mention --enable-memory-profiling configure option

git-svn-id: https://svn.r-project.org/R-dev-web/trunk@1985 c52295ea-58df-0310-926a-d16021944841

Author: tlumley

memory-profiling.html

@@ -22,13 +22,11 @@ <h1 align="center">Memory profiling in R</h1>
 the interval.  <code>summaryRprof</code> now has options to summarize this information. 
  This is available only on Unix at the moment.
 
-<li> If R is compiled with <code>R_MEMORY_PROFILING</code> defined, the function <code>tracemem</code> marks an object so that <code>Rf_duplicate</code> will print a stack trace when the object is duplicated. This is intended for tracking accidental copying of large objects. <code>untracemem</code> will untrace an object (though not all copies of it) and <code>tracingState</code> controls whether tracing information is printed. In the example below we see that <code>lm</code> does not duplicate its <code>data</code>  argument, but that <code>glm</code> does, and that <code>lm</code> does copy the response vector three times.
+<li> If R is compiled with <code>--enable-memory-profiling</code>, the function <code>tracemem</code> marks an object so that  a stack trace will be printed  when the object is duplicated, or when it is copied by coercion or arithmetic functions. This is intended for tracking accidental copying of large objects. <code>untracemem</code> will untrace an object (though not all copies of it) and <code>tracingState</code> controls whether tracing information is printed. In the example below we see that <code>lm</code> does not duplicate its <code>data</code>  argument, but that <code>glm</code> does, and that <code>lm</code> does copy the response vector.
 <pre>
 > data(trees)
-> ls()
-[1] "trees"
 > tracemem(trees)
-[1] "<0xa499a88>"
+[1] "<0x8bfff28>"
 > lm(log(Volume)~log(Height)+log(Girth),data=trees)
 
 Call:
@@ -38,9 +36,9 @@ <h1 align="center">Memory profiling in R</h1>
 (Intercept)  log(Height)   log(Girth)
      -6.632        1.117        1.983
 
-> glm(log(Volume)~log(Height)+log(Girth),data=trees)
-memtrace[0xa499a88->0x9f5c7a0]: glm
-memtrace[0x9f5c7a0->0x9f5c408]: glm
+>  glm(log(Volume)~log(Height)+log(Girth),data=trees)
+memtrace[0x8bfff28->0x8b6d820]: glm
+memtrace[0x8b6d820->0x89b4c10]: glm
 
 Call:  glm(formula = log(Volume) ~ log(Height) + log(Girth), data = trees)
 
@@ -51,20 +49,17 @@ <h1 align="center">Memory profiling in R</h1>
 Degrees of Freedom: 30 Total (i.e. Null);  28 Residual
 Null Deviance:      8.309
 Residual Deviance: 0.1855       AIC: -62.71
->  tracemem(trees$Volume)
-[1] "<0x9b45bc8>"
-> lm(Volume~Height+Girth,data=trees)
-memtrace[0x9b45bc8->0x9db3658]: eval eval model.frame.default model.frame eval eval lm
-memtrace[0x9db3658->0x9db39a0]: makepredictcall model.frame.default model.frame eval eval lm
-memtrace[0x9db3658->0x992b4a8]: <Anonymous> [[.data.frame [[ na.omit.data.frame na.omit model.frame.default model.frame eval eval lm
+> tracemem(trees$Volume)
+[1] "<0x895e230>"
+>  lm(Volume~Height+Girth,data=trees)
+memtrace[0x895e230->0x87a2898]: eval eval model.frame.default model.frame eval eval lm
 
 Call:
 lm(formula = Volume ~ Height + Girth, data = trees)
 
 Coefficients:
 (Intercept)       Height        Girth
    -57.9877       0.3393       4.7082
-
 </pre>
  <code>tracemem</code> cannot be used on functions, since it uses the same trace bit that <code>trace</code> uses, and will not work on objects such as environments that are passed by reference and not duplicated. The output for this could be made prettier and sent to a file: the main thing to decide is how to handle files when multiple objects may be being traced.
 
@@ -76,6 +71,6 @@ <h1 align="center">Memory profiling in R</h1>
 Valgrind can also help with memory profiling at the C level, using <code>--tool=massif</code>
 
 <hr>
-Thomas Lumley. 2006-5-10
+Thomas Lumley. 2006-8-1
 <p>