Re: [PATCH v13 3/8] iommu/vt-d: Add nested translation helper function

[Jacob Pan]

On Wed, 13 May 2020 22:54:24 -0700
Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote:

> > +	 * 1. CPU vs. IOMMU
> > +	 * 2. Guest vs. Host.
> > +	 */
> > +	switch (addr_width) {
> > +#ifdef CONFIG_X86
> > +	case ADDR_WIDTH_5LEVEL:
> > +		if (cpu_feature_enabled(X86_FEATURE_LA57) &&
> > +		    cap_5lp_support(iommu->cap)) {
> > +			pasid_set_flpm(pte, 1);
> > +		} else {
> > +			dev_err_ratelimited(dev, "5-level paging
> > not supported\n");
> > +			return -EINVAL;
> > +		}
> > +		break;  
> 
> The normal style would be to handle the early error return first:
> 
> 		if (!cpu_feature_enabled(X86_FEATURE_LA57) ||
> 		    !cap_5lp_support(iommu->cap)) {
> 			dev_err_ratelimited(dev,
> 				"5-level paging not supported\n");
> 			return -EINVAL;
> 		}
> 
> 		pasid_set_flpm(pte, 1);
> 		break;
> 

Sounds good. I was thinking in positive logic.

thanks