Merge branch 'x86-dma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 dma mapping updates from Ingo Molnar:
 "This tree, by Christoph Hellwig, switches over the x86 architecture to
  the generic dma-direct and swiotlb code, and also unifies more of the
  dma-direct code between architectures. The now unused x86-only
  primitives are removed"

* 'x86-dma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  dma-mapping: Don't clear GFP_ZERO in dma_alloc_attrs
  swiotlb: Make swiotlb_{alloc,free}_buffer depend on CONFIG_DMA_DIRECT_OPS
  dma/swiotlb: Remove swiotlb_{alloc,free}_coherent()
  dma/direct: Handle force decryption for DMA coherent buffers in common code
  dma/direct: Handle the memory encryption bit in common code
  dma/swiotlb: Remove swiotlb_set_mem_attributes()
  set_memory.h: Provide set_memory_{en,de}crypted() stubs
  x86/dma: Remove dma_alloc_coherent_gfp_flags()
  iommu/intel-iommu: Enable CONFIG_DMA_DIRECT_OPS=y and clean up intel_{alloc,free}_coherent()
  iommu/amd_iommu: Use CONFIG_DMA_DIRECT_OPS=y and dma_direct_{alloc,free}()
  x86/dma/amd_gart: Use dma_direct_{alloc,free}()
  x86/dma/amd_gart: Look at dev->coherent_dma_mask instead of GFP_DMA
  x86/dma: Use generic swiotlb_ops
  x86/dma: Use DMA-direct (CONFIG_DMA_DIRECT_OPS=y)
  x86/dma: Remove dma_alloc_coherent_mask()

2 files changed, 46 insertions, 66 deletions

diff --git a/lib/dma-direct.c b/lib/dma-direct.c
index c9e8e21cb334..c0bba30fef0a 100644
--- a/lib/dma-direct.c
+++ b/lib/dma-direct.c
@@ -9,6 +9,7 @@
 #include <linux/scatterlist.h>
 #include <linux/dma-contiguous.h>
 #include <linux/pfn.h>
+#include <linux/set_memory.h>
 
 #define DIRECT_MAPPING_ERROR		0
 
@@ -20,6 +21,14 @@
 #define ARCH_ZONE_DMA_BITS 24
 #endif
 
+/*
+ * For AMD SEV all DMA must be to unencrypted addresses.
+ */
+static inline bool force_dma_unencrypted(void)
+{
+	return sev_active();
+}
+
 static bool
 check_addr(struct device *dev, dma_addr_t dma_addr, size_t size,
 		const char *caller)
@@ -37,7 +46,9 @@ check_addr(struct device *dev, dma_addr_t dma_addr, size_t size,
 
 static bool dma_coherent_ok(struct device *dev, phys_addr_t phys, size_t size)
 {
-	return phys_to_dma(dev, phys) + size - 1 <= dev->coherent_dma_mask;
+	dma_addr_t addr = force_dma_unencrypted() ?
+		__phys_to_dma(dev, phys) : phys_to_dma(dev, phys);
+	return addr + size - 1 <= dev->coherent_dma_mask;
 }
 
 void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
@@ -46,6 +57,10 @@ void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
 	int page_order = get_order(size);
 	struct page *page = NULL;
+	void *ret;
+
+	/* we always manually zero the memory once we are done: */
+	gfp &= ~__GFP_ZERO;
 
 	/* GFP_DMA32 and GFP_DMA are no ops without the corresponding zones: */
 	if (dev->coherent_dma_mask <= DMA_BIT_MASK(ARCH_ZONE_DMA_BITS))
@@ -78,10 +93,15 @@ again:
 
 	if (!page)
 		return NULL;
-
-	*dma_handle = phys_to_dma(dev, page_to_phys(page));
-	memset(page_address(page), 0, size);
-	return page_address(page);
+	ret = page_address(page);
+	if (force_dma_unencrypted()) {
+		set_memory_decrypted((unsigned long)ret, 1 << page_order);
+		*dma_handle = __phys_to_dma(dev, page_to_phys(page));
+	} else {
+		*dma_handle = phys_to_dma(dev, page_to_phys(page));
+	}
+	memset(ret, 0, size);
+	return ret;
 }
 
 /*
@@ -92,9 +112,12 @@ void dma_direct_free(struct device *dev, size_t size, void *cpu_addr,
 		dma_addr_t dma_addr, unsigned long attrs)
 {
 	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	unsigned int page_order = get_order(size);
 
