Commit 143e0376a16462198a3412243e776ff9b69742d5

Daniel Mendler 2019-10-29T20:02:32

simplifications: basic arithmetic functions

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
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
diff --git a/mp_add_d.c b/mp_add_d.c
index 4508cc8..43d50e8 100644
--- a/mp_add_d.c
+++ b/mp_add_d.c
@@ -6,9 +6,7 @@
 /* single digit addition */
 mp_err mp_add_d(const mp_int *a, mp_digit b, mp_int *c)
 {
-   mp_err     err;
-   int ix, oldused;
-   mp_digit *tmpa, *tmpc;
+   int oldused;
 
    /* fast path for a == c */
    if (a == c) {
@@ -27,6 +25,7 @@ mp_err mp_add_d(const mp_int *a, mp_digit b, mp_int *c)
 
    /* grow c as required */
    if (c->alloc < (a->used + 1)) {
+      mp_err err;
       if ((err = mp_grow(c, a->used + 1)) != MP_OKAY) {
          return err;
       }
@@ -34,6 +33,7 @@ mp_err mp_add_d(const mp_int *a, mp_digit b, mp_int *c)
 
    /* if a is negative and |a| >= b, call c = |a| - b */
    if ((a->sign == MP_NEG) && ((a->used > 1) || (a->dp[0] >= b))) {
+      mp_err err;
       mp_int a_ = *a;
       /* temporarily fix sign of a */
       a_.sign = MP_ZPOS;
@@ -53,49 +53,34 @@ mp_err mp_add_d(const mp_int *a, mp_digit b, mp_int *c)
    /* old number of used digits in c */
    oldused = c->used;
 
-   /* source alias */
-   tmpa    = a->dp;
-
-   /* destination alias */
-   tmpc    = c->dp;
-
    /* if a is positive */
    if (a->sign == MP_ZPOS) {
       /* add digits, mu is carry */
+      int i;
       mp_digit mu = b;
-      for (ix = 0; ix < a->used; ix++) {
-         *tmpc   = *tmpa++ + mu;
-         mu      = *tmpc >> MP_DIGIT_BIT;
-         *tmpc++ &= MP_MASK;
+      for (i = 0; i < a->used; i++) {
+         c->dp[i] = a->dp[i] + mu;
+         mu = c->dp[i] >> MP_DIGIT_BIT;
+         c->dp[i] &= MP_MASK;
       }
       /* set final carry */
-      ix++;
-      *tmpc++  = mu;
+      c->dp[i] = mu;
 
       /* setup size */
       c->used = a->used + 1;
    } else {
       /* a was negative and |a| < b */
-      c->used  = 1;
+      c->used = 1;
 
       /* the result is a single digit */
-      if (a->used == 1) {
-         *tmpc++  =  b - a->dp[0];
-      } else {
-         *tmpc++  =  b;
-      }
-
-      /* setup count so the clearing of oldused
-       * can fall through correctly
-       */
-      ix       = 1;
+      c->dp[0] = (a->used == 1) ? b - a->dp[0] : b;
    }
 
    /* sign always positive */
    c->sign = MP_ZPOS;
 
    /* now zero to oldused */
-   MP_ZERO_DIGITS(tmpc, oldused - ix);
+   MP_ZERO_DIGITS(c->dp + c->used, oldused - c->used);
    mp_clamp(c);
 
    return MP_OKAY;
diff --git a/mp_div_2.c b/mp_div_2.c
index 60bd63d..573570d 100644
--- a/mp_div_2.c
+++ b/mp_div_2.c
@@ -6,12 +6,11 @@
 /* b = a/2 */
 mp_err mp_div_2(const mp_int *a, mp_int *b)
 {
-   int     x, oldused;
-   mp_digit r, rr, *tmpa, *tmpb;
-   mp_err err;
+   int x, oldused;
+   mp_digit r;
 
-   /* copy */
    if (b->alloc < a->used) {
+      mp_err err;
       if ((err = mp_grow(b, a->used)) != MP_OKAY) {
          return err;
       }
@@ -20,20 +19,14 @@ mp_err mp_div_2(const mp_int *a, mp_int *b)
    oldused = b->used;
    b->used = a->used;
 
