Tim Armstrong has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/8303 )

Change subject: IMPALA-1575: Part 1: eagerly release query exec resources
......................................................................


Patch Set 11:

(3 comments)

http://gerrit.cloudera.org:8080/#/c/8303/11/be/src/runtime/mem-tracker.cc
File be/src/runtime/mem-tracker.cc:

http://gerrit.cloudera.org:8080/#/c/8303/11/be/src/runtime/mem-tracker.cc@359
PS11, Line 359: was not available
> Okay, but my original question was whether the comment should be reworded f
Done


http://gerrit.cloudera.org:8080/#/c/8303/11/be/src/runtime/query-state.cc
File be/src/runtime/query-state.cc:

http://gerrit.cloudera.org:8080/#/c/8303/11/be/src/runtime/query-state.cc@95
PS11, Line 95:   if (!released_exec_resources_) ReleaseExecResources();
> What I proposed was not to handle it in Init(), but let the caller handle i
Should have read it more carefully, sorry. I try to avoid function APIs that 
require callers to undo side-effects of a failed function - that seems 
error-prone in general but I guess it is ok in this specific case.


http://gerrit.cloudera.org:8080/#/c/8303/11/be/src/runtime/runtime-state.cc
File be/src/runtime/runtime-state.cc:

http://gerrit.cloudera.org:8080/#/c/8303/11/be/src/runtime/runtime-state.cc@88
PS11, Line 88:   local_query_state_->AcquireExecResourceRefcount(); // 
Decremented in ReleaseResources().
> Why are exvaluating exprs query-wide resources. Aren't they per-thread reso
I'd consider per-thread execution resources to be a subset of query-wide 
execution resources. I.e. releasing a query's execution resources while threads 
are still consuming execution resources is not valid in my mind. The comments 
in QueryState weren't very clear about what holding a refcount meant so I tried 
to improve them.



--
To view, visit http://gerrit.cloudera.org:8080/8303
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I41ff374b0403f10a145f7fee9b3145953ee32341
Gerrit-Change-Number: 8303
Gerrit-PatchSet: 11
Gerrit-Owner: Tim Armstrong <[email protected]>
Gerrit-Reviewer: Dan Hecht <[email protected]>
Gerrit-Reviewer: Joe McDonnell <[email protected]>
Gerrit-Reviewer: Sailesh Mukil <[email protected]>
Gerrit-Reviewer: Tim Armstrong <[email protected]>
Gerrit-Comment-Date: Fri, 27 Oct 2017 20:57:53 +0000
Gerrit-HasComments: Yes

Reply via email to