Skip to contentSkip to conversation

swiotlb-xen cleanups v2

PATCH In review Christoph Hellwig 2 participants 26 messages Aug 26, 2019 – Aug 29, 2019

Subsystems (8)
lore
Share & tools
lore
#0 lore raw

Hi Xen maintainers and friends,

please take a look at this series that cleans up the parts of swiotlb-xen that deal with non-coherent caches.

Changes since v1:
- rewrite dma_cache_maint to be much simpler
- improve various comments and commit logs
- remove page-coherent.h entirely

#1 lore raw

PATCH 1/11 xen/arm: use dma-noncoherent.h calls for xen-swiotlb cache maintainance

Reuse the arm64 code that uses the dma-direct/swiotlb helpers for DMA non-coherent devices.

---
 arch/arm/include/asm/device.h              |  3 -
 arch/arm/include/asm/xen/page-coherent.h   | 93 ----------------------
 arch/arm/mm/dma-mapping.c                  |  8 +-
 arch/arm64/include/asm/xen/page-coherent.h | 75 -----------------
 drivers/xen/swiotlb-xen.c                  | 49 +-----------
 include/xen/arm/page-coherent.h            | 80 +++++++++++++++++++
 6 files changed, 83 insertions(+), 225 deletions(-)
arch/arm/include/asm/device.h +0 -3
index f6955b55c544..c675bc0d5aa8 100644
@@ -14,9 +14,6 @@ struct dev_archdata {
 #endif
 #ifdef CONFIG_ARM_DMA_USE_IOMMU
 	struct dma_iommu_mapping	*mapping;
-#endif
-#ifdef CONFIG_XEN
-	const struct dma_map_ops *dev_dma_ops;
 #endif
 	unsigned int dma_coherent:1;
 	unsigned int dma_ops_setup:1;
arch/arm/include/asm/xen/page-coherent.h +0 -93

95 lines not on this page · raw message

arch/arm/mm/dma-mapping.c +1 -7
index d42557ee69c2..738097396445 100644
@@ -1132,10 +1132,6 @@ static const struct dma_map_ops *arm_get_dma_map_ops(bool coherent)
 	 * 32-bit DMA.
 	 * Use the generic dma-direct / swiotlb ops code in that case, as that
 	 * handles bounce buffering for us.
-	 *
-	 * Note: this checks CONFIG_ARM_LPAE instead of CONFIG_SWIOTLB as the
-	 * latter is also selected by the Xen code, but that code for now relies
-	 * on non-NULL dev_dma_ops.  To be cleaned up later.
 	 */
 	if (IS_ENABLED(CONFIG_ARM_LPAE))
 		return NULL;
@@ -2363,10 +2359,8 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 	set_dma_ops(dev, dma_ops);
 
 #ifdef CONFIG_XEN
-	if (xen_initial_domain()) {
-		dev->archdata.dev_dma_ops = dev->dma_ops;
+	if (xen_initial_domain())
 		dev->dma_ops = xen_dma_ops;
-	}
 #endif
 	dev->archdata.dma_ops_setup = true;
 }
arch/arm64/include/asm/xen/page-coherent.h +0 -75

77 lines not on this page · raw message

drivers/xen/swiotlb-xen.c +2 -47

59 lines not on this page · raw message

include/xen/arm/page-coherent.h +80 -0

90 lines not on this page · raw message

Signature · 1 line
2.20.1
#2 lore raw

Re: [PATCH 01/11] xen/arm: use dma-noncoherent.h calls for xen-swiotlb cache maintainance

Quoted 2 lines · Christoph Hellwig

Reuse the arm64 code that uses the dma-direct/swiotlb helpers for DMA non-coherent devices.

This patch does a bunch of things not listed in the commit message, such as moving the static inline functions to include/xen/arm/page-coherent.h and removing xen_swiotlb_dma_mmap and xen_swiotlb_get_sgtable because unnecessary.

I would prefer if they were separate patches (for bisectability). It's OK if you want to keep it all in one patch but please list all changes the commit message.

In any case, I looked at the patch in details and it does all the right things -- it's correct.

Quoted 397 lines
---
 arch/arm/include/asm/device.h              |  3 -
 arch/arm/include/asm/xen/page-coherent.h   | 93 ----------------------
 arch/arm/mm/dma-mapping.c                  |  8 +-
 arch/arm64/include/asm/xen/page-coherent.h | 75 -----------------
 drivers/xen/swiotlb-xen.c                  | 49 +-----------
 include/xen/arm/page-coherent.h            | 80 +++++++++++++++++++
 6 files changed, 83 insertions(+), 225 deletions(-)
arch/arm/include/asm/device.h +0 -3
index f6955b55c544..c675bc0d5aa8 100644
@@ -14,9 +14,6 @@ struct dev_archdata {
 #endif
 #ifdef CONFIG_ARM_DMA_USE_IOMMU
 	struct dma_iommu_mapping	*mapping;
-#endif
-#ifdef CONFIG_XEN
-	const struct dma_map_ops *dev_dma_ops;
 #endif
 	unsigned int dma_coherent:1;
 	unsigned int dma_ops_setup:1;
arch/arm/include/asm/xen/page-coherent.h +0 -93
index 2c403e7c782d..27e984977402 100644
@@ -1,95 +1,2 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_ARM_XEN_PAGE_COHERENT_H
-#define _ASM_ARM_XEN_PAGE_COHERENT_H
-
-#include <linux/dma-mapping.h>
-#include <asm/page.h>
 #include <xen/arm/page-coherent.h>
-
-static inline const struct dma_map_ops *xen_get_dma_ops(struct device *dev)
-{
-	if (dev && dev->archdata.dev_dma_ops)
-		return dev->archdata.dev_dma_ops;
-	return get_arch_dma_ops(NULL);
-}
-
-static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flags, unsigned long attrs)
-{
-	return xen_get_dma_ops(hwdev)->alloc(hwdev, size, dma_handle, flags, attrs);
-}
-
-static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
-		void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs)
-{
-	xen_get_dma_ops(hwdev)->free(hwdev, size, cpu_addr, dma_handle, attrs);
-}
-
-static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs)
-{
-	unsigned long page_pfn = page_to_xen_pfn(page);
-	unsigned long dev_pfn = XEN_PFN_DOWN(dev_addr);
-	unsigned long compound_pages =
-		(1<<compound_order(page)) * XEN_PFN_PER_PAGE;
-	bool local = (page_pfn <= dev_pfn) &&
-		(dev_pfn - page_pfn < compound_pages);
-
-	/*
-	 * Dom0 is mapped 1:1, while the Linux page can span across
-	 * multiple Xen pages, it's not possible for it to contain a
-	 * mix of local and foreign Xen pages. So if the first xen_pfn
-	 * == mfn the page is local otherwise it's a foreign page
-	 * grant-mapped in dom0. If the page is local we can safely
-	 * call the native dma_ops function, otherwise we call the xen
-	 * specific function.
-	 */
-	if (local)
-		xen_get_dma_ops(hwdev)->map_page(hwdev, page, offset, size, dir, attrs);
-	else
-		__xen_dma_map_page(hwdev, page, dev_addr, offset, size, dir, attrs);
-}
-
-static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir, unsigned long attrs)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-	/*
-	 * Dom0 is mapped 1:1, while the Linux page can be spanned accross
-	 * multiple Xen page, it's not possible to have a mix of local and
-	 * foreign Xen page. Dom0 is mapped 1:1, so calling pfn_valid on a
-	 * foreign mfn will always return false. If the page is local we can
-	 * safely call the native dma_ops function, otherwise we call the xen
-	 * specific function.
-	 */
-	if (pfn_valid(pfn)) {
-		if (xen_get_dma_ops(hwdev)->unmap_page)
-			xen_get_dma_ops(hwdev)->unmap_page(hwdev, handle, size, dir, attrs);
-	} else
-		__xen_dma_unmap_page(hwdev, handle, size, dir, attrs);
-}
-
-static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-	if (pfn_valid(pfn)) {
-		if (xen_get_dma_ops(hwdev)->sync_single_for_cpu)
-			xen_get_dma_ops(hwdev)->sync_single_for_cpu(hwdev, handle, size, dir);
-	} else
-		__xen_dma_sync_single_for_cpu(hwdev, handle, size, dir);
-}
-
-static inline void xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-	if (pfn_valid(pfn)) {
-		if (xen_get_dma_ops(hwdev)->sync_single_for_device)
-			xen_get_dma_ops(hwdev)->sync_single_for_device(hwdev, handle, size, dir);
-	} else
-		__xen_dma_sync_single_for_device(hwdev, handle, size, dir);
-}
-
-#endif /* _ASM_ARM_XEN_PAGE_COHERENT_H */
arch/arm/mm/dma-mapping.c +1 -7
index d42557ee69c2..738097396445 100644
@@ -1132,10 +1132,6 @@ static const struct dma_map_ops *arm_get_dma_map_ops(bool coherent)
 	 * 32-bit DMA.
 	 * Use the generic dma-direct / swiotlb ops code in that case, as that
 	 * handles bounce buffering for us.
-	 *
-	 * Note: this checks CONFIG_ARM_LPAE instead of CONFIG_SWIOTLB as the
-	 * latter is also selected by the Xen code, but that code for now relies
-	 * on non-NULL dev_dma_ops.  To be cleaned up later.
 	 */
 	if (IS_ENABLED(CONFIG_ARM_LPAE))
 		return NULL;
@@ -2363,10 +2359,8 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 	set_dma_ops(dev, dma_ops);
 
