sigsegv on running basic scenario

[curb1968]

I'm attempting to use catch to perform unit testing for an embedded target running on a microblaze processor. I'm testing using QEMU and using gcc version 4.6.4 (20120924). I've provided my unit test code as well as the backtrace of the call tree immediately prior to the sigsegv.

/*
- Unit test application written for testing within QEMU
  */
  #define CATCH_CONFIG_RUNNER // This tells Catch that we will provide a main() - ONLY PROVIDE THIS ONCE!!!
  #include "catch.hpp"
# include <iostream>
# include <string>
# include <memory>   // unique_ptr

int main( int argc, char\* const argv[] )
{
    int result = Catch::Session().run( argc, argv );
    return result;
}
SCENARIO( "test the UT framework under qemu", "[CATCH]" )
{
    GIVEN( "UT has started" )
    {
        REQUIRE( (1 == 1) );
    }
}

Here's the stack trace under gdb:

> # 0  Catch::FreeFunctionTestCase::invoke (this=0x1019abc8)
> 
> 
> at ../deps/include/catch.hpp:5810
> 
> # 1  0x100090c0 in Catch::TestCase::invoke (this=0x1019c208)
> 
> 
> at ../deps/include/catch.hpp:6715
> 
> # 2  0x100251fc in Catch::RunContext::invokeActiveTestCase (this=0xbffffa28)
> 
> 
> at ../deps/include/catch.hpp:5412
> 
> # 3  0x10024dfc in Catch::RunContext::runCurrentTest (this=0xbffffa28, redirectedCout=..., redirectedCerr=...)
> 
> 
> at ../deps/include/catch.hpp:5384
> 
> # 4  0x100230d4 in Catch::RunContext::runTest (this=0xbffffa28, testCase=...)
> 
> 
> at ../deps/include/catch.hpp:5223
> 
> # 5  0x10025ad8 in Catch::Runner::runTests (this=0xbffffb38)
> 
> 
> at ../deps/include/catch.hpp:5543
> 
> # 6  0x100272cc in Catch::Session::run (this=0xbffffcbc)
> 
> 
> at ../deps/include/catch.hpp:5670
> 
> # 7  0x10027000 in Catch::Session::run (this=0xbffffcbc, argc=1, argv=0xbffffe04)
> 
> 
> at ../deps/include/catch.hpp:5650
> 
> # 8  0x10011738 in main (argc=1, argv=0xbffffe04) at ../UnitTest/unitTest.cpp:26
> 
> (gdb) n
> 
> Program received signal SIGSEGV, Segmentation fault.
> 0x00000008 in ?? ()
> (gdb) bt
> # 0  0x00000008 in ?? ()
> # 1  0x00000008 in ?? ()
> 
> Backtrace stopped: previous frame identical to this frame (corrupt stack?)
> (gdb) 
> # 0  0x00000008 in ?? ()
> # 1  0x00000008 in ?? ()

[horenmar]

I edited the issue to make the example code more legible.

Can anyone with QEMU confirm if this is happening with current Catch?

[philsquared]

This looks like it could be related to #661

[horenmar]

Since it has been ~5 months without new information, I am going to close this issue. If someone manages to reliably reproduce this with latest Catch, please open a new issue and link this one.