This is a generic bug for improving various code chunks in PV wrt floating point handling (e.g. comparison). One specific issue in practice was #1149, but recently another problem occurred with hantek-4032l voltage thresholds (same reason as #1149, a different fix will be needed). Relevant URLs: https://randomascii.wordpress.com/2012/02/25/comparing-floating-point-numbers-2012-edition/ https://www.boost.org/doc/libs/1_64_0/libs/math/doc/html/math_toolkit/float_comparison.html
The compiler option -Wfloat-equal can be used to find all (or at least most?) issues related to direct floating point comparison via "==". Other FP-related issues in addition to those might exist in the code-base, though.