 #ifdef CONFIG_XEN
-	if (xen_initial_domain()) {
-		dev->archdata.dev_dma_ops = dev->dma_ops;
+	if (xen_initial_domain())
 		dev->dma_ops = xen_dma_ops;
-	}
 #endif
 	dev->archdata.dma_ops_setup = true;
 }
arch/arm64/include/asm/xen/page-coherent.h +0 -75
index d88e56b90b93..27e984977402 100644
@@ -1,77 +1,2 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_ARM64_XEN_PAGE_COHERENT_H
-#define _ASM_ARM64_XEN_PAGE_COHERENT_H
-
-#include <linux/dma-mapping.h>
-#include <asm/page.h>
 #include <xen/arm/page-coherent.h>
-
-static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flags, unsigned long attrs)
-{
-	return dma_direct_alloc(hwdev, size, dma_handle, flags, attrs);
-}
-
-static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
-		void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs)
-{
-	dma_direct_free(hwdev, size, cpu_addr, dma_handle, attrs);
-}
-
-static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-
-	if (pfn_valid(pfn))
-		dma_direct_sync_single_for_cpu(hwdev, handle, size, dir);
-	else
-		__xen_dma_sync_single_for_cpu(hwdev, handle, size, dir);
-}
-
-static inline void xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-	if (pfn_valid(pfn))
-		dma_direct_sync_single_for_device(hwdev, handle, size, dir);
-	else
-		__xen_dma_sync_single_for_device(hwdev, handle, size, dir);
-}
-
-static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs)
-{
-	unsigned long page_pfn = page_to_xen_pfn(page);
-	unsigned long dev_pfn = XEN_PFN_DOWN(dev_addr);
-	unsigned long compound_pages =
-		(1<<compound_order(page)) * XEN_PFN_PER_PAGE;
-	bool local = (page_pfn <= dev_pfn) &&
-		(dev_pfn - page_pfn < compound_pages);
-
-	if (local)
-		dma_direct_map_page(hwdev, page, offset, size, dir, attrs);
-	else
-		__xen_dma_map_page(hwdev, page, dev_addr, offset, size, dir, attrs);
-}
-
-static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir, unsigned long attrs)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-	/*
-	 * Dom0 is mapped 1:1, while the Linux page can be spanned accross
-	 * multiple Xen page, it's not possible to have a mix of local and
-	 * foreign Xen page. Dom0 is mapped 1:1, so calling pfn_valid on a
-	 * foreign mfn will always return false. If the page is local we can
-	 * safely call the native dma_ops function, otherwise we call the xen
-	 * specific function.
-	 */
-	if (pfn_valid(pfn))
-		dma_direct_unmap_page(hwdev, handle, size, dir, attrs);
-	else
-		__xen_dma_unmap_page(hwdev, handle, size, dir, attrs);
-}
-
-#endif /* _ASM_ARM64_XEN_PAGE_COHERENT_H */
drivers/xen/swiotlb-xen.c +2 -47
index ae1df496bf38..b8808677ae1d 100644
@@ -547,51 +547,6 @@ xen_swiotlb_dma_supported(struct device *hwdev, u64 mask)
 	return xen_virt_to_bus(xen_io_tlb_end - 1) <= mask;
 }
 
-/*
- * Create userspace mapping for the DMA-coherent memory.
- * This function should be called with the pages from the current domain only,
- * passing pages mapped from other domains would lead to memory corruption.
- */
-static int
-xen_swiotlb_dma_mmap(struct device *dev, struct vm_area_struct *vma,
-		     void *cpu_addr, dma_addr_t dma_addr, size_t size,
-		     unsigned long attrs)
-{
-#ifdef CONFIG_ARM
-	if (xen_get_dma_ops(dev)->mmap)
-		return xen_get_dma_ops(dev)->mmap(dev, vma, cpu_addr,
-						    dma_addr, size, attrs);
-#endif
-	return dma_common_mmap(dev, vma, cpu_addr, dma_addr, size, attrs);
-}
-
-/*
- * This function should be called with the pages from the current domain only,
- * passing pages mapped from other domains would lead to memory corruption.
- */
-static int
-xen_swiotlb_get_sgtable(struct device *dev, struct sg_table *sgt,
-			void *cpu_addr, dma_addr_t handle, size_t size,
-			unsigned long attrs)
-{
-#ifdef CONFIG_ARM
-	if (xen_get_dma_ops(dev)->get_sgtable) {
-#if 0
-	/*
-	 * This check verifies that the page belongs to the current domain and
-	 * is not one mapped from another domain.
-	 * This check is for debug only, and should not go to production build
-	 */
-		unsigned long bfn = PHYS_PFN(dma_to_phys(dev, handle));
-		BUG_ON (!page_is_ram(bfn));
-#endif
-		return xen_get_dma_ops(dev)->get_sgtable(dev, sgt, cpu_addr,
-							   handle, size, attrs);
-	}
-#endif
-	return dma_common_get_sgtable(dev, sgt, cpu_addr, handle, size, attrs);
-}
-
 const struct dma_map_ops xen_swiotlb_dma_ops = {
 	.alloc = xen_swiotlb_alloc_coherent,
 	.free = xen_swiotlb_free_coherent,
@@ -604,6 +559,6 @@ const struct dma_map_ops xen_swiotlb_dma_ops = {
 	.map_page = xen_swiotlb_map_page,
 	.unmap_page = xen_swiotlb_unmap_page,
 	.dma_supported = xen_swiotlb_dma_supported,
-	.mmap = xen_swiotlb_dma_mmap,
-	.get_sgtable = xen_swiotlb_get_sgtable,
+	.mmap = dma_common_mmap,
+	.get_sgtable = dma_common_get_sgtable,
 };
include/xen/arm/page-coherent.h +80 -0
index 2ca9164a79bf..a840d6949a87 100644
@@ -2,6 +2,9 @@
 #ifndef _XEN_ARM_PAGE_COHERENT_H
 #define _XEN_ARM_PAGE_COHERENT_H
 
+#include <linux/dma-mapping.h>
+#include <asm/page.h>
+
 void __xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, unsigned long attrs);
@@ -13,4 +16,81 @@ void __xen_dma_sync_single_for_cpu(struct device *hwdev,
 void __xen_dma_sync_single_for_device(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir);
 
+static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
+		dma_addr_t *dma_handle, gfp_t flags, unsigned long attrs)
+{
+	return dma_direct_alloc(hwdev, size, dma_handle, flags, attrs);
+}
+
+static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
+		void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs)
+{
+	dma_direct_free(hwdev, size, cpu_addr, dma_handle, attrs);
+}
+
+static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	unsigned long pfn = PFN_DOWN(handle);
+
+	if (pfn_valid(pfn))
+		dma_direct_sync_single_for_cpu(hwdev, handle, size, dir);
+	else
+		__xen_dma_sync_single_for_cpu(hwdev, handle, size, dir);
+}
+
+static inline void xen_dma_sync_single_for_device(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	unsigned long pfn = PFN_DOWN(handle);
+	if (pfn_valid(pfn))
+		dma_direct_sync_single_for_device(hwdev, handle, size, dir);
+	else
+		__xen_dma_sync_single_for_device(hwdev, handle, size, dir);
+}
+
+static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
+	     dma_addr_t dev_addr, unsigned long offset, size_t size,
+	     enum dma_data_direction dir, unsigned long attrs)
+{
+	unsigned long page_pfn = page_to_xen_pfn(page);
+	unsigned long dev_pfn = XEN_PFN_DOWN(dev_addr);
+	unsigned long compound_pages =
+		(1<<compound_order(page)) * XEN_PFN_PER_PAGE;
+	bool local = (page_pfn <= dev_pfn) &&
+		(dev_pfn - page_pfn < compound_pages);
+
+	/*
+	 * Dom0 is mapped 1:1, while the Linux page can span across
+	 * multiple Xen pages, it's not possible for it to contain a
+	 * mix of local and foreign Xen pages. So if the first xen_pfn
+	 * == mfn the page is local otherwise it's a foreign page
+	 * grant-mapped in dom0. If the page is local we can safely
+	 * call the native dma_ops function, otherwise we call the xen
+	 * specific function.
+	 */
+	if (local)
+		dma_direct_map_page(hwdev, page, offset, size, dir, attrs);
+	else
+		__xen_dma_map_page(hwdev, page, dev_addr, offset, size, dir, attrs);
+}
+
+static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir, unsigned long attrs)
+{
+	unsigned long pfn = PFN_DOWN(handle);
+	/*
+	 * Dom0 is mapped 1:1, while the Linux page can be spanned accross
+	 * multiple Xen page, it's not possible to have a mix of local and
+	 * foreign Xen page. Dom0 is mapped 1:1, so calling pfn_valid on a
+	 * foreign mfn will always return false. If the page is local we can
+	 * safely call the native dma_ops function, otherwise we call the xen
+	 * specific function.
+	 */
+	if (pfn_valid(pfn))
+		dma_direct_unmap_page(hwdev, handle, size, dir, attrs);
+	else
+		__xen_dma_unmap_page(hwdev, handle, size, dir, attrs);
+}
+
 #endif /* _XEN_ARM_PAGE_COHERENT_H */
Signature · 1 line
2.20.1
#3 lore raw

PATCH 2/11 xen/arm: use dev_is_dma_coherent

In reply to Christoph Hellwig

Use the dma-noncoherent dev_is_dma_coherent helper instead of the home grown variant. Note that both are always initialized to the same value in arch_setup_dma_ops.

---
 arch/arm/include/asm/dma-mapping.h   |  6 ------
 arch/arm/xen/mm.c                    | 12 ++++++------
 arch/arm64/include/asm/dma-mapping.h |  9 ---------
 3 files changed, 6 insertions(+), 21 deletions(-)
arch/arm/include/asm/dma-mapping.h +0 -6
index dba9355e2484..bdd80ddbca34 100644
@@ -91,12 +91,6 @@ static inline dma_addr_t virt_to_dma(struct device *dev, void *addr)
 }
 #endif
 
-/* do not use this function in a driver */
-static inline bool is_device_dma_coherent(struct device *dev)
-{
-	return dev->archdata.dma_coherent;
-}
-
 /**
  * arm_dma_alloc - allocate consistent memory for DMA
  * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
arch/arm/xen/mm.c +6 -6
index d33b77e9add3..90574d89d0d4 100644
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only
 #include <linux/cpu.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-noncoherent.h>
 #include <linux/gfp.h>
 #include <linux/highmem.h>
 #include <linux/export.h>
@@ -99,7 +99,7 @@ void __xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, unsigned long attrs)
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
 		return;
@@ -112,7 +112,7 @@ void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
 		unsigned long attrs)
 
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
 		return;
@@ -123,7 +123,7 @@ void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
 void __xen_dma_sync_single_for_cpu(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
 }
@@ -131,7 +131,7 @@ void __xen_dma_sync_single_for_cpu(struct device *hwdev,
 void __xen_dma_sync_single_for_device(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	__xen_dma_page_cpu_to_dev(hwdev, handle, size, dir);
 }
@@ -159,7 +159,7 @@ bool xen_arch_need_swiotlb(struct device *dev,
 	 * memory and we are not able to flush the cache.
 	 */
 	return (!hypercall_cflush && (xen_pfn != bfn) &&
-		!is_device_dma_coherent(dev));
+		!dev_is_dma_coherent(dev));
 }
 
 int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
arch/arm64/include/asm/dma-mapping.h +0 -9
index bdcb0922a40c..67243255a858 100644
@@ -18,14 +18,5 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return NULL;
 }
 
-/*
- * Do not use this function in a driver, it is only provided for
- * arch/arm/mm/xen.c, which is used by arm64 as well.
- */
-static inline bool is_device_dma_coherent(struct device *dev)
-{
-	return dev->dma_coherent;
-}
-
 #endif	/* __KERNEL__ */
 #endif	/* __ASM_DMA_MAPPING_H */
Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 2/11
#4 lore raw

Re: [PATCH 02/11] xen/arm: use dev_is_dma_coherent

Quoted 6 lines · Christoph Hellwig

Use the dma-noncoherent dev_is_dma_coherent helper instead of the home grown variant. Note that both are always initialized to the same value in arch_setup_dma_ops.

