[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20230811131001.7b22a17d@gandalf.local.home>
Date: Fri, 11 Aug 2023 13:10:01 -0400
From: Steven Rostedt <rostedt@...dmis.org>
To: "Masami Hiramatsu (Google)" <mhiramat@...nel.org>
Cc: Florent Revest <revest@...omium.org>,
Alexei Starovoitov <alexei.starovoitov@...il.com>,
linux-trace-kernel@...r.kernel.org,
LKML <linux-kernel@...r.kernel.org>,
Martin KaFai Lau <martin.lau@...ux.dev>,
bpf <bpf@...r.kernel.org>, Sven Schnelle <svens@...ux.ibm.com>,
Alexei Starovoitov <ast@...nel.org>,
Jiri Olsa <jolsa@...nel.org>,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Alan Maguire <alan.maguire@...cle.com>,
Mark Rutland <mark.rutland@....com>,
Peter Zijlstra <peterz@...radead.org>,
Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [RFC PATCH v2 1/6] fprobe: Use fprobe_regs in fprobe entry
handler
On Thu, 10 Aug 2023 07:13:30 +0900
Masami Hiramatsu (Google) <mhiramat@...nel.org> wrote:
> >
> > I hope that makes my thoughts clearer? It's a hairy topic ahah
>
> Ah, I see your point.
>
> static void fprobe_init(struct fprobe *fp)
> {
> fp->nmissed = 0;
> if (fprobe_shared_with_kprobes(fp))
> fp->ops.func = fprobe_kprobe_handler;
> else
> fp->ops.func = fprobe_handler;
> fp->ops.flags |= FTRACE_OPS_FL_SAVE_REGS; <---- This flag!
> }
>
> So it should be FTRACE_OPS_FL_ARGS. Let me fix that.
Yes, this was the concern that I was bringing up, where I did not see an
advantage of fprobes over kprobes using ftrace, because they both were
saving all registers.
-- Steve
Powered by blists - more mailing lists