[PATCH][gdb/symtab] Fix gdb.base/vla-optimized-out.exp with clang

[Tom de Vries]

On 11/25/20 4:25 PM, Gary Benson wrote:
> Gary Benson wrote:
>> Thank you for picking this up Tom (de Vries).  Your patch looks good,
>> however I wanted to point out that the location expressions Clang
>> generates for gdb.base/vla-ptr.exp don't end with DW_OP_stack_value:
>>
>>   < 2><0x000000a3>  DW_TAG_variable
>>                       DW_AT_location    len 0x0002: 9168: DW_OP_fbreg -24
>>                       DW_AT_name      __vla_expr0
>>                       DW_AT_type      <0x00000118>
>>                       DW_AT_artificial        yes(1)
>>   < 2><0x000000af>  DW_TAG_variable
>>                       DW_AT_location  len 0x0002: 9160: DW_OP_fbreg -32
>>                       DW_AT_name      __vla_expr1
>>                       DW_AT_type      <0x00000118>
>>                       DW_AT_artificial        yes(1)
>>
>> It wasn't obvious to me how GDB with your patch would handle these.
>> Did you check your patch using that test?
> 

Yes.

> I forgot to add, that I did, and the final test failed with my setup:
> 
>   gdb.base/vla-ptr.exp
>   ...
>   print td_vla
>   GCC compiled:       $6 = {4, 5, 6, 7, 8}
>   Clang-12 compiled:  $6 = 0x7fffffffc0d0
> 
> But!  This is potentially a different issue.  I don't think that test
> accesses the array's size, and I've definitely seen typedef-related
> failures with Clang compared with GCC.
> 
> (I don't need you to debug this for me, I'm just making sure you're
> aware!)

Yes, I've seen this, I've filed a clang PR (
https://bugs.llvm.org/show_bug.cgi?id=48247 ), then committed an xfail
in gdb (
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=da39d3ba57639b715dea7ce68b6a4a3d70dfcb03
).

Thanks,
- Tom