Clang recently improved its DWARF support for C VLA types. The DWARF now looks like this:
0x00000051: DW_TAG_variable [4] DW_AT_location( fbreg -32 ) DW_AT_name( "__vla_expr" ) DW_AT_type( {0x000000d3} ( long unsigned int ) ) DW_AT_artificial( true ) ... 0x000000da: DW_TAG_array_type [10] * DW_AT_type( {0x000000cc} ( int ) ) 0x000000df: DW_TAG_subrange_type [11] DW_AT_type( {0x000000e9} ( __ARRAY_SIZE_TYPE__ ) ) DW_AT_count( {0x00000051} )
Without this patch LLDB will naively interpret the DIE offset 0x51 as the static size of the array, which is clearly wrong.
This patch uses LLDB's dynamic type mechanism to re-parse VLA types with an optional execution context, to dynamically resolve the size of the array correctly. These dynamic types are not being cached, since they are only valid in a single execution context.
See the testcase for an example:
4 int foo(int a) { 5 int vla[a]; 6 for (int i = 0; i < a; ++i) 7 vla[i] = i; 8 -> 9 pause(); // break here 10 return vla[a-1]; 11 } 12 (lldb) fr v vla (int [4]) vla = ([0] = 0, [1] = 1, [2] = 2, [3] = 3) (lldb) quit
Here it could accidentally follow a number as DIE reference, addressed by: D56068