Filter
Top Products
ARCHITECTURE AND INSTRUCTIONS
before being stored in a second string,
or
before being operated
upon
in some other
fashion. The translation
is
performed by
using
the
value in the
AL
register as
an
index
into a table pointed
at
by the BX register.
The translated value obtained from the table
then
replaces the value initially in the
AL
register.
Here
is
an
example
problem
solved by use
of
primitive string operations
and
iteration con-
trol operations to implement a complex
string operation:
An
input
driver must trans-
late a buffer
of
EBCDIC
characters
into
ASCII,
and
transfer characters until one
of
several
EBCDIC
control characters
is
encoun-
tered.
The
transferred
ASCII
string is
to
be
terminated with
an
EOT
character.
To
initialize the translation sequence,
SI
points
to
the beginning
of
the
EBCDIC
buffer,
DI
points
to
the beginning
of
the
receiving
ASCII
buffer, BX points
to
an
EBCDIC-to-ASCII translation table,
and
CX
contains the length
of
the
EBCDIC
buffer (possibly empty). The translation table
contains the
ASCII
equivalent for each
EBCDIC
character, perhaps with
ASCII
NULs
for illegal characters. The
EOT
code
is
placed
into
the
table
corresponding
to
EBCDIC
stop characters. The 8088 instruc-
tion sequence
to
implement this example
is
the following:
Next:
JCXZ Empty ;skip if
input
buffer
empty
LODS Ebcbuf ;fetch next EBCDIC character
XLAT
Table
;translate
it
to
ASCII
CMP
AL,
EOT ;test for the EOT
STOS
Ascbuf ;transfer ASCII character
LOOPNE Next ;continue if not EOT
Empty:
The
body
of
this loop requires seven bytes
of
code.
CONTROL TRANSFER INSTRUCTIONS
Four
classes
of
control
transfer operations
may be distinguished:
I) calls, jumps,
and
returns;
2)
conditional transfers;
3)
iteration control;
and
4) interrupts.
All
control
transfer operations cause the
pro-
gram
execution
to
continue
at
some new
location in memory, possibly in a new code
segment.
Calls, Jumps, and Returns
Two basic varieties
of
call jumps,
and
returns
are provided - those which transfer
control
within the current code segment,
and
those
which transfer control
to
an
arbitrary
code
segment, which then becomes the
current
code segment. Both direct
and
indirect
transfers are supported; indirect transfers
make use
of
the standard addressing modes.
The three transfer operations are described
below:
CALL
pushes the offset address
of
the next
instruction
onto
the stack (in the case
of
an
inter-segment transfer the CS segment regis-
ter
is
pushed first)
and
then transfers control
to
the target operand.
JMP
transfers control
to
the target operand.
RET
transfers
control
to
the
return
address
saved by a previous
CALL
operation,
and
optionally may adjust
theSP
register
to
dis-
card stacked parameters.
Intra-segment direct calls and
jumps
specify a
self-relative direct replacement, thus allowing
position independent code. A
short
jump
instruction (optional use) transfers
-128
to
+
127
bytes from the current instruction for
code compaction.
Conditional
Jumps
The conditional transfers of
control
perform
a
jump
continuing
upon
various Boolean
functions
of
the flag registers. The destination
must be within
-128
to +
127
bytes from the
instruction.
2-15