On Sun, Oct 30, 2022 at 04:22:34PM +0100, Mark Wielaard wrote:
> Hi,
>
> On Wed, Oct 26, 2022 at 10:39:09AM +0200, Jakub Jelinek wrote:
> > I must say I don't understand nor like this DW_LANG_Rust_old stuff at all.
> > Other languages don't do similar dances.
> > Look for D, or Go. Neither of them has any non-standard lang code as
> > fallback, they use the DWARF assigned DW_LANG_* code, and DW_LANG_C as
> > fallback. On most arches, DWARF 5 is the default anyway, or non-strict
> > DWARF at least. Where neither is enabled because of prehistoric or buggy
> > DWARF consumers, it is unlikely they'd handle Rust sanely anyway.
> > Just follow what Go does in the same function.
>
> DW_LANG_Rust_old was used by old rustc compilers <= 2016 before DWARF5
> assigned an official number. It might be recognized by some
> debuggers. But I agree that these days it doesn't really make sense to
> emit it. When producing strict DWARF it is also slightly odd to emit a
> non-standard language code. So I agree that it makes sense to do what
> Go does, always emit DW_LANG_Rust unless we emit strict DWARF for
> versions before 5 (and then just fall back to DW_LANG_C).
>
> The attached patch (against "upstream gccrs") does that. I kept the
> oldlang.rs testcase just to see that the -gstrict-dwarf -gdwarf-3 case
> does something sane.
>
> The only "issue" is that is_rust () depends on the comp_unit_die
> DW_AT_language being DW_LANG_Rust. But the only usage of is_rust
> already depends on strict DWARF.
>
> https://code.wildebeest.org/git/user/mjw/gccrs/commit/?h=no-Rust-old
> if someone wants to push that, to merge for a v4.
LGTM, thanks.
Jakub
