Re: [PATCH 2/4] tracing/eprobes: Do not hardcode $comm as a string

[Google]

On Fri, 19 Aug 2022 21:40:37 -0400
Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:

> From: "Steven Rostedt (Google)" <rostedt@xxxxxxxxxxx>
> 
> The variable $comm is hard coded as a string, which is true for both
> kprobes and uprobes, but for event probes (eprobes) it is a field name. In
> most cases the "comm" field would be a string, but there's no guarantee of
> that fact.
> 
> Do not assume that comm is a string. Not to mention, it currently forces
> comm fields to fault, as string processing for event probes is currently
> broken.

Indeed. There should be an event argument which names "comm".
Eprobe might refer it. BTW, does eprobe use any special common fields?
I originally introduced "$" variable for such special variables.

Thank you,
> 
> Cc: stable@xxxxxxxxxxxxxxx
> Fixes: 7491e2c44278 ("tracing: Add a probe that attaches to trace events")
> Signed-off-by: Steven Rostedt (Google) <rostedt@xxxxxxxxxxx>
> ---
>  kernel/trace/trace_probe.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c
> index dec657af363c..23dcd52ad45c 100644
> --- a/kernel/trace/trace_probe.c
> +++ b/kernel/trace/trace_probe.c
> @@ -622,9 +622,10 @@ static int traceprobe_parse_probe_arg_body(const char *argv, ssize_t *size,
>  
>  	/*
>  	 * Since $comm and immediate string can not be dereferenced,
> -	 * we can find those by strcmp.
> +	 * we can find those by strcmp. But ignore for eprobes.
>  	 */
> -	if (strcmp(arg, "$comm") == 0 || strncmp(arg, "\\\"", 2) == 0) {
> +	if (!(flags & TPARG_FL_TPOINT) &&
> +	    strcmp(arg, "$comm") == 0 || strncmp(arg, "\\\"", 2) == 0) {
>  		/* The type of $comm must be "string", and not an array. */
>  		if (parg->count || (t && strcmp(t, "string")))
>  			goto out;
> -- 
> 2.35.1


-- 
Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>