Re: [PATCH] sparc: iommu/pci_sun4v: fix page alignment in dma mapping

[John Paul Adrian Glaubitz]

Hi Stian,

please rephrase the subject to:

	sparc: Fix page alignment in dma mapping

Your patch is not specific to sun4v but also fixes generic sparc code,
so you should remove the part "iommu/pci_sun4v:" from the subject.

On Tue, 2026-02-17 at 17:21 +0100, stian@xxxxxx wrote:
> From: Stian Halseth <stian@xxxxxx>
> 
> 'phys' may include an offset within the page, while 'base_paddr' is already page-aligned.
> 
> This caused incorrect DMA mapping in dma_4u_map_phys and dma_4v_map_phys.
> 
> Fix both functions by masking phys with IO_PAGE_MASK or subtracting the page offset.

Please remove the newlines between your sentences, they're not necessary
and also trim the lines at 80 columns.

> Fixes: 38c0d0ebf520 ("sparc: Use physical address DMA mapping")
> 
> Reported-by: Stian Halseth <stian@xxxxxx>
> Suggested-by: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx>
> Signed-off-by: Stian Halseth <stian@xxxxxx>
> ---
>  arch/sparc/kernel/iommu.c     | 2 ++
>  arch/sparc/kernel/pci_sun4v.c | 2 +-
>  2 files changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/sparc/kernel/iommu.c b/arch/sparc/kernel/iommu.c
> index 46ef88bc9c26..f3755a388ac7 100644
> --- a/arch/sparc/kernel/iommu.c
> +++ b/arch/sparc/kernel/iommu.c
> @@ -312,6 +312,8 @@ static dma_addr_t dma_4u_map_phys(struct device *dev, phys_addr_t phys,
>  	if (direction != DMA_TO_DEVICE)
>  		iopte_protection |= IOPTE_WRITE;
>  
> +	phys -= offset_in_page(phys);
> +
>  	for (i = 0; i < npages; i++, base++, phys += IO_PAGE_SIZE)
>  		iopte_val(*base) = iopte_protection | phys;
>  
> diff --git a/arch/sparc/kernel/pci_sun4v.c b/arch/sparc/kernel/pci_sun4v.c
> index 791f0a76665f..2f30eeac4861 100644
> --- a/arch/sparc/kernel/pci_sun4v.c
> +++ b/arch/sparc/kernel/pci_sun4v.c
> @@ -411,7 +411,7 @@ static dma_addr_t dma_4v_map_phys(struct device *dev, phys_addr_t phys,
>  	iommu_batch_start(dev, prot, entry);
>  
>  	for (i = 0; i < npages; i++, phys += IO_PAGE_SIZE) {
> -		long err = iommu_batch_add(phys, mask);
> +		long err = iommu_batch_add(phys & IO_PAGE_MASK, mask);
>  		if (unlikely(err < 0L))
>  			goto iommu_map_fail;
>  	}

Looks good to me otherwise.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913