Re: [PATCH -tip] x86: uv - prevent NULL dereference in uv_system_init

[Cyrill Gorcunov <gorcunov@gmail.com>]

[Ingo Molnar - Fri, May 01, 2009 at 10:03:31PM +0200]
| 
| * Cyrill Gorcunov <gorcunov@openvz.org> wrote:
| 
| > We may reach NULL dereference oops if kmalloc failed.
| > Lets do panic better with sensible message.
| > 
| > Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
| > ---
| > 
| > Actually there is a dubious place as well at early_get_nodeid.
| > Is there a guarantee that we _never_ fail in early_ioremap?
| > 
| >  arch/x86/kernel/apic/x2apic_uv_x.c |    9 +++++++++
| >  1 file changed, 9 insertions(+)
| > 
| > Index: linux-2.6.git/arch/x86/kernel/apic/x2apic_uv_x.c
| > =====================================================================
| > --- linux-2.6.git.orig/arch/x86/kernel/apic/x2apic_uv_x.c
| > +++ linux-2.6.git/arch/x86/kernel/apic/x2apic_uv_x.c
| > @@ -584,15 +584,21 @@ void __init uv_system_init(void)
| >  
| >  	bytes = sizeof(struct uv_blade_info) * uv_num_possible_blades();
| >  	uv_blade_info = kmalloc(bytes, GFP_KERNEL);
| > +	if (!uv_blade_info)
| > +		goto err_nomem;
| 
| hm, i think a BUG_ON() might be shorter and more appropriate here. 
| We really shouldnt be running out of memory during system init.
| 
| 	Ingo
| 

Yeah, indeed! I was thinking of __GPF_NOFAIL here as well with
message on top like pr_debug("UV: allocating memory\n") or something
like that. It would make it even cleaner I guess. Hmm?

	-- Cyrill