Re: [PATCH] Handle DW_FORM_implicit_const for DW_AT_decl_line

[Tom de Vries <tdevries@suse.de>]

On 2/14/21 10:27 PM, Tom de Vries wrote:
> On 2/14/21 8:19 PM, Jakub Jelinek wrote:
>> On Sun, Feb 14, 2021 at 08:10:49PM +0100, Mark Wielaard wrote:
>>> Yes, this makes sense. Like you said, we do something like this already
>>> for decl/call_file. I think technically the die_eq_1 part could be put
>>> under the same switch case. But maybe you find it clearer to do it
>>> separately to mirror the checksum_die part?
>>
>> Yeah, I'm wondering if we just shouldn't hash and compare all constant class
>> forms with the same value the same (ok, except DW_FORM_data16 which is too
>> large).  For DW_FORM_data{1,2,4,8} we don't know if it is signed or
>> unsigned, so perhaps just treat values with the msb clear that way?
>> Or sure, if we can put in a list of attributes that always have unsigned
>> or always have signed constants, we can even improve that.
> 
> Make sense to me.

Filed as PR27421 - Improve DW_FORM_implicit_const handling in
checksum_die/die_eq_1 (
https://sourceware.org/bugzilla/show_bug.cgi?id=27421 ).

Thanks,
- Tom