s390/pci_dma: fix DMA table corruption with > 4 TB main memory

DMA addresses returned from map_page() are calculated by using an iommu
bitmap plus a start_dma offset. The size of this bitmap is based on the main
memory size. If we have more than (4 TB - start_dma) main memory, the DMA
address calculation will also produce addresses > 4 TB. Such addresses
cannot be inserted in the 3-level DMA page table, instead the entries
modulo 4 TB will be overwritten.

Fix this by restricting the iommu bitmap size to (4 TB - start_dma).
Also set zdev->end_dma to the actual end address of the usable
range, instead of the theoretical maximum as reported by the hardware,
which fixes a sanity check in dma_map() and also the IOMMU API domain
geometry aperture calculation.

Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Reviewed-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

arch/s390/include/asm/pci_dma.h

@@ -23,6 +23,8 @@ enum zpci_ioat_dtype {
 #define ZPCI_IOTA_FS_2G			2
 #define ZPCI_KEY			(PAGE_DEFAULT_KEY << 5)
 
+#define ZPCI_TABLE_SIZE_RT	(1UL << 42)
+
 #define ZPCI_IOTA_STO_FLAG	(ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_ST)
 #define ZPCI_IOTA_RTTO_FLAG	(ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_RT)
 #define ZPCI_IOTA_RSTO_FLAG	(ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_RS)

arch/s390/pci/pci.c

@@ -701,8 +701,7 @@ static int zpci_restore(struct device *dev)
 		goto out;
 
 	zpci_map_resources(pdev);
-	zpci_register_ioat(zdev, 0, zdev->start_dma + PAGE_OFFSET,
-			   zdev->start_dma + zdev->iommu_size - 1,
+	zpci_register_ioat(zdev, 0, zdev->start_dma, zdev->end_dma,
 			   (u64) zdev->dma_table);
 
 out:

arch/s390/pci/pci_dma.c

@@ -458,7 +458,19 @@ int zpci_dma_init_device(struct zpci_dev *zdev)
 		goto out_clean;
 	}
 
-	zdev->iommu_size = (unsigned long) high_memory - PAGE_OFFSET;
+	/*
+	 * Restrict the iommu bitmap size to the minimum of the following:
+	 * - main memory size
+	 * - 3-level pagetable address limit minus start_dma offset
+	 * - DMA address range allowed by the hardware (clp query pci fn)
+	 *
+	 * Also set zdev->end_dma to the actual end address of the usable
+	 * range, instead of the theoretical maximum as reported by hardware.
+	 */
+	zdev->iommu_size = min3((u64) high_memory,
+				ZPCI_TABLE_SIZE_RT - zdev->start_dma,
+				zdev->end_dma - zdev->start_dma + 1);
+	zdev->end_dma = zdev->start_dma + zdev->iommu_size - 1;
 	zdev->iommu_pages = zdev->iommu_size >> PAGE_SHIFT;
 	zdev->iommu_bitmap = vzalloc(zdev->iommu_pages / 8);
 	if (!zdev->iommu_bitmap) {
@@ -466,10 +478,7 @@ int zpci_dma_init_device(struct zpci_dev *zdev)
 		goto out_reg;
 	}
 
-	rc = zpci_register_ioat(zdev,
-				0,
-				zdev->start_dma + PAGE_OFFSET,
-				zdev->start_dma + zdev->iommu_size - 1,
+	rc = zpci_register_ioat(zdev, 0, zdev->start_dma, zdev->end_dma,
 				(u64) zdev->dma_table);
 	if (rc)
 		goto out_reg;