Basic System Problem Analysis - August 2003

Case Study: Hang Memory Dump continued

$152 ($86) nmdat > pm_semaphores

ADDRESS OF SEMAPHORE WAITED ON: $b.88ae19b0

$154 ($86) nmdat > rm_semaphore b.88ae19b0

List of pins waiting on semaphore at $b.88ae19b0

$60 $6e $76 $7e $86 $8e $92 $96 $9a $9e $a2 $a6 $aa $ae

Pin $35 has an exclusive lock on shareable semaphore at

$b.88ae19b0

RECORD

SEM_INFO_WORD :

SEM_STATE : 2

SEM_LOCK : 1

SEM_SPEC : 4

SEM_CLASS : 39

SEM_OWNER : 35

SEM_OWNER_COUNT : 1

SEM_WAIT_COUNT : e

SEM_HEAD_WAITER : d3818280

SEM_TAIL_WAITER : d382ba80

END

The “SEM_HEAD_WAITER” and “SEM_TAIL_WAITER” fields are actually PIB

pointers formatted using the type PIB_TYPE.

The RM_SEMPHORE macro is a lot easier as it displays the list of pins that are currently

blocked waiting for this semaphore.

If you had wanted to manually get the address of the semaphore then you would do it this

way:

$166 ($35) nmdat > lev 4

$167 ($35) nmdat > dv psp-28,2

VIRT $480.41853b88 $ 0000000b 88ae18d0

The semaphore address is passed, as a long pointer to SEM_BLOCK so it will be found

at PSP-28 and PSP-24.

This is another of those “how can I know this without the source code” problems. So that

is why I am sharing it here J