optsumry.gml

开放源码的编译器open watcom 1.6.0版的源代码

.mono __stack_conventions__
is a special macro that is defined by the 32-bit compiler when the
"sc" compiler option is specified to indicate that stack calling
conventions are to be used for code generation.
.np
.ix 'predefined macros' '__fpi__'
.ix '__fpi__'
The macro
.mono __fpi__
is a special macro that is defined
.if '&cmpclass' eq 'load-n-go' .do begin
when using &c2cmdup (requires a math coprocessor) as opposed to &ccmdup
(no math coprocessor required).
.do end
.el .do begin
by the compiler when one of the following floating-point options is
specified: "fpi" or "fpi87".
.do end
.*
.if '&cmpclass' ne 'load-n-go' .do begin
.*
.note [NO]DEPendency
.ixm 'DEPENDENCY option'
.ixm 'options' 'DEPENDENCY'
:cmt. .ixm 'NODEPENDENCY option'
:cmt. .ixm 'options' 'NODEPENDENCY'
The "dependency" option specifies that file dependencies are to be
included in the object file.
This is the default option.
This option is used by the &wide. to determine if an object file is
up-to-date with respect to the source files used to build it.
You can specify the "nodependency" option if you do not want file
dependencies to be included in the object file.
.*
.do end
.*
.if '&cmpclass' ne 'load-n-go' .do begin
.*
.note [NO]DEScriptor
.ixm 'DESCRIPTOR option'
.ixm 'options' 'DESCRIPTOR'
:cmt. .ixm 'NODESCRIPTOR option'
:cmt. .ixm 'options' 'NODESCRIPTOR'
The "descriptor" option specifies that string descriptors are to be
passed for character arguments.
This is the default option.
You can specify the "nodescriptor" option if you do not want string
descriptors to be passed for character arguments.
Instead, the pointer to the actual character data and the length will
be passed as two arguments.
The arguments for the length will be passed as additional arguments
following the normal argument list.
For character functions, the pointer to the data and the length of the
character function will be passed as the first two arguments.
.*
.do end
.*
.note DIsk
.ixm 'DISK option'
.ixm 'options' 'DISK'
When this option is used in conjunction with the "list" option, the
listing file is written to the current directory of the default disk.
The listing file name will be the same as the source file name but the
file extension will be :FNAME.&lst:eFNAME..
By default, listing files are written to disk.
The "disk" option will override any previously specified "type" or
"print" option.
.*
.if '&cmpclass' ne 'load-n-go' .do begin
.*
.note DT=<size>
.ixm 'DT=<size> option'
.ixm 'options' 'DT=<size>'
The "data threshold" option is used to set the minimum size for data
objects to be included in the default data segment.
Normally, all data objects smaller than 256 bytes in size are placed
in the default data segment.
When there is a large amount of static data, it is often useful to set
the data threshold size so that all objects of the specified size or
larger are placed into another segment.
For example, the option:
.millust begin
/DT=100
.millust end
.pc
causes all data objects of 100 bytes or more to be placed in a far
data segment.
The "data threshold" only applies to the large and huge memory models
where there can be more than one data segment.
The default data threshold value is 256.
.*
.do end
.*
.if '&cmpclass' eq 'load-n-go' .do begin
.*
.note [NO]EDit
.ixm 'EDIT option'
.ixm 'options' 'EDIT'
:cmt. .ixm 'NOEDIT option'
:cmt. .ixm 'options' 'NOEDIT'
.if '&target' eq 'NEC' .do begin
This option is not supported in the Japanese PC environment.
.do end
.el .do begin
This option is used to control whether the editor interface to &product
is to be used or not.
Specifying the "edit" option causes the editor user interface to be
used.
The default is "noedit" and causes the batch user interface to be used.
.np
The editor is supplied as the program :FNAME.weditf&exe:eFNAME..
When the "edit" option is specified, &product will load and run
the editor provided that it is located in the current directory
or a directory listed in the
.ev PATH
environment string.
.np
This option can only be specified when invoking &product from
the command line and not from the batch prompt (i.e. the
"WATFOR>" prompt).
.do end
.*
.do end
.*
.note [NO]ERrorfile
.ixm 'ERRORFILE option'
.ixm 'options' 'ERRORFILE'
:cmt. .ixm 'NOERRORFILE option'
:cmt. .ixm 'options' 'NOERRORFILE'
This option is used to control whether error messages are output to
a separate error file.
