Subject: Re: [PATCH v12 0/5] Add kprobe and kretprobe support for LoongArch
To:     Jeff Xie <xiehuan09@gmail.com>, Huacai Chen <chenhuacai@kernel.org>
References: <1674007261-9198-1-git-send-email-yangtiezhu@loongson.cn>
 <CAAhV-H4aTd6_cSy45KKjv-KrLTiwT4iG6+fkb84KfCrL3Y+hpg@mail.gmail.com>
 <CAAhV-H5WN5E=0Z9wpbXDc6VO7Nc+j7PGvnyAAGOmCRMJkdwSYw@mail.gmail.com>
 <CAEr6+ECO-=jfhzHrcdKGx0MsjMBMiN6wsBPCfv7CaXo_amAWWg@mail.gmail.com>
Cc:     WANG Xuerui <kernel@xen0n.name>,
        Masami Hiramatsu <mhiramat@kernel.org>,
        loongarch@lists.linux.dev, linux-kernel@vger.kernel.org
From:   Tiezhu Yang <yangtiezhu@loongson.cn>
Message-ID: <02806f85-bc09-d316-f058-3947353cb190@loongson.cn>
Date:   Thu, 2 Feb 2023 10:23:13 +0800
User-Agent: Mozilla/5.0 (X11; Linux mips64; rv:45.0) Gecko/20100101
 Thunderbird/45.4.0
MIME-Version: 1.0
In-Reply-To: <CAEr6+ECO-=jfhzHrcdKGx0MsjMBMiN6wsBPCfv7CaXo_amAWWg@mail.gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 02/01/2023 05:40 PM, Jeff Xie wrote:
> On Wed, Feb 1, 2023 at 12:56 PM Huacai Chen <chenhuacai@kernel.org> wrote:
>>
>> Hi, Jeff,
>>
>> Could you please pay some time to test this series? Thank you.
>
> Thanks for notifying me about the test.
>
> I have tested the patchset(based on the
> https://github.com/loongson/linux/tree/loongarch-next),
> I found that some functions can't  be probed e.g. scheduler_tick() or
> uart_write_wakeup()
> the two functions have the same point,  they are all run in the hardirq context.
>
> I don't know if it's related to the hardirq context, I haven't had
> time to study this patchset carefully.
> and they can be probed in the x86_64 arch.
>
> root@loongarch modules]# insmod ./kprobe_example.ko symbol=scheduler_tick
> insmod: can't insert './kprobe_example.ko': invalid parameter
>
> dmesg:
> [   39.806435] kprobe_init: register_kprobe failed, returned -22
>

Thanks for your test.

On my test environment, I can not reproduce the above issue,
here are the test results, it seems no problem.

[root@linux loongson]# dmesg -c
[root@linux loongson]# uname -m
loongarch64
[root@linux loongson]# modprobe kprobe_example symbol=scheduler_tick
[root@linux loongson]# rmmod kprobe_example
[root@linux loongson]# dmesg | tail -2
[ 3317.138086] handler_post: <scheduler_tick> p->addr = 
0x0000000065d12f66, estat = 0xc0000
[ 3317.154086] kprobe_exit: kprobe at 0000000065d12f66 unregistered

[root@linux loongson]# dmesg -c
[root@linux loongson]# uname -m
loongarch64
[root@linux loongson]# modprobe kprobe_example symbol=uart_write_wakeup
[root@linux loongson]# rmmod kprobe_example
[root@linux loongson]# dmesg | tail -2
[ 3433.502092] handler_post: <uart_write_wakeup> p->addr = 
0x0000000019718061, estat = 0xc0000
[ 3433.762085] kprobe_exit: kprobe at 0000000019718061 unregistered

Additionally, "register_kprobe failed, returned -22" means the symbol
can not be probed, here is the related code:

register_kprobe()
   check_kprobe_address_safe()

static int check_kprobe_address_safe(struct kprobe *p,
				     struct module **probed_mod)
{
	int ret;

	ret = check_ftrace_location(p);
	if (ret)
		return ret;
	jump_label_lock();
	preempt_disable();

	/* Ensure it is not in reserved area nor out of text */
	if (!(core_kernel_text((unsigned long) p->addr) ||
	    is_module_text_address((unsigned long) p->addr)) ||
	    in_gate_area_no_mm((unsigned long) p->addr) ||
	    within_kprobe_blacklist((unsigned long) p->addr) ||
	    jump_label_text_reserved(p->addr, p->addr) ||
	    static_call_text_reserved(p->addr, p->addr) ||
	    find_bug((unsigned long)p->addr)) {
		ret = -EINVAL;
		goto out;
	}
...
}

Thanks,
Tiezhu