This cl does two things

1) It shaves of a few cycles from the data dependency chain. By using "shrd" instead of a load.
2) The important loop is finding small copies (4-12) which are either "copy 1", or "copy 2" depending if the offset fits <2048. It turns out that this is a branch that is mispredicted often. Due to the long dependency chain the CPU is running with IPC~1 anyway so we can freely add instructions to instead emit copies branchfree. This reduces the branch misspredicts from 15% to 11% (for BM_ZFlat/6 txt1) and from 5.6% to 4% (for BM_ZFlat/10 or pb).

PiperOrigin-RevId: 303328967

snappy_unittest.cc

@@ -957,8 +957,9 @@ TEST(Snappy, ZeroOffsetCopyValidation) {
 namespace {
 
 int TestFindMatchLength(const char* s1, const char *s2, unsigned length) {
+  uint64 data;
   std::pair<size_t, bool> p =
-      snappy::internal::FindMatchLength(s1, s2, s2 + length);
+      snappy::internal::FindMatchLength(s1, s2, s2 + length, &data);
   CHECK_EQ(p.first < 8, p.second);
   return p.first;
 }