Re: [PATCH 1/8] Notifier chain update

On Wed, 18 Jan 2006, Benjamin LaHaise wrote:

> The notifier interface is supposed to be *light weight*.

Again, where is that documented?

> Adding locks
> that get taken on every call basically changes the concept entirely. The
> cache misses notifiers add are measurable overhead, with locks being far
> worse.

Which is worse: overhead due to cache misses or an oops caused by code
being called after it was unloaded?

Do you have a better proposal for a way to prevent blocking notifier
chains from being modified while in use? Or would you prefer to rewrite
all the callout routines that currently block, so that all the notifier
chains can be made atomic and we don't need the blocking notifier API?

Alan Stern

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages

  • Re: [Lse-tech] Re: [PATCH 0/7]: Fix for unsafe notifier chain
    ... >> Unregister is not the only problem. ... notifier chain, it has only one walking_handler_list array. ... can be traversed without locks. ... gives us lockfree traversal of notifier chains and supports multiple ...
    (Linux-Kernel)
  • Re: Notifier chains are unsafe
    ... > includes oops, panic, NMI and other unmaskable machine check events. ... > it follows that the callbacks cannot take any locks. ... own call-out functions. ... I am inclined to think that we have to split the notifier interface ...
    (Linux-Kernel)
  • Re: [Lse-tech] Re: [ckrm-tech] Re: [PATCH 00/01] Move Exit Connectors
    ... >>One unique part of the task notify is that it doesn't use locks. ... > Neither does Alan Stern's atomic notifier chain. ... > locks, because the atomic notifier chains can be called from anywhere, ...
    (Linux-Kernel)
  • Re: [PATCH 1/8] Notifier chain update
    ... > Read the kernel. ... In the patch we classified chains as blocking or atomic based on how they ... provision specifically for lightweight notifiers: ... > Notifiers should not have locks. ...
    (Linux-Kernel)
  • Re: [Lse-tech] Re: [ckrm-tech] Re: [PATCH 00/01] Move Exit Connectors
    ... >One unique part of the task notify is that it doesn't use locks. ... Neither does Alan Stern's atomic notifier chain. ... locks, because the atomic notifier chains can be called from anywhere, ...
    (Linux-Kernel)