The error file is a disk file of type :FNAME.&err:eFNAME. and
is produced if any diagnostic messages are issued by the compiler.
Specifying "noerrorfile" prevents the creation of an error file.
By default, an error file is created.
.np
If an error file exists before compilation begins, it will be erased.
If no diagnostic messages are produced then an error file will not be
created even though the "errorfile" option is selected.
This option has no effect on the inclusion of diagnostic messages in
the source listing file or the production of diagnostic messages on
the screen.
.*
.note [NO]EXPlicit
.ixm 'EXPLICIT option'
.ixm 'options' 'EXPLICIT'
:cmt. .ixm 'NOEXPLICIT option'
:cmt. .ixm 'options' 'NOEXPLICIT'
The "explicit" option requires the type of all symbols to be
explicitly declared.
An error message will be issued by the compiler if a symbol that does
not appear in a type declaration statement is encountered.
Specifying this option is equivalent to using the
.kw IMPLICIT NONE
statement.
By default, symbols do not have to be explicitly typed.
.*
.note [NO]EXtensions
.ixm 'EXTENSIONS option'
.ixm 'options' 'EXTENSIONS'
:cmt. .ixm 'NOEXTENSIONS option'
:cmt. .ixm 'options' 'NOEXTENSIONS'
This option is used to control the printing of extension messages.
This option may be specified on the command line or it may be placed
anywhere in the source input stream.
In a source file, the option appears as a comment line and takes the
following form.
.millust begin
*$[NO]EXtensions
.millust end
.pc
The "extensions" option enables the printing of extension messages, while
"noextensions" disables the printing of these messages.
By default, extension messages are
.if '&cmpclass' ne 'load-n-go' .do begin
not
.do end
printed.
.*
.note [NO]EZ
.ixm 'EZ option'
.ixm 'options' 'EZ'
:cmt. .ixm 'NOEZ option'
:cmt. .ixm 'options' 'NOEZ'
(32-bit only)
&cmpname will generate an object file in Phar Lap Easy OMF-386 (object
module format) instead of the default Microsoft OMF.
The default option is "noez".
.*
.if '&cmpclass' ne 'load-n-go' .do begin
.*
.note FO=<obj_default>
.ixm 'FO=<obj_default> option'
.ixm 'options' 'FO=<obj_default>'
By default, the object file name is constructed from the source file
name.
Using the "fo" option, the default object file drive, path, file name
and extension can be specified.
.exam begin
C>&ccmd32 report /fo=d:\programs\obj\
.exam end
.pc
A trailing "\" must be specified for directory names.
If, for example, the option was specified as "/fo=d:\programs\obj"
then the object file would be called
.fi D:\PROGRAMS\OBJ.OBJ.
.np
A default extension must be preceded by a period (".").
.exam begin
C>&ccmd32 report /fo=d:\programs\obj\.dbo
.exam end
.*
.do end
.*
.note [NO]FORmat
.ixm 'FORMAT option'
.ixm 'options' 'FORMAT'
:cmt. .ixm 'NOFORMAT option'
:cmt. .ixm 'options' 'NOFORMAT'
The "format" option suppresses the run-time checking that ensures that
the type of an input/output list item matches the format edit
descriptor in a format specification.
This allows an input/output list item of type INTEGER to be formatted
using an F, E or D edit descriptor.
It also allows an input/output list item of a floating-point type to
be formatted using an I edit descriptor.
Normally, this generates an error.
The "format" option is particularly useful for applications that use
integer arrays to store integer and floating-point data.
The default option is "noformat".
.*
.if '&cmpclass' eq 'load-n-go' .do begin
.*
.note [NO]FOVer
.ixm 'FOVER option'
.ixm 'options' 'FOVER'
:cmt. .ixm 'NOFOVER option'
:cmt. .ixm 'options' 'NOFOVER'
The "fover" option causes the compiler to issue an error when a
floating-point overflow occurs during the execution of the program.
Generally, floating-point overflows indicate an error in the program
that should be corrected.
For this reason, the default for this option is "fover" (issue an
error when a floating-point overflow occurs).
If you wish to ignore floating-point overflows, specify the "nofover"
option.
.*
.note [NO]FUNder
.ixm 'FUNDER option'
.ixm 'options' 'FUNDER'
:cmt. .ixm 'NOFUNDER option'
:cmt. .ixm 'options' 'NOFUNDER'
.ix 'debugging' 'traceback'
The "funder" option causes the compiler to issue an error when a
floating-point underflow occurs during the execution of the program.
