On Wed, Sep 22, 2021 at 05:14:24PM -0700, Florian Fainelli wrote:
> commit fcf044891c84e38fc90eb736b818781bccf94e38 upstream
>
> We do not need a SWIOTLB unless we have DRAM that is addressable beyond
> the arm_dma_limit. Compare max_pfn with arm_dma_pfn_limit to determine
> whether we do need a SWIOTLB to be initialized.
>
> Fixes: ad3c7b18c5b3 ("arm: use swiotlb for bounce buffering on LPAE configs")
> Signed-off-by: Florian Fainelli <f.fainelli@xxxxxxxxx>
> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
> ---
> arch/arm/mm/init.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
> index d54d69cf1732..75f3ab531bdf 100644
> --- a/arch/arm/mm/init.c
> +++ b/arch/arm/mm/init.c
> @@ -378,7 +378,11 @@ static void __init free_highpages(void)
> void __init mem_init(void)
> {
> #ifdef CONFIG_ARM_LPAE
> - swiotlb_init(1);
> + if (swiotlb_force == SWIOTLB_FORCE ||
> + max_pfn > arm_dma_pfn_limit)
> + swiotlb_init(1);
> + else
> + swiotlb_force = SWIOTLB_NO_FORCE;
> #endif
>
> set_max_mapnr(pfn_to_page(max_pfn) - mem_map);
> --
> 2.25.1
>
Both now queued up, thanks.
greg k-h
