Submitter | Simon Marchi (Code Review) |
---|---|
Date | Nov. 14, 2019, 2:56 a.m. |
Message ID | <gerrit.1573700165000.I323361b1c120bf8525613b74e7e5983910e002df@gnutoolchain-gerrit.osci.io> |
Download | mbox | patch |
Permalink | /patch/35869/ |
State | New |
Headers | show |
Comments
Andrew Burgess has posted comments on this change. Change URL: https://gnutoolchain-gerrit.osci.io/r/c/binutils-gdb/+/626 ...................................................................... Patch Set 1: Code-Review+2 Sorry for introducing this breakage. Oddly I tried on a GCC 10 built from git a few weeks ago and I don't see the build warning/error. Anyway, I tested that with this patch the original bug in unordered_remove is still detected, so this LGTM. Thanks for picking up after me :)
Patch
diff --git a/gdb/unittests/vec-utils-selftests.c b/gdb/unittests/vec-utils-selftests.c index 823bbb6..d0ead99 100644 --- a/gdb/unittests/vec-utils-selftests.c +++ b/gdb/unittests/vec-utils-selftests.c @@ -38,6 +38,15 @@ { std::vector<void *> var; + obj() = default; + + /* gcc complains if we provide an assignment operator but no copy + constructor, so provide even if don't really care for this test. */ + obj(const obj &other) + { + this->var = other.var; + } + obj &operator= (const obj &other) { if (this == &other)