"... In PAGE 4: ... Green Thread has the limitation that it can not be used for I/O devices that do not support asynchronous or nonblocking I/O fa- cilities. Execution time for several benchmarks Table4 presents the execution time for several well- known Java Benchmarks. The execution times vary by up to 48%, depending on the multi-threading mecha- nism.... ..."

"... In PAGE 11: ... We have evaluated the impact of the di erent annotations on the execution time by running a series of benchmarks (brie y described in Table 2) in parallel. Table3 shows the speedups obtained with respect to the sequential execution, i.... In PAGE 11: ... In order to make the annotation unconditional (as the rest of the annotators we are dealing with), we simply removed the conditional parallelism in the places where it was not being exploited. This is why it appears in Table3 under the name UMEL. All the benchmarks executed were parallelized automatically by CiaoPP, starting from their sequential code.... In PAGE 12: ... Rather, our main focus of attention is in the comparison among the speedups obtained using di erent annotators. A rst examination of the experimental results in Table3 allows inferring that in no case is UUDG worse than any other annotator, and in no case is UOUDG worse than (U)MEL. They should therefore be the annotators of choice... ..."

"... In PAGE 11: ... We have evaluated the impact of the di erent annotations on the execution time by running a series of benchmarks (brie y described in Table 2) in parallel. Table3 shows the speedups obtained with respect to the sequential execution, i.... In PAGE 11: ... In order to make the annotation unconditional (as the rest of the annotators we are dealing with), we simply removed the conditional parallelism in the places where it was not being exploited. This is why it appears in Table3 under the name UMEL. All the benchmarks executed were parallelized automatically by CiaoPP, starting from their sequential code.... In PAGE 12: ... Rather, our main focus of attention is in the comparison among the speedups obtained using di erent annotators. A rst examination of the experimental results in Table3 allows inferring that in no case is UUDG worse than any other annotator, and in no case is UOUDG worse than (U)MEL. They should therefore be the annotators of choice... ..."

"... In PAGE 27: ... The machine we used is a Sun UltraSparc T2000 (a Niagara) with 8 4-thread cores. In the performance results shown in Table2 , we did not use more than 8 cores since in that case, and due to access to shared units, speedups are sublinear even for completely independent tasks. The fork-join annotators we chose to compare with are MEL (Muthukumar et al.... In PAGE 28: ... In order to make the annotation unconditional (as the rest of the annotators we are dealing with), we simply removed the conditional parallelism in the places where it was not being exploited. This is why it appears in Table2 under the name UMEL. All the benchmarks executed were parallelized automatically by CiaoPP, start- ing from their sequential code.... In PAGE 29: ... Rather, our main fo- cus of attention is in the comparison among the speedups obtained using di erent annotators. A rst examination of the experimental results in Table2 , and also in Figure 12 allows inferring that in no case is UUDG worse than any other annotator, and in no case is UOUDG worse than (U)MEL. They should therefore be the annotators of choice if available.... ..."

"... In PAGE 9: ... Therefore, march eq does not perform any equivalence reasoning if no equivalence clauses are detected during the pre-processing phase (if no CoE exists), making march eq equivalent to its older brother march. Table2 shows that the integration of equivalence reasoning in march rarely results in a loss of performance: on some benchmarks like the random unsat and the quasigroup family no performance difference is noticed, since no equiv- alence clauses were detected. Most families containing equivalence clauses are solved faster due to the integration.... ..."

"... In PAGE 14: ... We have evaluated the impact of the different annotations on the execution time by running a series of benchmarks in parallel. Table1 shows the speedups obtained with respect to the sequential execution when using from 1 to 8 threads. The machine we used is a Sun UltraSparc Report No.... In PAGE 15: ... In order to make the annotation unconditional (as the rest of the annotators we are dealing with), we simply removed the conditional parallelism in the places where it was not being exploited. This is why it appears in Table1 under the name UMEL. The test programs we used are the following: 3We did not use more than 8 processors since in that case, and due to data contention and access to shared processor units, we have observed speedups to be sublinear (and difficult to predict) even for completely independent tasks.... In PAGE 16: ....e., computed w.r.t. the speed of the sequential version on one processor (hence the speedups are sometimes below 1 for one processor of the parallelized versions). A first examination of the experimental results in Table1 supports the discussion in Sec- tion 3.3: in no case UUDG is worse than any other annotator, and in no case UOUDG is worse than (U)MEL.... ..."