On 2025/9/24 07:07, Masami Hiramatsu wrote: > On Tue, 23 Sep 2025 17:20:00 +0800 > Menglong Dong <[email protected]> wrote: > > > The fprobe_entry() is used by fgraph_ops, so rename it to > > fprobe_fgraph_entry to be more distinctive. > > Sorry, NAK. fprobe is based on fgraph by design. > So "fprobe_fgraph" sounds redundant.
Hi, Masami. Did you see my next patch that use ftrace for the fprobe to obtain better performance? Hmm, it seems that the cover-letter is necessary :/ Thanks! Menglong Dong > > Thanks, > > > > > Signed-off-by: Menglong Dong <[email protected]> > > --- > > kernel/trace/fprobe.c | 8 ++++---- > > 1 file changed, 4 insertions(+), 4 deletions(-) > > > > diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c > > index 6a205903b1ed..1785fba367c9 100644 > > --- a/kernel/trace/fprobe.c > > +++ b/kernel/trace/fprobe.c > > @@ -254,8 +254,8 @@ static inline int __fprobe_kprobe_handler(unsigned long > > ip, unsigned long parent > > return ret; > > } > > > > -static int fprobe_entry(struct ftrace_graph_ent *trace, struct fgraph_ops > > *gops, > > - struct ftrace_regs *fregs) > > +static int fprobe_fgraph_entry(struct ftrace_graph_ent *trace, struct > > fgraph_ops *gops, > > + struct ftrace_regs *fregs) > > { > > unsigned long *fgraph_data = NULL; > > unsigned long func = trace->func; > > @@ -340,7 +340,7 @@ static int fprobe_entry(struct ftrace_graph_ent *trace, > > struct fgraph_ops *gops, > > /* If any exit_handler is set, data must be used. */ > > return used != 0; > > } > > -NOKPROBE_SYMBOL(fprobe_entry); > > +NOKPROBE_SYMBOL(fprobe_fgraph_entry); > > > > static void fprobe_return(struct ftrace_graph_ret *trace, > > struct fgraph_ops *gops, > > @@ -379,7 +379,7 @@ static void fprobe_return(struct ftrace_graph_ret > > *trace, > > NOKPROBE_SYMBOL(fprobe_return); > > > > static struct fgraph_ops fprobe_graph_ops = { > > - .entryfunc = fprobe_entry, > > + .entryfunc = fprobe_fgraph_entry, > > .retfunc = fprobe_return, > > }; > > static int fprobe_graph_active; > > >