-   /* source alias */
-   tmpa = a->dp + b->used - 1;
-
-   /* dest alias */
-   tmpb = b->dp + b->used - 1;
-
    /* carry */
    r = 0;
-   for (x = b->used - 1; x >= 0; x--) {
+   for (x = b->used; x --> 0;) {
       /* get the carry for the next iteration */
-      rr = *tmpa & 1u;
+      mp_digit rr = a->dp[x] & 1u;
 
       /* shift the current digit, add in carry and store */
-      *tmpb-- = (*tmpa-- >> 1) | (r << (MP_DIGIT_BIT - 1));
+      b->dp[x] = (a->dp[x] >> 1) | (r << (MP_DIGIT_BIT - 1));
 
       /* forward carry to next iteration */
       r = rr;
diff --git a/mp_div_2d.c b/mp_div_2d.c
index 9b396ac..e523465 100644
--- a/mp_div_2d.c
+++ b/mp_div_2d.c
@@ -6,23 +6,16 @@
 /* shift right by a certain bit count (store quotient in c, optional remainder in d) */
 mp_err mp_div_2d(const mp_int *a, int b, mp_int *c, mp_int *d)
 {
-   mp_digit D, r, rr;
-   int     x;
    mp_err err;
 
-   /* if the shift count is <= 0 then we do no work */
-   if (b <= 0) {
-      err = mp_copy(a, c);
-      if (d != NULL) {
-         mp_zero(d);
-      }
-      return err;
+   if (b < 0) {
+      return MP_VAL;
    }
 
-   /* copy */
    if ((err = mp_copy(a, c)) != MP_OKAY) {
       return err;
    }
+
    /* 'a' should not be used after here - it might be the same as d */
 
    /* get the remainder */
@@ -38,28 +31,25 @@ mp_err mp_div_2d(const mp_int *a, int b, mp_int *c, mp_int *d)
    }
 
    /* shift any bit count < MP_DIGIT_BIT */
-   D = (mp_digit)(b % MP_DIGIT_BIT);
-   if (D != 0u) {
-      mp_digit *tmpc, mask, shift;
+   b %= MP_DIGIT_BIT;
+   if (b != 0u) {
+      int x;
+      mp_digit r, mask, shift;
 
       /* mask */
-      mask = ((mp_digit)1 << D) - 1uL;
+      mask = ((mp_digit)1 << b) - 1uL;
 
       /* shift for lsb */
-      shift = (mp_digit)MP_DIGIT_BIT - D;
-
-      /* alias */
-      tmpc = c->dp + (c->used - 1);
+      shift = (mp_digit)(MP_DIGIT_BIT - b);
 
       /* carry */
       r = 0;
-      for (x = c->used - 1; x >= 0; x--) {
+      for (x = c->used; x --> 0;) {
          /* get the lower  bits of this word in a temp */
-         rr = *tmpc & mask;
+         mp_digit rr = c->dp[x] & mask;
 
          /* shift the current word and mix in the carry bits from the previous word */
-         *tmpc = (*tmpc >> D) | (r << shift);
-         --tmpc;
+         c->dp[x] = (c->dp[x] >> b) | (r << shift);
 
          /* set the carry to the carry bits of the current word found above */
          r = rr;
diff --git a/mp_div_d.c b/mp_div_d.c
index 98b6b24..472ab27 100644
--- a/mp_div_d.c
+++ b/mp_div_d.c
@@ -8,7 +8,6 @@ mp_err mp_div_d(const mp_int *a, mp_digit b, mp_int *c, mp_digit *d)
 {
    mp_int  q;
    mp_word w;
-   mp_digit t;
    mp_err err;
    int ix;
 
