lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [day] [month] [year] [list]
Date:	Wed, 28 Jan 2009 12:44:10 -0800
From:	Andrew Morton <akpm@...ux-foundation.org>
To:	Marcelo Tosatti <mtosatti@...hat.com>
Cc:	agraf@...e.de, Clemens Noss <cnoss@....de>,
	linux-kernel@...r.kernel.org, kvm@...r.kernel.org
Subject: Re: linux-next: Tree for January 23 (kvm)

(cc mailing lists)

On Wed, 28 Jan 2009 12:26:16 -0200
Marcelo Tosatti <mtosatti@...hat.com> wrote:

> On Wed, Jan 28, 2009 at 11:20:16AM +0100, Alexander Graf wrote:
> >
> > On 28.01.2009, at 10:51, Andrew Morton wrote:
> >
> >> On Fri, 23 Jan 2009 12:34:24 -0800 Randy Dunlap 
> >> <randy.dunlap@...cle.com> wrote:
> >>
> >>> Stephen Rothwell wrote:
> >>>> Hi all,
> >>>>
> >>>> News:  I will be on leave next week, so there will probably be no
> >>>> linux-next release until Feb 2.
> >>>>
> >>>> Changes since 20090122:
> >>>
> >>>
> >>> kvm changes cause this build error on i386:
> >>>
> >>> when kvm is built as module:
> >>> ERROR: "__moddi3" [arch/x86/kvm/kvm.ko] undefined!
> >>>
> >>> or when kvm is built into the kernel image:
> >>> lapic.c:(.text+0x19276): undefined reference to `__moddi3'
> >>>
> >>>
> >>> I guess it's this line:
> >>> 	ns = ktime_to_ns(remaining) % apic->timer.period;
> >>>
> >>
> >> dammit, I just spent N minutes hunting down the same thing in
> >> the Jan 26 linux-next :(
> >>
> >
> > Yeah, Clemens Foss posted a real fix on kvm@...r already.

Probably wasn't the best mailing list, as this bug affects all
developers and testers..  But whatever - thanks.

> commit efe4ff38b2bbc3944d0b2c9b6ec669b67cb7acbc
> Author: Clemens Noss <cnoss@....de>
> Date:   Mon Jan 26 02:55:16 2009 +0100
> 
>     KVM: x86: fix "__moddi3 undefined" build failure in lapic.c
>     
>     use mod_64 from arch/x86/kvm/i8254.c to fix
>     ERROR: "__moddi3" [arch/x86/kvm/kvm.ko] undefined!
>     
>     Signed-off-by: Clemens Noss <cnoss@....de>
>     Signed-off-by: Marcelo Tosatti <mtosatti@...hat.com>
> 
> diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
> index d8adc50..f0b67f2 100644
> --- a/arch/x86/kvm/lapic.c
> +++ b/arch/x86/kvm/lapic.c
> @@ -35,6 +35,12 @@
>  #include "kvm_cache_regs.h"
>  #include "irq.h"
>  
> +#ifndef CONFIG_X86_64
> +#define mod_64(x, y) ((x) - (y) * div64_u64(x, y))
> +#else
> +#define mod_64(x, y) ((x) % (y))
> +#endif
> +
>  #define PRId64 "d"
>  #define PRIx64 "llx"
>  #define PRIu64 "u"
> @@ -525,7 +531,7 @@ static u32 apic_get_tmcct(struct kvm_lapic *apic)
>  	if (ktime_to_ns(remaining) < 0)
>  		remaining = ktime_set(0, 0);
>  
> -	ns = ktime_to_ns(remaining) % apic->timer.period;
> +	ns = mod_64(ktime_to_ns(remaining), apic->timer.period);
>  	tmcct = div64_u64(ns, (APIC_BUS_CYCLE_NS * apic->timer.divide_count));
>  
>  	return tmcct;

Rather than cooking up a private implementation I think it would be
better to implement a kernel-wide mod64_u64() (?) facility and then to
use that in KVM.  There will presumably be other sites which will (or
already do) need such a thing.

However it'll probably be a bit hard, getting the signednesses of the
args and the return value right.

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

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