iommu: apple-dart: Support specifying the DMA aperture in the DT

Apple DARTs are often connected directly to devices that expect only a
portion of their address space to be used for DMA (for example, because
other ranges are mapped directly to something else). Add an
apple,dma-range property to allow specifying this range.

This range *can* be outside of the DART's IAS. In that case, it is
assumed that the hardware truncates addresses and the page tables will
only map the lower bits of the address. However, the specified range
cannot straddle an IAS boundary (you cannot cover more than IAS worth
of address space nor wrap).

This corresponds to the vm-base and vm-size properties on the Apple
device tree side of things.

Signed-off-by: Hector Martin <marcan@marcan.st>

Author: marcan

drivers/iommu/apple-dart.c

@@ -21,6 +21,7 @@
 #include <linux/io-pgtable.h>
 #include <linux/iommu.h>
 #include <linux/iopoll.h>
+#include <linux/minmax.h>
 #include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
@@ -226,6 +227,9 @@ struct apple_dart {
 	u32 locked : 1;
 	u32 four_level : 1;
 
+	dma_addr_t dma_min;
+	dma_addr_t dma_max;
+
 	struct iommu_group *sid2group[DART_MAX_STREAMS];
 	struct iommu_device iommu;
 
@@ -273,6 +277,7 @@ struct apple_dart_domain {
 	struct io_pgtable_ops *pgtbl_ops;
 
 	bool finalized;
+	u64 mask;
 	struct mutex init_lock;
 	struct apple_dart_atomic_stream_map stream_maps[MAX_DARTS_PER_DEVICE];
 
@@ -623,7 +628,7 @@ static phys_addr_t apple_dart_iova_to_phys(struct iommu_domain *domain,
 	if (!ops)
 		return 0;
 
-	return ops->iova_to_phys(ops, iova);
+	return ops->iova_to_phys(ops, iova & dart_domain->mask);
 }
 
 static int apple_dart_map_pages(struct iommu_domain *domain, unsigned long iova,
@@ -637,8 +642,8 @@ static int apple_dart_map_pages(struct iommu_domain *domain, unsigned long iova,
 	if (!ops)
 		return -ENODEV;
 
-	return ops->map_pages(ops, iova, paddr, pgsize, pgcount, prot, gfp,
-			      mapped);
+	return ops->map_pages(ops, iova & dart_domain->mask, paddr, pgsize,
+			      pgcount, prot, gfp, mapped);
 }
 
 static size_t apple_dart_unmap_pages(struct iommu_domain *domain,
@@ -649,7 +654,8 @@ static size_t apple_dart_unmap_pages(struct iommu_domain *domain,
 	struct apple_dart_domain *dart_domain = to_dart_domain(domain);
 	struct io_pgtable_ops *ops = dart_domain->pgtbl_ops;
 
-	return ops->unmap_pages(ops, iova, pgsize, pgcount, gather);
+	return ops->unmap_pages(ops, iova & dart_domain->mask, pgsize, pgcount,
+				gather);
 }
 
 static void
@@ -721,6 +727,8 @@ static int apple_dart_finalize_domain(struct apple_dart_domain *dart_domain,
 {
 	struct apple_dart *dart = cfg->stream_maps[0].dart;
 	struct io_pgtable_cfg pgtbl_cfg;
+	dma_addr_t dma_max = dart->dma_max;
+	u32 ias = min_t(u32, dart->ias, fls64(dma_max));
 	int ret = 0;
 	int i, j;
 