@@ -56,14 +55,12 @@ mp_err mp_div_d(const mp_int *a, mp_digit b, mp_int *c, mp_digit *d)
    q.used = a->used;
    q.sign = a->sign;
    w = 0;
-   for (ix = a->used - 1; ix >= 0; ix--) {
+   for (ix = a->used; ix --> 0;) {
+      mp_digit t = 0;
       w = (w << (mp_word)MP_DIGIT_BIT) | (mp_word)a->dp[ix];
-
       if (w >= b) {
          t = (mp_digit)(w / b);
          w -= (mp_word)t * (mp_word)b;
-      } else {
-         t = 0;
       }
       q.dp[ix] = t;
    }
@@ -78,7 +75,7 @@ mp_err mp_div_d(const mp_int *a, mp_digit b, mp_int *c, mp_digit *d)
    }
    mp_clear(&q);
 
-   return err;
+   return MP_OKAY;
 }
 
 #endif
diff --git a/mp_mul.c b/mp_mul.c
index 9c8f8ae..1a7091c 100644
--- a/mp_mul.c
+++ b/mp_mul.c
@@ -7,8 +7,8 @@
 mp_err mp_mul(const mp_int *a, const mp_int *b, mp_int *c)
 {
    mp_err err;
-   int min_len = MP_MIN(a->used, b->used),
-       max_len = MP_MAX(a->used, b->used),
+   int min = MP_MIN(a->used, b->used),
+       max = MP_MAX(a->used, b->used),
        digs = a->used + b->used + 1;
    mp_sign neg = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
 
@@ -20,16 +20,16 @@ mp_err mp_mul(const mp_int *a, const mp_int *b, mp_int *c)
         * Using it to cut the input into slices small enough for s_mp_mul_digs_fast
         * was actually slower on the author's machine, but YMMV.
         */
-       (min_len >= MP_KARATSUBA_MUL_CUTOFF) &&
-       ((max_len / 2) >= MP_KARATSUBA_MUL_CUTOFF) &&
+       (min >= MP_KARATSUBA_MUL_CUTOFF) &&
+       ((max / 2) >= MP_KARATSUBA_MUL_CUTOFF) &&
        /* Not much effect was observed below a ratio of 1:2, but again: YMMV. */
-       (max_len >= (2 * min_len))) {
+       (max >= (2 * min))) {
       err = s_mp_balance_mul(a,b,c);
    } else if (MP_HAS(S_MP_TOOM_MUL) &&
-              (min_len >= MP_TOOM_MUL_CUTOFF)) {
+              (min >= MP_TOOM_MUL_CUTOFF)) {
       err = s_mp_toom_mul(a, b, c);
    } else if (MP_HAS(S_MP_KARATSUBA_MUL) &&
-              (min_len >= MP_KARATSUBA_MUL_CUTOFF)) {
+              (min >= MP_KARATSUBA_MUL_CUTOFF)) {
       err = s_mp_karatsuba_mul(a, b, c);
    } else if (MP_HAS(S_MP_MUL_DIGS_FAST) &&
               /* can we use the fast multiplier?
@@ -39,7 +39,7 @@ mp_err mp_mul(const mp_int *a, const mp_int *b, mp_int *c)
                * digits won't affect carry propagation
                */
               (digs < MP_WARRAY) &&
-              (min_len <= MP_MAXFAST)) {
+              (min <= MP_MAXFAST)) {
       err = s_mp_mul_digs_fast(a, b, c, digs);
    } else if (MP_HAS(S_MP_MUL_DIGS)) {
       err = s_mp_mul_digs(a, b, c, digs);
diff --git a/mp_mul_2.c b/mp_mul_2.c
index cd5589d..45b6f1c 100644
--- a/mp_mul_2.c
+++ b/mp_mul_2.c
@@ -6,11 +6,12 @@
 /* b = a*2 */
 mp_err mp_mul_2(const mp_int *a, mp_int *b)
 {
-   int     x, oldused;
-   mp_err err;
+   int x, oldused;
+   mp_digit r;
 
    /* grow to accomodate result */
    if (b->alloc < (a->used + 1)) {
+      mp_err err;
       if ((err = mp_grow(b, a->used + 1)) != MP_OKAY) {
          return err;
       }
@@ -19,45 +20,35 @@ mp_err mp_mul_2(const mp_int *a, mp_int *b)
    oldused = b->used;
    b->used = a->used;
 
-   {
-      mp_digit r, rr, *tmpa, *tmpb;
+   /* carry */
+   r = 0;
+   for (x = 0; x < a->used; x++) {
 
