aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* More ref manual changeslloyd2010-06-141-179/+175
|
* Mention options for getting the moduleslloyd2010-06-141-17/+13
|
* Some reference manual updates. Still needs a lot of work.lloyd2010-06-141-298/+97
|
* Tick to 1.9.9-devlloyd2010-06-143-5/+7
|
* Remove -dev suffix from version1.9.8lloyd2010-06-141-1/+1
|
* Update log, readme for 1.9.8 releaselloyd2010-06-142-2/+2
|
* Expose a datestamp in build.h, provide function in version.h forlloyd2010-06-145-1/+19
| | | | getting runtime value.
* Use strncpy instead of strcpy in EGD readerlloyd2010-06-131-1/+1
|
* Change how alloc_mmap's TemporaryFile class works. Don't exposelloyd2010-06-131-10/+17
| | | | | | | | | | | | | | | | | | | | | | the name at all; instead unlink it at the end of the constructor, so by the time it is fully constructed it is purely an anonymous file descriptor. mkstemp has a weird interface and returns the final name of the file in its template argument. This prevented us from using a std::string, since c_str's return is const (and we can't use &string[0], because that might not be NULL-terminated). This previously required doing nasty things like explicit new/delete and using strcpy (the strcpy was what got me started on looking at this; OpenBSD complains about it, so I was trying to figure out a good way to remove it). Instead, use the idea from http://www.gotw.ca/gotw/042.htm, and use a std::vector to hold the mkstemp argument/result. That works consistently everywhere, and we don't need to rely on strcpy, and don't have to worry about memory leaks either. Only minor nit is having to add an explicit NULL terminator as the std::string doesn't contain it.
* Rename the --use-boost-python optin to --with-boost-python, and add alloyd2010-06-133-2/+8
| | | | | | | --without-boost-python to explicitly disable it. This makes it much easier to use at least in Gentoo's ebuild system, and perhaps with other packaging systems as well.
* Demo to_ber in the RSA example/test scriptlloyd2010-06-111-2/+15
|
* Add typedefs for the named EME encryptor/decryptor classes so codelloyd2010-06-111-0/+6
| | | | using the 1.8 names continues to work.
* Change how install_name is set on OS X. Not tested, taken from patchlloyd2010-06-111-1/+1
| | | | used by MacPorts; I assume they know what they are doing.
* Oops. On a private key, call PKCS8::encode so the full key is exportedlloyd2010-06-111-1/+1
|
* Add to_ber to the RSA objects so you can get the raw BER encoding.lloyd2010-06-111-0/+20
| | | | Requested by Thomas Capricelli.
* merge of '311b4c5a7dea00ed2291aa6d8e76aa66e4460699'lloyd2010-06-111-1/+3
|\ | | | | | | and '50fe98d85f1fbe231eaf5b71ae727cd1d87f414b'
| * Fix problem that prevented AES-NI code from getting loaded in at leastlloyd2010-06-111-1/+3
| | | | | | | | | | | | | | some cases. Add a westmere alias for "Core i5 CPU M 520", which is what uname (and thus, platform.processor()) returns on my laptop. Mostly for my benefit of course.
* | Work around GCC 3.x bug with reverse iteratorslloyd2010-06-111-2/+6
|/
* Document new code for wide multiply under VClloyd2010-06-111-0/+1
|
* Include generic mp_asmi.h for MSVClloyd2010-06-113-1/+3
| | | | | | | Don't use /EHc; it says "C" functions are nothrow, which is not true for bigint_sub2_rev. Include needed <intrin.h> for mp_asm.h
* Have to add it as explicit dep in bigintlloyd2010-06-111-1/+1
|
* Add (untested) support for VC++'s _umul128 intrinsic, which apparentlylloyd2010-06-112-0/+76
| | | | works on both x86-64 and ia64. Will allow using 64-bit limbs on Windows.
* For 64-bit InnoSetuplloyd2010-06-101-1/+5
|
* Improvments to the build instructions:lloyd2010-06-101-25/+38
| | | | | | | | | | | | | | | | | | Note that if you want to use Python 3, you have to use 2to3. This has come up a couple of times on the list. Add make check to the 'for the impatient' instructions. Mention --no-autoload. Said that we would guess the compiler; this isn't true with the Python version. Add examples of using ICC and Clang. Note that prebuilt Windows binaries are sometimes available. Talk about make install for Windows. Fix version #s in pkg-config output.
* Add a simple Windows install targetlloyd2010-06-101-5/+3
|
* Fix comparison to use IVs with a hypothetical negotiated TLS that useslloyd2010-06-092-2/+2
| | | | a larger major version #.
