When we use the generic dma-direct + remap code we also need to
initialize the atomic pool that is used for GFP_ATOMIC allocations on
non-coherent devices.
Fixes: ad3c7b18c5b3 ("arm: use swiotlb for bounce buffering on LPAE configs")
Signed-off-by: Christoph Hellwig <h...@lst.de>
---
arch/arm/mm/dma-mapping.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 6774b03aa405..e509365cc9ca 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -2423,4 +2423,10 @@ void arch_dma_free(struct device *dev, size_t size, void
*cpu_addr,
{
__arm_dma_free(dev, size, cpu_addr, dma_handle, attrs, false);
}
+
+static int __init atomic_pool_init(void)
+{
+ return dma_atomic_pool_init(GFP_DMA, pgprot_noncached(PAGE_KERNEL));
+}
+postcore_initcall(atomic_pool_init);
#endif /* CONFIG_SWIOTLB */
--
2.20.1
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
