[PATCH] riscv: kprobe: Optimize kprobe with accurate atomicity

Mark Rutland mark.rutland at arm.com
Mon Feb 20 02:35:57 PST 2023


On Fri, Feb 17, 2023 at 12:23:51AM +0900, Masami Hiramatsu wrote:
> On Tue, 31 Jan 2023 10:33:05 +0000
> Mark Rutland <mark.rutland at arm.com> wrote:
> 
> > On Tue, Jan 31, 2023 at 09:48:29AM +0800, Guo Ren wrote:
> > > On Mon, Jan 30, 2023 at 11:49 PM Mark Rutland <mark.rutland at arm.com> wrote:
> > Masami, Steve, and I had a chat at the tracing summit late last year (which
> > unfortunately, was not recorded), and what we'd like to do is get each
> > architecture to have FPROBE (and FTRACE_WITH_ARGS), at which point OPTPROBE
> > and KRETPROBE become redundant and could be removed.
> 
> No, the fprobe will replace the KRETPROBE but not OPTPROBE. The OPTPROBE
> is completely different one. Fprobe is used only for function entry, but
> optprobe is applied to the function body.

Sorry, I had OPTPROBE and KPROBE_ON_FTRACE confused in my head, and was
thinking that FPROBE would supersede KPROBE_ON_FTRACE and KRETPROBE.

> > i.e. we'd keep KPROBES as a "you can trace any instruction" feature, but in the
> > few cases where OPTPROBES can make things fater by using FTRACE, you should
> > just use that directly via FPROBE.
> 
> I think what you are saying is KPROBE_ON_FTRACE, and that will be replaced by
> FPROBES.

Yes, sorry for the confusion.

Mark.



More information about the linux-riscv mailing list