Quoted 102 lines
---
 arch/arm/include/asm/dma-mapping.h   |  6 ------
 arch/arm/xen/mm.c                    | 12 ++++++------
 arch/arm64/include/asm/dma-mapping.h |  9 ---------
 3 files changed, 6 insertions(+), 21 deletions(-)
arch/arm/include/asm/dma-mapping.h +0 -6
index dba9355e2484..bdd80ddbca34 100644
@@ -91,12 +91,6 @@ static inline dma_addr_t virt_to_dma(struct device *dev, void *addr)
 }
 #endif
 
-/* do not use this function in a driver */
-static inline bool is_device_dma_coherent(struct device *dev)
-{
-	return dev->archdata.dma_coherent;
-}
-
 /**
  * arm_dma_alloc - allocate consistent memory for DMA
  * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
arch/arm/xen/mm.c +6 -6
index d33b77e9add3..90574d89d0d4 100644
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only
 #include <linux/cpu.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-noncoherent.h>
 #include <linux/gfp.h>
 #include <linux/highmem.h>
 #include <linux/export.h>
@@ -99,7 +99,7 @@ void __xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, unsigned long attrs)
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
 		return;
@@ -112,7 +112,7 @@ void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
 		unsigned long attrs)
 
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
 		return;
@@ -123,7 +123,7 @@ void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
 void __xen_dma_sync_single_for_cpu(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
 }
@@ -131,7 +131,7 @@ void __xen_dma_sync_single_for_cpu(struct device *hwdev,
 void __xen_dma_sync_single_for_device(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
-	if (is_device_dma_coherent(hwdev))
+	if (dev_is_dma_coherent(hwdev))
 		return;
 	__xen_dma_page_cpu_to_dev(hwdev, handle, size, dir);
 }
@@ -159,7 +159,7 @@ bool xen_arch_need_swiotlb(struct device *dev,
 	 * memory and we are not able to flush the cache.
 	 */
 	return (!hypercall_cflush && (xen_pfn != bfn) &&
-		!is_device_dma_coherent(dev));
+		!dev_is_dma_coherent(dev));
 }
 
 int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
arch/arm64/include/asm/dma-mapping.h +0 -9
index bdcb0922a40c..67243255a858 100644
@@ -18,14 +18,5 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return NULL;
 }
 
-/*
- * Do not use this function in a driver, it is only provided for
- * arch/arm/mm/xen.c, which is used by arm64 as well.
- */
-static inline bool is_device_dma_coherent(struct device *dev)
-{
-	return dev->dma_coherent;
-}
-
 #endif	/* __KERNEL__ */
 #endif	/* __ASM_DMA_MAPPING_H */
Signature · 1 line
2.20.1
#5 lore raw

PATCH 3/11 xen/arm: simplify dma_cache_maint

In reply to Christoph Hellwig

Calculate the required operation in the caller, and pass it directly instead of recalculating it for each page, and use simple arithmetics to get from the physical address to Xen page size aligned chunks.

---
 arch/arm/xen/mm.c | 62 +++++++++++++++++------------------------------
 1 file changed, 22 insertions(+), 40 deletions(-)
arch/arm/xen/mm.c +22 -40
index 90574d89d0d4..14210ebdea1a 100644
@@ -35,64 +35,46 @@ unsigned long xen_get_swiotlb_free_pages(unsigned int order)
 	return __get_free_pages(flags, order);
 }
 
-enum dma_cache_op {
-       DMA_UNMAP,
-       DMA_MAP,
-};
 static bool hypercall_cflush = false;
 
-/* functions called by SWIOTLB */
-
-static void dma_cache_maint(dma_addr_t handle, unsigned long offset,
-	size_t size, enum dma_data_direction dir, enum dma_cache_op op)
+/* buffers in highmem or foreign pages cannot cross page boundaries */
+static void dma_cache_maint(dma_addr_t handle, size_t size, u32 op)
 {
 	struct gnttab_cache_flush cflush;
-	unsigned long xen_pfn;
-	size_t left = size;
 
-	xen_pfn = (handle >> XEN_PAGE_SHIFT) + offset / XEN_PAGE_SIZE;
-	offset %= XEN_PAGE_SIZE;
+	cflush.a.dev_bus_addr = handle & XEN_PAGE_MASK;
+	cflush.offset = xen_offset_in_page(handle);
+	cflush.op = op;
 
 	do {
-		size_t len = left;
-	
-		/* buffers in highmem or foreign pages cannot cross page
-		 * boundaries */
-		if (len + offset > XEN_PAGE_SIZE)
-			len = XEN_PAGE_SIZE - offset;
-
-		cflush.op = 0;
-		cflush.a.dev_bus_addr = xen_pfn << XEN_PAGE_SHIFT;
-		cflush.offset = offset;
-		cflush.length = len;
-
-		if (op == DMA_UNMAP && dir != DMA_TO_DEVICE)
-			cflush.op = GNTTAB_CACHE_INVAL;
-		if (op == DMA_MAP) {
-			if (dir == DMA_FROM_DEVICE)
-				cflush.op = GNTTAB_CACHE_INVAL;
-			else
-				cflush.op = GNTTAB_CACHE_CLEAN;
-		}
-		if (cflush.op)
-			HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
+		if (size + cflush.offset > XEN_PAGE_SIZE)
+			cflush.length = XEN_PAGE_SIZE - cflush.offset;
+		else
+			cflush.length = size;
+
+		HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
+
+		handle += cflush.length;
+		size -= cflush.length;
 
-		offset = 0;
-		xen_pfn++;
-		left -= len;
-	} while (left);
+		cflush.offset = 0;
+	} while (size);
 }
 
 static void __xen_dma_page_dev_to_cpu(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_UNMAP);
+	if (dir != DMA_TO_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 }
 
 static void __xen_dma_page_cpu_to_dev(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_MAP);
+	if (dir == DMA_FROM_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
+	else
+		dma_cache_maint(handle, size, GNTTAB_CACHE_CLEAN);
 }
 
 void __xen_dma_map_page(struct device *hwdev, struct page *page,
Signature · 1 line
2.20.1
#6 lore raw

Re: [PATCH 03/11] xen/arm: simplify dma_cache_maint

And this was still buggy I think, it really needs some real Xen/Arm testing which I can't do. Hopefully better version below:

--
From 5ad4b6e291dbb49f65480c9b769414931cbd485a Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <[email protected]>
Date: Wed, 24 Jul 2019 15:26:08 +0200
Subject: xen/arm: simplify dma_cache_maint

Calculate the required operation in the caller, and pass it directly instead of recalculating it for each page, and use simple arithmetics to get from the physical address to Xen page size aligned chunks.

---
 arch/arm/xen/mm.c | 61 ++++++++++++++++-------------------------------
 1 file changed, 21 insertions(+), 40 deletions(-)
arch/arm/xen/mm.c +21 -40
index 90574d89d0d4..2fde161733b0 100644
@@ -35,64 +35,45 @@ unsigned long xen_get_swiotlb_free_pages(unsigned int order)
 	return __get_free_pages(flags, order);
 }
 
-enum dma_cache_op {
-       DMA_UNMAP,
-       DMA_MAP,
-};
 static bool hypercall_cflush = false;
 
-/* functions called by SWIOTLB */
-
-static void dma_cache_maint(dma_addr_t handle, unsigned long offset,
-	size_t size, enum dma_data_direction dir, enum dma_cache_op op)
+/* buffers in highmem or foreign pages cannot cross page boundaries */
+static void dma_cache_maint(dma_addr_t handle, size_t size, u32 op)
 {
 	struct gnttab_cache_flush cflush;
-	unsigned long xen_pfn;
-	size_t left = size;
 
-	xen_pfn = (handle >> XEN_PAGE_SHIFT) + offset / XEN_PAGE_SIZE;
-	offset %= XEN_PAGE_SIZE;
+	cflush.a.dev_bus_addr = handle & XEN_PAGE_MASK;
+	cflush.offset = xen_offset_in_page(handle);
+	cflush.op = op;
 
 	do {
-		size_t len = left;
-	
-		/* buffers in highmem or foreign pages cannot cross page
-		 * boundaries */
-		if (len + offset > XEN_PAGE_SIZE)
-			len = XEN_PAGE_SIZE - offset;
-
-		cflush.op = 0;
-		cflush.a.dev_bus_addr = xen_pfn << XEN_PAGE_SHIFT;
-		cflush.offset = offset;
-		cflush.length = len;
-
-		if (op == DMA_UNMAP && dir != DMA_TO_DEVICE)
-			cflush.op = GNTTAB_CACHE_INVAL;
-		if (op == DMA_MAP) {
-			if (dir == DMA_FROM_DEVICE)
-				cflush.op = GNTTAB_CACHE_INVAL;
-			else
-				cflush.op = GNTTAB_CACHE_CLEAN;
-		}
-		if (cflush.op)
-			HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
+		if (size + cflush.offset > XEN_PAGE_SIZE)
+			cflush.length = XEN_PAGE_SIZE - cflush.offset;
+		else
+			cflush.length = size;
+
+		HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
 
-		offset = 0;
-		xen_pfn++;
-		left -= len;
-	} while (left);
+		cflush.offset = 0;
+		cflush.a.dev_bus_addr += cflush.length;
+		size -= cflush.length;
+	} while (size);
 }
 
 static void __xen_dma_page_dev_to_cpu(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_UNMAP);
+	if (dir != DMA_TO_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 }
 
 static void __xen_dma_page_cpu_to_dev(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_MAP);
+	if (dir == DMA_FROM_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
+	else
+		dma_cache_maint(handle, size, GNTTAB_CACHE_CLEAN);
 }
 
 void __xen_dma_map_page(struct device *hwdev, struct page *page,
Signature · 1 line
2.20.1
maintainer Reviewed-by
#7 lore raw

Re: [PATCH 03/11] xen/arm: simplify dma_cache_maint

Quoted 76 lines · Christoph Hellwig

And this was still buggy I think, it really needs some real Xen/Arm testing which I can't do. Hopefully better version below:

--

Quoted 1 line

From 5ad4b6e291dbb49f65480c9b769414931cbd485a Mon Sep 17 00:00:00 2001

From: Christoph Hellwig <[email protected]>
Date: Wed, 24 Jul 2019 15:26:08 +0200
Subject: xen/arm: simplify dma_cache_maint

Calculate the required operation in the caller, and pass it directly instead of recalculating it for each page, and use simple arithmetics to get from the physical address to Xen page size aligned chunks.

