Messages in this thread | | | Date | Mon, 24 Mar 2014 14:53:51 -0400 | From | Steven Rostedt <> | Subject | Re: [PATCH -tip v8 03/26] kprobes: Prohibit probing on .entry.text code |
| |
On Mon, 24 Mar 2014 10:48:02 +0900 Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> wrote:
> (2014/03/22 7:04), Steven Rostedt wrote: > > On Wed, 05 Mar 2014 20:59:04 +0900 > > Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> wrote: > > > > > >> struct kprobe_insn_cache { > >> struct mutex mutex; > >> diff --git a/kernel/kprobes.c b/kernel/kprobes.c > >> index ceeadfc..5b5ac76 100644 > >> --- a/kernel/kprobes.c > >> +++ b/kernel/kprobes.c > >> @@ -96,9 +96,6 @@ static raw_spinlock_t *kretprobe_table_lock_ptr(unsigned long hash) > >> static struct kprobe_blackpoint kprobe_blacklist[] = { > >> {"preempt_schedule",}, > >> {"native_get_debugreg",}, > >> - {"irq_entries_start",}, > >> - {"common_interrupt",}, > >> - {"mcount",}, /* mcount can be called from everywhere */ > > > > Is mcount in the entry.text section? Also, what about ftrace_caller and > > friends. > > I saw mcount(and ftrace_caller) in arch/x86/kernel/entry_*.S, and > almost all code in it are in ".entry.text" section as far as I can see, > or do I miss something? >
Ah, checking the System.map I guess we are OK.
-- Steve
| |