linux-kernel - Re: [PATCH] sh: kprobes: remove unused variables in kprobe_exceptions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAMuHMdU-JhAU-n7sPgHG7aOHW8qhLS-MDtiyiotdAcJ0d0ZFFg@mail.gmail.com>
Date: Mon, 19 May 2025 12:39:14 +0200
From: Geert Uytterhoeven <geert@...ux-m68k.org>
To: Mike Rapoport <rppt@...nel.org>
Cc: John Paul Adrian Glaubitz <glaubitz@...sik.fu-berlin.de>, Mike Rapoport <rppt@...il.com>, 
	Rich Felker <dalias@...c.org>, Yoshinori Sato <ysato@...rs.sourceforge.jp>, 
	linux-kernel@...r.kernel.org, linux-sh@...r.kernel.org, 
	kernel test robot <lkp@...el.com>, Masami Hiramatsu <mhiramat@...nel.org>
Subject: Re: [PATCH] sh: kprobes: remove unused variables in kprobe_exceptions_notify()

Hi Mike,

On Mon, 19 May 2025 at 12:02, Mike Rapoport <rppt@...nel.org> wrote:
> On Mon, May 19, 2025 at 10:48:20AM +0200, Geert Uytterhoeven wrote:
> > On Sat, 17 May 2025 at 11:30, Mike Rapoport <rppt@...nel.org> wrote:
> > > From: "Mike Rapoport (Microsoft)" <rppt@...nel.org>
> > >
> > > kbuild reports the following warning:
> > >
> > >    arch/sh/kernel/kprobes.c: In function 'kprobe_exceptions_notify':
> > > >> arch/sh/kernel/kprobes.c:412:24: warning: variable 'p' set but not used [-Wunused-but-set-variable]
> > >      412 |         struct kprobe *p = NULL;
> > >          |                        ^
> > >
> > > The variable 'p' is indeed unused since the commit fa5a24b16f94
> > > ("sh/kprobes: Don't call the ->break_handler() in SH kprobes code")
> > >
> > > Remove that variable along with 'kprobe_opcode_t *addr' which also
> > > becomes unused after 'p' is removed.
> > >
> > > Reported-by: kernel test robot <lkp@...el.com>
> > > Closes: https://lore.kernel.org/oe-kbuild-all/202505151341.EuRFR22l-lkp@intel.com/
> > > Fixes: fa5a24b16f94 ("sh/kprobes: Don't call the ->break_handler() in SH kprobes code")
> > > Signed-off-by: Mike Rapoport (Microsoft) <rppt@...nel.org>
> >
> > Thanks for your patch!
> >
> > "p" and "addr" are definitely unused (besides side-effects?), so
> > Reviewed-by: Geert Uytterhoeven <geert+renesas@...der.be>
> >
> > > --- a/arch/sh/kernel/kprobes.c
> > > +++ b/arch/sh/kernel/kprobes.c
> > > @@ -404,13 +404,10 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
> > >  int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
> > >                                        unsigned long val, void *data)
> > >  {
> > > -       struct kprobe *p = NULL;
> > >         struct die_args *args = (struct die_args *)data;
> > >         int ret = NOTIFY_DONE;
> > > -       kprobe_opcode_t *addr = NULL;
> > >         struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
> > >
> > > -       addr = (kprobe_opcode_t *) (args->regs->pc);
> > >         if (val == DIE_TRAP &&
> > >             args->trapnr == (BREAKPOINT_INSTRUCTION & 0xff)) {
> > >                 if (!kprobe_running()) {
> > > @@ -421,7 +418,6 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
> > >                                 ret = NOTIFY_DONE;
> > >                         }
> > >                 } else {
> > > -                       p = get_kprobe(addr);
> > >                         if ((kcb->kprobe_status == KPROBE_HIT_SS) ||
> > >                             (kcb->kprobe_status == KPROBE_REENTER)) {
> > >                                 if (post_kprobe_handler(args->regs))
> >
> > I have no idea what this code is supposed to do, and if it actually
> > works.  Red flags are that the assigned "p" was never used at all
> > since the inception of this function.
>
> "p" was used before fa5a24b16f94 ("sh/kprobes: Don't call the
> ->break_handler() in SH kprobes code"), but I can't say I understand that
> code either :)

Yes, the _variable_ "p" was used before, but not before assigning a
different value to it first.

I guess this is where Rust would be helpful? Although C can and does
give a warning here, too...

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@...ux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds