| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
|
|
|
|
|
|
|
| |
Move m_aliasBlock into Rijndael::Base. m_aliasBlock is now an extra data member for Dec because the aliased table is only used for Enc when unaligned data access is in effect. However, the SecBlock is not allocated in the Dec class so there is no runtime penalty.
Moving m_aliasBlock into Base also allowed us to remove the Enc::Enc() constructor, which always appeared as a wart in my eyes. Now m_aliasBlock is sized in UncheckedSetKey, so there's no need for the ctor initialization.
Also see https://stackoverflow.com/q/46561818/608639 on Stack Overflow. The SO question had an unusual/unexpected interaction with CMake, so the removal of the Enc::Enc() ctor should help the problem.
|
| |
|
|
|
|
| |
The problem was vec_sld is endian sensitive. The built-in required more than us setting up arguments to ensure the vsx load resulted in a big endian value. Thanks to Paul R on Stack Overflow for sharing the information that IBM did not provide. Also see http://stackoverflow.com/q/46341923/608639
|
|
|
|
| |
Arg... GetUserKey was performing a 32-bit word reverse. It was part of the problem on little endian machines
|
|
|
|
| |
We are going to attempt to perform key setup using Power8 in-core vector instructions
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
machines
The refactoring has no effect on little endian machines. However, on big endian GCC119 using GCC 7.1 the performance improved by 2.5x for ECB and CTR modes:
BEFORE:
<TR><TH>AES/CTR (128-bit key)<TD>2723<TD>1.4<TD>0.163<TD>670
<TR><TH>AES/CTR (192-bit key)<TD>2560<TD>1.5<TD>0.175<TD>719
<TR><TH>AES/CTR (256-bit key)<TD>2728<TD>1.4<TD>0.183<TD>749
<TR><TH>AES/CBC (128-bit key)<TD>1204<TD>3.2<TD>0.135<TD>554
<TR><TH>AES/CBC (192-bit key)<TD>1066<TD>3.7<TD>0.148<TD>605
<TR><TH>AES/CBC (256-bit key)<TD>948<TD>4.1<TD>0.155<TD>635
<TR><TH>AES/OFB (128-bit key)<TD>1019<TD>3.8<TD>0.158<TD>648
<TR><TH>AES/CFB (128-bit key)<TD>949<TD>4.1<TD>0.192<TD>787
<TR><TH>AES/ECB (128-bit key)<TD>3564<TD>1.1<TD>0.082<TD>337
AFTER:
<TR><TH>AES/CTR (128-bit key)<TD>6484<TD>0.6<TD>0.163<TD>677
<TR><TH>AES/CTR (192-bit key)<TD>5641<TD>0.7<TD>0.176<TD>728
<TR><TH>AES/CTR (256-bit key)<TD>5005<TD>0.8<TD>0.183<TD>761
<TR><TH>AES/CBC (128-bit key)<TD>1223<TD>3.2<TD>0.135<TD>559
<TR><TH>AES/CBC (192-bit key)<TD>1080<TD>3.7<TD>0.147<TD>611
<TR><TH>AES/CBC (256-bit key)<TD>966<TD>4.1<TD>0.155<TD>642
<TR><TH>AES/OFB (128-bit key)<TD>1057<TD>3.7<TD>0.158<TD>656
<TR><TH>AES/CFB (128-bit key)<TD>1217<TD>3.3<TD>0.186<TD>774
<TR><TH>AES/ECB (128-bit key)<TD>7289<TD>0.5<TD>0.082<TD>342
|
|
|
|
| |
This increases performance to about 1.6 cpb. We are about 0.5 cpb behind Botan, and about 1.0 cpb behind OpenSSL. However, it beats the snot out of C/C++, which runs at 20 to 30 cpb
|
| |
|
| |
|
|
|
|
| |
This is the forward direction on encryption only. Crypto++ uses the "Equivalent Inverse Cipher" (FIPS-197, Section 5.3.5, p.23), and it is not compatible with IBM hardware. The library library will need to re-work the decryption key scheduling routines. (We may be able to work around it another way, but I have not investigated it).
|
|
|
|
|
| |
The strategy of "cleanup under-aligned buffers" is not scaling well. Corner cases are still turing up. The library has some corner-case breaks, like old 32-bit Intels. And it still has not solved the AltiVec and Power8 alignment problems.