-      /* alias for source */
-      tmpa = a->dp;
-
-      /* alias for dest */
-      tmpb = b->dp;
-
-      /* carry */
-      r = 0;
-      for (x = 0; x < a->used; x++) {
+      /* get what will be the *next* carry bit from the
+       * MSB of the current digit
+       */
+      mp_digit rr = a->dp[x] >> (mp_digit)(MP_DIGIT_BIT - 1);
 
-         /* get what will be the *next* carry bit from the
-          * MSB of the current digit
-          */
-         rr = *tmpa >> (mp_digit)(MP_DIGIT_BIT - 1);
+      /* now shift up this digit, add in the carry [from the previous] */
+      b->dp[x] = ((a->dp[x] << 1uL) | r) & MP_MASK;
 
-         /* now shift up this digit, add in the carry [from the previous] */
-         *tmpb++ = ((*tmpa++ << 1uL) | r) & MP_MASK;
+      /* copy the carry that would be from the source
+       * digit into the next iteration
+       */
+      r = rr;
+   }
 
-         /* copy the carry that would be from the source
-          * digit into the next iteration
-          */
-         r = rr;
-      }
+   /* new leading digit? */
+   if (r != 0u) {
+      /* add a MSB which is always 1 at this point */
+      b->dp[b->used++] = 1;
+   }
 
-      /* new leading digit? */
-      if (r != 0u) {
-         /* add a MSB which is always 1 at this point */
-         *tmpb = 1;
-         ++(b->used);
-      }
+   /* now zero any excess digits on the destination
+    * that we didn't write to
+    */
+   MP_ZERO_DIGITS(b->dp + b->used, oldused - b->used);
 
-      /* now zero any excess digits on the destination
-       * that we didn't write to
-       */
-      MP_ZERO_DIGITS(b->dp + b->used, oldused - b->used);
-   }
    b->sign = a->sign;
    return MP_OKAY;
 }
diff --git a/mp_mul_2d.c b/mp_mul_2d.c
index 1ba53a0..f1016ea 100644
--- a/mp_mul_2d.c
+++ b/mp_mul_2d.c
@@ -6,17 +6,19 @@
 /* shift left by a certain bit count */
 mp_err mp_mul_2d(const mp_int *a, int b, mp_int *c)
 {
-   mp_digit d;
-   mp_err   err;
+   if (b < 0) {
+      return MP_VAL;
+   }
 
-   /* copy */
    if (a != c) {
+      mp_err err;
       if ((err = mp_copy(a, c)) != MP_OKAY) {
          return err;
       }
    }
 
    if (c->alloc < (c->used + (b / MP_DIGIT_BIT) + 1)) {
+      mp_err err;
       if ((err = mp_grow(c, c->used + (b / MP_DIGIT_BIT) + 1)) != MP_OKAY) {
          return err;
       }
@@ -24,35 +26,32 @@ mp_err mp_mul_2d(const mp_int *a, int b, mp_int *c)
 
    /* shift by as many digits in the bit count */
    if (b >= MP_DIGIT_BIT) {
+      mp_err err;
       if ((err = mp_lshd(c, b / MP_DIGIT_BIT)) != MP_OKAY) {
          return err;
       }
    }
 
    /* shift any bit count < MP_DIGIT_BIT */
-   d = (mp_digit)(b % MP_DIGIT_BIT);
-   if (d != 0u) {
-      mp_digit *tmpc, shift, mask, r, rr;
+   b %= MP_DIGIT_BIT;
+   if (b != 0u) {
+      mp_digit shift, mask, r;
       int x;
 
       /* bitmask for carries */
-      mask = ((mp_digit)1 << d) - (mp_digit)1;
+      mask = ((mp_digit)1 << b) - (mp_digit)1;
 
       /* shift for msbs */
-      shift = (mp_digit)MP_DIGIT_BIT - d;
-
-      /* alias */
-      tmpc = c->dp;
+      shift = (mp_digit)(MP_DIGIT_BIT - b);
 
       /* carry */
       r    = 0;
       for (x = 0; x < c->used; x++) {
          /* get the higher bits of the current word */
-         rr = (*tmpc >> shift) & mask;
+         mp_digit rr = (c->dp[x] >> shift) & mask;
 
