Yida Wu has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/20168 )

Change subject: IMPALA-12269: Codegen cache false negative because of function 
names hash
......................................................................


Patch Set 3: Code-Review+1

(1 comment)

Thanks Daniel for fixing this.
A question maybe not related to this fix, does the same function name always 
mean the same function in different modules? i.e, is GetSlotRef_xxx in module a 
always same as GetSlotRef_xxx in module b?

http://gerrit.cloudera.org:8080/#/c/20168/3/be/src/codegen/llvm-codegen.cc
File be/src/codegen/llvm-codegen.cc:

http://gerrit.cloudera.org:8080/#/c/20168/3/be/src/codegen/llvm-codegen.cc@1256
PS3, Line 1256: exported_fn_names
Here doing a similar thing to put all the names to a set. Seems the repetitive 
computation can be avoided. Do you think it good to let them use the same set 
(this one uses unordered_set, but we require an order)? Or even maybe change 
the fns_to_jit_compile_ to a map with the function name being the key?



--
To view, visit http://gerrit.cloudera.org:8080/20168
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ibf1d2b424c969fbba181ab90bf9c7bf22355f139
Gerrit-Change-Number: 20168
Gerrit-PatchSet: 3
Gerrit-Owner: Daniel Becker <[email protected]>
Gerrit-Reviewer: Csaba Ringhofer <[email protected]>
Gerrit-Reviewer: Daniel Becker <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Yida Wu <[email protected]>
Gerrit-Comment-Date: Tue, 11 Jul 2023 21:10:27 +0000
Gerrit-HasComments: Yes

Reply via email to