aboutsummaryrefslogtreecommitdiff
path: root/sysdeps/ia64/fpu/libm_lgammal.S
blob: d82e4fcaf996fee06f93fc34af31ba79e271d01e (plain)
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
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
5584
5585
5586
5587
5588
5589
5590
5591
5592
5593
5594
5595
5596
5597
5598
5599
5600
5601
5602
5603
5604
5605
5606
5607
5608
5609
5610
5611
5612
5613
5614
5615
5616
5617
5618
5619
5620
5621
5622
5623
5624
5625
5626
5627
5628
5629
5630
5631
5632
5633
5634
5635
5636
5637
5638
5639
5640
5641
5642
5643
5644
5645
5646
5647
5648
5649
5650
5651
5652
5653
5654
5655
5656
5657
5658
5659
5660
5661
5662
5663
5664
5665
5666
5667
5668
5669
5670
5671
5672
5673
5674
5675
5676
5677
5678
5679
5680
5681
5682
5683
5684
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
5703
5704
5705
5706
5707
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
5728
5729
5730
5731
5732
5733
5734
5735
5736
5737
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
5751
5752
5753
5754
5755
5756
5757
5758
5759
5760
5761
5762
5763
5764
5765
5766
5767
5768
5769
5770
5771
5772
5773
5774
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
5785
5786
5787
5788
5789
5790
5791
5792
5793
5794
5795
5796
5797
5798
5799
5800
5801
5802
5803
5804
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
5820
5821
5822
5823
5824
5825
5826
5827
5828
5829
5830
5831
5832
5833
5834
5835
5836
5837
5838
5839
5840
5841
5842
5843
5844
5845
5846
5847
5848
5849
5850
5851
5852
5853
5854
5855
5856
5857
5858
5859
5860
5861
5862
5863
5864
5865
5866
5867
5868
5869
5870
5871
5872
5873
5874
5875
5876
5877
5878
5879
5880
5881
5882
5883
5884
5885
5886
5887
5888
5889
5890
5891
5892
5893
5894
5895
5896
5897
5898
5899
5900
5901
5902
5903
5904
5905
5906
5907
5908
5909
5910
5911
5912
5913
5914
5915
5916
5917
5918
5919
5920
5921
5922
5923
5924
5925
5926
5927
5928
5929
5930
5931
5932
5933
5934
5935
5936
5937
5938
5939
5940
5941
5942
5943
5944
5945
5946
5947
5948
5949
5950
5951
5952
5953
5954
5955
5956
5957
5958
5959
5960
5961
5962
5963
5964
5965
5966
5967
5968
5969
5970
5971
5972
5973
5974
5975
5976
5977
5978
5979
5980
5981
5982
5983
5984
5985
5986
5987
5988
5989
5990
5991
5992
5993
5994
5995
5996
5997
5998
5999
6000
6001
6002
6003
6004
6005
6006
6007
6008
6009
6010
6011
6012
6013
6014
6015
6016
6017
6018
6019
6020
6021
6022
6023
6024
6025
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
6041
6042
6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
6058
6059
6060
6061
6062
6063
6064
6065
6066
6067
6068
6069
6070
6071
6072
6073
6074
6075
6076
6077
6078
6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
6094
6095
6096
6097
6098
6099
6100
6101
6102
6103
6104
6105
6106
6107
6108
6109
6110
6111
6112
6113
6114
6115
6116
6117
6118
6119
6120
6121
6122
6123
6124
6125
6126
6127
6128
6129
6130
6131
6132
6133
6134
6135
6136
6137
6138
6139
6140
6141
6142
6143
6144
6145
6146
6147
6148
6149
6150
6151
6152
6153
6154
6155
6156
6157
6158
6159
6160
6161
6162
6163
6164
6165
6166
6167
6168
6169
6170
6171
6172
6173
6174
6175
6176
6177
6178
6179
6180
6181
6182
6183
6184
6185
6186
6187
6188
6189
6190
6191
6192
6193
6194
6195
6196
6197
6198
6199
6200
6201
6202
6203
6204
6205
6206
6207
6208
6209
6210
6211
6212
6213
6214
6215
6216
6217
6218
6219
6220
6221
6222
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
6249
6250
6251
6252
6253
6254
6255
6256
6257
6258
6259
6260
6261
6262
6263
6264
6265
6266
6267
6268
6269
6270
6271
6272
6273
6274
6275
6276
6277
6278
6279
6280
6281
6282
6283
6284
6285
6286
6287
6288
6289
6290
6291
6292
6293
6294
6295
6296
6297
6298
6299
6300
6301
6302
6303
6304
6305
6306
6307
6308
6309
6310
6311
6312
6313
6314
6315
6316
6317
6318
6319
6320
6321
6322
6323
6324
6325
6326
6327
6328
6329
6330
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
6344
6345
6346
6347
6348
6349
6350
6351
6352
6353
6354
6355
6356
6357
6358
6359
6360
6361
6362
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
6390
6391
6392
6393
6394
6395
6396
6397
6398
6399
6400
6401
6402
6403
6404
6405
6406
6407
6408
6409
6410
6411
6412
6413
6414
6415
6416
6417
6418
6419
6420
6421
6422
6423
6424
6425
6426
6427
6428
6429
6430
6431
6432
6433
6434
6435
6436
6437
6438
6439
6440
6441
6442
6443
6444
6445
6446
6447
6448
6449
6450
6451
6452
6453
6454
6455
6456
6457
6458
6459
6460
6461
6462
6463
6464
6465
6466
6467
6468
6469
6470
6471
6472
6473
6474
6475
6476
6477
6478
6479
6480
6481
6482
6483
6484
6485
6486
6487
6488
6489
6490
6491
6492
6493
6494
6495
6496
6497
6498
6499
6500
6501
6502
6503
6504
6505
6506
6507
6508
6509
6510
6511
6512
6513
6514
6515
6516
6517
6518
6519
6520
6521
6522
6523
6524
6525
6526
6527
6528
6529
6530
6531
6532
6533
6534
6535
6536
6537
6538
6539
6540
6541
6542
6543
6544
6545
6546
6547
6548
6549
6550
6551
6552
6553
6554
6555
6556
6557
6558
6559
6560
6561
6562
6563
6564
6565
6566
6567
6568
6569
6570
6571
6572
6573
6574
6575
6576
6577
6578
6579
6580
6581
6582
6583
6584
6585
6586
6587
6588
6589
6590
6591
6592
6593
6594
6595
6596
6597
6598
6599
6600
6601
6602
6603
6604
6605
6606
6607
6608
6609
6610
6611
6612
6613
6614
6615
6616
6617
6618
6619
6620
6621
6622
6623
6624
6625
6626
6627
6628
6629
6630
6631
6632
6633
6634
6635
6636
6637
6638
6639
6640
6641
6642
6643
6644
6645
6646
6647
6648
6649
6650
6651
6652
6653
6654
6655
6656
6657
6658
6659
6660
6661
6662
6663
6664
6665
6666
6667
6668
6669
6670
6671
6672
6673
6674
6675
6676
6677
6678
6679
6680
6681
6682
6683
6684
6685
6686
6687
6688
6689
6690
6691
6692
6693
6694
6695
6696
6697
6698
6699
6700
6701
6702
6703
6704
6705
6706
6707
6708
6709
6710
6711
6712
6713
6714
6715
6716
6717
6718
6719
6720
6721
6722
6723
6724
6725
6726
6727
6728
6729
6730
6731
6732
6733
6734
6735
6736
6737
6738
6739
6740
6741
6742
6743
6744
6745
6746
6747
6748
6749
6750
6751
6752
6753
6754
6755
6756
6757
6758
6759
6760
6761
6762
6763
6764
6765
6766
6767
6768
6769
6770
6771
6772
6773
6774
6775
6776
6777
6778
6779
6780
6781
6782
6783
6784
6785
6786
6787
6788
6789
6790
6791
6792
6793
6794
6795
6796
6797
6798
6799
6800
6801
6802
6803
6804
6805
6806
6807
6808
6809
6810
6811
6812
6813
6814
6815
6816
6817
6818
6819
6820
6821
6822
6823
6824
6825
6826
6827
6828
6829
6830
6831
6832
6833
6834
6835
6836
6837
6838
6839
6840
6841
6842
6843
6844
6845
6846
6847
6848
6849
6850
6851
6852
6853
6854
6855
6856
6857
6858
6859
6860
6861
6862
6863
6864
6865
6866
6867
6868
6869
6870
6871
6872
6873
6874
6875
6876
6877
6878
6879
6880
6881
6882
6883
6884
6885
6886
6887
6888
6889
6890
6891
6892
6893
6894
6895
6896
6897
6898
6899
6900
6901
6902
6903
6904
6905
6906
6907
6908
6909
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
6931
6932
6933
6934
6935
6936
6937
6938
6939
6940
6941
6942
6943
6944
6945
6946
6947
6948
6949
6950
6951
6952
6953
6954
6955
6956
6957
6958
6959
6960
6961
6962
6963
6964
6965
6966
6967
6968
6969
6970
6971
6972
6973
6974
6975
6976
6977
6978
6979
6980
6981
6982
6983
6984
6985
6986
6987
6988
6989
6990
6991
6992
6993
6994
6995
6996
6997
6998
6999
7000
7001
7002
7003
7004
7005
7006
7007
7008
7009
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
7026
7027
7028
7029
7030
7031
7032
7033
7034
7035
7036
7037
7038
7039
7040
7041
7042
7043
7044
7045
7046
7047
7048
7049
7050
7051
7052
7053
7054
7055
7056
7057
7058
7059
7060
7061
7062
7063
7064
7065
7066
7067
7068
7069
7070
7071
7072
7073
7074
7075
7076
7077
7078
7079
7080
7081
7082
7083
7084
7085
7086
7087
7088
7089
7090
7091
7092
7093
7094
7095
7096
7097
7098
7099
7100
7101
7102
7103
7104
7105
7106
7107
7108
7109
7110
7111
7112
7113
7114
7115
7116
7117
7118
7119
7120
7121
7122
7123
7124
7125
7126
7127
7128
7129
7130
7131
7132
7133
7134
7135
7136
7137
7138
7139
7140
7141
7142
7143
7144
7145
7146
7147
7148
7149
7150
7151
7152
7153
7154
7155
7156
7157
7158
7159
7160
7161
7162
7163
7164
7165
7166
7167
7168
7169
7170
7171
7172
7173
7174
7175
7176
7177
7178
7179
7180
7181
7182
7183
7184
7185
7186
7187
7188
7189
7190
7191
7192
7193
7194
7195
7196
7197
7198
7199
7200
7201
7202
7203
7204
7205
7206
7207
7208
7209
7210
7211
7212
7213
7214
7215
7216
7217
7218
7219
7220
7221
7222
7223
7224
7225
7226
7227
7228
7229
7230
7231
7232
7233
7234
7235
7236
7237
7238
7239
7240
7241
7242
7243
7244
7245
7246
7247
7248
7249
7250
7251
7252
7253
7254
7255
7256
7257
7258
7259
7260
7261
7262
7263
7264
7265
7266
7267
7268
7269
7270
7271
7272
7273
7274
7275
7276
7277
7278
7279
7280
7281
7282
7283
7284
7285
7286
7287
7288
7289
7290
7291
7292
7293
7294
7295
7296
7297
7298
7299
7300
7301
7302
7303
7304
7305
7306
7307
7308
7309
7310
7311
7312
7313
7314
7315
7316
7317
7318
7319
7320
7321
7322
7323
7324
7325
7326
7327
7328
7329
7330
7331
7332
7333
7334
7335
7336
7337
7338
7339
7340
7341
7342
7343
7344
7345
7346
7347
7348
7349
7350
7351
7352
7353
7354
7355
7356
7357
7358
7359
7360
7361
7362
7363
7364
7365
7366
7367
7368
7369
7370
7371
7372
7373
7374
7375
7376
7377
7378
7379
7380
7381
7382
7383
7384
7385
7386
7387
7388
7389
7390
7391
7392
7393
7394
7395
7396
7397
7398
7399
7400
7401
7402
7403
7404
7405
7406
7407
7408
7409
7410
7411
7412
7413
7414
7415
7416
7417
7418
7419
7420
7421
7422
7423
7424
7425
7426
7427
7428
7429
7430
7431
7432
7433
7434
7435
7436
7437
7438
7439
7440
7441
7442
7443
7444
7445
7446
7447
7448
7449
7450
7451
7452
7453
7454
7455
7456
7457
7458
7459
7460
7461
7462
7463
7464
7465
7466
7467
7468
7469
7470
7471
7472
7473
7474
7475
7476
7477
7478
7479
7480
7481
7482
7483
7484
7485
7486
7487
7488
7489
7490
7491
7492
7493
7494
7495
7496
7497
7498
7499
7500
7501
7502
7503
7504
7505
7506
7507
7508
7509
7510
7511
7512
7513
7514
7515
7516
7517
7518
7519
7520
7521
7522
7523
7524
7525
7526
7527
7528
7529
7530
7531
7532
7533
7534
7535
7536
7537
7538
7539
7540
7541
7542
7543
7544
7545
7546
7547
7548
7549
7550
7551
7552
7553
7554
7555
7556
7557
7558
7559
7560
7561
7562
7563
7564
7565
7566
7567
7568
7569
7570
7571
7572
7573
7574
7575
7576
7577
7578
7579
7580
7581
7582
7583
7584
7585
7586
7587
7588
7589
7590
7591
7592
7593
7594
7595
7596
7597
7598
7599
7600
7601
7602
7603
7604
7605
7606
7607
7608
7609
7610
7611
7612
7613
7614
7615
7616
7617
7618
7619
7620
7621
7622
7623
7624
7625
7626
7627
7628
7629
7630
7631
7632
7633
7634
7635
7636
7637
7638
7639
7640
7641
7642
7643
7644
7645
7646
7647
7648
7649
7650
7651
7652
7653
7654
7655
7656
7657
7658
7659
7660
7661
7662
7663
7664
7665
7666
7667
7668
7669
7670
7671
7672
7673
7674
7675
7676
7677
.file "libm_lgammal.s"


// Copyright (c) 2002 - 2005, Intel Corporation
// All rights reserved.
//
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
//
// * Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
//
// * The name of Intel Corporation may not be used to endorse or promote
// products derived from this software without specific prior written
// permission.

// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,INCLUDING,BUT NOT
// LIMITED TO,THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL INTEL OR ITS
// CONTRIBUTORS BE LIABLE FOR ANY DIRECT,INDIRECT,INCIDENTAL,SPECIAL,
// EXEMPLARY,OR CONSEQUENTIAL DAMAGES (INCLUDING,BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,DATA,OR
// PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
// OF LIABILITY,WHETHER IN CONTRACT,STRICT LIABILITY OR TORT (INCLUDING
// NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
// SOFTWARE,EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// Intel Corporation is the author of this code,and requests that all
// problem reports or change requests be submitted to it directly at
// http://www.intel.com/software/products/opensource/libraries/num.htm.
//
//*********************************************************************
//
// History:
// 03/28/02  Original version
// 05/20/02  Cleaned up namespace and sf0 syntax
// 08/21/02  Added support of SIGN(GAMMA(x)) calculation
// 09/26/02  Algorithm description improved
// 10/21/02  Now it returns SIGN(GAMMA(x))=-1 for negative zero
// 02/10/03  Reordered header: .section, .global, .proc, .align
// 03/31/05  Reformatted delimiters between data tables
//
//*********************************************************************
//
// Function: __libm_lgammal(long double x, int* signgam, int szsigngam)
// computes the principal value of the logarithm of the GAMMA function
// of x. Signum of GAMMA(x) is stored to memory starting at the address
// specified by the signgam.
//
//*********************************************************************
//
// Resources Used:
//
//    Floating-Point Registers: f8 (Input and Return Value)
//                              f9-f15
//                              f32-f127
//
//    General Purpose Registers:
//      r2, r3, r8-r11, r14-r31
//      r32-r65
//      r66-r69 (Used to pass arguments to error handling routine)
//
//    Predicate Registers:      p6-p15
//
//*********************************************************************
//
// IEEE Special Conditions:
//
//    __libm_lgammal(+inf) = +inf
//    __libm_lgammal(-inf) = QNaN
//    __libm_lgammal(+/-0) = +inf
//    __libm_lgammal(x<0, x - integer) = QNaN
//    __libm_lgammal(SNaN) = QNaN
//    __libm_lgammal(QNaN) = QNaN
//
//*********************************************************************
//
// ALGORITHM DESCRIPTION
//
// Below we suppose that there is log(z) function which takes an long
// double argument and returns result as a pair of long double numbers
// lnHi and lnLo (such that sum lnHi + lnLo provides ~80 correct bits
// of significand). Algorithm description for such log(z) function
// see below.
// Also, it this algorithm description we use the following notational
// conventions:
// a) pair A = (Ahi, Alo) means number A represented as sum of Ahi and Alo
// b) C = A + B = (Ahi, Alo) + (Bhi, Blo) means multi-precision addition.
//    The result would be C = (Chi, Clo). Notice, that Clo shouldn't be
//    equal to Alo + Blo
// c) D = A*B = (Ahi, Alo)*(Bhi, Blo) = (Dhi, Dlo) multi-precisiion
//    multiplication.
//
// So, lgammal has the following computational paths:
// 1) |x| < 0.5
//    P = A1*|x| + A2*|x|^2 + ... + A22*|x|^22
//    A1, A2, A3 represented as a sum of two double precision
//    numbers and multi-precision computations are used for 3 higher
//    terms of the polynomial. We get polynomial as a sum of two
//    double extended numbers: P = (Phi, Plo)
//    1.1) x > 0
//         lgammal(x) = P - log(|x|) = (Phi, Plo) - (lnHi(|x|), lnLo(|x|))
//    1.2) x < 0
//         lgammal(x) = -P - log(|x|) - log(sin(Pi*x)/(Pi*x))
//         P and log(|x|) are computed by the same way as in 1.1;
//         - log(sin(Pi*x)/(Pi*x)) is approximated by a polynomial Plnsin.
//         Plnsin:= fLnSin2*|x|^2 + fLnSin4*|x|^4 + ... + fLnSin36*|x|^36
//         The first coefficient of Plnsin is represented as sum of two
//         double precision numbers (fLnSin2, fLnSin2L). Multi-precision
//         computations for higher two terms of Plnsin are used.
//         So, the final result is reconstructed by the following formula
//         lgammal(x) = (-(Phi, Plo) - (lnHi(|x|), lnLo(|x|))) -
//                      - (PlnsinHi,PlnsinLo)
//
// 2)    0.5 <= x <   0.75  -> t = x - 0.625
//     -0.75 <  x <= -0.5   -> t = x + 0.625
//      2.25 <= x <   4.0   -> t = x/2 - 1.5
//       4.0 <= x <   8.0   -> t = x/4 - 1.5
//       -0.5 < x <= -0.40625 -> t = x + 0.5
//       -2.6005859375 < x <= -2.5 -> t = x + 2.5
//       1.3125 <= x < 1.5625 -> t = x - LOC_MIN, where LOC_MIN is point in
//                                   which lgammal has local minimum. Exact
//                                   value can be found in the table below,
//                                   approximate value is ~1.46
//
//    lgammal(x) is approximated by the polynomial of 25th degree: P25(t)
//    P25(t) = A0 + A1*t + ... + A25*t^25 = (Phi, Plo) + t^4*P21(t),
//    where
//    (Phi, Plo) is sum of four highest terms of the polynomial P25(t):
//    (Phi, Plo) = ((A0, A0L) + (A1, A1L)*t) + t^2 *((A2, A2L) + (A3, A3L)*t),
//    (Ai, AiL) - coefficients represented as pairs of DP numbers.
//
//    P21(t) = (PolC(t)*t^8 + PolD(t))*t^8 + PolE(t),
//    where
//    PolC(t) = C21*t^5 + C20*t^4 + ... + C16,
//    C21 = A25, C20 = A24, ..., C16 = A20
//
//    PolD(t) = D7*t^7 + D6*t^6 + ... + D0,
//    D7 = A19, D6 = A18, ..., D0 = A12
//
//    PolE(t) = E7*t^7 + E6*t^6 + ... + E0,
//    E7 = A11, E6 = A10, ..., E0 = A4
//
//    Cis and Dis are represented as double precision numbers,
//    Eis are represented as double extended numbers.
//
// 3) 0.75 <=  x < 1.3125   -> t = x - 1.0
//    1.5625 <= x < 2.25   -> t = x - 2.0
//    lgammal(x) is approximated by the polynomial of 25th degree: P25(t)
//    P25(t) = A1*t + ... + A25*t^25, and computations are carried out
//    by similar way as in the previous case
//
// 4) 10.0 < x <= Overflow Bound ("positive Sterling" range)
//    lgammal(x) is approximated using Sterling's formula:
//    lgammal(x) ~ ((x*(lnHi(x) - 1, lnLo(x))) - 0.5*(lnHi(x), lnLo(x))) +
//                 + ((Chi, Clo) + S(1/x))
//    where
//    C = (Chi, Clo) - pair of double precision numbers representing constant
//    0.5*ln(2*Pi);
//    S(1/x) = 1/x * (B2 + B4*(1/x)^2 + ... + B20*(1/x)^18), B2, ..., B20 are
//    Bernulli numbers. S is computed in native precision and then added to
//    Clo;
//    lnHi(x) - 1 is computed in native precision and the multiprecision
//    multiplication (x, 0) *(lnHi(x) - 1, lnLo(x)) is used.
//
// 5) -INF < x <= -2^63, any negative integer < 0
//    All numbers in this range are integers -> error handler is called
//
// 6) -2^63 < x <= -0.75 ("negative Sterling" range), x is "far" from root,
//    lgammal(-t) for positive t is approximated using the following formula:
//    lgammal(-t) = -lgammal(t)-log(t)-log(|dT|)+log(sin(Pi*|dT|)/(Pi*|dT|))
//        where dT = -t -round_to_nearest_integer(-t)
//    Last item is approximated by the same polynomial as described in 1.2.
//    We split the whole range into three subranges due to different ways of
//    approximation of the first terms.
//    6.1) -2^63 < x < -6.0 ("negative Sterling" range)
//       lgammal(t) is approximated exactly as in #4. The only difference that
//       for -13.0 < x < -6.0 subrange instead of Bernulli numbers we use their
//       minimax approximation on this range.
//       log(t), log(|dT|) are approximated by the log routine mentioned above.
//    6.2) -6.0 < x <= -0.75, |x + 1|> 2^(-7)
//       log(t), log(|dT|) are approximated by the log routine mentioned above,
//       lgammal(t) is approximated by polynomials of the 25th degree similar
//       to ones from #2. Arguments z of the polynomials are as follows
//       a) 0.75 <= t < 1.0 - 2^(-7),  z = 2*t - 1.5
//       b) 1.0 - 2^(-7)  < t < 2.0,   z = t - 1.5
//       c) 2.0  < t < 3.0,   z = t/2 - 1.5
//       d) 3.0  < t < 4.0,   z = t/2 - 1.5. Notice, that range reduction is
//          the same as in case c) but the set of coefficients is different
//       e) 4.0  < t < 6.0,   z = t/4 - 1.5
//    6.3) |x + 1| <= 2^(-7)
//       log(1 + (x-1)) is approximated by Taylor series,
//       log(sin(Pi*|dT|)/(Pi*|dT|)) is still approximated by polynomial but
//       it has just 4th degree.
//       log(|dT|) is approximated by the log routine mentioned above.
//       lgammal(-x) is approximated by polynomial of 8th degree from (-x + 1).
//
// 7) -20.0 < x < -2.0, x falls in root "neighbourhood".
//    "Neighbourhood" means that |lgammal(x)| < epsilon, where epsilon is
//    different for every root (and it is stored in the table), but typically
//    it is ~ 0.15. There are 35 roots significant from "double extended"
//    point of view. We split all the roots into two subsets: "left" and "right"
//    roots. Considering [-(N+1), -N] range we call root as "left" one if it
//    lies closer to -(N+1) and "right" otherwise. There is no "left" root in
//    the [-20, -19] range (it exists, but is insignificant for double extended
//    precision). To determine if x falls in root "neighbourhood" we store
//    significands of all the 35 roots as well as epsilon values (expressed
//    by the left and right bound).
//    In these ranges we approximate lgammal(x) by polynomial series of 19th
//    degree:
//    lgammal(x) = P19(t) = A0 + A1*t + ...+ A19*t^19, where t = x - EDP_Root,
//    EDP_Root is the exact value of the corresponding root rounded to double
//    extended precision. So, we have 35 different polynomials which make our
//    table rather big. We may hope that x falls in root "neighbourhood"
//    quite rarely -> ther might be no need in frequent use of different
//    polynomials.
//    A0, A1, A2, A3 are represented as pairs of double precision numbers,
//    A4, A5 are long doubles, and to decrease the size of the table we
//    keep the rest of coefficients in just double precision
//
//*********************************************************************
// Algorithm for log(X) = (lnHi(X), lnLo(X))
//
//   ALGORITHM
//
//   Here we use a table lookup method. The basic idea is that in
//   order to compute logl(Arg) for an argument Arg in [1,2), we
//   construct a value G such that G*Arg is close to 1 and that
//   logl(1/G) is obtainable easily from a table of values calculated
//   beforehand. Thus
//
//      logl(Arg) = logl(1/G) + logl(G*Arg)
//                = logl(1/G) + logl(1 + (G*Arg - 1))
//
//   Because |G*Arg - 1| is small, the second term on the right hand
//   side can be approximated by a short polynomial. We elaborate
//   this method in four steps.
//
//   Step 0: Initialization
//
//   We need to calculate logl( X ). Obtain N, S_hi such that
//
//      X = 2^N * S_hi exactly
//
//   where S_hi in [1,2)
//
//   Step 1: Argument Reduction
//
//   Based on S_hi, obtain G_1, G_2, G_3 from a table and calculate
//
//      G := G_1 * G_2 * G_3
//      r := (G * S_hi - 1)
//
//   These G_j's have the property that the product is exactly
//   representable and that |r| < 2^(-12) as a result.
//
//   Step 2: Approximation
//
//
//   logl(1 + r) is approximated by a short polynomial poly(r).
//
//   Step 3: Reconstruction
//
//
//   Finally, logl( X ) is given by
//
//   logl( X )   =   logl( 2^N * S_hi )
//                 ~=~  N*logl(2) + logl(1/G) + logl(1 + r)
//                 ~=~  N*logl(2) + logl(1/G) + poly(r).
//
//   IMPLEMENTATION
//
//   Step 0. Initialization
//   ----------------------
//
//   Z := X
//   N := unbaised exponent of Z
//   S_hi := 2^(-N) * Z
//
//   Step 1. Argument Reduction
//   --------------------------
//
//   Let
//
//      Z = 2^N * S_hi = 2^N * 1.d_1 d_2 d_3 ... d_63
//
//   We obtain G_1, G_2, G_3 by the following steps.
//
//
//      Define          X_0 := 1.d_1 d_2 ... d_14. This is extracted
//                      from S_hi.
//
//      Define          A_1 := 1.d_1 d_2 d_3 d_4. This is X_0 truncated
//                      to lsb = 2^(-4).
//
//      Define          index_1 := [ d_1 d_2 d_3 d_4 ].
//
//      Fetch           Z_1 := (1/A_1) rounded UP in fixed point with
//      fixed point     lsb = 2^(-15).
//                      Z_1 looks like z_0.z_1 z_2 ... z_15
//                      Note that the fetching is done using index_1.
//                      A_1 is actually not needed in the implementation
//                      and is used here only to explain how is the value
//                      Z_1 defined.
//
//      Fetch           G_1 := (1/A_1) truncated to 21 sig. bits.
//      floating pt.    Again, fetching is done using index_1. A_1
//                      explains how G_1 is defined.
//
//      Calculate       X_1 := X_0 * Z_1 truncated to lsb = 2^(-14)
//                           = 1.0 0 0 0 d_5 ... d_14
//                      This is accomplished by integer multiplication.
//                      It is proved that X_1 indeed always begin
//                      with 1.0000 in fixed point.
//
//
//      Define          A_2 := 1.0 0 0 0 d_5 d_6 d_7 d_8. This is X_1
//                      truncated to lsb = 2^(-8). Similar to A_1,
//                      A_2 is not needed in actual implementation. It
//                      helps explain how some of the values are defined.
//
//      Define          index_2 := [ d_5 d_6 d_7 d_8 ].
//
//      Fetch           Z_2 := (1/A_2) rounded UP in fixed point with
//      fixed point     lsb = 2^(-15). Fetch done using index_2.
//                      Z_2 looks like z_0.z_1 z_2 ... z_15
//
//      Fetch           G_2 := (1/A_2) truncated to 21 sig. bits.
//      floating pt.
//
//      Calculate       X_2 := X_1 * Z_2 truncated to lsb = 2^(-14)
//                           = 1.0 0 0 0 0 0 0 0 d_9 d_10 ... d_14
//                      This is accomplished by integer multiplication.
//                      It is proved that X_2 indeed always begin
//                      with 1.00000000 in fixed point.
//
//
//      Define          A_3 := 1.0 0 0 0 0 0 0 0 d_9 d_10 d_11 d_12 d_13 1.
//                      This is 2^(-14) + X_2 truncated to lsb = 2^(-13).
//
//      Define          index_3 := [ d_9 d_10 d_11 d_12 d_13 ].
//
//      Fetch           G_3 := (1/A_3) truncated to 21 sig. bits.
//      floating pt.    Fetch is done using index_3.
//
//      Compute         G := G_1 * G_2 * G_3.
//
//      This is done exactly since each of G_j only has 21 sig. bits.
//
//      Compute
//
//               r := (G*S_hi - 1)
//
//
//  Step 2. Approximation
//  ---------------------
//
//   This step computes an approximation to logl( 1 + r ) where r is the
//   reduced argument just obtained. It is proved that |r| <= 1.9*2^(-13);
//   thus logl(1+r) can be approximated by a short polynomial:
//
//      logl(1+r) ~=~ poly = r + Q1 r^2 + ... + Q4 r^5
//
//
//  Step 3. Reconstruction
//  ----------------------
//
//   This step computes the desired result of logl(X):
//
//      logl(X)  =   logl( 2^N * S_hi )
//               =   N*logl(2) + logl( S_hi )
//               =   N*logl(2) + logl(1/G) +
//                      logl(1 + G*S_hi - 1 )
//
//   logl(2), logl(1/G_j) are stored as pairs of (single,double) numbers:
//   log2_hi, log2_lo, log1byGj_hi, log1byGj_lo. The high parts are
//   single-precision numbers and the low parts are double precision
//   numbers. These have the property that
//
//      N*log2_hi + SUM ( log1byGj_hi )
//
//   is computable exactly in double-extended precision (64 sig. bits).
//   Finally
//
//      lnHi(X) := N*log2_hi + SUM ( log1byGj_hi )
//      lnLo(X) := poly_hi + [ poly_lo +
//              ( SUM ( log1byGj_lo ) + N*log2_lo ) ]
//
//
//*********************************************************************
// General Purpose Registers
// scratch registers
rPolDataPtr              = r2
rLnSinDataPtr            = r3
rExpX                    = r8
rSignifX                 = r9
rDelta                   = r10
rSignExpX                = r11
GR_ad_z_1                = r14
r17Ones                  = r15
GR_Index1                = r16
rSignif1andQ             = r17
GR_X_0                   = r18
GR_X_1                   = r19
GR_X_2                   = r20
GR_Z_1                   = r21
GR_Z_2                   = r22
GR_N                     = r23
rExpHalf                 = r24
rExp8                    = r25
rX0Dx                    = r25
GR_ad_tbl_1              = r26
GR_ad_tbl_2              = r27
GR_ad_tbl_3              = r28
GR_ad_q                  = r29
GR_ad_z_1                = r30
GR_ad_z_2                = r31
// stacked registers
rPFS_SAVED               = r32
GR_ad_z_3                = r33
rSgnGamAddr              = r34
rSgnGamSize              = r35
rLogDataPtr              = r36
rZ1offsett               = r37
rTmpPtr                  = r38
rTmpPtr2                 = r39
rTmpPtr3                 = r40
rExp2                    = r41
rExp2tom7                = r42
rZ625                    = r42
rExpOne                  = r43
rNegSingularity          = r44
rXint                    = r45
rTbl1Addr                = r46
rTbl2Addr                = r47
rTbl3Addr                = r48
rZ2Addr                  = r49
rRootsAddr               = r50
rRootsBndAddr            = r51
rRoot                    = r52
rRightBound              = r53
rLeftBound               = r54
rSignifDx                = r55
rBernulliPtr             = r56
rLnSinTmpPtr             = r56
rIndex1Dx                = r57
rIndexPol                = r58
GR_Index3                = r59
GR_Index2                = r60
rSgnGam                  = r61
rXRnd                    = r62