          /* shift the current word and OR in the carry */
-         *tmpc = ((*tmpc << d) | r) & MP_MASK;
-         ++tmpc;
+         c->dp[x] = ((c->dp[x] << b) | r) & MP_MASK;
 
          /* set the carry to the carry bits of the current word */
          r = rr;
diff --git a/mp_mul_d.c b/mp_mul_d.c
index 399dc7b..3e5335f 100644
--- a/mp_mul_d.c
+++ b/mp_mul_d.c
@@ -6,10 +6,9 @@
 /* multiply by a digit */
 mp_err mp_mul_d(const mp_int *a, mp_digit b, mp_int *c)
 {
-   mp_digit u, *tmpa, *tmpc;
-   mp_word  r;
+   mp_digit u;
    mp_err   err;
-   int      ix, olduse;
+   int   ix, oldused;
 
    /* make sure c is big enough to hold a*b */
    if (c->alloc < (a->used + 1)) {
@@ -19,41 +18,35 @@ mp_err mp_mul_d(const mp_int *a, mp_digit b, mp_int *c)
    }
 
    /* get the original destinations used count */
-   olduse = c->used;
+   oldused = c->used;
 
    /* set the sign */
    c->sign = a->sign;
 
-   /* alias for a->dp [source] */
-   tmpa = a->dp;
-
-   /* alias for c->dp [dest] */
-   tmpc = c->dp;
-
    /* zero carry */
    u = 0;
 
    /* compute columns */
    for (ix = 0; ix < a->used; ix++) {
       /* compute product and carry sum for this term */
-      r       = (mp_word)u + ((mp_word)*tmpa++ * (mp_word)b);
+      mp_word r       = (mp_word)u + ((mp_word)a->dp[ix] * (mp_word)b);
 
       /* mask off higher bits to get a single digit */
-      *tmpc++ = (mp_digit)(r & (mp_word)MP_MASK);
+      c->dp[ix] = (mp_digit)(r & (mp_word)MP_MASK);
 
       /* send carry into next iteration */
       u       = (mp_digit)(r >> (mp_word)MP_DIGIT_BIT);
    }
 
    /* store final carry [if any] and increment ix offset  */
-   *tmpc++ = u;
-   ++ix;
-
-   /* now zero digits above the top */
-   MP_ZERO_DIGITS(tmpc, olduse - ix);
+   c->dp[ix] = u;
 
    /* set used count */
    c->used = a->used + 1;
+
+   /* now zero digits above the top */
+   MP_ZERO_DIGITS(c->dp + c->used, oldused - c->used);
+
    mp_clamp(c);
 
    return MP_OKAY;
diff --git a/mp_sub_d.c b/mp_sub_d.c
index 96a747c..c5cf726 100644
--- a/mp_sub_d.c
+++ b/mp_sub_d.c
@@ -6,9 +6,7 @@
 /* single digit subtraction */
 mp_err mp_sub_d(const mp_int *a, mp_digit b, mp_int *c)
 {
-   mp_digit *tmpa, *tmpc;
-   mp_err    err;
-   int       ix, oldused;
+   int oldused;
 
    /* fast path for a == c */
    if (a == c) {
@@ -26,6 +24,7 @@ mp_err mp_sub_d(const mp_int *a, mp_digit b, mp_int *c)
 
    /* grow c as required */
    if (c->alloc < (a->used + 1)) {
+      mp_err err;
       if ((err = mp_grow(c, a->used + 1)) != MP_OKAY) {
          return err;
       }
@@ -35,6 +34,7 @@ mp_err mp_sub_d(const mp_int *a, mp_digit b, mp_int *c)
     * addition [with fudged signs]
     */
    if (a->sign == MP_NEG) {
+      mp_err err;
       mp_int a_ = *a;
       a_.sign = MP_ZPOS;
       err     = mp_add_d(&a_, b, c);
@@ -46,24 +46,17 @@ mp_err mp_sub_d(const mp_int *a, mp_digit b, mp_int *c)
       return err;
    }
 
