Re: [PATCH v2] gdbserver: Check r_version < 1 for Linux debugger interface

["H.J. Lu via Libc-alpha" <libc-alpha@sourceware.org>]

On Mon, Aug 16, 2021 at 3:02 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Mon, Aug 16, 2021 at 1:30 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >
> > Update gdbserver to check r_version < 1 instead of r_version != 1 so
> > that r_version can be bumped for a new field in the glibc debugger
> > interface to support multiple namespaces.  Since so far, the gdbserver
> > only reads fields defined for r_version == 1, it is compatible with
> > r_version >= 1.
> >
> > All future glibc debugger interface changes will be backward compatible.
> > If there is ever the need for backward incompatible change to the glibc
> > debugger interface, a new DT_XXX element will be provided to access the
> > new incompatible interface.
> >
> >         PR gdb/11839
> >         * linux-low.cc (linux_process_target::qxfer_libraries_svr4):
> >         Check r_version < 1 instead of r_version != 1.
> > ---
> >  gdbserver/linux-low.cc | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/gdbserver/linux-low.cc b/gdbserver/linux-low.cc
> > index 5c6191d941c..fc7a995351d 100644
> > --- a/gdbserver/linux-low.cc
> > +++ b/gdbserver/linux-low.cc
> > @@ -6845,7 +6845,7 @@ linux_process_target::qxfer_libraries_svr4 (const char *annex,
> >           if (linux_read_memory (priv->r_debug + lmo->r_version_offset,
> >                                  (unsigned char *) &r_version,
> >                                  sizeof (r_version)) != 0
> > -             || r_version != 1)
> > +             || r_version < 1)
> >             {
> >               warning ("unexpected r_debug version %d", r_version);
> >             }
> > --
> > 2.31.1
> >
>
> Set r_version == 2 breaks GDB due to
>
> static CORE_ADDR
> solib_svr4_r_ldsomap (struct svr4_info *info)
> {
>   struct link_map_offsets *lmo = svr4_fetch_link_map_offsets ();
>   struct type *ptr_type = builtin_type (target_gdbarch ())->builtin_data_ptr;
>   enum bfd_endian byte_order = type_byte_order (ptr_type);
>   ULONGEST version = 0;
>
>   try
>     {
>       /* Check version, and return zero if `struct r_debug' doesn't have
>          the r_ldsomap member.  */
>       version
>         = read_memory_unsigned_integer (info->debug_base +
> lmo->r_version_offset,
>                                         lmo->r_version_size, byte_order);
>     }
>   catch (const gdb_exception_error &ex)
>     {
>       exception_print (gdb_stderr, ex);
>     }
>
>   if (version < 2 || lmo->r_ldsomap_offset == -1)
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
> glibc doesn't have r_ldsomap.  But r_ldsomap_offset is set
> unconditionally.   Shouldn't it be set only if the target debugger
> interface has it?
>
>     return 0;
>
>   return read_memory_typed_address (info->debug_base + lmo->r_ldsomap_offset,
>                                     ptr_type);
> }
>

I opened:

https://sourceware.org/bugzilla/show_bug.cgi?id=28236

-- 
H.J.