GR_SAVE_B0               = r63
GR_SAVE_GP               = r64
GR_SAVE_PFS              = r65
// output parameters when calling error handling routine
GR_Parameter_X           = r66
GR_Parameter_Y           = r67
GR_Parameter_RESULT      = r68
GR_Parameter_TAG         = r69

//********************************************************************
// Floating Point Registers
// CAUTION: due to the lack of registers there exist (below in the code)
// sometimes "unconventional" use of declared registers
//
fAbsX                    = f6
fDelX4                   = f6
fSignifX                 = f7
// macros for error handling routine
FR_X                     = f10 // first argument
FR_Y                     = f1  // second argument (lgammal has just one)
FR_RESULT                = f8  // result

// First 7 Bernulli numbers
fB2                      = f9
fLnDeltaL                = f9
fXSqr                    = f9
fB4                      = f10
fX4                      = f10
fB6                      = f11
fX6                      = f11
fB8                      = f12
fXSqrL                   = f12
fB10                     = f13
fRes7H                   = f13
fB12                     = f14
fRes7L                   = f14
fB14                     = f15

// stack registers
// Polynomial coefficients: A0, ..., A25
fA0                      = f32
fA0L                     = f33
fInvXL                   = f33
fA1                      = f34
fA1L                     = f35
fA2                      = f36
fA2L                     = f37
fA3                      = f38
fA3L                     = f39
fA4                      = f40
fA4L                     = f41
fRes6H                   = f41
fA5                      = f42
fB2L                     = f42
fA5L                     = f43
fMinNegStir              = f43
fRes6L                   = f43
fA6                      = f44
fMaxNegStir              = f44
fA7                      = f45
fLnDeltaH                = f45
fA8                      = f46
fBrnL                    = f46
fA9                      = f47
fBrnH                    = f47
fA10                     = f48
fRes5L                   = f48
fA11                     = f49
fRes5H                   = f49
fA12                     = f50
fDx6                     = f50
fA13                     = f51
fDx8                     = f51
fA14                     = f52
fDx4                     = f52
fA15                     = f53
fYL                      = f53
fh3Dx                    = f53
fA16                     = f54
fYH                      = f54
fH3Dx                    = f54
fA17                     = f55
fResLnDxL                = f55
fG3Dx                    = f55
fA18                     = f56
fResLnDxH                = f56
fh2Dx                    = f56
fA19                     = f57
fFloatNDx                = f57
fA20                     = f58
fPolyHiDx                = f58
fhDx                     = f58
fA21                     = f59
fRDxCub                  = f59
fHDx                     = f59
fA22                     = f60
fRDxSq                   = f60
fGDx                     = f60
fA23                     = f61
fPolyLoDx                = f61
fInvX3                   = f61
fA24                     = f62
fRDx                     = f62
fInvX8                   = f62
fA25                     = f63
fInvX4                   = f63
fPol                     = f64
fPolL                    = f65
// Coefficients of ln(sin(Pi*x)/Pi*x)
fLnSin2                  = f66
fLnSin2L                 = f67
fLnSin4                  = f68
fLnSin6                  = f69
fLnSin8                  = f70
fLnSin10                 = f71
fLnSin12                 = f72
fLnSin14                 = f73
fLnSin16                 = f74
fLnSin18                 = f75
fDelX8                   = f75
fLnSin20                 = f76
fLnSin22                 = f77
fDelX6                   = f77
fLnSin24                 = f78
fLnSin26                 = f79
fLnSin28                 = f80
fLnSin30                 = f81
fhDelX                   = f81
fLnSin32                 = f82
fLnSin34                 = f83
fLnSin36                 = f84
fXint                    = f85
fDxSqr                   = f85
fRes3L                   = f86
fRes3H                   = f87
fRes4H                   = f88
fRes4L                   = f89
fResH                    = f90
fResL                    = f91
fDx                      = f92
FR_MHalf                 = f93
fRes1H                   = f94
fRes1L                   = f95
fRes2H                   = f96
fRes2L                   = f97
FR_FracX                 = f98
fRcpX                    = f99
fLnSinH                  = f99
fTwo                     = f100
fMOne                    = f100
FR_G                     = f101
FR_H                     = f102
FR_h                     = f103
FR_G2                    = f104
FR_H2                    = f105
FR_poly_lo               = f106
FR_poly_hi               = f107
FR_h2                    = f108
FR_rsq                   = f109
FR_r                     = f110
FR_log2_hi               = f111
FR_log2_lo               = f112
fFloatN                  = f113
FR_Q4                    = f114
FR_G3                    = f115
FR_H3                    = f116
FR_h3                    = f117
FR_Q3                    = f118
FR_Q2                    = f119
FR_Q1                    = f120
fThirteen                = f121
fSix                     = f121
FR_rcub                  = f121
// Last three Bernulli numbers
fB16                     = f122
fB18                     = f123
fB20                     = f124
fInvX                    = f125
fLnSinL                  = f125
fDxSqrL                  = f126
fFltIntX                 = f126
fRoot                    = f127
fNormDx                  = f127

// Data tables
//==============================================================
RODATA
// ************* DO NOT CHANGE THE ORDER OF THESE TABLES *************
.align 16
LOCAL_OBJECT_START(lgammal_right_roots_data)
// List of all right roots themselves
data8 0x9D3FE4B007C360AB, 0x0000C000 // Range [-3, -2]
data8 0xC9306DE4F2CD7BEE, 0x0000C000 // Range [-4, -3]
data8 0x814273C2CCAC0618, 0x0000C001 // Range [-5, -4]
data8 0xA04352BF85B6C865, 0x0000C001 // Range [-6, -5]
data8 0xC00B592C4BE4676C, 0x0000C001 // Range [-7, -6]
data8 0xE0019FEF6FF0F5BF, 0x0000C001 // Range [-8, -7]
data8 0x80001A01459FC9F6, 0x0000C002 // Range [-9, -8]
data8 0x900002E3BB47D86D, 0x0000C002 // Range [-10, -9]
data8 0xA0000049F93BB992, 0x0000C002 // Range [-11, -10]
data8 0xB0000006B9915316, 0x0000C002 // Range [-12, -11]
data8 0xC00000008F76C773, 0x0000C002 // Range [-13, -12]
data8 0xD00000000B09230A, 0x0000C002 // Range [-14, -13]
data8 0xE000000000C9CBA5, 0x0000C002 // Range [-15, -14]
data8 0xF0000000000D73FA, 0x0000C002 // Range [-16, -15]
data8 0x8000000000006BA0, 0x0000C003 // Range [-17, -16]
data8 0x8800000000000655, 0x0000C003 // Range [-18, -17]
data8 0x900000000000005A, 0x0000C003 // Range [-19, -18]
data8 0x9800000000000005, 0x0000C003 // Range [-20, -19]
// List of bounds of ranges with special polynomial approximation near root
// Only significands of bounds are actually stored
data8 0xA000000000000000, 0x9800000000000000 // Bounds for root on [-3, -2]
data8 0xCAB88035C5EFBB41, 0xC7E05E31F4B02115 // Bounds for root on [-4, -3]
data8 0x817831B899735C72, 0x8114633941B8053A // Bounds for root on [-5, -4]
data8 0xA04E8B34C6AA9476, 0xA039B4A42978197B // Bounds for root on [-6, -5]
data8 0xC00D3D5E588A78A9, 0xC009BA25F7E858A6 // Bounds for root on [-7, -6]
data8 0xE001E54202991EB4, 0xE001648416CE897F // Bounds for root on [-8, -7]
data8 0x80001E56D13A6B9F, 0x8000164A3BAD888A // Bounds for root on [-9, -8]
data8 0x9000035F0529272A, 0x9000027A0E3D94F0 // Bounds for root on [-10, -9]
data8 0xA00000564D705880, 0xA000003F67EA0CC7 // Bounds for root on [-11, -10]
data8 0xB0000007D87EE0EF, 0xB0000005C3A122A5 // Bounds for root on [-12, -11]
data8 0xC0000000A75FE8B1, 0xC00000007AF818AC // Bounds for root on [-13, -12]
data8 0xD00000000CDFFE36, 0xD000000009758BBF // Bounds for root on [-14, -13]
data8 0xE000000000EB6D96, 0xE000000000ACF7B2 // Bounds for root on [-15, -14]
data8 0xF0000000000FB1F9, 0xF0000000000B87FB // Bounds for root on [-16, -15]
data8 0x8000000000007D90, 0x8000000000005C40 // Bounds for root on [-17, -16]
data8 0x8800000000000763, 0x880000000000056D // Bounds for root on [-18, -17]
data8 0x9000000000000069, 0x900000000000004D // Bounds for root on [-19, -18]
data8 0x9800000000000006, 0x9800000000000005 // Bounds for root on [-20, -19]
// List of all left roots themselves
data8 0xAFDA0850DEC8065E, 0x0000C000 // Range [-3, -2]
data8 0xFD238AA3E17F285C, 0x0000C000 // Range [-4, -3]
data8 0x9FBABBD37757E6A2, 0x0000C001 // Range [-5, -4]
data8 0xBFF497AC8FA06AFC, 0x0000C001 // Range [-6, -5]
data8 0xDFFE5FBB5C377FE8, 0x0000C001 // Range [-7, -6]
data8 0xFFFFCBFC0ACE7879, 0x0000C001 // Range [-8, -7]
data8 0x8FFFFD1C425E8100, 0x0000C002 // Range [-9, -8]
data8 0x9FFFFFB606BDFDCD, 0x0000C002 // Range [-10, -9]
data8 0xAFFFFFF9466E9F1B, 0x0000C002 // Range [-11, -10]
data8 0xBFFFFFFF70893874, 0x0000C002 // Range [-12, -11]
data8 0xCFFFFFFFF4F6DCF6, 0x0000C002 // Range [-13, -12]
data8 0xDFFFFFFFFF36345B, 0x0000C002 // Range [-14, -13]
data8 0xEFFFFFFFFFF28C06, 0x0000C002 // Range [-15, -14]
data8 0xFFFFFFFFFFFF28C0, 0x0000C002 // Range [-16, -15]
data8 0x87FFFFFFFFFFF9AB, 0x0000C003 // Range [-17, -16]
data8 0x8FFFFFFFFFFFFFA6, 0x0000C003 // Range [-18, -17]
data8 0x97FFFFFFFFFFFFFB, 0x0000C003 // Range [-19, -18]
data8 0x0000000000000000, 0x00000000 // pad to keep logic in the main path
// List of bounds of ranges with special polynomial approximation near root
// Only significands of bounds are actually stored
data8 0xB235880944CC758E, 0xADD2F1A9FBE76C8B // Bounds for root on [-3, -2]
data8 0xFD8E7844F307B07C, 0xFCA655C2152BDE4D // Bounds for root on [-4, -3]
data8 0x9FC4D876EE546967, 0x9FAEE4AF68BC4292 // Bounds for root on [-5, -4]
data8 0xBFF641FFBFCC44F1, 0xBFF2A47919F4BA89 // Bounds for root on [-6, -5]
data8 0xDFFE9C803DEFDD59, 0xDFFE18932EB723FE // Bounds for root on [-7, -6]
data8 0xFFFFD393FA47AFC3, 0xFFFFC317CF638AE1 // Bounds for root on [-8, -7]
data8 0x8FFFFD8840279925, 0x8FFFFC9DCECEEE92 // Bounds for root on [-9, -8]
data8 0x9FFFFFC0D34E2AF8, 0x9FFFFFA9619AA3B7 // Bounds for root on [-10, -9]
data8 0xAFFFFFFA41C18246, 0xAFFFFFF82025A23C // Bounds for root on [-11, -10]
data8 0xBFFFFFFF857ACB4E, 0xBFFFFFFF58032378 // Bounds for root on [-12, -11]
data8 0xCFFFFFFFF6934AB8, 0xCFFFFFFFF313EF0A // Bounds for root on [-13, -12]
data8 0xDFFFFFFFFF53A9E9, 0xDFFFFFFFFF13B5A5 // Bounds for root on [-14, -13]
data8 0xEFFFFFFFFFF482CB, 0xEFFFFFFFFFF03F4F // Bounds for root on [-15, -14]
data8 0xFFFFFFFFFFFF482D, 0xFFFFFFFFFFFF03F5 // Bounds for root on [-16, -15]
data8 0x87FFFFFFFFFFFA98, 0x87FFFFFFFFFFF896 // Bounds for root on [-17, -16]
data8 0x8FFFFFFFFFFFFFB3, 0x8FFFFFFFFFFFFF97 // Bounds for root on [-18, -17]
data8 0x97FFFFFFFFFFFFFC, 0x97FFFFFFFFFFFFFB // Bounds for root on [-19, -18]
LOCAL_OBJECT_END(lgammal_right_roots_data)

LOCAL_OBJECT_START(lgammal_0_Half_data)
// Polynomial coefficients for the lgammal(x), 0.0 < |x| < 0.5
data8 0xBFD9A4D55BEAB2D6, 0xBC8AA3C097746D1F //A3
data8 0x3FEA51A6625307D3, 0x3C7180E7BD2D0DCC //A2
data8 0xBFE2788CFC6FB618, 0xBC9E9346C4692BCC //A1
data8 0x8A8991563EC1BD13, 0x00003FFD //A4
data8 0xD45CE0BD52C27EF2, 0x0000BFFC //A5
data8 0xADA06587FA2BBD47, 0x00003FFC //A6
data8 0x9381D0ED2194902A, 0x0000BFFC //A7
data8 0x80859B3CF92D4192, 0x00003FFC //A8
data8 0xE4033517C622A946, 0x0000BFFB //A9
data8 0xCD00CE67A51FC82A, 0x00003FFB //A10
data8 0xBA44E2A96C3B5700, 0x0000BFFB //A11
data8 0xAAAD008FA46DBD99, 0x00003FFB //A12
data8 0x9D604AC65A41153D, 0x0000BFFB //A13
data8 0x917CECB864B5A861, 0x00003FFB //A14
data8 0x85A4810EB730FDE4, 0x0000BFFB //A15
data8 0xEF2761C38BD21F77, 0x00003FFA //A16
data8 0xC913043A128367DA, 0x0000BFFA //A17
data8 0x96A29B71FF7AFFAA, 0x00003FFA //A18
data8 0xBB9FFA1A5FE649BB, 0x0000BFF9 //A19
data8 0xB17982CD2DAA0EE3, 0x00003FF8 //A20
data8 0xDE1DDCBFFB9453F0, 0x0000BFF6 //A21
data8 0x87FBF5D7ACD9FA9D, 0x00003FF4 //A22
LOCAL_OBJECT_END(lgammal_0_Half_data)

LOCAL_OBJECT_START(Constants_Q)
// log2_hi, log2_lo, Q_4, Q_3, Q_2, and Q_1
data4  0x00000000,0xB1721800,0x00003FFE,0x00000000
data4  0x4361C4C6,0x82E30865,0x0000BFE2,0x00000000
data4  0x328833CB,0xCCCCCAF2,0x00003FFC,0x00000000
data4  0xA9D4BAFB,0x80000077,0x0000BFFD,0x00000000
data4  0xAAABE3D2,0xAAAAAAAA,0x00003FFD,0x00000000
data4  0xFFFFDAB7,0xFFFFFFFF,0x0000BFFD,0x00000000
LOCAL_OBJECT_END(Constants_Q)

LOCAL_OBJECT_START(Constants_Z_1)
// Z1 - 16 bit fixed
data4  0x00008000
data4  0x00007879
data4  0x000071C8
data4  0x00006BCB
data4  0x00006667
data4  0x00006187
data4  0x00005D18
data4  0x0000590C
data4  0x00005556
data4  0x000051EC
data4  0x00004EC5
data4  0x00004BDB
data4  0x00004925
data4  0x0000469F
data4  0x00004445
data4  0x00004211
LOCAL_OBJECT_END(Constants_Z_1)

LOCAL_OBJECT_START(Constants_G_H_h1)
// G1 and H1 - IEEE single and h1 - IEEE double
data4  0x3F800000,0x00000000,0x00000000,0x00000000
data4  0x3F70F0F0,0x3D785196,0x617D741C,0x3DA163A6
data4  0x3F638E38,0x3DF13843,0xCBD3D5BB,0x3E2C55E6
data4  0x3F579430,0x3E2FF9A0,0xD86EA5E7,0xBE3EB0BF
data4  0x3F4CCCC8,0x3E647FD6,0x86B12760,0x3E2E6A8C
data4  0x3F430C30,0x3E8B3AE7,0x5C0739BA,0x3E47574C
data4  0x3F3A2E88,0x3EA30C68,0x13E8AF2F,0x3E20E30F
data4  0x3F321640,0x3EB9CEC8,0xF2C630BD,0xBE42885B
data4  0x3F2AAAA8,0x3ECF9927,0x97E577C6,0x3E497F34
data4  0x3F23D708,0x3EE47FC5,0xA6B0A5AB,0x3E3E6A6E
data4  0x3F1D89D8,0x3EF8947D,0xD328D9BE,0xBDF43E3C
data4  0x3F17B420,0x3F05F3A1,0x0ADB090A,0x3E4094C3
data4  0x3F124920,0x3F0F4303,0xFC1FE510,0xBE28FBB2
data4  0x3F0D3DC8,0x3F183EBF,0x10FDE3FA,0x3E3A7895
data4  0x3F088888,0x3F20EC80,0x7CC8C98F,0x3E508CE5
data4  0x3F042108,0x3F29516A,0xA223106C,0xBE534874
LOCAL_OBJECT_END(Constants_G_H_h1)

LOCAL_OBJECT_START(Constants_Z_2)
// Z2 - 16 bit fixed
data4  0x00008000
data4  0x00007F81
data4  0x00007F02
data4  0x00007E85
data4  0x00007E08
data4  0x00007D8D
data4  0x00007D12
data4  0x00007C98
data4  0x00007C20
data4  0x00007BA8
data4  0x00007B31
data4  0x00007ABB
data4  0x00007A45
data4  0x000079D1
data4  0x0000795D
data4  0x000078EB
LOCAL_OBJECT_END(Constants_Z_2)

LOCAL_OBJECT_START(Constants_G_H_h2)
// G2 and H2 - IEEE single and h2 - IEEE double
data4  0x3F800000,0x00000000,0x00000000,0x00000000
data4  0x3F7F00F8,0x3B7F875D,0x22C42273,0x3DB5A116
data4  0x3F7E03F8,0x3BFF015B,0x21F86ED3,0x3DE620CF
data4  0x3F7D08E0,0x3C3EE393,0x484F34ED,0xBDAFA07E
data4  0x3F7C0FC0,0x3C7E0586,0x3860BCF6,0xBDFE07F0
data4  0x3F7B1880,0x3C9E75D2,0xA78093D6,0x3DEA370F
data4  0x3F7A2328,0x3CBDC97A,0x72A753D0,0x3DFF5791
data4  0x3F792FB0,0x3CDCFE47,0xA7EF896B,0x3DFEBE6C
data4  0x3F783E08,0x3CFC15D0,0x409ECB43,0x3E0CF156
data4  0x3F774E38,0x3D0D874D,0xFFEF71DF,0xBE0B6F97
data4  0x3F766038,0x3D1CF49B,0x5D59EEE8,0xBE080483
data4  0x3F757400,0x3D2C531D,0xA9192A74,0x3E1F91E9
data4  0x3F748988,0x3D3BA322,0xBF72A8CD,0xBE139A06
data4  0x3F73A0D0,0x3D4AE46F,0xF8FBA6CF,0x3E1D9202
data4  0x3F72B9D0,0x3D5A1756,0xBA796223,0xBE1DCCC4
data4  0x3F71D488,0x3D693B9D,0xB6B7C239,0xBE049391
LOCAL_OBJECT_END(Constants_G_H_h2)

LOCAL_OBJECT_START(Constants_G_H_h3)
// G3 and H3 - IEEE single and h3 - IEEE double
data4  0x3F7FFC00,0x38800100,0x562224CD,0x3D355595
data4  0x3F7FF400,0x39400480,0x06136FF6,0x3D8200A2
data4  0x3F7FEC00,0x39A00640,0xE8DE9AF0,0x3DA4D68D
data4  0x3F7FE400,0x39E00C41,0xB10238DC,0xBD8B4291
data4  0x3F7FDC00,0x3A100A21,0x3B1952CA,0xBD89CCB8
data4  0x3F7FD400,0x3A300F22,0x1DC46826,0xBDB10707
data4  0x3F7FCC08,0x3A4FF51C,0xF43307DB,0x3DB6FCB9
data4  0x3F7FC408,0x3A6FFC1D,0x62DC7872,0xBD9B7C47
data4  0x3F7FBC10,0x3A87F20B,0x3F89154A,0xBDC3725E
data4  0x3F7FB410,0x3A97F68B,0x62B9D392,0xBD93519D
data4  0x3F7FAC18,0x3AA7EB86,0x0F21BD9D,0x3DC18441
data4  0x3F7FA420,0x3AB7E101,0x2245E0A6,0xBDA64B95
data4  0x3F7F9C20,0x3AC7E701,0xAABB34B8,0x3DB4B0EC
data4  0x3F7F9428,0x3AD7DD7B,0x6DC40A7E,0x3D992337
data4  0x3F7F8C30,0x3AE7D474,0x4F2083D3,0x3DC6E17B
data4  0x3F7F8438,0x3AF7CBED,0x811D4394,0x3DAE314B
data4  0x3F7F7C40,0x3B03E1F3,0xB08F2DB1,0xBDD46F21
data4  0x3F7F7448,0x3B0BDE2F,0x6D34522B,0xBDDC30A4
data4  0x3F7F6C50,0x3B13DAAA,0xB1F473DB,0x3DCB0070
data4  0x3F7F6458,0x3B1BD766,0x6AD282FD,0xBDD65DDC
data4  0x3F7F5C68,0x3B23CC5C,0xF153761A,0xBDCDAB83
data4  0x3F7F5470,0x3B2BC997,0x341D0F8F,0xBDDADA40
data4  0x3F7F4C78,0x3B33C711,0xEBC394E8,0x3DCD1BD7
data4  0x3F7F4488,0x3B3BBCC6,0x52E3E695,0xBDC3532B
data4  0x3F7F3C90,0x3B43BAC0,0xE846B3DE,0xBDA3961E
data4  0x3F7F34A0,0x3B4BB0F4,0x785778D4,0xBDDADF06
data4  0x3F7F2CA8,0x3B53AF6D,0xE55CE212,0x3DCC3ED1
data4  0x3F7F24B8,0x3B5BA620,0x9E382C15,0xBDBA3103
data4  0x3F7F1CC8,0x3B639D12,0x5C5AF197,0x3D635A0B
data4  0x3F7F14D8,0x3B6B9444,0x71D34EFC,0xBDDCCB19
data4  0x3F7F0CE0,0x3B7393BC,0x52CD7ADA,0x3DC74502
data4  0x3F7F04F0,0x3B7B8B6D,0x7D7F2A42,0xBDB68F17
LOCAL_OBJECT_END(Constants_G_H_h3)

LOCAL_OBJECT_START(lgammal_data)
// Positive overflow value
data8 0xB8D54C8BFFFDEBF4, 0x00007FF1
LOCAL_OBJECT_END(lgammal_data)

LOCAL_OBJECT_START(lgammal_Stirling)
// Coefficients needed for Strirling's formula
data8 0x3FED67F1C864BEB4 // High part of 0.5*ln(2*Pi)
data8 0x3C94D252F2400510 // Low part of 0.5*ln(2*Pi)
//
// Bernulli numbers used in Striling's formula for -2^63 < |x| < -13.0
//(B1H, B1L) = 8.3333333333333333333262747254e-02
data8 0x3FB5555555555555, 0x3C55555555555555
data8 0xB60B60B60B60B60B, 0x0000BFF6 //B2 = -2.7777777777777777777777777778e-03
data8 0xD00D00D00D00D00D, 0x00003FF4 //B3 = 7.9365079365079365079365079365e-04
data8 0x9C09C09C09C09C0A, 0x0000BFF4 //B4 = -5.9523809523809523809523809524e-04
data8 0xDCA8F158C7F91AB8, 0x00003FF4 //B5 = 8.4175084175084175084175084175e-04
data8 0xFB5586CCC9E3E410, 0x0000BFF5 //B6 = -1.9175269175269175269175269175e-03
data8 0xD20D20D20D20D20D, 0x00003FF7 //B7 = 6.4102564102564102564102564103e-03
data8 0xF21436587A9CBEE1, 0x0000BFF9 //B8 = -2.9550653594771241830065359477e-02
data8 0xB7F4B1C0F033FFD1, 0x00003FFC //B9 = 1.7964437236883057316493849002e-01
data8 0xB23B3808C0F9CF6E, 0x0000BFFF //B10 = -1.3924322169059011164274322169e+00
// Polynomial coefficients for Stirling's formula, -13.0 < x < -6.0
data8 0x3FB5555555555555, 0x3C4D75060289C58B //A0
data8 0xB60B60B60B0F0876, 0x0000BFF6 //A1
data8 0xD00D00CE54B1256C, 0x00003FF4 //A2
data8 0x9C09BF46B58F75E1, 0x0000BFF4 //A3
data8 0xDCA8483BC91ACC6D, 0x00003FF4 //A4
data8 0xFB3965C939CC9FEE, 0x0000BFF5 //A5
data8 0xD0723ADE3F0BC401, 0x00003FF7 //A6
data8 0xE1ED7434E81F0B73, 0x0000BFF9 //A7
data8 0x8069C6982F993283, 0x00003FFC //A8
data8 0xC271F65BFA5BEE3F, 0x0000BFFD //A9
LOCAL_OBJECT_END(lgammal_Stirling)

LOCAL_OBJECT_START(lgammal_lnsin_data)
// polynomial approximation of -ln(sin(Pi*x)/(Pi*x)), 0 < x <= 0.5
data8 0x3FFA51A6625307D3, 0x3C81873332FAF94C //A2
data8 0x8A8991563EC241C3, 0x00003FFE //A4
data8 0xADA06588061805DF, 0x00003FFD //A6
data8 0x80859B57C338D0F7, 0x00003FFD //A8
data8 0xCD00F1C2D78754BD, 0x00003FFC //A10
data8 0xAAB56B1D3A1F4655, 0x00003FFC //A12
data8 0x924B6F2FBBED12B1, 0x00003FFC //A14
data8 0x80008E58765F43FC, 0x00003FFC //A16
data8 0x3FBC718EC115E429//A18
data8 0x3FB99CE544FE183E//A20
data8 0x3FB7251C09EAAD89//A22
data8 0x3FB64A970733628C//A24
data8 0x3FAC92D6802A3498//A26
data8 0x3FC47E1165261586//A28
data8 0xBFCA1BAA434750D4//A30
data8 0x3FE460001C4D5961//A32
data8 0xBFE6F06A3E4908AD//A34
data8 0x3FE300889EBB203A//A36
LOCAL_OBJECT_END(lgammal_lnsin_data)

LOCAL_OBJECT_START(lgammal_half_3Q_data)
// Polynomial coefficients for the lgammal(x), 0.5 <= x < 0.75
data8 0xBFF7A648EE90C62E, 0x3C713F326857E066 // A3, A0L
data8 0xBFF73E4B8BA780AE, 0xBCA953BC788877EF // A1, A1L
data8 0x403774DCD58D0291, 0xC0415254D5AE6623 // D0, D1
data8 0x40B07213855CBFB0, 0xC0B8855E25D2D229 // C20, C21
data8 0x3FFB359F85FF5000, 0x3C9BAECE6EF9EF3A // A2, A2L
data8 0x3FD717D498A3A8CC, 0xBC9088E101CFEDFA  // A0, A3L
data8 0xAFEF36CC5AEC3FF0, 0x00004002 // E6
data8 0xABE2054E1C34E791, 0x00004001 // E4
data8 0xB39343637B2900D1, 0x00004000 // E2
data8 0xD74FB710D53F58F6, 0x00003FFF // E0
data8 0x4070655963BA4256, 0xC078DA9D263C4EA3 // D6, D7
data8 0x405CD2B6A9B90978, 0xC065B3B9F4F4F171 // D4, D5
data8 0x4049BC2204CF61FF, 0xC05337227E0BA152 // D2, D3
data8 0x4095509A50C07A96, 0xC0A0747949D2FB45 // C18, C19
data8 0x4082ECCBAD709414, 0xC08CD02FB088A702 // C16, C17
data8 0xFFE4B2A61B508DD5, 0x0000C002 // E7
data8 0xF461ADB8AE17E0A5, 0x0000C001 // E5
data8 0xF5BE8B0B90325F20, 0x0000C000 // E3
data8 0x877B275F3FB78DCA, 0x0000C000 // E1
LOCAL_OBJECT_END(lgammal_half_3Q_data)

LOCAL_OBJECT_START(lgammal_half_3Q_neg_data)
// Polynomial coefficients for the lgammal(x), -0.75 < x <= -0.5
data8 0xC014836EFD94899C, 0x3C9835679663B44F // A3, A0L
data8 0xBFF276C7B4FB1875, 0xBC92D3D9FA29A1C0 // A1, A1L
data8 0x40C5178F24E1A435, 0xC0D9DE84FBC5D76A // D0, D1
data8 0x41D4D1B236BF6E93, 0xC1EBB0445CE58550 // C20, C21
data8 0x4015718CD67F63D3, 0x3CC5354B6F04B59C // A2, A2L
data8 0x3FF554493087E1ED, 0xBCB72715E37B02B9 // A0, A3L
data8 0xE4AC7E915FA72229, 0x00004009 // E6
data8 0xA28244206395FCC6, 0x00004007 // E4
data8 0xFB045F19C07B2544, 0x00004004 // E2
data8 0xE5C8A6E6A9BA7D7B, 0x00004002 // E0
data8 0x4143943B55BF5118, 0xC158AC05EA675406 // D6, D7
data8 0x4118F6833D19717C, 0xC12F51A6F375CC80 // D4, D5
data8 0x40F00C209483481C, 0xC103F1DABF750259 // D2, D3
data8 0x4191038F2D8F9E40, 0xC1A413066DA8AE4A // C18, C19
data8 0x4170B537EDD833DE, 0xC1857E79424C61CE // C16, C17
data8 0x8941D8AB4855DB73, 0x0000C00B // E7
data8 0xBB822B131BD2E813, 0x0000C008 // E5
data8 0x852B4C03B83D2D4F, 0x0000C006 // E3
data8 0xC754CA7E2DDC0F1F, 0x0000C003 // E1
LOCAL_OBJECT_END(lgammal_half_3Q_neg_data)

LOCAL_OBJECT_START(lgammal_2Q_4_data)
// Polynomial coefficients for the lgammal(x), 2.25 <= |x| < 4.0
data8 0xBFCA4D55BEAB2D6F, 0x3C7ABC9DA14141F5 // A3, A0L
data8 0x3FFD8773039049E7, 0x3C66CB7957A95BA4 // A1, A1L
data8 0x3F45C3CC79E91E7D, 0xBF3A8E5005937E97 // D0, D1
data8 0x3EC951E35E1C9203, 0xBEB030A90026C5DF // C20, C21
data8 0x3FE94699894C1F4C, 0x3C91884D21D123F1 // A2, A2L
data8 0x3FE62E42FEFA39EF, 0xBC66480CEB70870F // A0, A3L
data8 0xF1C2EAFF0B3A7579, 0x00003FF5 // E6
data8 0xB36AF863926B55A3, 0x00003FF7 // E4
data8 0x9620656185BB44CA, 0x00003FF9 // E2
data8 0xA264558FB0906AFF, 0x00003FFB // E0
data8 0x3F03D59E9666C961, 0xBEF91115893D84A6 // D6, D7
data8 0x3F19333611C46225, 0xBF0F89EB7D029870 // D4, D5
data8 0x3F3055A96B347AFE, 0xBF243B5153E178A8 // D2, D3
data8 0x3ED9A4AEF30C4BB2, 0xBED388138B1CEFF2 // C18, C19
data8 0x3EEF7945A3C3A254, 0xBEE36F32A938EF11 // C16, C17
data8 0x9028923F47C82118, 0x0000BFF5 // E7
data8 0xCE0DAAFB6DC93B22, 0x0000BFF6 // E5
data8 0xA0D0983B34AC4C8D, 0x0000BFF8 // E3
data8 0x94D6C50FEB8B0CE7, 0x0000BFFA // E1
LOCAL_OBJECT_END(lgammal_2Q_4_data)

