Re: [PATCH 25/26] tcg: exclude lookup_tb_ptr from helper instrumentation

[Richard Henderson]

On 1/10/23 09:39, Alex Bennée wrote:
> From: Emilio Cota <cota@braap.org>
>
> It is internal to TCG and therefore we know it does not
> access guest memory.
>
> Related: #1381
>
> Signed-off-by: Emilio Cota <cota@braap.org>
> Message-Id: <20230108164731.61469-4-cota@braap.org>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>   tcg/tcg.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/tcg/tcg.c b/tcg/tcg.c
> index da91779890..ee67eefc0c 100644
> --- a/tcg/tcg.c
> +++ b/tcg/tcg.c
> @@ -1652,8 +1652,10 @@ void tcg_gen_callN(void *func, TCGTemp *ret, int nargs, 
> TCGTemp **args)
>       op = tcg_op_alloc(INDEX_op_call, total_args);
>   
>   #ifdef CONFIG_PLUGIN
> -    /* detect non-plugin helpers */
> -    if (tcg_ctx->plugin_insn && unlikely(strncmp(info->name, "plugin_", 7))) {
> +    /* flag helpers that are not internal to TCG */
> +    if (tcg_ctx->plugin_insn &&
> +        strncmp(info->name, "plugin_", 7) &&
> +        strcmp(info->name, "lookup_tb_ptr")) {
>           tcg_ctx->plugin_insn->calls_helpers = true;
>       }
>   #endif

I think this should be detected with

  !(info->flags & TCG_CALL_NO_SIDE_EFFECTS)

i.e., side-effects, which in this case is the possibility of a fault.


r~