<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <title>Using Poly/ML - Chapter 1: Introduction</title> </head> <body> <h1><font SIZE="6"><a name="Intro1">Chapter 1: Introduction</a></font></h1> <p>This document describes Poly/ML version 4 and older. Certain features are significantly different in version 5, particularly the persistent store which has been removed.</p> <p><font SIZE="3">Poly/ML is an implementation of Standard ML, with a few non-standard extensions such as <a href="VectorArray4.html">arrays</a>, <a href="Processes5.html">processes</a>, a <a href="Make7.html">make system</a> and the fact that ML values may become '<a href="#persist1.6">persistent</a>'.</font></p> <h2><a name="structurePolyML1.1"><font SIZE="4">1.1 The structure</font><b><font FACE="Courier" SIZE="4">PolyML</font></b></a></h2> <p><font SIZE="3">Most non-standard system-level functions are held in the structure </font><b><font FACE="Courier" SIZE="3">PolyML</font></b><font SIZE="3">.<b> </b>This structure contains at least the following functions:</font></p> <p><font FACE="Courier" SIZE="3"><b>val commit: unit-> bool<br> val quit: unit -> unit<br> val exit: int -> unit<br> val use: string -> unit<br> val cd: string -> unit</b></font></p> <p><font FACE="Courier" SIZE="3"><b>val isLogging: unit -> bool<br> val startLog: string -> unit<br> val restartLog: string -> unit<br> val stopLog: unit -> unit<br> val logName: unit -> string<br> val writeLog: string -> unit</b></font></p> <p><font FACE="Courier" SIZE="3"><b>val exception_trace: (unit -> a) -> 'a</b></font></p> <p><font FACE="Courier" SIZE="3"><b>val print_depth: int -> unit</b></font></p> <p><font FACE="Courier" SIZE="3"><b>val error_depth: int -> unit</b></font></p> <p><font FACE="Courier" SIZE="3"><b>val timing: bool -> unit<br> val profiling: int -> unit</b></font></p> <p><font FACE="Courier" SIZE="3"><b>val make-database : string -> unit</b></font></p> <p><font FACE="Courier" SIZE="3"><b>val make: string unit<br> val depends: string unit</b></font></p> <p><font SIZE="3">The purpose of the above functions is explained in the following sections except for </font><b><font FACE="Courier" SIZE="3">make</font></b><font SIZE="3"> and </font><b><font FACE="Courier" SIZE="3">depends</font></b><font SIZE="3"> which are discussed in <a href="Make7.html">chapter 7</a>.</font></p> <p><font SIZE="3">In addition, the Motif Edition of Poly/ML supports the X Windows System. The functions which provide the Poly/ML interface to X are encapsulated in the structures <b>XWindows </b>and <b>Motif </b>which, because of their size and importance, are described in separate manuals.</font></p> <h2><a name="startsession1.2"><b><font SIZE="4">1.2 Starting a Poly/ML session</font></b></a></h2> <p><font SIZE="3">You start a Poly/ML session by running the driver program </font><b><font FACE="Courier" SIZE="3">poly</font></b><font SIZE="3"> with your Poly/ML database:</font></p> <p><font FACE="Courier" SIZE="3"><b>unix% poly ML_dbase</b></font></p> <p><font SIZE="3">This should produce (something like) the response:</font></p> <p><font FACE="Courier" SIZE="3"><b>Poly/ML RTS version Windows<br> Copyright (c) 1999 Cambridge University Technical Services Limited<br> Poly/ML 3.X<br> ></b></font></p> <p><font SIZE="3">(‘</font><strong><font FACE="Courier" SIZE="3">></font></strong><font SIZE="3">’ is the default Poly/ML prompt.) You can now enter an ML program, for example:</font></p> <p><font FACE="Courier" SIZE="3"><b>> fun fact 0 = 1<br> # | fact n = n * fact(n-1);<br> </b></font></p> <p><font FACE="Courier" SIZE="3"><b>val fact = fn : int -> int</b></font></p> <p><font FACE="Courier" SIZE="3"><b>> fact 20;<br> val it = 2432902008176640000 : int</b></font></p> <p><font FACE="Courier" SIZE="3"><b>></b></font></p> <p><font SIZE="3">(‘<strong><font face="Courier">#</font></strong>’ is the default Poly/ML secondary prompt, which is used whenever the current expression is incomplete.) The Poly/ML system may take slightly longer than normal to respond to the first thing you type; this is because the ML compiler is itself a collection of persistent objects which have to be paged into memory before they can be used.