LOCAL_OBJECT_START(lgammal_4_8_data)
// Polynomial coefficients for the lgammal(x), 4.0 <= |x| < 8.0
data8 0xBFD6626BC9B31B54, 0x3CAA53C82493A92B // A3, A0L
data8 0x401B4C420A50AD7C, 0x3C8C6E9929F789A3 // A1, A1L
data8 0x3F49410427E928C2, 0xBF3E312678F8C146 // D0, D1
data8 0x3ED51065F7CD5848, 0xBED052782A03312F // C20, C21
data8 0x3FF735973273D5EC, 0x3C831DFC65BF8CCF // A2, A2L
data8 0x401326643C4479C9, 0xBC6FA0498C5548A6 // A0, A3L
data8 0x9382D8B3CD4EB7E3, 0x00003FF6 // E6
data8 0xE9F92CAD8A85CBCD, 0x00003FF7 // E4
data8 0xD58389FE38258CEC, 0x00003FF9 // E2
data8 0x81310136363AE8AA, 0x00003FFC // E0
data8 0x3F04F0AE38E78570, 0xBEF9E2144BB8F03C // D6, D7
data8 0x3F1B5E992A6CBC2A, 0xBF10F3F400113911 // D4, D5
data8 0x3F323EE00AAB7DEE, 0xBF2640FDFA9FB637 // D2, D3
data8 0x3ED2143EBAFF067A, 0xBEBBDEB92D6FF35D // C18, C19
data8 0x3EF173A42B69AAA4, 0xBEE78B9951A2EAA5 // C16, C17
data8 0xAB3CCAC6344E52AA, 0x0000BFF5 // E7
data8 0x81ACCB8915B16508, 0x0000BFF7 // E5
data8 0xDA62C7221102C426, 0x0000BFF8 // E3
data8 0xDF1BD44C4083580A, 0x0000BFFA // E1
LOCAL_OBJECT_END(lgammal_4_8_data)

LOCAL_OBJECT_START(lgammal_loc_min_data)
// Polynomial coefficients for the lgammal(x), 1.3125 <= x < 1.5625
data8 0xBB16C31AB5F1FB71, 0x00003FFF // xMin - point of local minimum
data8 0xBFC2E4278DC6BC23, 0xBC683DA8DDCA9650 // A3, A0L
data8 0x3BD4DB7D0CA61D5F, 0x386E719EDD01D801 // A1, A1L
data8 0x3F4CC72638E1D93F, 0xBF4228EC9953CCB9 // D0, D1
data8 0x3ED222F97A04613E,0xBED3DDD58095CB6C  // C20, C21
data8 0x3FDEF72BC8EE38AB, 0x3C863AFF3FC48940 // A2, A2L
data8 0xBFBF19B9BCC38A41,  0xBC7425F1BFFC1442// A0, A3L
data8 0x941890032BEB34C3, 0x00003FF6 // E6
data8 0xC7E701591CE534BC, 0x00003FF7 // E4
data8 0x93373CBD05138DD4, 0x00003FF9 // E2
data8 0x845A14A6A81C05D6, 0x00003FFB // E0
data8 0x3F0F6C4DF6D47A13, 0xBF045DCDB5B49E19 // D6, D7
data8 0x3F22E23345DDE59C, 0xBF1851159AFB1735 // D4, D5
data8 0x3F37101EA4022B78, 0xBF2D721E6323AF13 // D2, D3
data8 0x3EE691EBE82DF09D, 0xBEDD42550961F730 // C18, C19
data8 0x3EFA793EDE99AD85, 0xBEF14000108E70BE // C16, C17
data8 0xB7CBC033ACE0C99C, 0x0000BFF5 // E7
data8 0xF178D1F7B1A45E27, 0x0000BFF6 // E5
data8 0xA8FCFCA8106F471C, 0x0000BFF8 // E3
data8 0x864D46FA898A9AD2, 0x0000BFFA // E1
LOCAL_OBJECT_END(lgammal_loc_min_data)

LOCAL_OBJECT_START(lgammal_03Q_1Q_data)
// Polynomial coefficients for the lgammal(x), 0.75 <= |x| < 1.3125
data8 0x3FD151322AC7D848, 0x3C7184DE0DB7B4EE // A4, A2L
data8 0x3FD9A4D55BEAB2D6, 0x3C9E934AAB10845F // A3, A1L
data8 0x3FB111289C381259, 0x3FAFFFCFB32AE18D // D2, D3
data8 0x3FB3B1D9E0E3E00D, 0x3FB2496F0D3768DF // D0, D1
data8 0xBA461972C057D439, 0x00003FFB         // E6
data8 0x3FEA51A6625307D3, 0x3C76ABC886A72DA2 // A2, A4L
data8 0x3FA8EFE46B32A70E, 0x3F8F31B3559576B6 // C17, C20
data8 0xE403383700387D85, 0x00003FFB // E4
data8 0x9381D0EE74BF7251, 0x00003FFC // E2
data8 0x3FAA2177A6D28177, 0x3FA4895E65FBD995 // C18, C19
data8 0x3FAAED2C77DBEE5D, 0x3FA94CA59385512C // D6, D7
data8 0x3FAE1F522E8A5941, 0x3FAC785EF56DD87E // D4, D5
data8 0x3FB556AD5FA56F0A, 0x3FA81F416E87C783 // E7, C16
data8 0xCD00F1C2DC2C9F1E, 0x00003FFB // E5
data8 0x3FE2788CFC6FB618, 0x3C8E52519B5B17CB // A1, A3L
data8 0x80859B57C3E7F241, 0x00003FFC // E3
data8 0xADA065880615F401, 0x00003FFC // E1
data8 0xD45CE0BD530AB50E, 0x00003FFC // E0
LOCAL_OBJECT_END(lgammal_03Q_1Q_data)

LOCAL_OBJECT_START(lgammal_13Q_2Q_data)
// Polynomial coefficients for the lgammal(x), 1.5625 <= |x| < 2.25
data8 0x3F951322AC7D8483, 0x3C71873D88C6539D // A4, A2L
data8 0xBFB13E001A557606, 0x3C56CB907018A101 // A3, A1L
data8 0xBEC11B2EC1E7F6FC, 0x3EB0064ED9824CC7 // D2, D3
data8 0xBEE3CBC963EC103A, 0x3ED2597A330C107D // D0, D1
data8 0xBC6F2DEBDFE66F38, 0x0000BFF0 // E6
data8 0x3FD4A34CC4A60FA6, 0x3C3AFC9BF775E8A0 // A2, A4L
data8 0x3E48B0C542F85B32, 0xBE347F12EAF787AB // C17, C20
data8 0xE9FEA63B6984FA1E, 0x0000BFF2 // E4
data8 0x9C562E15FC703BBF, 0x0000BFF5 // E2
data8 0xBE3C12A50AB0355E, 0xBE1C941626AE4717 // C18, C19
data8 0xBE7AFA8714342BC4,0x3E69A12D2B7761CB // D6, D7
data8 0xBE9E25EF1D526730, 0x3E8C762291889B99 // D4, D5
data8 0x3EF580DCEE754733, 0xBE57C811D070549C // E7, C16
data8 0xD093D878BE209C98, 0x00003FF1 // E5
data8 0x3FDB0EE6072093CE, 0xBC6024B9E81281C4 // A1, A3L
data8 0x859B57C31CB77D96, 0x00003FF4 // E3
data8 0xBD6EB756DB617E8D, 0x00003FF6 // E1
data8 0xF2027E10C7AF8C38, 0x0000BFF7 // E0
LOCAL_OBJECT_END(lgammal_13Q_2Q_data)

LOCAL_OBJECT_START(lgammal_8_10_data)
// Polynomial coefficients for the lgammal(x), 8.0 <= |x| < 10.0
// Multi Precision terms
data8 0x40312008A3A23E5C, 0x3CE020B4F2E4083A //A1
data8 0x4025358E82FCB70C, 0x3CD4A5A74AF7B99C //A0
// Native precision terms
data8 0xF0AA239FFBC616D2, 0x00004000 //A2
data8 0x96A8EA798FE57D66, 0x0000BFFF //A3
data8 0x8D501B7E3B9B9BDB, 0x00003FFE //A4
data8 0x9EE062401F4B1DC2, 0x0000BFFD //A5
data8 0xC63FD8CD31E93431, 0x00003FFC //A6
data8 0x8461101709C23C30, 0x0000BFFC //A7
data8 0xB96D7EA7EF3648B2, 0x00003FFB //A8
data8 0x86886759D2ACC906, 0x0000BFFB //A9
data8 0xC894B6E28265B183, 0x00003FFA //A10
data8 0x98C4348CAD821662, 0x0000BFFA //A11
data8 0xEC9B092226A94DF2, 0x00003FF9 //A12
data8 0xB9F169FF9B98CDDC, 0x0000BFF9 //A13
data8 0x9A3A32BB040894D3, 0x00003FF9 //A14
data8 0xF9504CCC1003B3C3, 0x0000BFF8 //A15
LOCAL_OBJECT_END(lgammal_8_10_data)

LOCAL_OBJECT_START(lgammal_03Q_6_data)
// Polynomial coefficients for the lgammal(x), 0.75 <= |x| < 1.0
data8 0xBFBC47DCA479E295, 0xBC607E6C1A379D55 //A3
data8 0x3FCA051C372609ED, 0x3C7B02D73EB7D831 //A0
data8 0xBFE15FAFA86B04DB, 0xBC3F52EE4A8945B5 //A1
data8 0x3FD455C4FF28F0BF, 0x3C75F8C6C99F30BB //A2
data8 0xD2CF04CD934F03E1, 0x00003FFA //A4
data8 0xDB4ED667E29256E1, 0x0000BFF9 //A5
data8 0xF155A33A5B6021BF, 0x00003FF8 //A6
data8 0x895E9B9D386E0338, 0x0000BFF8 //A7
data8 0xA001BE94B937112E, 0x00003FF7 //A8
data8 0xBD82846E490ED048, 0x0000BFF6 //A9
data8 0xE358D24EC30DBB5D, 0x00003FF5 //A10
data8 0x89C4F3652446B78B, 0x0000BFF5 //A11
data8 0xA86043E10280193D, 0x00003FF4 //A12
data8 0xCF3A2FBA61EB7682, 0x0000BFF3 //A13
data8 0x3F300900CC9200EC //A14
data8 0xBF23F42264B94AE8 //A15
data8 0x3F18EEF29895FE73 //A16
data8 0xBF0F3C4563E3EDFB //A17
data8 0x3F0387DBBC385056 //A18
data8 0xBEF81B4004F92900 //A19
data8 0x3EECA6692A9A5B81 //A20
data8 0xBEDF61A0059C15D3 //A21
data8 0x3ECDA9F40DCA0111 //A22
data8 0xBEB60FE788217BAF //A23
data8 0x3E9661D795DFC8C6 //A24
data8 0xBE66C7756A4EDEE5 //A25
// Polynomial coefficients for the lgammal(x), 1.0 <= |x| < 2.0
data8 0xBFC1AE55B180726B, 0xBC7DE1BC478453F5 //A3
data8 0xBFBEEB95B094C191, 0xBC53456FF6F1C9D9 //A0
data8 0x3FA2AED059BD608A, 0x3C0B65CC647D557F //A1
data8 0x3FDDE9E64DF22EF2, 0x3C8993939A8BA8E4 //A2
data8 0xF07C206D6B100CFF, 0x00003FFA //A4
data8 0xED2CEA9BA52FE7FB, 0x0000BFF9 //A5
data8 0xFCE51CED52DF3602, 0x00003FF8 //A6
data8 0x8D45D27872326619, 0x0000BFF8 //A7
data8 0xA2B78D6BCEBE27F7, 0x00003FF7 //A8
data8 0xBF6DC0996A895B6F, 0x0000BFF6 //A9
data8 0xE4B9AD335AF82D79, 0x00003FF5 //A10
data8 0x8A451880195362A1, 0x0000BFF5 //A11
data8 0xA8BE35E63089A7A9, 0x00003FF4 //A12
data8 0xCF7FA175FA11C40C, 0x0000BFF3 //A13
data8 0x3F300C282FAA3B02 //A14
data8 0xBF23F6AEBDA68B80 //A15
data8 0x3F18F6860E2224DD //A16
data8 0xBF0F542B3CE32F28 //A17
data8 0x3F039436218C9BF8 //A18
data8 0xBEF8AE6307677AEC //A19
data8 0x3EF0B55527B3A211 //A20
data8 0xBEE576AC995E7605 //A21
data8 0x3ED102DDC1365D2D //A22
data8 0xBEC442184F97EA54 //A23
data8 0x3ED4D2283DFE5FC6 //A24
data8 0xBECB9219A9B46787 //A25
// Polynomial coefficients for the lgammal(x), 2.0 <= |x| < 3.0
data8 0xBFCA4D55BEAB2D6F, 0xBC66F80E5BFD5AF5 //A3
data8 0x3FE62E42FEFA39EF, 0x3C7ABC9E3B347E3D //A0
data8 0x3FFD8773039049E7, 0x3C66CB9007C426EA //A1
data8 0x3FE94699894C1F4C, 0x3C918726EB111663 //A2
data8 0xA264558FB0906209, 0x00003FFB //A4
data8 0x94D6C50FEB902ADC, 0x0000BFFA //A5
data8 0x9620656184243D17, 0x00003FF9 //A6
data8 0xA0D0983B8BCA910B, 0x0000BFF8 //A7
data8 0xB36AF8559B222BD3, 0x00003FF7 //A8
data8 0xCE0DACB3260AE6E5, 0x0000BFF6 //A9
data8 0xF1C2C0BF0437C7DB, 0x00003FF5 //A10
data8 0x902A2F2F3AB74A92, 0x0000BFF5 //A11
data8 0xAE05009B1B2C6E4C, 0x00003FF4 //A12
data8 0xD5B71F6456D7D4CB, 0x0000BFF3 //A13
data8 0x3F2F0351D71BC9C6 //A14
data8 0xBF2B53BC56A3B793 //A15
data8 0xBF18B12DC6F6B861 //A16
data8 0xBF43EE6EB5215C2F //A17
data8 0xBF5474787CDD455E //A18
data8 0xBF642B503C9C060A //A19
data8 0xBF6E07D1AA254AA3 //A20
data8 0xBF71C785443AAEE8 //A21
data8 0xBF6F67BF81B71052 //A22
data8 0xBF63E4BCCF4FFABF //A23
data8 0xBF50067F8C671D5A //A24
data8 0xBF29C770D680A5AC //A25
// Polynomial coefficients for the lgammal(x), 4.0 <= |x| < 6.0
data8 0xBFD6626BC9B31B54, 0xBC85AABE08680902 //A3
data8 0x401326643C4479C9, 0x3CAA53C26F31E364 //A0
data8 0x401B4C420A50AD7C, 0x3C8C76D55E57DD8D //A1
data8 0x3FF735973273D5EC, 0x3C83A0B78E09188A //A2
data8 0x81310136363AAB6D, 0x00003FFC //A4
data8 0xDF1BD44C4075C0E6, 0x0000BFFA //A5
data8 0xD58389FE38D8D664, 0x00003FF9 //A6
data8 0xDA62C7221D5B5F87, 0x0000BFF8 //A7
data8 0xE9F92CAD0263E157, 0x00003FF7 //A8
data8 0x81ACCB8606C165FE, 0x0000BFF7 //A9
data8 0x9382D8D263D1C2A3, 0x00003FF6 //A10
data8 0xAB3CCBA4C853B12C, 0x0000BFF5 //A11
data8 0xCA0818BBCCC59296, 0x00003FF4 //A12
data8 0xF18912691CBB5BD0, 0x0000BFF3 //A13
data8 0x3F323EF5D8330339 //A14
data8 0xBF2641132EA571F7 //A15
data8 0x3F1B5D9576175CA9 //A16
data8 0xBF10F56A689C623D //A17
data8 0x3F04CACA9141A18D //A18
data8 0xBEFA307AC9B4E85D //A19
data8 0x3EF4B625939FBE32 //A20
data8 0xBECEE6AC1420F86F //A21
data8 0xBE9A95AE2E485964 //A22
data8 0xBF039EF47F8C09BB //A23
data8 0xBF05345957F7B7A9 //A24
data8 0xBEF85AE6385D4CCC //A25
// Polynomial coefficients for the lgammal(x), 3.0 <= |x| < 4.0
data8 0xBFCA4D55BEAB2D6F, 0xBC667B20FF46C6A8 //A3
data8 0x3FE62E42FEFA39EF, 0x3C7ABC9E3B398012 //A0
data8 0x3FFD8773039049E7, 0x3C66CB9070238D77 //A1
data8 0x3FE94699894C1F4C, 0x3C91873D8839B1CD //A2
data8 0xA264558FB0906D7E, 0x00003FFB //A4
data8 0x94D6C50FEB8AFD72, 0x0000BFFA //A5
data8 0x9620656185B68F14, 0x00003FF9 //A6
data8 0xA0D0983B34B7088A, 0x0000BFF8 //A7
data8 0xB36AF863964AA440, 0x00003FF7 //A8
data8 0xCE0DAAFB5497AFB8, 0x0000BFF6 //A9
data8 0xF1C2EAFA79CC2864, 0x00003FF5 //A10
data8 0x9028922A839572B8, 0x0000BFF5 //A11
data8 0xAE1E62F870BA0278, 0x00003FF4 //A12
data8 0xD4726F681E2ABA29, 0x0000BFF3 //A13
data8 0x3F30559B9A02FADF //A14
data8 0xBF243ADEB1266CAE //A15
data8 0x3F19303B6F552603 //A16
data8 0xBF0F768C288EC643 //A17
data8 0x3F039D5356C21DE1 //A18
data8 0xBEF81BCA8168E6BE //A19
data8 0x3EEC74A53A06AD54 //A20
data8 0xBEDED52D1A5DACDF //A21
data8 0x3ECCB4C2C7087342 //A22
data8 0xBEB4F1FAFDFF5C2F //A23
data8 0x3E94C80B52D58904 //A24
data8 0xBE64A328CBE92A27 //A25
LOCAL_OBJECT_END(lgammal_03Q_6_data)

LOCAL_OBJECT_START(lgammal_1pEps_data)
// Polynomial coefficients for the lgammal(x), 1 - 2^(-7) <= |x| < 1 + 2^(-7)
data8 0x93C467E37DB0C7A5, 0x00003FFE //A1
data8 0xD28D3312983E9919, 0x00003FFE //A2
data8 0xCD26AADF559A47E3, 0x00003FFD //A3
data8 0x8A8991563EC22E81, 0x00003FFD //A4
data8 0x3FCA8B9C168D52FE //A5
data8 0x3FC5B40CB0696370 //A6
data8 0x3FC270AC2229A65D //A7
data8 0x3FC0110AF10FCBFC //A8
// Polynomial coefficients for the log1p(x), - 2^(-7) <= |x| <  2^(-7)
data8 0x3FBC71C71C71C71C //P8
data8 0xBFC0000000000000 //P7
data8 0x3FC2492492492492 //P6
data8 0xBFC5555555555555 //P5
data8 0x3FC999999999999A //P4
data8 0xBFD0000000000000 //P3
data8 0x3FD5555555555555 //P2
data8 0xBFE0000000000000 //P1
// short version of "lnsin" polynomial
data8 0xD28D3312983E9918, 0x00003FFF //A2
data8 0x8A8991563EC241B6, 0x00003FFE //A4
data8 0xADA06588061830A5, 0x00003FFD //A6
data8 0x80859B57C31CB746, 0x00003FFD //A8
LOCAL_OBJECT_END(lgammal_1pEps_data)

LOCAL_OBJECT_START(lgammal_neg2andHalf_data)
// Polynomial coefficients for the lgammal(x), -2.005859375 <= x < -2.5
data8 0xBF927781D4BB093A, 0xBC511D86D85B7045 // A3, A0L
data8 0x3FF1A68793DEFC15, 0x3C9852AE2DA7DEEF // A1, A1L
data8 0x408555562D45FAFD, 0xBF972CDAFE5FEFAD // D0, D1
data8 0xC18682331EF492A5, 0xC1845E3E0D29606B // C20, C21
data8 0x4013141822E16979, 0x3CCF8718B6E75F6C // A2, A2L
data8 0xBFACCBF9F5ED0F15, 0xBBDD1AEB73297401 // A0, A3L
data8 0xCCCDB17423046445, 0x00004006 // E6
data8 0x800514E230A3A452, 0x00004005 // E4
data8 0xAAE9A48EC162E76F, 0x00004003 // E2
data8 0x81D4F88B3F3EA0FC, 0x00004002 // E0
data8 0x40CF3F3E35238DA0, 0xC0F8B340945F1A7E // D6, D7
data8 0x40BF89EC0BD609C6, 0xC095897242AEFEE2 // D4, D5
data8 0x40A2482FF01DBC5C, 0xC02095E275FDCF62 // D2, D3
data8 0xC1641354F2312A6A, 0xC17B3657F85258E9 // C18, C19
data8 0xC11F964E9ECBE2C9, 0xC146D7A90F70696C // C16, C17
data8 0xE7AECDE6AF8EA816, 0x0000BFEF // E7
data8 0xD711252FEBBE1091, 0x0000BFEB // E5
data8 0xE648BD10F8C43391, 0x0000BFEF // E3
data8 0x948A1E78AA00A98D, 0x0000BFF4 // E1
LOCAL_OBJECT_END(lgammal_neg2andHalf_data)

LOCAL_OBJECT_START(lgammal_near_neg_half_data)
// Polynomial coefficients for the lgammal(x), -0.5 < x < -0.40625
data8 0xBFC1AE55B180726C, 0x3C8053CD734E6A1D // A3, A0L
data8 0x3FA2AED059BD608A, 0x3C0CD3D2CDBA17F4 // A1, A1L
data8 0x40855554DBCD1E1E, 0x3F96C51AC2BEE9E1 // D0, D1
data8 0xC18682331EF4927D, 0x41845E3E0D295DFC // C20, C21
data8 0x4011DE9E64DF22EF, 0x3CA692B70DAD6B7B // A2, A2L
data8 0x3FF43F89A3F0EDD6, 0xBC4955AED0FA087D // A0, A3L
data8 0xCCCD3F1DF4A2C1DD, 0x00004006 // E6
data8 0x80028ADE33C7FCD9, 0x00004005 // E4
data8 0xAACA474E485507EF, 0x00004003 // E2
data8 0x80F07C206D6B0ECD, 0x00004002 // E0
data8 0x40CF3F3E33E83056, 0x40F8B340944633D9 // D6, D7
data8 0x40BF89EC059931F0, 0x409589723307AD20 // D4, D5
data8 0x40A2482FD0054824, 0x402095CE7F19D011 // D2, D3
data8 0xC1641354F2313614, 0x417B3657F8525354 // C18, C19
data8 0xC11F964E9ECFD21C, 0x4146D7A90F701836 // C16, C17
data8 0x86A9C01F0EA11E5A, 0x0000BFF5 // E7
data8 0xBF6D8469142881C0, 0x0000BFF6 // E5
data8 0x8D45D277BA8255F1, 0x0000BFF8 // E3
data8 0xED2CEA9BA528BCC3, 0x0000BFF9 // E1
LOCAL_OBJECT_END(lgammal_near_neg_half_data)

