2007-11-08 19:56:27

by Dave Hansen

[permalink] [raw]
Subject: 2.6.23 git current compile error on UP

drivers/kvm/kvm_main.c: In function `kvm_flush_remote_tlbs':
drivers/kvm/kvm_main.c:220: error: implicit declaration of function `smp_call_function_mask'
make[2]: *** [drivers/kvm/kvm_main.o] Error 1
make[1]: *** [drivers/kvm] Error 2

http://sr71.net/~dave/linux/config-kvm-up

Looks like that function calls smp_call_function_mask() which is never
defined for UP. Nobody else uses it that way, so I'm not sure what the
right fix is. I'm not even sure kvm_flush_remote_tlbs() is safe with
its raw_smp_processor_id() use. Is there a reason it can't get
preempted?

-- Dave


2007-11-08 20:09:50

by Laurent Vivier

[permalink] [raw]
Subject: Re: 2.6.23 git current compile error on UP

Le jeudi 08 novembre 2007 à 11:55 -0800, Dave Hansen a écrit :
> drivers/kvm/kvm_main.c: In function `kvm_flush_remote_tlbs':
> drivers/kvm/kvm_main.c:220: error: implicit declaration of function `smp_call_function_mask'
> make[2]: *** [drivers/kvm/kvm_main.o] Error 1
> make[1]: *** [drivers/kvm] Error 2
>
> http://sr71.net/~dave/linux/config-kvm-up
>
> Looks like that function calls smp_call_function_mask() which is never
> defined for UP. Nobody else uses it that way, so I'm not sure what the
> right fix is. I'm not even sure kvm_flush_remote_tlbs() is safe with
> its raw_smp_processor_id() use. Is there a reason it can't get
> preempted?

Avi has already posted a patch to correct this.

http://lkml.org/lkml/2007/10/24/95

Laurent


Attachments:
signature.asc (189.00 B)
Ceci est une partie de message num?riquement sign