summaryrefslogtreecommitdiff
path: root/chacha_simd.cpp
Commit message (Collapse)AuthorAgeFilesLines
* Use Altivec as minimum ISA of ChaCha Simon64 and Speck64Jeffrey Walton2020-04-051-11/+4
|
* Add 64-bit overload for VecLoadAlignedJeffrey Walton2020-04-051-4/+5
|
* Add XLC 12 loads and stores for AIX (PR #907)Jeffrey Walton2019-10-261-5/+5
| | | Add XLC 12 loads and stores for AIX
* Fix ARM headers and Android compile (PR #896)Jeffrey Walton2019-10-161-3/+2
| | | | | | | | | | | * Test fix ARM headers This problem has been festering for some time. The header file includes are slightly different than the ISA options. Some platforms need an include, others don't. * Fix cryptest-android.sh and cryptest-ios.sh * Fix MSVC ARM32 and ARM64 compile * Split ARM32 and ARM64 recipes in GNUmakefile
* Fix missing _mm_roti_epi32 and _mm_roti_epi64 under GCC (GH #859)Jeffrey Walton2019-07-021-0/+3
|
* Rework SSE2 and AVX2 loads and storesJeffrey Walton2019-06-091-44/+40
|
* Revert changes for lgtm findingsJeffrey Walton2019-06-091-41/+44
| | | | This broke SunCC to the point of no repair. SunCC is using AVX2 instructions for C++ and SSE2. Man this compiler sucks...
* Clear lgtm findingsJeffrey Walton2019-06-081-44/+41
|
* Use PowerPC unaligned loads and stores with Power8 (GH #825, PR #826)Jeffrey Walton2019-04-271-5/+5
| | | Use PowerPC unaligned loads and stores with Power8. Formerly we were using Power7 as the floor because the IBM POWER Architecture manuals said unaligned loads and stores were available. However, some compilers generate bad code for unaligned loads and stores using `-march=power7`, so bump to a known good.
* Use carryless multiplies for NIST b233 and k233 curves (GH #783, PR #784)Jeffrey Walton2019-01-161-14/+2
| | | Use carryless multiplies for NIST b233 and k233 curves.
* Cleanup headers after Microsoft ARM64 portJeffrey Walton2019-01-041-2/+0
|
* Fix <arm_neon.h> include for ARM64 with MSVC compiler (GH #776)Jeffrey Walton2019-01-041-1/+2
|
* Add PACK32x4 macro guardJeffrey Walton2019-01-031-5/+7
|
* Fix ChaCha NEON compile with MSVC compiler (GH #776)Jeffrey Walton2019-01-031-1/+12
|
* Fix ChaCha compiler crash for GCC 3.3Jeffrey Walton2018-12-291-5/+10
|
* Update commentsJeffrey Walton2018-11-181-1/+3
|
* Add BLAKE2s and ChaCha CORE SIMD function (GH #656)Jeffrey Walton2018-11-181-3/+22
| | | | | The CORE function provides the implementation for ChaCha_OperateKeystream_ALTIVEC, ChaCha_OperateKeystream_POWER7, BLAKE2_Compress32_ALTIVEC and BLAKE2_Compress32_POWER7. Depending on the options used to compile the source files, either POWER7 or ALTIVEC will be used. This is needed to support the "new toolchain, ancient hardware" use case.
* Rename PPC vector functions from VectorFunc to VecFuncJeffrey Walton2018-11-151-192/+193
|
* Make Altivec vector wraps friendly to downgradesJeffrey Walton2018-11-151-84/+49
| | | | The way the existing ppc_simd.h is written makes it hard to to switch between the old Altivec loads and stores and the new POWER7 loads and stores. This checkin rewrites the wrappers to use _ALTIVEC_, _ARCH_PWR7 and _ARCH_PWR8. The wrappers in this file now honor -maltivec, -mcpu-power7 and -mcpu=power8. It allows users to compile a source file, like chacha_simd.cpp, with a lower ISA and things just work for them.
* Whitespace check-inJeffrey Walton2018-11-141-6/+6
|
* Drop ChaCha requirements to POWER7Jeffrey Walton2018-11-141-8/+29
| | | | This costs about 0.6 cpb (700 MB/s on GCC112), but it makes the faster algorithm available to more machines. In the future we may want to provide both POWER7 and POWER8
* Rename files with dashes to underscores (GH #736)Jeffrey Walton2018-11-101-0/+1104
Also see https://groups.google.com/forum/#!topic/cryptopp-users/HBz-6gZZFOA on the mailing list