linux-kernel - Re: How can I check creator of probe point?

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: <543F5EFA.7050209@hitachi.com>
Date:	Thu, 16 Oct 2014 15:00:26 +0900
From:	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
To:	Gioh Kim <gioh.kim@....com>
Cc:	Ananth N Mavinakayanahalli <ananth@...ibm.com>,
	Anil S Keshavamurthy <anil.s.keshavamurthy@...el.com>,
	"David S. Miller" <davem@...emloft.net>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: How can I check creator of probe point?

Hello Gioh,

If you can use ftrace and perftool, you can also put probes on
caller site. perf-probe will give you source-code level probes.
Or, just trace kmalloc event with ftrace stacktrace option, which
gives you caller information so that you can filter your driver
by postprocessing.

Thank you,

(2014/10/14 11:16), Gioh Kim wrote:
> 
> Hi,
> 
> I am trying to find a way to make statistics for memory allocation of my device driver.
> I want to know how much memory it allocates and how many times it calls kmalloc().
> 
> So I am considering to use kprobe but I think it doesn't provide a way to identify who makes the probe point.
> Can I distinguish kmalloc() calling only from my driver?
> 
> For example I think it could be like this:
> 
> diff --git a/samples/kprobes/kretprobe_example.c b/samples/kprobes/kretprobe_example.c
> index 1041b67..5322e0a 100644
> --- a/samples/kprobes/kretprobe_example.c
> +++ b/samples/kprobes/kretprobe_example.c
> @@ -32,6 +32,7 @@ MODULE_PARM_DESC(func, "Function to kretprobe; this module will report the"
> 
>  /* per-instance private data */
>  struct my_data {
> +       unsigned long signature;
>         ktime_t entry_stamp;
>  };
> 
> @@ -43,8 +44,10 @@ static int entry_handler(struct kretprobe_instance *ri, struct pt_regs *regs)
>         if (!current->mm)
>                 return 1;       /* Skip kernel threads */
> 
> -       data = (struct my_data *)ri->data;
> -       data->entry_stamp = ktime_get();
> +       if (signature == 0xabcdabcd) {
> +               data = (struct my_data *)ri->data;
> +               data->entry_stamp = ktime_get();
> +       }
>         return 0;
>  }
> 
> @@ -60,10 +63,12 @@ static int ret_handler(struct kretprobe_instance *ri, struct pt_regs *regs)
>         s64 delta;
>         ktime_t now;
> 
> -       now = ktime_get();
> -       delta = ktime_to_ns(ktime_sub(now, data->entry_stamp));
> -       printk(KERN_INFO "%s returned %d and took %lld ns to execute\n",
> -                       func_name, retval, (long long)delta);
> +       if (signature == 0xabcdabcd) {
> +               now = ktime_get();
> +               delta = ktime_to_ns(ktime_sub(now, data->entry_stamp));
> +               printk(KERN_INFO "%s returned %d and took %lld ns to execute\n",
> +                      func_name, retval, (long long)delta);
> +       }
>         return 0;
>  }
> 
> 


-- 
Masami HIRAMATSU
Software Platform Research Dept. Linux Technology Research Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: masami.hiramatsu.pt@...achi.com


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/