Re: [PATCH v2] sctp: loading sctp when load sctp_probe

[Wang Weidong <weidong1991.wang@gmail.com>]

From: Wang Weidong <wangweidong1@huawei.com>

On 2013/12/16 22:32, Neil Horman wrote:
> On Mon, Dec 16, 2013 at 10:14:55PM +0800, Wang Weidong wrote:
>> From: Wang Weidong <wangweidong1@huawei.com>
>>
>> On 2013/12/16 21:48, Neil Horman wrote:
>>> On Mon, Dec 16, 2013 at 10:47:00AM +0800, Wang Weidong wrote:
>>>> On 2013/12/13 20:26, Neil Horman wrote:
>>>>> On Fri, Dec 13, 2013 at 11:00:10AM +0800, Wang Weidong wrote:
>>>>>> when I modprobe sctp_probe, it failed with "FATAL: ". I found that
>>>>>> sctp should load before sctp_probe register jprobe. So I add a
>>>>>> sctp_setup_jprobe for loading 'sctp' when first failed to register
>>>>>> jprobe, just do this similar to dccp_probe.
>>>>>>
>>>>>> v2: add MODULE_SOFTDEP and check of request_module, as suggested by Neil
>>>>>>
>>>>>> Signed-off-by: Wang Weidong <wangweidong1@huawei.com>
>>>>>> ---
>>>>>>   net/sctp/probe.c | 17 ++++++++++++++++-
>>>>>>   1 file changed, 16 insertions(+), 1 deletion(-)
>>>>>>
>>>>>> diff --git a/net/sctp/probe.c b/net/sctp/probe.c
>>>>>> index 53c452e..5e68b94 100644
>>>>>> --- a/net/sctp/probe.c
>>>>>> +++ b/net/sctp/probe.c
>>>>>> @@ -38,6 +38,7 @@
>>>>>>   #include <net/sctp/sctp.h>
>>>>>>   #include <net/sctp/sm.h>
>>>>>>
>>>>>> +MODULE_SOFTDEP("pre: sctp");
>>>>>>   MODULE_AUTHOR("Wei Yongjun <yjwei@cn.fujitsu.com>");
>>>>>>   MODULE_DESCRIPTION("SCTP snooper");
>>>>>>   MODULE_LICENSE("GPL");
>>>>>> @@ -182,6 +183,20 @@ static struct jprobe sctp_recv_probe = {
>>>>>>   	.entry	= jsctp_sf_eat_sack,
>>>>>>   };
>>>>>>
>>>>>> +static __init int sctp_setup_jprobe(void)
>>>>>> +{
>>>>>> +	int ret = register_jprobe(&sctp_recv_probe);
>>>>>> +
>>>>>> +	if (ret) {
>>>>>> +		if (request_module("sctp"))
>>>>>> +			goto out;
>>>>>> +		ret = register_jprobe(&sctp_recv_probe);
>>>>>> +	}
>>>>>> +
>>>>>> +out:
>>>>>> +	return ret;
>>>>>> +}
>>>>>> +
>>>>>>   static __init int sctpprobe_init(void)
>>>>>>   {
>>>>>>   	int ret = -ENOMEM;
>>>>>> @@ -202,7 +217,7 @@ static __init int sctpprobe_init(void)
>>>>>>   			 &sctpprobe_fops))
>>>>>>   		goto free_kfifo;
>>>>>>
>>>>>> -	ret = register_jprobe(&sctp_recv_probe);
>>>>>> +	ret = sctp_setup_jprobe();
>>>>> You don't need to create your own function for this, you can collapse it down to
>>>>> a call to try_then_request_module(regitser_jprobe(...), "sctp");
>>>>> Neil
>>>>>
>>>> Hi Neil,
>>>>
>>>> I try to use try_then_request_module(!regitser_jprobe(...), "sctp"); I found that if
>>>> I used this, I couldn't get the ture value which returned by register_jprobe(). Is the
>>>> returned value not important? The problem("if sctp.ko doesn't exist") you had pointed out
>>>> is exist as well.
>>>>
>>>> Regards.
>>>> Wang
>>>
>>> Not sure I follow.  There appear to be lots of examples in the kernel where
>>
>> I do this just like the dccp/probe.c do. So I think it can be an example.
>>
>
> Ah thats right, I remember doing this before now. My bad.
>
>>> exactly what you are trying to do is done.  try_then_request_module calls your
>>> requested method, and if it returns non-zero, calls request_module, then calls
>>
>> No actually. Just look the definition:
>> #define try_then_request_module(x, mod...) \
>>           ((x) ?: (__request_module(true, mod), (x)))
>> if x return non-zero, it will return the value, not call request_module.
>>
>>> your requested method again, returning the final result.  What problem are you
>>> running into?
>>> Neil
>>>
>>
>> So I only can use it like this:
>> try_then_request_module(!regitser_jprobe(...), "sctp");
>>
>> when register_jprobe return 0 indicates success, I get the value is 1. It is OK.
>>
>> when register_jprobe is non-zero indicates failed, then calls request_module,
>> and last calls register_jprobe. Here,
>> 1)maybe it will failed return non-zero, what I get the value is !register_jprobe
>> is 0 with losing the value.
>> 2)request_module failed with the sctp is not exist, it will do the register_jprobe
>> as well.
>>
>> Is there I am wrong somewhere?
>>
>
> It really seems like there should be a way to roll up what you want into the
> try_then_request_module macro.  What about:
>
> ret = try_then_request_module((register_jprobe(...) = 0), "sctp");
>
The first problem maybe exist. because the last value is x. So if use this
it always return (register_jprobe(...) = 0) while the value is only 1 or 0.

> That seems like it should work.  It will mask the actual return value of
> register_jprobe though I think, which kind of sucks.  Its almost like we need
> another variant of this macro that accepts an expected condition result so the
> macro can use it to compare, something like:
>
> #define try_then_request_module_cond(x, result, mod...) \
> 	((x) = (result) ?: __request_module(true, mod), (x)))
>
> that would let you set an expected comparison value for your macro, but get the
> result of the method if it still fails after the module is loaded, or if it
> fails to load.

I think this is a good idea.

>
> Thats probably a cleanup for a second patch.  I think your version two is
> probably as close as we're going to get right now.  So for v2:
>
> Acked-by: Neil Horman <nhorman@tuxdriver.com>
>
Thanks.

Will you send the cleanup patch?

Regards.
Wang