1206917Smarius 2206917SmariusKnown Floating-point Bugs Detected by TestFloat 3206917Smarius 4206917SmariusJohn R. Hauser 5206917Smarius1997 December 15 6206917Smarius 7206917Smarius 8206917Smarius------------------------------------------------------------------------------- 9206917SmariusIntroduction 10206917Smarius 11206917SmariusSeveral popular systems have bugs that TestFloat is very likely to run 12206917Smariusacross. The ones I know of are documented here. First off, TestFloat finds 13206917Smariusno errors in the following processors/machines: 14206917Smarius 15206917Smarius AMD 486 DX4's 16206917Smarius Sun UltraSPARC 1's and 2's 17206917Smarius 18206917SmariusOn the other hand, bugs are found in these processors/machines: 19206917Smarius 20206917Smarius Older Intel Pentiums (with the divide bug) 21206917Smarius Intel Pentium Pros 22206917Smarius Sun SPARCstation 1's and IPX's 23206917Smarius Sun SPARCstation 10's 24206917Smarius HP Precision Architecture processors, with HP-UX prior to version 10.10 25206917Smarius 26206917SmariusFor some reason, most of the bugs found involve conversions from floating- 27206917Smariuspoint to integer formats. 28206917Smarius 29206917SmariusThe bugs are shown as actual TestFloat error lines, along with a brief 30228975Suqsexplanation. The error lines given are not necessarily exhaustive and were 31206917Smariusnot necessarily output in the order shown. 32206917Smarius 33206917SmariusThis document does not pretend to be an authoritative bug listing for all 34206917Smariuscommercial processors. The vast majority of processors are absent from this 35206917Smariuslist because I have never run TestFloat on such machines and I thus have no 36206917Smariusknowledge of what bugs TestFloat might find in them. 37206917Smarius 38206917SmariusThe latest version of this file can be found at the Web page `http:// 39206917Smariushttp.cs.berkeley.edu/~jhauser/arithmetic/testfloat.html'. 40206917Smarius 41206917Smarius 42206917Smarius------------------------------------------------------------------------------- 43206917SmariusOlder Intel Pentiums (with the divide bug) 44206917Smarius 45206917SmariusThe following conversion problems are found on Pentiums that also suffer 46206917Smariusfrom the infamous floating-point divide bug. These bugs have been fixed on 47206917Smariusnewer Pentiums. (TestFloat does not find the divide bug.) 48206917Smarius 49206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 50206917Smariusfloatx80_to_int32 51206917Smarius 52206917Smarius-- A few small fractions are treated as though they were zero. 53206917Smarius 54206917Smarius Errors found in floatx80_to_int32, rounding nearest_even: 55206917Smarius 3FFB.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 56206917Smarius 3FFC.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 57206917Smarius 3FFC.C000000000000000 soft: 00000000 ....x syst: 00000000 ..... 58206917Smarius BFFB.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 59206917Smarius BFFC.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 60206917Smarius Errors found in floatx80_to_int32, rounding to_zero: 61206917Smarius 3FFB.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 62206917Smarius 3FFC.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 63206917Smarius 3FFC.C000000000000000 soft: 00000000 ....x syst: 00000000 ..... 64206917Smarius BFFB.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 65206917Smarius BFFC.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 66206917Smarius BFFC.C000000000000000 soft: 00000000 ....x syst: 00000000 ..... 67206917Smarius Errors found in floatx80_to_int32, rounding down: 68206917Smarius 3FFB.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 69206917Smarius 3FFC.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 70206917Smarius 3FFC.C000000000000000 soft: 00000000 ....x syst: 00000000 ..... 71206917Smarius BFFB.8000000000000000 soft: FFFFFFFF ....x syst: 00000000 ..... 72206917Smarius BFFC.8000000000000000 soft: FFFFFFFF ....x syst: 00000000 ..... 