-   /* setup regs */
    oldused = c->used;
-   tmpa    = a->dp;
-   tmpc    = c->dp;
 
    /* if a <= b simply fix the single digit */
    if (((a->used == 1) && (a->dp[0] <= b)) || (a->used == 0)) {
-      if (a->used == 1) {
-         *tmpc++ = b - *tmpa;
-      } else {
-         *tmpc++ = b;
-      }
-      ix      = 1;
+      c->dp[0] = (a->used == 1) ? b - a->dp[0] : b;
 
       /* negative/1digit */
       c->sign = MP_NEG;
       c->used = 1;
    } else {
+      int i;
       mp_digit mu = b;
 
       /* positive/size */
@@ -71,15 +64,15 @@ mp_err mp_sub_d(const mp_int *a, mp_digit b, mp_int *c)
       c->used = a->used;
 
       /* subtract digits, mu is carry */
-      for (ix = 0; ix < a->used; ix++) {
-         *tmpc    = *tmpa++ - mu;
-         mu       = *tmpc >> (MP_SIZEOF_BITS(mp_digit) - 1u);
-         *tmpc++ &= MP_MASK;
+      for (i = 0; i < a->used; i++) {
+         c->dp[i] = a->dp[i] - mu;
+         mu = c->dp[i] >> (MP_SIZEOF_BITS(mp_digit) - 1u);
+         c->dp[i] &= MP_MASK;
       }
    }
 
    /* zero excess digits */
-   MP_ZERO_DIGITS(tmpc, oldused - ix);
+   MP_ZERO_DIGITS(c->dp + c->used, oldused - c->used);
 
    mp_clamp(c);
    return MP_OKAY;
diff --git a/s_mp_add.c b/s_mp_add.c
index 9220719..1dd09f8 100644
--- a/s_mp_add.c
+++ b/s_mp_add.c
@@ -6,84 +6,65 @@
 /* low level addition, based on HAC pp.594, Algorithm 14.7 */
 mp_err s_mp_add(const mp_int *a, const mp_int *b, mp_int *c)
 {
-   const mp_int *x;
-   mp_err err;
-   int     olduse, min, max;
+   int oldused, min, max, i;
+   mp_digit u;
 
    /* find sizes, we let |a| <= |b| which means we have to sort
     * them.  "x" will point to the input with the most digits
     */
-   if (a->used > b->used) {
-      min = b->used;
-      max = a->used;
-      x = a;
-   } else {
-      min = a->used;
-      max = b->used;
-      x = b;
+   if (a->used < b->used) {
+      MP_EXCH(const mp_int *, a, b);
    }
 
+   min = b->used;
+   max = a->used;
+
    /* init result */
    if (c->alloc < (max + 1)) {
+      mp_err err;
       if ((err = mp_grow(c, max + 1)) != MP_OKAY) {
          return err;
       }
    }
 
    /* get old used digit count and set new one */
-   olduse = c->used;
+   oldused = c->used;
    c->used = max + 1;
 
-   {
-      mp_digit u, *tmpa, *tmpb, *tmpc;
-      int i;
-
-      /* alias for digit pointers */
-
-      /* first input */
-      tmpa = a->dp;
+   /* zero the carry */
+   u = 0;
+   for (i = 0; i < min; i++) {
+      /* Compute the sum at one digit, T[i] = A[i] + B[i] + U */
+      c->dp[i] = a->dp[i] + b->dp[i] + u;
 
-      /* second input */
-      tmpb = b->dp;
+      /* U = carry bit of T[i] */
+      u = c->dp[i] >> (mp_digit)MP_DIGIT_BIT;
 
-      /* destination */
-      tmpc = c->dp;
+      /* take away carry bit from T[i] */
+      c->dp[i] &= MP_MASK;
+   }
 
-      /* zero the carry */
-      u = 0;
-      for (i = 0; i < min; i++) {
-         /* Compute the sum at one digit, T[i] = A[i] + B[i] + U */
-         *tmpc = *tmpa++ + *tmpb++ + u;
+   /* now copy higher words if any, that is in A+B
+    * if A or B has more digits add those in
+    */
+   if (min != max) {
+      for (; i < max; i++) {
+         /* T[i] = A[i] + U */
+         c->dp[i] = a->dp[i] + u;
 
