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
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
"http://www.w3.org/TR/REC-html40/loose.dtd">
<html>
<head>
<meta content="text/html; charset=iso-8859-1" http-equiv="Content-Type">
<title>Dear Diary</title>
</head>
<body alink="#FF0000" vlink="#55188A" link="#0000EF" bgcolor="#FFFFFF" text="#000099">
<table border="0">
<tr align="left">
<td valign="top">
<table border="0">
<tr align="left"><td valign="top" >
<img width="120" height="171" src="uml-small.png">
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">
<font size="-1"><a href="index.html">Site Home Page</a></font>
<br>
<font size="-1"><a href="uses.html">What it's good for</a></font>
<br>
<font size="-1"><a href="case-studies.html">Case Studies</a></font>
<br>
<font size="-1"><a href="kernel.html">Kernel Capabilities</a></font>
<br>
<font size="-1"><a href="dl-sf.html">Downloading it</a></font>
<br>
<font size="-1"><a href="run.html">Running it</a></font>
<br>
<font size="-1"><a href="compile.html">Compiling</a></font>
<br>
<font size="-1"><a href="install.html">Installation</a></font>
<br>
<font size="-1"><a href="fs_making.html">Building filesystems</a></font>
<br>
<font size="-1"><a href="faq.html">Troubles</a></font>
<br>
<font size="-1"><a href="contrib.html">User Contributions</a></font>
<br>
<font size="-1"><a href="links.html">Related Links</a></font>
<br>
<font size="-1"><a href="todo.html">The ToDo list</a></font>
<br>
<font size="-1"><a href="projects.html">Projects</a></font>
<br>
<font size="-1"><a href="diary.html">Diary</a></font>
<br>
<font size="-1"><a href="thanks.html">Thanks</a></font>
<br>
<font size="-1"><a href="contacts.html">Contacts</a></font>
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">Tutorials<br>
<font size="-1"><a href="UserModeLinux-HOWTO.html">The HOWTO (html)</a></font>
<br>
<font size="-1"><a href="UserModeLinux-HOWTO.txt">The HOWTO (text)</a></font>
<br>
<font size="-1"><a href="hostfs.html">Host file access</a></font>
<br>
<font size="-1"><a href="input.html">Device inputs</a></font>
<br>
<font size="-1"><a href="shared_fs.html">Sharing filesystems</a></font>
<br>
<font size="-1"><a href="fs.html">Creating filesystems</a></font>
<br>
<font size="-1"><a href="networking.html">Virtual Networking</a></font>
<br>
<font size="-1"><a href="mconsole.html">Management Console</a></font>
<br>
<font size="-1"><a href="debugging.html">Kernel Debugging</a></font>
<br>
<font size="-1"><a href="gprof.html">gprof and gcov</a></font>
<br>
<font size="-1"><a href="xtut.html">Running X</a></font>
<br>
<font size="-1"><a href="trouble.html">Diagnosing problems</a></font>
<br>
<font size="-1"><a href="config.html">Configuration</a></font>
<br>
<font size="-1"><a href="slack_readme.html">Installing Slackware</a></font>
<br>
<font size="-1"><a href="arch-port.html">Porting UML</a></font>
<br>
<font size="-1"><a href="iomem.html">IO memory emulation</a></font>
<br>
<font size="-1"><a href="2G-2G.html">UML on 2G/2G hosts</a></font>
<br>
<font size="-1"><a href="lksct/index.html">Adding a UML system call</a></font>
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">How you can help<br>
<font size="-1"><a href="help-gen.html">Overview</a></font>
<br>
<font size="-1"><a href="help-doc.html">Documentation</a></font>
<br>
<font size="-1"><a href="help-userspace.html">Utilities</a></font>
<br>
<font size="-1"><a href="help-kernel-v1.html">Kernel bugs</a></font>
<br>
<font size="-1"><a href="projects.html">Kernel projects</a></font>
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">Screenshots<br>
<font size="-1"><a href="net.html">A virtual network</a></font>
<br>
<font size="-1"><a href="x.html">An X session</a></font>
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">Transcripts<br>
<font size="-1"><a href="login.html">A login session</a></font>
<br>
<font size="-1"><a href="debug-session.html">A debugging session</a></font>
<br>
<font size="-1"><a href="slackinst.html">Slackware installation</a></font>
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">Reference<br>
<font size="-1"><a href="switches.html">Kernel switches</a></font>
<br>
<font size="-1"><a href="slack_readme.html">Slackware README</a></font>
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">Papers<br>
<font size="-1"><a href="als2000/index.html">ALS 2000 paper (html)</a></font>
<br>
<font size="-1"><a href="als2000.tex">ALS 2000 paper (TeX)</a></font>
<br>
<font size="-1"><a href="slides/als2000/slides.html">ALS 2000 slides</a></font>
<br>
<font size="-1"><a href="slides/lca2001/lca.html">LCA 2001 slides</a></font>
<br>
<font size="-1"><a href="slides/ols2001/index.html">OLS 2001 paper (html)</a></font>
<br>
<font size="-1"><a href="ols2001.tex">OLS 2001 paper (TeX)</a></font>
<br>
<font size="-1"><a href="als2001/index.html">ALS 2001 paper (html)</a></font>
<br>
<font size="-1"><a href="als2001.tex">ALS 2001 paper (TeX)</a></font>
<br>
<font size="-1"><a href="slides/ists2002/umlsec.htm">UML security (html)</a></font>
<br>
<font size="-1"><a href="slides/lca2002/lca2002.htm">LCA 2002 (html)</a></font>
<br>
<font size="-1"><a href="slides/wvu2002/wvu2002.htm">WVU 2002 (html)</a></font>
</td></tr>
<tr align="left"><td valign="top" bgcolor="#e0e0e0">Fun and Games<br>
<font size="-1"><a href="http://user-mode-linux.sourceforge.net/cgi-bin/hangman">Kernel Hangman</a></font>
<br>
<font size="-1"><a href="sdotm.html">Disaster of the Month</a></font>
</td></tr>
</table>
</td>
<td valign="top" align="left">
<center>
<h3>Dear Diary</h3>
</center>
This page contains information about what's currently happening with the
project. I may update it once in a while if I feel like it.
<p>
<a name="4-25-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Apr 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
This was a fairly lazy week. I went down to West Virginia at the
invitation of David Krovich to give a talk at WVU. That went fairly
well - the chairman of the CSEE department was apparently impressed by
the number of students who attended. After dinner, I gave another
talk at a MORLUG meeting (MORLUG == Morgantown LUG) which consisted of
me firing up UMLs and demonstrating various neat things you can do
with it. This also seemed to go well.
<p>
In extra-curricular activities, we tried to go hiking on Sunday, but
got rained on heavily. It looks like nice country if only the clouds
would get out of the way so you could see something.
<p>
In UML news, I spent some time fixing the iomem support. It was
broken in such a way that I had a hard time believing it ever worked.
What I now think is more likely is that the VM system changed in a way
that broken iomem, but no one noticed. The problem was that the VM
system deals largely with page structs instead of raw page addresses
and the iomem regions had no sane mem_maps, so they had no page
structs.
<p>
I fixed this by changing the infrastructure to allow for segmented
physical memory consisting of regions which have their own separate
mem_maps. This will allow for plugging and unplugging of iomem
regions. I was hoping this would work for physical memory regions as
well (plugging anyway; unplugging is harder), and after a bunch of
failed experiments, decided that this wasn't going to be.
<p>
After getting this working, I decided to release 2.4.18-19. It also
contains James McMechan's partitioned device support and a bunch of
smaller bug fixes that were noticed by various people.
</blockquote>
<a name="4-15-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">15 Apr 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
More bug-bashing. My lists contain a total of 93 items, but a large
number of them are about to die. The umlgdb expect script from
Chandan Kudige will knock off the two related to reloading module
symbols. When I verify that UML can boot as a diskless client, that
will get rid of five more. James McMechan's ubd partition patch is
currently accounting for six items. I should be able to knock all
those off in the next couple of days.
<p>
I released an RPM last Tuesday to get the accumulated changes out to a
wider audience. I think I'm going to start releasing an RPM every two
or three weeks from now on. 2.4.x releases are too far apart for me now.
</blockquote>
<a name="4-06-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">6 Apr 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I spent the last few days on a bug-smashing spree. To-do items had
accumulated at an alarming pace over the last couple weeks, so I
decided to whack away at them a bit. I had over 110 items on my
various lists, so I knocked off as many as were easy to kill. I now have
99 items, so I got rid of more than 10 of them.
<p>
Prominent among them are
<ul>
<li>
floating point registers not being available to gdb inside UML or
stored in core files
</li>
<li>
hostfs not being able to create unix sockets
</li>
<li>
the daemon transport now gets its MAC from uml_switch
</li>
<li>
if the umid is set on the command line, it is put into host process
names and into xterm title bars
</li>
</ul>
There were some small patches from mulix, Sapan, and Daniel Phillips,
which all did useful things.
<p>
With that, I released 2.4.18-14 and uml_utilities_20020406. This may
be the basis of another full release (with an RPM instead of just a
patch). I need to do that soon since the 2.4.18 UML is getting fairly
old at this point.
</blockquote>
<a name="4-03-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">3 Apr 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
In the nearly two weeks since the last entry, I've pretty much knocked
off the console flow control bug (the one remaining piece is to make
sure that it works correctly when ptys deliver output SIGIO) and the
init hang on older machines. These are two of the oldest UML bugs.
<p>
The init hang is caused by init executing a cmov on a processor that
doesn't support it. What I ended up doing (at the suggestion of Alan
Cox who saw this on one of his boxes) is detect cmov support by
looking at /proc/cpuinfo, then panic if init gets a SIGILL on a cmov.
<p>
In other news, I got all of Bill Stearns' bootable filesystems over to
SF, so the links to them are likely to work. They're not on
ftp.nl.linux.org yet though.
</blockquote>
<a name="3-22-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">22 Mar 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I've decided that I'm out of major design changes I need to make and
can now concentrate on knocking items off the todo list. The one
published on the site has 44 things on it. I also keep a todo mail
folder containing pieces of mail that describe something well enough
that I can keep track of it. That folder contains 54 messages right
now, so I've got almost 100 things to do. Some are duplicates, and
some are already fixed and I haven't figured it yet, so the actual
number is somewhat smaller.
<p>
My latest run of the test suite succeeded in booting all of the
filesystems available from the UML site, which is probably the first
time that has ever been true.
</blockquote>
<a name="3-15-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">15 Mar 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I discovered a remnant bug. 'strace -p' didn't work. Fixing that was
easy enough, but I decided to clean up some code while I was in it.
That took more work than I expected, since it involved saving state in
the thread structure in signal handlers, and when you forget to
restore the old state, you get very obscure misbehaviors. I ended up
backing out the changes and putting them back in one line at a time
before it dawned on me that perhaps restoring the old stuff would be a
good plan.
<p>
So, I released 2.4.18-7 today. I also released another version of the
test suite. This one allows tests to be interactive and for the perl
test driver to interact with them.
<p>
In separate news, my paper proposal for OLS was accepted. So, it
looks like I'm on the hook for another paper.
</blockquote>
<a name="3-14-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">14 Mar 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
After much surgery, I finished the pt_regs to sigcontext work and put
UML back together. There were surprisingly few bugs to chase once it
compiled again. I fixed around three bugs, and haven't found any more
since.
<p>
I did some more work on the test suite. There is now a kernel build
test, for which I wrote a perl mconsole client. This prompted me to
add the option of the mconsole driver sending the name of its socket
to a socket specified on the command line. This allows scripts to
find out where to send mconsole commands without having to parse the
boot log. It also tells them when UML has booted to that point.
<p>
This test also exposed a bug in the mconsole driver which caused
commands not to be NULL-terminated.
</blockquote>
<a name="3-7-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">7 Mar 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
Bad news. I decided to do a UML-sucks/rocks-ometer on Google. Well,
<a href="http://www.google.com/search?num=100&hl=en&q=%22user+mode+linux%22+sucks">"sucks"</a> beats
<a href="http://www.google.com/search?num=100&hl=en&as_qdr=all&q=%22user+mode+linux%22+rocks">"rocks"</a> 125-83. To
make things worse, the first few "sucks" entries are me.
</blockquote>
<a name="3-4-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">4 Mar 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I made the 2.4.18 announcement on freshmeat today. That is the final
piece of this release.
<p>
I have known for a long time that UML is susceptible to bus errors in
random places in the code if it touches memory that it had previously
mmapped, but the host can't back with physical memory. This turned
out to be a lot easier to trigger than I expected. I run with tmpfs
mounted on /tmp for speed reasons. It has a maximum size of 1/2 RAM
by default, which is 128M for me. I got a UML to hit that limit and
crash with a bus error.
<p>
So, I posted an RFC to lkml asking for comments on my proposed
solution which was to add a hook to __alloc_pages to allow the
architecture to touch pages before they're returned to the caller.
Physical architectures have no need of this because they have a known
amount of physical memory and they know it's not going anywhere. UML
doesn't, so it needs to assure itself that an allocated page is real
and that accesses to it won't fault.
<p>
What I got was an argument with Alan Cox who persistently doesn't
understand what I'm talking about. He has appeared to believe that
I'm trying to get good behavior when the system is out of memory and
good behavior is impossible, that I want the host kernel to allocate
memory as soon as the address space is allocated, and a bunch of other
things that I don't begin to fathom. Peter Anvin hopped in briefly
with what look like similar problems.
<p>
So, I implemented what I wanted, and sent the patch in. Hopefully
that will clear out the confusion.
<p>
In actual development news, I'm converting UML from storing register
information in pt_regs structs to storing it in sigcontext structs.
What I'm doing is making UML return to userspace by way of host signal
returns rather then having the tracing thread teleport it back with
PTRACE_SETREGS. The main advantage is that it is guaranteed to
correctly restore floating point state, which has worried me for quite
a while. It has other advantages, like removing code from the tracing
thread, which will make it easier to finally eliminate at some point.
</blockquote>
<a name="3-1-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Mar 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I finished the release of 2.4.18 today with the bare kernel and the
RPM. No deb since Matt is taking care of that for me now.
<p>
I cleaned up the test driver some more. The local configuration
information is now stored in ~/.umltest, and it is possible to pass in
configuration options on the command line. This is in preparation for
automatically running the tests whenever a new patch is ready.
</blockquote>
<a name="2-25-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Feb 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released the 2.4.18 UML patch today. It was a no-effort update,
except that there were a bunch of placeholder entries for
the extended attribute system calls.
<p>
I'm going to sit on the full release for a few days while I work up a
new test suite and harness. Hopefully, I can get something automated
set up so that whenever I release a patch, it will get banged on
without me having to do it by hand.
</blockquote>
<a name="2-24-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">24 Feb 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I think I'm emerging from signal delivery hell. Everything seems to
work again with the exception of a memory corruption problem which may
not be new. I've discovered new and interesting ways to screw up.
<p>
It turns out that when you leave -ERESTARTSYS or -ERESTARTNOHAND in %eax
when leaving a system call, the host kernel will helpfully subtract 2
from eip. This is because a system call instruction is 2 bytes long
and it restarts them by executing that instruction over again. This
turns out to be a problem when UML itself is restarting one of its own
system calls. It delivers the signal first, so eip points to the
beginning of the handler, and if you don't put a zero or something in
eip, then you will fake the host into subtracting two from eip. That
puts it at the very end of the previous procedure, which will try to
return from a stack frame that never really existed in the first
place. This leads to very interesting debugging sessions.
<p>
It also turns out that ptrace "knows" that when it has intercepted the
start of a system call, eax contains -ENOSYS. It depends on this.
When I changed how UML gets the process state before a system call, I
changed the value in eax to something else, and ptrace stopped
working. It took a couple of days to figure this out.
<p>
But things seem back to normal now. My signal delivery exercisers
work OK, my stress tests work OK, UML works when compiled on 2.4 and
run on 2.2, and vice versa, which -13 didn't. So, I'm releasing -14
now and maybe this signal delivery rewrite will be done.
</blockquote>
<a name="2-17-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">17 Feb 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I submitted a paper proposal for OLS yesterday. This time, I decided
to rant and rave about how the host kernel needs to be fixed to better
support virtual machines. We'll see how that goes. At least it's
different from the standard UML song and dance I've been flogging for
the last couple of years.
<p>
I redid the UML signal delivery code in order to fix the pthreads hang
with the newer pthreads library. I prototyped this in a small
standalone process in Incheon Airport (Seoul's international airport)
on my way back from Brisbane and just got around to integrating it
into UML. It massively cleaned up a bunch of code and opens the way
for getting rid of some other problems that have existed for a while.
</blockquote>
<a name="2-11-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">11 Feb 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
My flight leaves tonight, so I spent the day wandering around central
Brisbane. Ben LaHaise happened to take the same CityCat ferry (which
is a system of catamarans used to ferry people up and down the
Brisbane River) as me down to the city. He tried to blow up the boat
with his umbrella, but, fortunately, an alert crew member stopped
him.