//!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
////////////// POLYNOMIAL COEFFICIENTS FOR "NEAR ROOTS" RANGES    /////////////
////////////// THIS PART OF TABLE SHOULD BE ADDRESSED REALLY RARE /////////////
//!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
LOCAL_OBJECT_START(lgammal_right_roots_polynomial_data)
// Polynomial coefficients for right root on [-3, -2]
// Lgammal is approximated by polynomial within [-.056244 ; .158208 ] range
data8 0xBBBD5E9DCD11030B, 0xB867411D9FF87DD4 //A0
data8 0x3FF83FE966AF535E, 0x3CAA21235B8A769A //A1
data8 0x40136EEBB002F55C, 0x3CC3959A6029838E //A2
data8 0xB4A5302C53C2BEDD, 0x00003FFF //A3
data8 0x8B8C6BE504F2DA1C, 0x00004002 //A4
data8 0xB99CFF02593B4D98, 0x00004001 //A5
data8 0x4038D32F682AA1CF //A6
data8 0x403809F04EE6C5B5 //A7
data8 0x40548EAA81634CEE //A8
data8 0x4059297ADB6BC03D //A9
data8 0x407286FB8EC5C9DA //A10
data8 0x407A92E05B744CFB //A11
data8 0x4091A9D4144258CD //A12
data8 0x409C4D01D24F367E //A13
data8 0x40B1871B9A426A83 //A14
data8 0x40BE51C48BD9A583 //A15
data8 0x40D2140D0C6153E7 //A16
data8 0x40E0FB2C989CE4A3 //A17
data8 0x40E52739AB005641 //A18
data8 0x41161E3E6DDF503A //A19
// Polynomial coefficients for right root on [-4, -3]
// Lgammal is approximated by polynomial within [-.172797 ; .171573 ] range
data8 0x3C172712B248E42E, 0x38CB8D17801A5D67 //A0
data8 0x401F20A65F2FAC54, 0x3CCB9EA1817A824E //A1
data8 0x4039D4D2977150EF, 0x3CDA42E149B6276A //A2
data8 0xE089B8926AE2D9CB, 0x00004005 //A3
data8 0x933901EBBB586C37, 0x00004008 //A4
data8 0xCCD319BED1CFA1CD, 0x0000400A //A5
data8 0x40D293C3F78D3C37 //A6
data8 0x40FBB97AA0B6DD02 //A7
data8 0x41251EA3345E5EB9 //A8
data8 0x415057F65C92E7B0 //A9
data8 0x41799C865241B505 //A10
data8 0x41A445209EFE896B //A11
data8 0x41D02D21880C953B //A12
data8 0x41F9FFDE8C63E16D //A13
data8 0x422504DC8302D2BE //A14
data8 0x425111BF18C95414 //A15
data8 0x427BCBE74A2B8EF7 //A16
data8 0x42A7256F59B286F7 //A17
data8 0x42D462D1586DE61F //A18
data8 0x42FBB1228D6C5118 //A19
// Polynomial coefficients for right root on [-5, -4]
// Lgammal is approximated by polynomial within [-.163171 ; .161988 ] range
data8 0x3C5840FBAFDEE5BB, 0x38CAC0336E8C490A //A0
data8 0x403ACA5CF4921642, 0x3CCEDCDDA5491E56 //A1
data8 0x40744415CD813F8E, 0x3CFBFEBC17E39146 //A2
data8 0xAACD88D954E3E1BD, 0x0000400B //A3
data8 0xCB68C710D75ED802, 0x0000400F //A4
data8 0x8130F5AB997277AC, 0x00004014 //A5
data8 0x41855E3DBF99EBA7 //A6
data8 0x41CD14FE49C49FC2 //A7
data8 0x421433DCE281F07D //A8
data8 0x425C8399C7A92B6F //A9
data8 0x42A45FBE67840F1A //A10
data8 0x42ED68D75F9E6C98 //A11
data8 0x433567291C27E5BE //A12
data8 0x437F5ED7A9D9FD28 //A13
data8 0x43C720A65C8AB711 //A14
data8 0x441120A6C1D40B9B //A15
data8 0x44596F561F2D1CBE //A16
data8 0x44A3507DA81D5C01 //A17
data8 0x44EF06A31E39EEDF //A18
data8 0x45333774C99F523F //A19
// Polynomial coefficients for right root on [-6, -5]
// Lgammal is approximated by polynomial within [-.156450 ; .156126 ] range
data8 0x3C71B82D6B2B3304, 0x3917186E3C0DC231 //A0
data8 0x405ED72E0829AE02, 0x3C960C25157980EB //A1
data8 0x40BCECC32EC22F9B, 0x3D5D8335A32F019C //A2
data8 0x929EC2B1FB931F17, 0x00004012 //A3
data8 0xD112EF96D37316DE, 0x00004018 //A4
data8 0x9F00BB9BB13416AB, 0x0000401F //A5
data8 0x425F7D8D5BDCB223 //A6
data8 0x42C9A8D00C776CC6 //A7
data8 0x433557FD8C481424 //A8
data8 0x43A209221A953EF0 //A9
data8 0x440EDC98D5618AB7 //A10
data8 0x447AABD25E367378 //A11
data8 0x44E73DE20CC3B288 //A12
data8 0x455465257B4E0BD8 //A13
data8 0x45C2011532085353 //A14
data8 0x462FEE4CC191945B //A15
data8 0x469C63AEEFEF0A7F //A16
data8 0x4709D045390A3810 //A17
data8 0x4778D360873C9F64 //A18
data8 0x47E26965BE9A682A //A19
// Polynomial coefficients for right root on [-7, -6]
// Lgammal is approximated by polynomial within [-.154582 ; .154521 ] range
data8 0x3C75F103A1B00A48, 0x391C041C190C726D //A0
data8 0x40869DE49E3AF2AA, 0x3D1C17E1F813063B //A1
data8 0x410FCE23484CFD10, 0x3DB6F38C2F11DAB9 //A2
data8 0xEF281D1E1BE2055A, 0x00004019 //A3
data8 0xFCE3DA92AC55DFF8, 0x00004022 //A4
data8 0x8E9EA838A20BD58E, 0x0000402C //A5
data8 0x4354F21E2FB9E0C9 //A6
data8 0x43E9500994CD4F09 //A7
data8 0x447F3A2C23C033DF //A8
data8 0x45139152656606D8 //A9
data8 0x45A8D45F8D3BF2E8 //A10
data8 0x463FD32110E5BFE5 //A11
data8 0x46D490B3BDBAE0BE //A12
data8 0x476AC3CAD905DD23 //A13
data8 0x48018558217AD473 //A14
data8 0x48970AF371D30585 //A15
data8 0x492E6273A8BEFFE3 //A16
data8 0x49C47CC9AE3F1073 //A17
data8 0x4A5D38E8C35EFF45 //A18
data8 0x4AF0123E89694CD8 //A19
// Polynomial coefficients for right root on [-8, -7]
// Lgammal is approximated by polynomial within [-.154217 ; .154208 ] range
data8 0xBCD2507D818DDD68, 0xB97F6940EA2871A0 //A0
data8 0x40B3B407AA387BCB, 0x3D6320238F2C43D1 //A1
data8 0x41683E85DAAFBAC7, 0x3E148D085958EA3A //A2
data8 0x9F2A95AF1E10A548, 0x00004022 //A3
data8 0x92F21522F482300E, 0x0000402E //A4
data8 0x90B51AB03A1F244D, 0x0000403A //A5
data8 0x44628E1C70EF534F //A6
data8 0x452393E2BC32D244 //A7
data8 0x45E5164141F4BA0B //A8
data8 0x46A712B3A8AF5808 //A9
data8 0x47698FD36CEDD0F2 //A10
data8 0x482C9AE6BBAA3637 //A11
data8 0x48F023821857C8E9 //A12
data8 0x49B2569053FC106F //A13
data8 0x4A74F646D5C1604B //A14
data8 0x4B3811CF5ABA4934 //A15
data8 0x4BFBB5DD6C84E233 //A16
data8 0x4CC05021086F637B //A17
data8 0x4D8450A345B0FB49 //A18
data8 0x4E43825848865DB2 //A19
// Polynomial coefficients for right root on [-9, -8]
// Lgammal is approximated by polynomial within [-.154160 ; .154158 ] range
data8 0x3CDF4358564F2B46, 0x397969BEE6042F81 //A0
data8 0x40E3B088FED67721, 0x3D82787BA937EE85 //A1
data8 0x41C83A3893550EF4, 0x3E542ED57E244DA8 //A2
data8 0x9F003C6DC56E0B8E, 0x0000402B //A3
data8 0x92BDF64A3213A699, 0x0000403A //A4
data8 0x9074F503AAD417AF, 0x00004049 //A5
data8 0x4582843E1313C8CD //A6
data8 0x467387BD6A7826C1 //A7
data8 0x4765074E788CF440 //A8
data8 0x4857004DD9D1E09D //A9
data8 0x4949792ED7530EAF //A10
data8 0x4A3C7F089A292ED3 //A11
data8 0x4B30125BF0AABB86 //A12
data8 0x4C224175195E307E //A13
data8 0x4D14DC4C8B32C08D //A14
data8 0x4E07F1DB2786197E //A15
data8 0x4EFB8EA1C336DACB //A16
data8 0x4FF03797EACD0F23 //A17
data8 0x50E4304A8E68A730 //A18
data8 0x51D3618FB2EC9F93 //A19
// Polynomial coefficients for right root on [-10, -9]
// Lgammal is approximated by polynomial within [-.154152 ; .154152 ] range
data8 0x3D42F34DA97ECF0C, 0x39FD1256F345B0D0 //A0
data8 0x4116261203919787, 0x3DC12D44055588EB //A1
data8 0x422EA8F32FB7FE99, 0x3ED849CE4E7B2D77 //A2
data8 0xE25BAF73477A57B5, 0x00004034 //A3
data8 0xEB021FD10060504A, 0x00004046 //A4
data8 0x8220A208EE206C5F, 0x00004059 //A5
data8 0x46B2C3903EC9DA14 //A6
data8 0x47D64393744B9C67 //A7
data8 0x48FAF79CCDC604DD //A8
data8 0x4A20975DB8061EBA //A9
data8 0x4B44AB9CBB38DB21 //A10
data8 0x4C6A032F60094FE9 //A11
data8 0x4D908103927634B4 //A12
data8 0x4EB516CA21D30861 //A13
data8 0x4FDB1BF12C58D318 //A14
data8 0x510180AAE094A553 //A15
data8 0x5226A8F2A2D45D57 //A16
data8 0x534E00B6B0C8B809 //A17
data8 0x5475022FE21215B2 //A18
data8 0x5596B02BF6C5E19B //A19
// Polynomial coefficients for right root on [-11, -10]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0x3D7AA9C2E2B1029C, 0x3A15FB37578544DB //A0
data8 0x414BAF825A0C91D4, 0x3DFB9DA2CE398747 //A1
data8 0x4297F3EC8AE0AF03, 0x3F34208B55FB8781 //A2
data8 0xDD0C97D3197F56DE, 0x0000403E //A3
data8 0x8F6F3AF7A5499674, 0x00004054 //A4
data8 0xC68DA1AF6D878EEB, 0x00004069 //A5
data8 0x47F1E4E1E2197CE0 //A6
data8 0x494A8A28E597C3EB //A7
data8 0x4AA4175D0D35D705 //A8
data8 0x4BFEE6F0AF69E814 //A9
data8 0x4D580FE7B3DBB3C6 //A10
data8 0x4EB2ECE60E4608AF //A11
data8 0x500E04BE3E2B4F24 //A12
data8 0x5167F9450F0FB8FD //A13
data8 0x52C342BDE747603F //A14
data8 0x541F1699D557268C //A15
data8 0x557927C5F079864E //A16
data8 0x56D4D10FEEDB030C //A17
data8 0x5832385DF86AD28A //A18
data8 0x598898914B4D6523 //A19
// Polynomial coefficients for right root on [-12, -11]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0xBD96F61647C58B03, 0xBA3ABB0C2A6C755B //A0
data8 0x418308A82714B70D, 0x3E1088FC6A104C39 //A1
data8 0x4306A493DD613C39, 0x3FB2341ECBF85741 //A2
data8 0x8FA8FE98339474AB, 0x00004049 //A3
data8 0x802CCDF570BA7942, 0x00004062 //A4
data8 0xF3F748AF11A32890, 0x0000407A //A5
data8 0x493E3B567EF178CF //A6
data8 0x4ACED38F651BA362 //A7
data8 0x4C600B357337F946 //A8
data8 0x4DF0F71A52B54CCF //A9
data8 0x4F8229F3B9FA2C70 //A10
data8 0x5113A4C4979B770E //A11
data8 0x52A56BC367F298D5 //A12
data8 0x543785CF31842DC0 //A13
data8 0x55C9FC37E3E40896 //A14
data8 0x575CD5D1BA556C82 //A15
data8 0x58F00A7AD99A9E08 //A16
data8 0x5A824088688B008D //A17
data8 0x5C15F75EF7E08EBD //A18
data8 0x5DA462EA902F0C90 //A19
// Polynomial coefficients for right root on [-13, -12]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0x3DC3191752ACFC9D, 0x3A26CB6629532DBF //A0
data8 0x41BC8CFC051191BD, 0x3E68A84DA4E62AF2 //A1
data8 0x43797926294A0148, 0x400F345FF3723CFF //A2
data8 0xF26D2AF700B82625, 0x00004053 //A3
data8 0xA238B24A4B1F7B15, 0x00004070 //A4
data8 0xE793B5C0A41A264F, 0x0000408C //A5
data8 0x4A9585BDDACE863D //A6
data8 0x4C6075953448088A //A7
data8 0x4E29B2F38D1FC670 //A8
data8 0x4FF4619B079C440F //A9
data8 0x51C05DAE118D8AD9 //A10
data8 0x538A8C7F87326AD4 //A11
data8 0x5555B6937588DAB3 //A12
data8 0x5721E1F8B6E6A7DB //A13
data8 0x58EDA1D7A77DD6E5 //A14
data8 0x5AB8A9616B7DC9ED //A15
data8 0x5C84942AA209ED17 //A16
data8 0x5E518FC34C6F54EF //A17
data8 0x601FB3F17BCCD9A0 //A18
data8 0x61E61128D512FE97 //A1
// Polynomial coefficients for right root on [-14, -13]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0xBE170D646421B3F5, 0xBAAD95F79FCB5097 //A0
data8 0x41F7328CBFCD9AC7, 0x3E743B8B1E8AEDB1 //A1
data8 0x43F0D0FA2DBDA237, 0x40A0422D6A227B55 //A2
data8 0x82082DF2D32686CC, 0x0000405F //A3
data8 0x8D64EE9B42E68B43, 0x0000407F //A4
data8 0xA3FFD82E08C5F1F1, 0x0000409F //A5
data8 0x4BF8C49D99123454 //A6
data8 0x4DFEC79DDF11342F //A7
data8 0x50038615A892F6BD //A8
data8 0x520929453DB32EF1 //A9
data8 0x54106A7808189A7F //A10
data8 0x5615A302D03C207B //A11
data8 0x581CC175AA736F5E //A12
data8 0x5A233E071147C017 //A13
data8 0x5C29E81917243F22 //A14
data8 0x5E3184B0B5AC4707 //A15
data8 0x6037C11DE62D8388 //A16
data8 0x6240787C4B1C9D6C //A17
data8 0x6448289235E80977 //A18
data8 0x664B5352C6C3449E //A19
// Polynomial coefficients for right root on [-15, -14]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0x3E562C2E34A9207D, 0x3ADC00DA3DFF7A83 //A0
data8 0x42344C3B2F0D90AB, 0x3EB8A2E979F24536 //A1
data8 0x4469BFFF28B50D07, 0x41181E3D05C1C294 //A2
data8 0xAE38F64DCB24D9F8, 0x0000406A //A3
data8 0xA5C3F52C1B350702, 0x0000408E //A4
data8 0xA83BC857BCD67A1B, 0x000040B2 //A5
data8 0x4D663B4727B4D80A //A6
data8 0x4FA82C965B0F7788 //A7
data8 0x51EAD58C02908D95 //A8
data8 0x542E427970E073D8 //A9
data8 0x56714644C558A818 //A10
data8 0x58B3EC2040C77BAE //A11
data8 0x5AF72AE6A83D45B1 //A12
data8 0x5D3B214F611F5D12 //A13
data8 0x5F7FF5E49C54E92A //A14
data8 0x61C2E917AB765FB2 //A15
data8 0x64066FD70907B4C1 //A16
data8 0x664B3998D60D0F9B //A17
data8 0x689178710782FA8B //A18
data8 0x6AD14A66C1C7BEC3 //A19
// Polynomial coefficients for right root on [-16, -15]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0xBE6D7E7192615BAE, 0xBB0137677D7CC719 //A0
data8 0x4273077763F6628C, 0x3F09250FB8FC8EC9 //A1
data8 0x44E6A1BF095B1AB3, 0x4178D5A74F6CB3B3 //A2
data8 0x8F8E0D5060FCC76E, 0x00004076 //A3
data8 0x800CC1DCFF092A63, 0x0000409E //A4
data8 0xF3AB0BA9D14CDA72, 0x000040C5 //A5
data8 0x4EDE3000A2F6D54F //A6
data8 0x515EC613B9C8E241 //A7
data8 0x53E003309FEEEA96 //A8
data8 0x5660ED908D7C9A90 //A9
data8 0x58E21E9B517B1A50 //A10
data8 0x5B639745E4374EE2 //A11
data8 0x5DE55BB626B2075D //A12
data8 0x606772B7506BA747 //A13
data8 0x62E9E581AB2E057B //A14
data8 0x656CBAD1CF85D396 //A15
data8 0x67EFF4EBD7989872 //A16
data8 0x6A722D2B19B7E2F9 //A17
data8 0x6CF5DEB3073B0743 //A18
data8 0x6F744AC11550B93A //A19
// Polynomial coefficients for right root on [-17, -16]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0xBEDCC6291188207E, 0xBB872E3FDD48F5B7 //A0
data8 0x42B3076EE7525EF9, 0x3F6687A5038CA81C //A1
data8 0x4566A1AAD96EBCB5, 0x421F0FEDFBF548D2 //A2
data8 0x8F8D4D3DE9850DBA, 0x00004082 //A3
data8 0x800BDD6DA2CE1859, 0x000040AE //A4
data8 0xF3A8EC4C9CDC1CE5, 0x000040D9 //A5
data8 0x505E2FAFDB812628 //A6
data8 0x531EC5B3A7508719 //A7
data8 0x55E002F77E99B628 //A8
data8 0x58A0ED4C9B4DAE54 //A9
data8 0x5B621E4A8240F90C //A10
data8 0x5E2396E5C8849814 //A11
data8 0x60E55B43D8C5CE71 //A12
data8 0x63A7722F5D45D01D //A13
data8 0x6669E4E010DCE45A //A14
data8 0x692CBA120D5E78F6 //A15
data8 0x6BEFF4045350B22E //A16
data8 0x6EB22C9807C21819 //A17
data8 0x7175DE20D04617C4 //A18
data8 0x74344AB87C6D655F //A19
// Polynomial coefficients for right root on [-18, -17]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0xBF28AEEE7B61D77C, 0xBBDBBB5FC57ABF79 //A0
data8 0x42F436F56B3B8A0C, 0x3FA43EE3C5C576E9 //A1
data8 0x45E98A22535D115D, 0x42984678BE78CC48 //A2
data8 0xAC176F3775E6FCFC, 0x0000408E //A3
data8 0xA3114F53A9FEB922, 0x000040BE //A4
data8 0xA4D168A8334ABF41, 0x000040EE //A5
data8 0x51E5B0E7EC7182BB //A6
data8 0x54E77D67B876EAB6 //A7
data8 0x57E9F7C30C09C4B6 //A8
data8 0x5AED29B0488614CA //A9
data8 0x5DF09486F87E79F9 //A10
data8 0x60F30B199979654E //A11
data8 0x63F60E02C7DCCC5F //A12
data8 0x66F9B8A00EB01684 //A13
data8 0x69FE2D3ED0700044 //A14
data8 0x6D01C8363C7DCC84 //A15
data8 0x700502B29C2F06E3 //A16
data8 0x730962B4500F4A61 //A17
data8 0x76103C6ED099192A //A18
data8 0x79100C7132CFD6E3 //A19
// Polynomial coefficients for right root on [-19, -18]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0x3F3C19A53328A0C3, 0x3BE04ADC3FBE1458 //A0
data8 0x4336C16C16C16C19, 0x3FE58CE3AC4A7C28 //A1
data8 0x46702E85C0898B70, 0x432C922E412CEC6E //A2
data8 0xF57B99A1C034335D, 0x0000409A //A3
data8 0x82EC9634223DF909, 0x000040CF //A4
data8 0x94F66D7557E2EA60, 0x00004103 //A5
data8 0x5376118B79AE34D0 //A6
data8 0x56BAE7106D52E548 //A7
data8 0x5A00BD48CC8E25AB //A8
data8 0x5D4529722821B493 //A9
data8 0x608B1654AF31BBC1 //A10
data8 0x63D182CC98AEA859 //A11
data8 0x6716D43D5EEB05E8 //A12
data8 0x6A5DF884FC172E1C //A13
data8 0x6DA3CA7EBB97976B //A14
data8 0x70EA416D0BE6D2EF //A15
data8 0x743176C31EBB65F2 //A16
data8 0x7777C401A8715CF9 //A17
data8 0x7AC1110C6D350440 //A18
data8 0x7E02D0971CF84865 //A19
// Polynomial coefficients for right root on [-20, -19]
// Lgammal is approximated by polynomial within [-.154151 ; .154151 ] range
data8 0xBFAB767F9BE21803, 0xBC5ACEF5BB1BD8B5 //A0
data8 0x4379999999999999, 0x4029241C7F5914C8 //A1
data8 0x46F47AE147AE147A, 0x43AC2979B64B9D7E //A2
data8 0xAEC33E1F67152993, 0x000040A7 //A3
data8 0xD1B71758E219616F, 0x000040DF //A4
data8 0x8637BD05AF6CF468, 0x00004118 //A5
data8 0x55065E9F80F293DE //A6
data8 0x588EADA78C44EE66 //A7
data8 0x5C15798EE22DEF09 //A8
data8 0x5F9E8ABFD644FA63 //A9
data8 0x6325FD7FE29BD7CD //A10
data8 0x66AFFC5C57E1F802 //A11
data8 0x6A3774CD7D5C0181 //A12
data8 0x6DC152724DE2A6FE //A13
data8 0x7149BB138EB3D0C2 //A14
data8 0x74D32FF8A70896C2 //A15
data8 0x785D3749F9C72BD7 //A16
data8 0x7BE5CCF65EBC4E40 //A17
data8 0x7F641A891B5FC652 //A18
data8 0x7FEFFFFFFFFFFFFF //A19
LOCAL_OBJECT_END(lgammal_right_roots_polynomial_data)

LOCAL_OBJECT_START(lgammal_left_roots_polynomial_data)
// Polynomial coefficients for left root on [-3, -2]
// Lgammal is approximated by polynomial within [.084641 ; -.059553 ] range
data8 0xBC0844590979B82E, 0xB8BC7CE8CE2ECC3B //A0
data8 0xBFFEA12DA904B18C, 0xBC91A6B2BAD5EF6E //A1
data8 0x4023267F3C265A51, 0x3CD7055481D03AED //A2
data8 0xA0C2D618645F8E00, 0x0000C003 //A3
data8 0xFA8256664F8CD2BE, 0x00004004 //A4
data8 0xC2C422C103F57158, 0x0000C006 //A5
data8 0x4084373F7CC70AF5 //A6
data8 0xC0A12239BDD6BB95 //A7
data8 0x40BDBA65E2709397 //A8
data8 0xC0DA2D2504DFB085 //A9
data8 0x40F758173CA5BF3C //A10
data8 0xC11506C65C267E72 //A11
data8 0x413318EE3A6B05FC //A12
data8 0xC1517767F247DA98 //A13
data8 0x41701237B4754D73 //A14
data8 0xC18DB8A03BC5C3D8 //A15
data8 0x41AB80953AC14A07 //A16
data8 0xC1C9B7B76638D0A4 //A17
data8 0x41EA727E3033E2D9 //A18
data8 0xC20812C297729142 //A19
//
// Polynomial coefficients for left root on [-4, -3]
// Lgammal is approximated by polynomial within [.147147 ; -.145158 ] range
data8 0xBC3130AE5C4F54DB, 0xB8ED23294C13398A //A0
data8 0xC034B99D966C5646, 0xBCE2E5FE3BC3DBB9 //A1
data8 0x406F76DEAE0436BD, 0x3D14974DDEC057BD //A2
data8 0xE929ACEA5979BE96, 0x0000C00A //A3
data8 0xF47C14F8A0D52771, 0x0000400E //A4
data8 0x88B7BC036937481C, 0x0000C013 //A5
data8 0x4173E8F3AB9FC266 //A6
data8 0xC1B7DBBE062FB11B //A7
data8 0x41FD2F76DE7A47A7 //A8
data8 0xC242225FE53B124D //A9
data8 0x4286D12AE2FBFA30 //A10
data8 0xC2CCFFC267A3C4C0 //A11
data8 0x431294E10008E014 //A12
data8 0xC357FAC8C9A2DF6A //A13
data8 0x439F2190AB9FAE01 //A14
data8 0xC3E44C1D8E8C67C3 //A15
data8 0x442A8901105D5A38 //A16
data8 0xC471C4421E908C3A //A17
data8 0x44B92CD4D59D6D17 //A18
data8 0xC4FB3A078B5247FA //A19
// Polynomial coefficients for left root on [-5, -4]
// Lgammal is approximated by polynomial within [.155671 ; -.155300 ] range
data8 0xBC57BF3C6E8A94C1, 0xB902FB666934AC9E //A0
data8 0xC05D224A3EF9E41F, 0xBCF6F5713913E440 //A1
data8 0x40BB533C678A3955, 0x3D688E53E3C72538 //A2
data8 0x869FBFF732E99B84, 0x0000C012 //A3
data8 0xBA9537AD61392DEC, 0x00004018 //A4
data8 0x89EAE8B1DEA06B05, 0x0000C01F //A5
data8 0x425A8C5C53458D3C //A6
data8 0xC2C5068B3ED6509B //A7
data8 0x4330FFA575E99B4E //A8
data8 0xC39BEC12DDDF7669 //A9
data8 0x44073825725F74F9 //A10
data8 0xC47380EBCA299047 //A11
data8 0x44E084DD9B666437 //A12
data8 0xC54C2DA6BF787ACF //A13
data8 0x45B82D65C8D6FA42 //A14
data8 0xC624D62113FE950A //A15
data8 0x469200CC19B45016 //A16
data8 0xC6FFDDC6DD938E2E //A17
data8 0x476DD7C07184B9F9 //A18
data8 0xC7D554A30085C052 //A19
// Polynomial coefficients for left root on [-6, -5]
// Lgammal is approximated by polynomial within [.157425 ; -.157360 ] range
data8 0x3C9E20A87C8B79F1, 0x39488BE34B2427DB //A0
data8 0xC08661F6A43A5E12, 0xBD3D912526D759CC //A1
data8 0x410F79DCB794F270, 0x3DB9BEE7CD3C1BF5 //A2
data8 0xEB7404450D0005DB, 0x0000C019 //A3
data8 0xF7AE9846DFE4D4AB, 0x00004022 //A4
data8 0x8AF535855A95B6DA, 0x0000C02C //A5
data8 0x43544D54E9FE240E //A6
data8 0xC3E8684E40CE6CFC //A7
data8 0x447DF44C1D803454 //A8
data8 0xC512AC305439B2BA //A9
data8 0x45A79226AF79211A //A10
data8 0xC63E0DFF7244893A //A11
data8 0x46D35216C3A83AF3 //A12
data8 0xC76903BE0C390E28 //A13
data8 0x48004A4DECFA4FD5 //A14
data8 0xC8954FBD243DB8BE //A15
data8 0x492BF3A31EB18DDA //A16
data8 0xC9C2C6A864521F3A //A17
data8 0x4A5AB127C62E8DA1 //A18
data8 0xCAECF60EF3183C57 //A19
// Polynomial coefficients for left root on [-7, -6]
// Lgammal is approximated by polynomial within [.157749 ; -.157739 ] range
data8 0x3CC9B9E8B8D551D6, 0x3961813C8E1E10DB //A0
data8 0xC0B3ABF7A5CEA91F, 0xBD55638D4BCB4CC4 //A1
data8 0x4168349A25504236, 0x3E0287ECE50CCF76 //A2
data8 0x9EC8ED6E4C219E67, 0x0000C022 //A3
data8 0x9279EB1B799A3FF3, 0x0000402E //A4
data8 0x90213EF8D9A5DBCF, 0x0000C03A //A5
data8 0x4462775E857FB71C //A6
data8 0xC52377E70B45FDBF //A7
data8 0x45E4F3D28EDA8C28 //A8
data8 0xC6A6E85571BD2D0B //A9
data8 0x47695BB17E74DF74 //A10
data8 0xC82C5AC0ED6A662F //A11
data8 0x48EFF8159441C2E3 //A12
data8 0xC9B22602C1B68AE5 //A13
data8 0x4A74BA8CE7B34100 //A14
data8 0xCB37C7E208482E4B //A15
data8 0x4BFB5A1D57352265 //A16
data8 0xCCC01CB3021212FF //A17
data8 0x4D841613AC3431D1 //A18
data8 0xCE431C9E9EE43AD9 //A19
// Polynomial coefficients for left root on [-8, -7]
// Lgammal is approximated by polynomial within [.157799 ; -.157798 ] range
data8 0xBCF9C7A33AD9478C, 0xB995B0470F11E5ED //A0
data8 0xC0E3AF76FE4C2F8B, 0xBD8DBCD503250511 //A1
data8 0x41C838E76CAAF0D5, 0x3E5D79F5E2E069C3 //A2
data8 0x9EF345992B262CE0, 0x0000C02B //A3
data8 0x92AE0292985FD559, 0x0000403A //A4
data8 0x90615420C08F7D8C, 0x0000C049 //A5
data8 0x45828139342CEEB7 //A6
data8 0xC67384066C31E2D3 //A7
data8 0x476502BC4DAC2C35 //A8
data8 0xC856FAADFF22ADC6 //A9
data8 0x49497243255AB3CE //A10
data8 0xCA3C768489520F6B //A11
data8 0x4B300D1EA47AF838 //A12
data8 0xCC223B0508AC620E //A13
data8 0x4D14D46583338CD8 //A14
data8 0xCE07E7A87AA068E4 //A15
data8 0x4EFB811AD2F8BEAB //A16
data8 0xCFF0351B51508523 //A17
data8 0x50E4364CCBF53100 //A18
data8 0xD1D33CFD0BF96FA6 //A19
// Polynomial coefficients for left root on [-9, -8]
// Lgammal is approximated by polynomial within [.157806 ; -.157806 ] range
data8 0x3D333E4438B1B9D4, 0x39E7B956B83964C1 //A0
data8 0xC11625EDFC63DCD8, 0xBDCF39625709EFAC //A1
data8 0x422EA8C150480F16, 0x3EC16ED908AB7EDD //A2
data8 0xE2598725E2E11646, 0x0000C034 //A3
data8 0xEAFF2346DE3EBC98, 0x00004046 //A4
data8 0x821E90DE12A0F05F, 0x0000C059 //A5
data8 0x46B2C334AE5366FE //A6
data8 0xC7D64314B43191B6 //A7
data8 0x48FAF6ED5899E01B //A8
data8 0xCA2096E4472AF37D //A9
data8 0x4B44AAF49FB7E4C8 //A10
data8 0xCC6A02469F2BD920 //A11
data8 0x4D9080626D2EFC07 //A12
data8 0xCEB515EDCF0695F7 //A13
data8 0x4FDB1AC69BF36960 //A14
data8 0xD1017F8274339270 //A15
data8 0x5226A684961BAE2F //A16
data8 0xD34E085C088404A5 //A17
data8 0x547511892FF8960E //A18
data8 0xD5968FA3B1ED67A9 //A19
// Polynomial coefficients for left root on [-10, -9]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0xBD355818A2B42BA2, 0xB9B7320B6A0D61EA //A0
data8 0xC14BAF7DA5F3770E, 0xBDE64AF9A868F719 //A1
data8 0x4297F3E8791F9CD3, 0x3F2A553E59B4835E //A2
data8 0xDD0C5F7E551BD13C, 0x0000C03E //A3
data8 0x8F6F0A3B2EB08BBB, 0x00004054 //A4
data8 0xC68D4D5AD230BA08, 0x0000C069 //A5
data8 0x47F1E4D8C35D1A3E //A6
data8 0xC94A8A191DB0A466 //A7
data8 0x4AA4174F65FE6AE8 //A8
data8 0xCBFEE6D90F94E9DD //A9
data8 0x4D580FD3438BE16C //A10
data8 0xCEB2ECD456D50224 //A11
data8 0x500E049F7FE64546 //A12
data8 0xD167F92D9600F378 //A13
data8 0x52C342AE2B43261A //A14
data8 0xD41F15DEEDA4B67E //A15
data8 0x55792638748AFB7D //A16
data8 0xD6D4D760074F6E6B //A17
data8 0x5832469D58ED3FA9 //A18
data8 0xD988769F3DC76642 //A19
// Polynomial coefficients for left root on [-11, -10]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0xBDA050601F39778A, 0xBA0D4D1CE53E8241 //A0
data8 0xC18308A7D8EA4039, 0xBE370C379D3EAD41 //A1
data8 0x4306A49380644E6C, 0x3FBBB143C0E7B5C8 //A2
data8 0x8FA8FB233E4AA6D2, 0x0000C049 //A3
data8 0x802CC9D8AEAC207D, 0x00004062 //A4
data8 0xF3F73EE651A37A13, 0x0000C07A //A5
data8 0x493E3B550A7B9568 //A6
data8 0xCACED38DAA060929 //A7
data8 0x4C600B346BAB3BC6 //A8
data8 0xCDF0F719193E3D26 //A9
data8 0x4F8229F24528B151 //A10
data8 0xD113A4C2D32FBBE2 //A11
data8 0x52A56BC13DC4474D //A12
data8 0xD43785CFAF5E3CE3 //A13
data8 0x55C9FC3EA5941202 //A14
data8 0xD75CD545A3341AF5 //A15
data8 0x58F009911F77C282 //A16
data8 0xDA8246294D210BEC //A17
data8 0x5C1608AAC32C3A8E //A18
data8 0xDDA446E570A397D5 //A19
// Polynomial coefficients for left root on [-12, -11]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0x3DEACBB3081C502E, 0x3A8AA6F01DEDF745 //A0
data8 0xC1BC8CFBFB0A9912, 0xBE6556B6504A2AE6 //A1
data8 0x43797926206941D7, 0x40289A9644C2A216 //A2
data8 0xF26D2A78446D0839, 0x0000C053 //A3
data8 0xA238B1D937FFED38, 0x00004070 //A4
data8 0xE793B4F6DE470538, 0x0000C08C //A5
data8 0x4A9585BDC44DC45D //A6
data8 0xCC60759520342C47 //A7
data8 0x4E29B2F3694C0404 //A8
data8 0xCFF4619AE7B6BBAB //A9
data8 0x51C05DADF52B89E8 //A10
data8 0xD38A8C7F48819A4A //A11
data8 0x5555B6932D687860 //A12
data8 0xD721E1FACB6C1B5B //A13
data8 0x58EDA1E2677C8F91 //A14
data8 0xDAB8A8EC523C1F71 //A15
data8 0x5C84930133F30411 //A16
data8 0xDE51952FDFD1EC49 //A17
data8 0x601FCCEC1BBD25F1 //A18
data8 0xE1E5F2D76B610920 //A19
// Polynomial coefficients for left root on [-13, -12]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0xBE01612F373268ED, 0xBA97B7A18CDF103B //A0
data8 0xC1F7328CBF7A4FAC, 0xBE89A25A6952F481 //A1
data8 0x43F0D0FA2DBDA237, 0x40A0422EC1CE6084 //A2
data8 0x82082DF2D32686C5, 0x0000C05F //A3
data8 0x8D64EE9B42E68B36, 0x0000407F //A4
data8 0xA3FFD82E08C630C9, 0x0000C09F //A5
data8 0x4BF8C49D99123466 //A6
data8 0xCDFEC79DDF1119ED //A7
data8 0x50038615A892D242 //A8
data8 0xD20929453DC8B537 //A9
data8 0x54106A78083BA1EE //A10
data8 0xD615A302C69E27B2 //A11
data8 0x581CC175870FF16F //A12
data8 0xDA233E0979E12B74 //A13
data8 0x5C29E822BC568C80 //A14
data8 0xDE31845DB5340FBC //A15
data8 0x6037BFC6D498D5F9 //A16
data8 0xE2407D92CD613E82 //A17
data8 0x64483B9B62367EB7 //A18
data8 0xE64B2DC830E8A799 //A1
// Polynomial coefficients for left root on [-14, -13]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0x3E563D0B930B371F, 0x3AE779957E14F012 //A0
data8 0xC2344C3B2F083767, 0xBEC0B7769AA3DD66 //A1
data8 0x4469BFFF28B50D07, 0x41181E3F13ED2401 //A2
data8 0xAE38F64DCB24D9EE, 0x0000C06A //A3
data8 0xA5C3F52C1B3506F2, 0x0000408E //A4
data8 0xA83BC857BCD6BA92, 0x0000C0B2 //A5
data8 0x4D663B4727B4D81A //A6
data8 0xCFA82C965B0F62E9 //A7
data8 0x51EAD58C02905B71 //A8
data8 0xD42E427970FA56AD //A9
data8 0x56714644C57D8476 //A10
data8 0xD8B3EC2037EC95F2 //A11
data8 0x5AF72AE68BBA5B3D //A12
data8 0xDD3B2152C67AA6B7 //A13
data8 0x5F7FF5F082861B8B //A14
data8 0xE1C2E8BE125A5B7A //A15
data8 0x64066E92FE9EBE7D //A16
data8 0xE64B4201CDF9F138 //A17
data8 0x689186351E58AA88 //A18
data8 0xEAD132A585DFC60A //A19
// Polynomial coefficients for left root on [-15, -14]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0xBE6D7DDE12700AC1, 0xBB1E025BF1667FB5 //A0
data8 0xC273077763F60AD5, 0xBF2A1698184C7A9A //A1
data8 0x44E6A1BF095B1AB3, 0x4178D5AE8A4A2874 //A2
data8 0x8F8E0D5060FCC767, 0x0000C076 //A3
data8 0x800CC1DCFF092A57, 0x0000409E //A4
data8 0xF3AB0BA9D14D37D1, 0x0000C0C5 //A5
data8 0x4EDE3000A2F6D565 //A6
data8 0xD15EC613B9C8C800 //A7
data8 0x53E003309FEECCAA //A8
data8 0xD660ED908D8B15C4 //A9
data8 0x58E21E9B51A1C4AE //A10
data8 0xDB639745DB82210D //A11
data8 0x5DE55BB60C68FCF6 //A12
data8 0xE06772BA3FCA23C6 //A13
data8 0x62E9E58B4F702C31 //A14
data8 0xE56CBA49B071ABE2 //A15
data8 0x67EFF31E4F2BA36A //A16
data8 0xEA7232C8804F32C3 //A17
data8 0x6CF5EFEE929A0928 //A18
data8 0xEF742EE03EC3E8FF //A19
// Polynomial coefficients for left root on [-16, -15]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0xBEDCC628FEAC7A1B, 0xBB80582C8BEBB198 //A0
data8 0xC2B3076EE752595E, 0xBF5388F55AFAE53E //A1
data8 0x4566A1AAD96EBCB5, 0x421F0FEFE2444293 //A2
data8 0x8F8D4D3DE9850DB2, 0x0000C082 //A3
data8 0x800BDD6DA2CE184C, 0x000040AE //A4
data8 0xF3A8EC4C9CDC7A43, 0x0000C0D9 //A5
data8 0x505E2FAFDB81263F //A6
data8 0xD31EC5B3A7506CD9 //A7
data8 0x55E002F77E999810 //A8
data8 0xD8A0ED4C9B5C2900 //A9
data8 0x5B621E4A8267C401 //A10
data8 0xDE2396E5BFCFDA7A //A11
data8 0x60E55B43BE6F9A79 //A12
data8 0xE3A772324C7405FA //A13
data8 0x6669E4E9B7E57A2D //A14
data8 0xE92CB989F8A8FB37 //A15
data8 0x6BEFF2368849A36E //A16
data8 0xEEB23234FE191D55 //A17
data8 0x7175EF5D1080B105 //A18
data8 0xF4342ED7B1B7BE31 //A19
// Polynomial coefficients for left root on [-17, -16]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0xBF28AEEE7B58C790, 0xBBC4448DE371FA0A //A0
data8 0xC2F436F56B3B89B1, 0xBF636755245AC63A //A1
data8 0x45E98A22535D115D, 0x4298467DA93DB784 //A2
data8 0xAC176F3775E6FCF2, 0x0000C08E //A3
data8 0xA3114F53A9FEB908, 0x000040BE //A4
data8 0xA4D168A8334AFE5A, 0x0000C0EE //A5
data8 0x51E5B0E7EC7182CF //A6
data8 0xD4E77D67B876D6B4 //A7
data8 0x57E9F7C30C098C83 //A8
data8 0xDAED29B0489EF7A7 //A9
data8 0x5DF09486F8A524B8 //A10
data8 0xE0F30B19910A2393 //A11
data8 0x63F60E02AB3109F4 //A12
data8 0xE6F9B8A3431854D5 //A13
data8 0x69FE2D4A6D94218E //A14
data8 0xED01C7E272A73560 //A15
data8 0x7005017D82B186B6 //A16
data8 0xF3096A81A69BD8AE //A17
data8 0x76104951BAD67D5C //A18
data8 0xF90FECC99786FD5B //A19
// Polynomial coefficients for left root on [-18, -17]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0x3F3C19A53328E26A, 0x3BE238D7BA036B3B //A0
data8 0xC336C16C16C16C13, 0xBFEACE245DEC56F3 //A1
data8 0x46702E85C0898B70, 0x432C922B64FD1DA4 //A2
data8 0xF57B99A1C0343350, 0x0000C09A //A3
data8 0x82EC9634223DF90D, 0x000040CF //A4
data8 0x94F66D7557E3237D, 0x0000C103 //A5
data8 0x5376118B79AE34D6 //A6
data8 0xD6BAE7106D52CE49 //A7
data8 0x5A00BD48CC8E11AB //A8
data8 0xDD4529722833E2DF //A9
data8 0x608B1654AF5F46AF //A10
data8 0xE3D182CC90D8723F //A11
data8 0x6716D43D46706AA0 //A12
data8 0xEA5DF888C5B428D3 //A13
data8 0x6DA3CA85888931A6 //A14
data8 0xF0EA40EF2AC7E070 //A15
data8 0x743175D1A251AFCD //A16
data8 0xF777CB6E2B550D73 //A17
data8 0x7AC11E468A134A51 //A18
data8 0xFE02B6BDD0FC40AA //A19
// Polynomial coefficients for left root on [-19, -18]
// Lgammal is approximated by polynomial within [.157807 ; -.157807 ] range
data8 0xBFAB767F9BE217FC, 0xBC4A5541CE0D8D0D //A0
data8 0xC379999999999999, 0xC01A84981B490BE8 //A1
data8 0x46F47AE147AE147A, 0x43AC2987BBC466EB //A2
data8 0xAEC33E1F67152987, 0x0000C0A7 //A3
data8 0xD1B71758E2196153, 0x000040DF //A4
data8 0x8637BD05AF6D420E, 0x0000C118 //A5
data8 0x55065E9F80F293B2 //A6
data8 0xD88EADA78C44BFA7 //A7
data8 0x5C15798EE22EC6CD //A8
data8 0xDF9E8ABFD67895CF //A9
data8 0x6325FD7FE13B0DE0 //A10
data8 0xE6AFFC5C3DE70858 //A11
data8 0x6A3774CE81C70D43 //A12
data8 0xEDC1527412D8129F //A13
data8 0x7149BABCDA8B7A72 //A14
data8 0xF4D330AD49071BB5 //A15
data8 0x785D4046F4C5F1FD //A16
data8 0xFBE59BFEDBA73FAF //A17
data8 0x7F64BEF2B2EC8DA1 //A18
data8 0xFFEFFFFFFFFFFFFF //A19
LOCAL_OBJECT_END(lgammal_left_roots_polynomial_data)


