4.1.2: - Fixed Cygwin DLL missing symbols problem. - Fixed GNU-stack assembler section on some platforms (Debian-ARM). - Fixed problem cause by include of . - Fixed SHA-384 and SHA-512 code Visual C++ compatibility. - Improved detection of IBM ICU library version; has to be >= 2.8. 4.1.1: - Fixed shared library version info. 4.1.0: - Added SHA-384 and SHA-512 algorithms. - Added HMAC-SHA-384 and HMAC-SHA-512 algorithms. - Added generic SSE2 optimization for the above algorithms. - Added more digest algorithms for PKCS#1 EMSA. - Optimized swap32 and swap64 routines on Linux. - Fixed missing definition in mpopt.h for s390x. - Fixed nostackexec configuration bug. - Fixed problem in Date::toString. - Fixed deadlock problem which occured in certain cases where security or crypto SPI constructor called getInstance for another security or crypto SPI. - Fixed a bug in the generic CBC encryption code; when called with nblocks == 1, the feedback was set incorrectly. - Fixed a bug in mpbsubmod; sometimes it takes multiple additions of the modulus to get a positive number. - Fixed PowerPC 64-bit configuration problem on Linux. 4.0.0: - Added a C++ API interface, modeled after Java's security & crypto API. - Added the new GNU noexecstack feature. - Added more x86_64 and s390x assembler routines. - Modified i2osp, so that it only requires as many octets as there are significant bytes in the multi-precision integers. - Fixed a bug in the creation of rsa keypairs; code was not correctly migrated to new calling sequence. The code now implements the method described in IEEE P.1363. - Fixed another bug in mpextgcd_w which sometimes returned incorrect results. - Fixed a bug in mprshiftlsz, which didn't work correctly when size = 1. - Fixed a configuration problem on Tru64 Unix. 3.1.0: - Added wiping of private key components of keypairs before freeing. - Fixed bug in mpextgcd_w which sometimes returned incorrect result. - Fixed error in PowerPC 64-bit assembler symbol definitions. 3.0.0: - Cleaned up installed header files. - Modified the API so that all keys can be passed as arrays of bytes. - Modified the API so that all key sizes are given in bits. - Modified the multi-precision integer library to work better on 64-bit machines. - Modified the assembly source generation mechanism, employing the m4 macro processor. - Added multi-precision integer vectorized assembler routines for Itanium. - Added multi-precision integer assembler routines for PowerPC 64-bit. - Added multi-precision integer assembler routines for Alpha. - Added multi-precision integer assembler routines for Opteron. - Added multi-precision integer assembler routines for IBM zSeries 64-bit. - Added multi-precision integer assembler routines for M68K. - Added Jeff Johnson's python bindings. - Added new unit tests. - Added new benchmarking programs. 2.3.0pre: - Modified the header files so that the library now uses self-contained autoconf-generated configuration files; a program employing BeeCrypt can now use the symbols already tested and defined instead of having to regenerate them (thus also eliminating the risk of inconsistencies). - Added the AES algorithm, with assembler routines for i586 and powerpc. - Added the DSA signature algorithm. - Added PowerPC assembler routines for blowfish. - Added Pentium4 SSE2 assembler multiplication routines. - Fixed the RSA CRT algorithm. - Fixed the gas/i386 mp32even and mp32odd routines. - Fixed a bug in modular inverse computation; thanks to Jeff Johnson of RedHat for pointing this out. - Fixed a bug in testing the result of a gcd operation in the mp32prndconone routine. - Fixed an ugly bug in base64 decoding. - Fixed compatibility with the latest automake & autoconf versions. - Replaces CPU optimization mechanism in configure script. 2.1.0: - Added support for automake, autoheader and libtool, which should make compiling the library even easier. - Changed DHAES API to conform to IEEE P.1363 submission and to allow for uneven key splitting. - Improved PKCS#5 padding routines. - Added a hash reset to the hashFunctionContextInit function. This was pointed out by Marko Kreen. - Fixed problem with configuring on i486-pc-linux-gnu. This was pointed out Steve O'Neill. - Fixed problem in the C version of mp32sub where carry would sometimes be missed. This was pointed out by Jon Sturgeon. - Revised entropy gathering system to do timeouts & asynchronous I/O where possible, to avoid hangs in case there's no noise on the audio device (i.e. digital silence), or when no data is available on devices such as /dev/random. - Changed mp32opt i386 assembler routines for slight performance improvement. - Changed mp32opt powerpc assembler routines for slight performance improvement. - Changed mp32opt sparcv9 assembler routines for slight performance improvement. - Added sparcv8 assembler routines for multi-precision integer multiplication. - Added arm assembler routines for multi-precision integer multiplication. - Added prototype 64-bit ia64 assembler routines for multi-precision integer operations. - Started writing the long-awaited documentation. 