<p>
Back to the Uni at the end of the afternoon, collect my bags, call a
taxi to the airport, and wait for the flight to Seoul. Then it's
another 14 hours to JFK and another couple back to CT and then I will
be hopelessly confused about what time it is for a couple of days.
</blockquote>
<a name="2-10-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">10 Feb 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
LCA is over as of yesterday, and I'm heading home tomorrow.
<p>
The slides from my talk (with notes) are available
<a href="slides/lca2002/lca2002.htm">here</a>.
It was pretty well attended and it seemed to be received fairly well.
LCA always seems to do some innovative things, one of which is to
rerun talks that a large number of people regretted missing. The
three talks that were chosen this year were a virtual reality talk
that a huge number of people wanted to see, Andrew van der Stock's
talk on code auditing, and one other that I can't remember. This was
good because I was one of the huge number of people that wanted to
see the virtual reality talk. However, it turned out that my talk was
the number four vote-getter, and this turned out to be relevant when
Andrew was nowhere to be seen when the reruns were about to happen.
<p>
So, I was happily debugging the ppc UML build with Anton Blanchard
when one of the organizers ran up and asked me if I could do mine
again. I did so, except I skipped over some of the heavier parts of
the talk to leave a good bit of time for a demo at the end.
<p>
This went well, except for the panics I got when I tried to have two
UMLs mount the same filesystem. I demoed three UMLs running (one
Debian, two Slackwares) with most of them (plus the host, I think)
displaying on the X server of one of the Slackware UMLs. I showed
various other aspects of UML like what it looks like from the host
side.
</blockquote>
<a name="2-1-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Feb 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I'm off to Australia tomorrow for
<link target="http://linux.conf.au" type="external">LCA 2002</link>.
This caps a fairly productive week of UML bug hunting:
<ul>
<li>
mistral and blinky started seeing a panic in fork. mistral figured
out how to reproduce it and tracked it down to the point where it had
something to do with kernel threads reference counting their mm's.
This was enough information for me to fix the bug, by giving kernel
threads NULL mm's.
</li>
<li>
While I was at ISTS giving a talk, I spent some free time looking at
the pthreads problem that people have been seeing for a while. It
turns out to be a problem with UML signal delivery. I had assumed
that the registers going into a signal handler didn't matter (except
for the IP and SP, of course) and that only the stack frame mattered.
So, the process registers at the start of a signal delivery are
initialized with a set that was captured from a UML thread at boot
time. This works fine usually, except that recent pthreads libraries
store some thread-private data in %gs, and the %gs value has to be
preserved in signal handlers. So, the UML signal delivery mechanism
needs to be reworked again.
</li>
<li>
The UML IO hangs that were reported this week were tracked down and
found to be a bug in the host's handling of SIGIO. It turns out to be
possible, on an SMP host, for SIGIO to be queued to a process after
that process has returned from the fcntl that registered a different
process as the SIGIO recipient. This breaks UML badly because SIGIOs
end up queued, but not delivered, to a process which is out of context
and sleeping.
</li>
</ul>
</blockquote>
<a name="1-30-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">30 Jan 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I gave a talk at ISTS yesterday on the UML security work that I did
last week. It's available both as
<a href="slides/ists2002/umlsec.htm">html-ized
slides</a> and as the
<a href="slides/ists2002/umlsec.sdd">original Star
Office presentation</a>. These are intended to provide a starting
point for anyone wanting to probe this for exploitable holes as well
as anyone who's curious about what was done.
<p>
I released 2.4.17-10 today. It contains a pile of bug fixes and a
bunch of changes which allow a UML patch to come close to compiling in
both 2.4 and 2.5 pools. I reverted a change which is causing problems
on SMP hosts. I decided that using sockets rather than pts devices to
communicate between the IO thread and UML was a good idea because
sockets are lighter weight and they're pretty much guaranteed to be
supported on the host, whereas there are lots of systems without pts
devices. However, there is some difference in how SIGIO is delivered
which causes UML to lose interrupts once in a while. The effect is
that it seems to hang on boot, but can be made to continue by banging
on the keyboard.
<p>
MTD is in the configuration now. So, UML supports MTD devices and
creating JFFS2 filesystems and mounting them seems to work, although
there are some nasty-looking error messages along the way. They don't
seem obviously related to UML though.
</blockquote>
<a name="1-25-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Jan 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
The security work is largely done. The exception is the lcall
prevention fix that's needed on the host. So, I released 2.4.17-9
today. It also contains a number of fixes and patches from other
people, the largest being the latest set of James McMechan's ubd
changes.
</blockquote>
<a name="1-23-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">23 Jan 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released 2.4.17-8 yesterday (and 2.4.17-7 earlier this week without
dignifying it with a diary entry). I spent a fair amount of time
tracking down some old debugging problems. The strace recipe on the
<a href="debugging.html">debugging page</a>
hasn't worked for a while, so I mostly fixed it. strace still doesn't
see system calls from new processes until they receive a signal.
<p>
I also figured out what was happening with using the gdb under ddd as an
external debugger. ddd periodically calls wait on gdb, and when UML
attaches it, gdb gets reparented away from ddd. wait starts returning
ECHILD, and ddd reacts by shutting down gdb's input, and gdb, in turn,
exits. To fix this, I think I'm going to have to have a 'gdb-parent='
switch that will make UML attach to the parent and fake normal return
values from wait.
<p>
In other news, there have been a couple of articles about UML
recently. NewsForge ran
<a href="http://www.newsforge.com/article.pl?sid=02/01/18/2025249">one</a>
yesterday. This is a followup on the article last week about Linux
virtual machines which completely failed to mention UML. Bill Stearns
also noticed <a href="http://www.securitywriters.org/texts/internet%20security/Virtual_HoneyNet.php">
this article</a> on using UML as the basis of a honeynet.
<p>
I've started finishing off the security work needed to make UML a
secure root jail. The 'jail' switch now checks for config options
which would make the UML inherently insecure and refuses to run if any
of them are enabled. Currently, the proscribed options are
CONFIG_MODULES, CONFIG_HOSTFS, and CONFIG_SMP. CONFIG_MODULES is
fairly obvious. If modules are enabled, then root can insert any code
at all into UML, and a nasty root would insert code that execs a shell
or something on the host. CONFIG_HOSTFS is forbidden to prevent
accidentally providing access to the host filesystem. This is a bit
dubious because hostfs is not inherently insecure, and I may relax
this one at some point. CONFIG_SMP is non-obvious. Lennert Buytenhek
noticed the relationship between SMP and security. 'jail' is
implemented by unprotecting kernel memory (by making it writable) on
entering the UML kernel, and write-protecting it on kernel exit. If a
process were to have two threads, one busy-waiting in userspace, and
the other sleeping in the kernel, kernel memory would be writable
because the sleeping thread would be in the kernel. So, the spinning
thread would wait for that to happen and write on whatever part of
kernel memory would let it escape. This will be fixed when UML gets a
separate address space for the kernel.
<p>
I've stared at /proc and /dev to find devices that provide access to
kernel memory. The only two that I spotted were /dev/mem and
/dev/kmem. These are disabled with a trick that someone on
#kernelnewbies told me about. Access to them is controlled by
CAP_SYS_RAWIO, so removing that capability from the bounding
capability set makes it impossible for any process to ever get it.
So, no process can ever open those devices. Even better, in my
limited testing, nothing seems to break badly as a result.
<p>
/proc/kcore looks suspicious, but it's a readonly file that seems to
fit out a memory image wrapped in an ELF header, so it's OK,
security-wise.
</blockquote>
<a name="1-18-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 Jan 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
2.4.17-6 is out. There are a lot of driver cleanups and bug fixes in
this patch. The IRQ hang is fixed. The default console and serial
line channel initialization strings are now configurable.
<p>
In other news, NewsForge ran an
<a href="http://www.newsforge.com/article.pl?sid=02/01/16/1930207">article
</a> about Linux virtual machines which covered everything
relevant, including some things that weren't virtual machines, except
for UML. They got some complaints about that, and later that day, I
got a piece of email from the guy who wrote the article wanting to
write a followup about UML. So, it looks like UML will be getting
another nice bit of publicity.
</blockquote>
<a name="1-13-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">13 Jan 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I made another patch against the 2.5.2-pre tree again today. This one
is against 2.5.2-pre11. It has been sent off to Linus so he can drop
it in his bit bucket.
<p>
In another development, it turns out that the O(1) scheduler breaks
UML by holding IRQs disabled across context switches. This results in
SIGIO (i.e. from disk IO completions) to be trapped in a process that
has gone out of context, and can't be woken up until something else
notices that the IO has completed, and of course it won't because the
SIGIO has been delivered to the wrong process.
</blockquote>
<a name="1-11-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">11 Jan 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
After spending five days tracking down a swap corruption bug, I
discovered that Rodrigo de Castro had explained to me exactly what it
was about a month ago. Unfortunately, at the time, I didn't know
enough about the swap code to decide whether he was making sense. Of
course he was, and I discovered that at the end of the great bug hunt.
<p>
So, with that fix, Lennert's latest SMP changes, and a bunch of
smaller stuff, I'm releasing 2.4.17-5 today.
<p>
Linus saw fit to silently drop UML into the bit bucket again, so I'll
make another patch soon and send it in.
</blockquote>
<a name="1-5-2002"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">5 Jan 2002</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released the 2.4.17-3 and 2.4.17-4 patches this week. The biggest
change has been the merging of Lennert's SMP fixes.
<p>
I made another attempt to get UML into the Linus tree. This patch is
against 2.5.2-pre9 and is the 2.4.17-4 patch. We'll see how well this
attempt fares.
</blockquote>
<a name="12-30-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">30 Dec 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I announced the full 2.4.17 release and the 2.4.17-2 patch today. The
patch is largely changes to allow UML to calculate current from the
stack. This is to make life easier for Lennert, who's trying to get
SMP working. It also contains a bunch of fixes for bugs that crop up
when host devices get closed from under UML consoles or serial lines.
<p>
Iain Young is making a decent stab at a UML/sparc64 port. He's got
the boilerplate filled in (albeit with some skeptical comments about
their correctness...) and he's trying to get the whole thing to compile.
<p>
Linus released 2.5.2-pre4 today with no sign of UML in the changelog.
I grabbed the patch to make sure it wasn't there. It wasn't. Grrr.
I'll give him another patch and then I'll spam him with it again.
</blockquote>
<a name="12-28-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">28 Dec 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
OK, so the diary has taken a bit of a holiday break. I released the
2.4.17 UML patch yesterday. The full release will be forthcoming. I
also ported that patch into 2.5.1, created the 2.5.1 patch, and sent
it to Linus. Hopefully, he will put it in without my having to resend
it too many times. I sent a little note off to LKML announcing this,
which got some favorable reaction, both on and off the list. Alan,
being his usual taciturn self sent a reply, which read, in its
entirety, "Cool".
<p>
This release had a lot of accumulated stuff in it. The biggest item
are the port channel, which let you attach any number of UML consoles
and serial lines to a host port, at which point you can access them by
telnetting to that port. I also redid the context switching mechanism
after thinking of a much simpler way of doing it. This should also be
much faster since it doesn't involve signals flying around, so context
switches are now invisible to the tracing thread.
</blockquote>
<a name="12-08-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">8 Dec 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released 2.4.15-3 today. It contains some fixes to previous patches
and a lot of changes to the gdb support. gdb now sees ^C immediately,
rather than an arbitrary amount of time after it's typed. I also
cleaned up that code quite a bit. This knocks a couple of items off
the <a href="todo.html">todo list</a>. It also
sets me up to fix the gdb shell hang, but I'll let these changes gel
for a bit before dealing with that.
</blockquote>
<a name="12-07-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">7 Dec 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I got the <a href="sdotm.html">Sysadmin Disaster of
the Month contest</a> going about a week
later than I should have. The thing that happened a week earlier was
the publication of an <a href="http://www.oreillynet.com/pub/a/linux/2001/11/29/UserModeLinux.html">
article</a> that I wrote for O'Reilly on using UML to simulate and
recover from disasters.
<p>
This month's disaster is a trashed root superblock. It involves
booting UML, zeroing out the superblock, and figuring out how to fix
the filesystem. I had some trouble coming up with a good example to
use for the contest. So, if I have similar troubles at the end of the
month, I might just trash a filesystem, make it available for
download, and the contest will be to figure out what's wrong with it
and fix it.
</blockquote>
<a name="12-04-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">4 Dec 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Back from Linux-Kongress. Back on US/Eastern. I think I never left
it, which made life (the staying awake part of it) difficult in
Holland.
<p>
As for highlights of the conference, we have:
<ul>
<li>
I and the UML project seem to have some name recognition. Everyone I
talked to seemed to have heard of both me and UML, which is very cool.
</li>
<li>
The talk went reasonably well. I gave the same talk as I did at ALS.
I had forgotten that it was somewhat tailored for ALS (i.e. I tried
to avoid talking about stuff that I had talked about at the previous
ALS), and it would have been somwhat different if I had prepared a new
talk for Linux-Kongress.
</li>
<li>
I met a pile of cool people, like
<ul>
<li>
Lennert Buytenhek (who, in recognition of his contributions to both
projects, was embarassed by both me and Rusty by being asked to stand
up for some audience appreciation, which must be a new Linux-Kongress
record)
</li>
<li>
Roman Zippel (who told me about a couple of ubd driver bugs, one of
which I knew of (a subtle rounding error), and one of which I didn't
(Greg Lonnon and I went to some trouble to put the COW header in
network byte order, but forgot to do the same for the block bitmap
(grrrr))
</li>
<li>
Bruce Walker (who's in charge of the Compaq SSI project, and who asked
me clustering questions during my talk, at which point, I (correctly)
guessed who he worked for and why he was asking)
</li>
<li>
Fabio Olive Leite (a Conectivite who gave me a nice Conectiva
filesystem image a while back, and whose name I unaccented to get it
through my XSL processor)
</li>
<li>
Philipp Reisner (who threatened an Alpha UML port a while back, but
gave up, so he's less cool than the others :-)
</li>
</ul>
</li>
<li>
The organizers took the speakers back to Amsterdam after the
conference to spend the day bumming around the city. We broke up into
small groups and went our separate ways. Our group spent much of our
time in two smallish cafe-type places just talking about random stuff.
</li>
<li>
The train trip from Schiphol airport (Amsterdam) to Enschede (near the
German border) was interesting for a number of people. It was exactly
as described for me (2+ hours, direct train, no problem), but some
track maintenance in Amsterdam's central station caused subsequent
trains to be cancelled, so other people had nightmares involving 5
trains and a bus totalling 5 hours.
</li>
</ul>
</blockquote>
<a name="11-26-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">26 Nov 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released 2.4.14-6 today in the interest of clearing the decks for
the 2.4.15 release. Before leaving for Thanksgiving, I had redone the
mconsole protocol to be packet-oriented. This allows a lot more
flexibility in what can be done with the protocol. As a result,
you'll need the new mconsole client for this release.
<p>
While down in CT, I redid the host channel support. It is all much
cleaner now, and makes it a lot simpler to knock a bunch of
related items off the todo list.
<p>
I decided to knock off the 2.4.15 patch today as well. It went cleanly,
aside from a ptrace cleanup and a new way of generating /proc/cpuinfo
which I had to support. I also put in the file corruption fix after
forgetting to and discovering that a boot/halt caused fsck to complain
a little.
</blockquote>
<a name="11-18-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 Nov 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
While I'm waiting for the meteors to arrive, I'm chasing and stomping
UML bugs. I cleaned up and released the proxy arp fixes that I did on
planes and in airports on my way to Oakland. Before, uml_net would
blindly add an arp entry to eth0 and nothing else. This is wrong if
there is no eth0, and it's also wrong if eth0 doesn't connect to the
local net or if there are other interfaces also attached to the local
net. uml_net now looks at the routing table and puts an arp entry on
every interface that talks to the local net.
<p>
I also noticed that slip support wasn't up to date, so I modernized it
and cleaned up the code while I was at it. You can now change the IP
address of a slip-based interface and the host configuration will be
updated just like the other transports.
<p>
I added some RT signal support. SA_SIGINFO is now supported, which
will hopefully fix some of the strange process behaviors that have
cropped up lately. If this fix doesn't do it, I chased down another
bug which was causing rt_sigsuspend and sigsuspend to return incorrect
values. This was causing the libc sigsuspend to hang, and its process
with it. This fixes the pthread_create hang that Greg Lonnon noticed,
plus the gdb hang, I think. I haven't checked that yet.
<p>
Those fixes are in 2.4.14-3um which I just released. You'll need the
latest utilities in order to use the network, since I bumped the
uml_net version again.
</blockquote>
<a name="11-14-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">14 Nov 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
OK, I'm back from ALS. My talk was on the first day, and it was
reasonably well attended considering the somewhat dismal overall
number of attendees. It was a half-hour talk, so about an hour
beforehand, I took my OLS slides, threw out more than half of them,
and updated the rest. That worked out reasonably well, but 30 minutes
makes for a very short talk without much detail.
<p>
Daniel Phillip's talk was the last of the conference, and was somewhat
interesting. He had a pile of raw data that he needed to turn into
slides, and all of the KDE presentation tools blew up on him in one
way or another. So, in the break before his slot, he grabbed Stephen
Tweedie, and they plus me and another guy went off to a local dim-sum
place. Daniel and Stephen sweated over Star Office on Stephen's
laptop making slides. In the event, they turned out rather well.
<p>
I left just after Daniel's talk, so I missed out on some of the
socializing that afternoon and evening. It turned out that Daniel and
Larry McVoy were talking about his clustering ideas (MetaLinux, or
ML), and it occurred to them that UML was not only a good simulation
tool for ML, but that it actually implements a good part of what Larry
has in mind. I found out about this later, and had a long talk with
Larry on Monday, in which he explained his plans. I had heard various
mumblings about it, and saw a
<a href="http://bitmover.com/ml">slide show</a>
that Larry has, and remained unenlightened. It turns out, that as far
as I can tell, the only way to find out what he's thinking is to have
him explain it in person. Anyway, I became enlightened after our
chat, and it looks like this could be a whole new area that UML could
branch into.
<p>
In actual development news, I fully released 2.4.14 today. En route
to Oakland, I fixed uml_net so it's smarter about doing proxy arp. It
figures out what devices are connected to the local net and only sets
proxy arp on those. As a side-effect, if the host is totally
isolated, then you don't get scary-looking error messages when it
tries to set proxy arp on eth0 and it turns out not to exist.
<p>
This happened to me at OLS when I tried to demo it after my talk. I got
this nasty message which convinced me that the network all of a sudden
didn't work, and I was all apologetic and had no idea what happened.
In reality, the network was fine, and I could have demoed it if I had
retained a bit more presence of mind.
<p>
This isn't in the 2.4.14 release because I'm not happy about the
cleanliness of the change. I'll probably clean it up for the next
2.4.14 patch.
<p>
On a much-delayed train from San Francisco to Mountain View (a
supposedly 1:13 hour trip that in reality required about 1:50 and two
trains), I also figured out why you can't talk to eth1 from the host
if you configure both an eth0 and eth1. It turned out to be the same
bug that other people had noticed causing dropped packets. I was
checking errno incorrectly. I had code that did this:
<tt>
<pre>
<font size="-1">
n = read(...);
if(errno == EAGAIN) return(0);
</font>
</pre>
</tt>
forgetting that successful system calls don't necessarily set errno to
zero. So, the eth1 read was succeeding, but errno was still EAGAIN
from the eth0 read.
<p>
In other news, beware of kernels built with gcc 3.0.2. I got a
complaint from Jens Axboe today about UML leaving all kinds of
not-quite-zombie processes lying around. I looked at it a bit and
guessed that the host kernel was messed up somehow. He looked at
that, decided I was right, and that the culprit was the latest gcc.
The interesting thing was that, until he ran UML on that kernel, it
looked just fine to him.
</blockquote>
<a name="11-06-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">6 Nov 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
In preparation for fixing the problem of the console driver losing
output, I ported the SIGIO handler to use poll instead of select.
This was mostly what 2.4.13-4 was. I later discovered a bug in it,
which is fixed in -5.
<p>
I then decided to fix the problem of UML not being able to be
interrupted and backgrounded. The problem was that all UML processes
are in the same process group, with all of them stopped except for the
one that's actually running. The problem is that when UML is
backgrounded, the shell sends a SIGCONT to the process group, which
wakes up every UML process, which is very bad.
<p>
I did some failed experiments with setpgrp/setsid and friends, and
discovered that a separate process group wouldn't work because then
those threads can't write to the terminal because they're in the wrong
process group.
<p>
So, I decided that out-of-context processes should be asleep rather
than stopped. This required redoing the task switching code. They
were stopped because the tracing thread intercepted a signal from them
when they went out of context and never continued them. Having them
sleep would require that the tracing thread stop doing that and that
the threads involved in a context switch arrange the transfer
themselves.
<p>
So, what is now done is that non-running processes are asleep in
sigsuspend, and they are woken up by the going-out-of-context process
sending a SIGTERM. Races are avoided by having the SIGTERM sent
inside a section of code that has blocked SIGTERM. SIGTERM is
re-enabled atomically with the sleep with sigsuspend.
<p>
So, that plus the poll fix is the contents of -5.
</blockquote>
<a name="11-03-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">3 Nov 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Time to patch-bomb Alan again. I sent in ten patches to get the ac
tree current with CVS. Here they are:
<ul>
<li name="2.4.13/misc-4">
<a href="ac-patches/2.4.13/misc-4">Miscellaneous fixes</a> - some build cleanup,
a config update, some name changes in the mconsole driver, exporting
of gprof symbols, and various other small cleanups and fixes
</li>
<li name="2.4.13/physmem">
A <a href="ac-patches/2.4.13/physmem">name change</a> from physmem
to uml_physmem
</li>
<li name="2.4.13/con">
A <a href="ac-patches/2.4.13/con">console driver fix</a> caused by
a tty struct leak
</li>
<li name="2.4.13/init">
Adam Heath's <a href="ac-patches/2.4.13/init">init changes</a> and
other cleanups
</li>
<li name="2.4.13/initrd">
A <a href="ac-patches/2.4.13/initrd">initrd cleanup</a>
</li>
<li name="2.4.13/net">
<a href="ac-patches/2.4.13/net">Update to the network backends</a> to make
them grab output from uml_net and printk it cleanly
</li>
<li name="2.4.13/ppc">
Some <a href="ac-patches/2.4.13/ppc">ppc updates</a> and fixes from Chris
</li>
<li name="2.4.13/sig">
A <a href="ac-patches/2.4.13/sig">signal handling fix</a> which eliminates
most of the process segfaults which people had been seeing
</li>
<li name="2.4.13/vm">
Jorgen Cederlof's <a href="ac-patches/2.4.13/vm">context switch speedup</a> ,
which also includes some VM fixes that I found
</li>
<li name="2.4.13/sync">
A <a href="ac-patches/2.4.13/sync">sychronization patch</a> which
includes a grab-bag of changes which I hadn't managed to get into the
ac tree yet
</li>
</ul>
</blockquote>
<a name="11-02-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">2 Nov 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
That last patch went into -ac6, so the ac UML builds and works again.
The next job is to get the ac tree up to date.
<p>
I released a new utilities tarball today. uml_net should now do proxy
arp correctly. uml_mconsole is now able to take a command on its
command line and execute it, rather than being strictly a command line
tool.
</blockquote>
<a name="10-30-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">30 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.13/build">
I decided to make the -ac UML build again, so I made
<a href="ac-patches/2.4.13/build">this patch</a> and sent it off to
Alan. The rest of the updates will be forthcoming.
</blockquote>
<a name="10-29-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">29 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Today is 2.4.14-3 day. I decided to remove the code in fix_range
which unmaps pages whose ptes say they're not present. That basically
caused it to try to uselessly unmap all of its unused address space.
So, I did that and it uncovered a bug. It turns out that swapped-out
pages weren't marked as needing to be remapped. Everything worked a
lot better with that fixed, and context switching should be a bit
faster now.
</blockquote>
<a name="10-28-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">28 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released 2.4.14-2 today. This contains the fix for the process
segfaults and the gdb problems people have been having. It also turns
on morlock's context switch optimization which I disabled until I
figured out the segfaults.
</blockquote>
<a name="10-26-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">26 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I finished releasing 2.4.13 today.
<p>
After some prodding from Greg Lonnon, and after he did some investigation,
I figured out what the problem with gdb inside UML is. The signal
handlers don't save their registers in the thread struct. This means
that when a SIGTRAP from a breakpoint comes in and it gets forwarded
to gdb, when it gets the registers to find out what the ip is, it gets
an old, bogus value. So, it doesn't recognize that as a breakpoint
and complains about a spurious SIGTRAP instead.
</blockquote>
<a name="10-25-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I spent the last few days chasing a process segfault problem. I
finally tracked it down today. It turns out that my rewrite of the
process signal delivery code was broken in the case of a signal being
delivered from an interrupt handler rather than a system call. It
grabs the process registers from the thread structure, saves them away
on the stack, and then restores them to the process when the handler
finishes.
<p>
However, interrupts don't save their registers in the thread
structure, so those registers represent the last system call, which
has already finished. And restoring those causes great confusion in
the process.
</blockquote>
<a name="10-20-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">20 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released 2.4.12-3um and 2.4.12-4um over the last week. -3 fixed a
couple of problems with -2, and -4 adds some miscellaneous fixes to
that. The major ones are that physical memory protection is optional
(controlled by the 'jail' switch) and that the network driver backends
now collect uml_net commands and output and nicely printk them instead
of having the output just dumped to the terminal. To support this,
uml_net now hangs on to the commands it runs and the output they
produce and send them back to UML. This required that the uml_net
interface be incremented, so it's now at 3. The new drivers require
the new uml_net, so if you grab the UML patch, also get the latest
utilities tarball too.
</blockquote>
<a name="10-13-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">13 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released 2.4.12-2um today. It's almost entirely changes sent in by
other people, dominated by Adam Heath's cleanups. There were also some
ppc fixes from Chris Emerson, and small fixes from other people.
<p>
I also released a new utilities tarball. The one change was to
uml_net, which does proxy arp in a different, and apparently more
robust way than it used to.
</blockquote>
<a name="10-11-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">11 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Linus released 2.4.11 and 2.4.12 two days apart. I had the 2.4.11
patch uploaded, and had started releasing packages when 2.4.12 came
out. So, 2.4.12 is out there and I'm doing the packages again.
</blockquote>
<a name="10-08-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">8 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I should have mentioned the latest -ac patches already since they've
been in Alan's tree for a few days, but I didn't, so here they are
<ul>
<li name="2.4.10/mmapper">
<a href="ac-patches/2.4.10/mmapper">an update</a> to the mmapper driver
</li>
<li name="2.4.10/misc-2">
yet another set of <a href="ac-patches/2.4.10/misc-2">miscellaneous fixes
</a>
</li>
<li name="2.4.10/addr">
a <a href="ac-patches/2.4.10/addr">reorganization</a> of the UML
address space
</li>
<li name="2.4.10/sigs">
a <a href="ac-patches/2.4.10/sigs">rewrite</a> of the signal
delivery mechanism to allow all of kernel physical memory to be
protected from userspace
</li>
<li name="2.4.10/ubd">
<a href="ac-patches/2.4.10/ubd">version 2</a> of the ubd COW header
</li>
</ul>
In other news, with the help of Paul, I tracked down an ancient
console driver bug that held on to a struct tty after it had been
freed and subsequently caused panics.
<p>
I released 2.4.10-7um today with that fix and some other minor changes.
</blockquote>
<a name="10-05-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">5 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Paul Larson found a test case for the signal problems that was
reproducable for me. So, with that in hand, I tracked down the bug,
and released 2.4.10-6um.
<p>
The bug turned out to be a result of moving where state is saved
before a signal is delivered to a process. The process registers and
some other things need to be saved on the process stack so they can be
restored later. The way it used to work is that
<ul>
<li>
handle_signal would figure out what the interrupted system call
eventually returns
</li>
<li>
that value is passed up the stack and stored in the process registers
stored in its task structure
</li>
<li>
the process would be sent a signal so it starts running on its process stack
</li>
<li>
the UML signal handler copies the register state from the task
structure to its own stack
</li>
<li>
it calls the process signal handler
</li>
<li>
and restores the registers back to the task struct
</li>
</ul>
What I implemented did this
<ul>
<li>
handle_signal figures out what the interrupted system call eventually
returns and constructs the process stack frame, copying the registers
from the task struct onto the stack
</li>
<li>
that value is passed up the stack and stored in the process registers
stored in its task structure
</li>
</ul>
The bug is that the second step happened too late. The registers
saved on the stack hold a bogus return value, and it's that value
which the system call eventually returns.
</blockquote>
<a name="10-03-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">3 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I decided to profile a stretch of UML thrashing. So, I took the
2.4.10-2ac UML (which I updated to the latest stuff, and which I'll be
sending to Alan shortly), gave it 128M of memory and 1G of swap, and
let a 'make -j' kernel build run for a couple of hours.
<a href="text/gprof1.out">These</a> are the
results. All of the system calls show up as <spontaneous>.
Somehow it wasn't linked against the profiling libc. I'll try to
figure out why not.
<p>
Some highlights:
<ul>
<li>
Protecting kernel memory from userspace seems to be expensive -
mprotect is the top item on the list.
</li>
<li>
wait4 is number two, which I don't entirely understand. That's the
tracing thread. It sleeps in wait, and wakes up when there's
something that needs doing, so I don't understand why it shows up,
unless it's somehow being charged for all of the context switching
that UML causes.
</li>
<li>
Then we have other low-level VM things, fix_range and
flush_tlb_kernel_vm. These manually walk address spaces to update
them. These two are unnecessarily inefficient and can probably be
knocked far down the list pretty easily.
</li>
<li>
Finally, we get into generic kernel things, which show clear signs of
heavy swapping - page_launder, swap_out_pmd, do_anonymous_page.
</li>
<li>
The first system call which shows up is sys_brk, way down the list,
followed distantly by sys_read and sys_close.
</li>
<li>
There were 312175 system calls total. The most frequently called were
sys_brk, sys_read, sys_open, sys_newfstat, and sys_stat64, not
unexpected for a kernel build.
</li>
<li>
kmalloc was called most often from load_elf_binary, select_bits_alloc,
and load_elf_interp. __get_free_pages was called most often from
handle_mm_fault, pipe_poll, and do_fork. It called _alloc_pages,
which was called most frequently from read_swap_cache_async,
do_anonymous_page, and do_wp_page.
</li>
</ul>
</blockquote>
<a name="10-01-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Oct 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.10/build-2">
I discovered a new way of breaking UML. A 'make -j' kernel build
drives the load above 150, and on 2.4.10 causes essentially a
livelock. I eventually regained control by sending SIGILL to all the
processes from the host. Plus, I got all kinds of interesting illegal
instruction and bus error deaths. These were absent on -ac2, probably
because that wasn't a totally up-to-date UML, so it was missing the
most recent bugs that I added. I'm going to track those bugs down by
updating UML in the -ac tree bit by bit and seeing which bit causes
these nasty little problems.
<p>
Speaking of -ac2, it needs a <a href="ac-patches/2.4.10/build-2">little
fix to ptrace</a> in order to build.
</blockquote>
<a name="09-25-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Today I redid the signal delivery code. Now all the saving and
restoring of state happens in kernelspace rather than on the process
stack like before. This allows the task structure to be protected
from processes. Since that was the only hole in the protection of
physical memory, that is now fully protected against being changed
from userspace.
</blockquote>
<a name="09-24-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">24 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I made a .deb and an RPM in preparation for releasing 2.4.10, and
Jacques Nilo reported that yesterday's fix wasn't enough. I had
forgot about an instance of MAP_SHARED | MAP_ANONYMOUS. So, I fixed
that, and that is 2.4.10-3um. And that is the basis of the official
2.4.10 UML release.
</blockquote>
<a name="09-24-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">24 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I thought of an easy fix for the stack capturing problem that
prevented UML from booting on 2.2 hosts. Basically, a new process is
created which stops itself, and when that happens, the parent grabs a
copy of the stack and uses it to create a context for future threads
to run in. On 2.2, the parent used ptrace to extract the contents of
the stack from the child word by word. I looked at that code and
decided it would be much easier to map the stack MAP_SHARED so it
would be shared between parent and child and the parent could just
memcpy it to a safe place rather than ptracing it out.
<p>
What I forgot was that, while 2.4 supports MAP_SHARED | MAP_ANONYMOUS,
2.2 doesn't. So, on 2.2 hosts, UML wouldn't even begin to boot.
<p>
The easy solution was to go back to MAP_PRIVATE | MAP_ANONYMOUS, but
clone the new process with CLONE_VM, making it a thread, which allows
the parent to copy the stack directly, since they're both in the same
address space.
<p>
This fix makes 2.4.10 usable, so I've released another patch and
updated CVS.
</blockquote>
<a name="09-23-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">23 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Linux released 2.4.10 today, so I updated UML as well. I decided not
to base this on the latest UML patch, since that it not entirely
healthy at the moment. My sigaltstack fixes broke UML totally on 2.2
hosts. So, 2.4.10-1um is 2.3.9-8um updated to 2.4.10.
<p>
CVS is not updated, but will be once I have the sigaltstack thing
fixed and that pool updated to 2.4.10.
</blockquote>
<a name="09-22-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">22 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
The last set of -ac patches went into -ac14. That brings Alan's tree
reasonably up-to-date.