//==============================================================
// Code
//==============================================================

.section .text
GLOBAL_LIBM_ENTRY(__libm_lgammal)
{ .mfi
      getf.exp           rSignExpX = f8
      // Test x for NaTVal, NaN, +/-0, +/-INF, denormals
      fclass.m           p6,p0  = f8,0x1EF
      addl               r17Ones = 0x1FFFF, r0 // exponent mask
}
{ .mfi
      addl               GR_ad_z_1 = @ltoff(Constants_Z_1#),gp
      fcvt.fx.s1         fXint = f8 // Convert arg to int (int repres. in FR)
      adds               rDelta = 0x3FC, r0
}
;;
{ .mfi
      getf.sig           rSignifX = f8
      fcmp.lt.s1         p15, p14 = f8, f0
      shl                rDelta = rDelta, 20 // single precision 1.5
}
{ .mfi
      ld8                GR_ad_z_1 = [GR_ad_z_1]// get pointer to Constants_Z_1
      fma.s1             fTwo = f1, f1, f1      // 2.0
      addl               rExp8 = 0x10002, r0    // exponent of 8.0
}
;;
{ .mfi
      alloc              rPFS_SAVED = ar.pfs, 0, 34, 4, 0 // get some registers
      fmerge.s           fAbsX = f1, f8                   // |x|
      and                rExpX = rSignExpX, r17Ones       // mask sign bit
}
{ .mib
      addl               rExpHalf = 0xFFFE, r0 // exponent of 0.5
      addl               rExp2 = 0x10000, r0 // exponent of 2.0
      // branch out if x is NaTVal, NaN, +/-0, +/-INF, or denormalized number
(p6)  br.cond.spnt       lgammal_spec
}
;;
_deno_back_to_main_path:
{ .mfi
      // Point to Constants_G_H_h1
      add                rTbl1Addr = 0x040, GR_ad_z_1
      frcpa.s1           fRcpX, p0 = f1, f8 // initial approximation of 1/x
      extr.u             GR_Index1 = rSignifX, 59, 4
}
{ .mib
(p14) cmp.ge.unc         p8, p0 = rExpX, rExp8 // p8 = 1 if x >= 8.0
      adds               rZ625 = 0x3F2, r0
(p8)  br.cond.spnt       lgammal_big_positive // branch out if x >= 8.0
}
;;
{ .mfi
      shladd             rZ1offsett = GR_Index1, 2, GR_ad_z_1  // Point to Z_1
      fmerge.se          fSignifX =  f1, f8 // sifnificand of x
      // Get high 15 bits of significand
      extr.u             GR_X_0 = rSignifX, 49, 15
}
{ .mib
      cmp.lt.unc         p9, p0 = rExpX, rExpHalf // p9 = 1 if |x| < 0.5
      // set p11 if 2 <= x < 4
(p14) cmp.eq.unc         p11, p0 = rExpX, rExp2
(p9)  br.cond.spnt       lgammal_0_half // branch out if |x| < 0.5
}
;;
{ .mfi
      ld4                GR_Z_1 = [rZ1offsett] // Load Z_1
      fms.s1             fA5L = f1, f1, f8 // for 0.75 <= x < 1.3125 path
      shl                rZ625 = rZ625, 20 // sinfle precision 0.625
}
{ .mib
      setf.s             FR_MHalf = rDelta
      // set p10 if x >= 4.0
(p14) cmp.gt.unc         p10, p0 = rExpX, rExp2
      // branch to special path for 4.0 <= x < 8
(p10) br.cond.spnt       lgammal_4_8
}
;;
{ .mfi
      // for 1.3125 <= x < 1.5625 path
      addl               rPolDataPtr= @ltoff(lgammal_loc_min_data),gp
      // argument of polynomial approximation for 1.5625 <= x < 2.25
      fms.s1             fB4 = f8, f1, fTwo
      cmp.eq             p12, p0 = rExpX, rExpHalf
}
{ .mib
      addl               rExpOne = 0xFFFF, r0 // exponent of 1.0
     // set p10 if significand of x >= 1.125
(p11) cmp.le             p11, p0 = 2, GR_Index1
(p11) br.cond.spnt       lgammal_2Q_4
}
;;
{ .mfi
      // point to xMin for 1.3125 <= x < 1.5625 path
      ld8                rPolDataPtr = [rPolDataPtr]
      fcvt.xf            fFltIntX = fXint // RTN(x)
(p14) cmp.eq.unc         p13, p7 = rExpX, rExpOne // p13 set if 1.0 <= x < 2.0
}
{ .mib
      setf.s             FR_FracX = rZ625
      // set p12 if |x| < 0.75
(p12) cmp.gt.unc         p12, p0 = 8, GR_Index1
      // branch out to special path for |x| < 0.75
(p12) br.cond.spnt       lgammal_half_3Q
}
;;
.pred.rel "mutex", p7, p13
{ .mfi
      getf.sig           rXRnd = fXint // integer part of the input value
      fnma.s1            fInvX = f8, fRcpX, f1 // start of 1st NR iteration
      // Get bits 30-15 of X_0 * Z_1
      pmpyshr2.u         GR_X_1 = GR_X_0,GR_Z_1,15
}
{ .mib
(p7)  cmp.eq             p6, p0 = rExpX, rExp2 // p6 set if 2.0 <= x < 2.25
(p13) cmp.le             p6, p0 = 9, GR_Index1
      // branch to special path 1.5625 <= x < 2.25
(p6)  br.cond.spnt       lgammal_13Q_2Q
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      shladd             GR_ad_tbl_1 = GR_Index1, 4, rTbl1Addr // Point to G_1
      fma.s1             fSix = fTwo, fTwo, fTwo // 6.0
      add                GR_ad_q = -0x60, GR_ad_z_1   // Point to Constants_Q
}
{ .mib
      add                rTmpPtr3 = -0x50, GR_ad_z_1
(p13) cmp.gt             p7, p0 = 5, GR_Index1
      // branch to special path 0.75 <= x < 1.3125
(p7)  br.cond.spnt       lgammal_03Q_1Q
}
;;
{ .mfi
      add                rTmpPtr = 8, GR_ad_tbl_1
      fma.s1             fRoot = f8, f1, f1 // x + 1
       // Absolute value of int arg. Will be used as index in table with roots
      sub                rXRnd = r0, rXRnd
}
{ .mib
      ldfe               fA5L = [rPolDataPtr], 16 // xMin
      addl               rNegSingularity = 0x3003E, r0
(p14) br.cond.spnt       lgammal_loc_min
}
;;
{ .mfi
      ldfps              FR_G, FR_H = [GR_ad_tbl_1], 8 // Load G_1, H_1
      nop.f              0
      add                rZ2Addr = 0x140, GR_ad_z_1  // Point to Constants_Z_2
}
{ .mib
      ldfd               FR_h = [rTmpPtr] // Load h_1
      // If arg is less or equal to -2^63
      cmp.geu.unc        p8,p0 = rSignExpX, rNegSingularity
      // Singularity for x < -2^63 since all such arguments are integers
      // branch to special code which deals with singularity
(p8)  br.cond.spnt       lgammal_singularity
}
;;
{ .mfi
      ldfe               FR_log2_hi = [GR_ad_q], 32 // Load log2_hi
      nop.f              0
      extr.u             GR_Index2 = GR_X_1, 6, 4 // Extract bits 6-9 of X_1
}
{ .mfi
      ldfe               FR_log2_lo = [rTmpPtr3], 32 // Load log2_lo
      fms.s1             fDx = f8, f1, fFltIntX // x - RTN(x)
      // index in table with roots and bounds
      adds               rXint = -2, rXRnd
}
;;
{ .mfi
      ldfe               FR_Q4 = [GR_ad_q], 32      // Load Q4
      nop.f              0
      // set p12 if x may be close to negative root: -19.5 < x < -2.0
      cmp.gtu            p12, p0 = 18, rXint
}
{ .mfi
      shladd             GR_ad_z_2 = GR_Index2, 2, rZ2Addr  // Point to Z_2
      fma.s1             fRcpX = fInvX, fRcpX, fRcpX // end of 1st NR iteration
      // Point to Constants_G_H_h2
      add                rTbl2Addr = 0x180, GR_ad_z_1
}
;;
{ .mfi
      shladd             GR_ad_tbl_2 = GR_Index2, 4, rTbl2Addr // Point to G_2
      // set p9 if x is integer and negative
      fcmp.eq.s1         p9, p0 = f8,fFltIntX
      // Point to Constants_G_H_h3
      add                rTbl3Addr = 0x280, GR_ad_z_1
}
{ .mfi
      ld4                GR_Z_2 = [GR_ad_z_2] // Load Z_2
      nop.f              0
      sub                GR_N = rExpX, rExpHalf, 1
}
;;
{ .mfi
      ldfe               FR_Q3 = [rTmpPtr3], 32 // Load Q3
      nop.f              0
      // Point to lnsin polynomial coefficients
      adds               rLnSinDataPtr = 864, rTbl3Addr
}
{ .mfi
      ldfe               FR_Q2 = [GR_ad_q],32 // Load Q2
      nop.f              0
      add                rTmpPtr = 8, GR_ad_tbl_2
}
;;
{ .mfi
      ldfe               FR_Q1 = [rTmpPtr3] // Load Q1
      fcmp.lt.s1         p0, p15 = fAbsX, fSix // p15 is set when x < -6.0
      // point to table with roots and bounds
      adds               rRootsBndAddr = -1296, GR_ad_z_1
}
{ .mfb
      // Put integer N into rightmost significand
      setf.sig           fFloatN = GR_N
      fma.s1             fThirteen = fSix, fTwo, f1 // 13.0
      // Singularity if -2^63 < x < 0 and x is integer
      // branch to special code which deals with singularity
(p9)  br.cond.spnt       lgammal_singularity
}
;;
{ .mfi
      ldfps              FR_G2, FR_H2 = [GR_ad_tbl_2]  // Load G_2, H_2
      // y = |x|/2^(exponent(x)) - 1.5
      fms.s1             FR_FracX = fSignifX, f1, FR_MHalf
      // Get bits 30-15 of X_1 * Z_2
      pmpyshr2.u         GR_X_2 = GR_X_1,GR_Z_2,15
}
{ .mfi
      ldfd               FR_h2 = [rTmpPtr] // Load h_2
      fma.s1             fDxSqr = fDx, fDx, f0 // deltaX^2
      adds               rTmpPtr3 = 128, rLnSinDataPtr
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      getf.exp           rRoot = fRoot // sign and biased exponent of (x + 1)
      nop.f              0
      // set p6 if -4 < x <= -2
      cmp.eq             p6, p0 = rExpX, rExp2
}
{ .mfi
      ldfpd              fLnSin2, fLnSin2L = [rLnSinDataPtr], 16
      fnma.s1            fInvX = f8, fRcpX, f1 // start of 2nd NR iteration
      sub                rIndexPol = rExpX, rExpHalf // index of polynom
}
;;
{ .mfi
      ldfe               fLnSin4 = [rLnSinDataPtr], 96
      // p10 is set if x is potential "right" root
      // p11 set for possible "left" root
      fcmp.lt.s1         p10, p11 = fDx, f0
      shl                rIndexPol = rIndexPol, 6  // (i*16)*4
}
{ .mfi
      ldfpd              fLnSin18, fLnSin20 = [rTmpPtr3], 16
      nop.f              0
      mov                rExp2tom7 = 0x0fff8 // Exponent of 2^-7
}
;;
{ .mfi
      getf.sig           rSignifDx = fDx // Get significand of RTN(x)
      nop.f              0
      // set p6 if -4 < x <= -3.0
(p6)  cmp.le.unc         p6, p0 = 0x8, GR_Index1
}
{ .mfi
      ldfpd              fLnSin22, fLnSin24 = [rTmpPtr3], 16
      nop.f              0
      // mask sign bit in the exponent of (x + 1)
      and                rRoot = rRoot, r17Ones
}
;;
{ .mfi
      ldfe               fLnSin16 = [rLnSinDataPtr], -80
      nop.f              0
      extr.u             GR_Index3 = GR_X_2, 1, 5 // Extract bits 1-5 of X_2
}
{ .mfi
      ldfpd              fLnSin26, fLnSin28 = [rTmpPtr3], 16
      nop.f              0
      and                rXRnd = 1, rXRnd
}
;;
{ .mfi
      shladd             GR_ad_tbl_3 = GR_Index3, 4, rTbl3Addr // Point to G_3
      fms.s1             fDxSqrL = fDx, fDx, fDxSqr // low part of deltaX^2
      // potential "left" root
(p11) adds               rRootsBndAddr = 560, rRootsBndAddr
}
{ .mib
      ldfpd              fLnSin30, fLnSin32 = [rTmpPtr3], 16
      // set p7 if |x+1| < 2^-7
      cmp.lt             p7, p0 =  rRoot, rExp2tom7
      // branch to special path for |x+1| < 2^-7
(p7)  br.cond.spnt       _closeToNegOne
}
;;
{ .mfi
      ldfps              FR_G3, FR_H3 = [GR_ad_tbl_3], 8 // Load G_3, H_3
      fcmp.lt.s1         p14, p0 = fAbsX, fThirteen // set p14 if x > -13.0
      // base address of polynomial on range [-6.0, -0.75]
      adds               rPolDataPtr = 3440, rTbl3Addr
}
{ .mfi
      // (i*16)*4 + (i*16)*8 - offsett of polynomial on range [-6.0, -0.75]
      shladd             rTmpPtr = rIndexPol, 2, rIndexPol
      fma.s1             fXSqr = FR_FracX, FR_FracX, f0 // y^2
      // point to left "near root" bound
(p12) shladd             rRootsBndAddr = rXint, 4, rRootsBndAddr
}
;;
{ .mfi
      ldfpd              fLnSin34, fLnSin36 = [rTmpPtr3], 16
      fma.s1             fRcpX = fInvX, fRcpX, fRcpX // end of 2nd NR iteration
      // add special offsett if -4 < x <= -3.0
(p6)  adds               rPolDataPtr = 640, rPolDataPtr
}
{ .mfi
      // point to right "near root" bound
      adds               rTmpPtr2 = 8, rRootsBndAddr
      fnma.s1            fMOne = f1, f1, f0 // -1.0
      // Point to Bernulli numbers
      adds               rBernulliPtr = 544, rTbl3Addr
}
;;
{ .mfi
      // left bound of "near root" range
(p12) ld8                rLeftBound = [rRootsBndAddr]
      fmerge.se          fNormDx = f1, fDx // significand of DeltaX
      // base address + offsett for polynomial coeff. on range [-6.0, -0.75]
      add                rPolDataPtr = rPolDataPtr, rTmpPtr
}
{ .mfi
      // right bound of "near root" range
(p12) ld8                rRightBound = [rTmpPtr2]
      fcvt.xf            fFloatN = fFloatN
      // special "Bernulli" numbers for Stirling's formula for -13 < x < -6
(p14) adds               rBernulliPtr = 160, rBernulliPtr
}
;;
{ .mfi
      ldfd               FR_h3 = [GR_ad_tbl_3] // Load h_3
      fmpy.s1            FR_G = FR_G, FR_G2 // G = G_1 * G_2
      adds               rTmpPtr3 = -160, rTmpPtr3
}
{ .mfb
      adds               rTmpPtr = 80, rPolDataPtr
      fadd.s1            FR_H = FR_H, FR_H2 // H = H_1 + H_2
      // p15 is set if -2^63 < x < 6.0 and x is not an integer
      // branch to path with implementation using Stirling's formula for neg. x
(p15) br.cond.spnt       _negStirling
}
;;
{ .mfi
      ldfpd              fA3, fA3L = [rPolDataPtr], 16 // A3
      fma.s1             fDelX4 = fDxSqr, fDxSqr, f0 // deltaX^4
      // Get high 4 bits of signif
      extr.u             rIndex1Dx = rSignifDx, 59, 4
}
{ .mfi
      ldfe               fA5 = [rTmpPtr], -16 // A5
      fadd.s1            FR_h = FR_h, FR_h2 // h = h_1 + h_2
      adds               rLnSinTmpPtr = 16, rLnSinDataPtr
}
;;
{ .mfi
      ldfpd              fA0, fA0L = [rPolDataPtr], 16 // A0
      fma.s1             fLnSin20 = fLnSin20, fDxSqr, fLnSin18
      // Get high 15 bits of significand
      extr.u             rX0Dx = rSignifDx, 49, 15
}
{ .mfi
      ldfe               fA4 = [rTmpPtr], 192 // A4
      fms.s1             fXSqrL = FR_FracX, FR_FracX, fXSqr // low part of y^2
      shladd             GR_ad_z_1 = rIndex1Dx, 2, GR_ad_z_1 // Point to Z_1
}
;;
{ .mfi
      ldfpd              fA1, fA1L = [rPolDataPtr], 16 // A1
      fma.s1             fX4 = fXSqr, fXSqr, f0 // y^4
      adds               rTmpPtr2 = 32, rTmpPtr
}
{ .mfi
      ldfpd              fA18, fA19 = [rTmpPtr], 16 // A18, A19
      fma.s1             fLnSin24 = fLnSin24, fDxSqr, fLnSin22
      nop.i              0
}
;;
{ .mfi
      ldfe               fLnSin6 = [rLnSinDataPtr], 32
      fma.s1             fLnSin28 = fLnSin28, fDxSqr, fLnSin26
      nop.i              0
}
{ .mfi
      ldfe               fLnSin8 = [rLnSinTmpPtr], 32
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA20, fA21 = [rTmpPtr], 16 // A20, A21
      fma.s1             fLnSin32 = fLnSin32, fDxSqr, fLnSin30
      nop.i              0
}
{ .mfi
      ldfpd              fA22, fA23 = [rTmpPtr2], 16 // A22, A23
      fma.s1             fB20 = f1, f1, FR_MHalf // 2.5
(p12) cmp.ltu.unc        p6, p0 = rSignifX, rLeftBound
}
;;
{ .mfi
      ldfpd              fA2, fA2L = [rPolDataPtr], 16 // A2
      fmpy.s1            FR_G = FR_G, FR_G3 // G = (G_1 * G_2) * G_3
      // set p6 if x falls in "near root" range
(p6)  cmp.geu.unc        p6, p0 = rSignifX, rRightBound
}
{ .mfb
      adds               rTmpPtr3 = -64, rTmpPtr
      fadd.s1            FR_H = FR_H, FR_H3 // H = (H_1 + H_2) + H_3
      // branch to special path if x falls in "near root" range
(p6)  br.cond.spnt       _negRoots
}
;;
{ .mfi
      ldfpd              fA24, fA25 = [rTmpPtr2], 16 // A24, A25
      fma.s1             fLnSin36 = fLnSin36, fDxSqr, fLnSin34
(p11) cmp.eq.unc         p7, p0 = 1,rXint // p7 set if  -3.0 < x < -2.5
}
{ .mfi
      adds               rTmpPtr = -48, rTmpPtr
      fma.s1             fLnSin20 = fLnSin20, fDxSqr, fLnSin16
      addl               rDelta = 0x5338, r0 // significand of -2.605859375
}
;;
{ .mfi
      getf.exp           GR_N =  fDx // Get N = exponent of DeltaX
      fma.s1             fX6 = fX4, fXSqr, f0 // y^6
      // p7 set if -2.605859375 <= x < -2.5
(p7)  cmp.gt.unc         p7, p0 = rDelta, GR_X_0
}
{ .mfb
      ld4                GR_Z_1 = [GR_ad_z_1] // Load Z_1
      fma.s1             fDelX8 = fDelX4, fDelX4, f0 // deltaX^8
      // branch to special path for -2.605859375 <= x < -2.5
(p7)  br.cond.spnt       _neg2andHalf
}
;;
{ .mfi
      ldfpd              fA14, fA15 = [rTmpPtr3], 16 // A14, A15
      fadd.s1            FR_h = FR_h, FR_h3 // h = (h_1 + h_2) + h_3
      adds               rTmpPtr2 = 128 , rPolDataPtr
}
{ .mfi
      ldfpd              fA16, fA17 = [rTmpPtr], 16 // A16, A17
      fma.s1             fLnSin28 = fLnSin28, fDelX4, fLnSin24
      adds               rPolDataPtr = 144 , rPolDataPtr
}
;;
{ .mfi
      ldfe               fLnSin10 = [rLnSinDataPtr], 32
      fma.s1             fRes1H = fA3, FR_FracX, f0 // (A3*y)hi
      and                GR_N = GR_N, r17Ones // mask sign bit
}
{ .mfi
      ldfe               fLnSin12 = [rLnSinTmpPtr]
      fma.s1             fDelX6 = fDxSqr, fDelX4, f0 // DeltaX^6
      shladd             GR_ad_tbl_1 = rIndex1Dx, 4, rTbl1Addr // Point to G_1
}
;;
{ .mfi
      ldfe               fA13 = [rPolDataPtr], -32 // A13
      fma.s1             fA4 = fA5, FR_FracX, fA4   // A5*y + A4
      // Get bits 30-15 of X_0 * Z_1
      pmpyshr2.u         GR_X_1 = rX0Dx, GR_Z_1, 15
}
{ .mfi
      ldfe               fA12 = [rTmpPtr2], -32 // A12
      fms.s1             FR_r = FR_G, fSignifX, f1 // r = G * S_hi - 1
      sub                GR_N = GR_N, rExpHalf, 1 // unbisaed exponent of DeltaX
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
.pred.rel "mutex",p10,p11
{ .mfi
      ldfe               fA11 = [rPolDataPtr], -32 // A11
      // High part of log(|x|) = Y_hi = N * log2_hi + H
      fma.s1             fResH = fFloatN, FR_log2_hi, FR_H
(p10) cmp.eq             p8, p9 = rXRnd, r0
}
{ .mfi
      ldfe               fA10 = [rTmpPtr2], -32 // A10
      fma.s1             fRes6H = fA1, FR_FracX, f0 // (A1*y)hi
(p11) cmp.eq             p9, p8 = rXRnd, r0
}
;;
{ .mfi
      ldfe               fA9 = [rPolDataPtr], -32 // A9
      fma.s1             fB14 = fLnSin6, fDxSqr, f0 // (LnSin6*deltaX^2)hi
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
{ .mfi
      ldfe               fA8 = [rTmpPtr2], -32 // A8
      fma.s1             fA18 = fA19, FR_FracX, fA18
      nop.i              0
}
;;
{ .mfi
      ldfe               fA7 = [rPolDataPtr] // A7
      fma.s1             fA23 = fA23, FR_FracX, fA22
      nop.i              0
}
{ .mfi
      ldfe               fA6 = [rTmpPtr2] // A6
      fma.s1             fA21 = fA21, FR_FracX, fA20
      nop.i              0
}
;;
{ .mfi
      ldfe               fLnSin14 = [rLnSinDataPtr]
      fms.s1             fRes1L = fA3, FR_FracX, fRes1H // delta((A3*y)hi)
      extr.u             GR_Index2 = GR_X_1, 6, 4 // Extract bits 6-9 of X_1
}
{ .mfi
      setf.sig           fFloatNDx = GR_N
      fadd.s1            fPol = fRes1H, fA2 // (A3*y + A2)hi
      nop.i              0
}
;;
{ .mfi
      ldfps              FR_G, FR_H = [GR_ad_tbl_1], 8 // Load G_1, H_1
      fma.s1             fRes2H = fA4, fXSqr, f0 // ((A5 + A4*y)*y^2)hi
      nop.i              0
}
{ .mfi
      shladd             GR_ad_z_2 = GR_Index2, 2, rZ2Addr  // Point to Z_2
      fma.s1             fA25 = fA25, FR_FracX, fA24
      shladd             GR_ad_tbl_2 = GR_Index2, 4, rTbl2Addr  // Point to G_2
}
;;
.pred.rel "mutex",p8,p9
{ .mfi
      ld4                GR_Z_2 = [GR_ad_z_2] // Load Z_2
      fms.s1             fRes6L = fA1, FR_FracX, fRes6H // delta((A1*y)hi)
      // sign of GAMMA(x) is negative
(p8)  adds               rSgnGam = -1, r0
}
{ .mfi
      adds               rTmpPtr = 8, GR_ad_tbl_2
      fadd.s1            fRes3H = fRes6H, fA0 // (A1*y + A0)hi
      // sign of GAMMA(x) is positive
(p9)  adds               rSgnGam = 1, r0
}
;;
{ .mfi
      ldfps              FR_G2, FR_H2 = [GR_ad_tbl_2] // Load G_2, H_2
      // (LnSin6*deltaX^2 + LnSin4)hi
      fadd.s1            fLnSinH = fB14, fLnSin4
      nop.i              0
}
{ .mfi
      ldfd               FR_h2 = [rTmpPtr] // Load h_2
      fms.s1             fB16 = fLnSin6, fDxSqr, fB14 // delta(LnSin6*deltaX^2)
      nop.i              0
}
;;
{ .mfi
      ldfd               fhDelX = [GR_ad_tbl_1] // Load h_1
      fma.s1             fA21 = fA21, fXSqr, fA18
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36, fDelX4, fLnSin32
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes1L = fA3L, FR_FracX, fRes1L // (A3*y)lo
      // Get bits 30-15 of X_1 * Z_
      pmpyshr2.u         GR_X_2 = GR_X_1,GR_Z_2,15
}
{ .mfi
      nop.m              0
      fsub.s1            fPolL = fA2, fPol
      nop.i              0
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      nop.m              0
     // delta(((A5 + A4*y)*y^2)hi)
      fms.s1             fRes2L = fA4, fXSqr, fRes2H
      nop.i              0
}
{ .mfi
      nop.m              0
      // (((A5 + A4*y)*y^2) + A3*y + A2)hi
      fadd.s1            fRes4H = fRes2H, fPol
      nop.i              0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fma.s1             fRes6L = fA1L, FR_FracX, fRes6L // (A1*y)lo
      nop.i              0
}
{ .mfi
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fsub.s1            fRes3L = fA0, fRes3H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fLnSinL = fLnSin4, fLnSinH
      nop.i              0
}
{ .mfi
      nop.m              0
      // ((LnSin6*deltaX^2 + LnSin4)*deltaX^2)hi
      fma.s1             fB18 = fLnSinH, fDxSqr, f0
      nop.i              0
}
;;
{ .mfi
      adds               rTmpPtr = 8, rTbl3Addr
      fma.s1             fB16 = fLnSin6, fDxSqrL, fB16 // (LnSin6*deltaX^2)lo
      extr.u             GR_Index3 = GR_X_2, 1, 5 // Extract bits 1-5 of X_2
}
{ .mfi
      nop.m              0
      fma.s1             fA25 = fA25, fXSqr, fA23
      nop.i              0
}
;;
{ .mfi
      shladd             GR_ad_tbl_3 = GR_Index3, 4, rTbl3Addr // Point to G_3
      fadd.s1            fPolL = fPolL, fRes1H
      nop.i              0
}
{ .mfi
      shladd             rTmpPtr = GR_Index3, 4, rTmpPtr // Point to G_3
      fadd.s1            fRes1L = fRes1L, fA2L // (A3*y)lo + A2lo
      nop.i              0
}
;;
{ .mfi
      ldfps              FR_G3, FR_H3 = [GR_ad_tbl_3] // Load G_3, H_3
      fma.s1             fRes2L = fA4, fXSqrL, fRes2L // ((A5 + A4*y)*y^2)lo
      nop.i              0
}
{ .mfi
      ldfd               FR_h3 = [rTmpPtr] // Load h_3
      fsub.s1            fRes4L = fPol, fRes4H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // ((((A5 + A4*y)*y^2) + A3*y + A2)*y^2)hi
      fma.s1             fRes7H = fRes4H, fXSqr, f0
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA15 = fA15, FR_FracX, fA14
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fRes3L, fRes6H
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes6L = fRes6L, fA0L // (A1*y)lo + A0lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fLnSinL = fLnSinL, fB14

      nop.i              0
}
{ .mfi
      nop.m              0
      // delta((LnSin6*deltaX^2 + LnSin4)*deltaX^2)
      fms.s1             fB20 = fLnSinH, fDxSqr, fB18
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fPolL = fPolL, fRes1L // (A3*y + A2)lo

