New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
clang: disable RTTI support with cxx_no_rtti_flags
case for parity with g++
#2148
clang: disable RTTI support with cxx_no_rtti_flags
case for parity with g++
#2148
Conversation
0ba1c6a
to
c6c9bd0
Compare
@ngie-eign could you please rebase/resolve conflicts |
c6c9bd0
to
afdb3e2
Compare
@gennadiycivil: done! |
How do we turn on C++11 today? The fact we failed to set |
@EricWF C++ 11 set. Some set in travis configs, some elsewhere, it is generally not well organized. |
@gennadiycivil It looks like we're using |
Yes. I'll remove that explicit setting. |
Ping? |
6dd1559
to
15a6eb3
Compare
@gennadiycivil: pong :). |
15a6eb3
to
3443810
Compare
@gennadiycivil: ping? |
28c9d9b
to
07f1af0
Compare
@EricWF, @gennadiycivil: ping? |
3443810
to
7475ba5
Compare
I still don't understand why we weed to pass |
I edited the comment a bit to just leave rtti , as C++11 is no longer in scope... Please advise? Thanks again |
@EricWF: the concern I have is that downstream projects might disable RTTI in their builds, resulting in unexpected behavior when compiling with clang, or the default may change due to an upstream change or local modification. If either of these things happen, googletest may be compiled improperly, resulting in confusing behavior and test results, forcing one or more engineers to have to analyze why googletest isn’t functioning as expected. |
@ngie-eign I am not sure I agree with this premise. I don't believe googletest can presume to know what the downstream client means to when they choose disable RTTI . If the theoretical "they" chose to do this there must have been a reason. |
…with g++ Add `cxx_no_rtti_flags` for clang in order to test the case with RTTI off, similar to g++. This increases test coverage with the non-RTTI case with clang. This something I used in when investigating test failures on FreeBSD, as the tests that rely are failing with googletest 1.8.1 on the OS/platform, as described in issue google#2172. Signed-off-by: Enji Cooper <yaneurabeya@gmail.com>
7475ba5
to
6d393d7
Compare
cxx_no_rtti_flags
case for parity with g++
Ok, I've walked by the changes to assume the implicit |
…-flag PiperOrigin-RevId: 242038269
Should be in now 5ba69d5 |
We should have a test that catches this. I'll post something soon to add it to Travis. |
Add
cxx_no_rtti_flags
for clang in order to test the case with RTTI off,similar to g++.
This increases test coverage with the non-RTTI case with clang. This something
I used in when investigating test failures on FreeBSD, as the tests that rely
are failing with googletest 1.8.1 on the OS/platform, as described in
issue #2172.
Signed-off-by: Enji Cooper yaneurabeya@gmail.com