</font></p> <h2><a name="updatedb1.3"><font SIZE="4">1.3 Updating the Database</font></a></h2> <p><font SIZE="3">The database can be updated from the local heap with</font></p> <p><font FACE="Courier" SIZE="3"><b>PolyML.commit ();</b></font></p> <p><font SIZE="3">All changes to the local heap since the last commit (or since the start of the session, if there was no previous commit) are written out to the database, making them permanent. See section 1.6, below, for a more detailed explanation of this. The function </font><b><font FACE="Courier" SIZE="3">commit</font></b><font SIZE="3"> returns a bool; this is </font><b><font FACE="Courier" SIZE="3">true</font></b><font SIZE="3"> in the initial Poly/ML session - the one in which </font><b><font FACE="Courier" SIZE="3">commit</font></b><font SIZE="3"> was called - but is </font><font FACE="Courier" SIZE="3"><b>false</b></font><font SIZE="3"> in any subsequent Poly/ML session started using the saved database.</font></p> <h2><a name="quitting1.4"><font SIZE="4"><b>1.4 Quitting a Poly/ML session</b></font></a></h2> <p><font SIZE="3">You can quit the Poly/ML session without updating the database by:</font></p> <p><font FACE="Courier" SIZE="3"><b>PolyML.quit ();</b></font></p> <p><font SIZE="3">As this does not update the database, all work done since the last commit will be lost. Alternatively, typing</font></p> <p><font FACE="Courier" SIZE="3"><b><control-D></b></font></p> <p><font SIZE="3">(or whatever is the local end-of-file indicator) performs a commit, then quits the current Poly/ML session.</font></p> <p><font SIZE="3">If you are running Poly/ML as part of a Unix batch job, it may be useful to set the Unix return code when quitting Poly/ML. This can be done using</font></p> <p><font FACE="Courier" SIZE="3"><b>PolyML.exit n;</b></font></p> <p><font SIZE="3">where <b>n </b>is an integer. This quits the Poly/ML session, without updating the database, and sets the Unix return code to <b>n</b> <b>mod 256</b></font></p> <p><b><font SIZE="3">Note: </font><font FACE="Courier" SIZE="3">PolyML.quit()</font></b><font SIZE="3"> is the same as </font><b><font FACE="Courier" SIZE="3">PolyML.exit 0</font></b><font SIZE="3"> but has been retained for backwards compatibility with previous versions of Poly/ML.</font></p> <h2><a name="including1.5"><font SIZE="4"><b>1.5 Including files </b></font></a></h2> <p><font SIZE="3">You can include pieces of ML program from external files using <b>PolyML. use:</b></font></p> <p><font FACE="Courier" SIZE="3"><b>PolyML.use "myfile";</b></font></p> <p><font SIZE="3">This will attempt to read the file </font><b><font FACE="Courier New" SIZE="3">myfile</font></b><font SIZE="3">. If </font><b><font SIZE="3" face="Courier">myfile</font></b><font SIZE="3"> does not exist, it will read </font><b><font SIZE="3" face="Courier">myfile.ML</font></b><font SIZE="3"> instead (and if that doesn't exist, it will try </font><font SIZE="3" face="Courier"><b>myfile.sml</b></font><font SIZE="3">). Included files may also contain nested calls to </font><font SIZE="3" face="Courier"><b>use</b></font><font SIZE="3">. File names are interpreted with respect to the current working directory. Initially, the current working directory is the same as the Unix directory in which the Poly/ML session was started. The current working directory may be changed using </font><font SIZE="3" face="Courier"><b>PolyML.cd</b></font><font SIZE="3">, for example:</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.cd "../examples";</b></font></p> <p><font SIZE="3">selects the '</font><font SIZE="3" face="Courier"><b>examples</b></font><font SIZE="3">'<b> </b>subdirectory of the parent directory of the old working directory as the new working directory.</font></p> <h2><a name="persist1.6"><font SIZE="4"><b>1.6 Persistence</b></font></a></h2> <p><font SIZE="3">Persistence means that values created in one Poly/ML session may be accessed in subsequent Poly/ML sessions without the need to explicitly reload them and without the need for recompilation.</font></p> <p><font SIZE="3">When Poly/ML values are initially created, they are transient (non-persistent). Transient values are stored in the local heap and vanish at the end of the Poly/ML session which created them. By contrast, persistent values are stored in a disk file called a database and are accessible by all future Poly/ML sessions. When a persistent value is used during a Poly/ML session, it is paged into memory using the Unix demand paging mechanism. The function <b>PolyML.commit </b>causes transient values to become persistent, by copying them from the local heap to the database.