Hi Everyone, We cleared the last four known (and duplicatable) issues in Crypto++ 5.6.3. The last two were tricky because they involved bad code generation due to optimizations and inlining. One was also hardware, compiler and platform specific, so it was not easy to track down. In fact, I spent about two weeks trying to duplicate Debian's SHA hang (also see https://github.com/weidai11/cryptopp/issues/47).
It turned out we needed real Core2 Duo hardware and a particular version of Debian Sid installed in a particular way. László Böszörményi provided a tarball of his configuration, and we supplied the Core2 Duo hardware. It turned out Debian's issue was related to an issue reported by Zireael N and problems with inlined code for `ResizeBuffer` from `modes.h` (https://github.com/weidai11/cryptopp/pull/46). When Zireael helped clear his issue, the Debian problem magically disappeared. Crypto++ 5.6.3 has some residual undefined behavior, and folks can elect to remove it by *undefining* MAINTAIN_BACKWARDS_COMPATIBILITY_562. We are going to place one last RC for everyone to take stabs at, and then release 5.6.3. We are also going to release a Crypto++ 5.7 that's nearly identical to 5.6.3. The difference will be the undefined behavior is removed by default in 5.7. Otherwise, there's no difference. After that, we will move directly into 6.0 and the interface changes. Does anyone have any thoughts or objections? -- -- 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.
