amd_iommu_init.c: possible problem with IVHD_FLAG_HT_TUN_EN

amd_iommu_init.c: possible problem with IVHD_FLAG_HT_TUN_EN

[Julia Lawall]

The file arch/x86/kernel/amd_iommu_init.c contains the following code:

(starting at line 586 in a recent linux-next)

	h->flags & IVHD_FLAG_HT_TUN_EN ?
		iommu_feature_enable(iommu, CONTROL_HT_TUN_EN) :
		iommu_feature_disable(iommu, CONTROL_HT_TUN_EN);

	h->flags & IVHD_FLAG_PASSPW_EN ?
		iommu_feature_enable(iommu, CONTROL_PASSPW_EN) :
		iommu_feature_disable(iommu, CONTROL_PASSPW_EN);

	h->flags & IVHD_FLAG_RESPASSPW_EN ?
		iommu_feature_enable(iommu, CONTROL_RESPASSPW_EN) :
		iommu_feature_disable(iommu, CONTROL_RESPASSPW_EN);

	h->flags & IVHD_FLAG_ISOC_EN ?
		iommu_feature_enable(iommu, CONTROL_ISOC_EN) :
		iommu_feature_disable(iommu, CONTROL_ISOC_EN);

The only possible value of IVHD_FLAG_HT_TUN_EN, however, seems to be 0, as
defined in the same file, and thus the first test is never true.  Is this
what is intended, or should the test be expressed in another way?

julia

This problem was found using the following semantic match:
(http://www.emn.fr/x-info/coccinelle/)

@r expression@
identifier C;
expression E;
position p;
@@

(
 E & C@p && ...
|
 E & C@p || ...
)

@s@
identifier r.C;
position p1;
@@

#define C 0

@t@
identifier r.C;
expression E != 0;
@@

#define C E

@script:python depends on s && !t@
p << r.p;
C << r.C;
@@

cocci.print_main("and with 0", p)

Re: amd_iommu_init.c: possible problem with IVHD_FLAG_HT_TUN_EN

[Joerg Roedel]

On Sat, May 02, 2009 at 03:20:23PM +0200, Julia Lawall wrote:
> The file arch/x86/kernel/amd_iommu_init.c contains the following code:
> 
> (starting at line 586 in a recent linux-next)
> 
> 	h->flags & IVHD_FLAG_HT_TUN_EN ?
> 		iommu_feature_enable(iommu, CONTROL_HT_TUN_EN) :
> 		iommu_feature_disable(iommu, CONTROL_HT_TUN_EN);
> 
> 	h->flags & IVHD_FLAG_PASSPW_EN ?
> 		iommu_feature_enable(iommu, CONTROL_PASSPW_EN) :
> 		iommu_feature_disable(iommu, CONTROL_PASSPW_EN);
> 
> 	h->flags & IVHD_FLAG_RESPASSPW_EN ?
> 		iommu_feature_enable(iommu, CONTROL_RESPASSPW_EN) :
> 		iommu_feature_disable(iommu, CONTROL_RESPASSPW_EN);
> 
> 	h->flags & IVHD_FLAG_ISOC_EN ?
> 		iommu_feature_enable(iommu, CONTROL_ISOC_EN) :
> 		iommu_feature_disable(iommu, CONTROL_ISOC_EN);
> 
> The only possible value of IVHD_FLAG_HT_TUN_EN, however, seems to be 0, as
> defined in the same file, and thus the first test is never true.  Is this
> what is intended, or should the test be expressed in another way?

Very true. Another bug because I mixed up bit shifts with bit masks :-(
Thanks for reporting this. I will send a patch to fix it.

Joerg

-- 
           | Advanced Micro Devices GmbH
 Operating | Karl-Hammerschmidt-Str. 34, 85609 Dornach bei München
 System    | 
 Research  | Geschäftsführer: Jochen Polster, Thomas M. McCoy, Giuliano Meroni
 Center    | Sitz: Dornach, Gemeinde Aschheim, Landkreis München
           | Registergericht München, HRB Nr. 43632

--
To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html