73206917Smarius BFFC.C000000000000000 soft: FFFFFFFF ....x syst: 00000000 ..... 74206917Smarius Errors found in floatx80_to_int32, rounding up: 75206917Smarius 3FFB.8000000000000000 soft: 00000001 ....x syst: 00000000 ..... 76206917Smarius 3FFC.8000000000000000 soft: 00000001 ....x syst: 00000000 ..... 77206917Smarius 3FFC.C000000000000000 soft: 00000001 ....x syst: 00000000 ..... 78206917Smarius BFFB.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 79206917Smarius BFFC.8000000000000000 soft: 00000000 ....x syst: 00000000 ..... 80206917Smarius 81206917Smarius 3FFB.8000000000000000 is the fraction 1/16; 3FFC.8000000000000000 is 1/8; 82206917Smarius and 3FFC.C000000000000000 is 3/16. Both positive and negative inputs are 83206917Smarius affected. 84206917Smarius 85206917Smarius-- Some (all?) positive floating-point values between 2^32 - 1/2 86206917Smarius (401E.FFFFFFFF00000000) and 2^32 (401F.0000000000000000) are rounded to 87206917Smarius zero when the rounding mode is nearest/even or up. 88206917Smarius 89206917Smarius Errors found in floatx80_to_int32, rounding nearest_even: 90206917Smarius 401E.FFFFFFFF80000000 soft: 7FFFFFFF v.... syst: 00000000 ....x 91206917Smarius 401E.FFFFFFFFC00001FE soft: 7FFFFFFF v.... syst: 00000000 ....x 92206917Smarius 401E.FFFFFFFFF8000000 soft: 7FFFFFFF v.... syst: 00000000 ....x 93206917Smarius 401E.FFFFFFFFFEC00000 soft: 7FFFFFFF v.... syst: 00000000 ....x 94206917Smarius 401E.FFFFFFFFFF002000 soft: 7FFFFFFF v.... syst: 00000000 ....x 95206917Smarius 401E.FFFFFFFFFFC00000 soft: 7FFFFFFF v.... syst: 00000000 ....x 96206917Smarius 401E.FFFFFFFFFFE00000 soft: 7FFFFFFF v.... syst: 00000000 ....x 97206917Smarius 401E.FFFFFFFFFFFD7FFE soft: 7FFFFFFF v.... syst: 00000000 ....x 98206917Smarius 401E.FFFFFFFFFFFFFFFE soft: 7FFFFFFF v.... syst: 00000000 ....x 99206917Smarius 401E.FFFFFFFFFFFFFFFF soft: 7FFFFFFF v.... syst: 00000000 ....x 100206917Smarius Errors found in floatx80_to_int32, rounding up: 101206917Smarius 401E.FFFFFFFF00800000 soft: 7FFFFFFF v.... syst: 00000000 ....x 102206917Smarius 401E.FFFFFFFF80000000 soft: 7FFFFFFF v.... syst: 00000000 ....x 103206917Smarius 401E.FFFFFFFFEFFFC000 soft: 7FFFFFFF v.... syst: 00000000 ....x 104206917Smarius 401E.FFFFFFFFFC000000 soft: 7FFFFFFF v.... syst: 00000000 ....x 105206917Smarius 401E.FFFFFFFFFE7FFFFF soft: 7FFFFFFF v.... syst: 00000000 ....x 106206917Smarius 401E.FFFFFFFFFFF00000 soft: 7FFFFFFF v.... syst: 00000000 ....x 107206917Smarius 401E.FFFFFFFFFFFE0800 soft: 7FFFFFFF v.... syst: 00000000 ....x 108206917Smarius 401E.FFFFFFFFFFFF7FFB soft: 7FFFFFFF v.... syst: 00000000 ....x 109206917Smarius 401E.FFFFFFFFFFFFFFFE soft: 7FFFFFFF v.... syst: 00000000 ....x 110206917Smarius 401E.FFFFFFFFFFFFFFFF soft: 7FFFFFFF v.... syst: 00000000 ....x 111206917Smarius 112206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 113206917Smarius 114206917Smarius 115206917Smarius------------------------------------------------------------------------------- 116206917SmariusIntel Pentium Pros 117206917Smarius 118206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 119206917Smariusfloatx80_to_int32 120206917Smarius 121206917Smarius-- The inexact flag is sometimes raised instead of the invalid flag for 122206917Smarius floating-point inputs under -(2^32) (C01F.0000000000000000). This bug is 123206917Smarius sporadic. It appears to be deterministic but dependent on the sequence 124206917Smarius of operations executed. 125206917Smarius 126206917Smarius Errors found in floatx80_to_int32, rounding nearest_even: 127206917Smarius C01F.C000000000000002 soft: 80000000 v.... syst: 80000000 ....x 128206917Smarius C021.F00000000000003F soft: 80000000 v.... syst: 80000000 ....x 129206917Smarius Errors found in floatx80_to_int32, rounding to_zero: 130206917Smarius C021.F00000000000003F soft: 80000000 v.... syst: 80000000 ....x 131206917Smarius Errors found in floatx80_to_int32, rounding up: 132206917Smarius C01F.C000000000000007 soft: 80000000 v.... syst: 80000000 ....x 133206917Smarius C01F.C000000000001000 soft: 80000000 v.... syst: 80000000 ....x 134206917Smarius 135206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 136206917Smarius 137206917Smarius 138206917Smarius------------------------------------------------------------------------------- 139206917SmariusSun SPARCstation 1's and IPX's 140206917Smarius 141206917SmariusSome older SPARCstations appear confused about whether underflow tininess is 142206917Smariusdetected before or after rounding. For conversions from double precision 143206917Smariusto single precision, tininess is detected after rounding, while for all 144206917Smariusquadruple-precision operations it is detected before rounding. Single- and 145206917Smariusdouble-precision multipies go both ways: 146206917Smarius 147206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 148206917Smariusfloat32_mul, float64_mul 149206917Smarius 150206917Smarius-- For multiplies, underflow tininess is detected _before_ rounding if one 151206917Smarius of the inputs is subnormal, and _after_ rounding otherwise. If tininess 152206917Smarius is assumed to be detected before rounding, the following errors are 153206917Smarius generated: 154206917Smarius 155206917Smarius Errors found in float32_mul, rounding nearest_even: 156206917Smarius 001.000001 07E.7FFFFE soft: 001.000000 ...ux syst: 001.000000 ....x 157206917Smarius 001.000001 87E.7FFFFE soft: 801.000000 ...ux syst: 801.000000 ....x 158206917Smarius 001.000002 07E.7FFFFC soft: 001.000000 ...ux syst: 001.000000 ....x 159206917Smarius 001.000002 87E.7FFFFC soft: 801.000000 ...ux syst: 801.000000 ....x 160206917Smarius 001.000004 07E.7FFFF8 soft: 001.000000 ...ux syst: 001.000000 ....x 161206917Smarius Errors found in float32_mul, rounding down: 162206917Smarius 001.000001 87E.7FFFFE soft: 801.000000 ...ux syst: 801.000000 ....x 163206917Smarius 001.000002 87E.7FFFFC soft: 801.000000 ...ux syst: 801.000000 ....x 164206917Smarius 001.000004 87E.7FFFF8 soft: 801.000000 ...ux syst: 801.000000 ....x 165206917Smarius 001.000008 87E.7FFFF0 soft: 801.000000 ...ux syst: 801.000000 ....x 166206917Smarius 001.000010 87E.7FFFE0 soft: 801.000000 ...ux syst: 801.000000 ....x 167206917Smarius Errors found in float32_mul, rounding up: 168206917Smarius 001.000001 07E.7FFFFE soft: 001.000000 ...ux syst: 001.000000 ....x 169206917Smarius 001.000002 07E.7FFFFC soft: 001.000000 ...ux syst: 001.000000 ....x 170206917Smarius 001.000004 07E.7FFFF8 soft: 001.000000 ...ux syst: 001.000000 ....x 171206917Smarius 001.000008 07E.7FFFF0 soft: 001.000000 ...ux syst: 001.000000 ....x 172206917Smarius 001.000010 07E.7FFFE0 soft: 001.000000 ...ux syst: 001.000000 ....x 173206917Smarius Errors found in float64_mul, rounding nearest_even: 174206917Smarius 001.0000000000001 3FE.FFFFFFFFFFFFE 175206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 176206917Smarius 001.0000000000001 BFE.FFFFFFFFFFFFE 177206917Smarius soft: 801.0000000000000 ...ux syst: 801.0000000000000 ....x 178206917Smarius 001.0000000000002 3FE.FFFFFFFFFFFFC 179206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 180206917Smarius 001.0000000000002 BFE.FFFFFFFFFFFFC 181206917Smarius soft: 801.0000000000000 ...ux syst: 801.0000000000000 ....x 182206917Smarius 001.0000000000004 3FE.FFFFFFFFFFFF8 183206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 184206917Smarius Errors found in float64_mul, rounding down: 185206917Smarius 001.0000000000001 BFE.FFFFFFFFFFFFE 186206917Smarius soft: 801.0000000000000 ...ux syst: 801.0000000000000 ....x 187206917Smarius 001.0000000000002 BFE.FFFFFFFFFFFFC 188206917Smarius soft: 801.0000000000000 ...ux syst: 801.0000000000000 ....x 189206917Smarius 001.0000000000004 BFE.FFFFFFFFFFFF8 190206917Smarius soft: 801.0000000000000 ...ux syst: 801.0000000000000 ....x 191206917Smarius 001.0000000000008 BFE.FFFFFFFFFFFF0 192206917Smarius soft: 801.0000000000000 ...ux syst: 801.0000000000000 ....x 193206917Smarius 001.0000000000010 BFE.FFFFFFFFFFFE0 194206917Smarius