Re: kmalloc without GFP_xxx?

[Steven Rostedt <rostedt@goodmis.org>]

On Wed, 29 Jun 2005, Manfred Spraul wrote:

> Hi,
>
> One question from Linux-Tag was about the lack of documentation about/in
> the kernel. I try to maintain docbook entries when I modify code, even
> though I think it's mostly wasted time: Virtually noone reads it anyway,
> instead armchair logic on lkml.

Hmm, I do like to read the comments, see below.

>
> Steven wrote:
>
> >Here we see that task 2 can spin with interrupts off, while the first task
> >is servicing an interrupt, and God forbid if the IRQ handler sends some
> >kind of SMP signal to the CPU running task 2 since that would be a
> >deadlock.  Granted, this is a hypothetical situation, but makes using
> >spin_lock with interrupts enabled a little scary.
> >
> >
> Not, it's not even a hypothetical situation. It's an explicitely
> forbidden situation: SMP signals are sent with smp_call_function and the
> documentation to that function clearly says:

When I said _hypothetical_ I ment it.  That's basically stating that the
situation wont happen, but lets pretend that it will. And no, SMP signals
(on intel anyway) are sent with send_IPI_* which even smp_call_function
uses.

>  *
>  * You must not call this function with disabled interrupts or from a
>  * hardware interrupt handler or from a bottom half handler.
>  */
>

And if you had read my other emails you would have noticed that I
even mentioned this particular comment. When I said:

"This is probably the reason it is not allowed to call most IPIs from
interrupt or bottom half context."

Also, a comment doesn't force this, and there's no test in
smp_call_function that prevents a user from calling this form a
bottom_half!

-- Steve

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