Re: [PATCH] net: tipc: Convert timers to use timer_setup()

[Kees Cook]

On Mon, Oct 30, 2017 at 2:57 AM, Jon Maloy <jon.maloy@xxxxxxxxxxxx> wrote:
>
>
>> -----Original Message-----
>> From: keescook@xxxxxxxxxx [mailto:keescook@xxxxxxxxxx] On Behalf Of
>> Kees Cook
>> Sent: Friday, October 27, 2017 06:58
>> To: Jon Maloy <jon.maloy@xxxxxxxxxxxx>
>> Cc: David S. Miller <davem@xxxxxxxxxxxxx>; Ying Xue
>> <ying.xue@xxxxxxxxxxxxx>; netdev@xxxxxxxxxxxxxxx; tipc-
>> discussion@xxxxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
>> Subject: Re: [PATCH] net: tipc: Convert timers to use timer_setup()
>>
>> On Tue, Oct 24, 2017 at 11:44 AM, Jon Maloy <jon.maloy@xxxxxxxxxxxx>
>> wrote:
>> > NAK. It doesn't sound like a good idea to send this to net. Especially since
>> one of these timers has already been refactored in net-next.
>>
>> Hi! I'm not sure what you mean about the one timer issue. I don't see any
>> use of timer_setup() in net/tipc (and no recent conversions to the older
>> setup_timer() API). What's the preferred path for landing this API conversion
>> in net/tipc/?
>>
>> And, just to note, these changes are almost entirely mechanical. The only
>> "special" case is in tipc_sk_timeout() where the argument needs to be
>> slightly adjusted to fetch the tsk from the sk again.
>
> I am referring to commit 0d5fcebf3c37 ("tipc: refactor tipc_sk_timeout() function") which is in "net-next" but not yet in "net".
> I guess David M is the one who decides whether you deliver to net or net-next, but in the former case you will have to wait until David merges in the above commit, and then add your change again.

Ah, thanks, I see the change now. These patches passed in the night,
it seems. I've sent a v2 now for net-next.

Thanks!

-Kees

-- 
Kees Cook
Pixel Security