soft: 801.0000000000000 ...ux syst: 801.0000000000000 ....x 195206917Smarius Errors found in float64_mul, rounding up: 196206917Smarius 001.0000000000001 3FE.FFFFFFFFFFFFE 197206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 198206917Smarius 001.0000000000002 3FE.FFFFFFFFFFFFC 199206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 200206917Smarius 001.0000000000004 3FE.FFFFFFFFFFFF8 201206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 202206917Smarius 001.0000000000008 3FE.FFFFFFFFFFFF0 203206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 204206917Smarius 001.0000000000010 3FE.FFFFFFFFFFFE0 205206917Smarius soft: 001.0000000000000 ...ux syst: 001.0000000000000 ....x 206206917Smarius 207206917Smarius If we assume tininess should be detected after rounding, we get the 208206917Smarius following errors: 209206917Smarius 210206917Smarius Errors found in float32_mul, rounding nearest_even: 211206917Smarius 000.7FFC00 07F.000400 soft: 001.000000 ....x syst: 001.000000 ...ux 212206917Smarius 000.7FFC00 87F.000400 soft: 801.000000 ....x syst: 801.000000 ...ux 213206917Smarius 000.7FFE00 07F.000200 soft: 001.000000 ....x syst: 001.000000 ...ux 214206917Smarius 000.7FFE00 87F.000200 soft: 801.000000 ....x syst: 801.000000 ...ux 215206917Smarius 000.7FFF00 07F.000100 soft: 001.000000 ....x syst: 001.000000 ...ux 216206917Smarius Errors found in float32_mul, rounding down: 217206917Smarius 000.7FFC00 87F.000400 soft: 801.000000 ....x syst: 801.000000 ...ux 218206917Smarius 000.7FFE00 87F.000200 soft: 801.000000 ....x syst: 801.000000 ...ux 219206917Smarius 000.7FFF00 87F.000100 soft: 801.000000 ....x syst: 801.000000 ...ux 220206917Smarius 000.7FFF80 87F.000080 soft: 801.000000 ....x syst: 801.000000 ...ux 221206917Smarius 000.7FFFC0 87F.000040 soft: 801.000000 ....x syst: 801.000000 ...ux 222206917Smarius Errors found in float32_mul, rounding up: 223206917Smarius 000.7FFC00 07F.000400 soft: 001.000000 ....x syst: 001.000000 ...ux 224206917Smarius 000.7FFE00 07F.000200 soft: 001.000000 ....x syst: 001.000000 ...ux 225206917Smarius 000.7FFF00 07F.000100 soft: 001.000000 ....x syst: 001.000000 ...ux 226206917Smarius 000.7FFF80 07F.000080 soft: 001.000000 ....x syst: 001.000000 ...ux 227206917Smarius 000.7FFFC0 07F.000040 soft: 001.000000 ....x syst: 001.000000 ...ux 228206917Smarius Errors found in float64_mul, rounding nearest_even: 229206917Smarius 000.FFFFFFE000000 3FF.0000002000000 230206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 231206917Smarius 000.FFFFFFE000000 BFF.0000002000000 232206917Smarius soft: 801.0000000000000 ....x syst: 801.0000000000000 ...ux 233206917Smarius 000.FFFFFFF000000 3FF.0000001000000 234206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 235206917Smarius 000.FFFFFFF000000 BFF.0000001000000 236206917Smarius soft: 801.0000000000000 ....x syst: 801.0000000000000 ...ux 237206917Smarius 000.FFFFFFF800000 3FF.0000000800000 238206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 239206917Smarius Errors found in float64_mul, rounding down: 240206917Smarius 000.FFFFFFE000000 BFF.0000002000000 241206917Smarius soft: 801.0000000000000 ....x syst: 801.0000000000000 ...ux 242206917Smarius 000.FFFFFFF000000 BFF.0000001000000 243206917Smarius soft: 801.0000000000000 ....x syst: 801.0000000000000 ...ux 244206917Smarius 000.FFFFFFF800000 BFF.0000000800000 245206917Smarius soft: 801.0000000000000 ....x syst: 801.0000000000000 ...ux 246206917Smarius 000.FFFFFFFC00000 BFF.0000000400000 247206917Smarius soft: 801.0000000000000 ....x syst: 801.0000000000000 ...ux 248206917Smarius 000.FFFFFFFE00000 BFF.0000000200000 249206917Smarius soft: 801.0000000000000 ....x syst: 801.0000000000000 ...ux 250206917Smarius Errors found in float64_mul, rounding up: 251206917Smarius 000.FFFFFFE000000 3FF.0000002000000 252206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 253206917Smarius 000.FFFFFFF000000 3FF.0000001000000 254206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 255206917Smarius 000.FFFFFFF800000 