NAME
core - format of core image file
DESCRIPTION
UNIX writes out a core image of a terminated process when
any of various errors occur. See signal (II) for the list
of reasons; the most common are memory violations, illegal
instructions, bus errors, and user-generated quit signals.
The core image is called ``core'' and is written in the
process's working directory (provided it can be; normal
access controls apply).
The first 1024 bytes of the core image are a copy of the
system's per-user data for the process, including the regis-
ters as they were at the time of the fault. The remainder
represents the actual contents of the user's core area when
the core image was written. If the text segment is write-
protected and shared, it is not dumped; otherwise the entire
address space is dumped.
The format of the information in the first 1024 bytes is
described by the user structure of the system. The impor-
tant stuff not detailed therein is the locations of the reg-
isters. Here are their offsets. The parenthesized numbers
for the floating registers are used if the floating-point
hardware is in single precision mode, as indicated in the
status register.
fpsr 0004
fr0 0006 (0006)
fr1 0036 (0022)
fr2 0046 (0026)
fr3 0056 (0032)
fr4 0016 (0012)
fr5 0026 (0016)
r0 1772
r1 1766
r2 1750
r3 1752
r4 1754
r5 1756
sp 1764
pc 1774
ps 1776
In general the debuggers db (I) and cdb (I) are sufficient
to deal with core images.
SEE ALSO
cdb (I), db (I), signal (II)
1