Re: [PATCH v2] gdb/dwarf: fix reading subprogram with DW_AT_specification (PR gdb/26693)

[Simon Marchi <simon.marchi@polymtl.ca>]

On 2020-10-22 10:18 a.m., Tom Tromey wrote:
>>> gdb currently goes through a lot of work to support DWARF like this, but
>>> I wish it didn't have to.  It complicates the DWARF reader -- but IIRC
>>> both gcc and clang just emit specializations with names like
>>> "apply<int>".  So, I wonder if we could drop support for this at some
>>> point.
> 
> Simon> The object that contained this kind of DWARF, that made GDB crash, was
> Simon> compiled with ICC.  So unless we decide to drop support for ICC, I don't
> Simon> think we can get rid of it any time soon.
> 
> Well, we could support it in a more limited way - like, these functions
> would still appear but without template parameters in their names.
> 
> We could try to get a clarification from DWARF as to whether icc or
> gcc/clang are correct here, then file bug(s) against the compiler.

Ok, I see.  I don't have enough knowledge about this to have an informed opinion at the moment.

> Simon> I would write this, does that sound good?
> 
> Simon> /* Load the DIEs associated with PER_CU into memory.
> 
> Simon>    In some cases, the caller, while reading partial symbols, will need to load
> Simon>    the full symbols for the CU for some reason.  It will already have a
> Simon>    dwarf2_cu object for THIS_CU and pass it as EXISTING_CU, so it can be re-used
> Simon>    rather than creating a new one.  */
> 
> Looks great, thank you.

Thanks, pushed to master and gdb-10-branch.

Simon