[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Fwd: Sleeping after preempt_disable() possible?

Hi,

I didn't know the likely and unlikely macros, so I searched in
http://lxr.linux.no and I'm a bit confussed about how they work...

(in particular I don't know what the bang bang does...I tried with
google, but I couldn't find information... I made a search into the
gcc manual nothing about !! was found...)

#define likely(x) __builtin_expect(!!(x), 1)

and

#define __builtin_expect(x, expected_value) (x)

The exit_state field of task_struct is defined as:

long exit_state;

Can somebody give me a little explanation?

Thanks in advance


---------- Forwarded message ----------
From: Rik van Riel <riel@xxxxxxxxxxx>
Date: Sep 2, 2006 1:56 AM
Subject: Re: Sleeping after preempt_disable() possible?
To: Rick Brown <rick.brown.3@xxxxxxxxx>
Cc: "bora.sahin@xxxxxxxxxxxx" <bora.sahin@xxxxxxxxxxxx>,
kernelnewbies@xxxxxxxxxxxx, linux-newbie@xxxxxxxxxxxxxxx

Rick Brown wrote:

I was curious that after a call to preempt_disable(), can I call a
function that MAY sleep?

Bad idea. That can make the kernel do bad things...

Just take a look at schedule() in kernel/sched.c:

asmlinkage void __sched schedule(void)
{
...
       if (unlikely(in_atomic() && !current->exit_state)) {
                printk(KERN_ERR "BUG: scheduling while atomic: "
                        "%s/0x%08x/%d\n",
                        current->comm, preempt_count(), current->pid);
                dump_stack();
        }

Of course, in_atomic() checks whether you disabled preemption.

--
What is important?  What you want to be true, or what is true?

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/