      nop.i              0
}
{ .mfi
      nop.m              0
      // ((LnSin6*deltaX^2 + LnSin4)*deltaX^2 + LnSin2)hi
      fadd.s1            fLnSin6 = fB18, fLnSin2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fRes2H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA17 = fA17, FR_FracX, fA16
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // delta(((((A5 + A4*y)*y^2) + A3*y + A2)*y^2)
      fms.s1             fRes7L = fRes4H, fXSqr, fRes7H
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fPol = fRes7H, fRes3H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fRes3L, fRes6L // (A1*y + A0)lo
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA25 = fA25, fX4, fA21
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (LnSin6*deltaX^2 + LnSin4)lo
      fadd.s1            fLnSinL = fLnSinL, fB16
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB20 = fLnSinH, fDxSqrL, fB20
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fLnSin4 = fLnSin2, fLnSin6
      nop.i              0
}
{ .mfi
      nop.m              0
      // (((LnSin6*deltaX^2 + LnSin4)*deltaX^2 + LnSin2)*DeltaX^2)hi
      fma.s1             fLnSinH = fLnSin6, fDxSqr, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // ((A5 + A4*y)*y^2)lo + (A3*y + A2)lo
      fadd.s1            fRes2L = fRes2L, fPolL
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA17 = fA17, fXSqr, fA15
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // ((((A5 + A4*y)*y^2) + A3*y + A2)*y^2)lo
      fma.s1             fRes7L = fRes4H, fXSqrL, fRes7L
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fPolL = fRes3H, fPol
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA13 = fA13, FR_FracX, fA12
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA11 = fA11, FR_FracX, fA10
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // ((LnSin6*deltaX^2 + LnSin4)*deltaX^2)lo
      fma.s1             fB20 = fLnSinL, fDxSqr, fB20
      nop.i              0
}
{ .mfi
      nop.m              0
      fmpy.s1            FR_G = FR_G, FR_G2 // G = G_1 * G_2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fLnSin4 = fLnSin4, fB18
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fLnSinL = fLnSin6, fDxSqr, fLnSinH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (((A5 + A4*y)*y^2) + A3*y + A2)lo
      fadd.s1            fRes4L = fRes4L, fRes2L
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fhDelX = fhDelX, FR_h2 // h = h_1 + h_2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes7L = fRes7L, fRes3L
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fPolL = fPolL, fRes7H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fcvt.xf            fFloatNDx = fFloatNDx
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            FR_H = FR_H, FR_H2 // H = H_1 + H_2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fmpy.s1            FR_G = FR_G, FR_G3 // G = (G_1 * G_2) * G_3
      nop.i              0
}
{ .mfi
      nop.m              0
      // ((LnSin6*deltaX^2 + LnSin4)*deltaX^2)lo + (LnSin2)lo
      fadd.s1            fLnSin2L = fLnSin2L, fB20
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA25 = fA25, fX4, fA17
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA13 = fA13, fXSqr, fA11
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA9 = fA9, FR_FracX, fA8
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA7 = fA7, FR_FracX, fA6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36, fDelX8, fLnSin28
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin14 = fLnSin14, fDxSqr, fLnSin12
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin10 = fLnSin10, fDxSqr, fLnSin8
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            FR_H = FR_H, FR_H3 // H = (H_1 + H_2) + H_3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRDx = FR_G, fNormDx, f1 // r = G * S_hi - 1
      nop.i              0
}
{ .mfi
      nop.m              0
      // poly_lo = r * Q4 + Q3
      fma.s1             FR_poly_lo = FR_r, FR_Q4, FR_Q3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fmpy.s1            FR_rsq = FR_r, FR_r // rsq = r * r
      nop.i              0
}
{ .mfi
      nop.m              0
      // ((((A5 + A4*y)*y^2) + A3*y + A2)*y^2)lo + (A1*y + A0)lo
      fma.s1             fRes7L = fRes4L, fXSqr, fRes7L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA25 = fA25, fX4, fA13
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA9 = fA9, fXSqr, fA7
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // h = N * log2_lo + h
      fma.s1             FR_h = fFloatN, FR_log2_lo, FR_h
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fhDelX = fhDelX, FR_h3 // h = (h_1 + h_2) + h_3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36, fDelX6, fLnSin20
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin14 = fLnSin14, fDelX4, fLnSin10
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = r * Q4 + Q3
      fma.s1             fPolyLoDx = fRDx, FR_Q4, FR_Q3
      nop.i              0
}
{ .mfi
      nop.m              0
      fmpy.s1            fRDxSq = fRDx, fRDx // rsq = r * r
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Y_hi = N * log2_hi + H
      fma.s1             fResLnDxH = fFloatNDx, FR_log2_hi, FR_H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             FR_rcub = FR_rsq, FR_r, f0 // rcub = r^3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA9 = fA25, fX4, fA9
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fPolL = fPolL, fRes7L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fLnSin4 = fLnSin4, fLnSin2L
      nop.i              0
}
{ .mfi
      nop.m              0
      // h = N * log2_lo + h
      fma.s1             fhDelX = fFloatNDx, FR_log2_lo, fhDelX
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36, fDelX8, fLnSin14
      nop.i              0
}
{ .mfi
      nop.m              0
      // ((LnSin6*deltaX^2 + LnSin4)*deltaX^2 + LnSin2)lo
      fma.s1             fLnSinL = fLnSin6, fDxSqrL, fLnSinL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo * r + Q2
      fma.s1             fPolyLoDx = fPolyLoDx, fRDx, FR_Q2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fRDxCub = fRDxSq, fRDx, f0 // rcub = r^3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      famax.s0              fRes5H = fPol, fResH
      nop.i              0
}
{ .mfi
      nop.m              0
      //  High part of (lgammal(|x|) + log(|x|))
      fadd.s1            fRes1H = fPol, fResH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo * r + Q2
      fma.s1             FR_poly_lo = FR_poly_lo, FR_r, FR_Q2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPolL = fA9, fX6, fPolL // P25lo
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      famin.s0              fRes5L = fPol, fResH
      nop.i              0
}
{ .mfi
      nop.m              0
      // High part of -(LnSin + log(|DeltaX|))
      fnma.s1            fRes2H = fResLnDxH, f1, fLnSinH
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      // (((LnSin6*deltaX^2 + LnSin4)*deltaX^2 + LnSin2)*DeltaX^2)lo
      fma.s1             fLnSinL = fLnSin4, fDxSqr, fLnSinL
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36, fDelX6, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             fPolyHiDx = FR_Q1, fRDxSq, fRDx
      nop.i              0
}
{ .mfi
      nop.m              0
      // poly_lo = poly_lo*r^3 + h
      fma.s1             fPolyLoDx = fPolyLoDx, fRDxCub, fhDelX
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fRes1L = fRes5H, fRes1H
      nop.i              0
}
{ .mfi
      nop.m              0
      //  -(lgammal(|x|) + log(|x|))hi
      fnma.s1            fRes1H = fRes1H, f1, f0

      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             FR_poly_hi = FR_Q1, FR_rsq, FR_r
      nop.i              0
}
{ .mfi
      nop.m              0
      // poly_lo = poly_lo*r^3 + h
      fma.s1             FR_poly_lo = FR_poly_lo, FR_rcub, FR_h
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes2L = fResLnDxH, fMOne, fRes2H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSinL = fLnSin36, fDxSqr, fLnSinL
      nop.i              0
}
{ .mfi
      nop.m              0
      // Y_lo = poly_hi + poly_lo
      fadd.s1            fResLnDxL = fPolyHiDx, fPolyLoDx
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fRes5L
      nop.i              0
}
{ .mfi
      nop.m              0
      // high part of the final result
      fadd.s1            fYH = fRes2H, fRes1H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Y_lo = poly_hi + poly_lo
      fadd.s1            fResL = FR_poly_hi, FR_poly_lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      famax.s0              fRes4H = fRes2H, fRes1H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      famin.s0              fRes4L = fRes2H, fRes1H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (LnSin)lo + (log(|DeltaX|))lo
      fsub.s1            fLnSinL = fLnSinL, fResLnDxL
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fLnSinH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      //(lgammal(|x|))lo + (log(|x|))lo
      fadd.s1            fPolL = fResL, fPolL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fYL = fRes4H, fYH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Low part of -(LnSin + log(|DeltaX|))
      fadd.s1            fRes2L = fRes2L, fLnSinL
      nop.i              0
}
{ .mfi
      nop.m              0
      //  High part of (lgammal(|x|) + log(|x|))
      fadd.s1            fRes1L = fRes1L, fPolL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fYL = fYL, fRes4L
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fRes2L = fRes2L, fRes1L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // low part of the final result
      fadd.s1            fYL = fYL, fRes2L
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      // final result for -6.0 < x <= -0.75, non-integer, "far" from roots
      fma.s0             f8 = fYH, f1, fYL
      // exit here for -6.0 < x <= -0.75, non-integer, "far" from roots
      br.ret.sptk        b0
}
;;

// here if |x+1| < 2^(-7)
.align 32
_closeToNegOne:
{ .mfi
      getf.exp           GR_N =  fDx // Get N = exponent of x
      fmerge.se          fAbsX =  f1, fDx // Form |deltaX|
      // Get high 4 bits of significand of deltaX
      extr.u             rIndex1Dx = rSignifDx, 59, 4
}
{ .mfi
      addl               rPolDataPtr= @ltoff(lgammal_1pEps_data),gp
      fma.s1             fA0L = fDxSqr, fDxSqr, f0 // deltaX^4
      // sign of GAMMA is positive if p10 is set to 1
(p10) adds               rSgnGam = 1, r0
}
;;
{ .mfi
      shladd             GR_ad_z_1 = rIndex1Dx, 2, GR_ad_z_1 // Point to Z_1
      fnma.s1            fResL = fDx, f1, f0 // -(x+1)
      // Get high 15 bits of significand
      extr.u             GR_X_0 = rSignifDx, 49, 15
}
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      shladd             GR_ad_tbl_1 = rIndex1Dx, 4, rTbl1Addr // Point to G_1
}
;;
{ .mfi
      ld4                GR_Z_1 = [GR_ad_z_1] // Load Z_1
      nop.f              0
      and                GR_N = GR_N, r17Ones // mask sign bit
}
{ .mfi
      adds               rTmpPtr = 8, GR_ad_tbl_1
      nop.f              0
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
;;
{ .mfi
      ldfps              FR_G, FR_H = [GR_ad_tbl_1],8 // Load G_1, H_1
      nop.f              0
      adds               rTmpPtr2 = 96, rPolDataPtr
}
{ .mfi
      ldfd               FR_h = [rTmpPtr] // Load h_1
      nop.f              0
      // unbiased exponent of deltaX
      sub                GR_N = GR_N, rExpHalf, 1
}
;;
{ .mfi
      adds               rTmpPtr3 = 192, rPolDataPtr
      nop.f              0
      // sign of GAMMA is negative if p11 is set to 1
(p11) adds               rSgnGam = -1, r0
}
{ .mfi
      ldfe               fA1 = [rPolDataPtr], 16 // A1
      nop.f              0
      nop.i              0
}
;;
{.mfi
      ldfe               fA2 = [rPolDataPtr], 16 // A2
      nop.f              0
      // Get bits 30-15 of X_0 * Z_1
      pmpyshr2.u         GR_X_1 = GR_X_0,GR_Z_1,15
}
{ .mfi
      ldfpd              fA20, fA19 = [rTmpPtr2], 16 // P8, P7
      nop.f              0
      nop.i              0
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      ldfe               fA3 = [rPolDataPtr], 16 // A3
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA18, fA17 = [rTmpPtr2], 16 // P6, P5
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfe               fA4 = [rPolDataPtr], 16 // A4
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA16, fA15 = [rTmpPtr2], 16 // P4, p3
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA5L, fA6 = [rPolDataPtr], 16 // A5, A6
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA14, fA13 = [rTmpPtr2], 16 // P2, P1
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA7, fA8 = [rPolDataPtr], 16 // A7, A8
      nop.f              0
      extr.u             GR_Index2 = GR_X_1, 6, 4 // Extract bits 6-9 of X_1
}
{ .mfi
      ldfe               fLnSin2 = [rTmpPtr2], 16
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      shladd             GR_ad_z_2 = GR_Index2, 2, rZ2Addr  // Point to Z_2
      nop.f              0
      shladd             GR_ad_tbl_2 = GR_Index2, 4, rTbl2Addr // Point to G_2
}
{ .mfi
      ldfe               fLnSin4 = [rTmpPtr2], 32
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ld4                GR_Z_2 = [GR_ad_z_2] // Load Z_2
      nop.f              0
      adds               rTmpPtr = 8, GR_ad_tbl_2
}
{ .mfi
      // Put integer N into rightmost significand
      setf.sig           fFloatN = GR_N
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfe               fLnSin6 = [rTmpPtr3]
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfe               fLnSin8 = [rTmpPtr2]
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfps              FR_G2, FR_H2 = [GR_ad_tbl_2],8 // Load G_2, H_2
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfd               FR_h2 = [rTmpPtr] // Load h_2
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fma.s1             fResH = fA20, fResL, fA19 //polynomial for log(|x|)
     // Get bits 30-15 of X_1 * Z_2
      pmpyshr2.u         GR_X_2 = GR_X_1,GR_Z_2,15
}
{ .mfi
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s1             fA2 = fA2, fDx, fA1 // polynomial for lgammal(|x|)
      nop.i              0
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      nop.m              0
      fma.s1             fA18 = fA18, fResL, fA17 //polynomial for log(|x|)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA16 = fA16, fResL, fA15 //polynomial for log(|x|)
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA4 = fA4, fDx, fA3 // polynomial for lgammal(|x|)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA14 = fA14, fResL, fA13 //polynomial for log(|x|)
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA6 = fA6, fDx, fA5L // polynomial for lgammal(|x|)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPol = fA8, fDx, fA7 // polynomial for lgammal(|x|)
      extr.u             GR_Index3 = GR_X_2, 1, 5 // Extract bits 1-5 of X_2
}
;;
{ .mfi
      shladd             GR_ad_tbl_3 = GR_Index3, 4, rTbl3Addr // Point to G_3
      // loqw part of lnsin polynomial
      fma.s1             fRes3L = fLnSin4, fDxSqr, fLnSin2
      nop.i              0
}
;;
{ .mfi
      ldfps              FR_G3, FR_H3 = [GR_ad_tbl_3], 8 // Load G_3, H_3
      fcvt.xf            fFloatN = fFloatN // N as FP number
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fResH = fResH, fDxSqr, fA18 // High part of log(|x|)
      nop.i              0
}
;;
{ .mfi
      ldfd               FR_h3 = [GR_ad_tbl_3] // Load h_3
      fma.s1             fA4 = fA4, fDxSqr, fA2 // Low part of lgammal(|x|)
      nop.i              0
}
{ .mfi
      nop.m              0
      // high part of lnsin polynomial
      fma.s1             fRes3H = fLnSin8, fDxSqr, fLnSin6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fmpy.s1            FR_G = FR_G, FR_G2 // G = G_1 * G_2
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            FR_H = FR_H, FR_H2 // H = H_1 + H_2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            FR_h = FR_h, FR_h2 // h = h_1 + h_2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA16 = fA16, fDxSqr, fA14 // Low part of log(|x|)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, fDxSqr, fA6 // High part of lgammal(|x|)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResH = fResH, fA0L, fA16 // log(|x|)/deltaX^2 - deltaX
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fmpy.s1            FR_G = FR_G, FR_G3 // G = (G_1 * G_2) * G_3
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            FR_H = FR_H, FR_H3 // H = (H_1 + H_2) + H_3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            FR_h = FR_h, FR_h3 // h = (h_1 + h_2) + h_3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResH = fResH, fDxSqr, fResL // log(|x|)
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, fA0L, fA4 // lgammal(|x|)/|x|
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             FR_r = FR_G, fAbsX, f1 // r = G * S_hi - 1
      nop.i              0
}
{ .mfi
      nop.m              0
      // high part of log(deltaX)= Y_hi = N * log2_hi + H
      fma.s1             fRes4H = fFloatN, FR_log2_hi, FR_H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // h = N * log2_lo + h
      fma.s1             FR_h = fFloatN, FR_log2_lo, FR_h
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResH = fPol, fDx, fResH // lgammal(|x|) + log(|x|)
      nop.i              0
}
{ .mfi
      nop.m              0
      // lnsin/deltaX^2
      fma.s1             fRes3H = fRes3H, fA0L, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = r * Q4 + Q3
      fma.s1             FR_poly_lo = FR_r, FR_Q4, FR_Q3
      nop.i              0
}
{ .mfi
      nop.m              0
      fmpy.s1            FR_rsq = FR_r, FR_r // rsq = r * r
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // lnSin - log(|x|) - lgammal(|x|)
      fms.s1             fResH = fRes3H, fDxSqr, fResH
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      // poly_lo = poly_lo * r + Q2
      fma.s1             FR_poly_lo = FR_poly_lo, FR_r, FR_Q2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             FR_rcub = FR_rsq, FR_r, f0 // rcub = r^3
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             FR_poly_hi = FR_Q1, FR_rsq, FR_r
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      // poly_lo = poly_lo*r^3 + h
      fma.s1             FR_poly_lo = FR_poly_lo, FR_rcub, FR_h
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      // low part of log(|deltaX|) = Y_lo = poly_hi + poly_lo
      fadd.s1            fRes4L = FR_poly_hi, FR_poly_lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fResH = fResH, fRes4L
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      // final result for |x+1|< 2^(-7) path
      fsub.s0            f8 = fResH, fRes4H
      // exit for |x+1|< 2^(-7) path
      br.ret.sptk        b0
}
;;


// here if -2^63 < x < -6.0 and x is not an integer
// Also we are going to filter out cases when x falls in
// range which is "close enough" to negative root. Rhis case
// may occur only for -19.5 < x since other roots of lgamma are
// insignificant from double extended point of view (they are closer
// to RTN(x) than one ulp(x).
.align 32
_negStirling:
{ .mfi
      ldfe               fLnSin6 = [rLnSinDataPtr], 32
      fnma.s1            fInvX = f8, fRcpX, f1 // start of 3rd NR iteration
      // Get high 4 bits of significand of deltaX
      extr.u             rIndex1Dx = rSignifDx, 59, 4
}
{ .mfi
      ldfe               fLnSin8 = [rTmpPtr3], 32
      fadd.s1            FR_h = FR_h, FR_h2 // h = h_1 + h_2
(p12) cmp.ltu.unc        p6, p0 = rSignifX, rLeftBound
}
;;
{ .mfi
      ldfe               fLnSin10 = [rLnSinDataPtr], 32
      fmpy.s1            FR_G = FR_G, FR_G3 // G = (G_1 * G_2) * G_3
      // Get high 15 bits of significand
      extr.u             GR_X_0 = rSignifDx, 49, 15
}
{ .mfi
      shladd             GR_ad_z_1 = rIndex1Dx, 2, GR_ad_z_1  // Point to Z_1
      fadd.s1            FR_H = FR_H, FR_H3 // H = (H_1 + H_2) + H_3
      // set p6 if x falls in "near root" range
(p6)  cmp.geu.unc        p6, p0 = rSignifX, rRightBound
}
;;
{ .mfi
      getf.exp           GR_N =  fDx // Get N = exponent of x
      fma.s1             fDx4 = fDxSqr, fDxSqr, f0 // deltaX^4
      adds               rTmpPtr = 96, rBernulliPtr
}
{ .mfb
      ld4                GR_Z_1 = [GR_ad_z_1] // Load Z_1
      fma.s1             fLnSin34 = fLnSin34, fDxSqr, fLnSin32
      // branch to special path if x falls in "near root" range
(p6)  br.cond.spnt       _negRoots
}
;;
.pred.rel "mutex",p10,p11
{ .mfi
      ldfe               fLnSin12 = [rTmpPtr3]
      fma.s1             fLnSin26 = fLnSin26, fDxSqr, fLnSin24
(p10) cmp.eq             p8, p9 = rXRnd, r0
}
{ .mfi
      ldfe               fLnSin14 = [rLnSinDataPtr]
      fma.s1             fLnSin30 = fLnSin30, fDxSqr, fLnSin28
(p11) cmp.eq             p9, p8 = rXRnd, r0
}
;;
{ .mfi
      ldfpd              fB2, fB2L = [rBernulliPtr], 16
      fma.s1             fLnSin18 = fLnSin18, fDxSqr, fLnSin16
      shladd             GR_ad_tbl_1 = rIndex1Dx, 4, rTbl1Addr // Point to G_1

}
{ .mfi
      ldfe               fB14 = [rTmpPtr], 16
      fma.s1             fLnSin22 = fLnSin22, fDxSqr, fLnSin20
      and                GR_N = GR_N, r17Ones // mask sign bit
}
;;
{ .mfi
      ldfe               fB4 = [rBernulliPtr], 16
      fma.s1             fInvX = fInvX, fRcpX, fRcpX // end of 3rd NR iteration
      // Get bits 30-15 of X_0 * Z_1
      pmpyshr2.u         GR_X_1 = GR_X_0,GR_Z_1,15
}
{ .mfi
      ldfe               fB16 = [rTmpPtr], 16
      fadd.s1            FR_h = FR_h, FR_h3 // h = (h_1 + h_2) + h_3
      adds               rTmpPtr2 = 8, GR_ad_tbl_1
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      ldfe               fB6 = [rBernulliPtr], 16
      fms.s1             FR_r = FR_G, fSignifX, f1 // r = G * S_hi - 1
      adds               rTmpPtr3 = -48, rTmpPtr
}
{ .mfi
      ldfe               fB18 = [rTmpPtr], 16
      // High part of the log(|x|) = Y_hi = N * log2_hi + H
      fma.s1             fResH = fFloatN, FR_log2_hi, FR_H
      sub                GR_N = GR_N, rExpHalf, 1 // unbiased exponent of deltaX
}
;;
.pred.rel "mutex",p8,p9
{ .mfi
      ldfe               fB8 = [rBernulliPtr], 16
      fma.s1             fLnSin36 = fLnSin36, fDx4, fLnSin34
      // sign of GAMMA(x) is negative
(p8)  adds               rSgnGam = -1, r0
}
{ .mfi
      ldfe               fB20 = [rTmpPtr], -160
      fma.s1             fRes5H = fLnSin4, fDxSqr, f0
      // sign of GAMMA(x) is positive
(p9)  adds               rSgnGam = 1, r0

}
;;
{ .mfi
      ldfe               fB10 = [rBernulliPtr], 16
      fma.s1             fLnSin30 = fLnSin30, fDx4, fLnSin26
(p14) adds               rTmpPtr = -160, rTmpPtr
}
{ .mfi
      ldfe               fB12 = [rTmpPtr3], 16
      fma.s1             fDx8 = fDx4, fDx4, f0 // deltaX^8
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
;;
{ .mfi
      ldfps              fGDx, fHDx = [GR_ad_tbl_1], 8 // Load G_1, H_1
      fma.s1             fDx6 = fDx4, fDxSqr, f0 // deltaX^6
      extr.u             GR_Index2 = GR_X_1, 6, 4 // Extract bits 6-9 of X_1
}
{ .mfi
      ldfd               fhDx = [rTmpPtr2] // Load h_1
      fma.s1             fLnSin22 = fLnSin22, fDx4, fLnSin18
      nop.i              0
}
;;
{ .mfi
      // Load two parts of C
      ldfpd              fRes1H, fRes1L = [rTmpPtr], 16
      fma.s1             fRcpX = fInvX, fInvX, f0  // (1/x)^2
      shladd             GR_ad_tbl_2 = GR_Index2, 4, rTbl2Addr // Point to G_2
}
{ .mfi
      shladd             GR_ad_z_2 = GR_Index2, 2, rZ2Addr  // Point to Z_2
      fma.s1             FR_h = fFloatN, FR_log2_lo, FR_h// h = N * log2_lo + h
      nop.i              0
}
;;
{ .mfi
      ld4                GR_Z_2 = [GR_ad_z_2] // Load Z_2
      fnma.s1            fInvXL = f8, fInvX, f1 // relative error of 1/x
      nop.i              0
}
{ .mfi
      adds               rTmpPtr2 = 8, GR_ad_tbl_2
      fma.s1             fLnSin8 = fLnSin8, fDxSqr, fLnSin6
      nop.i              0
}
;;
{ .mfi
      ldfps              FR_G2, FR_H2 = [GR_ad_tbl_2],8   // Load G_2, H_2
      // poly_lo = r * Q4 + Q3
      fma.s1             FR_poly_lo = FR_r, FR_Q4, FR_Q3
      nop.i              0
}
{ .mfi
      ldfd               fh2Dx = [rTmpPtr2] // Load h_2
      fmpy.s1            FR_rsq = FR_r, FR_r // rsq = r * r
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA1L = fB2, fInvX, f0 // (B2*(1/x))hi
      nop.i              0
}
{ .mfi
      // Put integer N into rightmost significand
      setf.sig           fFloatNDx = GR_N
      fms.s1             fRes4H = fResH, f1, f1  // ln(|x|)hi - 1
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2H = fRes5H, fLnSin2//(lnSin4*DeltaX^2 + lnSin2)hi
      // Get bits 30-15 of X_1 * Z_2
      pmpyshr2.u         GR_X_2 = GR_X_1,GR_Z_2,15
}
{ .mfi
      nop.m              0
      fms.s1             fRes5L = fLnSin4, fDxSqr, fRes5H
      nop.i              0
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      nop.m              0
      fma.s1             fInvX4 = fRcpX, fRcpX, f0 // (1/x)^4
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB6 = fB6, fRcpX, fB4
      nop.i              0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fma.s1             fB18 = fB18, fRcpX, fB16
      nop.i              0
}
{ .mfi
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s1             fInvXL = fInvXL, fInvX, f0 // low part of 1/x
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo * r + Q2
      fma.s1             FR_poly_lo = FR_poly_lo, FR_r, FR_Q2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             FR_rcub = FR_rsq, FR_r, f0 // rcub = r^3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes3H = fRes4H, f8, f0 // (-|x|*(ln(|x|)-1))hi
      extr.u             GR_Index3 = GR_X_2, 1, 5  // Extract bits 1-5 of X_2
}
{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             FR_poly_hi = FR_Q1, FR_rsq, FR_r
      nop.i              0
}
;;
{ .mfi
      shladd             GR_ad_tbl_3 = GR_Index3, 4, rTbl3Addr // Point to G_3
      fms.s1             fA2L = fB2, fInvX, fA1L // delta(B2*(1/x))
      nop.i              0
}
{ .mfi
      nop.m              0
      fnma.s1            fBrnH = fRes1H, f1, fA1L // (-C - S(1/x))hi
      nop.i              0
}
;;
{ .mfi
      ldfps              fG3Dx, fH3Dx = [GR_ad_tbl_3],8 // Load G_3, H_3
      fma.s1             fInvX8 = fInvX4, fInvX4, f0 // (1/x)^8
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB10 = fB10, fRcpX, fB8
      nop.i              0
}
;;

{ .mfi
      ldfd               fh3Dx = [GR_ad_tbl_3] // Load h_3
      fma.s1             fB20 = fB20, fInvX4, fB18
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB14 = fB14, fRcpX, fB12
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36, fDx8, fLnSin30
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin12 = fLnSin12, fDxSqr, fLnSin10
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fRes2L = fLnSin2, fRes2H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fRes2H, fDxSqr, f0 // high part of LnSin
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fnma.s1            fResH = fResH, FR_MHalf, fResH // -0.5*ln(|x|)hi
      nop.i              0
}
{ .mfi
      nop.m              0
      fmpy.s1            fGDx = fGDx, FR_G2 // G = G_1 * G_2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo*r^3 + h
      fma.s1             FR_poly_lo = FR_poly_lo, FR_rcub, FR_h
      nop.i              0
}
{ .mfi
      nop.m              0
      // B2lo*(1/x)hi+ delta(B2*(1/x))
      fma.s1             fA2L = fB2L, fInvX, fA2L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fB20 = fB20, fInvX4, fB14
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB10 = fB10, fInvX4, fB6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fcvt.xf            fFloatNDx = fFloatNDx
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin14 = fLnSin14, fDx4, fLnSin12
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36, fDx8, fLnSin22
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fRes3L = fRes4H, f8, fRes3H // delta(-|x|*(ln(|x|)-1))
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fmpy.s1            fGDx = fGDx, fG3Dx // G = (G_1 * G_2) * G_3
      nop.i              0
}
{ .mfi
      nop.m              0
      // (-|x|*(ln(|x|)-1) - 0.5ln(|x|))hi
      fadd.s1            fRes4H = fRes3H, fResH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA2L = fInvXL, fB2, fA2L //(B2*(1/x))lo
      nop.i              0
}
{ .mfi
      nop.m              0
      // low part of log(|x|) = Y_lo = poly_hi + poly_lo
      fadd.s1            fResL = FR_poly_hi, FR_poly_lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fB20 = fB20, fInvX8, fB10
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fInvX3 = fInvX, fRcpX, f0 // (1/x)^3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fHDx = fHDx, FR_H2 // H = H_1 + H_2
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes5L = fRes5L, fLnSin2L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fRes5H
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fhDx = fhDx, fh2Dx // h = h_1 + h_2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fBrnL = fRes1H, fMOne, fBrnH
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             FR_r = fGDx, fNormDx, f1 // r = G * S_hi - 1
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes3L = fResL, f8 , fRes3L // (-|x|*(ln(|x|)-1))lo
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fRes4L = fRes3H, fRes4H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // low part of "Bernulli" polynomial
      fma.s1             fB20 = fB20, fInvX3, fA2L
      nop.i              0
}
{ .mfi
      nop.m              0
      fnma.s1            fResL = fResL, FR_MHalf, fResL // -0.5*ln(|x|)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fHDx = fHDx, fH3Dx // H = (H_1 + H_2) + H_3
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fPolL = fRes2H, fDxSqr, fPol
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fhDx = fhDx, fh3Dx // h = (h_1 + h_2) + h_3
      nop.i              0
}
{ .mfi
      nop.m              0
      // (-|x|*(ln(|x|)-1) - 0.5ln(|x|) - C - S(1/x))hi
      fadd.s1            fB14 = fRes4H, fBrnH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = r * Q4 + Q3
      fma.s1             FR_poly_lo = FR_r, FR_Q4, FR_Q3
      nop.i              0
}
{ .mfi
      nop.m              0
      fmpy.s1            FR_rsq = FR_r, FR_r // rsq = r * r
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fResH
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fBrnL = fBrnL, fA1L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (-|x|*(ln(|x|)-1))lo + (-0.5ln(|x|))lo
      fadd.s1            fRes3L = fRes3L, fResL
      nop.i              0
}
{ .mfi
      nop.m              0
      fnma.s1            fB20 = fRes1L, f1, fB20 // -Clo - S(1/x)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fRes5L // (lnSin4*DeltaX^2 + lnSin2)lo
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPolL = fDxSqrL, fRes2H, fPolL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin14 = fLnSin14, fDx4, fLnSin8
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin36,  fDx8, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo * r + Q2
      fma.s1             FR_poly_lo = FR_poly_lo, FR_r, FR_Q2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             FR_rcub = FR_rsq, FR_r, f0 // rcub = r^3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             FR_poly_hi = FR_Q1, FR_rsq, FR_r
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fB12 = fRes4H, fB14
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (-|x|*(ln(|x|)-1) - 0.5ln(|x|))lo
      fadd.s1            fRes4L = fRes4L, fRes3L
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fBrnL = fBrnL, fB20 // (-C - S(1/x))lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // high part of log(|DeltaX|) = Y_hi = N * log2_hi + H
      fma.s1             fLnDeltaH = fFloatNDx, FR_log2_hi, fHDx
      nop.i              0
}
{ .mfi
      nop.m              0
      // h = N * log2_lo + h
      fma.s1             fhDx = fFloatNDx, FR_log2_lo, fhDx
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPolL = fRes2L, fDxSqr, fPolL
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin14 = fLnSin36,  fDxSqr, fLnSin14
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (-|x|*(ln(|x|)-1) - 0.5ln(|x|))lo + (- C - S(1/x))lo
      fadd.s1            fBrnL = fBrnL, fRes4L
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fB12 = fB12, fBrnH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo*r^3 + h
      fma.s1             FR_poly_lo = FR_poly_lo, FR_rcub, fhDx
      nop.i              0
}
{ .mfi
      nop.m              0
      fnma.s1            fRes1H = fLnDeltaH, f1, fPol//(-ln(|DeltaX|) + LnSin)hi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPolL = fDxSqrL, fRes2L, fPolL
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin36 = fLnSin14, fDx6, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (-|x|*(ln(|x|)-1) - 0.5ln(|x|) - C - S(1/x))lo
      fadd.s1            fB12 = fB12, fBrnL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // low part of log(|DeltaX|) =  Y_lo = poly_hi + poly_lo
      fadd.s1            fLnDeltaL= FR_poly_hi, FR_poly_lo
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fRes1L = fLnDeltaH, fMOne, fRes1H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fPolL = fPolL, fLnSin36
      nop.i              0
}
{ .mfi
      nop.m              0
      //(-|x|*(ln(|x|)-1)-0.5ln(|x|) - C - S(1/x))hi + (-ln(|DeltaX|) + LnSin)hi
      fadd.s1            f8 = fRes1H, fB14
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      //max((-|x|*(ln(|x|)-1)-0.5ln(|x|) - C - S(1/x))hi,
      //    (-ln(|DeltaX|) + LnSin)hi)
      famax.s1           fMaxNegStir = fRes1H, fB14
      nop.i              0
}
{ .mfi
      nop.m              0
      //min((-|x|*(ln(|x|)-1)-0.5ln(|x|) - C - S(1/x))hi,
      //    (-ln(|DeltaX|) + LnSin)hi)
      famin.s1           fMinNegStir = fRes1H, fB14
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fPol
      nop.i              0
}
{ .mfi
      nop.m              0
      // (-ln(|DeltaX|))lo + (LnSin)lo
      fnma.s1            fPolL = fLnDeltaL, f1, fPolL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            f9 = fMaxNegStir, f8 // delta1
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fPolL // (-ln(|DeltaX|) + LnSin)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            f9 = f9, fMinNegStir
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fB12
      nop.i              0
}
;;
{ .mfi
      // low part of the result
      fadd.s1            f9 = f9, fRes1L
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      // final result for -2^63 < x < -6.0 path
      fma.s0             f8 = f8, f1, f9
      // exit here  for -2^63 < x < -6.0 path
      br.ret.sptk        b0
}
;;

