Re: [PATCH 1/8] Notifier chain update

[Jes Sorensen]

>>>>> "Benjamin" == Benjamin LaHaise <bcrl@xxxxxxxxx> writes:

Benjamin> On Wed, Jan 18, 2006 at 03:23:20PM -0500, Alan Stern wrote:
>> You can't use RCU protection around code that may sleep.  Whether
>> the code remains loaded in the kernel or is part of a removable
>> module doesn't enter into it.

Benjamin> A notifier callee should not be sleeping, if anything it
Benjamin> should be putting its work onto a workqueue and completing
Benjamin> it when it gets scheduled if it has to do something that
Benjamin> blocks.

There are cases where all the notifier client would want to do is
kmalloc(); sum_up_some_data(); copy_to_user(); kfree(); 
very often in the exit() path - doing that via a workqueue seems
overly complex.

Cheers,
Jes
-
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/