---
 arch/arm/xen/mm.c | 61 ++++++++++++++++-------------------------------
 1 file changed, 21 insertions(+), 40 deletions(-)
arch/arm/xen/mm.c +9 -34
index 90574d89d0d4..2fde161733b0 100644
@@ -35,64 +35,45 @@ unsigned long xen_get_swiotlb_free_pages(unsigned int order)
 	return __get_free_pages(flags, order);
 }
 
-enum dma_cache_op {
-       DMA_UNMAP,
-       DMA_MAP,
-};
 static bool hypercall_cflush = false;
 
-/* functions called by SWIOTLB */
-
-static void dma_cache_maint(dma_addr_t handle, unsigned long offset,
-	size_t size, enum dma_data_direction dir, enum dma_cache_op op)
+/* buffers in highmem or foreign pages cannot cross page boundaries */
+static void dma_cache_maint(dma_addr_t handle, size_t size, u32 op)
 {
 	struct gnttab_cache_flush cflush;
-	unsigned long xen_pfn;
-	size_t left = size;
 
-	xen_pfn = (handle >> XEN_PAGE_SHIFT) + offset / XEN_PAGE_SIZE;
-	offset %= XEN_PAGE_SIZE;
+	cflush.a.dev_bus_addr = handle & XEN_PAGE_MASK;
+	cflush.offset = xen_offset_in_page(handle);
+	cflush.op = op;
 
 	do {
-		size_t len = left;
-	
-		/* buffers in highmem or foreign pages cannot cross page
-		 * boundaries */
-		if (len + offset > XEN_PAGE_SIZE)
-			len = XEN_PAGE_SIZE - offset;
-
-		cflush.op = 0;
-		cflush.a.dev_bus_addr = xen_pfn << XEN_PAGE_SHIFT;
-		cflush.offset = offset;
-		cflush.length = len;
-
-		if (op == DMA_UNMAP && dir != DMA_TO_DEVICE)
-			cflush.op = GNTTAB_CACHE_INVAL;
-		if (op == DMA_MAP) {
-			if (dir == DMA_FROM_DEVICE)
-				cflush.op = GNTTAB_CACHE_INVAL;
-			else
-				cflush.op = GNTTAB_CACHE_CLEAN;
-		}
-		if (cflush.op)
-			HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
+		if (size + cflush.offset > XEN_PAGE_SIZE)
+			cflush.length = XEN_PAGE_SIZE - cflush.offset;
+		else
+			cflush.length = size;

arch/arm/xen/mm.c unsigned long xen_get_swiotlb_free_pages(unsigned int order) · line 59

Quoted 10 lines
arch/arm/xen/mm.c +5 -4
+
+		HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
 
-		offset = 0;
-		xen_pfn++;
-		left -= len;
-	} while (left);
+		cflush.offset = 0;
+		cflush.a.dev_bus_addr += cflush.length;
+		size -= cflush.length;

Yes that's better

Quoted 25 lines
arch/arm/xen/mm.c +7 -2
+	} while (size);
 }
 
 static void __xen_dma_page_dev_to_cpu(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_UNMAP);
+	if (dir != DMA_TO_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 }
 
 static void __xen_dma_page_cpu_to_dev(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_MAP);
+	if (dir == DMA_FROM_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
+	else
+		dma_cache_maint(handle, size, GNTTAB_CACHE_CLEAN);
 }
 
 void __xen_dma_map_page(struct device *hwdev, struct page *page,
Signature · 1 line
2.20.1
#8 lore raw

Re: [PATCH 03/11] xen/arm: simplify dma_cache_maint

In reply to Christoph Hellwig
Quoted 57 lines · Christoph Hellwig

Calculate the required operation in the caller, and pass it directly instead of recalculating it for each page, and use simple arithmetics to get from the physical address to Xen page size aligned chunks.

---
 arch/arm/xen/mm.c | 62 +++++++++++++++++------------------------------
 1 file changed, 22 insertions(+), 40 deletions(-)
arch/arm/xen/mm.c +5 -28
index 90574d89d0d4..14210ebdea1a 100644
@@ -35,64 +35,46 @@ unsigned long xen_get_swiotlb_free_pages(unsigned int order)
 	return __get_free_pages(flags, order);
 }
 
-enum dma_cache_op {
-       DMA_UNMAP,
-       DMA_MAP,
-};
 static bool hypercall_cflush = false;
 
-/* functions called by SWIOTLB */
-
-static void dma_cache_maint(dma_addr_t handle, unsigned long offset,
-	size_t size, enum dma_data_direction dir, enum dma_cache_op op)
+/* buffers in highmem or foreign pages cannot cross page boundaries */
+static void dma_cache_maint(dma_addr_t handle, size_t size, u32 op)
 {
 	struct gnttab_cache_flush cflush;
-	unsigned long xen_pfn;
-	size_t left = size;
 
-	xen_pfn = (handle >> XEN_PAGE_SHIFT) + offset / XEN_PAGE_SIZE;
-	offset %= XEN_PAGE_SIZE;
+	cflush.a.dev_bus_addr = handle & XEN_PAGE_MASK;
+	cflush.offset = xen_offset_in_page(handle);
+	cflush.op = op;
 
 	do {
-		size_t len = left;
-	
-		/* buffers in highmem or foreign pages cannot cross page
-		 * boundaries */
-		if (len + offset > XEN_PAGE_SIZE)
-			len = XEN_PAGE_SIZE - offset;
-
-		cflush.op = 0;
-		cflush.a.dev_bus_addr = xen_pfn << XEN_PAGE_SHIFT;
-		cflush.offset = offset;
-		cflush.length = len;
-
-		if (op == DMA_UNMAP && dir != DMA_TO_DEVICE)
-			cflush.op = GNTTAB_CACHE_INVAL;
-		if (op == DMA_MAP) {
-			if (dir == DMA_FROM_DEVICE)

arch/arm/xen/mm.c unsigned long xen_get_swiotlb_free_pages(unsigned int order) · line 53

Quoted 10 lines
arch/arm/xen/mm.c +4 -6
-				cflush.op = GNTTAB_CACHE_INVAL;
-			else
-				cflush.op = GNTTAB_CACHE_CLEAN;
-		}
-		if (cflush.op)
-			HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
+		if (size + cflush.offset > XEN_PAGE_SIZE)
+			cflush.length = XEN_PAGE_SIZE - cflush.offset;
+		else
+			cflush.length = size;

isn't it missing a:

  cflush.a.dev_bus_addr = handle & XEN_PAGE_MASK;

here?

Quoted 35 lines
arch/arm/xen/mm.c +12 -6
+		HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
+
+		handle += cflush.length;
+		size -= cflush.length;
 
-		offset = 0;
-		xen_pfn++;
-		left -= len;
-	} while (left);
+		cflush.offset = 0;
+	} while (size);
 }
 
 static void __xen_dma_page_dev_to_cpu(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_UNMAP);
+	if (dir != DMA_TO_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 }
 
 static void __xen_dma_page_cpu_to_dev(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_MAP);
+	if (dir == DMA_FROM_DEVICE)
+		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
+	else
+		dma_cache_maint(handle, size, GNTTAB_CACHE_CLEAN);
 }
 
 void __xen_dma_map_page(struct device *hwdev, struct page *page,
Signature · 1 line
2.20.1
#9 lore raw

PATCH 4/11 xen/arm: remove xen_dma_ops

In reply to Christoph Hellwig

arm and arm64 can just use xen_swiotlb_dma_ops directly like x86, no need for a pointer indirection.

---
 arch/arm/mm/dma-mapping.c    | 3 ++-
 arch/arm/xen/mm.c            | 4 ----
 arch/arm64/mm/dma-mapping.c  | 3 ++-
 include/xen/arm/hypervisor.h | 2 --
 4 files changed, 4 insertions(+), 8 deletions(-)
arch/arm/mm/dma-mapping.c +2 -1
index 738097396445..2661cad36359 100644
@@ -35,6 +35,7 @@
 #include <asm/mach/map.h>
 #include <asm/system_info.h>
 #include <asm/dma-contiguous.h>
+#include <xen/swiotlb-xen.h>
 
 #include "dma.h"
 #include "mm.h"
@@ -2360,7 +2361,7 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 
 #ifdef CONFIG_XEN
 	if (xen_initial_domain())
-		dev->dma_ops = xen_dma_ops;
+		dev->dma_ops = &xen_swiotlb_dma_ops;
 #endif
 	dev->archdata.dma_ops_setup = true;
 }
arch/arm/xen/mm.c +0 -4
index 14210ebdea1a..9b3a6c0ca681 100644
@@ -163,16 +163,12 @@ void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 }
 EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
 
-const struct dma_map_ops *xen_dma_ops;
-EXPORT_SYMBOL(xen_dma_ops);
-
 int __init xen_mm_init(void)
 {
 	struct gnttab_cache_flush cflush;
 	if (!xen_initial_domain())
 		return 0;
 	xen_swiotlb_init(1, false);
-	xen_dma_ops = &xen_swiotlb_dma_ops;
 
 	cflush.op = 0;
 	cflush.a.dev_bus_addr = 0;
arch/arm64/mm/dma-mapping.c +2 -1
index bd2b039f43a6..4b244a037349 100644
@@ -8,6 +8,7 @@
 #include <linux/cache.h>
 #include <linux/dma-noncoherent.h>
 #include <linux/dma-iommu.h>
+#include <xen/swiotlb-xen.h>
 
 #include <asm/cacheflush.h>
 
@@ -64,6 +65,6 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 
 #ifdef CONFIG_XEN
 	if (xen_initial_domain())
-		dev->dma_ops = xen_dma_ops;
+		dev->dma_ops = &xen_swiotlb_dma_ops;
 #endif
 }
include/xen/arm/hypervisor.h +0 -2
index 2982571f7cc1..43ef24dd030e 100644
@@ -19,8 +19,6 @@ static inline enum paravirt_lazy_mode paravirt_get_lazy_mode(void)
 	return PARAVIRT_LAZY_NONE;
 }
 
-extern const struct dma_map_ops *xen_dma_ops;
-
 #ifdef CONFIG_XEN
 void __init xen_early_init(void);
 #else
Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 4/11
#10 lore raw

Re: [PATCH 04/11] xen/arm: remove xen_dma_ops

Quoted 5 lines · Christoph Hellwig

arm and arm64 can just use xen_swiotlb_dma_ops directly like x86, no need for a pointer indirection.

