/usr/include/dieharder/rgb_lagged_sums.h is in libdieharder-dev 3.31.1-7.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 | /*
* This is an example header file for a test. For existing tests these
* headers are in the library includes already, but this one is an example
* suitable for use as a template.
*/
/*
* The function needs a prototype. In order to use the standard test
* creation/destruction/execution routines, the prototype should have
* precisely this form.
*/
int rgb_lagged_sums(Test **test,int irun);
/*
* This is default data for the test at hand. The first field is
* the test name. The second is the test description. The third
* is the default number of p-values generated by a run for display
* in a histogram and to generate a cumulative test p-value using
* Kuiper-Kolmogorov-Smirnov. The fourth is the number of "samples"
* accumulated per test, if relevant (some tests do not permit this
* to be varied). The fifth and final Dtest parameter is the number
* of statistics generated by the test (per test invocation) -- usually
* this will be one but for several it is two and could be more.
*/
static Dtest rgb_lagged_sums_dtest __attribute__((unused)) = {
"RGB Lagged Sum Test",
"rgb_lagged_sum",
"\
#==================================================================\n\
# RGB Lagged Sums Test\n\
# This package contains many very lovely tests. Very few of them,\n\
# however, test for lagged correlations -- the possibility that\n\
# the random number generator has a bitlevel correlation after\n\
# some fixed number of intervening bits.\n\
#\n\
# The lagged sums test is therefore very simple. One simply adds up\n\
# uniform deviates sampled from the rng, skipping lag samples in between\n\
# each rand used. The mean of tsamples samples thus summed should be\n\
# 0.5*tsamples. The standard deviation should be sqrt(tsamples/12).\n\
# The experimental values of the sum are thus converted into a\n\
# p-value (using the erf()) and a ks-test applied to psamples of them.\n\
#==================================================================\n",
100,
1000000,
1,
rgb_lagged_sums,
0
};
|