          /* U = carry bit of T[i] */
-         u = *tmpc >> (mp_digit)MP_DIGIT_BIT;
+         u = c->dp[i] >> (mp_digit)MP_DIGIT_BIT;
 
          /* take away carry bit from T[i] */
-         *tmpc++ &= MP_MASK;
-      }
-
-      /* now copy higher words if any, that is in A+B
-       * if A or B has more digits add those in
-       */
-      if (min != max) {
-         for (; i < max; i++) {
-            /* T[i] = X[i] + U */
-            *tmpc = x->dp[i] + u;
-
-            /* U = carry bit of T[i] */
-            u = *tmpc >> (mp_digit)MP_DIGIT_BIT;
-
-            /* take away carry bit from T[i] */
-            *tmpc++ &= MP_MASK;
-         }
+         c->dp[i] &= MP_MASK;
       }
+   }
 
-      /* add carry */
-      *tmpc++ = u;
+   /* add carry */
+   c->dp[i] = u;
 
-      /* clear digits above oldused */
-      MP_ZERO_DIGITS(tmpc, olduse - c->used);
-   }
+   /* clear digits above oldused */
+   MP_ZERO_DIGITS(c->dp + c->used, oldused - c->used);
 
    mp_clamp(c);
    return MP_OKAY;
diff --git a/s_mp_sqr.c b/s_mp_sqr.c
index 61106ed..4a20306 100644
--- a/s_mp_sqr.c
+++ b/s_mp_sqr.c
@@ -7,10 +7,8 @@
 mp_err s_mp_sqr(const mp_int *a, mp_int *b)
 {
    mp_int   t;
-   int      ix, iy, pa;
+   int      ix, pa;
    mp_err   err;
-   mp_word  r;
-   mp_digit u, tmpx, *tmpt;
 
    pa = a->used;
    if ((err = mp_init_size(&t, (2 * pa) + 1)) != MP_OKAY) {
@@ -21,10 +19,13 @@ mp_err s_mp_sqr(const mp_int *a, mp_int *b)
    t.used = (2 * pa) + 1;
 
    for (ix = 0; ix < pa; ix++) {
+      mp_digit u;
+      int iy;
+
       /* first calculate the digit at 2*ix */
       /* calculate double precision result */
-      r = (mp_word)t.dp[2*ix] +
-          ((mp_word)a->dp[ix] * (mp_word)a->dp[ix]);
+      mp_word r = (mp_word)t.dp[2*ix] +
+                  ((mp_word)a->dp[ix] * (mp_word)a->dp[ix]);
 
       /* store lower part in result */
       t.dp[ix+ix] = (mp_digit)(r & (mp_word)MP_MASK);
@@ -32,32 +33,27 @@ mp_err s_mp_sqr(const mp_int *a, mp_int *b)
       /* get the carry */
       u           = (mp_digit)(r >> (mp_word)MP_DIGIT_BIT);
 
-      /* left hand side of A[ix] * A[iy] */
-      tmpx        = a->dp[ix];
-
-      /* alias for where to store the results */
-      tmpt        = t.dp + ((2 * ix) + 1);
-
       for (iy = ix + 1; iy < pa; iy++) {
          /* first calculate the product */
-         r       = (mp_word)tmpx * (mp_word)a->dp[iy];
+         r       = (mp_word)a->dp[ix] * (mp_word)a->dp[iy];
 
          /* now calculate the double precision result, note we use
           * addition instead of *2 since it's easier to optimize
           */
-         r       = (mp_word)*tmpt + r + r + (mp_word)u;
+         r       = (mp_word)t.dp[ix + iy] + r + r + (mp_word)u;
 
          /* store lower part */
-         *tmpt++ = (mp_digit)(r & (mp_word)MP_MASK);
+         t.dp[ix + iy] = (mp_digit)(r & (mp_word)MP_MASK);
 