Quoted 85 lines
---
 arch/arm/mm/dma-mapping.c    | 3 ++-
 arch/arm/xen/mm.c            | 4 ----
 arch/arm64/mm/dma-mapping.c  | 3 ++-
 include/xen/arm/hypervisor.h | 2 --
 4 files changed, 4 insertions(+), 8 deletions(-)
arch/arm/mm/dma-mapping.c +2 -1
index 738097396445..2661cad36359 100644
@@ -35,6 +35,7 @@
 #include <asm/mach/map.h>
 #include <asm/system_info.h>
 #include <asm/dma-contiguous.h>
+#include <xen/swiotlb-xen.h>
 
 #include "dma.h"
 #include "mm.h"
@@ -2360,7 +2361,7 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 
 #ifdef CONFIG_XEN
 	if (xen_initial_domain())
-		dev->dma_ops = xen_dma_ops;
+		dev->dma_ops = &xen_swiotlb_dma_ops;
 #endif
 	dev->archdata.dma_ops_setup = true;
 }
arch/arm/xen/mm.c +0 -4
index 14210ebdea1a..9b3a6c0ca681 100644
@@ -163,16 +163,12 @@ void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 }
 EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
 
-const struct dma_map_ops *xen_dma_ops;
-EXPORT_SYMBOL(xen_dma_ops);
-
 int __init xen_mm_init(void)
 {
 	struct gnttab_cache_flush cflush;
 	if (!xen_initial_domain())
 		return 0;
 	xen_swiotlb_init(1, false);
-	xen_dma_ops = &xen_swiotlb_dma_ops;
 
 	cflush.op = 0;
 	cflush.a.dev_bus_addr = 0;
arch/arm64/mm/dma-mapping.c +2 -1
index bd2b039f43a6..4b244a037349 100644
@@ -8,6 +8,7 @@
 #include <linux/cache.h>
 #include <linux/dma-noncoherent.h>
 #include <linux/dma-iommu.h>
+#include <xen/swiotlb-xen.h>
 
 #include <asm/cacheflush.h>
 
@@ -64,6 +65,6 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 
 #ifdef CONFIG_XEN
 	if (xen_initial_domain())
-		dev->dma_ops = xen_dma_ops;
+		dev->dma_ops = &xen_swiotlb_dma_ops;
 #endif
 }
include/xen/arm/hypervisor.h +0 -2
index 2982571f7cc1..43ef24dd030e 100644
@@ -19,8 +19,6 @@ static inline enum paravirt_lazy_mode paravirt_get_lazy_mode(void)
 	return PARAVIRT_LAZY_NONE;
 }
 
-extern const struct dma_map_ops *xen_dma_ops;
-
 #ifdef CONFIG_XEN
 void __init xen_early_init(void);
 #else
Signature · 1 line
2.20.1
#11 lore raw

PATCH 5/11 xen: remove the exports for xen_{create,destroy}_contiguous_region

In reply to Christoph Hellwig

These routines are only used by swiotlb-xen, which cannot be modular.

---
 arch/arm/xen/mm.c     | 2 --
 arch/x86/xen/mmu_pv.c | 2 --
 2 files changed, 4 deletions(-)
arch/arm/xen/mm.c +0 -2
index 9b3a6c0ca681..b7d53415532b 100644
@@ -155,13 +155,11 @@ int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
 	*dma_handle = pstart;
 	return 0;
 }
-EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
 
 void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 {
 	return;
 }
-EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
 
 int __init xen_mm_init(void)
 {
arch/x86/xen/mmu_pv.c +0 -2
index 26e8b326966d..c8dbee62ec2a 100644
@@ -2625,7 +2625,6 @@ int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
 	*dma_handle = virt_to_machine(vstart).maddr;
 	return success ? 0 : -ENOMEM;
 }
-EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
 
 void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 {
@@ -2660,7 +2659,6 @@ void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 
 	spin_unlock_irqrestore(&xen_reservation_lock, flags);
 }
-EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
 
 static noinline void xen_flush_tlb_all(void)
 {
Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 5/11
#12 lore raw

Re: [PATCH 05/11] xen: remove the exports for xen_{create,destroy}_contiguous_region

Quoted 3 lines · Christoph Hellwig

These routines are only used by swiotlb-xen, which cannot be modular.

Quoted 46 lines
---
 arch/arm/xen/mm.c     | 2 --
 arch/x86/xen/mmu_pv.c | 2 --
 2 files changed, 4 deletions(-)
arch/arm/xen/mm.c +0 -2
index 9b3a6c0ca681..b7d53415532b 100644
@@ -155,13 +155,11 @@ int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
 	*dma_handle = pstart;
 	return 0;
 }
-EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
 
 void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 {
 	return;
 }
-EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
 
 int __init xen_mm_init(void)
 {
arch/x86/xen/mmu_pv.c +0 -2
index 26e8b326966d..c8dbee62ec2a 100644
@@ -2625,7 +2625,6 @@ int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
 	*dma_handle = virt_to_machine(vstart).maddr;
 	return success ? 0 : -ENOMEM;
 }
-EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
 
 void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 {
@@ -2660,7 +2659,6 @@ void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 
 	spin_unlock_irqrestore(&xen_reservation_lock, flags);
 }
-EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
 
 static noinline void xen_flush_tlb_all(void)
 {
Signature · 1 line
2.20.1
#13 lore raw

PATCH 6/11 swiotlb-xen: always use dma-direct helpers to alloc coherent pages

In reply to Christoph Hellwig

x86 currently calls alloc_pages, but using dma-direct works as well there, with the added benefit of using the CMA pool if available. The biggest advantage is of course to remove a pointless bit of architecture specific code.

---
 arch/x86/include/asm/xen/page-coherent.h | 16 ----------------
 drivers/xen/swiotlb-xen.c                |  7 +++----
 include/xen/arm/page-coherent.h          | 12 ------------
 3 files changed, 3 insertions(+), 32 deletions(-)
arch/x86/include/asm/xen/page-coherent.h +0 -16
index 116777e7f387..8ee33c5edded 100644
@@ -5,22 +5,6 @@
 #include <asm/page.h>
 #include <linux/dma-mapping.h>
 
-static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flags,
-		unsigned long attrs)
-{
-	void *vstart = (void*)__get_free_pages(flags, get_order(size));
-	*dma_handle = virt_to_phys(vstart);
-	return vstart;
-}
-
-static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
-		void *cpu_addr, dma_addr_t dma_handle,
-		unsigned long attrs)
-{
-	free_pages((unsigned long) cpu_addr, get_order(size));
-}
-
 static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, unsigned long attrs) { }
drivers/xen/swiotlb-xen.c +3 -4
index b8808677ae1d..f9dd4cb6e4b3 100644
@@ -299,8 +299,7 @@ xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	 * address. In fact on ARM virt_to_phys only works for kernel direct
 	 * mapped RAM memory. Also see comment below.
 	 */
-	ret = xen_alloc_coherent_pages(hwdev, size, dma_handle, flags, attrs);
-
+	ret = dma_direct_alloc(hwdev, size, dma_handle, flags, attrs);
 	if (!ret)
 		return ret;
 
@@ -319,7 +318,7 @@ xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	else {
 		if (xen_create_contiguous_region(phys, order,
 						 fls64(dma_mask), dma_handle) != 0) {
-			xen_free_coherent_pages(hwdev, size, ret, (dma_addr_t)phys, attrs);
+			dma_direct_free(hwdev, size, ret, (dma_addr_t)phys, attrs);
 			return NULL;
 		}
 		SetPageXenRemapped(virt_to_page(ret));
@@ -351,7 +350,7 @@ xen_swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr,
 	    TestClearPageXenRemapped(virt_to_page(vaddr)))
 		xen_destroy_contiguous_region(phys, order);
 
-	xen_free_coherent_pages(hwdev, size, vaddr, (dma_addr_t)phys, attrs);
+	dma_direct_free(hwdev, size, vaddr, (dma_addr_t)phys, attrs);
 }
 
 /*
include/xen/arm/page-coherent.h +0 -12
index a840d6949a87..0e244f4fec1a 100644
@@ -16,18 +16,6 @@ void __xen_dma_sync_single_for_cpu(struct device *hwdev,
 void __xen_dma_sync_single_for_device(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir);
 
-static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flags, unsigned long attrs)
-{
-	return dma_direct_alloc(hwdev, size, dma_handle, flags, attrs);
-}
-
-static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
-		void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs)
-{
-	dma_direct_free(hwdev, size, cpu_addr, dma_handle, attrs);
-}
-
 static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
Signature · 1 line
2.20.1
#14 lore raw

Re: [PATCH 06/11] swiotlb-xen: always use dma-direct helpers to alloc coherent pages

+ Boris, Juergen

Quoted 17 lines · Christoph Hellwig

x86 currently calls alloc_pages, but using dma-direct works as well there, with the added benefit of using the CMA pool if available. The biggest advantage is of course to remove a pointless bit of architecture specific code.

---
 arch/x86/include/asm/xen/page-coherent.h | 16 ----------------
 drivers/xen/swiotlb-xen.c                |  7 +++----
 include/xen/arm/page-coherent.h          | 12 ------------
 3 files changed, 3 insertions(+), 32 deletions(-)
arch/x86/include/asm/xen/page-coherent.h +0 -0
index 116777e7f387..8ee33c5edded 100644

arch/x86/include/asm/xen/page-coherent.h line 13

Quoted 9 lines
arch/x86/include/asm/xen/page-coherent.h +0 -6
@@ -5,22 +5,6 @@
 #include <asm/page.h>
 #include <linux/dma-mapping.h>
 
-static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flags,
-		unsigned long attrs)
-{
-	void *vstart = (void*)__get_free_pages(flags, get_order(size));
-	*dma_handle = virt_to_phys(vstart);

This is where we need Boris and Juergen's opinion. From an ARM POV it looks OK.

