impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tim Armstrong (Code Review)" <>
Subject [Impala-ASF-CR] IMPALA-4026: Implement double-buffering for BlockingQueue
Date Tue, 20 Sep 2016 20:14:24 GMT
Tim Armstrong has posted comments on this change.

Change subject: IMPALA-4026: Implement double-buffering for BlockingQueue

Patch Set 5: Code-Review+1

File be/src/util/blocking-queue.h:

PS5, Line 75: NotitfyAll

PS5, Line 98:  This may
            :     // imply that some writers may be sleeping on a partially empty queue
Maybe "If this race occurs, a writer can stay blocked on a partially empty queue until the
next BlockingGet() call." And I think it would be clearer if the consequence of the race was
explained before why it was benign.

PS5, Line 99: Given the major
            :     // use case is with HDFS scan node which has multiple producers and one
consumer, it's
            :     // expected that some producers can make progress.
Maybe more simply: "This should only occur when producers are faster than consumers, so is
unlikely to affect overall throughput".

To view, visit
To unsubscribe, visit

Gerrit-MessageType: comment
Gerrit-Change-Id: Ib9f4cf351455efefb0f3bb791cf9bc82d1421d54
Gerrit-PatchSet: 5
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Michael Ho <>
Gerrit-Reviewer: Alex Behm <>
Gerrit-Reviewer: Chen Huang <>
Gerrit-Reviewer: Dan Hecht <>
Gerrit-Reviewer: Michael Ho <>
Gerrit-Reviewer: Mostafa Mokhtar <>
Gerrit-Reviewer: Tim Armstrong <>
Gerrit-HasComments: Yes

View raw message