          /* get carry */
          u       = (mp_digit)(r >> (mp_word)MP_DIGIT_BIT);
       }
       /* propagate upwards */
       while (u != 0uL) {
-         r       = (mp_word)*tmpt + (mp_word)u;
-         *tmpt++ = (mp_digit)(r & (mp_word)MP_MASK);
+         r       = (mp_word)t.dp[ix + iy] + (mp_word)u;
+         t.dp[ix + iy] = (mp_digit)(r & (mp_word)MP_MASK);
          u       = (mp_digit)(r >> (mp_word)MP_DIGIT_BIT);
+         ++iy;
       }
    }
 
diff --git a/s_mp_sub.c b/s_mp_sub.c
index bef1fce..05386e5 100644
--- a/s_mp_sub.c
+++ b/s_mp_sub.c
@@ -6,64 +6,51 @@
 /* low level subtraction (assumes |a| > |b|), HAC pp.595 Algorithm 14.9 */
 mp_err s_mp_sub(const mp_int *a, const mp_int *b, mp_int *c)
 {
-   int    olduse, min, max;
-   mp_err err;
-
-   /* find sizes */
-   min = b->used;
-   max = a->used;
+   int oldused = c->used, min = b->used, max = a->used, i;
+   mp_digit u;
 
    /* init result */
    if (c->alloc < max) {
+      mp_err err;
       if ((err = mp_grow(c, max)) != MP_OKAY) {
          return err;
       }
    }
-   olduse = c->used;
-   c->used = max;
-
-   {
-      mp_digit u, *tmpa, *tmpb, *tmpc;
-      int i;
-
-      /* alias for digit pointers */
-      tmpa = a->dp;
-      tmpb = b->dp;
-      tmpc = c->dp;
-
-      /* set carry to zero */
-      u = 0;
-      for (i = 0; i < min; i++) {
-         /* T[i] = A[i] - B[i] - U */
-         *tmpc = (*tmpa++ - *tmpb++) - u;
-
-         /* U = carry bit of T[i]
-          * Note this saves performing an AND operation since
-          * if a carry does occur it will propagate all the way to the
-          * MSB.  As a result a single shift is enough to get the carry
-          */
-         u = *tmpc >> (MP_SIZEOF_BITS(mp_digit) - 1u);
 
-         /* Clear carry from T[i] */
-         *tmpc++ &= MP_MASK;
-      }
+   c->used = max;
 
-      /* now copy higher words if any, e.g. if A has more digits than B  */
-      for (; i < max; i++) {
-         /* T[i] = A[i] - U */
-         *tmpc = *tmpa++ - u;
+   /* set carry to zero */
+   u = 0;
+   for (i = 0; i < min; i++) {
+      /* T[i] = A[i] - B[i] - U */
+      c->dp[i] = (a->dp[i] - b->dp[i]) - u;
+
+      /* U = carry bit of T[i]
+       * Note this saves performing an AND operation since
+       * if a carry does occur it will propagate all the way to the
+       * MSB.  As a result a single shift is enough to get the carry
+       */
+      u = c->dp[i] >> (MP_SIZEOF_BITS(mp_digit) - 1u);
+
+      /* Clear carry from T[i] */
+      c->dp[i] &= MP_MASK;
+   }
 
-         /* U = carry bit of T[i] */
-         u = *tmpc >> (MP_SIZEOF_BITS(mp_digit) - 1u);
+   /* now copy higher words if any, e.g. if A has more digits than B  */
+   for (; i < max; i++) {
+      /* T[i] = A[i] - U */
+      c->dp[i] = a->dp[i] - u;
 
-         /* Clear carry from T[i] */
-         *tmpc++ &= MP_MASK;
-      }
+      /* U = carry bit of T[i] */
+      u = c->dp[i] >> (MP_SIZEOF_BITS(mp_digit) - 1u);
 
-      /* clear digits above used (since we may not have grown result above) */
-      MP_ZERO_DIGITS(tmpc, olduse - c->used);
+      /* Clear carry from T[i] */
+      c->dp[i] &= MP_MASK;
    }
 
+   /* clear digits above used (since we may not have grown result above) */
+   MP_ZERO_DIGITS(c->dp + c->used, oldused - c->used);
+
    mp_clamp(c);
    return MP_OKAY;
 }