Support SSE 4.2 qfind under MSVC

Summary:MSVC has support in the compiler for the intrinsics required, but both refuses to tell us that, and also gives them proper names.
The code already checks for runtime support, this just enables compiling the SSE 4.2 version in the first place.

Reviewed By: yfeldblum

Differential Revision: D3104296

fb-gh-sync-id: 9143240bede9b756817691fdd86818001267dac1
fbshipit-source-id: 9143240bede9b756817691fdd86818001267dac1

diff --git a/folly/Portability.h b/folly/Portability.h
index fb4a69968497ef9e8975e78e6fcaee0cb053ceb4..5bca1f54c8bdbd7f260cd847b62ca761c4f1dace 100644 (file)
--- a/folly/Portability.h
+++ b/folly/Portability.h
@@ -285,10 +285,9 @@ namespace std { typedef ::max_align_t max_align_t; }
 // Hide a GCC specific thing that breaks MSVC if left alone.
 # define __extension__
 
-#ifdef _M_IX86_FP
-# define FOLLY_SSE _M_IX86_FP
-# define FOLLY_SSE_MINOR 0
-#endif
+// We have compiler support for the newest of the new, but
+// MSVC doesn't tell us that.
+#define __SSE4_2__ 1
 
 #endif
 

diff --git a/folly/detail/RangeSse42.cpp b/folly/detail/RangeSse42.cpp
index a7aeb8d4963c741a0727836863ac23843becf461..b963e5e774b707c112bb9ec4430b0d84fa3fe744 100644 (file)
--- a/folly/detail/RangeSse42.cpp
+++ b/folly/detail/RangeSse42.cpp
@@ -14,49 +14,35 @@
  * limitations under the License.
  */
 
-
-
 #include "RangeSse42.h"
 
 #include <glog/logging.h>
 #include <folly/Portability.h>
 
-
-
 //  Essentially, two versions of this file: one with an SSE42 implementation
 //  and one with a fallback implementation. We determine which version to use by
 //  testing for the presence of the required headers.
 //
 //  TODO: Maybe this should be done by the build system....
 #if !FOLLY_SSE_PREREQ(4, 2)
-
-
-
 namespace folly {
-
 namespace detail {
-
 size_t qfind_first_byte_of_sse42(const StringPieceLite haystack,
                                  const StringPieceLite needles) {
   CHECK(false) << "Function " << __func__ << " only works with SSE42!";
   return qfind_first_byte_of_nosse(haystack, needles);
 }
-
 }
-
 }
-
-
-
 # else
-
-
-
 #include <cstdint>
 #include <limits>
 #include <string>
+
 #include <emmintrin.h>
+#include <nmmintrin.h>
 #include <smmintrin.h>
+
 #include <folly/Likely.h>
 
 //  GCC 4.9 with ASAN has a problem: a function with no_sanitize_address calling
@@ -68,10 +54,14 @@ size_t qfind_first_byte_of_sse42(const StringPieceLite haystack,
     __GNUC_PREREQ(4, 9)
 # define _mm_load_si128(p) (*(p))
 # define _mm_loadu_si128(p) ((__m128i)__builtin_ia32_loaddqu((const char*)(p)))
+# ifdef _mm_cmpestri
+#  undef _mm_cmpestri
+# endif
+# define _mm_cmpestri(a, b, c, d, e) \
+  __builtin_ia32_pcmpestri128((__v16qi)(a), b, (__v16qi)(c), d, e)
 #endif
 
 namespace folly {
-
 namespace detail {
 
 // It's okay if pages are bigger than this (as powers of two), but they should
@@ -116,8 +106,8 @@ size_t qfind_first_byte_of_needles16(const StringPieceLite haystack,
   // do an unaligned load for first block of haystack
   auto arr1 = _mm_loadu_si128(
       reinterpret_cast<const __m128i*>(haystack.data()));
-  auto index = __builtin_ia32_pcmpestri128((__v16qi)arr2, needles.size(),
-                                           (__v16qi)arr1, haystack.size(), 0);
+  auto index = _mm_cmpestri(arr2, needles.size(),
+                            arr1, haystack.size(), 0);
   if (index < 16) {
     return index;
   }
@@ -127,9 +117,9 @@ size_t qfind_first_byte_of_needles16(const StringPieceLite haystack,
   for (; i < haystack.size(); i+= 16) {
     auto arr1 = _mm_load_si128(
         reinterpret_cast<const __m128i*>(haystack.data() + i));
-    auto index = __builtin_ia32_pcmpestri128(
-        (__v16qi)arr2, needles.size(),
-        (__v16qi)arr1, haystack.size() - i, 0);
+    auto index = _mm_cmpestri(
+        arr2, needles.size(),
+        arr1, haystack.size() - i, 0);
     if (index < 16) {
       return i + index;
     }
@@ -172,17 +162,17 @@ size_t scanHaystackBlock(const StringPieceLite haystack,
   // This load is safe because needles.size() >= 16
   auto arr2 = _mm_loadu_si128(
       reinterpret_cast<const __m128i*>(needles.data()));
-  size_t b = __builtin_ia32_pcmpestri128(
-    (__v16qi)arr2, 16, (__v16qi)arr1, haystack.size() - blockStartIdx, 0);
+  size_t b = _mm_cmpestri(
+      arr2, 16, arr1, haystack.size() - blockStartIdx, 0);
 
   size_t j = nextAlignedIndex(needles.data());
   for (; j < needles.size(); j += 16) {
     arr2 = _mm_load_si128(
         reinterpret_cast<const __m128i*>(needles.data() + j));
 
-    auto index = __builtin_ia32_pcmpestri128(
-      (__v16qi)arr2, needles.size() - j,
-      (__v16qi)arr1, haystack.size() - blockStartIdx, 0);
+    auto index = _mm_cmpestri(
+      arr2, needles.size() - j,
+      arr1, haystack.size() - blockStartIdx, 0);
     b = std::min<size_t>(index, b);
   }
 
@@ -229,11 +219,6 @@ size_t qfind_first_byte_of_sse42(const StringPieceLite haystack,
 
   return std::string::npos;
 }
-
 }
-
 }
-
-
-
 #endif