</font></p> <p><font SIZE="3">If a Poly/ML session attempts to modify a persistent object, the database is not updated directly. Instead, a copy of the modified object is made in the local heap. This means that the modification itself is transient - the current Poly/ML session will use the modified value, but subsequent sessions will see the old value, unless the database is updated using <b>PolyML. commit. </b>Once a modification has been committed, it is permanent; there is no way to revert to the earlier state of the database.</font></p> <h2><a name="debugging1.7"><font SIZE="4"><b>1.7 Debugging</b></font></a></h2> <p><font SIZE="3">Occasionally, an ML program will produce an unexpected result. When you attempt to discover the reason for this, the first thing you should do is to ensure that values are printed in enough detail to be useful. The amount of detail printed can be altered by </font><b><font SIZE="3" face="Courier">PolyML.print_depth</font></b><font SIZE="3">. For example</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.print_depth 10;</b></font></p> <p><font SIZE="3">will print sub-expressions nested up to 10 levels deep. Setting the print depth to 0 disables printing entirely.</font></p> <p><font SIZE="3">The amount of detail printed in ML compiler error messages and when exceptions are raised can be controlled by </font><font FACE="Courier New" SIZE="3"><b>PolyML.error_depth</b></font><font SIZE="3">.<b> </b>For example</font></p> <p><b><font SIZE="3" face="Courier">PolyML.error_depth 10;</font></b></p> <p><font SIZE="3">will change the level of detail to 10 (the default is 5).</font></p> <p><font SIZE="3">If an ML program raises an exception which gets propagated to the top level, it may be difficult to discover which function raised the exception. The function </font><b><font SIZE="3" face="Courier">PolyML.exception_trace</font></b><font SIZE="3"> is provided to solve this difficulty. For example, if we define</font></p> <p><font SIZE="3" face="Courier"><b>exception Badfact;<br> fun badfact 0 = raise Badfact<br> | badfact n = n * badfact(n-1); </b></font></p> <p><font SIZE="3">and then try to evaluate</font></p> <p><font SIZE="3" face="Courier"><b>map badfact [1,2,3];</b></font></p> <p><font SIZE="3">Poly/ML responds with the error message</font></p> <p><font SIZE="3" face="Courier"><b>Exception- Badfact raised<br> Exception failure raised</b></font></p> <p><font SIZE="3">which doesn't indicate which function is at fault. To discover where the exception was raised, we should evaluate</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.exception_trace(fn () => map badfact [1,2,3]);</b></font></p> <p><font SIZE="3">which prints a trace of the state of the stack at the time the exception was raised:</font></p> <p><font SIZE="3" face="Courier"><b>Exception trace for exception - Badfact<br> ML-badfact<br> ML-badfact<br> ML-map()<br> R<br> End of trace<br> Exception- Badfact raised<br> Exception failure raised</b></font></p> <p><font SIZE="3">If an ML program appears to be in an infinite loop, it can be interrupted by typing</font></p> <p><font SIZE="3" face="Courier"><b><a name="ControlC"></a><control-C></b></font></p> <p><font SIZE="3">You now have six options, each indicated by typing a single character ’</font><font SIZE="3" face="Courier"><b>?</b></font><font SIZE="3">’, ‘</font><font SIZE="3" face="Courier"><b>s</b></font><font SIZE="3">’, ‘</font><font SIZE="3" face="Courier"><b>t</b></font><font SIZE="3">’, ‘</font><font SIZE="3" face="Courier"><b>c</b></font><font SIZE="3">’, ‘</font><font SIZE="3" face="Courier"><b>f</b></font><font SIZE="3">’ or ‘</font><font SIZE="3" face="Courier"><b>q</b></font><font SIZE="3">’<b>. </b>Typing:</font></p> <p><font SIZE="3" face="Courier"><b>‘?’</b></font><font SIZE="3"> lists the other five options.</font></p> <p><b><font SIZE="3" face="Courier">‘s’</font></b><font SIZE="3"> switches from the current ML shell to the other. The second ML shell is distinguished by having <b>'2>' </b>as its prompt. The second shell inherits all declarations from the first shell, but any declarations made in it are local. When you are in one shell all processing in the other shell is suspended. It is sometimes useful to switch shells to change the value of </font><b><font SIZE="3" face="Courier">PolyML.print_depth</font></b><font SIZE="3"> and then switch shells back again to continue.