+	if (force_dma_unencrypted())
+		set_memory_encrypted((unsigned long)cpu_addr, 1 << page_order);
 	if (!dma_release_from_contiguous(dev, virt_to_page(cpu_addr), count))
-		free_pages((unsigned long)cpu_addr, get_order(size));
+		free_pages((unsigned long)cpu_addr, page_order);
 }
 
 static dma_addr_t dma_direct_map_page(struct device *dev, struct page *page,
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index c43ec2271469..47aeb04c1997 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -31,6 +31,7 @@
 #include <linux/gfp.h>
 #include <linux/scatterlist.h>
 #include <linux/mem_encrypt.h>
+#include <linux/set_memory.h>
 
 #include <asm/io.h>
 #include <asm/dma.h>
@@ -156,22 +157,6 @@ unsigned long swiotlb_size_or_default(void)
 	return size ? size : (IO_TLB_DEFAULT_SIZE);
 }
 
-void __weak swiotlb_set_mem_attributes(void *vaddr, unsigned long size) { }
-
-/* For swiotlb, clear memory encryption mask from dma addresses */
-static dma_addr_t swiotlb_phys_to_dma(struct device *hwdev,
-				      phys_addr_t address)
-{
-	return __sme_clr(phys_to_dma(hwdev, address));
-}
-
-/* Note that this doesn't work with highmem page */
-static dma_addr_t swiotlb_virt_to_bus(struct device *hwdev,
-				      volatile void *address)
-{
-	return phys_to_dma(hwdev, virt_to_phys(address));
-}
-
 static bool no_iotlb_memory;
 
 void swiotlb_print_info(void)
@@ -209,12 +194,12 @@ void __init swiotlb_update_mem_attributes(void)
 
 	vaddr = phys_to_virt(io_tlb_start);
 	bytes = PAGE_ALIGN(io_tlb_nslabs << IO_TLB_SHIFT);
-	swiotlb_set_mem_attributes(vaddr, bytes);
+	set_memory_decrypted((unsigned long)vaddr, bytes >> PAGE_SHIFT);
 	memset(vaddr, 0, bytes);
 
 	vaddr = phys_to_virt(io_tlb_overflow_buffer);
 	bytes = PAGE_ALIGN(io_tlb_overflow);
-	swiotlb_set_mem_attributes(vaddr, bytes);
+	set_memory_decrypted((unsigned long)vaddr, bytes >> PAGE_SHIFT);
 	memset(vaddr, 0, bytes);
 }
 
@@ -355,7 +340,7 @@ swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs)
 	io_tlb_start = virt_to_phys(tlb);
 	io_tlb_end = io_tlb_start + bytes;
 
-	swiotlb_set_mem_attributes(tlb, bytes);
+	set_memory_decrypted((unsigned long)tlb, bytes >> PAGE_SHIFT);
 	memset(tlb, 0, bytes);
 
 	/*
@@ -366,7 +351,8 @@ swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs)
 	if (!v_overflow_buffer)
 		goto cleanup2;
 
-	swiotlb_set_mem_attributes(v_overflow_buffer, io_tlb_overflow);
+	set_memory_decrypted((unsigned long)v_overflow_buffer,
+			io_tlb_overflow >> PAGE_SHIFT);
 	memset(v_overflow_buffer, 0, io_tlb_overflow);
 	io_tlb_overflow_buffer = virt_to_phys(v_overflow_buffer);
 
@@ -622,7 +608,7 @@ map_single(struct device *hwdev, phys_addr_t phys, size_t size,
 		return SWIOTLB_MAP_ERROR;
 	}
 
-	start_dma_addr = swiotlb_phys_to_dma(hwdev, io_tlb_start);
+	start_dma_addr = __phys_to_dma(hwdev, io_tlb_start);
 	return swiotlb_tbl_map_single(hwdev, start_dma_addr, phys, size,
 				      dir, attrs);
 }
@@ -706,6 +692,7 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
 	}
 }
 
+#ifdef CONFIG_DMA_DIRECT_OPS
 static inline bool dma_coherent_ok(struct device *dev, dma_addr_t addr,
 		size_t size)
 {
@@ -726,12 +713,12 @@ swiotlb_alloc_buffer(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		goto out_warn;
 
 	phys_addr = swiotlb_tbl_map_single(dev,
-			swiotlb_phys_to_dma(dev, io_tlb_start),
+			__phys_to_dma(dev, io_tlb_start),
 			0, size, DMA_FROM_DEVICE, 0);
 	if (phys_addr == SWIOTLB_MAP_ERROR)
 		goto out_warn;
 