2.0.0: - Changed mp32barrett struct and operations to be multithread-safe; this required a change in API. - Changed hashFunction struct to incorporate internal block size parameter. - Changed HMAC algorithm and file names to match names in RFC 2104. - Changed SHA-1 C code for slightly faster results. - Changed detection of entropy devices. - Changed most void-returning functions to return int for error conditions. - Changed beecrypt-java class names in javaglue. - Added RSA keypair generation. - Added RSA private & public key operations. - Added SHA-256 hash function. - Added HMAC-MD5 and HMAC-SHA-256 keyed hash functions. - Added PKCS#5 padding. - Added DHAES encryption scheme. - Added Microsoft Visual C support, added Makefile.mak for this purpose. - Added Solaris/Sparc Forte C 64 bit support. - Added configure --disable-optimized option (disables assembler & processor-specific optimizations). - Fixed bug in SHA-1 assembler code for Pentium, where local variables were used below the current stack pointer; this could cause a problem if the routine was interrupted. This was pointed out by Richard Clayton. - Fixed bug in (certain cases of) modular inverse computation. - Fixed buffer overrun in base64 encoding. This was pointed out by Jon Sturgeon. - Fixed various minor bugs. - Renamed text files to match automake conventions. 1.1.2: - Fixed bugs in discrete logarithm domain parameter generator. The code to make a generator of order q and (p-1) was wrong. This was pointed out by Susumu Yamamoto. - Added MD5 hash function. 1.1.1: - Changed autoconfig script for easier porting. - Changed sources for easier compilation on Microsoft Visual C++; no assembler-optimization on this platform yet. - Fixed bug in javaglue when passing null IV to blockcipher. - Shared library is now linked dynamically, with shared object name and version. - Tested on Alpha Linux. - Tested on Alpha FreeBSD. - Added support for Compaq Alpha Tru64 Unix. - Added initial support for QNX. 1.1.0: - Added glue for interfacing from BeeCrypt Java Cryptography Provider. - Changed blockcipher struct to support interfacing with Java. - Added better blockcipher IV handling. - Multi-pass block processing is now possible with blockEncrypt/blockDecrypt. - Updated config.sub and config.guess to latest version from sources.redhat.com - Changed opening of entropy devices to blocking read-only mode instead of non-blocking read-write. - Added win32 'wincrypt' entropy source. - Added win32 'console' entropy source. - Added FreeBSD support. - Added PowerPC assembler optimized multiprecision subtraction routines. - Added initial ia64 support. - Added initial Darwin support (everything compiles, but the shared library doesn't build yet). 1.0.2: - Fixed Windows 2000 entropy bug; instead of using the first waveIn device, entropy now uses WAVE_MAPPER. - Added sparcv9 mp32addsqrtrc GNU assembler routine. - Added more hashFunctionContext and keyedHashFunctionContext functions. 1.0.1: - Added a sliding window modular exponentiation, about 30% faster than left-to-right exponentiation. - Fixed bugs in fips180opt.gas.i586.s (Linux SHA-1 assembler code for Pentium/Pentium Pro) - the Windows/Metrowerks version was okay. 1.0.0: - Added Win32 support; compiled as DLL with MetroWerks CodeWarrior Pro 5, it runs fine on Windows 95, 98, NT 4.0 (if you have a soundcard with a microphone port). Note that there is a know issue on Windows 2000, see BUGS. - Global code overhaul to support Win32 - Added more assembler routines, including SHA-1 for Pentium Pro (60% faster) - Added cleanup function to randomGenerator - Added missing functions in endianness.c - Fixed bug in entropy.c where devices might stay open - Eliminated mutex.h include file; it was more clear to do everything conditionally than to expand the macros in this file to encompass the Win32 API calls. 0.9.5: - Added PowerPC assembler optimization for multiprecision integers, 80% faster on our PowerMac 7200/90 - Fixed /dev/random entropy provider - Changed name SHA1 to SHA-1 in fips180 for consistency 0.9.4a: - Added missing file 'blowfishopt.o' 0.9.4: - Changes to configure script, to distinguish between different processors of the x86 family - Changes to blowfish code, 586/686 assembler optimization added, 30% faster on Pentium/PentiumPro - Changes to blowfish code, eliminated static blowfishSetupEncrypt; incorporated into regular encrypt - Changes to Makefile to selectively use blowfish assember code, depending on cpu type - Added missing routines 'mp32bzero' and 'mp32bnpowmod' to mp32barrett.c - Fixed 'const register' to 'register const' in mp32.c - Minor fixes in included header files 0.9.3: - Initial public release