fsx test results on arla 0.35.6

[Henry B. Hotz]

A bit ago Apple released a filesystem test tool written by NeXT.  I 
finally got to running it on OSX 10.1.2/arla-0.35.6.  Afraid it 
failed pretty quickly, though I'm not qualified to judge how serious 
the problem(s) is/are.

>[machotz:h/hotz/test] hotz% /usr/local/dist/fsx -c 1 tfil
>Chance of close/open is 1 in 1
>skipping zero size read
>truncating to largest ever: 0x8daf
>truncating to largest ever: 0x34c73
>Size error: expected 0x32456 stat 0x33000 seek 0x33000
>LOG DUMP (16 total operations):
>1(1 mod 256): SKIPPED (no operation)
>                 CLOSE/OPEN
>2(2 mod 256): MAPWRITE 0x34bc6 thru 0x3ffff     (0xb43a bytes)
>                 CLOSE/OPEN
>3(3 mod 256): MAPREAD   0x39e56 thru 0x3cb26    (0x2cd1 bytes)
>                 CLOSE/OPEN
>4(4 mod 256): MAPREAD   0x1bfc6 thru 0x1f146    (0x3181 bytes)
>                 CLOSE/OPEN
>5(5 mod 256): MAPWRITE 0x162e9 thru 0x168de     (0x5f6 bytes)
>                 CLOSE/OPEN
>6(6 mod 256): TRUNCATE DOWN     from 0x40000 to 0x8daf
>                 CLOSE/OPEN
>7(7 mod 256): WRITE     0xca1b thru 0x16d63     (0xa349 bytes) HOLE
>                 CLOSE/OPEN
>8(8 mod 256): READ      0x3bbe thru 0x4f04      (0x1347 bytes)
>                 CLOSE/OPEN
>9(9 mod 256): TRUNCATE UP       from 0x16d64 to 0x34c73
>                 CLOSE/OPEN
>10(10 mod 256): MAPWRITE 0x25fd3 thru 0x269b3   (0x9e1 bytes)
>                 CLOSE/OPEN
>11(11 mod 256): MAPWRITE 0x29666 thru 0x2fa62   (0x63fd bytes)
>                 CLOSE/OPEN
>12(12 mod 256): WRITE   0x3f1a5 thru 0x3ffff    (0xe5b bytes) HOLE
>                 CLOSE/OPEN
>13(13 mod 256): TRUNCATE DOWN   from 0x40000 to 0x289b2
>                 CLOSE/OPEN
>14(14 mod 256): TRUNCATE DOWN   from 0x289b2 to 0x19c72
>                 CLOSE/OPEN
>15(15 mod 256): READ    0x18e5a thru 0x19c71    (0xe18 bytes)
>                 CLOSE/OPEN
>16(16 mod 256): MAPWRITE 0x2c988 thru 0x32455   (0x5ace bytes)
>                 CLOSE/OPEN
>Correct content saved for comparison
>(maybe hexdump "tfil" vs "tfil.fsxgood")
>[machotz:h/hotz/test] hotz%

-- 
The opinions expressed in this message are mine,
not those of Caltech, JPL, NASA, or the US Government.
Henry.B.Hotz at jpl.nasa.gov, or hbhotz at oxy.edu