// here if x falls in neighbourhood of any negative root
// "neighbourhood" typically means that |lgammal(x)| < 0.17
// on the [-3.0,-2.0] range |lgammal(x)| has even less
// magnitude
// rXint contains index of the root
// p10 is set if root belongs to "right" ones
// p11 is set if root belongs to "left" ones
// lgammal(x) is approximated by polynomial of
// 19th degree from (x - root) argument
.align 32
_negRoots:
{ .mfi
      addl          rPolDataPtr= @ltoff(lgammal_right_roots_polynomial_data),gp
      nop.f              0
      shl                rTmpPtr2 = rXint, 7 // (i*16)*8
}
{ .mfi
      adds               rRootsAddr = -288, rRootsBndAddr
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfe               fRoot = [rRootsAddr] // FP representation of root
      nop.f              0
      shl                rTmpPtr = rXint, 6  // (i*16)*4
}
{ .mfi
(p11) adds               rTmpPtr2 = 3536, rTmpPtr2
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      shladd             rTmpPtr = rXint, 4, rTmpPtr // (i*16) + (i*16)*4
}
{ .mfi
      adds               rTmpPtr3 = 32, rTmpPtr2
      nop.f              0
      nop.i              0
}
;;
.pred.rel "mutex",p10,p11
{ .mfi
      add                rTmpPtr3 = rTmpPtr, rTmpPtr3
      nop.f              0
(p10) cmp.eq             p8, p9 = rXRnd, r0
}
{ .mfi
      // (i*16) + (i*16)*4 + (i*16)*8
      add                rTmpPtr = rTmpPtr, rTmpPtr2
      nop.f              0
(p11) cmp.eq             p9, p8 = rXRnd, r0
}
;;
{ .mfi
      add                rTmpPtr2 = rPolDataPtr, rTmpPtr3
      nop.f              0
      nop.i              0
}
{ .mfi
      add                rPolDataPtr = rPolDataPtr, rTmpPtr // begin + offsett
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA0, fA0L = [rPolDataPtr], 16 // A0
      nop.f              0
      adds               rTmpPtr = 112, rTmpPtr2
}
{ .mfi
      ldfpd              fA2, fA2L = [rTmpPtr2], 16 // A2
      nop.f              0
      cmp.eq             p12, p13 = 4, rSgnGamSize
}
;;
{ .mfi
      ldfpd              fA1, fA1L = [rPolDataPtr], 16 // A1
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfe               fA3 = [rTmpPtr2], 128 // A4
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA12, fA13 = [rTmpPtr], 16 // A12, A13
      nop.f              0
      adds               rTmpPtr3 = 64, rPolDataPtr
}
{ .mfi
      ldfpd              fA16, fA17 = [rTmpPtr2], 16 // A16, A17
      nop.f              0
      adds               rPolDataPtr = 32, rPolDataPtr
}
;;
.pred.rel "mutex",p8,p9
{ .mfi
      ldfpd              fA14, fA15 = [rTmpPtr], 16 // A14, A15
      nop.f              0
      // sign of GAMMA(x) is negative
(p8)  adds               rSgnGam = -1, r0
}
{ .mfi
      ldfpd              fA18, fA19 = [rTmpPtr2], 16 // A18, A19
      nop.f              0
      // sign of GAMMA(x) is positive
(p9)  adds               rSgnGam = 1, r0
}
;;
{ .mfi
      ldfe               fA4 = [rPolDataPtr], 16 // A4
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA6, fA7 = [rTmpPtr3], 16 // A6, A7
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfe               fA5 = [rPolDataPtr], 16 // A5
      // if x equals to (rounded) root exactly
      fcmp.eq.s1         p6, p0 = f8, fRoot
      nop.i              0
}
{ .mfi
      ldfpd              fA8, fA9 = [rTmpPtr3], 16 // A8, A9
      fms.s1             FR_FracX = f8, f1, fRoot
      nop.i              0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p12) st4                [rSgnGamAddr] = rSgnGam
      nop.f              0
      nop.i              0
}
{ .mfb
      // store signgam if size of variable is 8 bytes
(p13) st8                [rSgnGamAddr] = rSgnGam
      // answer if x equals to (rounded) root exactly
(p6)  fadd.s0            f8 = fA0, fA0L
      // exit if x equals to (rounded) root exactly
(p6)  br.ret.spnt        b0
}
;;
{ .mmf
      ldfpd              fA10, fA11 = [rTmpPtr3], 16 // A10, A11
      nop.m              0
      nop.f              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResH = fA2, FR_FracX, f0 // (A2*x)hi
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA4L = FR_FracX, FR_FracX, f0 // x^2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA17 = fA17, FR_FracX, fA16
      nop.i              0
}
{.mfi
      nop.m              0
      fma.s1             fA13 = fA13, FR_FracX, fA12
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA19 = fA19, FR_FracX, fA18
      nop.i              0
}
{.mfi
      nop.m              0
      fma.s1             fA15 = fA15, FR_FracX, fA14
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fPol = fA7, FR_FracX, fA6
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fA9 = fA9, FR_FracX, fA8
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fResL = fA2, FR_FracX, fResH // delta(A2*x)
      nop.i              0
}
{.mfi
      nop.m              0
      fadd.s1            fRes1H = fResH, fA1 // (A2*x + A1)hi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA11 = fA11, FR_FracX, fA10
      nop.i              0
}
{.mfi
      nop.m              0
      fma.s1             fA5L = fA4L, fA4L, f0 // x^4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA19 = fA19, fA4L, fA17
      nop.i              0
}
{.mfi
      nop.m              0
      fma.s1             fA15 = fA15, fA4L, fA13
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, FR_FracX, fA5
      nop.i              0
}
{.mfi
      nop.m              0
      fma.s1             fA3L = fA4L, FR_FracX, f0 // x^3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // delta(A2*x) + A2L*x = (A2*x)lo
      fma.s1             fResL = fA2L, FR_FracX, fResL
      nop.i              0
}
{.mfi
      nop.m              0
      fsub.s1            fRes1L = fA1, fRes1H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA11 = fA11, fA4L, fA9
      nop.i              0
}
{.mfi
      nop.m              0
      fma.s1             fA19 = fA19, fA5L, fA15
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fPol = fPol, FR_FracX, fA4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fResL = fResL, fA1L // (A2*x)lo + A1
      nop.i              0
}
{.mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fResH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes2H = fRes1H, FR_FracX, f0 // ((A2*x + A1)*x)hi
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fA19 = fA19, fA5L, fA11
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fPol = fPol, FR_FracX, fA3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fResL // (A2*x + A1)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // delta((A2*x + A1)*x)
      fms.s1             fRes2L = fRes1H, FR_FracX, fRes2H
      nop.i              0
}
{.mfi
      nop.m              0
      fadd.s1            fRes3H = fRes2H, fA0 // ((A2*x + A1)*x + A0)hi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA19 = fA19, fA5L, f0
      nop.i              0
}

;;
{ .mfi
      nop.m              0
      fma.s1             fRes2L = fRes1L, FR_FracX, fRes2L // ((A2*x + A1)*x)lo
      nop.i              0
}
{.mfi
      nop.m              0
      fsub.s1            fRes3L = fRes2H, fRes3H
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fPol = fA19, FR_FracX, fPol
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fRes3L, fA0
      nop.i              0
}
{.mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fA0L // ((A2*x + A1)*x)lo + A0L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fRes3L, fRes2L // (((A2*x + A1)*x) + A0)lo
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fRes3L = fPol, fA3L, fRes3L
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      // final result for arguments which are close to negative roots
      fma.s0             f8 = fRes3H, f1, fRes3L
      // exit here for arguments which are close to negative roots
      br.ret.sptk        b0
}
;;

// here if |x| < 0.5
.align 32
lgammal_0_half:
{ .mfi
      ld4                GR_Z_1 = [rZ1offsett] // Load Z_1
      fma.s1             fA4L = f8, f8, f0 // x^2
      addl               rPolDataPtr    = @ltoff(lgammal_0_Half_data), gp
}
{ .mfi
      shladd             GR_ad_tbl_1 = GR_Index1, 4, rTbl1Addr// Point to G_1
      nop.f              0
      addl               rLnSinDataPtr    = @ltoff(lgammal_lnsin_data), gp
}
;;
{ .mfi
      ldfps              FR_G, FR_H = [GR_ad_tbl_1],8 // Load G_1, H_1
      nop.f              0
      // Point to Constants_Z_2
      add                GR_ad_z_2 = 0x140, GR_ad_z_1
}
{ .mfi
      add                GR_ad_q = -0x60, GR_ad_z_1 // Point to Constants_Q
      nop.f              0
      // Point to Constants_G_H_h2
      add                GR_ad_tbl_2 = 0x180, GR_ad_z_1
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      // Point to Constants_G_H_h3
      add                GR_ad_tbl_3 = 0x280, GR_ad_z_1
}
{ .mfi
      ldfd               FR_h = [GR_ad_tbl_1] // Load h_1
      nop.f              0
      sub                GR_N = rExpX, rExpHalf, 1
}
;;
{ .mfi
      ld8                rLnSinDataPtr    = [rLnSinDataPtr]
      nop.f              0
      // Get bits 30-15 of X_0 * Z_1
      pmpyshr2.u         GR_X_1 = GR_X_0,GR_Z_1,15
}
{ .mfi
      ldfe               FR_log2_hi = [GR_ad_q],16 // Load log2_hi
      nop.f              0
      sub                GR_N = r0, GR_N
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      ldfe               FR_log2_lo = [GR_ad_q], 16 // Load log2_lo
      nop.f              0
      add                rTmpPtr2 = 320, rPolDataPtr
}
{ .mfi
      add                rTmpPtr = 32, rPolDataPtr
      nop.f              0
      // exponent of 0.25
      adds               rExp2 = -1, rExpHalf
}
;;
{ .mfi
      ldfpd              fA3, fA3L = [rPolDataPtr], 16 // A3
      fma.s1             fA5L = fA4L, fA4L, f0 // x^4
      nop.i              0
}
{ .mfi
      ldfpd              fA1, fA1L = [rTmpPtr], 16 // A1
      fms.s1             fB8 = f8, f8, fA4L // x^2 - <x^2>
      // set p6 if -0.5 < x <= -0.25
(p15) cmp.eq.unc         p6, p0 = rExpX, rExp2
}
;;
{ .mfi
      ldfpd              fA2, fA2L = [rPolDataPtr], 16 // A2
      nop.f              0
      // set p6 if -0.5 < x <= -0.40625
(p6)  cmp.le.unc         p6, p0 = 10, GR_Index1
}
{ .mfi
      ldfe               fA21 = [rTmpPtr2], -16 // A21
      // Put integer N into rightmost significand
      nop.f              0
      adds               rTmpPtr = 240, rTmpPtr
}
;;
{ .mfi
      setf.sig           fFloatN = GR_N
      nop.f              0
      extr.u             GR_Index2 = GR_X_1, 6, 4 // Extract bits 6-9 of X_1
}
{ .mfi
      ldfe               FR_Q4 = [GR_ad_q], 16 // Load Q4
      nop.f              0
      adds               rPolDataPtr = 304, rPolDataPtr
}
;;
{ .mfi
      ldfe               fA20 = [rTmpPtr2], -32 // A20
      nop.f              0
      shladd             GR_ad_z_2 = GR_Index2, 2, GR_ad_z_2  // Point to Z_2
}
{ .mfi
      ldfe               fA19 = [rTmpPtr], -32 // A19
      nop.f              0
      shladd             GR_ad_tbl_2 = GR_Index2, 4, GR_ad_tbl_2// Point to G_2
}
;;
{ .mfi
      ldfe               fA17 = [rTmpPtr], -32 // A17
      nop.f              0
      adds               rTmpPtr3 = 8, GR_ad_tbl_2
}
{ .mfb
      ldfe               fA18 = [rTmpPtr2], -32 // A18
      nop.f              0
      // branch to special path for -0.5 < x <= 0.40625
(p6)  br.cond.spnt       lgammal_near_neg_half
}
;;
{ .mmf
      ld4                GR_Z_2 = [GR_ad_z_2] // Load Z_2
      ldfe               fA15 = [rTmpPtr], -32 // A15
      fma.s1             fB20 = fA5L, fA5L, f0 // x^8
}
;;
{ .mmf
      ldfe               fA16 = [rTmpPtr2], -32 // A16
      ldfe               fA13 = [rTmpPtr], -32 // A13
      fms.s1             fB16 = fA4L, fA4L, fA5L
}
;;
{ .mmf
      ldfps              FR_G2, FR_H2 = [GR_ad_tbl_2], 8 // Load G_2, H_2
      ldfd               FR_h2 = [rTmpPtr3] // Load h_2
      fmerge.s           fB10 = f8, fA5L // sign(x) * x^4
}
;;
{ .mmi
      ldfe               fA14 = [rTmpPtr2], -32 // A14
      ldfe               fA11 = [rTmpPtr], -32 // A11
      // Get bits 30-15 of X_1 * Z_2
      pmpyshr2.u         GR_X_2 = GR_X_1,GR_Z_2,15
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      ldfe               fA12 = [rTmpPtr2], -32 // A12
      fma.s1             fRes4H = fA3, fAbsX, f0
      adds               rTmpPtr3 = 16, GR_ad_q
}
{ .mfi
      ldfe               fA9 = [rTmpPtr], -32 // A9
      nop.f              0
      nop.i              0
}
;;
{ .mmf
      ldfe               fA10 = [rTmpPtr2], -32 // A10
      ldfe               fA7 = [rTmpPtr], -32 // A7
      fma.s1             fB18 = fB20, fB20, f0 // x^16
}
;;
{ .mmf
      ldfe               fA8 = [rTmpPtr2], -32 // A8
      ldfe               fA22 = [rPolDataPtr], 16 // A22
      fcvt.xf            fFloatN = fFloatN
}
;;
{ .mfi
      ldfe               fA5 = [rTmpPtr], -32 // A5
      fma.s1             fA21 = fA21, fAbsX, fA20 // v16
      extr.u             GR_Index3 = GR_X_2, 1, 5 // Extract bits 1-5 of X_2
}
{ .mfi
      ldfe               fA6 = [rTmpPtr2], -32 // A6
      nop.f              0
      nop.i              0
}
;;
{ .mmf
      // Point to G_3
      shladd             GR_ad_tbl_3 = GR_Index3, 4, GR_ad_tbl_3
      ldfe               fA4 = [rTmpPtr2], -32 // A4
      fma.s1             fA19 = fA19, fAbsX, fA18 // v13
}
;;
.pred.rel "mutex",p14,p15
{ .mfi
      ldfps              FR_G3, FR_H3 = [GR_ad_tbl_3],8 // Load G_3, H_3
      fms.s1             fRes4L = fA3, fAbsX, fRes4H
(p14) adds               rSgnGam = 1, r0
}
{ .mfi
      cmp.eq             p6, p7 = 4, rSgnGamSize
      fadd.s1            fRes2H = fRes4H, fA2
(p15) adds               rSgnGam = -1, r0
}
;;

{ .mfi
      ldfd               FR_h3 = [GR_ad_tbl_3] // Load h_3
      fma.s1             fA17 = fA17, fAbsX, fA16 // v12
      nop.i              0
}
;;
{ .mfi
      ldfe               FR_Q3 = [GR_ad_q], 32 // Load Q3
      fmpy.s1            FR_G = FR_G, FR_G2 // G = G_1 * G_2
      nop.i              0
}
{ .mfi
      ldfe               FR_Q2 = [rTmpPtr3], 16 // Load Q2
      fadd.s1            FR_H = FR_H, FR_H2 // H = H_1 + H_2
      nop.i              0
}
;;
{ .mfi
      ldfe               FR_Q1 = [GR_ad_q] // Load Q1
      fma.s1             fA15 = fA15, fAbsX, fA14 // v8
      nop.i              0
}
{ .mfi
      adds               rTmpPtr3 = 32, rLnSinDataPtr
      fadd.s1            FR_h = FR_h, FR_h2 // h = h_1 + h_2
      nop.i              0
}
;;
{ .mmf
      ldfpd              fLnSin2, fLnSin2L = [rLnSinDataPtr], 16
      ldfe               fLnSin6 = [rTmpPtr3], 32
      fma.s1             fA13 = fA13, fAbsX, fA12 // v7

}
;;
{ .mfi
      ldfe               fLnSin4 = [rLnSinDataPtr], 32
      fma.s1             fRes4L = fA3L, fAbsX, fRes4L
      nop.i              0
}
{ .mfi
      ldfe               fLnSin10 = [rTmpPtr3], 32
      fsub.s1            fRes2L = fA2, fRes2H
      nop.i              0
}
;;
{ .mfi
      ldfe               fLnSin8 = [rLnSinDataPtr], 32
      fma.s1             fResH = fRes2H, fAbsX, f0
      nop.i              0
}
{ .mfi
      ldfe               fLnSin14 = [rTmpPtr3], 32
      fma.s1             fA22 = fA22, fA4L, fA21 // v15
      nop.i              0
}
;;
{ .mfi
      ldfe               fLnSin12 = [rLnSinDataPtr], 32
      fma.s1             fA9 = fA9, fAbsX, fA8 // v4
      nop.i              0
}
{ .mfi
      ldfd               fLnSin18 = [rTmpPtr3], 16
      fma.s1             fA11 = fA11, fAbsX, fA10 // v5
      nop.i              0
}
;;
{ .mfi
      ldfe               fLnSin16 = [rLnSinDataPtr], 24
      fma.s1             fA19 = fA19, fA4L, fA17 // v11
      nop.i              0
}
{ .mfi
      ldfd               fLnSin22 = [rTmpPtr3], 16
      fma.s1             fPolL = fA7, fAbsX, fA6
      nop.i              0
}
;;
{ .mfi
      ldfd               fLnSin20 = [rLnSinDataPtr], 16
      fmpy.s1            FR_G = FR_G, FR_G3 // G = (G_1 * G_2) * G_3
      nop.i              0
}
{ .mfi
      ldfd               fLnSin26 = [rTmpPtr3], 16
      fadd.s1            FR_H = FR_H, FR_H3 // H = (H_1 + H_2) + H_3
      nop.i              0
}
;;
{ .mfi
      ldfd               fLnSin24 = [rLnSinDataPtr], 16
      fadd.s1            fRes2L = fRes2L, fRes4H
      nop.i              0
}
{ .mfi
      ldfd               fLnSin30 = [rTmpPtr3], 16
      fadd.s1            fA2L = fA2L, fRes4L
      nop.i              0
}
;;
{ .mfi
      ldfd               fLnSin28 = [rLnSinDataPtr], 16
      fms.s1             fResL = fRes2H, fAbsX, fResH
      nop.i              0
}
{ .mfi
      ldfd               fLnSin34 = [rTmpPtr3], 8
      fadd.s1            fRes2H = fResH, fA1
      nop.i              0
}
;;
{ .mfi
      ldfd               fLnSin32 = [rLnSinDataPtr]
      fma.s1             fA11 = fA11, fA4L, fA9 // v3
      nop.i              0
}
{ .mfi
      ldfd               fLnSin36 = [rTmpPtr3]
      fma.s1             fA15 = fA15, fA4L, fA13 // v6
      nop.i              0
}
;;

{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fadd.s1            FR_h = FR_h, FR_h3 // h = (h_1 + h_2) + h_3
      nop.i              0
}
{ .mfi
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s1             fA5 = fA5, fAbsX, fA4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             FR_r = FR_G, fSignifX, f1 // r = G * S_hi - 1
      nop.i              0
}
{ .mfi
      nop.m               0
      // High part of the log(|x|): Y_hi = N * log2_hi + H
      fms.s1             FR_log2_hi = fFloatN, FR_log2_hi, FR_H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fA3L = fRes2L, fA2L
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA22 = fA22, fA5L, fA19
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fRes2L = fA1, fRes2H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fRes3H = fRes2H, f8, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA15 = fA15, fA5L, fA11 // v2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin18 = fLnSin18, fA4L, fLnSin16
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // h = N * log2_lo + h
      fms.s1             FR_h = fFloatN, FR_log2_lo, FR_h
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPolL = fPolL, fA4L, fA5
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = r * Q4 + Q3
      fma.s1             FR_poly_lo = FR_r, FR_Q4, FR_Q3
      nop.i              0
}
{ .mfi
      nop.m              0
      fmpy.s1            FR_rsq = FR_r, FR_r // rsq = r * r
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResL = fA3L, fAbsX, fResL
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin30 = fLnSin30, fA4L, fLnSin28
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fResH
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fRes3L = fRes2H, f8, fRes3H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1H = fRes3H, FR_log2_hi
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fB20, fA22, fA15
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin34 = fLnSin34, fA4L, fLnSin32
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin14 = fLnSin14, fA4L, fLnSin12
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      // poly_lo = poly_lo * r + Q2
      fma.s1             FR_poly_lo = FR_poly_lo, FR_r, FR_Q2
      nop.i              0
}
{ .mfi
      nop.m              0
      fnma.s1            FR_rcub = FR_rsq, FR_r, f0 // rcub = r^3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             FR_poly_hi = FR_Q1, FR_rsq, FR_r
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fA1L = fA1L, fResL
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      fma.s1             fLnSin22 = fLnSin22, fA4L, fLnSin20
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin26 = fLnSin26, fA4L, fLnSin24
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      fsub.s1            fRes1L = FR_log2_hi, fRes1H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, fA5L, fPolL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin34 = fLnSin36, fA5L, fLnSin34
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin18 = fLnSin18, fA5L, fLnSin14
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin6 = fLnSin6, fA4L, fLnSin4
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin10 = fLnSin10, fA4L, fLnSin8
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             FR_poly_hi = FR_Q1, FR_rsq, FR_r
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fA1L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo*r^3 + h
      fma.s1             FR_poly_lo = FR_poly_lo, FR_rcub, FR_h
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB2 = fLnSin2, fA4L, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fRes3H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, fB10, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin26 = fLnSin26, fA5L, fLnSin22
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin34 = fLnSin34, fA5L, fLnSin30
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin10 = fLnSin10, fA5L, fLnSin6
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin2L = fLnSin2L, fA4L, f0
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      fma.s1             fRes3L = fRes2L, f8, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Y_lo = poly_hi + poly_lo
      fsub.s1            FR_log2_lo = FR_poly_lo, FR_poly_hi
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fB4 = fLnSin2, fA4L, fB2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2H = fRes1H, fPol
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin34 = fLnSin34, fB20, fLnSin26
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fLnSin18 = fLnSin18, fB20, fLnSin10
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fLnSin2L = fB8, fLnSin2, fLnSin2L
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      fadd.s1            FR_log2_lo = FR_log2_lo, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fRes2L = fRes1H, fRes2H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fB6 = fLnSin34, fB18, fLnSin18
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fB4 = fLnSin2L, fB4
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, FR_log2_lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fPol
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fB12 = fB6, fA5L, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fRes1L
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      fms.s1             fB14 = fB6, fA5L, fB12
      nop.i              0
}
{ .mfb
      nop.m              0
      fadd.s1            fLnSin30 = fB2, fB12
      // branch out if x is negative
(p15) br.cond.spnt       _O_Half_neg
}
;;
{ .mfb
      nop.m              0
      // sign(x)*Pol(|x|) - log(|x|)
      fma.s0             f8 = fRes2H, f1, fRes2L
      // it's an answer already for positive x
      // exit if 0 < x < 0.5
      br.ret.sptk        b0
}
;;

// here if x is negative and |x| < 0.5
.align 32
_O_Half_neg:
{ .mfi
      nop.m              0
      fma.s1             fB14 = fB16, fB6, fB14
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fLnSin16 = fB2, fLnSin30
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fResH = fLnSin30, fRes2H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fLnSin16 = fLnSin16, fB12
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fB4 = fB14, fB4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fLnSin16 = fB4, fLnSin16
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fResL = fRes2H, fResH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fResL = fResL, fLnSin30
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fLnSin16 = fLnSin16, fRes2L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fResL = fResL, fLnSin16
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      // final result for -0.5 < x < 0
      fma.s0             f8 = fResH, f1, fResL
      // exit for -0.5 < x < 0
      br.ret.sptk        b0
}
;;

// here if x >= 8.0
// there are two computational paths:
// 1) For x >10.0 Stirling's formula is used
// 2) Polynomial approximation for 8.0 <= x <= 10.0
.align 32
lgammal_big_positive:
{ .mfi
      addl               rPolDataPtr = @ltoff(lgammal_data), gp
      fmerge.se          fSignifX =  f1, f8
      // Get high 15 bits of significand
      extr.u             GR_X_0 = rSignifX, 49, 15
}
{.mfi
      shladd             rZ1offsett = GR_Index1, 2, GR_ad_z_1  // Point to Z_1
      fnma.s1            fInvX = f8, fRcpX, f1 // start of 1st NR iteration
      adds               rSignif1andQ = 0x5, r0
}
;;
{.mfi
      ld4                GR_Z_1 = [rZ1offsett] // Load Z_1
      nop.f              0
      shl                rSignif1andQ = rSignif1andQ, 61 // significand of 1.25
}
{  .mfi
      cmp.eq             p8, p0 = rExpX, rExp8 // p8 = 1 if 8.0 <= x < 16
      nop.f              0
      adds               rSgnGam = 1, r0 // gamma is positive at this range
}
;;
{ .mfi
      shladd             GR_ad_tbl_1 = GR_Index1, 4, rTbl1Addr// Point to G_1
      nop.f              0
      add                GR_ad_q = -0x60, GR_ad_z_1 // Point to Constants_Q
}
{ .mlx
      ld8                rPolDataPtr    = [rPolDataPtr]
      movl               rDelta = 0x3FF2000000000000
}
;;
{ .mfi
      ldfps              FR_G, FR_H = [GR_ad_tbl_1],8 // Load G_1, H_1
      nop.f              0
      add                GR_ad_z_2 = 0x140, GR_ad_z_1 // Point to Constants_Z_2
}
{ .mfi
      // Point to Constants_G_H_h2
      add                GR_ad_tbl_2 = 0x180, GR_ad_z_1
      nop.f              0
      // p8 = 1 if 8.0 <= x <= 10.0
(p8)  cmp.leu.unc        p8, p0 = rSignifX, rSignif1andQ
}
;;
{ .mfi
      ldfd               FR_h = [GR_ad_tbl_1] // Load h_1
      nop.f              0
      // Get bits 30-15 of X_0 * Z_1
      pmpyshr2.u         GR_X_1 = GR_X_0,GR_Z_1,15
}
{ .mfb
(p8)  setf.d             FR_MHalf = rDelta
      nop.f              0
(p8)  br.cond.spnt       lgammal_8_10 // branch out if 8.0 <= x <= 10.0
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      ldfe               fA1 = [rPolDataPtr], 16 // Load overflow threshold
      fma.s1             fRcpX = fInvX, fRcpX, fRcpX // end of 1st NR iteration
      // Point to Constants_G_H_h3
      add                GR_ad_tbl_3 = 0x280, GR_ad_z_1
}
{ .mlx
      nop.m              0
      movl               rDelta = 0xBFE0000000000000 // -0.5 in DP
}
;;
{ .mfi
      ldfe               FR_log2_hi = [GR_ad_q],16 // Load log2_hi
      nop.f              0
      sub                GR_N = rExpX, rExpHalf, 1 // unbiased exponent of x
}
;;
{ .mfi
      ldfe               FR_log2_lo = [GR_ad_q],16 // Load log2_lo
      nop.f              0
      nop.i              0
}
{ .mfi
      setf.d             FR_MHalf = rDelta
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      // Put integer N into rightmost significand
      setf.sig           fFloatN = GR_N
      nop.f              0
      extr.u             GR_Index2 = GR_X_1, 6, 4 // Extract bits 6-9 of X_1
}
{ .mfi
      ldfe               FR_Q4 = [GR_ad_q], 16 // Load Q4
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      shladd             GR_ad_z_2 = GR_Index2, 2, GR_ad_z_2  // Point to Z_2
      nop.f              0
      shladd             GR_ad_tbl_2 = GR_Index2, 4, GR_ad_tbl_2// Point to G_2
}
{ .mfi
      ldfe               FR_Q3 = [GR_ad_q], 16 // Load Q3
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ld4                GR_Z_2 = [GR_ad_z_2] // Load Z_2
      fnma.s1            fInvX = f8, fRcpX, f1 // start of 2nd NR iteration
      nop.i              0
}
;;
{ .mfi
      ldfps              FR_G2, FR_H2 = [GR_ad_tbl_2], 8 // Load G_2, H_2
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfd               FR_h2 = [GR_ad_tbl_2] // Load h_2
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfe               FR_Q2 = [GR_ad_q],16 // Load Q2
      nop.f              0
      // Get bits 30-15 of X_1 * Z_2
      pmpyshr2.u         GR_X_2 = GR_X_1,GR_Z_2,15
}
;;
//
//    For performance, don't use result of pmpyshr2.u for 4 cycles.
//
{ .mfi
      ldfe               FR_Q1 = [GR_ad_q] // Load Q1
      fcmp.gt.s1         p7,p0 = f8, fA1 // check if x > overflow threshold
      nop.i              0
}
;;
{.mfi
      ldfpd              fA0, fA0L = [rPolDataPtr], 16 // Load two parts of C
      fma.s1             fRcpX = fInvX, fRcpX, fRcpX // end of 2nd NR iteration
      nop.i              0
}
;;
{ .mfb
      ldfpd              fB2, fA1 = [rPolDataPtr], 16
      nop.f              0
(p7)  br.cond.spnt       lgammal_overflow // branch if x > overflow threshold
}
;;
{.mfi
      ldfe               fB4 = [rPolDataPtr], 16
      fcvt.xf            fFloatN = fFloatN
      extr.u             GR_Index3 = GR_X_2, 1, 5 // Extract bits 1-5 of X_2
}
;;
{ .mfi
      shladd             GR_ad_tbl_3 = GR_Index3, 4, GR_ad_tbl_3// Point to G_3
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfe               fB6 = [rPolDataPtr], 16
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfps              FR_G3, FR_H3 = [GR_ad_tbl_3], 8 // Load G_3, H_3
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfd               FR_h3 = [GR_ad_tbl_3] // Load h_3
      fmpy.s1            FR_G = FR_G, FR_G2 // G = G_1 * G_2
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            FR_H = FR_H, FR_H2 // H = H_1 + H_2
      nop.i              0
}
;;

{ .mfi
      ldfe               fB8 = [rPolDataPtr], 16
      fadd.s1            FR_h = FR_h, FR_h2 // h = h_1 + h_2
      nop.i              0
}
{ .mfi
      nop.m              0
      fnma.s1            fInvX = f8, fRcpX, f1 // start of 3rd NR iteration
      nop.i              0
}
;;
{ .mfi
      ldfe               fB10 = [rPolDataPtr], 16
      nop.f              0
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
;;
{ .mfi
      ldfe               fB12 = [rPolDataPtr], 16
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfe               fB14 = [rPolDataPtr], 16
      nop.f              0
      nop.i              0
}
;;

