[PATCH] D60644: [compiler-rt][builtins][sanitizers] Guard test cases with macros with GLIBC* macros

[Hubert Tong via Phabricator via llvm-commits]

hubert.reinterpretcast added inline comments.


================
Comment at: compiler-rt/test/sanitizer_common/TestCases/Posix/getpw_getgr.cc:76
+  // Run this if we're using the C++11 ABI. There seems to be an issue
+  // that maybe related to _GLIBCXX_USE_CXX11_ABI and the string assigned.
+#if _GLIBCXX_USE_CXX11_ABI
----------------
amyk wrote:
> hubert.reinterpretcast wrote:
> > Since we now have a better understanding of the issue, the solution seems to be that the test should not be using `std::string`.
> > 
> > ```
> > diff -U0 a/getpw_getgr.cc b/getpw_getgr.cc
> > --- a/getpw_getgr.cc    2019-03-08 12:56:26.499543426 -0500
> > +++ b/getpw_getgr.cc    2019-04-25 10:06:56.240191290 -0400
> > @@ -11 +10,0 @@
> > -#include <string>
> > @@ -13 +12 @@
> > -std::string any_group;
> > +std::unique_ptr<char []> any_group;
> > @@ -51,2 +50,5 @@
> > -  if (any_group.empty())
> > -    any_group = result->gr_name;
> > +  if (!any_group) {
> > +    auto len = strlen(result->gr_name);
> > +    any_group.reset(new char[len + 1]);
> > +    memcpy(any_group.get(), result->gr_name, len + 1);
> > +  }
> > @@ -75 +77 @@
> > -  test<group>(&getgrnam, any_group.c_str());
> > +  test<group>(&getgrnam, any_group.get());
> > @@ -94 +96 @@
> > -  test_r<group>(&getgrnam_r, any_group.c_str());
> > +  test_r<group>(&getgrnam_r, any_group.get());
> > ```
> Yes. I saw the reply on the mailing list; thank you for that. I was going to look into something like this. I will try these changes and report back. 
> 
> This could be a silly question, but are there any other alternatives to `unique_ptr`, or is there a specific reason we should use this? 
`unique_ptr` is resource leak resistant. That is why using it is a good idea.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D60644/new/

https://reviews.llvm.org/D60644