Tim Armstrong has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/8814 )

Change subject: IMPALA-6290: limit ScannerContext to 1 buffer at a time
......................................................................


Patch Set 6:

(3 comments)

http://gerrit.cloudera.org:8080/#/c/8814/6/be/src/exec/scanner-context.h
File be/src/exec/scanner-context.h:

http://gerrit.cloudera.org:8080/#/c/8814/6/be/src/exec/scanner-context.h@162
PS6, Line 162:     /// Equivalent to java.io.DataInput.readBoolean()
> May help to document what returning false means for the Read*() functions.
Done


http://gerrit.cloudera.org:8080/#/c/8814/6/be/src/exec/scanner-context.cc
File be/src/exec/scanner-context.cc:

http://gerrit.cloudera.org:8080/#/c/8814/6/be/src/exec/scanner-context.cc@94
PS6, Line 94:     boundary_buffer_pos_ = nullptr;
            :     boundary_buffer_bytes_left_ = 0;
            :     boundary_buffer_->Reset();
> Why weren't we doing it before ?
I don't think it's strictly necessary, since we shouldn't be doing anything 
with the stream once this is called with done=true and the memory gets freed 
with the MemPool, but it seemed worth doing defensively.


http://gerrit.cloudera.org:8080/#/c/8814/6/be/src/exec/scanner-context.cc@99
PS6, Line 99: ReturnIoBuffer()
> I wonder if it warrants poisoning the IOBuffer returned to catch any cases
I added poison of cached I/O buffers, since it's only a few lines of change. 
We'll get it for free in follow-on patches once it's switched to the buffer 
pool but nice to get the test coverage sooner.



--
To view, visit http://gerrit.cloudera.org:8080/8814
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I74c5960a75f7d88b0e1de4199af731fb13e592f0
Gerrit-Change-Number: 8814
Gerrit-PatchSet: 6
Gerrit-Owner: Tim Armstrong <[email protected]>
Gerrit-Reviewer: Lars Volker <[email protected]>
Gerrit-Reviewer: Michael Ho <[email protected]>
Gerrit-Reviewer: Tim Armstrong <[email protected]>
Gerrit-Comment-Date: Tue, 09 Jan 2018 21:51:59 +0000
Gerrit-HasComments: Yes

Reply via email to