Re: [PATCH] Fix PR86413

[Jason Merrill <jason at redhat dot com>]

OK.

On Fri, Jul 6, 2018 at 8:14 PM, Richard Biener <rguenther@suse.de> wrote:
>
> The following fixes
>
> FAIL: gcc.dg/guality/pr48437.c   -O2 -flto -fno-use-linker-plugin
> -flto-partition=none  line 14 i == 0
>
> because we now prune non-local/VAR_DECLs from BLOCK trees during
> free-lang-data (after we emitted early dwarf).  gen_block_die
> isn't prepared for that and now refuses to add low/high-PC
> attributes to blocks that got all BLOCK_VARS stripped that way.
>
> The fix is to simply always emit them for early generated DIEs
> (so we only ever elide the DIE creation during early dwarf).
>
> Note this would allow us to prune BLOCK_VARS completely after
> early dwarf generation (but we need to keep the BLOCK tree itself
> for scoping obviously).
>
> Bootstrap / regtest running on x86_64-unknown-linux-gnu, ok?
>
> Thanks,
> Richard.
>
> 2018-07-06  Richard Biener  <rguenther@suse.de>
>
>         PR debug/86413
>         * dwarf2out.c (gen_block_die): For an early generated DIE
>         always output high/low PC attributes.
>
> diff --git a/gcc/dwarf2out.c b/gcc/dwarf2out.c
> index a7c4620cfc3..95232177d83 100644
> --- a/gcc/dwarf2out.c
> +++ b/gcc/dwarf2out.c
> @@ -25622,6 +25622,11 @@ gen_block_die (tree stmt, dw_die_ref context_die)
>      /* The outer scopes for inlinings *must* always be represented.  We
>         generate DW_TAG_inlined_subroutine DIEs for them.  (See below.) */
>      must_output_die = 1;
> +  else if (BLOCK_DIE (stmt))
> +    /* If we already have a DIE then it was filled early.  Meanwhile
> +       we might have pruned all BLOCK_VARS as optimized out but we
> +       still want to generate high/low PC attributes so output it.  */
> +    must_output_die = 1;
>    else
>      {
>        /* Determine if this block directly contains any "significant"