Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH v3 1/2] printk: Fix panic caused by passing log_buf_len to
 command line
To:     Steven Rostedt <rostedt@goodmis.org>
CC:     <pmladek@suse.com>, <sergey.senozhatsky@gmail.com>,
        <linux-kernel@vger.kernel.org>
References: <1537630852-247674-1-git-send-email-zhe.he@windriver.com>
 <20180922121905.3e4159eb@vmware.local.home>
From:   He Zhe <zhe.he@windriver.com>
Message-ID: <e8d31c43-145b-df78-6e78-9e6ad26c3a31@windriver.com>
Date:   Sun, 23 Sep 2018 14:51:12 +0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.1
MIME-Version: 1.0
In-Reply-To: <20180922121905.3e4159eb@vmware.local.home>
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 2018年09月23日 00:19, Steven Rostedt wrote:
> On Sat, 22 Sep 2018 23:40:51 +0800
> <zhe.he@windriver.com> wrote:
>
>> From: He Zhe <zhe.he@windriver.com>
>>
>> log_buf_len_setup does not check input argument before passing it to
>> simple_strtoull. The argument would be a NULL pointer if "log_buf_len",
>> without its value, is set in command line and thus causes the following
>> panic.
>>
>> PANIC: early exception 0xe3 IP 10:ffffffffaaeacd0d error 0 cr2 0x0
>> [    0.000000] CPU: 0 PID: 0 Comm: swapper Not tainted 4.19.0-rc4-yocto-standard+ #1
>> [    0.000000] RIP: 0010:_parse_integer_fixup_radix+0xd/0x70
>> ...
>> [    0.000000] Call Trace:
>> [    0.000000]  simple_strtoull+0x29/0x70
>> [    0.000000]  memparse+0x26/0x90
>> [    0.000000]  log_buf_len_setup+0x17/0x22
>> [    0.000000]  do_early_param+0x57/0x8e
>> [    0.000000]  parse_args+0x208/0x320
>> [    0.000000]  ? rdinit_setup+0x30/0x30
>> [    0.000000]  parse_early_options+0x29/0x2d
>> [    0.000000]  ? rdinit_setup+0x30/0x30
>> [    0.000000]  parse_early_param+0x36/0x4d
>> [    0.000000]  setup_arch+0x336/0x99e
>> [    0.000000]  start_kernel+0x6f/0x4ee
>> [    0.000000]  x86_64_start_reservations+0x24/0x26
>> [    0.000000]  x86_64_start_kernel+0x6f/0x72
>> [    0.000000]  secondary_startup_64+0xa4/0xb0
>>
>> This patch adds a check to prevent the panic.
>>
>> Signed-off-by: He Zhe <zhe.he@windriver.com>
>> Cc: stable@vger.kernel.org
> I just tried this on a 2.6.32 kernel, and it crashes there. I guess
> this goes farther back than git history goes.
>
> Perhaps it should be commented that this bug has been here since
> creation of (git) time.

I did a try on 2.6.32. It passed. Actually this bug only happens on
early_param(not __setup) which is introduced since v3.0. The oldest
LTS version is 3.16 now. Should I send v4 and add a statement about
the supported version range in commit log?

>
>
>> Cc: pmladek@suse.com
>> Cc: sergey.senozhatsky@gmail.com
>> Cc: rostedt@goodmis.org
>> ---
>> v2:
>> Split out the addition of pr_fmt and the unsigned update
> Which unsigned update? As it does switch to unsigned to "unsigned int",
> but that change is fine to me with this.

No problem. It's the history of v2.

In v1 you suggested "unsigned int size" should be in a separate patch and
I did that in v2. Then Sergey suggested "unsigned int size" should be in the
1/2 patch to avoid checkpatch.pl warning. With your conformation, I add it
back here in v3.

Thanks,
Zhe

>
>> v3:
>> Use more clear error info
>> Change unsigned to unsigned in to avoid checkpatch.pl warning
>>
>>  kernel/printk/printk.c | 9 ++++++++-
>>  1 file changed, 8 insertions(+), 1 deletion(-)
>>
>> diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
>> index 9bf5404..d9821c0 100644
>> --- a/kernel/printk/printk.c
>> +++ b/kernel/printk/printk.c
>> @@ -1048,7 +1048,14 @@ static void __init log_buf_len_update(unsigned size)
>>  /* save requested log_buf_len since it's too early to process it */
>>  static int __init log_buf_len_setup(char *str)
>>  {
>> -	unsigned size = memparse(str, &str);
>> +	unsigned int size;
> I'm OK with the int update too, as its low risk.
>
> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
>
> -- Steve
>
>> +
>> +	if (!str) {
>> +		pr_err("boot command line parameter value not provided\n");
>> +		return -EINVAL;
>> +	}
>> +
>> +	size = memparse(str, &str);
>>  
>>  	log_buf_len_update(size);
>>  
>