V. List File Format

It identifies the assembler, contains the copyright information and shows the actual page number at the right margin. After the page header, source lines are output in the list file format. When the maximum number of lines per page is reached, another page header is output after a form feed character. If your printer doesn't support form feeds, the page header can be suppressed with the $NOPAGING control. The number of lines per page can be adjusted to the paper format with the $PAGELENGTH control. The width of the page header (and all other lines) can be set with the $PAGEWIDTH control.

The file header appears only on the first page. It identifies the assembler, lists all input and output files and marks the columns for the line headings. A typical file header is looking as shown below:

Directly after the file header starts the listing of the source code lines. Every source code line is preceded by a line heading. The line heading consists of four columns: line number, include file or macro level, line address, and generated code.
By default the line headings contain tab characters to save disk space. If your printer or file browser doesn't support tabs, they can be expanded to blanks with the $NOTABS control.

The column “Line” contains the global line number. It is not necessarily the local line number within the particular source file, but a global line number that is counted over the main source, all include files, and all macro expansion lines.
Since include files and macros can be nested arbitrarily, the global line number is terminated by a ':' character for the main source and all include file levels, and with a '+' character for macro expansion levels.

The column “I” flags the level of include file or macro nesting. In the main source, this column is empty. The first include file gets level 1. If this include file includes another include file, this one gets level 2, and so on. This is also valid for nested macro calls. If a macro is called in the main source, its expansion lines get level 1. If this macro calls another one, it gets level 2, and so forth.
Include file and macro levels can be nested in any sequence and to any depth!

The column “Addr” shows the start address of the listed line in the currently active segment (8051 address space). All addresses are represented as hex numbers. The addresses in the CODE and XDATA segments are four-digit numbers. Addresses in all other segments are two-digit numbers. For lines that cannot be assigned to a particular segment, the “Addr” field is left blank.

The “Code” column may contain up to four bytes of generated code, which is sufficient for all 8051 instructions. The code is listed in hex byte quantities starting from the left margin of the “Code” column. However, the code generated for DB and DW instructions may be longer than four bytes. In these cases, the source code line is followed by additional line headings until the whole code of the line is listed.
The “Code” column does not always contain code that consumes space in the 8051 CODE segment. In contrast to many other assemblers, ASEM-51 lists the evaluation results of all expressions that may appear in pseudo instructions or assembler controls. These values are listed in hex representation at the right margin of the “Code” column. The segment type of those expressions is flagged with one single character at the left margin of the “Code” column:

            C         CODE
            D         DATA
            I         IDATA
            X         XDATA
            B         BIT
            N         typeless number
            R         register

The “Source” column finally contains the original source code line. A typical source code listing is looking as follows:

If an error is detected in a source line, its position is flagged with a ^ character as good as possible, and a comprehensive error message is inserted. This is looking as shown below:

The error diagnosis at the end of program lists the register banks used, and the total number of errors detected throughout the assembly:

A register bank counts as “used”, if the program had switched to that bank with a USING instruction, or one of the special assembler symbols AR0 ... AR7 has been used, while the bank was active. The message

means, that no bank has been used explicitly, and that the program code may, but need not, be register bank independent.

After the source code listing and error diagnosis, the symbol table or cross-reference listing starts. By default, a symbol table is generated. The symbol table lists all the symbols of a program in alphabetical order with their symbol name, segment type, hex value and first definition line. Predefined symbols are listed without a definition line number.
The symbol table listing can be suppressed with the $NOSYMBOLS control. A typical symbol table listing is looking as shown below:

If the $XREF control is specified, a cross-reference listing is generated instead of a symbol table. The corresponding cross-reference listing for the symbol table above is looking as follows:

It lists all the symbols of the program in alphabetical order, with their symbol name, all definitions including definition lines, segment types, and numerical values. Furthermore, all symbol references are listed as well. The SYMBOL column contains the symbol name, while the columns TYPE, VALUE, and DEFINED may contain the segment types, numerical values, and definition lines of one, more, or no symbol defintions.
Register symbols have the symbol type “REGISTER”, module names have the symbol type “MODULE”, macro names have the symbol type “MACRO”, and symbols that have been referenced but not defined, are flagged with “undef.” in the TYPE column. Starting from column REFERENCED up to the right margin, there is a number of columns (depending on the page width), containing all line numbers of symbol references (if any).
The cross-reference listing does not distinguish, whether multiple definitions of, or references to a particular symbol are legal or not. For this, refer to the error messages in the source listing.