<p>
I released 2.4.9-8um yesterday and 2.4.9-9um today. -8 was some bug
fixes and cleanup. -9 was fixing sigaltstack and doing a lot of cleanup
and rearrangement of the signal delivery code. This sets me up to redo
the entire signal delivery mechanism so I can finish protecting all of the
kernel's physical memory from userspace.
</blockquote>
<a name="09-18-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.9/personality">
That last batch of patches went into -ac12. So, the next batch is off
to him, plus <a href="ac-patches/2.4.9/personality">one from Andrea
Arcangeli</a> which fixes a declaration which is needed to
compile UML successfully.
<p>
Once these are in, the -ac tree is almost up to date. It'll be one
CVS release behind, which is OK because there are some tweaks I want
to make to the address space reorg. So, I'll get that right and send
it in rather than sending it in two pieces.
</blockquote>
<a name="09-15-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">15 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.9/hz">
I released 2.4.9-6um last night. It contains the already-mentioned
COW header changes. It also occurred to me that I can fix the
mlockall bug by sticking UML at the top of the address space where
it's supposed to be anyway. So, I went ahead and did that. This
allowed me to get rid of the vmas that UML needed to stick in each mm
to prevent mmap from reallocating areas of virtual memory that UML is
living in. This, plus the fact that these vmas had no ptes, caused
mlockall to cause major damage to UML by trying to unmap it. Putting
UML above TASK_SIZE causes it to be ignored by mmap, and the problem
just disappears. This also let me get rid of the nasty address space
reservation code that was needed in order to prevent libc from mapping
stuff in where UML wanted to put stuff.
<p>
In other news, I'm back in the ac patch business. First is a
<a href="ac-patches/2.4.9/hz">patch</a> that I've been sitting on
all week which defined hz_to_std and allows UML to build again. Then,
we have
<ul>
<li name="2.4.9/build">
some <a href="ac-patches/2.4.9/build">build cleanups</a>
</li>
<li name="2.4.9/misc">
a couple <a href="ac-patches/2.4.9/misc">miscellaneous fixes</a>
</li>
<li name="2.4.9/hostfs">
a small <a href="ac-patches/2.4.9/hostfs">hostfs patch</a> from
Jorgen Cederlof
</li>
<li name="2.4.9/sysrq">
mistral's <a href="ac-patches/2.4.9/sysrq">sysrq patch</a>
</li>
</ul>
These are now all off to Alan. I've got some more which will wait
till those have gone in, in order to minimize conflicts:
<ul>
<li name="2.4.9/inline">
a bunch of 'extern inline' declarations <a href="ac-patches/2.4.9/inline">
turned</a> into 'static inline'
</li>
<li name="2.4.9/mem">
the first <a href="ac-patches/2.4.9/mem">installment</a> of
protecting the kernel from userspace
</li>
<li name="2.4.9/misc-2">
another <a href="ac-patches/2.4.9/misc-2">miscellaneous fixes patch</a>
</li>
<li name="2.4.9/tuntap">
the <a href="ac-patches/2.4.9/tuntap">TUN/TAP transport</a>
</li>
<li name="2.4.9/misc-3">
yet another <a href="ac-patches/2.4.9/misc-3">miscellaneous fixes patch</a>
</li>
<li name="2.4.9/ppc">
some <a href="ac-patches/2.4.9/ppc">ppc updates</a> from Chris Emerson
</li>
<li name="2.4.9/malloc">
calls to malloc, calloc, and free are now <a href="ac-patches/2.4.9/malloc">
converted</a> into calls to kmalloc and kfree when the slab
allocator is running
</li>
</ul>
</blockquote>
<a name="09-14-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">14 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Greg Lonnon and I have been fiddling with the COW file header format.
I had already discovered that blindly copying the backing file path
provided by the user into the header is a problem when it is a
relative path. That COW file won't be usable by a UML run in a
different level of the directory hierarchy because, from there, the
relative path stored in the header doesn't refer to the backing file.
The fix is to write an absolute pathname into the header.
<p>
Greg had a couple of other good ideas which we thought should be
implemented earlier rather than later
<ul>
<li>
The header should be able to hold a MAXPATHLEN-sized backing file name
rather than the current 256 bytes.
</li>
<li>
It should be in network byte order. This will allow COW files to be
moved between big-endian and little-endian hosts. Whether the
underlying filesystem can be mounted in UML after the move depends on
whether the filesystem has its metadata byte-swapped correctly. But,
at least the COW header won't prevent it from working.
</li>
</ul>
These two are not backward compatible, so we bumped the COW header
version and made these changes in the version 2 header. The driver
can read both V1 and V2 headers but it will only write V2 headers.
<p>
The absolute pathname change is in 2.4.9-5um since it was small and
backward compatible. The other two will be introduced in 2.4.9-6um.
<p>
The <a href="http://lists.sourceforge.net/lists/listinfo/user-mode-linux-user">uml-user list</a> had a couple of
interesting posts from UML users today
<ul>
<li post="6617693"> <a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/6617693">Martin Volf</a> did a Slackware 8.0
installation inside UML and wrote a
<a href="http://mv.web.wo.cz/uml/umlslackware.html">
page</a> describing how he did it.
</li>
<li post="6620124">Tim Robinson had some problems with the TUN/TAP transport and
<a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/6620124">posted</a> a nice diagnosis of them.
</li>
</ul>
</blockquote>
<a name="09-10-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">10 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote post="6593569">
Been playing with the tools and website lately. I added a bunch of
new <a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/6589530">features</a> to the mconsole
client (and <a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/6590956">promptly</a> had to
<a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/6593569">fix</a> it), and fixed uml_net
building on 2.2.
<p>
I also restructured the web site build somewhat to make it more manageable.
</blockquote>
<a name="09-06-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">6 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I tracked down the process segfault problem. It was caused by a newly
forked child inheriting some pages that were swapped out, but hadn't
been unmapped. The code that it ran on its first quantum didn't
update its address space correctly, so those pages remained mapped.
<p>
Having chased that problem down, I'm releasing 2.4.9-4um with that fix
plus Chris Emerson's latest ppc changes.
</blockquote>
<a name="09-01-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Sep 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
After much ado, I revamped the UML
<a href="dl-sf.html">download page</a>. It
essentially replaces the Sourceforge project download page. I did
this in order to be able to let people select the mirror they want to
download from and to be able to put explanatory information on the same
page as the download link. If it is missing stuff that you'd like to
see, regardless of whether it's on the SF download page,
<a href="contacts.html">I'd like to know</a> about it.
<p>
There are a couple things that aren't working right now - the
'Changelog's don't link to anything, and most of the SourceForge root
filesystem links don't work. I'm in the process of copying the
filesystems over to SF to fix this.
<p>
It's now pretty trivial for me to add mirrors, so if you have a box
available (particularly if it's in a part of the world not
well-covered by the UML global mirror system),
<a href="contacts.html">let me know</a>.
</blockquote>
<a name="08-30-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">30 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Thanks to what looks like an all-night debugging session on the part
of Yon Uriarte, the TUN/TAP backend now works. You'll need the latest
uml_net for this. It wasn't setting IFF_NO_PI, which was causing
extra cruft to be stuck on the front of the packet, which probably
required the broken nastiness I had to add to the driver. Adding that
and backing out all the skbuff fiddling made everything work a lot
better.
<p>
So, I released 2.4.9-3um with the fixed driver in it, plus new entries
in config.release, defconfig, and Configure.help.
</blockquote>
<a name="08-28-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">28 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I implemented a TUN/TAP backend for the network driver. It involved
more work than I expected. A lot of it was due to restructuring other
code in order to keep the code relatively clean.
<p>
I haven't done any stessing or timing of it, but I did happen to
notice that pings over TUN/TAP are about 10x faster than pings over
ethertap. The absence of the helper handling each packet on the way
to the kernel is no doubt a big piece of that. At some point, I'll do
some bandwidth measurements against ethertap to see how much better it
is. Hopefully a lot.
</blockquote>
<a name="08-26-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">26 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
UML development took a bit of a break while I got busy with other
stuff.
<p>
In UML news, I started work on my ALS paper, got a first draft
ready, and sent it off for review. I also did a bunch of web site
work. I've been letting things fall behind for lack of time to deal
with them, so I decided to swallow my pride and start asking for
help. This necessarily involves describing what needs doing, so I
wrote most of it up, and the results are
<a href="help-gen.html">here</a>,
<a href="help-doc.html">here</a>,
<a href="help-userspace.html">here</a>,
<a href="help-kernel-v1.html">here</a>, and
<a href="projects.html">here</a>.
<p>
I also made a pass over the site, fixing a bunch of hopelessly
outdated and wrong things, and probably leaving some things which are
only moderately outdated and wrong.
</blockquote>
<a name="08-16-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">16 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I've been having fun playing with crashme. It's a great little tool.
It generates buffers full of random data and then executes them. It
runs differently on UML than on the host, which it shouldn't. The
problems I've tracked down so far are signal handling bugs. UML
wasn't handling write faults correctly when the accessed memory was
readonly, and it wasn't properly segfaulting processes to which
signals couldn't be delivered (because their stack pointers were
garbage). This last was the bug I was chasing a couple days ago.
There are still problems. The first process (crashme +2000 666 100)
runs just as it does on the host, but the next one (crashme +2000 667
100) doesn't. On the host, the segfault handler somehow gets bus
errors in libc, which I don't understand, and that doesn't happen
under UML.
<p>
On IRC yesterday, Lennert Buytenhek clued me in on how to reliably
segfault processes and crash UML. He was running 8 "du /". That
didn't work for me, but 16 of them does. The segfaults are on pages
that are mapped in but shouldn't be (their ptes say that they should
be mapped out, and somehow that didn't happen). So, those pages were
presumably allocated for something else, and contain garbage from the
perspective of the process that should have unmapped them, and so it
segfaults.
<p>
The panic looks like memory corruption. I turned on slab debugging,
and it looks like that makes the panic go away.
<p>
Well, Linus released 2.4.9 today, so it's time for me to go into my
UML release routine. When I did the obligatory kernel build on 2.4.9,
one of the crashme fixes turned out to be bogus. It was doing the
segfault-during-signal-delivery check too early, so it caught fixable
segfaults that happened because the stack needed extending or was
readonly.
</blockquote>
<a name="08-14-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">14 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
2.4.8-2um is out as of yesterday. I made the freshmeat announcement
of 2.4.8 this morning.
<p>
I chased the crashme bug a little. Somehow, a signal is marked as
being pending, but it's never actually delivered and reset. So, no
further signals can be delivered to that process from then on. This
makes it unkillable and unstoppable.
<p>
I also took the first step towards making UML secure against nasty
users. UML physical memory, except for the task structure and kernel
stack, are protected from userspace access. I still need to protect the
task structure and kernel virtual memory. The task structure is a bit
tricky because of the signal delivery code. It runs on the process
stack and is considered to be userspace code. However, it needs to be
able to modify the task structure to restore state that it saved
before the signal delivery. So, if the task structure isn't writable,
this isn't possible. Further thought on the subject is necessary.
</blockquote>
<a name="08-13-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">13 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Well, Linus released 2.4.8 just as I was heading up north for a
weekend of camping and climbing mountains. He does this on purpose.
He released 2.4.3 when I had just arrived in San Jose for the Kernel
Summit.
<p>
Anyway, this was a relatively simple patch. It just dropped in and
worked, except that hostfs was already broken. My calculation of the
stat64 inode field was wrong. It looked at the kernel version to
decide what was in the userspace headers. I discovered the error of
my ways when I booted up a Debian UML to produce the 2.4.8 .deb. This
is a 2.2 filesystem (with .st_ino in stat64) with a 2.4 kernel (which
implied .__st_ino in stat64). hostfs did not build. I changed the
Makefile to just grep the appropriate header instead.
<p>
So, this fix will be the substance of 2.4.8-2um.
</blockquote>
<a name="08-09-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">9 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
The remaining differences between my pool and the ac tree are a couple
of patches that didn't go in for some reason, cleanups of printks and
some includes.
<p>
Daemonizing UML does work. I just checked it, and the only case where
it does something strange is if you background it without nohupping it
and log out. The tracing thread dies from the SIGHUP, but all the
other threads survive.
<p>
I released 2.4.7-5um today. It contains a few recent patches from
other people. I figured out how to turn -fno-common back on. I tried
all kinds of linker tricks to throw errno.o out of the binary. Then I
discovered that the linking that had already taken place had destroyed
any notion of what objects anything originally came from. So,
instead, I added -Derrno=kernel_errno to all the kernelspace gcc
lines, which translates all the kernel uses of errno to kernel_errno,
and leaves the libc errno alone. That's actually a better solution
than throwing out one of the errnos because that would leave open the
possibility that the kernel and userspace uses of libc could step on
each other. Now that they're using different symbols, that's not a problem.
</blockquote>
<a name="08-07-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">7 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Yesterday's patches are off to Alan.
<p>
In other news, daemonizing UML seems to be broken again. Grrr. That
seems to break now and then for no apparent reason.
<p>
ac9 is out with my patches in it. So, time to make the final diff
between Alan's stuff and mine to get him totally caught up with me.
</blockquote>
<a name="08-06-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">6 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Yesterday's patches are in ac8. So, two more patches will bring the
ac tree completely up to date:
<ul>
<li name="2.4.7/net-8">
A <a href="ac-patches/2.4.7/net-8">network driver update</a> which
adds the ability for the drivers to tell the helper about any IP
address changes. This allows the host configuration (routing and
proxy arp) to stay in sync with the interface address changing inside
UML. If you're in the habit of getting UML from the ac tree, you'll
need the latest uml_net in order to use the network when this patch
goes in because it makes an incompatible change in the helper interface.
</li>
<li name="2.4.7/misc-8">
Another batch of (surprise!) <a href="ac-patches/2.4.7/misc-8">miscellaneous
fixes</a> , including some cleanup of stack permission setting, the
apparently gratuitous locals that are needed to pursuade -pg to work
properly, a couple of symbol exports for GFS, a fix that ensures that
the pid file contains the correct pid, and yet another squashed warning.
</li>
</ul>
With these in, I'll be able to diff the ac tree against mine to see
what divergences there are. I know there are some, because I
occasionally see patches fail to apply because of context conflicts
which shouldn't be there. So, there will be one more patch to clean
those up, and Alan will be completely in sync with me.
</blockquote>
<a name="08-05-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">5 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Patch time again. This time I'm making them up ahead of ac7 coming
out. So, we have
<ul>
<li name="2.4.7/hostfs">
A <a href="ac-patches/2.4.7/hostfs">hostfs update</a> , which brings
the ac tree completely up-to-date. Normally, I bundle a couple of cvs
updates into a small number of patches and send them off to Alan.
With hostfs, I decided to give him the latest stuff, since there have
been a bunch of changes spread over a number of cvs updates. This is
fairly easy since hostfs is a completely self-contained piece of code.
</li>
<li name="2.4.7/net">
A <a href="ac-patches/2.4.7/net">network driver update</a> , which
fixes a crash and makes net devices pluggable via the mconsole.
There's some restructuring and cleanup in this patch. Also, mconsole
actions move into keventd context from softirq context. This is
because alloc_netdevice does a GFP_KERNEL kmalloc, which has to be
done in process context.
</li>
<li name="2.4.7/misc-7">
Yet another <a href="ac-patches/2.4.7/misc-7">batch of miscellaneous
fixes</a> , including renaming CONFIG_IOMEM to CONFIG_MMAPPER,
some cleanup in the ubd driver, and removal of a number of warnings.
</li>
<li name="2.4.7/ppc">
The <a href="ac-patches/2.4.7/ppc">complete merge of the ppc
port</a> , which reorganizes the headers somewhat. For some
headers, there are now header.h, which is a symlink to
header-$(SUBARCH).h, which includes header-generic.h and is allowed to
do whatever it wants before and after. This provides the flexibility
needed to do things like undef stuff after the include and rename
things beforehand.
</li>
</ul>
These all will bring Alan up to my 2.4.7 release, except for hostfs,
which will be completely up to date. Since I'm up to 2.4.7-4um, and
2.4.7-2um was just a hostfs fix, I might be able to bring the ac tree
up to date with one more set of patches.
<p>
Alan released ac7 this afternoon, as I prophesied, so those patches
are off to him. I'll be looking for them in ac8.
<p>
I failed to resist temptation. I looked at the diffs between the ac
tree once those patches are in and my current stuff and I noticed a
big wad of documentation. So, I rolled that up and sent it to Alan.
</blockquote>
<a name="08-04-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">4 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.7/build-5">
OK, I'm back in the business of sending Alan patches. I sent in a
<a href="ac-patches/2.4.7/build-5">small patch</a> which fixes the
things that broke when 2.4.7 came out. So, UML now builds and works in
the -ac tree again. It made 2.4.7-ac6 an hour or so after I sent it over.