* merge of '0ab1e77862bca53dec5ac1f2f9dbe994378e91f7'lloyd2010-06-081-2/+4
|\ | | | | | | and 'b9e4e0dcc98d3266c2d7e4fd631038babdfd933b'
| * In BigInt::bits, cache sig_words() result instead of calling twicelloyd2010-06-071-2/+4
| |
* | Exclude Python/Perl wrappers from Doxygen outputlloyd2010-06-071-1/+1
| |
* | Make Filter::new_msg and finish_msg private; only used by Pipe, which is a ↵lloyd2010-06-071-12/+13
| | | | | | | | friend
* | Add Doxygen header comments for XTS modeslloyd2010-06-071-4/+4
| |
* | Fix buildlloyd2010-06-071-0/+1
| |
* | Use "/*" instead of "/**" in starting comments at the begining of a file.lloyd2010-06-0785-97/+96
|/ | | | | This caused Doxygen to think this was markup meant for it, which really caused some clutter in the namespace page.
* Small cleanupslloyd2010-06-021-9/+8
|
* Document fixing 512/768 bit DL groups via DSA kosherizerlloyd2010-06-021-0/+1
|
* Show inherited members; makes Doxygen output much more sensiblelloyd2010-06-021-1/+1
|
* Add constructor and destructor for pipe_wrapper to handle init and closelloyd2010-06-021-5/+5
|
* Put PKCS hash ids in anon namespacelloyd2010-06-021-21/+22
|
* OpenBSD doesn't have MSG_NOSIGNAL; you need to set up a signal handlerlloyd2010-06-011-1/+0
| | | | | | | | to catch SIGPIPE instead. Simply avoid building the unix_socket module there. Yet another reason to move to a fully async/event-based interface that doesn't interact with sockets directly.
* If you didn't specify a qbits for the DSA kosherizer, then it wouldlloyd2010-05-281-19/+20
| | | | | | choose 256 bits unless the pbits was exactly 1024. That would mean you for pbits = 512/768, the FIPS 186-3 size check would fail and it wouldn't work. Pointed out by Rickard Bellgrim.
* Hid --enable-isa and instead expose --enable-{sse2,ssse3,aes-ni,altivec}lloyd2010-05-262-10/+10
| | | | | | | | | | | | in the help. Unfortunately we can't just remove --enable-isa, because for the callback to work the target list has to already exist, and it only does by virtue of the default=[] param to the enable-isa setup. We could just use append_const, except then we can't run on Python 2.4, and the latest release of RHEL only has 2.4 :( Rename aes_ni to aes-ni in configuration-speak
* Forming TARGET_ARCH, filter characters as with TARGET_CPU, otherwiselloyd2010-05-261-2/+3
| | | | bogus macros will result if there is a dash in the arch name (eg HITACHI-SH)
* Remove FORK-256; it's obscure and has been definitively broken.lloyd2010-05-259-1136/+2
| | | | | More commentary posted to the list: http://lists.randombit.net/pipermail/botan-devel/2010-May/001123.html
* Change BlockCipher::parallelism() to return the native parallelism oflloyd2010-05-2511-27/+31
| | | | | | | | | | | | | | | | | | | | the implementation rather than the preferred one. Update all implementations. Add a new function parallel_bytes() which returns parallelism() * BLOCK_SIZE * BUILD_TIME_CONSTANT This is because i noticed all current calls of parallelism() just multiplied the result by the block size already, so this simplified that code. The build time constant is set to 4, which was the previous default return value of parallelism(). However the SIMD versions returned 2*native paralellism rather than 4*, so this increases the buffer sizes used for those algorithms. The constant multiple lives in buildh.in and build.h, and is named BOTAN_BLOCK_CIPHER_PAR_MULT.
* Add a couple of small patches from Thomas Capricelli <[email protected]>lloyd2010-05-213-14/+27
| | | | that enable botan to be built under the clang C++ compiler.
* Use -dumpversion instead of -v for detecting the GCC version. Thislloyd2010-05-201-7/+8
| | | | | | | | | | | | work at least as far back as 3.2.3. And GCC's manual for -dumpversion says "Print the compiler version (for example, `3.0')", which suggests it works at least that far back, which is good enough (almost certainly GCC 2.95 can't compile botan, in any case). Also print the detected version. Make an error (being unable to executed GCC) a warning instead of a info message.
* Replace print to stderr with logging.error call on unhandled exception.lloyd2010-05-201-1/+1
| | | | This should work with both Python 2 and Python 3.
* merge of '540ae85af1cc9245c325ef716fcc5c5b334251d0'lloyd2010-05-1922-84/+60
|\ | | | | | | and 'ce3d40d9f2e90346189ca6dfed2a1f38804d5c10'
| * Note other changeslloyd2010-05-191-0/+3
| |
| * Add a build.h macro BOTAN_GCC_VERSION which is set to major*100+minor*10+patchlloyd2010-05-132-2/+10
| | | | | | | | | | | | | | | | if we are compiling under GCC, or 0 otherwise. Use it in cpuid.cpp for use of GCC's cpuid.h header file. If we don't have a method of calling cpuid, print a warning.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-29 18:27:28 +0000