Re: [PATCH v3 06/24] Unit tests for gdbserver memory tagging remote packets

[Simon Marchi via Gdb-patches <gdb-patches@sourceware.org>]

On 2020-11-09 12:04 p.m., Luis Machado via Gdb-patches wrote:
> Updates on v2:
> 
> - Update unit tests to cope with additional tag type field in the remote
> packets.
> 
> --
> 
> Add some unit testing to exercise the functions handling the qMemTags and
> QMemTags packets as well as feature support.
> 
> gdbserver/ChangeLog:
> 
> YYYY-MM-DD  Luis Machado  <luis.machado@linaro.org>
> 
> 	* server.cc (test_memory_tagging_functions): New function.
> 	(captured_main): Register test_memory_tagging_functions.
> ---
>  gdbserver/server.cc | 100 ++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 100 insertions(+)
> 
> diff --git a/gdbserver/server.cc b/gdbserver/server.cc
> index 7eddd0e01f..6812b624f8 100644
> --- a/gdbserver/server.cc
> +++ b/gdbserver/server.cc
> @@ -3647,6 +3647,103 @@ detach_or_kill_for_exit_cleanup ()
>      }
>  }
>  
> +#if GDB_SELF_TEST
> +
> +namespace selftests {
> +
> +static void
> +test_memory_tagging_functions (void)
> +{
> +  /* Setup testing.  */
> +  gdb::char_vector packet;
> +  gdb::byte_vector tags, bv;
> +  std::string expected;
> +  packet.resize (32000);
> +  CORE_ADDR addr;
> +  size_t len;
> +  int type;
> +
> +  /* Test parsing a qMemTags request.  */
> +
> +  /* Invalid request, addr len and type unchanged.  */
> +  addr = 0xff;
> +  len = 255;
> +  type = 255;
> +  strcpy (packet.data (), "qMemTags_wrong:0,0:0");
> +  SELF_CHECK (parse_fmemtags_request (packet.data (), &addr, &len, &type) != 0);
> +  SELF_CHECK (addr == 0xff && len == 255 && type == 255);

As I said earlier, I would have made parse_fmemtags_request assert that the
packet begins with "qMemTags:" (instead of returning an error value), but it's
really a nit and I can live with both.

Otherwise this patch LGTM.

Simon