For now we are backing out the changes and investigating other strategies
|
|
|
|
| |
This commit supports the upcoming AltiVec and Power8 processor. This commit affects a number of classes due to the ubiquitous use of AES. The commit adds debug asserts to warn of under-aligned and misaligned buffers in debug builds.
|
|
|
|
| |
This commit supports the upcoming AltiVec and Power8 processor. This commit affects a number of classes due to the ubiquitous use of AES. The commit provides the data alignment requirements.
|
| |
|
|
|
|
| |
Currently the CRYPTOPP_BOOL_XXX macros set the macro value to 0 or 1. If we remove setting the 0 value (the #else part of the expression), then the self tests speed up by about 0.3 seconds. I can't explain it, but I have observed it repeatedly.
This check-in prepares for the removal in Upstream master
|
| |
|
|
|
|
|
| |
#461)
Split source files to support Base Implementation + SIMD implementation
|
| |
|
|
|
|
|
|
|
|
| |
Reworked SHA class internals to align all the implementations. Formerly all hashes were software based, IterHashBase handled endian conversions, IterHashBase repeatedly called the single block SHA{N}::Transform. The rework added SHA{N}::HashMultipleBlocks, and the SHA classes attempt to always use it.
Now SHA{N}::Transform calls into SHA{N}_HashMultipleBlocks, which is a free standing function. An added wrinkle is hardware wants little endian data and software presents big endian data, so HashMultipleBlocks accepts a ByteOrder for the incoming data. Hardware based SHA{N}_HashMultipleBlocks can often perform the endian swap much easier by setting an EPI mask so it was profitable to defer to hardware when available.
The rework also removed the hacked-in pointers to implementations. The class now looks more like AES, GCM, etc.
|
| |
|
| |
|
|
|
|
|
| |
When MSVC init_seg or GCC init_priority is available, we don't need to use the Singleton. We only need to create a file scope class variable and place it in the segment for MSVC or provide the attribute for GCC.
An additional upside is we cleared all the memory leaks that used to be reported by MSVC for debug builds.
|
| |
|
| |
|
|
|
|
| |
We added asserts due to Coverity findings. We beieve the findings were false positives
|
| |
|
| |
|
| |
|
|
|
|
| |
instead of 256+ bytes. Search within that 4K space to put the 256-byte aligned Locals struct in a place which does not have 4K cache conflicts with the Te temporary buffer. This permits us to call _malloca() or alloca() once per call of this function. This commit also makes sure that the Microsoft-only _freea() occurs at the correct location instead of at a pointer to the middle of the stack, when the memory allocated by _malloca() or alloca() is not 256-byte aligned.
|
|
|
|
| |
trap.h and CRYPTOPP_ASSERT has existed for over a year in Master. We deferred on the cut-over waiting for a minor version bump (5.7). We have to use it now due to CVE-2016-7420
|
| |
|
| |
|
|
|
|
| |
std::swap(__m128i, __m128i)
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
In a 32-bit Windows program compiled with Visual C++ 2013 Update 5, we
sometimes get crashes because of an exception "0xC000001D: Illegal Instruction"
on the pextrd (_mm_extract_epi32) instruction. Explicitly check for SSE4
support instead of only AES-NI before using this SSE4.1 instruction.
|
| |
|
| |
|
|
|
|
| |
This reverts commit 5cce8c33cabd92afecfb0ec31f841d3082d15b3d. It was an incomplete remediation because it was applied to one area of the code, and not both areas of the code. The complete remediation will be checked-in next.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
platforms when CRYPTOPP_NO_UNALIGNED_DATA_ACCESS defined
|
| |
|
|
|
|
| |
appear to have entered on November 15, 2015 with commit 48809d4e85c12581
|