Home
last modified time | relevance | path

Searched refs:AVX2 (Results 1 – 22 of 22) sorted by relevance

/freebsd-12-stable/crypto/openssl/doc/man3/
DOPENSSL_ia32cap.pod93 =item bit #64+5 denoting availability of AVX2 instructions;
121 ':~0x20' would disable AVX2 code paths, and ':0' - all post-AVX
133 AVX2 | 2.22 | 2.10 | 3.1
/freebsd-12-stable/contrib/llvm-project/clang/lib/Basic/Targets/
DX86.cpp334 .Case("+avx2", AVX2) in handleTargetFeatures()
762 case AVX2: in getTargetDefines()
795 case AVX2: in getTargetDefines()
944 .Case("avx2", SSELevel >= AVX2) in hasFeature()
DX86.h61 AVX2, enumerator
/freebsd-12-stable/contrib/llvm-project/llvm/lib/Target/X86/
DX86Subtarget.h62 NoSSE, SSE1, SSE2, SSE3, SSSE3, SSE41, SSE42, AVX, AVX2, AVX512F enumerator
637 bool hasAVX2() const { return X86SSELevel >= AVX2; } in hasAVX2()
DX86.td124 def FeatureAVX2 : SubtargetFeature<"avx2", "X86SSELevel", "AVX2",
125 "Enable AVX2 instructions",
412 // Gather is available since Haswell (AVX2 set). So technically, we can
413 // generate Gathers on all AVX2 processors. But the overhead on HSW is high.
DX86InstrFormats.td767 // AVX2 Instruction Templates:
768 // Instructions introduced in AVX2 (no SSE equivalent forms)
770 // AVX28I - AVX2 instructions with T8PD prefix.
771 // AVX2AIi8 - AVX2 instructions with TAPD prefix and ImmT = Imm8.
DX86CallingConv.td225 // since the boolean types in AVX/AVX2 are promoted by default.
546 // since the boolean types in AVX/AVX2 are promoted by default.
839 // since the boolean types in AVX/AVX2 are promoted by default.
DX86Schedule.td488 // AVX2.
DX86SchedSandyBridge.td572 // AVX2/FMA is not supported on that architecture, but we should define the basic
DX86InstrSSE.td539 // have AVX2. Execution domain fixing will convert to integer if AVX2 is
4985 // AVX2 Patterns
5032 // AVX2 Register-Memory patterns
6986 // AVX2 adds register forms
7423 // AVX2 Instructions
7426 /// AVX2_blend_rmi - AVX2 blend with 8-bit immediate
DX86SchedBroadwell.td585 // AVX2.
DX86SchedSkylakeClient.td585 // AVX2.
DX86ScheduleZnver3.td1359 // AVX2.
DX86SchedSkylakeServer.td586 // AVX2.
/freebsd-12-stable/sys/contrib/libsodium/
Dconfigure.ac451 AC_MSG_CHECKING(for AVX2 instructions set)
454 # error NativeClient detected - Avoiding AVX2 opcodes
464 AC_DEFINE([HAVE_AVX2INTRIN_H], [1], [AVX2 is available])
469 # error NativeClient detected - Avoiding AVX2 opcodes
DChangeLog100 - An AVX2 optimized implementation of the Argon2 round function was
125 - AVX2 implementations of Salsa20 and ChaCha20 have been added. They
141 - AVX2 detection is more reliable.
161 - AVX2 detection was fixed, resulting in faster Blake2b hashing on
188 - A speed-record AVX2 implementation of BLAKE2b was added (thanks to
/freebsd-12-stable/contrib/llvm-project/llvm/lib/Analysis/
DVFABIDemangling.cpp40 .Case("d", VFISAKind::AVX2) in tryParseISA()
/freebsd-12-stable/contrib/llvm-project/llvm/include/llvm/Support/
DX86TargetParser.def114 X86_FEATURE_COMPAT(AVX2, "avx2")
/freebsd-12-stable/contrib/llvm-project/llvm/include/llvm/Analysis/
DVectorUtils.h49 AVX2, // x86 AVX2 enumerator
/freebsd-12-stable/contrib/llvm-project/clang/include/clang/Basic/
DBuiltinsX86.def559 // AVX2
/freebsd-12-stable/contrib/llvm-project/llvm/include/llvm/IR/
DIntrinsicsX86.td1333 // AVX2
/freebsd-12-stable/crypto/openssl/
DCHANGES1508 There is an overflow bug in the AVX2 Montgomery multiplication procedure
1519 This only affects processors that support the AVX2 but not ADX extensions
3475 *) Support for new and upcoming Intel processors, including AVX2,