<p>
Also, in the interest of getting the ac tree more caught up with my CVS, I
sent Alan a batch of fixes which bring him up to 2.4.6-4um:
<ul>
<li name="2.4.7/umid">
<a href="ac-patches/2.4.7/umid">umid fixes</a> from Henrik Nordstrom
which create a directory based on the umid rather than having that be
the pid file. The pid file and the mconsole socket are now in that
directory.
</li>
<li name="2.4.7/misc-6">
Another batch of <a href="ac-patches/2.4.7/misc-6">small fixes</a> -
a Makefile fix, mconsole cleanups and an update to create the socket in
the umid directory.
</li>
<li name="2.4.7/config">
Some <a href="ac-patches/2.4.7/config">config changes</a> , also
from Henrik Nordstrom. These change the network config names to be
more explicitly UML-specific. The config.in is also cleaned up so
that it resembles the i386 config more closely.
</li>
<li name="2.4.7/iomem">
Greg Lonnon's <a href="ac-patches/2.4.7/iomem">example iomem
driver</a> , plus a couple of generic UML fixes that were needed
in order to make it work.
</li>
<li name="2.4.7/uaccess">
A <a href="ac-patches/2.4.7/uaccess">uaccess fix</a> which required
a surprising amount of surgery to fix. The copy_{to,from}_macros
previously regarded a fault location of 0 as meaning that the copy has
succeeded without faulting. When the address passed into the kernel
was NULL, this of course broke badly. It had a very interesting
side-effect in the case I saw. After running the command that
exercised the bug, every command on the system started failing to
start because libc was corrupted. This was something of a
head-scratcher. I eventually figured out that I was causing the
command to open NULL, the fault went undetected, and the buffer that
was supposed to have had the filename copied into it had
the filename of libc in it from a previous use. So, libc was opened
for writing with fairly severe results.
</li>
</ul>
</blockquote>
<a name="08-03-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">3 Aug 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
The deb build problem turned out to be me accidentally redefining VERSION
in the upper layers of the build process. That value overrode a VERSION
in the kernel build, which resulted in a totally bogus KERNELRELEASE, which
confused a macro which tested it badly enough that it broke the build.
Simple to fix once I figured it out.
<p>
I discovered another hostfs bug on my way back from OLS. ls didn't work
and I found two bugs as a result. The easy one was that hostfs_readdir
was filling in the directory inode rather than the file inode for every
directory entry it passed back to vfs. This was fixed by having read_dir
pass the inode back up so it could be use to fill in the entry properly.
<p>
The more interesting one is that there was a source-incompatible change
made in the stat64 struct between 2.2 and 2.4. The st_ino field changed
its name to __st_ino and a new st_ino field was added at the end. The
inode appears in the same place (the st_ino/__st_ino field) making it
binary compatible. So, after changing to use the 2.4 field name (and
breaking hostfs on 2.2), I changed the hostfs build to figure out what
name to use and passing that in on the compile line to hostfs_user.c.
<p>
In other news, we (me, Rodrigo de Castro, and Livio Baldini Soares) have
decided that -pg support in gcc is broken in multiple ways. rcastro and
livio complained a couple weeks ago about UML's gprof support not working.
I finally had a look at it, and found that it was broken, but not in the
way they described.
<p>
UML crashed in a very inconvient place, and when I finally got in there
enough to figure out what was happening, it turned out that mcount was
segfaulting when it dereferenced ebp because ebp was NULL. The reason
for that turned out to be that in some procedures, mcount is absolutely
the first thing they do. Everything else calls mcount after the new
stack frame has been set up and ebp has a valid value in it (the old esp).
When the procedure is the main procedure for a thread, then ebp turns
out to be NULL.
<p>
The difference between the two sets of procedures seems to be that the
good ones have local variables and the bad ones don't. So, to work
around this bug, I added a useless, but non-optimizable, local to the
affected trampolines.
<p>
Having done that, rcastro and livio were still complaining about UML
crashing. So, I looked at it with rcastro using gdbbot (and livio did so
later and discovered the same thing). -pg was trashing edx for some
reason. A constant (which varies from procedure to procedure) is
dumped into it. This suggests that it's used for the profiling
bookkeeping somehow, but looking at the assembly, we don't see how.
mcount carefully pushes it and restores it, which is not typical of
something that is going to be used for something. The problem is that
FASTCALL procedures (which are regparam(3)) pass arguments in eax,
edx, and ecx. So, dumping this constant into edx trashes the second
argument to the procedure. A workaround for this bug would seem to be
to disable FASTCALL (and I guess that gprof support stopped working
when I enabled FASTCALL to fix a different bug).
<p>
I released 2.4.7-4um today. The main new thing is that you can change
the IP address of a ethertap eth0 device and the host configuration
will change to match. This required a bit of infrastructure which I
wanted for other reasons. The uml_net interface is now versioned,
which I've been meaning to do for a while. uml_net now goes away
cleanly when UML is killed messily. Before, it would hang around,
occupying the tap device, and when UML was rerun, the new uml_net
would emit non-intuitive error messages.
<p>
I also made hostfs build and run again on 2.2 with a bit of Makefile
hackery.
</blockquote>
<a name="07-28-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">28 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
That hostfs problem turned out to be different than I thought. Livio
Soares started chasing the problem and found that the hostfs_user
close_file didn't actually close anything. It took a pointer to a
file descriptor and closed the pointer (or at least tried to) rather
than the descriptor that it pointed to. Fixing that made hostfs
behave a lot better.
<p>
Having fixed that, I finished the page cache work for UML and it can
now successfully do the deb build through hostfs without getting the
md5sum mismatches it was getting before. Having said that, I've
started seeing a compilation problem when building UML through hostfs
that I don't get on the host.
<p>
On to OLS. My talk was in the second slot of the first day, which was
nice. It's good to get your talk over early so you can do the rest of
the conference without worrying about it. It went pretty well. I had
hoped to fit a demo in at the end, but the talk basically went the
full 90 minutes. So I did a real short watch-it-boot-up demo
afterwards while most of the crowd was filing out of the room.
<p>
There was a talk on porting Linux to the i-series IBM boxes (aka
AS-400) which was fairly interesting. They ported Linux/ppc to a
hypervisor running on OS-400, making it fairly similar to the UML
port, being a port to an OS rather than to bare hardware. Dave
Boucher, who gave the talk, made a number of comments comparing it to
UML, which was nice. He also grabbed me during lunch today to quiz me
about the COW ubd driver. It turns out that he can't do that so
easily because OS-400 doesn't have sparse files, so he can't drop
blocks down in the same location in the COW file as in the backing
file because that would allocate space. I suggested a block directory
instead of a bitmap at the beginning of the COW file and dropping
changed blocks down sequentially, but he seemed unconvinced for some
reason.
<p>
A number of people told me either they or people they knew were using
UML for various things. The FreeS/WAN project as a whole seems
extremely interested in UML for running tests on their stuff over a
virtual network. A PPPoE maintainer complained about the ethertap
transport not being intuitively obvious on 2.4. And there were a
bunch of other people who were less specific about what their interest
in UML was who were either using it or were intending to.
<p>
In other news, I discovered that the mcast network transport didn't
work when the box had no ethernet card in it. Being at OLS, I showed
this to Harald Welte and we stared at the code a bit, then asked Andi
Kleen about it. The underlying problem turned out to be that there was
no route to any multicast address because there was no interface on
the system that supported multicast. The fix seems to be to add
multicast support to the loopback device, preferably, and if that's
not possible for some reason, to the dummy device.
</blockquote>
<a name="07-22-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">22 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
2.4.7-1um is released. A change which made kernel threads sychronize with
the parent at startup caused a hang at boot. The cause was a long-standing
bug which caused initdata not to be shared between processes. Andrea
noticed the problem as well, and found the fix.
<p>
That bug was fixed and I released everything. I released it with a fairly
big hostfs problem that I didn't notice until the middle of the release
process. I changed how it opens and closes files, with the result that it
closes them later than it used to. So, it isn't too hard to get hostfs
very confused by running UML out of file descriptors.
</blockquote>
<a name="07-21-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">21 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
2.4.7 appeared yesterday. I'm looking it over to see what's new. One
interesting thing is that Alan is sending over some bits of UML which
change the generic kernel. These don't affect anything besides UML,
so they're harmless. On the other hand, they eliminate some generic
files from my patch, which is nice. It makes the UML patch appear
purer.
</blockquote>
<a name="07-16-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">16 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Yesterday's patches are in 2.4.6-ac5. So, time to send in another
batch. This will get him up to my 2.4.6-2um. Today's batch contains
another batch of <a href="ac-patches/2.4.6/misc-6">
random fixes</a> and Greg Lonnon's
<a href="ac-patches/2.4.6/cow">ubd COW</a>
patch. See <a href="shared_fs.html">this
page</a> for more information on the ubd COW driver.
<p>
I also checked in all the userspace stuff, including the deb builder,
recent changes to the tools, and the website, which I hadn't checked
in for quite a while.
</blockquote>
<a name="07-15-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">15 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Those two pesky patches finally made it to Alan OK and were included
in 2.4.6-ac4. This gets the ac tree up to 2.4.5-8um. The next batch
will bring him up to 2.4.5-10um. It includes a bunch of
<a href="ac-patches/2.4.6/misc-5">miscellaneous
fixes</a>, the first merge of the
<a href="ac-patches/2.4.6/iomem">iomem</a> patch,
and an <a href="ac-patches/2.4.6/mconsole-5">mconsole
update</a> which makes gdb and the ubd driver hot-pluggable and
runs mconsole stuff from a tasklet rather than inside the interrupt.
<p>
With some more symlink abuse, I merged the last of Chris Emerson's ppc
port patch.
</blockquote>
<a name="07-14-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">14 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Two of the three patches I sent to Alan were broken again. However, I
figured out why. My devious little mail reader was breaking lines
when it sent out the mail, which was way too late for me to eyeball it
to make sure it wasn't messing up. Turning off this behavior results
in much better patches at the other end of the line.
<p>
I played with the UML .deb builder and got a workable .deb out of it.
I think I figured out why the process gets a checksum error at the
very end - it's on hostfs, and hostfs reads through the page cache,
but doesn't write through it. I'll have to check with a filesystem
guru on this, but it sounds right to me. Putting the process on a
normal block device results in good checksums.
</blockquote>
<a name="07-13-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">13 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I put out a couple more patches. Highlights include
<ul>
<li>
Greg Lonnon's <a href="iomem.html">iomem</a> driver now
works
</li>
<li>
network devices can now be
<a href="mconsole.html#config">added to</a> and
<a href="mconsole.html#remove">removed</a> from a
running system via the <a href="mconsole.html">mconsole</a>
</li>
<li>
most of the ppc port is merged in
</li>
</ul>
</blockquote>
<a name="07-09-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">9 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Thanks to Simon Blake, I tracked down an interesting bug last night.
The UML build turns off __i386__ in order to throw out some very
hardware-specific code that UML definitely doesn't want.
This also turns off the i386 definition of FASTCALL, which invokes an
in-register parameter passing convention that gcc supports. This
wouldn't be a problem, except that UML borrows code from the i386 port
which assumes that this convention is being used.
<p>
In the case that I was looking at, rw_down_write_failed() was getting
its semaphore address from the wrong place and using a random
userspace address as its semaphore. This could cause all kinds of
interesting side-effects, like kernel corruption from two threads
using two different random addresses as the same semaphore or process
memory corruption from the kernel writing semaphore stuff into its
memory. Hopefully, this fix will eliminate some of the strange
crashes that people ocassionally see with UML.
<p>
<a href="text/fastcall.txt">Here</a> is a more
detailed description of the bug and its side-effects.
<p>
I sent the two broken patches (the
<a href="ac-patches/2.4.6/mconsole">mconsole</a>
and <a href="ac-patches/2.4.6/64-bits">64-bit</a>
patches, see the 30 Jun 2001 entry for
descriptions) to Alan again. Hopefuly they aren't broken this time.
Also, I
<a href="ac-patches/2.4.6/build-2">fixed</a>
a few build problems that turned up lately.
</blockquote>
<a name="07-07-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">7 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I integrated Greg Lonnon's ubd COW patch today. It allows multiple
UMLs to <a href="shared_fs.html">share a filesystem</a>
read-write by storing the changes in a
private file. This private file can be considered to overlay the
read-only shared file. All writes go into the private file, and reads
come from the private file if it has a valid block and from the shared
file if not.
<p>
This allows a huge savings in disk space for people running many UMLs
with large filesystems. It probably will help performance, since the
caching requirements on the host are similarly reduced.
</blockquote>
<a name="07-04-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">4 Jul 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Two of the last three patches I sent Alan somehow got corrupted. I
suspect that what happened was I added spaces accidentally while
reading the patch in my mail composition window by trying to page it
with the space bar, then messed by the patch when deleting the
spaces. So, I'll send them in again.
<p>
Rik van Riel has been visiting for the last couple of days. He was in
Boston for Usenix, and was visiting
<a href="http://www.emc.com">EMC</a> and
<a href="http://www.missioncriticallinux.com">MCLX
</a> (where a number of my former coworkers from DEC now work)
after the show. Since I live a couple of hours north of Boston, I
invited him up. In doing so, I acquired the responsibility of getting
him to Logan airport at the same time that 2M people were going into
Boston to see the 4th of July fireworks and concert. I ended up
putting him on a bus that ran from outside the city straight to the
airport. I haven't heard anything from him since, so I suppose that's
good news.
<p>
2.4.6 was released last night. It turns out to be a piece of cake.
The well-known softirq fix is the only thing that needed changing. I
stuck that in, and it built and ran through my tests without a
problem. The patch is released, and I'll probably finish the rest
tomorrow.
</blockquote>
<a name="06-30-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">30 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I sent Alan patches which will bring him up to 2.4.5-8um, which include:
<ul>
<li name="2.4.5/misc-22">
a collection of <a href="ac-patches/2.4.5/misc-22">small fixes</a> ,
including ^S/^Q support for the console, some ubd driver cleanups, and
the TASK_UNINTERRUPTIBLE fix
</li>
<li name="2.4.5/64-bits">
Lennert's reimplementation of the <a href="ac-patches/2.4.5/64-bits">64-bit
file support</a> - the first try used libc's magic support for
popping the 64-bit interfaces under the 32-bit names. That broke UML
modules badly. This version explicitly uses the 64-bit interfaces and
seems a lot healthier.
</li>
<li name="2.4.5/mconsole">
Lennert's <a href="ac-patches/2.4.5/mconsole">management
console</a> patch. This version has support for getting the
kernel version, halting and rebooting the system, and turning the
debugger on and off.
</li>
</ul>
<p>
Last night, the f00f bug was bugging me, so I fixed it. It turned out
that the tracing thread was routing SIGILL and SIGBUS incorrectly.
Fixing that causes f00f to SIGILL properly.
</blockquote>
<a name="06-29-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">29 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I found and fixed the TASK_UNINTERRUPTIBLE hang last night. It turned
out to be caused by an interrupted write in the block driver. The
driver didn't check the return value, so didn't notice that an IO
request it sent to the IO thread didn't go anywhere. That shut down
the disk IO system, which ultimately results in the whole system being
deadlocked waiting for IO that's never going to happen.
<p>
That, plus a few other things, are checked in as 2.4.5-11um.
<p>
In other news, Bill Stearns, who's always looking for more devious
things to inflict on UML, happened across the
<a href="http://sourceforge.net/projects/ltp">
Linux Test Project</a>
and decided to run it on UML. UML did pretty well. There were three
failures, two of which also fail on the host. The other is the f00f
test, which causes UML to hang. I applied the obvious fix of relaying
SIGILL from UML to the process. That fixed the hang, but after a long
pause, the test's SIGILL handler apparently gets called twice.
</blockquote>
<a name="06-26-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">26 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Those last two patches made it into ac19. Time to start thinking
about bringing the ac tree up to -7um.
<p>
I finally got Greg Lonnon's iomem match into UML. This allows a
process outside UML to communicate with one inside (or with a UML
driver) through a mmapped file.
<p>
I've also been chasing the TASK_UNINTERRUPTIBLE hang that a few people
have been seeing. It happens most easily under UML, apparently. I'm
using a recipe discovered by mistral to reproduce it (two infinite
loops each diffing two kernel pools). The longest it's taken to
reproduce is about 30 minutes. It hung on boot once. The others have
been in the 5-10 minute range.
<p>
I had a long chat with Al Viro last night with him telling me what he
wanted to see from gdb and me providing it. He ended up being puzzled
about what was happening. Following a suggestion from Daniel
Phillips, I've started instrumenting buffer_heads and pages to see
what happened to the ones involved in the hang.
</blockquote>
<a name="06-25-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
It's -ac patch time again. I boiled the -5um to -6um changes down to
two patches:
<ul>
<li name="2.4.5/misc-17">
a <a href="ac-patches/2.4.5/misc-17">miscellaneous fixes</a> which
adds some IP address sanity checking to the ethertap backend, fixes
a couple of process signal delivery races, cleans up the associated
thread data a little, fixes a swap bug (which caused swapped-out pages
to never be unmapped from their processes), and gets rid of the last
vestige of the mm_changes code.
