OpenRISC: DMA

Simple DMA implementation.  Allows for allocation of coherent memory
(simply uncached) for DMA operations.
Signed-off-by: Jonas Bonn <jonas@southpole.se>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>

arch/openrisc/include/asm/dma-mapping.h

+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others.  All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <jonas@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_DMA_MAPPING_H
+#define __ASM_OPENRISC_DMA_MAPPING_H
+
+/*
+ * See Documentation/PCI/PCI-DMA-mapping.txt and
+ * Documentation/DMA-API.txt for documentation.
+ *
+ * This file is written with the intention of eventually moving over
+ * to largely using asm-generic/dma-mapping-common.h in its place.
+ */
+
+#include <linux/dma-debug.h>
+#include <asm-generic/dma-coherent.h>
+#include <linux/kmemcheck.h>
+
+#define DMA_ERROR_CODE		(~(dma_addr_t)0x0)
+
+int dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+void *or1k_dma_alloc_coherent(struct device *dev, size_t size,
+			      dma_addr_t *dma_handle, gfp_t flag);
+void or1k_dma_free_coherent(struct device *dev, size_t size, void *vaddr,
+			    dma_addr_t dma_handle);
+dma_addr_t or1k_map_page(struct device *dev, struct page *page,
+			 unsigned long offset, size_t size,
+			 enum dma_data_direction dir,
+			 struct dma_attrs *attrs);
+void or1k_unmap_page(struct device *dev, dma_addr_t dma_handle,
+		     size_t size, enum dma_data_direction dir,
+		     struct dma_attrs *attrs);
+void or1k_sync_single_for_cpu(struct device *dev,
+			      dma_addr_t dma_handle, size_t size,
+			      enum dma_data_direction dir);
+void or1k_sync_single_for_device(struct device *dev,
+			         dma_addr_t dma_handle, size_t size,
+			         enum dma_data_direction dir);
+
+static inline void *dma_alloc_coherent(struct device *dev, size_t size,
+					dma_addr_t *dma_handle, gfp_t flag)
+{
+	void *memory;
+
+	memory = or1k_dma_alloc_coherent(dev, size, dma_handle, flag);
+
+	debug_dma_alloc_coherent(dev, size, *dma_handle, memory);
+	return memory;
+}
+
+static inline void dma_free_coherent(struct device *dev, size_t size,
+				     void *cpu_addr, dma_addr_t dma_handle)
+{
+	debug_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+	or1k_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+}
+
+static inline dma_addr_t dma_map_single(struct device *dev, void *ptr,
+					size_t size,
+					enum dma_data_direction dir)
+{
+	dma_addr_t addr;
+
+	kmemcheck_mark_initialized(ptr, size);
+	BUG_ON(!valid_dma_direction(dir));
+	addr = or1k_map_page(dev, virt_to_page(ptr),
+			     (unsigned long)ptr & ~PAGE_MASK, size,
+			     dir, NULL);
+	debug_dma_map_page(dev, virt_to_page(ptr),
+			   (unsigned long)ptr & ~PAGE_MASK, size,
+			   dir, addr, true);
+	return addr;
+}
+
+static inline void dma_unmap_single(struct device *dev, dma_addr_t addr,
+					  size_t size,
+					  enum dma_data_direction dir)
+{
+	BUG_ON(!valid_dma_direction(dir));
+	or1k_unmap_page(dev, addr, size, dir, NULL);
+	debug_dma_unmap_page(dev, addr, size, dir, true);
+}
+
+static inline void dma_sync_single_for_cpu(struct device *dev, dma_addr_t addr,
+					   size_t size,
+					   enum dma_data_direction dir)
+{
+	BUG_ON(!valid_dma_direction(dir));
+	or1k_sync_single_for_cpu(dev, addr, size, dir);
+	debug_dma_sync_single_for_cpu(dev, addr, size, dir);
+}
+
+static inline void dma_sync_single_for_device(struct device *dev,
+					      dma_addr_t addr, size_t size,
+					      enum dma_data_direction dir)
+{
+	BUG_ON(!valid_dma_direction(dir));
+	or1k_sync_single_for_device(dev, addr, size, dir);
+	debug_dma_sync_single_for_device(dev, addr, size, dir);
+}
+
+static inline int dma_supported(struct device *dev, u64 dma_mask)
+{
+	/* Support 32 bit DMA mask exclusively */
+	return dma_mask == 0xffffffffULL;
+}
+
+static inline int dma_set_mask(struct device *dev, u64 dma_mask)
+{
+	if (!dev->dma_mask || !dma_supported(dev, dma_mask))
+		return -EIO;
+
+	*dev->dma_mask = dma_mask;
+
+	return 0;
+}
+#endif	/* __ASM_OPENRISC_DMA_MAPPING_H */