Generally, floating-point underflows do not indicate an error in the
program.
Typically, underflows occur when an iterative calculation converges to
zero.
For this reason, the default for this option is "nofunder" (ignore
floating-point underflows).
If you wish the compiler to issue an error when a floating-point
underflow occurs, specify the "funder" option.
.*
.do end
.*
.if '&cmpclass' ne 'load-n-go' .do begin
.*
.note FPC
.ixm 'FPC option'
.ixm 'options' 'FPC'
All floating-point arithmetic is done with calls to a floating-point
emulation library.
This option should be used when speed of floating-point emulation is
favoured over code size.
.*
.note FPI
.ixm 'FPI option'
.ixm 'options' 'FPI'
(16-bit only)
&cmpname will generate in-line 80x87 numeric data processor
instructions into the object code for floating-point operations.
Depending on which library the code is linked against, these
instructions will be left as is or they will be replaced by special
interrupt instructions.
In the latter case, floating-point will be emulated if an 80x87 is not
present.
This is the default floating-point option if none is specified.
.np
(32-bit only)
&cmpname will generate in-line 80x87 numeric data processor
instructions into the object code for floating-point operations.
When any module containing floating-point operations is compiled with
the "fpi" option, coprocessor emulation software will be included in
the application when it is linked.
Thus, an 80x87 coprocessor need not be present at run-time.
This is the default floating-point option if none is specified.
.*
.note FPI87
.ixm 'FPI87 option'
.ixm 'options' 'FPI87'
(16-bit only)
&cmpname will generate in-line 80x87 numeric data processor
instructions into the object code for floating-point operations.
An 80x87 coprocessor must be present at run-time.
If the "2" option is used in conjunction with this option,
&cmpname will generate 287/387 compatible instructions;
otherwise &cmpname will generate 8087 compatible instructions.
.np
(32-bit only)
&cmpname will generate in-line 80x87 numeric data processor
instructions into the object code for floating-point operations.
When the "fpi87" option is used exclusively, coprocessor emulation
software is not included in the application when it is linked.
A 80x87 coprocessor must be present at run-time.
.ix 'numeric data processor' 'option'
.ix 'math coprocessor' 'option'
.ix 'floating-point' 'option'
.autonote 16-bit Notes:
.note
.ix 'floating-point' 'consistency of options'
When any module in an application is compiled with a particular
"floating-point" option, then all modules must be compiled with
the same option.
.note
Different math libraries are provided for applications which have been
compiled with a particular floating-point option.
See the chapter entitled :HDREF refid='fplibr'..
.endnote
.autonote 32-bit Notes:
.note
.ix 'floating-point' 'consistency of options'
When any module in an application is compiled with the "fpc" option,
then all modules must be compiled with the "fpc" option.
.note
When any module in an application is compiled with the "fpi" or
"fpi87" option, then all modules must be compiled with one of these
two options.
.note
If you wish to have floating-point emulation software included in
the application, you should select the "fpi" option.
A 387 coprocessor need not be present at run-time.
.note
Different math libraries are provided for applications which have been
compiled with a particular floating-point option.
See the chapter entitled :HDREF refid='fplibr'..
.endnote
.*
.note FP2
.ixm 'FP2 option'
.ixm 'options' 'FP2'
&cmpname will generate in-line 80x87 numeric data processor
instructions into the object code for floating-point operations.
For &company compilers generating 16-bit, this is the default.
For 32-bit applications, use this option if you wish to support those
few 386 systems that are equipped with an 80287 numeric data processor
("fp3" is the default for &company compilers generating 32-bit code).
However, for 32-bit applications, the use of this option will reduce
execution performance.
.*
.note FP3
.ixm 'FP3 option'
.ixm 'options' 'FP3'
&cmpname will generate in-line 387-compatible numeric data processor
instructions into the object code for floating-point operations.
For 16-bit applications, the use of this option will limit the range
of systems on which the application will run but there are execution
performance improvements.
.*
.note FP5
.ixm 'FP5 option'
.ixm 'options' 'FP5'
&cmpname will generate in-line 387-compatible numeric data processor
instructions into the object code for floating-point operations.
The sequence of floating-point instructions will be optimized for
greatest possible performance on the Intel Pentium processor.
For 16-bit applications, the use of this option will limit the range
of systems on which the application will run but there are execution