</li>
<li name="2.4.5/timer">
a <a href="ac-patches/2.4.5/timer">timer</a> patch which attempts
to eliminate missing clock ticks by never disabling the timer and
keeps track of ticks which happen when it's not safe to call the timer
IRQ. This improves things, but it doesn't eliminate missing ticks
under load.
</li>
</ul>
</blockquote>
<a name="06-22-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">22 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.5/ac17-build">
Some time around ac16 or ac17, someone added a call to
linux_booted_ok() which the ports have to implement. So, I sent
<a href="ac-patches/2.4.5/ac17-build">the patch</a> to Alan today.
<p>
And a short bit later, I got a reply saying not to bother. The
linux_booted_ok thing was a temporary test that's going to be
removed. So, it won't appear in my pool, but if you absolutely want
to run the ac16/ac17 UML, apply that patch.
<p>
I spent the better part of the afternoon in IRC trying to figure out
the hang that mistral is seeing. No joy, but I did learn more about
the problem. I'll attack it again later.
</blockquote>
<a name="06-21-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">21 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
gdbbot got its first test yesterday when I looked at the problem that
Chris Emerson is having with UML/ppc hanging during boot. I didn't
find the problem, but was able to check that signal delivery (which
was what I thought was broken) was working fine. The next step will
be to do a post-mortem on the hang.
</blockquote>
<a name="06-20-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">20 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I wrote a IRC gateway for gdb. This allows a gdb (like the UML kernel
debugger) to be controlled from an IRC channel. The intent is that if
someone sees a bug that I can't reproduce, but want to look at, that
person's UML gdb can be attached to an IRC channel where I can poke
around and see what's going on.
<p>
I also integrated Lennert's management console patch. This is a very
low-level interface to the kernel (like the i386 SysRq interface).
The main use for it right now is to hot-plug devices. At this point,
only the ubd driver and gdb support this. So, you can add and remove
block devices from your UML without having to reboot it. You can
switch gdb in and out the same way. I will also do the consoles,
serial lines, and network interfaces at some point as well.
</blockquote>
<a name="06-15-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">15 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
The two patches I sent to Alan yesterday are in 2.4.5-ac15. Alan
horribly mangled Harald Welte's name, unfortunately.
<p>
Today was a patch bashing day. I merged in a good number of the
patches in my queue.
<p>
Today was also the (extended) deadline for abstracts for ALS2001. So,
I sent one in. This is the most explicit that I've been so far about
my future development plans for UML. So, if you want to see how wierd
things are going to get, read all about it
<a href="text/als2001.txt">here</a>.
</blockquote>
<a name="06-14-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">14 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
IBM put out a
<a href="http://www-1.ibm.com/linux/news/LTCWhitepaper.shtml">
Linux security whitepaper</a>
in which UML gets a pretty lame mention (down towards the bottom, there's
some prose which is basically lifted from my site). Thanks to Bill Stearns
for spotting it.
<p>
I'm finally getting around to sending off the latest stuff to Alan. The
ac tree is now two cvs updates behind. The first set will be the -5um update,
which is basically
<ul>
<li name="2.4.5/mcast">
the <a href="ac-patches/2.4.5/mcast">mcast transport</a> plus some other
network cleanup
</li>
<li name="2.4.5/misc-13">
some <a href="ac-patches/2.4.5/misc-13">random fixes</a> , including an
updated defconfig, making the console xterms go away when the machine shuts
down, making a read-only hostfs really read-only, hooking up a couple of new
system calls, allowing UML to boot on hosts with a 2G/2G address space split
</li>
</ul>
</blockquote>
<a name="06-12-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">12 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Banged out a bunch of bugs. I started booting UML with 24 megs and
plenty of swap, and running a whole bunch of stuff on it to overload
it and put it heavily into swap. This turned up a couple of signal
delivery races and a swapping bug. The signal races would cause
various strange behavior. Mostly what I saw was hangs with an
infinite sequence of sigreturns. The swap bug caused pages not to be
unmapped when they were swapped out. Obviously, this is very bad.
With the help of rcastro, I fiddled my page table macros to fix this.
I'm still seeing process segfaults. It looks like pages are being
swapped out and swapped back in with the wrong data.
</blockquote>
<a name="06-08-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">8 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I fixed a bunch of buglets, like the console xterms not going away,
readonly hostfs not being readonly, merged Harald Welte's mcast
network transport, and a few other things, and checked them into CVS.
I also checked in the tools, so everything ought to be up-to-date and
consistent at this point.
<p>
I also have the .deb build procedure working, I think. The
uncertainty is due to the fact that I think there's a hostfs data
corruption problem. My development box runs Red Hat, and I couldn't
find RPMs for the Debian tools, so I just installed them in my Debian
filesystem (apt-get rocks, BTW :-), mount the source pool inside a
Debian UML via hostfs, and run the debian build procedure there. The
problem is that the gzipped source tarball has its md5sum recorded at
the beginning of the build and checked again at the end, and they
don't match. I also ran md5sum three times in a row on that file
while the builder was running, and got three different answers. So,
it looks like I have some debugging to do there.
</blockquote>
<a name="06-03-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">3 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
True to yesterday's promise, I sent Alan three more patches
<ul>
<li name="2.4.5/ethertap_mtu">
A <a href="ac-patches/2.4.5/ethertap_mtu">fix</a> for the ethertap driver
which fixes mishandling of large packets
</li>
<li name="2.4.5/segv">
More careful handling of <a href="ac-patches/2.4.5/segv">seg faults</a>
</li>
<li name="2.4.5/umn">
Deletion of the <a href="ac-patches/2.4.5/umn">old umn driver</a>
</li>
</ul>
</blockquote>
<a name="06-02-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">2 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
From the changelog, it looks like yesterday's patches are in ac7. Time to
start generating more...
</blockquote>
<a name="06-01-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Jun 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Another day, another set of patches for Alan. Today, the lucky winners are
<ul>
<li name="2.4.5/misc">
Some <a href="ac-patches/2.4.5/misc">cleanups</a> that I had missed before
</li>
<li name="2.4.5/ether">
Ethernet addresses passed in on the command line are no longer allowed to
be <a href="ac-patches/2.4.5/ether">broadcast addresses</a>
</li>
<li name="2.4.5/copy">
Chris Emerson's longjmp-based <a href="ac-patches/2.4.5/copy">user copy routines
</a>
</li>
<li name="2.4.5/hostfs">
Jorgen Cederlof's support for <a href="ac-patches/2.4.5/hostfs">device
nodes</a> in hostfs
</li>
<li name="2.4.5/ethertap">
A fixlet in the ethertap driver to <a href="ac-patches/2.4.5/ethertap">prevent
a hang</a> if the helper can't be execed
</li>
<li name="2.4.5/time">
Livio Soares' fix to <a href="ac-patches/2.4.5/time">allow the time to be
changed</a>
</li>
<li name="2.4.5/portable">
More <a href="ac-patches/2.4.5/portable">portability changes</a> to
generic and i386 code
</li>
<li name="2.4.5/ppc">
Almost all of the <a href="ac-patches/2.4.5/ppc">remaining ppc changes</a>
</li>
<li name="2.4.5/printk">
A big <a href="ac-patches/2.4.5/printk">printk cleanup</a> which attaches
KERN_* to most UML printks
</li>
</ul>
</blockquote>
<a name="05-30-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">30 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Alan put yesterday's patch into ac5, so UML should build and run
again. Thanks to Arjan van de Ven for telling me about that.
</blockquote>
<a name="05-29-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">29 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.5/ac4-fixes">
It turns out that I messed up the patches somewhat. So, this is the
<a href="ac-patches/2.4.5/ac4-fixes">patch for ac4</a> . With it, UML
will build and run again.
</blockquote>
<a name="05-28-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">28 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Alan apparently put all 10 of yesterday's patches in 2.4.5-ac3 (but
seems not to have dignified the added comment with a changelog entry).
<p>
I wrote up the new networking. Check it out
<a href="networking.html">here</a>.
</blockquote>
<a name="05-27-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">27 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I got 2.4.5 merged in, and mostly released. I'll probably finish it and
announce it tomorrow.
<p>
I also synced up with Alan by sending him a whole pack of patches, to wit:
<ul>
<li name="2.4.5/ppc-merge">
Chris Emerson's <a href="ac-patches/2.4.5/ppc-merge">ppc changes</a>
</li>
<li name="2.4.5/new-network">
the new <a href="ac-patches/2.4.5/new-network">network drivers</a>
</li>
<li name="2.4.5/console_crash">
a fix for a <a href="ac-patches/2.4.5/console_crash">console crash</a>
caused by typing at it too soon
</li>
<li name="2.4.5/configs">
include and __SMP__ <a href="ac-patches/2.4.5/configs">cleanups</a>
</li>
<li name="2.4.5/hostfs-fixlets">
a few <a href="ac-patches/2.4.5/hostfs-fixlets">hostfs fixes</a> from Henrik
Nordstrom
</li>
<li name="2.4.5/more-portable">
some more <a href="ac-patches/2.4.5/more-portable">portability changes</a>
needed by Chris
</li>
<li name="2.4.5/irq-note">
a <a href="ac-patches/2.4.5/irq-note">note</a> to myself to fix something at
some point
</li>
<li name="2.4.5/syscall-cleanup">
some <a href="ac-patches/2.4.5/syscall-cleanup">system call cleanups</a>
from Roman Zippel
</li>
<li name="2.4.5/vm-hole">
a <a href="ac-patches/2.4.5/vm-hole">fix</a> for a crash caused by a hole
between UML's text and data
</li>
<li name="2.4.5/debugger-tweaks">
a couple of <a href="ac-patches/2.4.5/debugger-tweaks">debugger fixes</a>
</li>
</ul>
</blockquote>
<a name="05-26-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">26 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I got the networking cleaned up enough that I'm happy for the general public
to use it. There are three host transports, ethertap, the routing daemon,
and slip. You can have the helper do the host setup for you or not. If you
do, then getting the network running is a matter of a command line switch,
ifconfiging the device, and setting routes inside UML. This is a huge
usability improvement over the previous situation.
<p>
This is all checked in, and I'm currently building 2.4.5, which I'll release
in the next day or two.
</blockquote>
<a name="05-18-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I fixed the slip interface, cleaned out some unused code which had become
a portability problem, and fixed the fix for the crash caused by someone
typing at the console too soon. It is all checked in to CVS.
</blockquote>
<a name="05-17-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">17 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I grabbed 2.4.4-ac11 to see if Henrik's patch was in there, and it was. So
I don't have to worry about it any more. I guess it made ac9, but Henrik
didn't get credit for it in the ac changelog.
<p>
In other news, the ethertap interface is working reasonably well. It couldn't
do HTTP until I figured out that the mtu on host tap device needed to be 16
bytes less than the UML eth0 mtu. The helper is now more helpful. In order
to talk to the rest of the world through it, you basically just have to
ifconfig the device inside UML and add a route to the outside world, and you're
done. Much better than what we had before.
</blockquote>
<a name="05-13-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">13 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Five of yesterday's six patches made it into 2.4.4-ac9. The lonely exception
was Henrik's hostfs blocksize fix.
</blockquote>
<a name="05-12-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">12 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I decided to clean out my patch backlog a bit. So, I merged and sent to Alan
the following patches:
<ul>
<li name="2.4.4/makefiles">
most of Roman Zippel's <a href="ac-patches/2.4.4/makefiles">Makefile
cleanup</a> , minus some bugs and some unistd.h stuff that I want to
look at more carefully
</li>
<li name="2.4.4/portable">
Chris Emerson's <a href="ac-patches/2.4.4/portable">portability changes</a>
to generic code and i386 code
</li>
<li name="2.4.4/rz-clean">
a bunch of <a href="ac-patches/2.4.4/rz-clean">non-Makefile cleanups</a>
from Roman Zippel
</li>
<li name="2.4.4/hostfs-blocksize">
Henrik Nordstrum's
<a href="ac-patches/2.4.4/hostfs-blocksize">one-line fix</a>
that allows root hostfs to work when something has changed the ubd block size
</li>
<li name="2.4.4/starting_exec">
the <a href="ac-patches/2.4.4/starting_exec">removal of
thread.starting_exec</a> ,
which was ancient history, is now longer needed, and was causing hangs under
heavy load
</li>
<li name="2.4.4/stack-race">
a <a href="ac-patches/2.4.4/stack-race">fix for a race</a> while getting
the stack snapshot - this was seen on SMP boxes because the child gets to run
sooner than on a UP box
</li>
</ul>
</blockquote>
<a name="05-11-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">11 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote post="5732981">
Chris Emerson got UML/ppc booting to a shell prompt! His uml-devel post
is <a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/5732981">here</a> . This is the first UML port,
and it showed me how to make UML portable. There aren't really all that many
non-portable things in UML, so a port doesn't take all that much code. Based
on his work, I'm going to write up a UML porting guide, which will be found
<a href="arch-port.html">here</a> when it's done. If that
link is dead, keep trying until I have something to put there.
<p>
In other news, I fiddled the ethertap driver backend so that the read hang
has gone. With some help from Bill Stearns, I also figured out how to talk
to the rest of my network through the ethertap device.
</blockquote>
<a name="05-9-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">9 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I got the ethertap backend to the network driver working today and I
submitted it to <a href="http://sourceforge.net/cvs/?group_id=429">CVS</a> . I haven't been able to get it
to talk to anything but the host over the tap device, but it communicates
with the host just fine.
</blockquote>
<a name="05-4-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">4 May 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
My 2.4.4 fixes, except for Andrew Morton's exitcall fix, are in 2.4.4-ac3.
<p>
I wrote and submitted my OLS paper yesterday, two days late. It's also
posted on this site, as <a href="ols2001.tex">TeX</a>
and <a href="slides/ols2001/index.html">HTML</a>
<p>
On the network driver front, I've got the unified front-end plus the slip
back-end working. I've started working on the ethertap back-end. After that
will come the socket and TUN/TAP back-ends. This stuff is in CVS, but I
haven't updated the patch because the ethernet driver is broken, and I don't
want a bunch of complaints from people who grabbed the latest patch without
knowing what was in it.
<p>
Update: Andrew's patch made it into 2.4.4-ac5. I was beginning to wonder.
That cleans out my pending ac patches.
</blockquote>
<a name="04-28-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">28 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.4/exitcall">
Linus released 2.4.4 yesterday, so I released the 2.4.4 UML today. No major
changes - I dropped in the semaphore changes I was keeping in my ac tree, and
I <a href="ac-patches/2.4.4/pgd_alloc">added an mm argument to
pgd_alloc()</a> .
When I built the RPM (which uses a
different configuration), I noticed that
<a href="ac-patches/2.4.4/blocksize">hostfs didn't compile any more</a> , and
<a href="ac-patches/2.4.4/attach">UML didn't compile with CONFIG_PT_PROXY turned
off</a> . These fixes aren't in CVS or the patch yet.
<p>
Those changes, plus <a href="ac-patches/2.4.4/exitcall">Andrew Morton's exitcall
fix</a> , are off to Alan.
</blockquote>
<a name="04-27-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">27 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote post="5641552">
The last batch of patches I sent to Alan made it into ac14.
<p>
I started looking at the two network drivers today. I think it won't
be too hard to merge them. They're pretty similar, since they're both
derived from the same code base, and the differences seem to be
orthogonal. They don't seem to have done the same things in
fundamentally different ways. I <a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/5644151">posted</a>
my impressions for the devel list to comment on.
<p>
Andrew Morton looked at the shutdown crash that
<a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/5611967">people</a>
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/5633700">started</a>
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/5634016">seeing</a>
lately and <a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/5641389">figured out</a> that it
was caused by /proc being unregistered before something else tried to
remove its proc entries when it was unregistered. He sent in a patch
which reversed the __exitcall order, and Henrik Nordstrom
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/5641552">reported</a> that it fixed the
crash for him.
</blockquote>
<a name="04-22-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">22 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote post="5599515" name="2.4.3/build-ac12">
The fixes I made on Thursday were broken. The initrd fix
<a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/5603373">introduced a name clash</a> with a
function in hostfs, and the sleep fix made
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/5599515">sleep always hang</a> . I
didn't notice because I was fixated on getting UML to boot from an initrd
image, and that wasn't obviously showing the problem.
<p>
Anyhow, I made the fixes, submitted them to CVS, updated the patch, and sent
fixes off to Alan. I hadn't sent in Thursday's changes to Alan, so the
patches are the real thing, not just patches to the patches.
<p>
The patches sent off to Alan today
<a href="ac-patches/2.4.3/initrd">add initrd support
</a> , <a href="ac-patches/2.4.3/timer">fix the sleep bug</a> , and
<a href="ac-patches/2.4.3/build-ac12">make UML build and work with the generic rw
semaphores</a> .