-	*dma_handle = swiotlb_phys_to_dma(dev, phys_addr);
+	*dma_handle = __phys_to_dma(dev, phys_addr);
 	if (dma_coherent_ok(dev, *dma_handle, size))
 		goto out_unmap;
 
@@ -759,28 +746,6 @@ out_warn:
 	return NULL;
 }
 
-void *
-swiotlb_alloc_coherent(struct device *hwdev, size_t size,
-		       dma_addr_t *dma_handle, gfp_t flags)
-{
-	int order = get_order(size);
-	unsigned long attrs = (flags & __GFP_NOWARN) ? DMA_ATTR_NO_WARN : 0;
-	void *ret;
-
-	ret = (void *)__get_free_pages(flags, order);
-	if (ret) {
-		*dma_handle = swiotlb_virt_to_bus(hwdev, ret);
-		if (dma_coherent_ok(hwdev, *dma_handle, size)) {
-			memset(ret, 0, size);
-			return ret;
-		}
-		free_pages((unsigned long)ret, order);
-	}
-
-	return swiotlb_alloc_buffer(hwdev, size, dma_handle, attrs);
-}
-EXPORT_SYMBOL(swiotlb_alloc_coherent);
-
 static bool swiotlb_free_buffer(struct device *dev, size_t size,
 		dma_addr_t dma_addr)
 {
@@ -799,15 +764,7 @@ static bool swiotlb_free_buffer(struct device *dev, size_t size,
 				 DMA_ATTR_SKIP_CPU_SYNC);
 	return true;
 }
-
-void
-swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr,
-		      dma_addr_t dev_addr)
-{
-	if (!swiotlb_free_buffer(hwdev, size, dev_addr))
-		free_pages((unsigned long)vaddr, get_order(size));
-}
-EXPORT_SYMBOL(swiotlb_free_coherent);
+#endif
 
 static void
 swiotlb_full(struct device *dev, size_t size, enum dma_data_direction dir,
@@ -867,10 +824,10 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
 	map = map_single(dev, phys, size, dir, attrs);
 	if (map == SWIOTLB_MAP_ERROR) {
 		swiotlb_full(dev, size, dir, 1);
-		return swiotlb_phys_to_dma(dev, io_tlb_overflow_buffer);
+		return __phys_to_dma(dev, io_tlb_overflow_buffer);
 	}
 
-	dev_addr = swiotlb_phys_to_dma(dev, map);
+	dev_addr = __phys_to_dma(dev, map);
 
 	/* Ensure that the address returned is DMA'ble */
 	if (dma_capable(dev, dev_addr, size))
@@ -879,7 +836,7 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
 	attrs |= DMA_ATTR_SKIP_CPU_SYNC;
 	swiotlb_tbl_unmap_single(dev, map, size, dir, attrs);
 
-	return swiotlb_phys_to_dma(dev, io_tlb_overflow_buffer);
+	return __phys_to_dma(dev, io_tlb_overflow_buffer);
 }
 
 /*
@@ -1009,7 +966,7 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems,
 				sg_dma_len(sgl) = 0;
 				return 0;
 			}
-			sg->dma_address = swiotlb_phys_to_dma(hwdev, map);
+			sg->dma_address = __phys_to_dma(hwdev, map);
 		} else
 			sg->dma_address = dev_addr;
 		sg_dma_len(sg) = sg->length;
@@ -1073,7 +1030,7 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
 int
 swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
 {
-	return (dma_addr == swiotlb_phys_to_dma(hwdev, io_tlb_overflow_buffer));
+	return (dma_addr == __phys_to_dma(hwdev, io_tlb_overflow_buffer));
 }
 
 /*
@@ -1085,7 +1042,7 @@ swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
 int
 swiotlb_dma_supported(struct device *hwdev, u64 mask)
 {
-	return swiotlb_phys_to_dma(hwdev, io_tlb_end - 1) <= mask;
+	return __phys_to_dma(hwdev, io_tlb_end - 1) <= mask;
 }
 
 #ifdef CONFIG_DMA_DIRECT_OPS