dtrace discuss - Nov 2008 - Seg fault in printa

printa() in the below script results in segmentation fault. If %A in
dtrace:::END is changed to %a, there is not segmentation fault but the
output is in hex form (Ex: OUT: 8 0xffffffff75f6a3f4 14) and not
readable.
How to avoid the seg fault?
Is there a means to use %a for arg1 and then extract symbol
information (offline via any other tool?) mapping to the hex code to
generate a readable output?

#!/usr/sbin/dtrace -s

#pragma D option quiet

profile:::profile-1001hz
/pid == $target/
{      @pc[tid, arg1] = count();    }

dtrace:::END
{      printa("OUT: %d %A %@d\n", at pc);      }

An addendum:

Crash due to signal: 11
> pstack output
-----------------  lwp# 1 / thread# 1  --------------------
 ffffffff7eba6110 _ndoprnt (ffffffff7f474760, ffffffff7f800ec8,
ffffffff7f800da8, 0, 128, ffffffff7f801700) + 4
 ffffffff7ebaa0a8 snprintf (ffffffff7f801700, 0, ffffffff7f474760,
ffffffff7f800f1c, ffffffff7f801300, 44) + 88
 ffffffff7f461b78 dtrace_uaddr2str (100112b60, 1c8a, ffffffff79f19604,
ffffffff7f801fb0, 128, ffffffff7f801700) + 108
 ffffffff7f458594 pfprint_uaddr (100112b60, 10010b470,
ffffffff7ffff174, 1c8a, 128, 8) + d8
 ffffffff7f45a524 dt_printf_format (ffffffff7ffff175, 1005a0308,
100e6e500, ffffffff7ffff175, ffffffff7f576000, ffffffff7f4584bc) + 6f4
 ffffffff7f45b00c dt_fprinta (0, ffffffff7ffff428, 100e6e500, 4,
1005a0280, 1005a02a8) + 88
 ffffffff7f428c38 dt_aggregate_walk_sorted (100112b60,
ffffffff7f45af84, ffffffff7ffff428, 3e8, 7d, 100e6fd00) + 108
 ffffffff7f45b224 dtrace_fprinta (100112b60, 10010b470, 100e6e500, 1,
1001118e8, 1) + 11c
 ffffffff7f4367f0 dt_consume_cpu (100112b60, 10010b470, 0,
ffffffff7f576000, 1, 1009d4e28) + 78c
 ffffffff7f43755c dtrace_consume (100112b60, 10010b470, 1000044f8,
10000446c, 1c, 0) + 2f8
 ffffffff7f4623e0 dtrace_work (100112b60, 10010b470, 1000044f8,
10000446c, 0, 1) + 108
 0000000100006c1c main (10010b, 10010b390, 10010b378, 100000, 10010b,
100000) + 1fb0
 0000000100001fdc _start (0, 0, 0, 0, 0, 0) + 17c
-----------------  lwp# 2 / thread# 2  --------------------
 ffffffff7ebd43f0 write    (102, ffffffff6fffbf48, 8)
 ffffffff7f45c068 dt_proc_control (0, 1fc000, 8, ffffffff6fffbf48, 1, 1) + 1f0
 ffffffff7ebd2e7c _lwp_start (0, 0, 0, 0, 0, 0)



On Sat, Nov 22, 2008 at 10:31 AM, S h i v <shivakumar.gn at gmail.com>
wrote:
> printa() in the below script results in segmentation fault. If %A in
> dtrace:::END is changed to %a, there is not segmentation fault but the
> output is in hex form (Ex: OUT: 8 0xffffffff75f6a3f4 14) and not
> readable.
> How to avoid the seg fault?
> Is there a means to use %a for arg1 and then extract symbol
> information (offline via any other tool?) mapping to the hex code to
> generate a readable output?
>
> #!/usr/sbin/dtrace -s
>
> #pragma D option quiet
>
> profile:::profile-1001hz
> /pid == $target/
> {      @pc[tid, arg1] = count();    }
>
> dtrace:::END
> {      printa("OUT: %d %A %@d\n", at pc);      }
>

I''m following this up with Shiv directly and I''ll post to
the alias if anything interesting falls out of it.

Jon.
> printa() in the below script results in segmentation fault. If %A in
> dtrace:::END is changed to %a, there is not segmentation fault but the
> output is in hex form (Ex: OUT: 8 0xffffffff75f6a3f4 14) and not
> readable.
> How to avoid the seg fault?
> Is there a means to use %a for arg1 and then extract symbol
> information (offline via any other tool?) mapping to the hex code to
> generate a readable output?
>
> #!/usr/sbin/dtrace -s
>
> #pragma D option quiet
>
> profile:::profile-1001hz
> /pid == $target/
> {      @pc[tid, arg1] = count();    }
>
> dtrace:::END
> {      printa("OUT: %d %A %@d\n", at pc);      }
> _______________________________________________
> dtrace-discuss mailing list
> dtrace-discuss at opensolaris.org
>