</blockquote>
<a name="04-18-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
The patches I sent in a couple days ago are all in ac10 by the looks of
Alan's change log.
<p>
I figured out how initrd support is supposed to work, and implemented the
necessary stuff in UML. I booted a RH initrd image far enough to convince
myself that it works.
<p>
I also figured out the sleep hang. It turns out to be a race between the
registration of the timer irq and the first time the timer interrupt calls
do_IRQ. The timer was enabled before the registration, so if an interrupt
happened in that window, do_IRQ would bail out early, leaving the irq
permanently marked as in progress and pending. This locked out all future
timer interrupts from going through the irq system, so counters would never
be decremented, and sleeps would never wake up.
</blockquote>
<a name="04-17-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">17 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.3/umn">
The UML build fixes made ac6. However, the rw semaphores in ac7 broke UML
again. I sent Alan the <a href="ac-patches/2.4.3/build-ac7">fix</a>
for that yesterday, and it made ac9 later in the
afternoon. I also sent a few other patches which
<a href="ac-patches/2.4.3/gcov">fix
the gcov and gprof support</a> ,
<a href="ac-patches/2.4.3/debugger">add support
for external debuggers</a> , and
<a href="ac-patches/2.4.3/umn">clean up the umn
driver a bit</a> .
</blockquote>
<a name="04-12-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">12 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
They didn't make -ac5. Oh well.
<p>
I had a chat on <a href="http://kernelnewbies.org">#kernelnewbies</a>
with Rodrigo de Castro, who's using UML
for his <a href="http://tutu.ime.usp.br/index.html">compressed
caching</a> project. He understands swapping better
than I, and told me why my new pte bits were breaking it. So, I fixed
it, and swapping now seems to work.
</blockquote>
<a name="04-11-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">11 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote name="2.4.3/build">
Sent Alan the <a href="ac-patches/2.4.3/build">patches</a> necessary
for UML to build and run in his tree.
I got back a reply which said in its entirety, "ok", which I think is good.
Maybe they will make -ac5.
</blockquote>
<a name="04-10-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">10 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
UML is now in 2.4.3-ac4. I was on IRC with Alan and a bunch of other
hackers when he merged it. He looked like he was going to start
asking a bunch of embarassing questions about my locking, but he was
concerned only about one thing, and that was a special case that
didn't need locking.
<p>
Too bad it doesn't build. The patch that Alan merged was against the
Linus 2.4.3 tree, which differs in a few respects from the current -ac tree.
</blockquote>
<a name="04-8-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">8 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Released 2.4.3 a week or so late. Blame Linus for releasing it the
night before the kernel summit officially started. We were all in San
Jose and not able to react.
</blockquote>
<a name="04-4-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">4 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
A couple more summit tidbits that I forgot to mention in my last entry:
<ul>
<li>
Willy is thinking about using UML as a testbed for NUMA support. He
wants to fire up a number of virtual machines and have them hook
themselves together so they can access each other's memory through
device files. This would allow people who don't have access to the
fancy hardware to develop and debug Linux support for these boxes.
</li>
<li>
UML may appear in the -ac trees at some point. He wanted to include it, but I
had sounded fairly negative towards that in the past. What I don't
want just yet is for UML to hit the Linus tree. Alan said he doesn't
send stuff to Linus if the author doesn't want it sent, which is fine
by me.
</li>
</ul>
</blockquote>
<a name="04-02-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">2 Apr 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Back from the kernel summit. I wanted to get a feel for whether four
things that I wanted from the host kernel were reasonable. I got two
OKs and two dings. That's fine, since the OKs were the important
ones. Here's the run-down:
<ul>
<li>Userspace manipulation of address spaces : I want to be able to
create, populate, release, and switch between mm_structs. This will
speed up UML context switches, and greatly clean up that code. I
asked Linus, and he said OK to the fairly static things that I want to
do. Apparently, there are serious complications when fiddling with the
address space of another process, but that's not what I want to do.
</li>
<li>System call interception via signals : In order to avoid the
context switching between threads involved in virtualizing a system
call, I want to have a process intercept its own system calls by
having the host kernel deliver a signal whenever it makes a system
call. The handler would be the current syscall_handler, which would
read the arguments from its sigcontext_struct. This would change a
system call virtualization from four context switches to a signal
delivery and return. I infer Alan's OK on this from my describing it
in his presence and him not objecting.
</li>
<li>Notification when a UML thread sleeps in the kernel due to a
page fault : For the sake of cleanliness and completeness, I want to
be able to have UML know when a thread is sleeping in the kernel and
be able to call schedule when that happens. This would let UML do as
much work as possible given its state of memory residence. Alan
rejected this on the grounds that UML would be the only sane user of
this mechanism.
</li>
<li>Full kernel preemption : This was implicitly rejected as a UML
need by Alan's rejection of the previous item. If UML is to call
schedule whenever it sleeps, the whole kernel needs to be preemptible
because the swapped-out page might be a kernel page. This doesn't at
all mean that preemption isn't going to happen. Rather, it means that
UML doesn't have a particular need for it.
</li>
</ul>
Other tidbits:
<ul>
<li>A number of people consider UML a very neat hack, including Ben
Lahaise, Andrea Arcangeli, and Eric Raymond.
</li>
<li>Alan turns out to be a UML user. For the last month or so, he's
been booting his kernels as UML kernels before booting them as native
kernels. This is in part because recovery from a totally messed up
kernel is a lot easier with UML than with a native kernel. UML is
also his ptrace test case. It apparently does things with ptrace that
nothing else tries.
</li>
<li>Al Viro is thinking about porting UML to Plan 9. He asked me
about what it would take. He had thought through the ptrace
requirement, and I told him about the mmap requirement, which is the
next hurdle. Plan 9 apparently doesn't have mmap. He's going to
think about how to do that.
</li>
</ul>
<p>On the trip over, I did some debugging, and I also threw in some
patches. There is now a "umid=<name>" switch for providing a virtual
machine with an identifier. This causes a pid file to be created
using that name, which is something that makes controlling multiple
UMLs through a nice UI a lot easier. This file will be replaced with
a socket to the machine console that Lennert is working on.
<p>I also implemented __exitcall, which declares a procedure which is
to be called on machine shutdown. This was prompted by the need to
remove the pid files when the virtual machine goes away. I also
converted other existing cleanups to use this mechanism.
</blockquote>
<a name="03-25-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Mar 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I checked in a bunch of changes again. Henrik Nordstrom provoked me into
making it possible to use hostfs as a root directory by sending me a patch
that did it, but which was wrong (IMHO). He did it in the same way that
nfsroot and initrd support is done, which is by adding a special block of
code to fs/super.c inside CONFIG_HOSTFS_ROOT. That works fine, but I didn't
want to annoy Al Viro. What I did instead was to add a second registration
of hostfs as a device (not a virtual) filesystem and change the ubd driver
to support being given a directory rather than a file or block device. What
happens is that when a read request comes in to the ubd driver, it is
guaranteed to be a request for the superblock. The driver constructs a fake
superblock with the directory name in it. hostfs recognizes that and claims
the mount as its own. After that, it goes back to being a normal virtual
filesystem and doesn't bother the block driver again. The involvement of the
ubd driver is a bit of a kludge, but it works well on the command line, and I
can't think of anything better besides some kind of general support for
virtual root filesystems that cover nfs and initrd as well as hostfs.
<p>
There were also a number of patches from other people: Lennert Buytenhek's
modify_ldt patch, a bunch from Greg Lonnon, one from Gordon McNutt, and
a buffer overrun patch from Henrik Nordstrom.
</blockquote>
<a name="03-23-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">23 Mar 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I spent a few days fixing the infinite recursive context switch bug. That
was a lot more complicated than I expected. The fix involved replacing the
shadow page tables that represent the mappings on the host for each process
with bits in the pte that say whether it is up-to-date or not. These bits
are set in the little functions that change ptes and cleared in fix_range
after it's updated the mappings. Since the process page tables are per-mm
and not per-process, a mapping that was changed in a multi-threaded process
would only be updated for one of the threads. This meant that UML processes
that share a memory context also need to share a memory context on the host.
This in turn complicated exec, since it now needs to create a new host process
in order to get out of a shared UML address space. I implemented this a few
times, and on about the third try, I got something that works.
<p>
So, aside from eliminating a nasty bug, this also makes the modify_ldt fix
more useful, since it now should work properly without any extra code, and
opens the way to more efficient context switching between threads, since they
won't have to go through the remapping that processes need.
</blockquote>
<a name="03-18-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 Mar 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Lots of bugs have been fixed. I got a little list of hostfs
complaints from Al Viro, which I think I fixed. hostfs is now pretty
solid. I fixed the naming problem which cropped up if you held a file
open, then moved its directory and accessed that file by its new
name. You'd get 'file not found'. This is because I stored the full
host pathname in the inode, and when you changed its name while
holding it open by the old name, the inode continued to contain the
old bogus name. This was fixed by having anything that needs a
filename walk the dentry tree back up to the root, constructing the
current filename. The other major problem was that readdir didn't
work, resulting in missing files when a directory was copied. These
are fixed. What remains is to get rid of some interfaces which will
complain about not being implemented.
<p>
The signal delivery race is fixed. That induced me to clean up a lot
of old, crufty code in the kernel entry and exit paths. That's
sensitive code, and a few bugs in it caused some very selective and
very strange behavior.
<p>
I've put together an RPM for UML just in time for the April Linux
Magazine to hit the streets with my article in it. This is good,
because the article claims that RPMs are available, which they weren't
at the time that I wrote it. This also goes some way towards
simplifying the network mess. The RPM installs the umn_helper, which
lets the umn device run without any help from the user. It also
installs the eth tools, which are otherwise hard to find unless you
pull them from cvs.
</blockquote>
<a name="02-25-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Feb 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
CVS update today. I fixed a few bugs and cleaned up a bunch of
things.
<p>
I've started keeping an up-to-date TODO list. This will help
me not forget anything important. I post it to the -devel list
occasionally to prompt people to send in whatever gripes they have.
</blockquote>
<a name="02-24-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">24 Feb 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I'm releasing 2.4.2 today. It has a number of bug fixes and no
significant functionality changes.
<p>
A number of bugs have cropped up lately. The most significant is a
race when a process signal handler returns. There is a narrow window
in which an interrupt can cause a crash. The fix is to implement
sigreturn like the other arches and run almost all of the kernel code
on the kernel stack rather than the process stack as I'm doing now.
</blockquote>
<a name="02-08-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">8 Feb 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I managed to reproduce a number of panics and fixed all but one of
them. The key was hitting UML with a high-concurrency ab run with
requests that fire off perl scripts which make mySQL requests, with
not too much memory, so that it is at least starting to swap.
<p>This reproduced two bugs, one was caused by a failed memory
allocation in the middle of setting up a tracing thread request. The
failure caused a schedule, which caused a switch request, which blew
away the first, partially-set-up request. When the process was
rescheduled, its request was garbage, confusing the tracing thread
into detaching it. This was fixed by moving the allocation to before
the request started being set up.
<p>
The other one, which isn't fixed yet, is caused by the shadow page
tables maintained by arch/um/kernel/tlb.c. It occasionally needs to
allocate a page table when it sets up ptes for a new range of memory.
However, if the context switch that it's dealing with was forced by
low memory, then that allocation will fail, causing a recursive
context switch, and recursion continues until either the stack guard
page is hit, or, in the case of a kernel thread, the task structure is
polluted. I'm going to fix this by following a suggestion by prumpf,
which is to use some spare bits in the pte rather than a separate page
table to figure out what parts of the address space need updating.
<p>
And, panic number three, which is also fixed, was caused a faulty
notion of when a thread is in kernel space. The old way was to look
at whether the thread is being traced. That fails when a breapoint
was put in a signal handler before it requested that tracing be turned
off. The fix is to look at the current stack pointer. However, that
causes problems when a signal is being delivered to a process. In
this case, there is kernel code running on the process stack. So, a
flag was added to the thread structure when this is happening.
</blockquote>
<a name="01-29-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">29 Jan 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Back from Sydney. The talk went pretty well, it was well attended,
and there was a fair amount of interest in UML there. Rik van Riel
and I wandered around Syndey until the following Friday.
<p>The slides from my talk are available
<a href="slides/lca2001/lca.html">
here</a>.
<p>While I was in .au, my OLS paper proposal was accepted, so it
looks like I'll be doing my song and dance in Ottawa this summer.
</blockquote>
<a name="01-03-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">3 Jan 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
Updated the web site with a couple pages describing hostfs and the new
console/serial line input specification.
<p>
The hostfs memory corruption problems are fixed. slab debug found
them for me. They turned out to be two string buffer overrun bugs.
I'll release a patch with the fixes pretty soon.
</blockquote>
<a name="01-01-2001"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Jan 2001</font>
</b>
</td>
</tr>
</table>
<blockquote>
I released the uml patch for 2.4.0-prerelease. You can find it
<a href="ftp.nl.linux.org:/pub/uml/uml-patch-2.4.0-prerelease.bz2">here</a>.
I'm going to make the full release tomorrow, hopefully after fixing
the hostfs crash and getting socket inputs to work.
<p>
Today is the deadline for a first draft of my Linux Magazine article
and for OLS paper proposals. I sent them both in last night. We'll
see what happens.
</blockquote>
<a name="12-27-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">27 Dec 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
hostfs now pretty much works. I built UML from inside itself on
hostfs. I fixed some bugs in the write code, added enough mmap
support to run binaries from hostfs, and implemented statfs. However,
there is some as-yet explained memory corruption going on.
<p>
In an attempt to reproduce the MySQL problems that a couple people are
seeing, I moved some of my work, which is heavy on MySQL and perl,
into UML. I've seen no problems, which is disappointing because I'm
not any closer to finding the bug, but also nice because it shows that
it's possible to do real work inside it.
<p>
I've also been banging on the ethernet driver trying to reproduce the
server buffer overflow that I saw earlier. No dice there either.
<p>
The swapoff bug is now fixed. It turned out to be a bad idea to give
kernel threads both a non-NULL mm and active_mm. That code has been
that way for ages. I have no idea when or why it became a bad thing.
</blockquote>
<a name="12-09-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">9 Dec 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
hostfs is now almost all working. mknod doesn't work, and you can't
run binaries out of a hostfs filesystem.
<p>
I also fixed that pesky linking failure that people have seen seeing
sporadically for a while. I noticed that profiling was turned
on in the latest case that showed up in my inbox. I did a profiling
build of my own and lo! it failed to link. Since I could reproduce
it, I was out of excuses for not fixing it, and so I did. You can see
a full explanation of the problem
<a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/4805000/">here
</a>.
<p>
Those changes plus a couple of smaller ones are now in CVS. They
aren't in the latest patch because the SourceForge upload system has
been seriously b0rked. I'll update the patch when I can.
</blockquote>
<a name="12-07-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">7 Dec 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
I fixed the known bugs in the block driver. The
<blockquote>
<tt>
<font color="#000000">
dd if=/dev/ubd/0 of=/dev/null
</font>
</tt>
</blockquote>
hang was due to the driver returning to the block layer rather than
continuing to process the queue when it found an out-of-range I/O
request. The dbench corruption was due to the elevator rearranging
the request queue while a request was in flight. When that request
finished, the interrupt handler was supposed to retire it by removing
it from the head of the queue. The problem is that the elevator put
some other request at the head, and that request was retired without
ever being done. Meanwhile, the original request was pushed back in
the queue somewhere, and it got done twice.
<p>
Dan Aloni has started the Windows port. He got most of the
kernel to compile. There are a number of undefined symbols from files
that don't compile yet. Overall, though, it's looking pretty good.
</blockquote>
<a name="11-30-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">30 Nov 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
I updated the site a little. The major changes involved the "ARCH=um"
build change. The
<a href="compile.html">compilation</a>
page is now
very explicit about that and there's a
<a href="faq.html">FAQ</a>
entry for it.
<p>
I fixed up the block driver a little. In the past, if you did
<blockquote>
<tt>
<font color="#000000">
dd if=/dev/ubd/0 of=/dev/null
</font>
</tt>
</blockquote>
when it ran off the end of the device, it could apparently hang. This
is fixed. The problem with dbench is not fixed, but I made the
driver's synchronous mode accessible from the command line with the
"ubd=sync" switch. In sychronous mode, the driver has no problems
with dbench.
</blockquote>
<a name="11-18-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 Nov 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
I've got hostfs starting to work reasonably. ls now works, you can cd
around and cat things. You can't write anything, create files, or
execute them yet.
</blockquote>
<a name="11-17-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">17 Nov 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
After a bit of a hiatus, I did a CVS update. A number of buglets
relating to running UML as a daemon were fixed. The build was cleaned
up - I had hard-coded "gcc" instead of "$(CC)" in my Makefiles, the
top-level Makefile is now able to do native and user-mode builds, and
I cleaned up the drivers and fs Makefiles so that they let Rules.mk do
all the hard work.
