These two methods expose weak fingerprints (SHA-1), and should be changed to SHA-256 or removed.
The argument in favor of changing to SHA-256 is that it allows an optimization where, if you have an X509Certificate, you can test equality by comparing the fingerprints first (e.g. X509Certificate::LessThan).
The argument in favor of removing it is that it's 64-bytes per X509Certificate object, of which there may be many, and the only major consumer of X509Certificate::LessThan is the CertStoreImpl, which is proposed to be removed (in favor of passing the full certificates to the renderer)
Comment 1 by est...@chromium.org
, May 20 2016