|
|
|
|
|
|
|
|
|
|
xen-devel
Re: [Xen-devel] Re: [PATCH 2/2] Introduce xen_swiotlb variable that is s
> > This doesn't apply to the xen/dom0/swiotlb-new branch.
> >
>
> OK, its against xen/master.
Do you want me to send patches against the swiotlb-new branch instead in the
future?
>
> >> swiotlb = 1;
> >> #endif
> >> - if (swiotlb_force)
> >> + if (swiotlb_force || !xen_swiotlb)
> >>
> >>
> > Are you sure this is right? This will always enable swiotlb if
> > !xen_swiotlb.
That is wrong. Thanks for spotting that.
> >
>
> I went with this instead:
>
> diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
> index 9640e17..e2d739e 100644
> --- a/arch/x86/kernel/pci-swiotlb.c
> +++ b/arch/x86/kernel/pci-swiotlb.c
> @@ -45,6 +45,9 @@ static struct dma_map_ops swiotlb_dma_ops = {
> void __init pci_swiotlb_init(void)
> {
> /* don't initialize swiotlb if iommu=off (no_iommu=1) */
> + if (xen_swiotlb)
> + return;
> +
> #ifdef CONFIG_X86_64
> if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN)
> swiotlb = 1;
>
> Does that work?
Yes. That will do it.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
|
|
|
|