Quoted 71 lines
arch/x86/include/asm/xen/page-coherent.h +0 -10
-	return vstart;
-}
-
-static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
-		void *cpu_addr, dma_addr_t dma_handle,
-		unsigned long attrs)
-{
-	free_pages((unsigned long) cpu_addr, get_order(size));
-}
-
 static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, unsigned long attrs) { }
drivers/xen/swiotlb-xen.c +3 -4
index b8808677ae1d..f9dd4cb6e4b3 100644
@@ -299,8 +299,7 @@ xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	 * address. In fact on ARM virt_to_phys only works for kernel direct
 	 * mapped RAM memory. Also see comment below.
 	 */
-	ret = xen_alloc_coherent_pages(hwdev, size, dma_handle, flags, attrs);
-
+	ret = dma_direct_alloc(hwdev, size, dma_handle, flags, attrs);
 	if (!ret)
 		return ret;
 
@@ -319,7 +318,7 @@ xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	else {
 		if (xen_create_contiguous_region(phys, order,
 						 fls64(dma_mask), dma_handle) != 0) {
-			xen_free_coherent_pages(hwdev, size, ret, (dma_addr_t)phys, attrs);
+			dma_direct_free(hwdev, size, ret, (dma_addr_t)phys, attrs);
 			return NULL;
 		}
 		SetPageXenRemapped(virt_to_page(ret));
@@ -351,7 +350,7 @@ xen_swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr,
 	    TestClearPageXenRemapped(virt_to_page(vaddr)))
 		xen_destroy_contiguous_region(phys, order);
 
-	xen_free_coherent_pages(hwdev, size, vaddr, (dma_addr_t)phys, attrs);
+	dma_direct_free(hwdev, size, vaddr, (dma_addr_t)phys, attrs);
 }
 
 /*
include/xen/arm/page-coherent.h +0 -12
index a840d6949a87..0e244f4fec1a 100644
@@ -16,18 +16,6 @@ void __xen_dma_sync_single_for_cpu(struct device *hwdev,
 void __xen_dma_sync_single_for_device(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir);
 
-static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flags, unsigned long attrs)
-{
-	return dma_direct_alloc(hwdev, size, dma_handle, flags, attrs);
-}
-
-static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
-		void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs)
-{
-	dma_direct_free(hwdev, size, cpu_addr, dma_handle, attrs);
-}
-
 static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
Signature · 1 line
2.20.1
#15 lore raw

PATCH 7/11 swiotlb-xen: use the same foreign page check everywhere

In reply to Christoph Hellwig

xen_dma_map_page uses a different and more complicated check for foreign pages than the other three cache maintainance helpers. Switch it to the simpler pfn_valid method a well, and document the scheme with a single improved comment in xen_dma_map_page.

---
 include/xen/arm/page-coherent.h | 31 +++++++++----------------------
 1 file changed, 9 insertions(+), 22 deletions(-)
include/xen/arm/page-coherent.h +9 -22
index 0e244f4fec1a..07c104dbc21f 100644
@@ -41,23 +41,17 @@ static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, unsigned long attrs)
 {
-	unsigned long page_pfn = page_to_xen_pfn(page);
-	unsigned long dev_pfn = XEN_PFN_DOWN(dev_addr);
-	unsigned long compound_pages =
-		(1<<compound_order(page)) * XEN_PFN_PER_PAGE;
-	bool local = (page_pfn <= dev_pfn) &&
-		(dev_pfn - page_pfn < compound_pages);
+	unsigned long pfn = PFN_DOWN(dev_addr);
 
 	/*
-	 * Dom0 is mapped 1:1, while the Linux page can span across
-	 * multiple Xen pages, it's not possible for it to contain a
-	 * mix of local and foreign Xen pages. So if the first xen_pfn
-	 * == mfn the page is local otherwise it's a foreign page
-	 * grant-mapped in dom0. If the page is local we can safely
-	 * call the native dma_ops function, otherwise we call the xen
-	 * specific function.
+	 * Dom0 is mapped 1:1, and while the Linux page can span across multiple
+	 * Xen pages, it is not possible for it to contain a mix of local and
+	 * foreign Xen pages.  Calling pfn_valid on a foreign mfn will always
+	 * return false, so if pfn_valid returns true the pages is local and we
+	 * can use the native dma-direct functions, otherwise we call the Xen
+	 * specific version.
 	 */
-	if (local)
+	if (pfn_valid(pfn))
 		dma_direct_map_page(hwdev, page, offset, size, dir, attrs);
 	else
 		__xen_dma_map_page(hwdev, page, dev_addr, offset, size, dir, attrs);
@@ -67,14 +61,7 @@ static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
 	unsigned long pfn = PFN_DOWN(handle);
-	/*
-	 * Dom0 is mapped 1:1, while the Linux page can be spanned accross
-	 * multiple Xen page, it's not possible to have a mix of local and
-	 * foreign Xen page. Dom0 is mapped 1:1, so calling pfn_valid on a
-	 * foreign mfn will always return false. If the page is local we can
-	 * safely call the native dma_ops function, otherwise we call the xen
-	 * specific function.
-	 */
+
 	if (pfn_valid(pfn))
 		dma_direct_unmap_page(hwdev, handle, size, dir, attrs);
 	else
Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 7/11
#16 lore raw

Re: [PATCH 07/11] swiotlb-xen: use the same foreign page check everywhere

Quoted 6 lines · Christoph Hellwig

xen_dma_map_page uses a different and more complicated check for foreign pages than the other three cache maintainance helpers. Switch it to the simpler pfn_valid method a well, and document the scheme with a single improved comment in xen_dma_map_page.

Quoted 59 lines
---
 include/xen/arm/page-coherent.h | 31 +++++++++----------------------
 1 file changed, 9 insertions(+), 22 deletions(-)
include/xen/arm/page-coherent.h +9 -22
index 0e244f4fec1a..07c104dbc21f 100644
@@ -41,23 +41,17 @@ static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, unsigned long attrs)
 {
-	unsigned long page_pfn = page_to_xen_pfn(page);
-	unsigned long dev_pfn = XEN_PFN_DOWN(dev_addr);
-	unsigned long compound_pages =
-		(1<<compound_order(page)) * XEN_PFN_PER_PAGE;
-	bool local = (page_pfn <= dev_pfn) &&
-		(dev_pfn - page_pfn < compound_pages);
+	unsigned long pfn = PFN_DOWN(dev_addr);
 
 	/*
-	 * Dom0 is mapped 1:1, while the Linux page can span across
-	 * multiple Xen pages, it's not possible for it to contain a
-	 * mix of local and foreign Xen pages. So if the first xen_pfn
-	 * == mfn the page is local otherwise it's a foreign page
-	 * grant-mapped in dom0. If the page is local we can safely
-	 * call the native dma_ops function, otherwise we call the xen
-	 * specific function.
+	 * Dom0 is mapped 1:1, and while the Linux page can span across multiple
+	 * Xen pages, it is not possible for it to contain a mix of local and
+	 * foreign Xen pages.  Calling pfn_valid on a foreign mfn will always
+	 * return false, so if pfn_valid returns true the pages is local and we
+	 * can use the native dma-direct functions, otherwise we call the Xen
+	 * specific version.
 	 */
-	if (local)
+	if (pfn_valid(pfn))
 		dma_direct_map_page(hwdev, page, offset, size, dir, attrs);
 	else
 		__xen_dma_map_page(hwdev, page, dev_addr, offset, size, dir, attrs);
@@ -67,14 +61,7 @@ static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
 	unsigned long pfn = PFN_DOWN(handle);
-	/*
-	 * Dom0 is mapped 1:1, while the Linux page can be spanned accross
-	 * multiple Xen page, it's not possible to have a mix of local and
-	 * foreign Xen page. Dom0 is mapped 1:1, so calling pfn_valid on a
-	 * foreign mfn will always return false. If the page is local we can
-	 * safely call the native dma_ops function, otherwise we call the xen
-	 * specific function.
-	 */
+
 	if (pfn_valid(pfn))
 		dma_direct_unmap_page(hwdev, handle, size, dir, attrs);
 	else
Signature · 1 line
2.20.1
#17 lore raw

PATCH 8/11 swiotlb-xen: simplify cache maintainance

In reply to Christoph Hellwig

Now that we know we always have the dma-noncoherent.h helpers available if we are on an architecture with support for non-coherent devices, we can just call them directly, and remove the calls to the dma-direct routines, including the fact that we call the dma_direct_map_page routines but ignore the value returned from it. Instead we now have Xen wrappers for the arch_sync_dma_for_{device,cpu} helpers that call the special Xen versions of those routines for foreign pages.

Note that the new helpers get the physical address passed in addition to the dma address to avoid another translation for the local cache maintainance. The pfn_valid checks remain on the dma address as in the old code, even if that looks a little funny.

---
 arch/arm/xen/mm.c                        | 64 ++++++----------------
 arch/x86/include/asm/xen/page-coherent.h | 11 ----
 drivers/xen/swiotlb-xen.c                | 20 +++----
 include/xen/arm/page-coherent.h          | 69 ++----------------------
 4 files changed, 31 insertions(+), 133 deletions(-)
arch/arm/xen/mm.c +17 -47
index b7d53415532b..7096652f5a1e 100644
@@ -61,63 +61,33 @@ static void dma_cache_maint(dma_addr_t handle, size_t size, u32 op)
 	} while (size);
 }
 
-static void __xen_dma_page_dev_to_cpu(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir)
+/*
+ * Dom0 is mapped 1:1, and while the Linux page can span across multiple Xen
+ * pages, it is not possible for it to contain a mix of local and foreign Xen
+ * pages.  Calling pfn_valid on a foreign mfn will always return false, so if
+ * pfn_valid returns true the pages is local and we can use the native
+ * dma-direct functions, otherwise we call the Xen specific version.
+ */
+void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
 {
-	if (dir != DMA_TO_DEVICE)
+	if (pfn_valid(PFN_DOWN(handle)))
+		arch_sync_dma_for_cpu(dev, paddr, size, dir);
+	else if (dir != DMA_TO_DEVICE)
 		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 }
 
-static void __xen_dma_page_cpu_to_dev(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir)
+void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
 {
-	if (dir == DMA_FROM_DEVICE)
+	if (pfn_valid(PFN_DOWN(handle)))
+		arch_sync_dma_for_device(dev, paddr, size, dir);
+	else if (dir == DMA_FROM_DEVICE)
 		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 	else
 		dma_cache_maint(handle, size, GNTTAB_CACHE_CLEAN);
 }
 
