Autosite_Users_Manual - 第110页

Trans lati on Form ats B-12 AutoSite User Manual Spectrum Format, Codes 12 or 13 In this format, bytes are recorded in ASCII codes with binary digits represented by 1s and 0 s. D uring output, each byte is preceded by a …

Translation Formats

AutoSite User Manual B-11

A paper tape generated by a programmer contains a 5-byte, arrow-

shaped header followed by a null and a 4-nibble byte count. The start

code, an 8-bit rubout, follows the byte count. The end of data is signaled

by two nulls and a 2-byte sumcheck of the data field. Refer to Figure B-4.

If the data output has a byte count GREATER than or equal to 64K, an

alternate arrow-shaped header is used. This alternate header (shown

below) is followed by an 8-nibble byte count, sandwiched between a null

and a rubout. The byte count shown here is 40000H (256K decimal). If the

byte count is LESS than 64K, the regular arrowhead is used instead. Data

that are input using Formatted Binary format will accept either version of

this format.

In addition, a third variation of this binary format is accepted on

download. This variation does not have an arrowhead and is accepted

only on input. The rubout begins the format and is immediately followed

by the data. There is no byte count or sumcheck.

DEC Binary Format, Code 11

Data transmission in the DEC Binary format is a stream of 8-bit data

words with no control characters except the start code. The start code is

one null preceded by at least one rubout. The DEC Binary format does

not have addresses.

Figure B-4

An Example of Formatted Binary

Format

RUBOUT (FF)

8 NIBBLE BYTE COUNT

NULL (00)

ARROW

HEAD

0483-2

DATA

Translation Formats

B-12 AutoSite User Manual

Spectrum Format, Codes 12 or 13

In this format, bytes are recorded in ASCII codes with binary digits

represented by 1s and 0s. During output, each byte is preceded by a

decimal address.

Figure B-5 shows sample data bytes coded in the Spectrum format. Bytes

are sandwiched between the space and carriage return characters and are

normally separated by line feeds. The start code is a nonprintable STX,

CTRL-B (or hex 02), and the end code is a nonprintable ETX, CTRL-C (or

hex 03).

Note: Data without a start or end code may be input to or output from the

programmer by use of the alternate data translation format code, 13

Figure B-5

An Example of Spectrum Format

0000 11111111

0001 11111111

0002 11111111

0003 11111111

0004 11111111

0005 11111111

0006 11111111

0007 11111111

0008 11111111

0009 11111111

0010 11111111

0011 11111111

0012 11111111

0013 11111111

0014 11111111

0015 11111111

End code is a

nonprintable EXT

4 or 8 data bits appear between the

space and the carriage return

Address Code is 4

decimal digits

Optional Start Code

is a nonprintable STX

0077-2

Translation Formats

AutoSite User Manual B-13

POF (Programmer Object File) Format, Code 14

The POF (Programmer Object File) format provides a highly compact

data format to enable translation of high bit count logic devices

efficiently. This format currently applies to MAX™ devices, such as the

Altera 5032.

The information contained in the file is grouped into “packets.” Each

packet contains a “tag,” identifying what sort of data the package

contains plus the data itself. This system of packeting information allows

for future definitions as required.

The POF is composed of a header and a list of packets. The packets have

variable lengths and structures, but the first six bytes of every packet

always adhere to the following structure.

struct PACKET_HEAD

{

short tag; /*tag number - type of packet */

long length; /*number of bytes in rest of packet */

}

A POF is read by the program examining each packet and if the tag value

is recognized, then the packet is used. If a tag value is not recognized, the

packet is ignored.

Any packet except the terminator packet may appear multiple times

within a POF. Packets do not need to occur in numerical tag sequence.

The POF reader software is responsible for the interpretation and action

taken as a result of any redundant data in the file, including the detection

of error conditions.

The POF format currently uses the following packet types.

Note: In the following packet type descriptions, one of the terms — Used,

Skipped, or Read — will appear after the tag and name.

Used: The information in this packet is used by the programmer.

Skipped: This information is not used by the programmer.

Read: This information is read by the programmer but has no direct

application.

Creator_ID

tag=1 Used

This packet contains a version ID string from the program which created

the POF.

Device_Name

tag=2 Used

This packet contains the ASCII name of the target device to be

programmed, for example, PM9129.