</font></p> <p><b><font SIZE="3" face="Courier">‘t’</font></b><font SIZE="3"> prints a trace of the state of the execution stack.</font></p> <p><b><font SIZE="3" face="Courier">‘c’</font></b><font SIZE="3"> allows the execution of the program to continue normally.</font></p> <p><font SIZE="3" face="Courier"><b>‘f’ </b></font><font SIZE="3">continues the execution of the program, but with the exception </font><b><font SIZE="3" face="Courier">Interrupt</font></b><font SIZE="3"> raised. If, as is normally the case, the program doesn't handle this exception, the exception will be propagated to the top level and so halt the program.</font></p> <p><b><font SIZE="3" face="Courier">‘q’</font></b><font SIZE="3"> is a last resort - it quits the current Poly/ML session without updating the database (i.e. it is equivalent to </font><font SIZE="3" face="Courier"><b>PolyML.quit()</b></font><font SIZE="3">).</font></p> <h2><font SIZE="4"><b><a name="timing1.8"></a>1.8 Timing and Profiling</b></font></h2> <p><font SIZE="3">Poly/ML has a couple of functions to measure run-time efficiency. The first measures the amount of time taken to evaluate each top-level expression.</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.timing true;</b></font></p> <p><font SIZE="3">activates the timing facility. The time taken to evaluate each top-level ML expression will be printed after the expression's value. Typing</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.timing false;</b></font></p> <p><font SIZE="3">cancels this.</font></p> <p><font SIZE="3">The second function allows more detailed profiling of individual functions. Typing </font></p> <p><b><font SIZE="3" face="Courier">PolyML.profiling 1;</font></b></p> <p><font SIZE="3">profiles the amount of <b>user time </b>used by each function. Note that this figure excludes any <b>system time </b>that may have been used. After each top-level expression has been evaluated, a table is printed, showing how many ticks (of a 50 Hz clock) were spent inside each function. Typing</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.profiling 2;</b></font></p> <p><font SIZE="3">profiles the amount of <b>space </b>used by each function. The table printed shows how many words (1 word = 4 bytes) of storage were allocated by each function. Profiling for space increases program execution time by a large factor (about 10) due to the overhead of recording every storage request. Typing</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.profiling 3;</b></font></p> <p><font SIZE="3">profiles the number of <b>emulation traps </b>executed to support Poly/ML's arbitrary-precision arithmetic. This is important because each emulation trap requires the execution of an operating system trap-handler which, on some operating systems, may add significantly to the amount of system time used. Finally, typing</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.profiling 0;</b></font></p> <p><font SIZE="3">disables the profiling facility.</font></p> <h2><a name="logging1.9"><font SIZE="4"><b>1.9 Session Logging</b></font></a></h2> <p><font SIZE="3">Poly/ML provides a session logging facility. When this facility is enabled, it records the text of all successful declarations in a user-specified logfile. This allows the user to recreate the current environment in a subsequent Poly/ML session by compiling the logfile with </font><font SIZE="3" face="Courier"><b>PolyML.use</b></font><font SIZE="3">. By selectively turning the logging facility on and off, it is also possible to create a cleaned-up version of the recording.</font></p> <p><font SIZE="3">A declaration (or command) is successful if</font></p> <p><font SIZE="3">1. It successfully compiles.</font></p> <p><font SIZE="3">2. It executes without raising an exception.</font></p> <p><font SIZE="3">When a declaration has been compiled, the text of that declaration is stored in a log buffer. If the execution terminates without raising an exception the log buffer is written to the logfile; if not, the buffer is discarded. Executing </font><b><font SIZE="3" face="Courier">PolyML.writeLog</font></b><font SIZE="3"> (see below) will also cause the log buffer to be discarded.