On Wed, Mar 14, 2018 at 06:52:04PM +0100, Christoph Hellwig wrote: > We want to phase out looking at the magic GFP_DMA flag in the dma mapping > routines, so switch the gart driver to use the coherent_dma_mask instead, > which is used to select the GFP_DMA flag in the caller. > Reviewed-by: Konrad Rzeszutek Wilk <konrad.w...@oracle.com>
> Signed-off-by: Christoph Hellwig <h...@lst.de> > --- > arch/x86/kernel/amd_gart_64.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c > index 52e3abcf3e70..79ac6caaaabb 100644 > --- a/arch/x86/kernel/amd_gart_64.c > +++ b/arch/x86/kernel/amd_gart_64.c > @@ -484,7 +484,7 @@ gart_alloc_coherent(struct device *dev, size_t size, > dma_addr_t *dma_addr, > unsigned long align_mask; > struct page *page; > > - if (force_iommu && !(flag & GFP_DMA)) { > + if (force_iommu && dev->coherent_dma_mask > DMA_BIT_MASK(24)) { > flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); > page = alloc_pages(flag | __GFP_ZERO, get_order(size)); > if (!page) > -- > 2.14.2 >