[jira] [Comment Edited] (IMPALA-4065) Inline comparator calls into TopN::InsertBatch()

["Tianyi Wang (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/IMPALA-4065?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16498761#comment-16498761 ] 

Tianyi Wang edited comment on IMPALA-4065 at 6/2/18 12:45 AM:
--------------------------------------------------------------

[~tarmstrong] I still don't get what replacing has to do with inlining. priority_queue is a bunch of templates. If nothing is inlined by clang++, aren't the call sites still there in the IR module?


was (Author: tianyiwang):
[~tarmstrong] I still don't get what replacing has to do with inlining. If nothing is inlined, aren't the call sites still there in the IR module?

> Inline comparator calls into TopN::InsertBatch()
> ------------------------------------------------
>
>                 Key: IMPALA-4065
>                 URL: https://issues.apache.org/jira/browse/IMPALA-4065
>             Project: IMPALA
>          Issue Type: Bug
>          Components: Backend
>    Affects Versions: Impala 2.7.0
>            Reporter: Tim Armstrong
>            Assignee: Zoram Thanga
>            Priority: Minor
>              Labels: codegen, ramp-up
>
> This is the more interesting follow-on from IMPALA-3815. We should inline the Compare() calls in the codegen'd TopN code to avoid the indirect function pointer call.
> The tricky aspect is that the Compare() calls are called from std::priority_queue, and we don't have a way to force-inline those functions at the moment.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscribe@impala.apache.org
For additional commands, e-mail: issues-all-help@impala.apache.org