Tim Armstrong has posted comments on this change. Change subject: IMPALA-2809: Improve ByteSwap with builtin function or SSE or AVX2. ......................................................................
Patch Set 5: (3 comments) http://gerrit.cloudera.org:8080/#/c/3081/5/be/src/benchmarks/bswap-benchmark.cc File be/src/benchmarks/bswap-benchmark.cc: Line 36: // Impala: the old Impala implementation. > Old Impala implementation refers to some code has been discarded in previou It's ok to include the old ImpalaByteSwap() here, but we should benchmark it against the new BitUtil::ByteSwap() http://gerrit.cloudera.org:8080/#/c/3081/5/be/src/util/bit-util.inline.h File be/src/util/bit-util.inline.h: Line 179: SIMDFuncTable[func_idx](dst, src); I don't think a jump table would be faster than a single direct branch (the function pointer call is an indirect branch, which is usually slower than a direct branch). If there are many branches (e.g. > 5 or 6) then it might be faster. Unless you have a benchmark showing a difference. Line 185: if(i >= 16) { The formatting of if () { and while () { is missing spaces. in many places. -- To view, visit http://gerrit.cloudera.org:8080/3081 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I392ed5a8d5683f30f161282c228c1aedd7b648c1 Gerrit-PatchSet: 5 Gerrit-Project: Impala Gerrit-Branch: cdh5-trunk Gerrit-Owner: Youwei Wang <[email protected]> Gerrit-Reviewer: Alex Behm <[email protected]> Gerrit-Reviewer: Jim Apple <[email protected]> Gerrit-Reviewer: Marcel Kornacker <[email protected]> Gerrit-Reviewer: Mostafa Mokhtar <[email protected]> Gerrit-Reviewer: Tim Armstrong <[email protected]> Gerrit-Reviewer: Youwei Wang <[email protected]> Gerrit-HasComments: Yes
