bkleiner
commented
2 years ago huh, good catch! should not be fatal cause the filterstate should land in zero-initialized memory anyways, so merging to develop should suffice.
Closed nopjmp closed 2 years ago
bkleiner
commented
2 years ago huh, good catch! should not be fatal cause the filterstate should land in zero-initialized memory anyways, so merging to develop should suffice.
While reviewing the code to familiarize myself with it, I found the second pass in both develop and master is it initialized using filter_lp_pt1_init. I've fixed this on develop, but it could be easily fixed in the master branch in the same fashion.