3FF.0000000800000 256206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 257206917Smarius 000.FFFFFFFC00000 3FF.0000000400000 258206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 259206917Smarius 000.FFFFFFFE00000 3FF.0000000200000 260206917Smarius soft: 001.0000000000000 ....x syst: 001.0000000000000 ...ux 261206917Smarius 262206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 263206917Smarius 264206917Smarius 265206917Smarius------------------------------------------------------------------------------- 266206917SmariusSun SPARCstation 10's 267206917Smarius 268206917SmariusLike other SPARCstations, some SPARCstation 10's are inconsistent regarding 269206917Smariusunderflow tininess, detecting it after rounding for single- and double- 270206917Smariusprecision operations and before rounding for quadruple-precision operations. 271206917SmariusThe following bug has also been observed. 272206917Smarius 273206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 274206917Smariusfloat32_to_int32_round_to_zero, float64_to_int32_round_to_zero 275206917Smarius 276206917Smarius-- Single- and double-precision NaNs are converted to the integer zero. 277206917Smarius (The invalid exception flag is raised correctly.) 278206917Smarius 279206917Smarius Errors found in float32_to_int32_round_to_zero: 280206917Smarius 8FF.5D36AC soft: 7FFFFFFF v.... syst: 00000000 v.... 281206917Smarius 0FF.7FFFC0 soft: 7FFFFFFF v.... syst: 00000000 v.... 282206917Smarius 8FF.7C0000 soft: 7FFFFFFF v.... syst: 00000000 v.... 283206917Smarius 0FF.2AB7ED soft: 7FFFFFFF v.... syst: 00000000 v.... 284206917Smarius 0FF.03FFFF soft: 7FFFFFFF v.... syst: 00000000 v.... 285206917Smarius Errors found in float64_to_int32_round_to_zero: 286206917Smarius 7FF.45AD84DB2524A soft: 7FFFFFFF v.... syst: 00000000 v.... 287206917Smarius 7FF.CFEE063EE0512 soft: 7FFFFFFF v.... syst: 00000000 v.... 288206917Smarius 7FF.89FF03AB7DBA2 soft: 7FFFFFFF v.... syst: 00000000 v.... 289206917Smarius 7FF.FFFFFFFFFF800 soft: 7FFFFFFF v.... syst: 00000000 v.... 290206917Smarius FFF.68A6410E91BF6 soft: 7FFFFFFF v.... syst: 00000000 v.... 291206917Smarius 292206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 293206917Smarius 294206917Smarius 295206917Smarius------------------------------------------------------------------------------- 296206917SmariusHP Precision Architecture processors, with HP-UX prior to version 10.10 297206917Smarius 298206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 299206917Smariusfloat32_to_int32_round_to_zero, float64_to_int32_round_to_zero 300206917Smarius 301206917Smarius-- When the floating-point value is too large, the overflow and inexact 302206917Smarius exception flags are raised instead of the invalid flag. 303206917Smarius 304206917Smarius Errors found in float32_to_int32_round_to_zero: 305206917Smarius 89E.000007 soft: 80000000 v.... syst: 80000000 ..o.x 306206917Smarius 0A2.000020 soft: 7FFFFFFF v.... syst: 7FFFFFFF ..o.x 307206917Smarius 8FA.7C0000 soft: 80000000 v.... syst: 80000000 ..o.x 308206917Smarius Errors found in float64_to_int32_round_to_zero: 309206917Smarius 7FD.0448700002F1C soft: 7FFFFFFF v.... syst: 7FFFFFFF ..o.x 310206917Smarius DAA.F000000000000 soft: 80000000 v.... syst: 80000000 ..o.x 311206917Smarius 41E.063DA00005E65 soft: 7FFFFFFF v.... syst: 7FFFFFFF ..o.x 312206917Smarius 47E.FFFF800000000 soft: 7FFFFFFF v.... syst: 7FFFFFFF ..o.x 313206917Smarius 51F.0000000000004 soft: 7FFFFFFF v.... syst: 7FFFFFFF ..o.x 314206917Smarius DDA.0000001FFFFFF soft: 80000000 v.... syst: 80000000 ..o.x 315206917Smarius D70.00000000003FF soft: 80000000 v.... syst: 80000000 ..o.x 316206917Smarius C7E.0000100000000 soft: 80000000 v.... syst: 80000000 ..o.x 317206917Smarius 47E.000000000007F soft: 7FFFFFFF v.... syst: 7FFFFFFF ..o.x 318206917Smarius D57.000000000FFFF soft: 80000000 v.... syst: 80000000 ..o.x 319206917Smarius 320206917Smarius 321206917Smarius- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 322206917Smarius 323206917Smarius 324