-void __xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs)
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
-		return;
-
-	__xen_dma_page_cpu_to_dev(hwdev, dev_addr, size, dir);
-}
-
-void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir,
-		unsigned long attrs)
-
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
-		return;
-
-	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
-}
-
-void __xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
-}
-
-void __xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	__xen_dma_page_cpu_to_dev(hwdev, handle, size, dir);
-}
-
 bool xen_arch_need_swiotlb(struct device *dev,
 			   phys_addr_t phys,
 			   dma_addr_t dev_addr)
arch/x86/include/asm/xen/page-coherent.h +0 -11
index 8ee33c5edded..c9c8398a31ff 100644
@@ -2,17 +2,6 @@
 #ifndef _ASM_X86_XEN_PAGE_COHERENT_H
 #define _ASM_X86_XEN_PAGE_COHERENT_H
 
-#include <asm/page.h>
-#include <linux/dma-mapping.h>
-
-static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs) { }
-
-static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir,
-		unsigned long attrs) { }
-
 static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
 
drivers/xen/swiotlb-xen.c +10 -10
index f9dd4cb6e4b3..a642e284f1e2 100644
@@ -28,6 +28,7 @@
 
 #include <linux/memblock.h>
 #include <linux/dma-direct.h>
+#include <linux/dma-noncoherent.h>
 #include <linux/export.h>
 #include <xen/swiotlb-xen.h>
 #include <xen/page.h>
@@ -390,6 +391,7 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 	if (map == (phys_addr_t)DMA_MAPPING_ERROR)
 		return DMA_MAPPING_ERROR;
 
+	phys = map;
 	dev_addr = xen_phys_to_bus(map);
 
 	/*
@@ -401,14 +403,9 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 		return DMA_MAPPING_ERROR;
 	}
 
-	page = pfn_to_page(map >> PAGE_SHIFT);
-	offset = map & ~PAGE_MASK;
 done:
-	/*
-	 * we are not interested in the dma_addr returned by xen_dma_map_page,
-	 * only in the potential cache flushes executed by the function.
-	 */
-	xen_dma_map_page(dev, page, dev_addr, offset, size, dir, attrs);
+	if (!dev_is_dma_coherent(dev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))
+		xen_dma_sync_for_device(dev, dev_addr, phys, size, dir);
 	return dev_addr;
 }
 
@@ -428,7 +425,8 @@ static void xen_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
 
 	BUG_ON(dir == DMA_NONE);
 
-	xen_dma_unmap_page(hwdev, dev_addr, size, dir, attrs);
+	if (!dev_is_dma_coherent(hwdev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))
+		xen_dma_sync_for_cpu(hwdev, dev_addr, paddr, size, dir);
 
 	/* NOTE: We use dev_addr here, not paddr! */
 	if (is_xen_swiotlb_buffer(dev_addr))
@@ -448,7 +446,8 @@ xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr,
 {
 	phys_addr_t paddr = xen_bus_to_phys(dma_addr);
 
-	xen_dma_sync_single_for_cpu(dev, dma_addr, size, dir);
+	if (!dev_is_dma_coherent(dev))
+		xen_dma_sync_for_cpu(dev, dma_addr, paddr, size, dir);
 
 	if (is_xen_swiotlb_buffer(dma_addr))
 		swiotlb_tbl_sync_single(dev, paddr, size, dir, SYNC_FOR_CPU);
@@ -463,7 +462,8 @@ xen_swiotlb_sync_single_for_device(struct device *dev, dma_addr_t dma_addr,
 	if (is_xen_swiotlb_buffer(dma_addr))
 		swiotlb_tbl_sync_single(dev, paddr, size, dir, SYNC_FOR_DEVICE);
 
-	xen_dma_sync_single_for_device(dev, dma_addr, size, dir);
+	if (!dev_is_dma_coherent(dev))
+		xen_dma_sync_for_device(dev, dma_addr, paddr, size, dir);
 }
 
 /*
include/xen/arm/page-coherent.h +4 -65

74 lines not on this page · raw message

Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 8/11
#18 lore raw

Re: [PATCH 08/11] swiotlb-xen: simplify cache maintainance

Quoted 13 lines · Christoph Hellwig

Now that we know we always have the dma-noncoherent.h helpers available if we are on an architecture with support for non-coherent devices, we can just call them directly, and remove the calls to the dma-direct routines, including the fact that we call the dma_direct_map_page routines but ignore the value returned from it. Instead we now have Xen wrappers for the arch_sync_dma_for_{device,cpu} helpers that call the special Xen versions of those routines for foreign pages.

Note that the new helpers get the physical address passed in addition to the dma address to avoid another translation for the local cache maintainance. The pfn_valid checks remain on the dma address as in the old code, even if that looks a little funny.

Quoted 8 lines
---
 arch/arm/xen/mm.c                        | 64 ++++++----------------
 arch/x86/include/asm/xen/page-coherent.h | 11 ----
 drivers/xen/swiotlb-xen.c                | 20 +++----
 include/xen/arm/page-coherent.h          | 69 ++----------------------
 4 files changed, 31 insertions(+), 133 deletions(-)

WOW nice! Now I really can see why this series was worth doing :-)

Quoted 256 lines
arch/arm/xen/mm.c +17 -47
index b7d53415532b..7096652f5a1e 100644
@@ -61,63 +61,33 @@ static void dma_cache_maint(dma_addr_t handle, size_t size, u32 op)
 	} while (size);
 }
 
-static void __xen_dma_page_dev_to_cpu(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir)
+/*
+ * Dom0 is mapped 1:1, and while the Linux page can span across multiple Xen
+ * pages, it is not possible for it to contain a mix of local and foreign Xen
+ * pages.  Calling pfn_valid on a foreign mfn will always return false, so if
+ * pfn_valid returns true the pages is local and we can use the native
+ * dma-direct functions, otherwise we call the Xen specific version.
+ */
+void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
 {
-	if (dir != DMA_TO_DEVICE)
+	if (pfn_valid(PFN_DOWN(handle)))
+		arch_sync_dma_for_cpu(dev, paddr, size, dir);
+	else if (dir != DMA_TO_DEVICE)
 		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 }
 
-static void __xen_dma_page_cpu_to_dev(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir)
+void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
 {
-	if (dir == DMA_FROM_DEVICE)
+	if (pfn_valid(PFN_DOWN(handle)))
+		arch_sync_dma_for_device(dev, paddr, size, dir);
+	else if (dir == DMA_FROM_DEVICE)
 		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 	else
 		dma_cache_maint(handle, size, GNTTAB_CACHE_CLEAN);
 }
 
-void __xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs)
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
-		return;
-
-	__xen_dma_page_cpu_to_dev(hwdev, dev_addr, size, dir);
-}
-
-void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir,
-		unsigned long attrs)
-
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
-		return;
-
-	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
-}
-
-void __xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
-}
-
-void __xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	if (dev_is_dma_coherent(hwdev))
-		return;
-	__xen_dma_page_cpu_to_dev(hwdev, handle, size, dir);
-}
-
 bool xen_arch_need_swiotlb(struct device *dev,
 			   phys_addr_t phys,
 			   dma_addr_t dev_addr)
arch/x86/include/asm/xen/page-coherent.h +0 -11
index 8ee33c5edded..c9c8398a31ff 100644
@@ -2,17 +2,6 @@
 #ifndef _ASM_X86_XEN_PAGE_COHERENT_H
 #define _ASM_X86_XEN_PAGE_COHERENT_H
 
-#include <asm/page.h>
-#include <linux/dma-mapping.h>
-
-static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs) { }
-
-static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir,
-		unsigned long attrs) { }
-
 static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
 		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
 
drivers/xen/swiotlb-xen.c +10 -10
index f9dd4cb6e4b3..a642e284f1e2 100644
@@ -28,6 +28,7 @@
 
 #include <linux/memblock.h>
 #include <linux/dma-direct.h>
+#include <linux/dma-noncoherent.h>
 #include <linux/export.h>
 #include <xen/swiotlb-xen.h>
 #include <xen/page.h>
@@ -390,6 +391,7 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 	if (map == (phys_addr_t)DMA_MAPPING_ERROR)
 		return DMA_MAPPING_ERROR;
 
+	phys = map;
 	dev_addr = xen_phys_to_bus(map);
 
 	/*
@@ -401,14 +403,9 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 		return DMA_MAPPING_ERROR;
 	}
 
-	page = pfn_to_page(map >> PAGE_SHIFT);
-	offset = map & ~PAGE_MASK;
 done:
-	/*
-	 * we are not interested in the dma_addr returned by xen_dma_map_page,
-	 * only in the potential cache flushes executed by the function.
-	 */
-	xen_dma_map_page(dev, page, dev_addr, offset, size, dir, attrs);
+	if (!dev_is_dma_coherent(dev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))
+		xen_dma_sync_for_device(dev, dev_addr, phys, size, dir);
 	return dev_addr;
 }
 
@@ -428,7 +425,8 @@ static void xen_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
 
 	BUG_ON(dir == DMA_NONE);
 
-	xen_dma_unmap_page(hwdev, dev_addr, size, dir, attrs);
+	if (!dev_is_dma_coherent(hwdev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))
+		xen_dma_sync_for_cpu(hwdev, dev_addr, paddr, size, dir);
 
 	/* NOTE: We use dev_addr here, not paddr! */
 	if (is_xen_swiotlb_buffer(dev_addr))
@@ -448,7 +446,8 @@ xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr,
 {
 	phys_addr_t paddr = xen_bus_to_phys(dma_addr);
 
-	xen_dma_sync_single_for_cpu(dev, dma_addr, size, dir);
+	if (!dev_is_dma_coherent(dev))
+		xen_dma_sync_for_cpu(dev, dma_addr, paddr, size, dir);
 
 	if (is_xen_swiotlb_buffer(dma_addr))
 		swiotlb_tbl_sync_single(dev, paddr, size, dir, SYNC_FOR_CPU);
@@ -463,7 +462,8 @@ xen_swiotlb_sync_single_for_device(struct device *dev, dma_addr_t dma_addr,
 	if (is_xen_swiotlb_buffer(dma_addr))
 		swiotlb_tbl_sync_single(dev, paddr, size, dir, SYNC_FOR_DEVICE);
 
-	xen_dma_sync_single_for_device(dev, dma_addr, size, dir);
+	if (!dev_is_dma_coherent(dev))
+		xen_dma_sync_for_device(dev, dma_addr, paddr, size, dir);
 }
 