<p>
I'm also back on hostfs. I fixed the mm problems that it uncovered.
It can now do ls on the top-level directory.
</blockquote>
<a name="11-01-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Nov 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
Linus finally released the final test10 yesterday, so I made my
release last night, with a freshmeat announcement this morning. The
stack overflow problems in test9 are fixed by doubling the stack
size. There is also an inaccessible page between the two stack pages
and the task structure, so there shouldn't be any task structure corruption.
<p>
There were a number of other fixes. At the last minute, I found and
fixed a nasty race which resulted in the kernel tracing its own system
calls, resulting in some nasty stack corruption which made it hard to
figure out what happened. UML can now run when its main console is
not a terminal (i.e. /dev/null). That didn't work because it flipped
the terminal between raw and cooked mode, complaining via printk if
the ioctls failed. That led to an infinite recursion of printk
error messages which ultimately resulted in a segfault. I also made
it possible to mount host devices again. That was broke when I made
the block driver check IO requests against the device size so it could
report errors for out of bounds IO. It turns out not to be possible
to get the size of the media behind a block special file, as far as I
can tell. So, as far as the block driver was concerned, block devices
had zero size, and all IO was out of bounds.
<p>
I also started work on the hostfs filesystem. This is a virtual
filesystem which provides access to the host filesystem. The theory
is straightforward - vfs calls are converted into the equivalent
system calls on the host - but this uncovered a subtle memory
management bug. If a libc routine which mallocs memory is called, and the
break is increased, that extra memory only exists in that process. If
the kernel in another process tries using that memory (or tries
calling malloc at all), it will fault. What needs to happen is for
the context switching code to see if malloc has increased the size of
the data segment and map the new memory into the newly running
process. This also raises some SMP issues because when the new memory
is mapped in, the other processors will need to be told about it so
they can also map it. The same is true of the kernel's virtual memory.
</blockquote>
<a name="10-20-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">20 Oct 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
At long last, I added a
<a href="links.html">page</a> for related
projects and other interesting links.
<p>
In other news, it turns out that Michael Vines wrote a Linux
executable runner for Windows that does what a UML port to Windows
would have to do and he has GPL-ed it and made it available for anyone
who wants to incorporate it into a UML Windows port. See the
<a href="todo.html">todo</a> page for a link to his stuff.
</blockquote>
<a name="10-17-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">17 Oct 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
Back from ALS. The talk went pretty well. I'll put the slides up on
the site at some point.
<p>
I fixed the stack overflow problems that people were seeing. The
stack is now two pages long, with an inaccessible third page
protecting the task structure, which is on the fourth. Now, any stack
overflows will segfault rather than polluting the task structure,
making them a lot easier to debug. This is in CVS along with a few
other changes.
</blockquote>
<a name="10-02-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">2 Oct 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
Bill Stearns decided to go overboard on root_fs production. He's been
fiddling with the mkrootfs script so that it can handle distros other
than Red Hat 6.x. He's done Red Hat 7.0, Mandrake, and Immunix.
These are all now available from the project
<a href="http://sourceforge.net/project/showfiles.php?group_id=429 ">
download page</a> . Caldera, Conectiva, and SuSE are in the works.
</blockquote>
<a name="9-26-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">26 Sep 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>
SGI released a new version of XFS for test5 and I tried to apply it to
my test8 um pool, the idea being that I could play with xfs in
userspace. The patch went in ok, with some rejects that were not too
hard to figure out. After some work, I got it to build. It didn't
boot, though. There were some changes in ll_blk_rw.c that I didn't
understand, and it looks like they are what resulted in the block
device getting a NULL buffer to do I/O into.
<p>
So, maybe I'll give XFS another try when SGI gets it slightly more
up-to-date.
</blockquote>
<a name="9-25-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Sep 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>
I found out why the kernel debugging interface doesn't handle
breakpoints very well. Setting breakpoints results in process
segfaults, floating point exceptions, and other strange behavior. It
turns out that do_syscall stored the current register state in the
thread structure while determining whether the process was doing a
system call. If the process hit a breakpoint in the kernel instead,
then that overwrote the state that was stored when the system call was
called. When the system call returned, that bogus state was restored,
and the process was essentially teleported back into the kernel just
after the breakpoint, leading to all kinds of strange behavior. With
that problem fixed, things work much better. The kernel debugger
seems to be basically healthy, and works just like on a normal process.
<p>
While I was fixing breakpoints, I decided to see why gdb inside a
virtual machine crashes it whenever it sets a breakpoint. There turn
out to be a number of problems. First, SIGTRAP wasn't being delivered
to the debuggee when it hit a breakpoint. This made it hard for gdb
to find out that the breakpoint had been hit, and to remove it
temporarily so the debuggee could get by it. Then, it turned out that
PTRACE_SINGLESTEP wasn't implemented. This is used by gdb to execute
the instruction which had the breakpoint and stop on the next one.
There were one or two other buglets, but now that they are fixed, gdb
seems happy with breakpoints.
</blockquote>
<a name="9-23-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">23 Sep 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
So, I've been a little lax. Here's what's happened in the last few weeks:
two bugs were fixed, the reboot bug and and shell segfault bug. That's it.
</blockquote>
<a name="9-01-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">1 Sep 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>I realized that I am starting to lose track of bugs and functionality
requests, so I dusted off the project's
<a href="http://sourceforge.net/tracker/?group_id=429&atid=100429 ">bug tracking</a>
system and put everything that I know of in it. I'm also using the
<a href="http://sourceforge.net/tracker/?group_id=429&atid=300429 ">patch manager</a>
to store the fixes. The idea is that I'll put fixes there and close
them when I make a release that contains the fix.
<p>I fixed a context-switching bug
<a href="http://www.geocrawler.com/archives/3/597/2000/8/0/4251403/">noticed
</a> by Lennert Buytenhek. The problem turned out to be a race while
updating the address space of the process being restarted. If the
interrupt handler needed data from the kernel's vm area, and that area
hadn't yet been updated, then the kernel would crash. The
<a href="http://sourceforge.net/tracker/?group_id=429&atid=300429 ">
fix</a> was to disable signals during that period of the context switch.
</blockquote>
<a name="8-31-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">31 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>I put in
<a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/4256313/">
Andrea's LFS patch</a>. While I was in there, I cleaned
that code up somewhat. That is some of the oldest code still
remaining, and it really needed some work. I also put in the fix for
the crash caused by a
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/4262186/">
module creating a kernel thread</a>. No word yet on
whether it's the right fix, though.
<p>Also, Laurent Bonnaud volunteered to update the ancient filesystem in
the Debian package to potato. This is very cool. It is something
that I've been wanting to do for a long time.
</blockquote>
<a name="8-25-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">25 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>I'm releasing test7 today.
<p>My ALS2000 paper is now available as
<a href="als2000/index.html">
HTML</a> and <a href="als2000.tex">TeX</a>.
<p>I redid the RH mkrootfs script. It now prompts for the info it
needs. It also works for RH6.1 and probably RH6.2, although I didn't
test that.
</blockquote>
<a name="8-23-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">23 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>Finished my ALS paper and sent it in. That's a load off my mind.
Made some more CVS checkins. This makes the various debugging options
configurable, although I haven't tested the gprof and gcov
configurations. I also added some compatibility code to make the
Debian install happier. It now recognizes that uml can have disks,
but the disk recognizer gets stuck in state 'D' for reasons I haven't
figured out yet.
<p>Linus just released test7, so I am building it right now (right
now as I'm writing this, and not right now as you're reading it, because
I've got no idea when you're reading this). A quick check of the
patch shows no changes that I need to worry about, so this looks like
a drop-it-in-and-it-just-works patch. We'll see...
<p>Things look good. I booted it up, and ran a few things, and
they all worked. So, I'll run the stress testers on it tomorrow, and
if that checks out, I'll release it.
</blockquote>
<a name="8-21-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">21 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
I've made the
<a href="debugging.html">ptrace proxy</a>, gprof, and
gcov support configurable. Also started playing with the Debian 2.2
install. It starts up ok, which the 2.1 install doesn't. It looks
like I'll have to fake some /proc/ide entries before it will deign to
admit that the virtual machine has disks. Right now, it's punting me
into the diskless install.
</blockquote>
<a name="8-17-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">17 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
Fixed a network driver bug which caused a crash when ab was run
against it. This might also fix the ping flood problem. The fix is
in cvs. It will appear for real in test7.
</blockquote>
<a name="8-15-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">15 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
I found out what was causing uml not to boot. It turned out to be a
casting bug which was making the compiler do pointer arithmetic rather
than integer arithmetic. This was a long-standing bug, and test6
changed things so it got hit more heavily. So, assuming that it's not
too badly broken now, I'll release test6 for real.
</blockquote>
<a name="8-08-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">8 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
Revamped the website. It will be put up as soon as Linus releases
test6 and I've integrated the changes in. This is because this site
talks about stuff which isn't really going to be released until then.
</blockquote>
<a name="8-07-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">7 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>Checked in changes which make the new debugging interface
more or less work. I also added a 'debug' command line switch which
starts the kernel in the debugger, so you have control of it from the start.
<p>There are some problems with it. Commands attached to breakpoints
cause segfaults for some reason. It also can't step across a context
switch.
<p>I also put in Rusty's patches. They completely revamp the config
mechanism. For some reason, there also seems to be very complete
networking/netfilter converage.
</blockquote>
<a name="8-05-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">5 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>The ptrace proxy is more or less working. I've checked it in to CVS and
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/4146641/">
announced</a> it on my devel list.
</blockquote>
<a name="8-04-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">4 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>Got a couple of patches from Rusty. I'm apparently going
to graduate to a complete port once I've applied one of them :-) It
is nice. It gives me what looks like a complete config process rather
than the one I've kludged together. He also sent in enough exports to
allow his stuff to be modular inside a UML.
<p>I'm integrating in Lars Brinkhoff's ptrace proxy. It's partially
working - enough that I can attach to the running thread, poke around
it, set breakpoints, etc. This without needing to detach it from the
uml tracing thread. I can't ^C gdb and have it stop the kernel
wherever it happens to be. It also doesn't seem to be following
threads as one goes to sleep and another starts up. Once these work, this
will be a huge improvement in uml kernel debugging.
</blockquote>
<a name="8-02-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">2 Aug 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>Bill Stearns
<a href="http://www.geocrawler.com/lists/3/SourceForge/597/0/4126255/">
pointed out</a> a reproducable way of crashing the kernel
yesterday. It turns out that irq_save/irq_restore were completely wrong.
irq_save was enabling signals when it should have been disabling them.
This could explain a lot of the problems I saw in test4.
<p>I
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/4131747/">
checked the fix into CVS</a> today.
</blockquote>
<a name="7-28-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">28 Jul 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>Linus released test5 last night, so I'm putting out the user-mode
version today. There's nothing new in it. The virtual ethernet is in
the patch, but not enabled in the binary kernels and off by default in
defconfig.
<p>The stress testing of this kernel produced no strange happenings.
Maybe the segfaults and other stuff in the last release weren't my
fault (heh).
</blockquote>
<a name="7-27-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">27 Jul 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>Started integrating Jim Leu's virtual ethernet driver. It
basically works, but it misbehaves a fair bit. It's unclear whether
that's the kernel's fault or the driver's.
</blockquote>
<a name="7-18-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">18 Jul 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>I/O, I/O, off to OLS I go
<p>I'm leaving OLS a bit early because I've got a hiking weekend
coming up - Carter Dome on Saturday and possibly Moriah on Sunday.
So, no one had better expect any work from me until at least next
week...
<p>Plus I'm getting <a href="http://home.ici.net/~erice/Outdoors/Kije.jpg">
Kije</a> next week.
</blockquote>
<a name="7-17-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">17 Jul 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>Announced test4 on freshmeat today.
<p>Also discovered a few more problems which I didn't see on test2 or test3:
<li>the occasional process segfault which I've mentioned already</li>
<li>a devfs segfault - I did this by displaying an xterm out to the
host; when I logged out, the kernel paniced with memory corruption in
devfs</li>
<li>X clients sometimes can't display against a local X server - strace says
that they're stuck in select</li>
<li>strace also displayed their read masks as '[?]', which doesn't
seem right</li>
<p>Patches for these will be forthcoming when I find fixes.
</blockquote>
<a name="7-14-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">14 Jul 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>
<p>Back from SF. Not only did Linus release test3 on Tuesday (as I
discovered when I was checking things out just before leaving for the
airport), but he also released test4 yesterday. So, it looks like I'm
going to be skipping test3 and going straight to test4.
<p>The test3 changes are pretty minor, but enough to prevent the um
test2 patch from going in. task.priority changed to task.nice, there
were some minor locking changes, devfs_mk_dir lost a parameter, and
kernel/timer.c doesn't compile because of that field change.
With those things fixed, the kernel boots fine.
<p>I also decided to get rid of the
<tt>
<pre>
<font size="-1">
pid 16 (mount) - segv changing ip to 0x10025ff2 for address 0x8064000
</font>
</pre>
</tt>
messages that appear at boot time. They are debugging messages to
convince me that the new uaccess macros are working. But they looked
abnormal and worried people so they are now gone. Don't worry, be
happy.
<p>The test3 kernel runs my stress tests (lmbench and a kernel build)
fine, so I'm checking it in to CVS and
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/4036235/">
announcing</a> it on the devel list.
<p>On to test4. The timer.c bug got fixed. Otherwise, the patch went
in cleanly. It compiles and the resulting kernel boots cleanly.
Unfortunately, lmbench segfaults. I put in some debugging code, and
lmbench stops segfaulting. On to the kernel build. That works fine.
I try a couple more lmbench runs. They work fine. Oh well.
<p>I'll consider this releasable. Maybe someone else can find a
better way to reproduce the problem. Check this stuff into CVS, and
out goes the
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/4036801/">
announcement</a>.
<p>I also updated all of the downloadable stuff and
<a href="http://kernelnotes.org/lnxlists/linux-kernel/lk_0007_02/msg01130.html">
announced</a> it.
</blockquote>
<a name="7-3-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">3 Jul 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>I fixed the double panic bug. That was caused by a stacksize limit
that was not a multiple of 4 meg. The reason that matters is that
check_range (in arch/um/kernel/tlb.c), which is used to remap address
spaces during a context switch, assumes that remappable areas and
non-remappable areas are under different pgdirs, which represent 4 meg
apiece. Non-remappable areas are areas of address space which don't
belong to the process. Kernel text, data, and physical and virtual
memory, plus the original stack, fall into this category. They are
represented by vmas in the process mm_struct, but don't have page
table entries. If check_range runs into one of these areas in the
course of looking at something else, the lack of ptes for it will
cause it to be unmapped. Since the process stack is placed just
outside the stacksize limit, if that limit is (say) less than 4 meg,
when check_range checks it for remapping, it will also run into the
main stack provided by the host kernel and unmap it. The panic
happened when a process tried to change its name, which is stored in
that initial stack.
<p>If you see this problem, you can change your stacksize limit to a
multiple of 4 meg, or apply
<a href="stacksize.txt">this patch</a> to
the kernel.
<p>Those two fixes are now checked into
<a href="http://sourceforge.net/cvs/?group_id=429">CVS</a> .
<a href="http://www.geocrawler.com/lists/3/SourceForge/709/0/3976060/">
Here's</a> the devel list post describing the changes.
</blockquote>
<a name="7-2-2000"><table width="100%" bgcolor="#e0e0e0">
<tr>
<td>
<b>
<font color="black">2 Jul 2000</font>
</b>
</td>
</tr>
</table>
<blockquote>UML doesn't run on recent 2.3/2.4 kernels and I figured out
why. The signal frame size increased due to some extra x86 state that needed
to be saved. UML is responsible for making sure that there is enough stack
available when it asks the host kernel to send a signal to one of its
processes. To do this, it pokes the stack (by reading and writing a word)
a little below the current stack pointer. If there is nothing mapped there,
the seg fault handler will map a page in and all will be well. The offset
that it used to poke was a hard-coded 512 bytes, which I got by looking at
the amount of stack state the syscall handler needed (312 bytes) and adding
a bit. However, it turns out that the new stack frames are much bigger than
that, so the 512 bytes wasn't enough. Fixing this makes UML run on new
kernels. If you are seeing this problem, apply
<a href="probe-stack.txt">this patch</a> to the 2.4.0-test2 pool.
<p>I'm also chasing a bug which causes a panic like this:<br>
<tt>
<pre>
<font size="-1">
Kernel panic: Double fault on 0xbffff874 - panicing because it wasn't
fixed the first time
</font>
</pre>
</tt>
</blockquote>
</td>
</tr>
</table>
<center>
<font size="-1">
Hosted at </font>
<a href="http://sourceforge.net">
<IMG alt="SourceForge Logo" border="0" height="31" width="88" src="http://sourceforge.net/sflogo.php?group_id=429">
</a>
</center>
</body>
</html>
|
