdiv_q_1.asm | 107 C ebx counter, limbs, negative
129 C ebx counter, limbs, negative
195 C ebx counter, limbs, negative
|
| H A D | dive_1.asm | 148 C ebx counter, limbs, negative
189 C ebx counter, limbs, negative
211 C ebx counter, limbs, negative
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/x86_64/fastsse/ |
| H A D | com-palignr.asm | 58 C limbs. We use the SSSE3 palignr instruction when rp - up = 8 (mod 16). That
62 C For operands of < COM_SSE_THRESHOLD limbs, we use a plain 64-bit loop, taken
155 C quite separate: up-rp < 5 or up-up > 15 limbs
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/arm/neon/ |
| H A D | lshiftc.asm | 88 vld1.32 {d19}, [ap], r12 C load initial 2 limbs
160 C Load last 2 - 3 limbs, store last 4 - 5 limbs
|
| H A D | lorrshift.asm | 112 vld1.32 {d19}, [ap], r12 C load initial 2 limbs
175 C Load last 2 - 3 limbs, store last 4 - 5 limbs
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/x86/k7/mmx/ |
| H A D | lshift.asm | 34 C K7: 1.21 cycles/limb (at 16 limbs/loop).
127 C eax loop counter, limbs
255 decl %eax C size-2, two last limbs handled at end
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/arm/v7a/cora15/neon/ |
| H A D | com.asm | 79 C Handle last 0-7 limbs. Note that rp is aligned after loop, but not when we
|
| H A D | copyd.asm | 89 C Copy last 0-7 limbs. Note that rp is aligned after loop, but not when we
|
| H A D | copyi.asm | 75 C Copy last 0-7 limbs. Note that rp is aligned after loop, but not when we
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/alpha/ev67/ |
| H A D | popcount.asm | 45 C The main loop processes two limbs at a time. An odd size is handled by
|
| /netbsd-current/external/lgpl3/gmp/dist/mpn/pa32/hppa1_1/pa7100/ |
| H A D | add_n.asm | 47 add %r20,%r19,%r28 C add first limbs ignoring cy
|