[PATCH v7 01/19] iommu/dma: Implement PCI allocation optimisation

[Robin Murphy]

On 09/01/17 13:45, Eric Auger wrote:
> From: Robin Murphy <robin.murphy at arm.com>
> 
> Whilst PCI devices may have 64-bit DMA masks, they still benefit from
> using 32-bit addresses wherever possible in order to avoid DAC (PCI) or
> longer address packets (PCIe), which may incur a performance overhead.
> Implement the same optimisation as other allocators by trying to get a
> 32-bit address first, only falling back to the full mask if that fails.

Oops, this was just some development work which got promoted to my misc
branch for safe keeping; it really has nothing to do with this series.

I'd managed to overlook that one of the __alloc_iova() conflicts hit the
MSI cookie patch, sorry. Things are now in a more appropriate order in
my tree.

Robin.

> Signed-off-by: Robin Murphy <robin.murphy at arm.com>
> ---
>  drivers/iommu/dma-iommu.c | 21 +++++++++++++++------
>  1 file changed, 15 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
> index 2db0d64..a59ca47 100644
> --- a/drivers/iommu/dma-iommu.c
> +++ b/drivers/iommu/dma-iommu.c
> @@ -204,19 +204,28 @@ int dma_direction_to_prot(enum dma_data_direction dir, bool coherent)
>  }
>  
>  static struct iova *__alloc_iova(struct iommu_domain *domain, size_t size,
> -		dma_addr_t dma_limit)
> +		dma_addr_t dma_limit, struct device *dev)
>  {
>  	struct iova_domain *iovad = cookie_iovad(domain);
>  	unsigned long shift = iova_shift(iovad);
>  	unsigned long length = iova_align(iovad, size) >> shift;
> +	struct iova *iova = NULL;
>  
>  	if (domain->geometry.force_aperture)
>  		dma_limit = min(dma_limit, domain->geometry.aperture_end);
> +
> +	/* Try to get PCI devices a SAC address */
> +	if (dma_limit > DMA_BIT_MASK(32) && dev_is_pci(dev))
> +		iova = alloc_iova(iovad, length, DMA_BIT_MASK(32) >> shift,
> +				  true);
>  	/*
>  	 * Enforce size-alignment to be safe - there could perhaps be an
>  	 * attribute to control this per-device, or at least per-domain...
>  	 */
> -	return alloc_iova(iovad, length, dma_limit >> shift, true);
> +	if (!iova)
> +		iova = alloc_iova(iovad, length, dma_limit >> shift, true);
> +
> +	return iova;
>  }
>  
>  /* The IOVA allocator knows what we mapped, so just unmap whatever that was */
> @@ -369,7 +378,7 @@ struct page **iommu_dma_alloc(struct device *dev, size_t size, gfp_t gfp,
>  	if (!pages)
>  		return NULL;
>  
> -	iova = __alloc_iova(domain, size, dev->coherent_dma_mask);
> +	iova = __alloc_iova(domain, size, dev->coherent_dma_mask, dev);
>  	if (!iova)
>  		goto out_free_pages;
>  
> @@ -440,7 +449,7 @@ static dma_addr_t __iommu_dma_map(struct device *dev, phys_addr_t phys,
>  	struct iova_domain *iovad = cookie_iovad(domain);
>  	size_t iova_off = iova_offset(iovad, phys);
>  	size_t len = iova_align(iovad, size + iova_off);
> -	struct iova *iova = __alloc_iova(domain, len, dma_get_mask(dev));
> +	struct iova *iova = __alloc_iova(domain, len, dma_get_mask(dev), dev);
>  
>  	if (!iova)
>  		return DMA_ERROR_CODE;
> @@ -598,7 +607,7 @@ int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
>  		prev = s;
>  	}
>  
> -	iova = __alloc_iova(domain, iova_len, dma_get_mask(dev));
> +	iova = __alloc_iova(domain, iova_len, dma_get_mask(dev), dev);
>  	if (!iova)
>  		goto out_restore_sg;
>  
> @@ -675,7 +684,7 @@ static struct iommu_dma_msi_page *iommu_dma_get_msi_page(struct device *dev,
>  	if (!msi_page)
>  		return NULL;
>  
> -	iova = __alloc_iova(domain, iovad->granule, dma_get_mask(dev));
> +	iova = __alloc_iova(domain, iovad->granule, dma_get_mask(dev), dev);
>  	if (!iova)
>  		goto out_free_page;
>  
>