</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.isLogging ();</b></font></p> <p><font SIZE="3">returns <b><font face="Courier">true</font> </b>is session logging is currently enabled.</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.startLog filename;</b></font></p> <p><font SIZE="3">enables the session logging facility, opening </font><b><font SIZE="3" face="Courier">filename</font></b><font SIZE="3"> as a new logfile. If logging is already enabled, it raises the exception </font><b><font SIZE="3" face="Courier">Io "Already logging"</font></b><font SIZE="3">.</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.restartLog filename;</b></font></p> <p><font SIZE="3">is like </font><font SIZE="3" face="Courier"><b>startLog</b></font><font SIZE="3">,<b> </b>except that it opens </font><b><font SIZE="3" face="Courier">filename</font></b><font SIZE="3"> in append mode, allowing an existing logfile to be extended.</font></p> <p><font SIZE="3" face="Courier"><b>PolyML.stopLog();</b></font></p> <p><font SIZE="3">disables the logging facility and closes the currently active logfile. If logging is already disabled, it raises the<b> </b>exception </font><b><font SIZE="3" face="Courier">Io "Not logging"</font><font SIZE="3">.</font></b></p> <p><font SIZE="3" face="Courier"><b>PolyML.logName ();</b></font></p> <p><font SIZE="3">returns the name of the currently active logfile. If logging is currently disabled, it raises the exception </font><font SIZE="3" face="Courier"><b>Io "Not logging"</b></font><font SIZE="3">.</font></p> <p><b><font SIZE="3" face="Courier">PolyML.writeLog commandString;</font></b></p> <p><font SIZE="3">writes the string </font><b><font SIZE="3" face="Courier">commandString</font></b><font SIZE="3"> to the logfile. If logging is currently disabled, this function has no effect. The first call of this function within any top-level declaration also has the effect of discarding the temporary log buffer; this means that the current declaration will not be recorded in the logfile.</font></p> <p><font SIZE="3">The reason for this rather unexpected side-effect is to allow user-written interactive programs to write logs which allow their actions to be replayed in batch-mode. If the command which starts the interactive program were logged, then the batch file would require interactive user input. Instead the interactive program should use </font><b><font SIZE="3" face="Courier">writeLog</font></b><font SIZE="3"> to output a sequence of equivalent batch-mode commands.</font></p> <h2><a name="remotely1.10"><b><font SIZE="4">1.10 Using Poly/ML remotely</font></b></a></h2> <p><font SIZE="3">Programs which use the X Window System may run remotely; this means that the program may be run on one machine but display its output (with full windowing facilities - not just a 'dumb terminal') on a completely different machine.</font></p> <p><font SIZE="3">For example, suppose we have logged into the machine </font><b><font SIZE="3" face="Courier">modus</font></b><font SIZE="3">, which is running X. Then we could run the </font><b><font SIZE="3" face="Courier">baseCalc</font></b><font SIZE="3"> demo using</font></p> <p><b><font SIZE="3" face="Courier">modus% poly ML_dbase<br> > PolyML.use "baseCalc";</font></b></p> <p><font SIZE="3">which will pop-up a desk calculator widget, written in Poly/ML, using the </font><b><font SIZE="3" face="Courier">modus</font></b><font SIZE="3"> display. This runs the demo <i>locally.</i></font></p> <p><font SIZE="3">Note: if </font><b><font SIZE="3" face="Courier">modus</font></b><font SIZE="3"> is not running X, then running the demo will raise an exception.</font></p> <p><font SIZE="3">Now suppose that we are sitting in front of </font><b><font SIZE="3" face="Courier">modus</font></b><font SIZE="3"> but want to run Poly/ML on </font><b><font SIZE="3" face="Courier">ponens</font></b><font SIZE="3">. First we have to log into <b>ponens </b>and set the </font><b><font FACE="Courier New" SIZE="3">DISPLAY</font></b><font SIZE="3"> variable to tell X that we want our output to appear on </font><b><font FACE="Courier" SIZE="3">modus</font></b><font SIZE="3">.