arch/openrisc/kernel/dma.c

+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others.  All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <phoenix@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <jonas@southpole.se>
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ *
+ * DMA mapping callbacks...
+ * As alloc_coherent is the only DMA callback being used currently, that's
+ * the only thing implemented properly.  The rest need looking into...
+ */
+
+#include <linux/dma-mapping.h>
+#include <linux/dma-debug.h>
+
+#include <asm/cpuinfo.h>
+#include <asm/spr_defs.h>
+#include <asm/tlbflush.h>
+
+static int page_set_nocache(pte_t *pte, unsigned long addr,
+			    unsigned long next, struct mm_walk *walk)
+{
+	unsigned long cl;
+
+	pte_val(*pte) |= _PAGE_CI;
+
+	/*
+	 * Flush the page out of the TLB so that the new page flags get
+	 * picked up next time there's an access
+	 */
+	flush_tlb_page(NULL, addr);
+
+	/* Flush page out of dcache */
+	for (cl = __pa(addr); cl < __pa(next); cl += cpuinfo.dcache_block_size)
+		mtspr(SPR_DCBFR, cl);
+
+	return 0;
+}
+
+static int page_clear_nocache(pte_t *pte, unsigned long addr,
+			      unsigned long next, struct mm_walk *walk)
+{
+	pte_val(*pte) &= ~_PAGE_CI;
+
+	/*
+	 * Flush the page out of the TLB so that the new page flags get
+	 * picked up next time there's an access
+	 */
+	flush_tlb_page(NULL, addr);
+
+	return 0;
+}
+
+/*
+ * Alloc "coherent" memory, which for OpenRISC means simply uncached.
+ *
+ * This function effectively just calls __get_free_pages, sets the
+ * cache-inhibit bit on those pages, and makes sure that the pages are
+ * flushed out of the cache before they are used.
+ *
+ */
+void *or1k_dma_alloc_coherent(struct device *dev, size_t size,
+			      dma_addr_t *dma_handle, gfp_t gfp)
+{
+	unsigned long va;
+	void *page;
+	struct mm_walk walk = {
+		.pte_entry = page_set_nocache,
+		.mm = &init_mm
+	};
+
+	page = alloc_pages_exact(size, gfp);
+	if (!page)
+		return NULL;
+
+	/* This gives us the real physical address of the first page. */
+	*dma_handle = __pa(page);
+
+	va = (unsigned long)page;
+
+	/*
+	 * We need to iterate through the pages, clearing the dcache for
+	 * them and setting the cache-inhibit bit.
+	 */
+	if (walk_page_range(va, va + size, &walk)) {
+		free_pages_exact(page, size);
+		return NULL;
+	}
+
+	return (void *)va;
+}
+
+void or1k_dma_free_coherent(struct device *dev, size_t size, void *vaddr,
+			    dma_addr_t dma_handle)
+{
+	unsigned long va = (unsigned long)vaddr;
+	struct mm_walk walk = {
+		.pte_entry = page_clear_nocache,
+		.mm = &init_mm
+	};
+
+	/* walk_page_range shouldn't be able to fail here */
+	WARN_ON(walk_page_range(va, va + size, &walk));
+
+	free_pages_exact(vaddr, size);
+}
+
+dma_addr_t or1k_map_page(struct device *dev, struct page *page,
+			 unsigned long offset, size_t size,
+			 enum dma_data_direction dir,
+			 struct dma_attrs *attrs)
+{
+	unsigned long cl;
+	dma_addr_t addr = page_to_phys(page) + offset;
+
+	switch (dir) {
+	case DMA_TO_DEVICE:
+		/* Flush the dcache for the requested range */
+		for (cl = addr; cl < addr + size;
+		     cl += cpuinfo.dcache_block_size)
+			mtspr(SPR_DCBFR, cl);
+		break;
+	case DMA_FROM_DEVICE:
+		/* Invalidate the dcache for the requested range */
+		for (cl = addr; cl < addr + size;
+		     cl += cpuinfo.dcache_block_size)
+			mtspr(SPR_DCBIR, cl);
+		break;
+	default:
+		/*
+		 * NOTE: If dir == DMA_BIDIRECTIONAL then there's no need to
+		 * flush nor invalidate the cache here as the area will need
+		 * to be manually synced anyway.
+		 */
+		break;
+	}
+
+	return addr;
+}
+
+void or1k_unmap_page(struct device *dev, dma_addr_t dma_handle,
+		     size_t size, enum dma_data_direction dir,
+		     struct dma_attrs *attrs)
+{
+	/* Nothing special to do here... */
+}
+
+void or1k_sync_single_for_cpu(struct device *dev,
+			      dma_addr_t dma_handle, size_t size,
+			      enum dma_data_direction dir)
+{
+	unsigned long cl;
+	dma_addr_t addr = dma_handle;
+
+	/* Invalidate the dcache for the requested range */
+	for (cl = addr; cl < addr + size; cl += cpuinfo.dcache_block_size)
+		mtspr(SPR_DCBIR, cl);
+}
+
+void or1k_sync_single_for_device(struct device *dev,
+			         dma_addr_t dma_handle, size_t size,
+			         enum dma_data_direction dir)
+{
+	unsigned long cl;
+	dma_addr_t addr = dma_handle;
+
+	/* Flush the dcache for the requested range */
+	for (cl = addr; cl < addr + size; cl += cpuinfo.dcache_block_size)
+		mtspr(SPR_DCBFR, cl);
+}
+
+/* Number of entries preallocated for DMA-API debugging */
+#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)
+
+static int __init dma_init(void)
+{
+	dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
+
+	return 0;
+}
+
+fs_initcall(dma_init);