Merge #13386: SHA256 implementations based on Intel SHA Extensions

66b2cf1ccfad545a8ec3f2a854e23f647322bf30 Use immintrin.h everywhere for intrinsics (Pieter Wuille)
4c935e2eee456ff66cdfb908b0edffdd1e8a6c04 Add SHA256 implementation using using Intel SHA intrinsics (Pieter Wuille)
268400d3188200c9e3dcd3482c4853354388a721 [Refactor] CPU feature detection logic for SHA256 (Pieter Wuille)

Pull request description:

  Based on #13191.

  This adds SHA256 implementations that use Intel's SHA Extension instructions (using intrinsics). This needs GCC 4.9 or Clang 3.4.

  In addition to #13191, two extra implementations are provided:
  * (a) A variable-length SHA256 implementation using SHA extensions.
  * (b) A 2-way 64-byte input double-SHA256 implementation using SHA extensions.

  Benchmarks for 9001-element Merkle tree root computation on an AMD Ryzen 1800X system:
  * Using generic C++ code (pre-#10821): 6.1ms
  * Using SSE4 (master, #10821): 4.6ms
  * Using 4-way SSE4 specialized for 64-byte inputs (#13191): 2.8ms
  * Using 8-way AVX2 specialized for 64-byte inputs (#13191): 2.1ms
  * Using 2-way SHA-NI specialized for 64-byte inputs (this PR): 0.56ms

  Benchmarks for 32-byte SHA256 on the same system:
  * Using SSE4 (master, #10821): 190ns
  * Using SHA-NI (this PR): 53ns

  Benchmarks for 1000000-byte SHA256 on the same system:
  * Using SSE4 (master, #10821): 2.5ms
  * Using SHA-NI (this PR): 0.51ms

Tree-SHA512: 2b319e33b22579f815d91f9daf7994a5e1e799c4f73c13e15070dd54ba71f3f6438ccf77ae9cbd1ce76f972d9cbeb5f0edfea3d86f101bbc1055db70e42743b7

1 files changed, 20 insertions, 8 deletions

diff --git a/configure.ac b/configure.ac
index 9fba9d0851..490b638ee1 100644
--- a/configure.ac
+++ b/configure.ac
@@ -320,6 +320,7 @@ fi
 AX_CHECK_COMPILE_FLAG([-msse4.2],[[SSE42_CXXFLAGS="-msse4.2"]],,[[$CXXFLAG_WERROR]])
 AX_CHECK_COMPILE_FLAG([-msse4.1],[[SSE41_CXXFLAGS="-msse4.1"]],,[[$CXXFLAG_WERROR]])
 AX_CHECK_COMPILE_FLAG([-mavx -mavx2],[[AVX2_CXXFLAGS="-mavx -mavx2"]],,[[$CXXFLAG_WERROR]])
+AX_CHECK_COMPILE_FLAG([-msse4 -msha],[[SHANI_CXXFLAGS="-msse4 -msha"]],,[[$CXXFLAG_WERROR]])
 
 TEMP_CXXFLAGS="$CXXFLAGS"
 CXXFLAGS="$CXXFLAGS $SSE42_CXXFLAGS"
@@ -348,11 +349,7 @@ CXXFLAGS="$CXXFLAGS $SSE41_CXXFLAGS"
 AC_MSG_CHECKING(for SSE4.1 intrinsics)
 AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
     #include <stdint.h>
-    #if defined(_MSC_VER)
     #include <immintrin.h>
-    #elif defined(__GNUC__)
-    #include <x86intrin.h>
-    #endif
   ]],[[
     __m128i l = _mm_set1_epi32(0);
     return _mm_extract_epi32(l, 3);
@@ -367,11 +364,7 @@ CXXFLAGS="$CXXFLAGS $AVX2_CXXFLAGS"
 AC_MSG_CHECKING(for AVX2 intrinsics)
 AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
     #include <stdint.h>
-    #if defined(_MSC_VER)
     #include <immintrin.h>
-    #elif defined(__GNUC__) && defined(__AVX2__)
-    #include <x86intrin.h>
-    #endif
   ]],[[
     __m256i l = _mm256_set1_epi32(0);
     return _mm256_extract_epi32(l, 7);
@@ -381,6 +374,23 @@ AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
 )
 CXXFLAGS="$TEMP_CXXFLAGS"
 
+TEMP_CXXFLAGS="$CXXFLAGS"
+CXXFLAGS="$CXXFLAGS $SHANI_CXXFLAGS"
+AC_MSG_CHECKING(for SHA-NI intrinsics)
+AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[
+    #include <stdint.h>
+    #include <immintrin.h>
+  ]],[[
+    __m128i i = _mm_set1_epi32(0);
+    __m128i j = _mm_set1_epi32(1);
+    __m128i k = _mm_set1_epi32(2);
+    return _mm_extract_epi32(_mm_sha256rnds2_epu32(i, i, k), 0);
+  ]])],
+ [ AC_MSG_RESULT(yes); enable_shani=yes; AC_DEFINE(ENABLE_SHANI, 1, [Define this symbol to build code that uses SHA-NI intrinsics]) ],
+ [ AC_MSG_RESULT(no)]
+)
+CXXFLAGS="$TEMP_CXXFLAGS"
+
 CPPFLAGS="$CPPFLAGS -DHAVE_BUILD_INFO -D__STDC_FORMAT_MACROS"
 
 AC_ARG_WITH([utils],
@@ -1309,6 +1319,7 @@ AM_CONDITIONAL([HARDEN],[test x$use_hardening = xyes])
 AM_CONDITIONAL([ENABLE_HWCRC32],[test x$enable_hwcrc32 = xyes])
 AM_CONDITIONAL([ENABLE_SSE41],[test x$enable_sse41 = xyes])
 AM_CONDITIONAL([ENABLE_AVX2],[test x$enable_avx2 = xyes])
+AM_CONDITIONAL([ENABLE_SHANI],[test x$enable_shani = xyes])
 AM_CONDITIONAL([USE_ASM],[test x$use_asm = xyes])
 
 AC_DEFINE(CLIENT_VERSION_MAJOR, _CLIENT_VERSION_MAJOR, [Major version])
@@ -1353,6 +1364,7 @@ AC_SUBST(SANITIZER_LDFLAGS)
 AC_SUBST(SSE42_CXXFLAGS)
 AC_SUBST(SSE41_CXXFLAGS)
 AC_SUBST(AVX2_CXXFLAGS)
+AC_SUBST(SHANI_CXXFLAGS)
 AC_SUBST(LIBTOOL_APP_LDFLAGS)
 AC_SUBST(USE_UPNP)
 AC_SUBST(USE_QRCODE)