[Bug target/66810] New: [6] ICE on tree error_mark with -miamcu

[julia.koval at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66810

            Bug ID: 66810
           Summary: [6] ICE on tree error_mark with -miamcu
           Product: gcc
           Version: 6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: julia.koval at intel dot com
  Target Milestone: ---

Created attachment 35934
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35934&action=edit
proposed patch

$ cat test.c
int vv;

void
i (void)
{
  static int a[vv];
}

$ gcc test.c -m32
test.c: In function ‘i’:
test.c:6:14: error: storage size of ‘a’ isn’t constant
   static int a[vv];
              ^
$ gcc test.c -m32 -miamcu
test.c: In function ‘i’:
test.c:6:14: error: storage size of ‘a’ isn’t constant
   static int a[vv];
              ^
At top level:
cc1: internal compiler error: tree check: expected class ‘type’, have
‘exceptional’ (error_mark) in iamcu_alignment, at config/i386/i386.c:27402
0xdba407 tree_class_check_failed(tree_node const*, tree_code_class, char
const*, int, char const*)
        .././../TRUNK/gcc/gcc/tree.c:9480
0xe39f48 tree_class_check(tree_node*, tree_code_class, char const*, int, char
const*)
        .././../TRUNK/gcc/gcc/tree.h:2981
0xe39f48 iamcu_alignment
        .././../TRUNK/gcc/gcc/config/i386/i386.c:27402
0xe6e4aa ix86_data_alignment(tree_node*, int, bool)
        .././../TRUNK/gcc/gcc/config/i386/i386.c:27456
0xe0cbc1 align_variable(tree_node*, bool)
        .././../TRUNK/gcc/gcc/varasm.c:1030
0xe1e7a2 varpool_node::analyze()
        .././../TRUNK/gcc/gcc/varpool.c:531
0x72ac47 analyze_functions
        .././../TRUNK/gcc/gcc/cgraphunit.c:1083
0x72b720 symbol_table::finalize_compilation_unit()
        .././../TRUNK/gcc/gcc/cgraphunit.c:2485
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See <http://gcc.gnu.org/bugs.html> for instructions.

There is a test, reproducing this:
gcc.dg/c11-thread-local-2.c