Re: [PATCH v4 2/3] ipv4: Namespaceify tcp_fastopen_key knob

[David Miller]

From: äæå <yanhaishuang@xxxxxxxxxxxxxxxxxxxx>
Date: Tue, 26 Sep 2017 09:25:51 +0800

>> On 2017å9æ26æ, at äå7:24, David Miller <davem@xxxxxxxxxxxxx> wrote:
>> 
>> From: Haishuang Yan <yanhaishuang@xxxxxxxxxxxxxxxxxxxx>
>> Date: Fri, 22 Sep 2017 21:48:43 +0800
>> 
>>> @@ -9,13 +9,18 @@
>>> #include <net/inetpeer.h>
>>> #include <net/tcp.h>
>>> 
>>> -struct tcp_fastopen_context __rcu *tcp_fastopen_ctx;
>>> -
>>> -static DEFINE_SPINLOCK(tcp_fastopen_ctx_lock);
>>> -
>>> -void tcp_fastopen_init_key_once(bool publish)
>>> +void tcp_fastopen_init_key_once(struct net *net)
>> 
>> Why did you remove the 'publish' logic from this function?
>> 
> 
> I think this logic is not necessary now, in proc_tcp_fastopen_key, I have removed 
> tcp_fastopen_init_key_once(false) where the âpublishâ is false:
> 
> -		/* Generate a dummy secret but don't publish it. This
> -		 * is needed so we don't regenerate a new key on the
> -		 * first invocation of tcp_fastopen_cookie_gen
> -		 */
> -		tcp_fastopen_init_key_once(false);
> -		tcp_fastopen_reset_cipher(user_key, TCP_FASTOPEN_KEY_LENGTH);
> +		tcp_fastopen_reset_cipher(net, user_key, TCP_FASTOPEN_KEY_LENGTH);
> 
> It said we don't regenerate a new key on first invocation of tcp_fastopen_cookie_gen, 
> but in tcp_fastopen_cookie_genïit didnât  call tcp_fastopen_init_key_once since
> from commit dfea2aa654243 (tcp: Do not call tcp_fastopen_reset_cipher from interrupt context)ï
> 
> And in other places where call tcp_fastopen_init_key_once, the âpublishâ is always true:

Ok, this simplification seems legitimate.

But it is unrelated to this namespacification.  So it should be in a separate patch,
and should be documented well in the commit message using the great explanation you
gave to me above.

Please respin this series, with this patch #2 split up into two changes.

Thank you.