</font></p> <p><font FACE="Courier" SIZE="3"><b>modus% rlogin ponens<br> ponens% setenv DISPLAY modus:0.0</b></font></p> <p><font SIZE="3">If we now run the </font><b><font FACE="Courier" SIZE="3">baseCalc</font></b><font SIZE="3"> demo on </font><b><font FACE="Courier" SIZE="3">ponens</font></b><font SIZE="3">, the calculator widget will appear on </font><font FACE="Courier" SIZE="3"><b>modus</b></font><font SIZE="3">; we are running the demo remotely.</font></p> <p><font FACE="Courier" SIZE="3"><b>ponens% poly ML_dbase<br> > PolyML.use "baseCalc";</b></font></p> <p><font SIZE="3">Try it!</font></p> <h2><font SIZE="4"><b><a name="flags1.11">1.11 Flags</a></b></font></h2> <p><font SIZE="3">The Poly/ML driver program </font><b><font FACE="Courier" SIZE="3">poly</font></b><font SIZE="3"> accepts several flags from the Unix command line. Typing</font></p> <p><font FACE="Courier" SIZE="3"><b>unix% poly</b></font></p> <p><font SIZE="3">(with no database supplied) will print out a complete list of the available flags. The most important of these are the <b>-r </b>flag and the <b>-h </b>flag. Using the <b>-r </b>flag sets read-only mode. For example:</font></p> <p><font FACE="Courier" SIZE="3"><b>unix% poly -r ML_dbase</b></font></p> <p><font SIZE="3">will start Poly/ML using </font><b><font FACE="Courier" SIZE="3">ML_dbase</font></b><font SIZE="3">, but will not allow </font><b><font FACE="Courier" SIZE="3">PolyML.commit</font></b><font SIZE="3"> to update the database. Using the <b>-h </b>flag changes the maximum local heap size allowed in Poly/ML. For example: </font></p> <p><b><font FACE="Courier" SIZE="3">unix% poly -h 5000 ML_dbase </font></b></p> <p><font SIZE="3">will prevent the local heap from growing larger than 5000K bytes. The default (if no </font><b><font FACE="Courier" SIZE="3">-h</font></b><font SIZE="3"> flag is present) is to allow a maximum local heap size of 6144K. See <a href="HeapParms9.html">chapter 9</a> for a discussion of the heap flags.</font></p> <p><font SIZE="3">Using the </font><b><font FACE="Courier" SIZE="3">-noDisplay</font></b><font SIZE="3"> flag runs Poly/ML in non-X mode, disabling the X interface. When Poly/ML is run in non-X mode, any X call will raise the exception </font><b><font FACE="Courier" SIZE="3">XWindows.XWindows</font><font SIZE="3">.</font></b></p> <h3><a name="discgarb1.12"><font SIZE="4"><b>1.12 The disc garbage collector</b></font></a></h3> <p><font SIZE="3">Each update to the database increases its size, even if fewer objects are accessible than before the update. The solution to this problem is to run poly with the -d option.</font></p> <p><font FACE="Courier" SIZE="3"><b>unix% poly -d ML_dbase</b></font></p> <p><font SIZE="3">This program performs a complete garbage collection of the database in two phases. The first phase simply copies all of the objects from the database to local memory. This phase may be safely interrupted since it does not change the database. When all the objects are in local memory the database is removed and created again. The objects are then copied into it using a copying garbage collector. This phase must not be interrupted because it will leave the database in an incomplete state. Database garbage collection is very heavy on machine resources. It will need a large amount of virtual memory to contain the database, and will page very heavily unless provided with a lot of local memory.</font></p> <p><font SIZE="3">Performing this garbage collection has two benefits. Firstly it will place objects that refer to each other near to each other in the database to improve locality and to reduce the paging overhead of the system. Secondly, only those objects that are accessible are copied, so that the database is reduced to its minimum size.</font></p> <p><font SIZE="3">Alternatively, you can run the </font><font SIZE="3">program with the </font><b><font FACE="Courier" SIZE="3">-c</font></b><font SIZE="3"> flag:</font></p> <p><font FACE="Courier" SIZE="3"><b>unix% poly -d -c ML_dbase</b></font></p> <p><font SIZE="3">This adds an extra common-expression elimination phase to the garbage collection process. Whenever the garbage collector finds two immutable objects that represent the same value, it eliminates one of them from the database. Running with the -c<b> </b>flag roughly doubles <b>discgarb</b>'s virtual memory requirements but makes a useful additional contribution to reducing the size of the database. </font></p> <p><font size="2"><b>Copyright (c) 2000 CUTS and contributers.</b></font></p> </body> </html>
