[Bug gdb/29286] SUMMARY: ThreadSanitizer: data race gdb/dwarf2/read.c:4100 in dw_expand_symtabs_matching_file_matcher

["tromey at sourceware dot org" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=29286

--- Comment #9 from Tom Tromey <tromey at sourceware dot org> ---
(In reply to Tom de Vries from comment #7)
> Created attachment 14171 [details]
> gdb.log
> 
> While testing, found another one.
> 
> Replicating the same fix works for the test-case:
> ...
> diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
> index 6e581a6a8ea..f1a5b71b66e 100644
> --- a/gdb/dwarf2/read.c
> +++ b/gdb/dwarf2/read.c
> @@ -4280,6 +4280,11 @@
> dwarf2_base_index_functions::find_pc_sect_compunit_symtab
>      warning (_("(Internal error: pc %s in read in CU, but not in symtab.)"),
>              paddress (objfile->arch (), pc));
>  
> +  cooked_index_vector *table
> +    = (static_cast<cooked_index_vector *>
> +       (per_objfile->per_bfd->index_table.get ()));
> +  table->wait ();
> +
>    result = recursively_find_pc_sect_compunit_symtab
>      (dw2_instantiate_symtab (data, per_objfile, false), pc);

This will probably have to be a new override method in
cooked_index_functions that waits and then calls the
superclass method.  The base functions can't assume
there is a cooked index.

-- 
You are receiving this mail because:
You are on the CC list for the bug.