@@ -741,7 +749,7 @@ static int apple_dart_finalize_domain(struct apple_dart_domain *dart_domain,
 
 	pgtbl_cfg = (struct io_pgtable_cfg){
 		.pgsize_bitmap = dart->pgsize,
-		.ias = dart->ias,
+		.ias = ias,
 		.oas = dart->oas,
 		.coherent_walk = 1,
 		.iommu_dev = dart->dev,
@@ -764,13 +772,21 @@ static int apple_dart_finalize_domain(struct apple_dart_domain *dart_domain,
 		/* If the DART is locked, we need to keep the translation level count. */
 		if (dart->hw->tcr_4level && dart->ias > 36) {
 			if (readl(dart->regs + DART_TCR(dart, sid)) & dart->hw->tcr_4level) {
-				if (dart->ias < 37) {
+				if (ias < 37) {
 					dev_info(dart->dev, "Expanded to ias=37 due to lock\n");
 					pgtbl_cfg.ias = 37;
 				}
-			} else if (dart->ias > 36) {
+			} else if (ias > 36) {
 				dev_info(dart->dev, "Limited to ias=36 due to lock\n");
 				pgtbl_cfg.ias = 36;
+				if (dart->dma_min == 0 && dma_max == DMA_BIT_MASK(dart->ias)) {
+					dma_max = DMA_BIT_MASK(pgtbl_cfg.ias);
+				} else if ((dart->dma_min ^ dma_max) & ~DMA_BIT_MASK(36)) {
+					dev_err(dart->dev,
+						"Invalid DMA range for locked 3-level PT\n");
+					ret = -ENOMEM;
+					goto done;
+				}
 			}
 		}
 	}
@@ -782,10 +798,16 @@ static int apple_dart_finalize_domain(struct apple_dart_domain *dart_domain,
 		goto done;
 	}
 
+	if (pgtbl_cfg.pgsize_bitmap == SZ_4K)
+		dart_domain->mask = DMA_BIT_MASK(min_t(u32, dart->ias, 32));
+	else if (pgtbl_cfg.apple_dart_cfg.n_levels == 3)
+		dart_domain->mask = DMA_BIT_MASK(min_t(u32, dart->ias, 36));
+	else if (pgtbl_cfg.apple_dart_cfg.n_levels == 4)
+		dart_domain->mask = DMA_BIT_MASK(min_t(u32, dart->ias, 47));
+
 	dart_domain->domain.pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
-	dart_domain->domain.geometry.aperture_start = 0;
-	dart_domain->domain.geometry.aperture_end =
-		(dma_addr_t)DMA_BIT_MASK(pgtbl_cfg.ias);
+	dart_domain->domain.geometry.aperture_start = dart->dma_min;
+	dart_domain->domain.geometry.aperture_end = dma_max;
 	dart_domain->domain.geometry.force_aperture = true;
 
 	dart_domain->finalized = true;
@@ -1298,6 +1320,7 @@ static int apple_dart_probe(struct platform_device *pdev)
 	struct resource *res;
 	struct apple_dart *dart;
 	struct device *dev = &pdev->dev;
+	u64 dma_range[2];
 
 	dart = devm_kzalloc(dev, sizeof(*dart), GFP_KERNEL);
 	if (!dart)
@@ -1360,6 +1383,26 @@ static int apple_dart_probe(struct platform_device *pdev)
 		break;
 	}
 
+	dart->dma_min = 0;
+	dart->dma_max = DMA_BIT_MASK(dart->ias);
+
+	ret = of_property_read_u64_array(dev->of_node, "apple,dma-range", dma_range, 2);
+	if (ret == -EINVAL) {
+		ret = 0;
+	} else if (ret) {
+		goto err_clk_disable;
+	} else {
+		dart->dma_min = dma_range[0];
+		dart->dma_max = dma_range[0] + dma_range[1] - 1;
+		if ((dart->dma_min ^ dart->dma_max) & ~DMA_BIT_MASK(dart->ias)) {
+			dev_err(&pdev->dev, "Invalid DMA range for ias=%d\n",
+				dart->ias);
+			goto err_clk_disable;
+		}
+		dev_info(&pdev->dev, "Limiting DMA range to %pad..%pad\n",
+			 &dart->dma_min, &dart->dma_max);
+	}
+
 	if (dart->num_streams > DART_MAX_STREAMS) {
 		dev_err(&pdev->dev, "Too many streams (%d > %d)\n",
 			dart->num_streams, DART_MAX_STREAMS);