deletions | additions       diff --git a/experim.tex b/experim.tex  index a8d0e34..908f198 100644  --- a/experim.tex  +++ b/experim.tex 
...
\begin{itemize}  \item {\bf Message 1}: how much does a never-firing OSR point impact code quality? We run a program with one or more OSR points, and we measure the slowdown given by factors such as cache effects (due to code bloat), register pressure, etc. due to the presence of the OSR points.  \item {\bf Message 2}: what is the overhead of an OSR transition to the same function? We run a program with a controlled OSR transition, e.g., with a counter that fires the OSR. Here we measure the impact of the actual OSR call [we already tried this with the repeated addition microbenchmark simple\_loop\_SSA.ll].  \item {\bf Message 3}: what is the overhead of the library for inserting OSR points? We compute: 1) the average time per OSR transition; 2) the number of transferred live variables; 3) the total benchmark time with an always-firing OSR at each iteration of the hottest loop; 4) the total benchmark time without OSR instrumentation (baseline).  \end{itemize}  \paragraph{Impact on code quality} 
...
For iterative benchmarks, we insert an OSR point in the body of their hottest loops. We classify a loop as hottest when its body is executed for a very high cumulative number of iterations (e.g., from a few thousands up to billions) and it either calls the method with the highest {\em self} time in the program, or it performs the most computational-intensive operations for the program in its own body. These loops are natural candidates for OSR point insertion, as they can be used - as in the Jikes RVM - to enable more dynamic inlining opportunities, with the benefits from several control-flow (e.g., dead code elimination) and data-flow (e.g., constant propagation) optimizations based on the run-time values of the live variables. In the shootout benchmarks, the number of such loops is typically 1 (2 for {\tt spectral-norm}).  For {\tt b-trees} - the only benchmark in our suite showing a recursive pattern - we insert an OSR point in the body of the method that accounts for the largest {\em self} execution time of the program. Such an OSR point might be useful to trigger recompilation of the code at a higher degree of optimization, or to enable some form of dynamic optimization (for instance, in a recursive search algorithm we might want to inline the comparator method provided by the user at the call).