Hi Everyone, The fix for SHA::Transform was checked in at https://github.com/weidai11/cryptopp/commit/2aff92ddb6e679fc. Also see https://github.com/weidai11/cryptopp/issues/455.
SHA basically got overhauled. Formerly IterHashBase dispatched a single block to digested when Update() was called. IterHashBase also performed endian conversions. Eventually the static SHA::Transform was called to hash a single block at a time. We added freestanding HashMultipleBlocks that handles multiple blocks. Effectively, Multiple blocks is our core now, and not a single block. We have two or three variants of it - a CXX, a SHA-NI, and an ARM version. IterHashBase no longer handles dispatching. We added a class override for HashMultipleBlocks that calls into the free standing function. The freestanding HashMultipleBlocks also has a ByteOrder parameter to tell us how the incoming data is presented. Now we can select a mask on the fly as needed to arrange the data as required. SHA-1 is pushing data at 2.3 cpb on ARM servers, 4.0 cpb on ARM gadgets, and 1.8 cpb on desktops. The numbers look good. I think we've pulled all the performance out of them we can. Jeff -- -- You received this message because you are subscribed to the "Crypto++ Users" Google Group. To unsubscribe, send an email to cryptopp-users-unsubscr...@googlegroups.com. More information about Crypto++ and this group is available at http://www.cryptopp.com. --- You received this message because you are subscribed to the Google Groups "Crypto++ Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to cryptopp-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
