Netdev List
 help / color / mirror / Atom feed
* Netlink send from the interrupt context
@ 2005-11-21 11:16 Robert Iakobashvili
  0 siblings, 0 replies; 3+ messages in thread
From: Robert Iakobashvili @ 2005-11-21 11:16 UTC (permalink / raw)
  To: netdev

Are there netlink socket netlink_unicast () and netlink_broadcast ()
interrupt safe?

If not, where is the problem and the direction to make them safe?
If it is not easy, what could be a workaroud?

Thank you in advance.
----------------------------------------------------------------
Robert Iakobashvili  - NAVIGARE NECESSE EST
coroberti at gmail dot com
---------------------------------------------------------------

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Netlink send from the interrupt context
       [not found]           ` <7e63f56c0511230032x5097569dr3ddac4b05bb6019c@mail.gmail.com>
@ 2005-11-23  8:33             ` Robert Iakobashvili
  2005-11-23  9:04               ` Yu Zhiguo
  0 siblings, 1 reply; 3+ messages in thread
From: Robert Iakobashvili @ 2005-11-23  8:33 UTC (permalink / raw)
  To: Yu Zhiguo; +Cc: netdev, netfilter-devel

On 11/23/05, Robert Iakobashvili <coroberti@gmail.com> wrote:
> On 11/23/05, Yu Zhiguo <yuzg@nanjing-fnst.com> wrote:
> > Please refer to the function netlink_broadcast(),
> >    int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 pid, u32 group, int allocation)
> > I think if allocation=GFP_ATOMIC, it can be used in interrupt context.
> > What's your opinion?

Allocation is OK and it uses only spinlocks in all functions called,
no semaphores.

My concern actually rises from here:
http://www.ussg.iu.edu/hypermail/linux/kernel/0407.3/0455.html

Sincerely,
----------------------------------------------------------------
Robert Iakobashvili  - NAVIGARE NECESSE EST
coroberti at gmail dot com
---------------------------------------------------------------

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Netlink send from the interrupt context
  2005-11-23  8:33             ` Netlink send from the interrupt context Robert Iakobashvili
@ 2005-11-23  9:04               ` Yu Zhiguo
  0 siblings, 0 replies; 3+ messages in thread
From: Yu Zhiguo @ 2005-11-23  9:04 UTC (permalink / raw)
  To: Robert Iakobashvili; +Cc: netdev, netfilter-devel

yes, it use spinlocks rather than semaphore, and it can use GFP_ATOMIC to clone the message, 
so I think it will not go sleep.

regards,
----- Original Message ----- 

> On 11/23/05, Robert Iakobashvili <coroberti@gmail.com> wrote:
>> On 11/23/05, Yu Zhiguo <yuzg@nanjing-fnst.com> wrote:
>> > Please refer to the function netlink_broadcast(),
>> >    int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 pid, u32 group, int allocation)
>> > I think if allocation=GFP_ATOMIC, it can be used in interrupt context.
>> > What's your opinion?
> 
> Allocation is OK and it uses only spinlocks in all functions called,
> no semaphores.
> 
> My concern actually rises from here:
> http://www.ussg.iu.edu/hypermail/linux/kernel/0407.3/0455.html
> 
> Sincerely,
> ----------------------------------------------------------------
> Robert Iakobashvili  - NAVIGARE NECESSE EST
> coroberti at gmail dot com
> ---------------------------------------------------------------

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2005-11-23  9:04 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <7e63f56c0511210034h49917721k74ddb251b9bb1815@mail.gmail.com>
     [not found] ` <00a901c5ee79$389e5040$6704a8c0@skqianyi>
     [not found]   ` <7e63f56c0511222328w6368f7bcl7877fe3aa62c2807@mail.gmail.com>
     [not found]     ` <02e501c5f002$e5094d00$6704a8c0@skqianyi>
     [not found]       ` <7e63f56c0511230004j79c7f080j56e25251387ebe94@mail.gmail.com>
     [not found]         ` <031201c5f007$0ce2e1c0$6704a8c0@skqianyi>
     [not found]           ` <7e63f56c0511230032x5097569dr3ddac4b05bb6019c@mail.gmail.com>
2005-11-23  8:33             ` Netlink send from the interrupt context Robert Iakobashvili
2005-11-23  9:04               ` Yu Zhiguo
2005-11-21 11:16 Robert Iakobashvili

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox