3 I've been running the SPEC CPU 2006 benchmarks under valgrind (doing some
4 work on my BBV generating plugin).
6 There are two benchmarks that have issues, and I thought I'd share them
7 here for future reference.
9 1). zeusmp - does not run
11 It has a 1GB data segment, which valgrind cannot handle on a 32-bit
14 2). dealII - runs forever, never ending
16 It took a while, but I tracked this down to a 64bit/80bit
19 The code in the QGauss<1>::QGauss() function has some code like this:
21 const long double tolerance = std::max (static_cast<long double>
22 (std::numeric_limits<double>::epsilon() / 100),
23 static_cast<long double>(std::numeric_limits<long
24 double>::epsilon() *5));
30 } while (abs(p1/pp) > tolerance);
33 The tolerance in this case is being set to ~2.22e-18, but the
34 abs(p1/pp) value never gets below ~2.586e-17 under valgrind.
36 [This is because Valgrind only uses 64-bit FP values on x86, not 80-bit
39 This is similar to an issue that happens with the "art"
40 benchmark on SPEC CPU 2000, but in the "art" case it only
41 makes the code take longer to finish; this "dealII" problem
42 makes the benchmark loop forever.