Magellan Linux

  • Back to homepage
    • /[magellan-source]/trunk/mkinitrd-magellan/busybox/shell/ash_doc.txt

    Annotation of /trunk/mkinitrd-magellan/busybox/shell/ash_doc.txt

    Parent Directory Parent Directory | Revision Log Revision Log

    Revision 984 - (hide annotations) (download)
    Sun May 30 11:32:42 2010 UTC (14 years ago) by niro
    File MIME type: text/plain
    File size: 4320 byte(s) 
    -updated to busybox-1.16.1 and enabled blkid/uuid support in default config
    1 niro 816 Wait + signals
    2    
    3     We had some bugs here which are hard to test in testsuite.
    4    
    5     Bug 1280 (http://busybox.net/bugs/view.php?id=1280):
    6     was misbehaving in interactive ash. Correct behavior:
    7    
    8     $ sleep 20 &
    9     $ wait
    10     ^C
    11     $ wait
    12     ^C
    13     $ wait
    14     ^C
    15     ...
    16    
    17 niro 984
    18 niro 816 Bug 1984 (http://busybox.net/bugs/view.php?id=1984):
    19     traps were not triggering:
    20    
    21     trap_handler_usr () {
    22     echo trap usr
    23     }
    24     trap_handler_int () {
    25     echo trap int
    26     }
    27     trap trap_handler_usr USR1
    28     trap trap_handler_int INT
    29     sleep 3600 &
    30     echo "Please do: kill -USR1 $$"
    31     echo "or: kill -INT $$"
    32     while true; do wait; echo wait interrupted; done
    33 niro 984
    34    
    35     Bug 189 (https://bugs.busybox.net/show_bug.cgi?id=189)
    36    
    37     func() {
    38     sleep 1
    39     }
    40     while (true); do
    41     func
    42     echo Looping
    43     done
    44    
    45     ^C was observed to make ash processes geometrically multiply (!) instead
    46     of exiting. (true) in subshell does not seem to matter, as another user
    47     reports the same with:
    48    
    49     trap "echo USR1" USR1
    50     while true; do
    51     echo Sleeping
    52     sleep 5
    53     done
    54    
    55     Compat note.
    56     Bash version 3.2.0(1) exits this script at the receipt of SIGINT
    57     _only_ if it had two last children die from it.
    58     The following trace was obtained while periodically running
    59     "killall -SIGINT sleep; sleep 0.1; kill -SIGINT <bash_PID>":
    60    
    61     23:48:32.376707 clone(...) = 13528
    62     23:48:32.388706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
    63     23:48:32.459761 --- SIGINT (Interrupt) @ 0 (0) ---
    64     kill -SIGINT <bash_PID> is ignored, back to waiting:
    65     23:48:32.463706 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 13528
    66     sleep exited with 0
    67     23:48:37.377557 --- SIGCHLD (Child exited) @ 0 (0) ---
    68     23:48:37.378451 clone(...) = 13538
    69     23:48:37.390708 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13538
    70     sleep was killed by "killall -SIGINT sleep"
    71     23:48:38.523944 --- SIGCHLD (Child exited) @ 0 (0) ---
    72     23:48:38.524861 clone(...) = 13542
    73     23:48:38.538706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
    74     23:48:38.624761 --- SIGINT (Interrupt) @ 0 (0) ---
    75     kill -SIGINT <bash_PID> is ignored, back to waiting:
    76     23:48:38.628706 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 13542
    77     sleep exited with 0
    78     23:48:43.525674 --- SIGCHLD (Child exited) @ 0 (0) ---
    79     23:48:43.526563 clone(...) = 13545
    80     23:48:43.538709 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13545
    81     sleep was killed by "killall -SIGINT sleep"
    82     23:48:44.466848 --- SIGCHLD (Child exited) @ 0 (0) ---
    83     23:48:44.467735 clone(...) = 13549
    84     23:48:44.481706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
    85     23:48:44.567757 --- SIGINT (Interrupt) @ 0 (0) ---
    86     kill -SIGINT <bash_PID> is ignored, back to waiting:
    87     23:48:44.571706 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 13549
    88     sleep exited with 0
    89     23:48:49.468553 --- SIGCHLD (Child exited) @ 0 (0) ---
    90     23:48:49.469445 clone(...) = 13551
    91     23:48:49.481708 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13551
    92     sleep was killed by "killall -SIGINT sleep"
    93     23:48:50.515837 --- SIGCHLD (Child exited) @ 0 (0) ---
    94     23:48:50.516718 clone(...) = 13555
    95     23:48:50.530706 waitpid(-1, 0xffc832ec, 0) = ? ERESTARTSYS (To be restarted)
    96     23:48:50.615761 --- SIGINT (Interrupt) @ 0 (0) ---
    97     kill -SIGINT <bash_PID> is ignored, back to waiting:
    98     23:48:50.619705 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 13555
    99     sleep was killed by "killall -SIGINT sleep".
    100     This is the second one in a row. Kill ourself:
    101     23:48:51.504604 kill(13515, SIGINT) = 0
    102     23:48:51.504689 --- SIGINT (Interrupt) @ 0 (0) ---
    103     23:48:51.504915 +++ killed by SIGINT +++
    104    
    105     As long as there is at least one "sleep 5" which exited successfully
    106     (not killed by SIGINT), bash continues. This is not documented anywhere
    107     AFAIKS.
    108    
    109     Why keyboard ^C acts differently?
    110    
    111     00:08:07.655985 clone(...) = 14270
    112     00:08:07.669707 waitpid(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 14270
    113     00:08:12.656872 --- SIGCHLD (Child exited) @ 0 (0) ---
    114     00:08:12.657743 clone(...) = 14273
    115     00:08:12.671708 waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGINT}], 0) = 14273
    116     00:08:13.810778 --- SIGINT (Interrupt) @ 0 (0) ---
    117     00:08:13.818705 kill(14269, SIGINT) = 0
    118     00:08:13.820103 --- SIGINT (Interrupt) @ 0 (0) ---
    119     00:08:13.820925 +++ killed by SIGINT +++
    120    
    121     Perhaps because at the moment bash got SIGINT it had no children?
    122     (it did not manage to spawn new sleep yet, see the trace)