{ .mfi
      ldfe               fB16 = [rPolDataPtr], 16
      // get double extended coefficients from two doubles
      // two doubles are needed in Stitling's formula for negative x
      fadd.s1            fB2 = fB2, fA1
      nop.i              0
}
;;
{ .mfi
      ldfe               fB18 = [rPolDataPtr], 16
      fma.s1             fInvX = fInvX, fRcpX, fRcpX // end of 3rd NR iteration
      nop.i              0
}
;;
{ .mfi
      ldfe               fB20 = [rPolDataPtr], 16
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fmpy.s1            FR_G = FR_G, FR_G3 // G = (G_1 * G_2) * G_3
      nop.i              0
}
{ .mfi
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fadd.s1            FR_H = FR_H, FR_H3 // H = (H_1 + H_2) + H_3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            FR_h = FR_h, FR_h3 // h = (h_1 + h_2) + h_3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRcpX = fInvX, fInvX, f0 // 1/x^2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA0L = fB2, fInvX, fA0L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             FR_r = fSignifX, FR_G, f1 // r = G * S_hi - 1
      nop.i              0
}
{ .mfi
      nop.m              0
      // High part of the log(x): Y_hi = N * log2_hi + H
      fma.s1             fRes2H = fFloatN, FR_log2_hi, FR_H
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      // h = N * log2_lo + h
      fma.s1             FR_h = fFloatN, FR_log2_lo, FR_h
      nop.i              0
}
{ .mfi
      nop.m              0
      // High part of the log(x): Y_hi = N * log2_hi + H
      fma.s1             fRes1H = fFloatN, FR_log2_hi, FR_H
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fPol = fB18, fRcpX, fB16 // v9
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA2L = fRcpX, fRcpX, f0   // v10
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fA3 = fB6, fRcpX, fB4     // v3
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA4 = fB10, fRcpX, fB8    // v4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes2H =fRes2H, f1, f1 //  log_Hi(x) -1
      nop.i              0
}
{ .mfi
      nop.m              0
      // poly_lo = r * Q4 + Q3
      fma.s1             FR_poly_lo = FR_r, FR_Q4, FR_Q3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes1H = fRes1H, FR_MHalf, f0 // -0.5*log_Hi(x)
      nop.i              0
}
{ .mfi
      nop.m              0
      fmpy.s1            FR_rsq = FR_r, FR_r // rsq = r * r
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA7 = fB14, fRcpX, fB12  // v7
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA8 = fA2L, fB20, fPol   // v8
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA2 = fA4, fA2L, fA3    // v2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA4L = fA2L, fA2L, f0    // v5
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResH = fRes2H, f8, f0 // (x*(ln(x)-1))hi
      nop.i              0
}
{ .mfi
      nop.m              0
      // poly_lo = poly_lo * r + Q2
      fma.s1             FR_poly_lo = FR_poly_lo, FR_r, FR_Q2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             FR_rcub = FR_rsq, FR_r, f0 // rcub = r^3
      nop.i              0
}
{ .mfi
      nop.m              0
      // poly_hi = Q1 * rsq + r
      fma.s1             FR_poly_hi = FR_Q1, FR_rsq, FR_r
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA11 = fRcpX, fInvX, f0 // 1/x^3
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA6 = fA8, fA2L, fA7   // v6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fResL = fRes2H, f8, fResH // d(x*(ln(x)-1))
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes3H = fResH, fRes1H // (x*(ln(x)-1) -0.5ln(x))hi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // poly_lo = poly_lo*r^3 + h
      fma.s1             FR_poly_lo = FR_poly_lo, FR_rcub, FR_h
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPol = fA4L, fA6, fA2   // v1
      nop.i              0
}
{ .mfi
      nop.m              0
      // raise inexact exception
      fma.s0             FR_log2_lo = FR_log2_lo, FR_log2_lo, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4H = fRes3H, fA0 // (x*(ln(x)-1) -0.5ln(x))hi + Chi
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fRes3L = fResH, fRes3H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Y_lo = poly_hi + poly_lo
      fadd.s1            fRes2L = FR_poly_hi, FR_poly_lo
      nop.i              0
}
;;

{ .mfi
      nop.m              0
      fma.s1             fA0L = fPol, fA11, fA0L // S(1/x) + Clo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fRes3L, fRes1H
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fRes4L = fRes3H, fRes4H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResL = fRes2L, f8 , fResL // lo part of x*(ln(x)-1)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Clo + S(1/x) - 0.5*logLo(x)
      fma.s1             fA0L = fRes2L, FR_MHalf, fA0L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fA0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Clo + S(1/x) - 0.5*logLo(x) + (x*(ln(x)-1))lo
      fadd.s1            fA0L = fA0L, fResL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fA0L
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      fma.s0             f8 = fRes4H, f1, fRes4L
      // exit for x > 10.0
      br.ret.sptk        b0
}
;;
// here if 8.0 <= x <= 10.0
// Result = P15(y), where y = x/8.0 - 1.5
.align 32
lgammal_8_10:
{ .mfi
      addl               rPolDataPtr    = @ltoff(lgammal_8_10_data), gp
      fms.s1             FR_FracX = fSignifX, f1, FR_MHalf // y = x/8.0 - 1.5
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
;;
{ .mfi
      ld8                rLnSinDataPtr = [rPolDataPtr]
      nop.f              0
      nop.i              0
}
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      adds               rZ1offsett = 32, rLnSinDataPtr
      nop.f              0
      nop.i              0
}
{ .mfi
      adds               rLnSinDataPtr = 48, rLnSinDataPtr
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA1, fA1L = [rPolDataPtr], 16 // A1
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfe               fA2 = [rZ1offsett], 32 // A5
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA0, fA0L = [rPolDataPtr], 16 // A0
      fma.s1             FR_rsq = FR_FracX, FR_FracX, f0 // y^2
      nop.i              0
}
{ .mfi
      ldfe               fA3 = [rLnSinDataPtr],32 // A5
      nop.f              0
      nop.i              0
}
;;
{ .mmf
      ldfe               fA4 = [rZ1offsett], 32 // A4
      ldfe               fA5 = [rLnSinDataPtr], 32 // A5
      nop.f              0
}
;;
{ .mmf
      ldfe               fA6 = [rZ1offsett], 32 // A6
      ldfe               fA7 = [rLnSinDataPtr], 32 // A7
      nop.f              0
}
;;
{ .mmf
      ldfe               fA8 = [rZ1offsett], 32 // A8
      ldfe               fA9 = [rLnSinDataPtr], 32 // A9
      nop.f              0
}
;;
{ .mmf
      ldfe               fA10 = [rZ1offsett], 32 // A10
      ldfe               fA11 = [rLnSinDataPtr], 32 // A11
      nop.f              0
}
;;
{ .mmf
      ldfe               fA12 = [rZ1offsett], 32 // A12
      ldfe               fA13 = [rLnSinDataPtr], 32 // A13
      fma.s1             FR_Q4 = FR_rsq, FR_rsq, f0 // y^4
}
;;
{ .mmf
      ldfe               fA14 = [rZ1offsett], 32 // A14
      ldfe               fA15 = [rLnSinDataPtr], 32 // A15
      nop.f              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes1H = FR_FracX, fA1, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA3 = fA3, FR_FracX, fA2 // v4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA5 = fA5, FR_FracX, fA4 // v5
      nop.i              0
}
;;
{ .mfi
      // store sign of GAMMA(x) if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fma.s1             fA3L = FR_Q4, FR_Q4, f0 // v9 = y^8
      nop.i              0
}
{ .mfi
      // store sign of GAMMA(x) if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s1             fA7 = fA7, FR_FracX, fA6 // v7
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA9 = fA9, FR_FracX, fA8 // v8
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes1L = FR_FracX, fA1, fRes1H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA11 = fA11, FR_FracX, fA10 // v12
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA13 = fA13, FR_FracX, fA12 // v13
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fRes2H = fRes1H, f1, fA0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA15 = fA15, FR_FracX, fA14 // v16
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA5 = fA5, FR_rsq, fA3 // v3
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA9 = fA9, FR_rsq, fA7 // v6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes1L = FR_FracX, fA1L, fRes1L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes2L = fA0, f1, fRes2H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA13 = fA13, FR_rsq, fA11 // v11
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA9 = fA9, FR_Q4, fA5 // v2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes1L = fRes1L, f1, fA0L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes2L = fRes2L, f1, fRes1H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA15 = fA15, FR_Q4, fA13 // v10
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes2L = fRes1L, f1, fRes2L
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fA3L, fA15, fA9
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             f8 = FR_rsq , fPol, fRes2H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, FR_rsq, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes1L = fRes2H, f1, f8
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes1L = fRes1L, f1, fPol
      nop.i              0
}
;;
{.mfi
      nop.m              0
      fma.s1             fRes1L = fRes1L, f1, fRes2L
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      fma.s0             f8 = f8, f1, fRes1L
      // exit for 8.0 <= x <= 10.0
      br.ret.sptk        b0
}
;;

// here if 4.0 <=x < 8.0
.align 32
lgammal_4_8:
{ .mfi
      addl               rPolDataPtr= @ltoff(lgammal_4_8_data),gp
      fms.s1             FR_FracX = fSignifX, f1, FR_MHalf
      adds               rSgnGam = 1, r0
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      nop.i              0
}
;;

{ .mfb
      adds               rTmpPtr = 160, rPolDataPtr
      nop.f              0
      // branch to special path which computes polynomial of 25th degree
      br.sptk            lgamma_polynom25
}
;;

// here if 2.25 <=x < 4.0
.align 32
lgammal_2Q_4:
{ .mfi
      addl               rPolDataPtr= @ltoff(lgammal_2Q_4_data),gp
      fms.s1             FR_FracX = fSignifX, f1, FR_MHalf
      adds               rSgnGam = 1, r0
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      nop.i              0
}
;;

{ .mfb
      adds               rTmpPtr = 160, rPolDataPtr
      nop.f              0
      // branch to special path which computes polynomial of 25th degree
      br.sptk            lgamma_polynom25
}
;;

// here if 0.5 <= |x| < 0.75
.align 32
lgammal_half_3Q:
.pred.rel "mutex", p14, p15
{ .mfi
(p14) addl               rPolDataPtr= @ltoff(lgammal_half_3Q_data),gp
     // FR_FracX = x - 0.625 for positive x
(p14) fms.s1             FR_FracX = f8, f1, FR_FracX
(p14) adds               rSgnGam = 1, r0
}
{ .mfi
(p15) addl               rPolDataPtr= @ltoff(lgammal_half_3Q_neg_data),gp
     // FR_FracX = x + 0.625 for negative x
(p15) fma.s1             FR_FracX = f8, f1, FR_FracX
(p15) adds               rSgnGam = -1, r0
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
       nop.f              0
       nop.i              0
}
;;
{ .mfb
      adds               rTmpPtr = 160, rPolDataPtr
      nop.f              0
      // branch to special path which computes polynomial of 25th degree
      br.sptk            lgamma_polynom25
}
;;
// here if 1.3125 <= x < 1.5625
.align 32
lgammal_loc_min:
{ .mfi
      adds               rSgnGam = 1, r0
      nop.f              0
      nop.i              0
}
{ .mfb
      adds               rTmpPtr = 160, rPolDataPtr
      fms.s1             FR_FracX = f8, f1, fA5L
      br.sptk            lgamma_polynom25
}
;;
// here if -2.605859375 <= x < -2.5
// special polynomial approximation used since neither "near root"
// approximation nor reflection formula give satisfactory accuracy on
// this range
.align 32
_neg2andHalf:
{ .mfi
      addl               rPolDataPtr= @ltoff(lgammal_neg2andHalf_data),gp
      fma.s1             FR_FracX = fB20, f1, f8 // 2.5 + x
      adds               rSgnGam = -1, r0
}
;;
{.mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      nop.i              0
}
;;
{ .mfb
      adds               rTmpPtr = 160, rPolDataPtr
      nop.f              0
      // branch to special path which computes polynomial of 25th degree
      br.sptk            lgamma_polynom25
}
;;

// here if -0.5 < x <= -0.40625
.align 32
lgammal_near_neg_half:
{ .mmf
      addl               rPolDataPtr= @ltoff(lgammal_near_neg_half_data),gp
      setf.exp           FR_FracX = rExpHalf
      nop.f              0
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      adds               rSgnGam = -1, r0
}
;;
{ .mfb
      adds               rTmpPtr = 160, rPolDataPtr
      fma.s1             FR_FracX = FR_FracX, f1, f8
      // branch to special path which computes polynomial of 25th degree
      br.sptk            lgamma_polynom25
}
;;

// here if there an answer is P25(x)
// rPolDataPtr, rTmpPtr point to coefficients
// x is in FR_FracX register
.align 32
lgamma_polynom25:
{ .mfi
      ldfpd              fA3, fA0L = [rPolDataPtr], 16 // A3
      nop.f              0
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
{ .mfi
      ldfpd              fA18, fA19 = [rTmpPtr], 16 // D7, D6
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA1, fA1L = [rPolDataPtr], 16 // A1
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA16, fA17 = [rTmpPtr], 16 // D4, D5
      nop.f              0
}
;;
{ .mfi
      ldfpd              fA12, fA13 = [rPolDataPtr], 16 // D0, D1
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA14, fA15 = [rTmpPtr], 16 // D2, D3
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA24, fA25 = [rPolDataPtr], 16 // C21, C20
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA22, fA23 = [rTmpPtr], 16 // C19, C18
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA2, fA2L = [rPolDataPtr], 16 // A2
      fma.s1             fA4L = FR_FracX, FR_FracX, f0 // x^2
      nop.i              0
}
{ .mfi
      ldfpd              fA20, fA21 = [rTmpPtr], 16 // C17, C16
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfe               fA11 = [rTmpPtr], 16 // E7
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA0, fA3L = [rPolDataPtr], 16 // A0
      nop.f              0
      nop.i              0
};;
{ .mfi
      ldfe               fA10 = [rPolDataPtr], 16 // E6
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfe               fA9 = [rTmpPtr], 16 // E5
      nop.f              0
      nop.i              0
}
;;
{ .mmf
      ldfe               fA8 = [rPolDataPtr], 16 // E4
      ldfe               fA7 = [rTmpPtr], 16 // E3
      nop.f              0
}
;;
{ .mmf
      ldfe               fA6 = [rPolDataPtr], 16 // E2
      ldfe               fA5 = [rTmpPtr], 16 // E1
      nop.f              0
}
;;
{ .mfi
      ldfe               fA4 = [rPolDataPtr], 16 // E0
      fma.s1             fA5L = fA4L, fA4L, f0 // x^4
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fB2 = FR_FracX, FR_FracX, fA4L // x^2 - <x^2>
      nop.i              0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fma.s1             fRes4H = fA3, FR_FracX, f0 // (A3*x)hi
      nop.i              0
}
{ .mfi
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s1             fA19 = fA19, FR_FracX, fA18 // D7*x + D6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResH = fA1, FR_FracX, f0 // (A1*x)hi
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB6 = fA1L, FR_FracX, fA0L // A1L*x + A0L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA17 = fA17, FR_FracX, fA16 // D5*x + D4
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA15 = fA15, FR_FracX, fA14 // D3*x + D2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA25 = fA25, FR_FracX, fA24 // C21*x + C20
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA13 = fA13, FR_FracX, fA12 // D1*x + D0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA23 = fA23, FR_FracX, fA22 // C19*x + C18
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA21 = fA21, FR_FracX, fA20 // C17*x + C16
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes4L = fA3, FR_FracX, fRes4H // delta((A3*x)hi)
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes2H = fRes4H, fA2 // (A3*x + A2)hi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fResL = fA1, FR_FracX, fResH // d(A1*x)
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes1H = fResH, fA0 // (A1*x + A0)hi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA19 = fA19, fA4L, fA17 // Dhi
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA11 = fA11, FR_FracX, fA10 // E7*x + E6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // Doing this to raise inexact flag
      fma.s0             fA10 = fA0, fA0, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA15 = fA15, fA4L, fA13 // Dlo
      nop.i              0
}
{ .mfi
      nop.m              0
      // (C21*x + C20)*x^2 + C19*x + C18
      fma.s1             fA25 = fA25, fA4L, fA23
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA9 = fA9, FR_FracX, fA8 // E5*x + E4
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA7 = fA7, FR_FracX, fA6 // E3*x + E2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes4L = fA3L, FR_FracX, fRes4L // (A3*x)lo
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fRes2L = fA2, fRes2H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fResL = fResL, fB6 // (A1L*x + A0L) + d(A1*x)
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fRes1L = fA0, fRes1H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA5 = fA5, FR_FracX, fA4 // E1*x + E0
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB8 = fA5L, fA5L, f0  // x^8
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // ((C21*x + C20)*x^2 + C19*x + C18)*x^2 + C17*x + C16
      fma.s1             fA25 = fA25, fA4L, fA21
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA19 = fA19, fA5L, fA15 // D
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA11 = fA11, fA4L, fA9 // Ehi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fRes4H
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fA2L // (A3*x)lo + A2L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes3H = fRes2H, fA4L, f0 //  ((A3*x + A2)*x^2)hi
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fResH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes3L = fRes2H, fB2, f0 // (A3*x + A2)hi*d(x^2)
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA7 = fA7, fA4L, fA5 // Elo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA25 = fA25, fB8, fA19 // C*x^8 + D
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fRes4L // (A3*x + A2)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fB4 = fRes2H, fA4L, fRes3H // d((A3*x + A2)*x^2))
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fResL // (A1*x + A0)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fB20 = fRes3H, fRes1H // Phi
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA11 = fA11, fA5L, fA7 // E
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      //  ( (A3*x + A2)lo*<x^2> + (A3*x + A2)hi*d(x^2))
      fma.s1             fRes3L = fRes2L, fA4L, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // d((A3*x + A2)*x^2)) + (A1*x + A0)lo
      fadd.s1            fRes1L = fRes1L, fB4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fB18 = fRes1H, fB20
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fA25, fB8, fA11
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fB18 = fB18, fRes3H
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fRes4H = fPol, fA5L, fB20
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPolL = fPol, fA5L, f0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fB18 = fB18, fRes1L // Plo
      nop.i              0
}
{ .mfi
      nop.m              0
      fsub.s1            fRes4L = fB20, fRes4H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fB18 = fB18, fPolL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fB18
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      fma.s0             f8 = fRes4H, f1, fRes4L
      // P25(x) computed, exit here
      br.ret.sptk        b0
}
;;


// here if 0.75 <= x < 1.3125
.align 32
lgammal_03Q_1Q:
{ .mfi
      addl               rPolDataPtr= @ltoff(lgammal_03Q_1Q_data),gp
      fma.s1             FR_FracX = fA5L, f1, f0 // x
      adds               rSgnGam = 1, r0
}
{ .mfi
      nop.m              0
      fma.s1             fB4 = fA5L, fA5L, f0 // x^2
      nop.i              0
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      nop.i              0
}
;;
{ .mfb
      adds               rTmpPtr = 144, rPolDataPtr
      nop.f              0
      br.sptk            lgamma_polynom24x
}
;;

// here if 1.5625 <= x < 2.25
.align 32
lgammal_13Q_2Q:
{ .mfi
      addl               rPolDataPtr= @ltoff(lgammal_13Q_2Q_data),gp
      fma.s1             FR_FracX = fB4, f1, f0 // x
      adds               rSgnGam = 1, r0
}
{ .mfi
      nop.m              0
      fma.s1             fB4 = fB4, fB4, f0 // x^2
      nop.i              0
}
;;
{ .mfi
      ld8                rPolDataPtr = [rPolDataPtr]
      nop.f              0
      nop.i              0
}
;;
{ .mfb
      adds               rTmpPtr = 144, rPolDataPtr
      nop.f              0
      br.sptk            lgamma_polynom24x
}
;;

// here if result is Pol24(x)
// x is in FR_FracX,
// rPolDataPtr, rTmpPtr point to coefficients
.align 32
lgamma_polynom24x:
{ .mfi
      ldfpd              fA4, fA2L = [rPolDataPtr], 16
      nop.f              0
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
{ .mfi
      ldfpd              fA23, fA24 = [rTmpPtr], 16 // C18, C19
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA3, fA1L = [rPolDataPtr], 16
      fma.s1             fA5L = fB4, fB4, f0 // x^4
      nop.i              0
}
{ .mfi
      ldfpd              fA19, fA20 = [rTmpPtr], 16 // D6, D7
      fms.s1             fB2 = FR_FracX, FR_FracX, fB4 // x^2 - <x^2>
      nop.i              0
}
;;
{ .mmf
      ldfpd              fA15, fA16 = [rPolDataPtr], 16 // D2, D3
      ldfpd              fA17, fA18 = [rTmpPtr], 16 // D4, D5
      nop.f              0
}
;;
{ .mmf
      ldfpd              fA13, fA14 = [rPolDataPtr], 16 // D0, D1
      ldfpd              fA12, fA21 = [rTmpPtr], 16 // E7, C16
      nop.f              0
}
;;
{ .mfi
      ldfe               fA11 = [rPolDataPtr], 16 // E6
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfe               fA10 = [rTmpPtr], 16 // E5
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA2, fA4L = [rPolDataPtr], 16
      nop.f              0
      nop.i              0
}
{ .mfi
      ldfpd              fA1, fA3L = [rTmpPtr], 16
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      ldfpd              fA22, fA25 = [rPolDataPtr], 16 // C17, C20
      fma.s1             fA0 = fA5L, fA5L, f0 // x^8
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA0L = fA5L, FR_FracX, f0 // x^5
      nop.i              0
}
;;
{ .mmf
      ldfe               fA9 = [rPolDataPtr], 16 // E4
      ldfe               fA8 = [rTmpPtr], 16 // E3
      nop.f              0
}
;;
{ .mmf
      ldfe               fA7 = [rPolDataPtr], 16 // E2
      ldfe               fA6 = [rTmpPtr], 16 // E1
      nop.f              0
}
;;
{ .mfi
      ldfe               fA5 = [rTmpPtr], 16 // E0
      fma.s1             fRes4H = fA4, fB4, f0 // A4*<x^2>
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fA24, FR_FracX, fA23 // C19*x + C18
      nop.i              0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fma.s1             fRes1H = fA3, fB4, f0 // A3*<x^2>
      nop.i              0
}
{ .mfi
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s1             fA1L = fA3, fB2,fA1L //  A3*d(x^2) + A1L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA20 = fA20, FR_FracX, fA19 // D7*x + D6
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA18 = fA18, FR_FracX, fA17 // D5*x + D4
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA16 = fA16, FR_FracX, fA15 // D3*x + D2
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA14 = fA14, FR_FracX, fA13 // D1*x + D0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA2L = fA4, fB2,fA2L //  A4*d(x^2) + A2L
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA12 = fA12, FR_FracX, fA11 // E7*x + E6
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes2L = fA4, fB4, fRes4H  // delta(A4*<x^2>)
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes2H = fRes4H, fA2       // A4*<x^2> + A2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fms.s1             fRes3L = fA3, fB4, fRes1H  // delta(A3*<x^2>)
      nop.i              0
}
{ .mfi
      nop.m              0
      fadd.s1            fRes3H = fRes1H, fA1       // A3*<x^2> + A1
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA20 = fA20, fB4, fA18 // (D7*x + D6)*x^2 + D5*x + D4
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA22 = fA22, FR_FracX, fA21 // C17*x + C16
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA16 = fA16, fB4, fA14 // (D3*x + D2)*x^2 + D1*x + D0
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPol = fA25, fB4, fPol // C20*x^2 + C19*x + C18
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA2L = fA4L, fB4, fA2L //  A4L*<x^2> + A4*d(x^2) + A2L
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA1L = fA3L, fB4, fA1L //  A3L*<x^2> + A3*d(x^2) + A1L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fRes4L = fA2, fRes2H // d1
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fResH = fRes2H, fB4, f0 // (A4*<x^2> + A2)*x^2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fRes1L = fA1, fRes3H // d1
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fB6 = fRes3H, FR_FracX, f0 // (A3*<x^2> + A1)*x
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fA10 = fA10, FR_FracX, fA9  // E5*x + E4
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA8 = fA8, FR_FracX, fA7 // E3*x + E2
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // (C20*x^2 + C19*x + C18)*x^2 + C17*x + C16
      fma.s1             fPol = fPol, fB4, fA22
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA6 = fA6, FR_FracX, fA5 // E1*x + E0
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // A4L*<x^2> + A4*d(x^2) + A2L + delta(A4*<x^2>)
      fadd.s1            fRes2L = fA2L, fRes2L
      nop.i              0
}
{ .mfi
      nop.m              0
      // A3L*<x^2> + A3*d(x^2) + A1L + delta(A3*<x^2>)
      fadd.s1            fRes3L = fA1L, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fRes4H // d2
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fResL = fRes2H, fB4, fResH  // d(A4*<x^2> + A2)*x^2)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes1L = fRes1L, fRes1H // d2
      nop.i              0
}
{ .mfi
      nop.m              0
      fms.s1             fB8 = fRes3H, FR_FracX, fB6 // d((A3*<x^2> + A1)*x)
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fB10 = fResH, fB6 // (A4*x^4 + .. + A1*x)hi
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA12 = fA12, fB4, fA10 // Ehi
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      // ((D7*x + D6)*x^2 + D5*x + D4)*x^4 + (D3*x + D2)*x^2 + D1*x + D0
      fma.s1             fA20 = fA20, fA5L, fA16
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fA8 = fA8, fB4, fA6 // Elo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes2L = fRes2L, fRes4L // (A4*<x^2> + A2)lo
      nop.i              0
}
{ .mfi
      nop.m              0
      // d(A4*<x^2> + A2)*x^2) + A4*<x^2> + A2)*d(x^2)
      fma.s1             fResL = fRes2H, fB2, fResL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fRes3L, fRes1L   // (A4*<x^2> + A2)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fB12 = fB6, fB10
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, fA0, fA20 // PolC*x^8 + PolD
      nop.i              0
}
{ .mfi
      nop.m              0
      fma.s1             fPolL = fA12, fA5L, fA8 // E
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fResL = fB4, fRes2L, fResL // ((A4*<x^2> + A2)*x^2)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes3L = fRes3L, FR_FracX, fB8 // ((A3*<x^2> + A1)*x)lo
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fB12 = fB12, fResH
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fPol = fPol, fA0, fPolL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fRes3L, fResL
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes2H = fPol, fA0L, fB10
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes3L = fB12, fRes3L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fsub.s1            fRes4L = fB10, fRes2H
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fma.s1             fRes4L = fPol, fA0L, fRes4L
      nop.i              0
}
;;
{ .mfi
      nop.m              0
      fadd.s1            fRes4L = fRes4L, fRes3L
      nop.i              0
}
;;
{ .mfb
      nop.m              0
      // final result for all paths for which the result is Pol24(x)
      fma.s0             f8 = fRes2H, f1, fRes4L
      // here is the exit for all paths for which the result is Pol24(x)
      br.ret.sptk        b0
}
;;


//  here if x is natval, nan, +/-inf, +/-0, or denormal
.align 32
lgammal_spec:
{ .mfi
      nop.m              0
      fclass.m           p9, p0 =  f8, 0xB // +/-denormals
      nop.i              0
};;
{ .mfi
      nop.m              0
      fclass.m           p6, p0 =  f8, 0x1E1 // Test x for natval, nan, +inf
      nop.i              0
};;
{ .mfb
      nop.m              0
      fclass.m           p7, p0 =  f8, 0x7 // +/-0
(p9)  br.cond.sptk       lgammal_denormal_input
};;
{ .mfb
      nop.m              0
      nop.f              0
      // branch out if x is natval, nan, +inf
(p6)  br.cond.spnt       lgammal_nan_pinf
};;
{ .mfb
      nop.m              0
      nop.f              0
(p7)  br.cond.spnt       lgammal_singularity
};;
// if we are still here then x = -inf
{ .mfi
      cmp.eq             p6, p7 = 4, rSgnGamSize
      nop.f              0
      adds               rSgnGam = 1, r0
};;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      nop.f              0
      nop.i              0
}
{ .mfb
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s0             f8 = f8,f8,f0 // return +inf, no call to error support
      br.ret.spnt        b0
};;

// here if x is NaN, NatVal or +INF
.align 32
lgammal_nan_pinf:
{ .mfi
      cmp.eq             p6, p7 = 4, rSgnGamSize
      nop.f              0
      adds               rSgnGam = 1, r0
}
;;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      fma.s0             f8 = f8,f1,f8 // return x+x if x is natval, nan, +inf
      nop.i              0
}
{ .mfb
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      nop.f              0
      br.ret.sptk        b0
}
;;

// here if x denormal or unnormal
.align 32
lgammal_denormal_input:
{ .mfi
      nop.m              0
      fma.s0             fResH = f1, f1, f8 // raise denormal exception
      nop.i              0
}
{ .mfi
      nop.m              0
      fnorm.s1           f8 = f8 // normalize input value
      nop.i              0
}
;;
{ .mfi
      getf.sig           rSignifX = f8
      fmerge.se          fSignifX =  f1, f8
      nop.i              0
}
{ .mfi
      getf.exp           rSignExpX = f8
      fcvt.fx.s1         fXint = f8 // Convert arg to int (int repres. in FR)
      nop.i              0
}
;;
{ .mfi
      getf.exp           rSignExpX = f8
      fcmp.lt.s1         p15, p14 = f8, f0
      nop.i              0
}
;;
{ .mfb
      and                rExpX = rSignExpX, r17Ones
      fmerge.s           fAbsX = f1, f8 // |x|
      br.cond.sptk       _deno_back_to_main_path
}
;;


// here if overflow (x > overflow_bound)
.align 32
lgammal_overflow:
{ .mfi
      addl               r8 = 0x1FFFE, r0
      nop.f              0
      cmp.eq             p6, p7 = 4, rSgnGamSize
}
{ .mfi
      adds               rSgnGam = 1, r0
      nop.f              0
      nop.i              0
}
;;
{ .mfi
      setf.exp           f9 = r8
      fmerge.s           FR_X = f8,f8
      mov                GR_Parameter_TAG = 102 // overflow
};;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      nop.f              0
      nop.i              0
}
{ .mfb
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      fma.s0             FR_RESULT = f9,f9,f0 // Set I,O and +INF result
      br.cond.sptk       __libm_error_region
};;

// here if x is negative integer or +/-0 (SINGULARITY)
.align 32
lgammal_singularity:
{ .mfi
      adds               rSgnGam = 1, r0
      fclass.m           p8,p0 = f8,0x6 // is x -0?
      mov                GR_Parameter_TAG = 103 // negative
}
{ .mfi
      cmp.eq             p6, p7 = 4, rSgnGamSize
      fma.s1             FR_X = f0,f0,f8
      nop.i              0
};;
{ .mfi
(p8)  sub                rSgnGam = r0, rSgnGam
      nop.f              0
      nop.i              0
}
{ .mfi
      nop.m              0
      nop.f              0
      nop.i              0
};;
{ .mfi
      // store signgam if size of variable is 4 bytes
(p6)  st4                [rSgnGamAddr] = rSgnGam
      nop.f              0
      nop.i              0
}
{ .mfb
      // store signgam if size of variable is 8 bytes
(p7)  st8                [rSgnGamAddr] = rSgnGam
      frcpa.s0           FR_RESULT, p0 = f1, f0
      br.cond.sptk       __libm_error_region
};;

GLOBAL_LIBM_END(__libm_lgammal)



LOCAL_LIBM_ENTRY(__libm_error_region)
.prologue
{ .mfi
        add   GR_Parameter_Y=-32,sp             // Parameter 2 value
        nop.f 0
.save   ar.pfs,GR_SAVE_PFS
        mov  GR_SAVE_PFS=ar.pfs                 // Save ar.pfs
}
{ .mfi
.fframe 64
        add sp=-64,sp                           // Create new stack
        nop.f 0
        mov GR_SAVE_GP=gp                       // Save gp
};;
{ .mmi
        stfe [GR_Parameter_Y] = FR_Y,16         // Save Parameter 2 on stack
        add GR_Parameter_X = 16,sp              // Parameter 1 address
.save   b0, GR_SAVE_B0
        mov GR_SAVE_B0=b0                       // Save b0
};;
.body
{ .mib
        stfe [GR_Parameter_X] = FR_X            // Store Parameter 1 on stack
        add   GR_Parameter_RESULT = 0,GR_Parameter_Y
        nop.b 0                                 // Parameter 3 address
}
{ .mib
        stfe [GR_Parameter_Y] = FR_RESULT      // Store Parameter 3 on stack
        add   GR_Parameter_Y = -16,GR_Parameter_Y
        br.call.sptk b0=__libm_error_support#  // Call error handling function
};;
{ .mmi
        add   GR_Parameter_RESULT = 48,sp
        nop.m 999
        nop.i 999
};;
{ .mmi
        ldfe  f8 = [GR_Parameter_RESULT]       // Get return result off stack
.restore sp
        add   sp = 64,sp                       // Restore stack pointer
        mov   b0 = GR_SAVE_B0                  // Restore return address
};;
{ .mib
        mov   gp = GR_SAVE_GP                  // Restore gp
        mov   ar.pfs = GR_SAVE_PFS             // Restore ar.pfs
        br.ret.sptk     b0                     // Return
};;

LOCAL_LIBM_END(__libm_error_region#)

.type   __libm_error_support#,@function
.global __libm_error_support#