Created attachment 32145 [details] preprocessed file foo.ii With the following code snippet <snip> void foo(const unsigned N) { int p[N][N]; auto go = [&p] (void) { p[0][0] = -1; }; go(); } </snip> I get an internal compiler error. #> /usr/local/bin/g++ --std=c++11 foo.cpp foo.cpp: In lambda function: foo.cpp:5:35: internal compiler error: in expand_expr_real_1, at expr.c:9459 auto go = [&p] (void) { p[0][0] = -1; }; ^ #> /usr/local/bin/g++ -v Using built-in specs. COLLECT_GCC=/usr/local/bin/g++ COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/x86_64-unknown-linux-gnu/4.9.0/lto-wrapper Target: x86_64-unknown-linux-gnu Configured with: ./configure --disable-multilib --enable-languages=c,c++ Thread model: posix gcc version 4.9.0 20140216 (experimental) (GCC)
Started with r180707.
We should find the actual commit, likely *before* r180707.
*** Bug 71628 has been marked as a duplicate of this bug. ***
*** Bug 64000 has been marked as a duplicate of this bug. ***
*** Bug 69756 has been marked as a duplicate of this bug. ***
Jason, in 8, at variance with 7, for this testcase and its Dups we issue a proper error message mentioning N3639 and no ICE. I think we can resolve the bug, right?
*** Bug 67608 has been marked as a duplicate of this bug. ***
(In reply to Paolo Carlini from comment #6) > Jason, in 8, at variance with 7, for this testcase and its Dups we issue a > proper error message mentioning N3639 and no ICE. I think we can resolve the > bug, right? Yes, my fix was marked with PR 84560. Actually allowing this code would be an extension beyond what was allowed under the abandoned array TS. *** This bug has been marked as a duplicate of bug 84560 ***