Re: [PATCH] ggc, jit: forcibly clear GTY roots in jit

[David Malcolm <dmalcolm@redhat.com>]

On Tue, 2023-09-12 at 13:36 -0400, Antoni Boucher wrote:
> In the mean time, here's a (Rust) reproducer for the issue:
> 
> fn main() {
>     for _ in 0..5 {
>         let context = Context::default();
>         context.add_command_line_option("-flto");
>        
> context.set_optimization_level(OptimizationLevel::Aggressive);
>         context.add_driver_option("-nostdlib");
> 
>         let int_type = context.new_type::<i32>();
> 
>         let function = context.new_function(None,
> FunctionType::Exported, int_type, &[], "main", false);
>         let block = function.new_block("start");
>         let value = context.new_rvalue_from_int(int_type, 42);
>         block.end_with_return(None, value);
> 
>         context.compile_to_file(OutputKind::Executable, "my_exe");
>     }
> }

Can we get this in bugzilla please?  If you generate a .c version of
the context (via gcc_jit_context_dump_reproducer_to_file) I can try to
debug it.

Thanks
Dave