Hi,

I am testing the FM_ACCCHECK capability of my GLDv3 network driver's FMA 
implementation. The driver, qlge, runs on an adapter with two Network ports and 
two fibre channel ports. So, this adapter has a total of four independent PCI 
functions.

I noticed that, the th_define command hangs when I try to do register access 
check of a particular driver instance, for example, instance 0.
bash-4.0# th_define -n qlge -i 0 -f 1

^C

However, if I do not specify which instance to check, ie to check all 
instances, then it does not hang.
bash-4.0# th_define -n qlge -f 1
Multiple instances match access criteria (only allowed when logging):
        inst    access  rnumber offset  length
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     0       0x0     0x7fffffff
        0       0x3     2       0x3000  0x10
        0       0x3     2       0x3000  0x10
        0       0x3     2       0x3000  0x10
        0       0x3     2       0x0     0x100
        0       0x3     2       0x0     0x100
        0       0x3     2       0x0     0x100
        0       0x3     2       0x2000  0x800
        0       0x3     2       0x2000  0x800
        0       0x3     2       0x2000  0x800
        0       0x3     3       0x0     0x100000
        0       0x3     3       0x0     0x100000
        0       0x3     3       0x0     0x100000
        0       0x40    0       0x0     0x0
        0       0x40    0       0x0     0x0
        0       0x40    0       0x0     0x0
        0       0x40    0       0x0     0x0
        0       0x40    0       0x0     0x0
        0       0x40    0       0x0     0x0
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     0       0x0     0x7fffffff
        1       0x3     2       0x3000  0x10
        1       0x3     2       0x3000  0x10
        1       0x3     2       0x3000  0x10
        1       0x3     2       0x0     0x100
        1       0x3     2       0x0     0x100
        1       0x3     2       0x0     0x100
        1       0x3     2       0x2000  0x800
        1       0x3     2       0x2000  0x800
        1       0x3     2       0x2000  0x800
        1       0x3     3       0x0     0x100000
        1       0x3     3       0x0     0x100000
        1       0x3     3       0x0     0x100000
        1       0x40    0       0x0     0x0
        1       0x40    0       0x0     0x0
        1       0x40    0       0x0     0x0
        1       0x40    0       0x0     0x0
        1       0x40    0       0x0     0x0
        1       0x40    0       0x0     0x0

Is this a bug of th_define?

Tom
-- 
This message posted from opensolaris.org
_______________________________________________
driver-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/driver-discuss

Reply via email to