[review] testsuite, cp: increase the coverage of testing pass-by-ref arguments

Tankut Baris Aktemur (Code Review) gerrit@gnutoolchain-gerrit.osci.io
Thu Oct 31 13:19:00 GMT 2019


Tankut Baris Aktemur has posted comments on this change.

Change URL: https://gnutoolchain-gerrit.osci.io/r/c/binutils-gdb/+/142
......................................................................


Patch Set 1:

> Patch Set 1:
> 
> Thank you for the patch.
> 
> This looks pretty good to me.
> 
> I wanted to verify that the tests failed without your series.
> 

Yes, tests fail without the patches in the series.  To give more details on this:

Without the series, using g++ 8.3.0 or 7.4.0, the pass/fail result is 378/546.

With the series;
* using g++ 8.3.0 or 7.4.0, it is 924/0.
* using g++ 5.5.0 or 6.5.0, it is 699/225.
* using clang++ 7.0.0 or 8.0.0, it is 727/197.

For the latter two, the main reason for failures is that the compiler does not emit the DW_AT_defaulted and DW_AT_deleted attributes.  There are more passes with clang because it emits DW_AT_calling_convention, which gcc does not, and the patch I sent favors compiler's hint if it detects a conflict between the DW_AT_calling_convention and the inferred result.

> Also, I think the generated file could just be generated each test run.
> Then this wouldn't need a special mode to regenerate the file.
> I think there are other tests that do this already.

Done.  I'll send this with the next revision; I want to revise the other patches in the series, too, and send the v2 in one go.

Thanks.


-- 
Gerrit-Project: binutils-gdb
Gerrit-Branch: master
Gerrit-Change-Id: Ie8ab1f260c6ad5ee4eb34b2c1597ce24af04abb6
Gerrit-Change-Number: 142
Gerrit-PatchSet: 1
Gerrit-Owner: Tankut Baris Aktemur <tankut.baris.aktemur@intel.com>
Gerrit-Reviewer: Tankut Baris Aktemur <tankut.baris.aktemur@intel.com>
Gerrit-CC: Tom Tromey <tromey@sourceware.org>
Gerrit-Comment-Date: Thu, 31 Oct 2019 13:19:28 +0000
Gerrit-HasComments: No
Gerrit-Has-Labels: No
Gerrit-MessageType: comment



More information about the Gdb-patches mailing list