Michael Ho has posted comments on this change. Change subject: IMPALA-4231: fix codegen time regression ......................................................................
Patch Set 1: (4 comments) http://gerrit.cloudera.org:8080/#/c/4623/1//COMMIT_MSG Commit Message: PS1, Line 26: y, which is fixed by this : patch. Does it recover all the performance ? http://gerrit.cloudera.org:8080/#/c/4623/1/be/src/exec/partitioned-hash-join-builder-ir.cc File be/src/exec/partitioned-hash-join-builder-ir.cc: PS1, Line 32: inline bool PhjBuilder::AppendRow( I thought the destructor is inlined so it only calls FreeMessage() which is not supposed to cross-compiled. So, I suppose the regression is not the codegen time but the runtime of this function, right ? http://gerrit.cloudera.org:8080/#/c/4623/1/be/src/runtime/raw-value.inline.h File be/src/runtime/raw-value.inline.h: PS1, Line 215: this may need to be inlined again eventually for loop unrolling to work. http://gerrit.cloudera.org:8080/#/c/4623/1/be/src/util/bloom-filter.h File be/src/util/bloom-filter.h: PS1, Line 176: inline void IR_ALWAYS_INLINE void ALWAYS_INLINE -- To view, visit http://gerrit.cloudera.org:8080/4623 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: Idf0fdedabd488550b6db90167a30c582949d608d Gerrit-PatchSet: 1 Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-Owner: Tim Armstrong <[email protected]> Gerrit-Reviewer: Michael Ho Gerrit-Reviewer: Michael Ho <[email protected]> Gerrit-HasComments: Yes