 /*
include/xen/arm/page-coherent.h +4 -65
index 07c104dbc21f..635492d41ebe 100644
@@ -2,70 +2,9 @@
 #ifndef _XEN_ARM_PAGE_COHERENT_H
 #define _XEN_ARM_PAGE_COHERENT_H
 
-#include <linux/dma-mapping.h>
-#include <asm/page.h>
-
-void __xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs);
-void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir,
-		unsigned long attrs);
-void __xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir);
-void __xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir);
-
-static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-
-	if (pfn_valid(pfn))
-		dma_direct_sync_single_for_cpu(hwdev, handle, size, dir);
-	else
-		__xen_dma_sync_single_for_cpu(hwdev, handle, size, dir);
-}
-
-static inline void xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-	if (pfn_valid(pfn))
-		dma_direct_sync_single_for_device(hwdev, handle, size, dir);
-	else
-		__xen_dma_sync_single_for_device(hwdev, handle, size, dir);
-}
-
-static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
-	     dma_addr_t dev_addr, unsigned long offset, size_t size,
-	     enum dma_data_direction dir, unsigned long attrs)
-{
-	unsigned long pfn = PFN_DOWN(dev_addr);
-
-	/*
-	 * Dom0 is mapped 1:1, and while the Linux page can span across multiple
-	 * Xen pages, it is not possible for it to contain a mix of local and
-	 * foreign Xen pages.  Calling pfn_valid on a foreign mfn will always
-	 * return false, so if pfn_valid returns true the pages is local and we
-	 * can use the native dma-direct functions, otherwise we call the Xen
-	 * specific version.
-	 */
-	if (pfn_valid(pfn))
-		dma_direct_map_page(hwdev, page, offset, size, dir, attrs);
-	else
-		__xen_dma_map_page(hwdev, page, dev_addr, offset, size, dir, attrs);
-}
-
-static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir, unsigned long attrs)
-{
-	unsigned long pfn = PFN_DOWN(handle);
-
-	if (pfn_valid(pfn))
-		dma_direct_unmap_page(hwdev, handle, size, dir, attrs);
-	else
-		__xen_dma_unmap_page(hwdev, handle, size, dir, attrs);
-}
+void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
+void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
 
 #endif /* _XEN_ARM_PAGE_COHERENT_H */
Signature · 1 line
2.20.1
#19 lore raw

PATCH 9/11 swiotlb-xen: remove page-coherent.h

In reply to Christoph Hellwig

The only thing left of page-coherent.h is two functions implemented by the architecture for non-coherent DMA support that are never called for fully coherent architectures. Just move the prototypes for those to swiotlb-xen.h instead.

---
 arch/arm/include/asm/xen/page-coherent.h   |  2 --
 arch/arm64/include/asm/xen/page-coherent.h |  2 --
 arch/x86/include/asm/xen/page-coherent.h   | 11 -----------
 drivers/xen/swiotlb-xen.c                  |  3 ---
 include/Kbuild                             |  1 -
 include/xen/arm/page-coherent.h            | 10 ----------
 include/xen/swiotlb-xen.h                  |  6 ++++++
 7 files changed, 6 insertions(+), 29 deletions(-)
 delete mode 100644 arch/arm/include/asm/xen/page-coherent.h
 delete mode 100644 arch/arm64/include/asm/xen/page-coherent.h
 delete mode 100644 arch/x86/include/asm/xen/page-coherent.h
 delete mode 100644 include/xen/arm/page-coherent.h
arch/arm/include/asm/xen/page-coherent.h deleted +0 -2
deleted file mode 100644
index 27e984977402..000000000000
@@ -1,2 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#include <xen/arm/page-coherent.h>
arch/arm64/include/asm/xen/page-coherent.h deleted +0 -2
deleted file mode 100644
index 27e984977402..000000000000
@@ -1,2 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#include <xen/arm/page-coherent.h>
arch/x86/include/asm/xen/page-coherent.h deleted +0 -11
deleted file mode 100644
index c9c8398a31ff..000000000000
@@ -1,11 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_X86_XEN_PAGE_COHERENT_H
-#define _ASM_X86_XEN_PAGE_COHERENT_H
-
-static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
-
-static inline void xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
-
-#endif /* _ASM_X86_XEN_PAGE_COHERENT_H */
drivers/xen/swiotlb-xen.c +0 -3
index a642e284f1e2..95911ff9c11c 100644
@@ -35,9 +35,6 @@
 #include <xen/xen-ops.h>
 #include <xen/hvc-console.h>
 
-#include <asm/dma-mapping.h>
-#include <asm/xen/page-coherent.h>
-
 #include <trace/events/swiotlb.h>
 /*
  * Used to do a quick range check in swiotlb_tbl_unmap_single and
include/Kbuild +0 -1
index c38f0d46b267..cce5cf6abf89 100644
@@ -1189,7 +1189,6 @@ header-test-			+= video/vga.h
 header-test-			+= video/w100fb.h
 header-test-			+= xen/acpi.h
 header-test-			+= xen/arm/hypercall.h
-header-test-			+= xen/arm/page-coherent.h
 header-test-			+= xen/arm/page.h
 header-test-			+= xen/balloon.h
 header-test-			+= xen/events.h
include/xen/arm/page-coherent.h deleted +0 -10
deleted file mode 100644
index 635492d41ebe..000000000000
@@ -1,10 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _XEN_ARM_PAGE_COHERENT_H
-#define _XEN_ARM_PAGE_COHERENT_H
-
-void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
-void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
-
-#endif /* _XEN_ARM_PAGE_COHERENT_H */
include/xen/swiotlb-xen.h +6 -0
index 5e4b83f83dbc..a7c642872568 100644
@@ -2,8 +2,14 @@
 #ifndef __LINUX_SWIOTLB_XEN_H
 #define __LINUX_SWIOTLB_XEN_H
 
+#include <linux/dma-mapping.h>
 #include <linux/swiotlb.h>
 
+void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
+void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
+
 extern int xen_swiotlb_init(int verbose, bool early);
 extern const struct dma_map_ops xen_swiotlb_dma_ops;
 
Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 9/11
#20 lore raw

Re: [PATCH 09/11] swiotlb-xen: remove page-coherent.h

Quoted 6 lines · Christoph Hellwig

The only thing left of page-coherent.h is two functions implemented by the architecture for non-coherent DMA support that are never called for fully coherent architectures. Just move the prototypes for those to swiotlb-xen.h instead.

Quoted 111 lines
---
 arch/arm/include/asm/xen/page-coherent.h   |  2 --
 arch/arm64/include/asm/xen/page-coherent.h |  2 --
 arch/x86/include/asm/xen/page-coherent.h   | 11 -----------
 drivers/xen/swiotlb-xen.c                  |  3 ---
 include/Kbuild                             |  1 -
 include/xen/arm/page-coherent.h            | 10 ----------
 include/xen/swiotlb-xen.h                  |  6 ++++++
 7 files changed, 6 insertions(+), 29 deletions(-)
 delete mode 100644 arch/arm/include/asm/xen/page-coherent.h
 delete mode 100644 arch/arm64/include/asm/xen/page-coherent.h
 delete mode 100644 arch/x86/include/asm/xen/page-coherent.h
 delete mode 100644 include/xen/arm/page-coherent.h
arch/arm/include/asm/xen/page-coherent.h deleted +0 -2
deleted file mode 100644
index 27e984977402..000000000000
@@ -1,2 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#include <xen/arm/page-coherent.h>
arch/arm64/include/asm/xen/page-coherent.h deleted +0 -2
deleted file mode 100644
index 27e984977402..000000000000
@@ -1,2 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#include <xen/arm/page-coherent.h>
arch/x86/include/asm/xen/page-coherent.h deleted +0 -11
deleted file mode 100644
index c9c8398a31ff..000000000000
@@ -1,11 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_X86_XEN_PAGE_COHERENT_H
-#define _ASM_X86_XEN_PAGE_COHERENT_H
-
-static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
-
-static inline void xen_dma_sync_single_for_device(struct device *hwdev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
-
-#endif /* _ASM_X86_XEN_PAGE_COHERENT_H */
drivers/xen/swiotlb-xen.c +0 -3
index a642e284f1e2..95911ff9c11c 100644
@@ -35,9 +35,6 @@
 #include <xen/xen-ops.h>
 #include <xen/hvc-console.h>
 
-#include <asm/dma-mapping.h>
-#include <asm/xen/page-coherent.h>
-
 #include <trace/events/swiotlb.h>
 /*
  * Used to do a quick range check in swiotlb_tbl_unmap_single and
include/Kbuild +0 -1
index c38f0d46b267..cce5cf6abf89 100644
@@ -1189,7 +1189,6 @@ header-test-			+= video/vga.h
 header-test-			+= video/w100fb.h
 header-test-			+= xen/acpi.h
 header-test-			+= xen/arm/hypercall.h
-header-test-			+= xen/arm/page-coherent.h
 header-test-			+= xen/arm/page.h
 header-test-			+= xen/balloon.h
 header-test-			+= xen/events.h
include/xen/arm/page-coherent.h deleted +0 -10
deleted file mode 100644
index 635492d41ebe..000000000000
@@ -1,10 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _XEN_ARM_PAGE_COHERENT_H
-#define _XEN_ARM_PAGE_COHERENT_H
-
-void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
-void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
-
-#endif /* _XEN_ARM_PAGE_COHERENT_H */
include/xen/swiotlb-xen.h +6 -0
index 5e4b83f83dbc..a7c642872568 100644
@@ -2,8 +2,14 @@
 #ifndef __LINUX_SWIOTLB_XEN_H
 #define __LINUX_SWIOTLB_XEN_H
 
+#include <linux/dma-mapping.h>
 #include <linux/swiotlb.h>
 
+void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
+void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
+		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
+
 extern int xen_swiotlb_init(int verbose, bool early);
 extern const struct dma_map_ops xen_swiotlb_dma_ops;
 
Signature · 1 line
2.20.1
#21 lore raw

PATCH 10/11 swiotlb-xen: merge xen_unmap_single into xen_swiotlb_unmap_page

In reply to Christoph Hellwig

No need for a no-op wrapper.

---
 drivers/xen/swiotlb-xen.c | 15 ++++-----------
 1 file changed, 4 insertions(+), 11 deletions(-)
drivers/xen/swiotlb-xen.c +4 -11
index 95911ff9c11c..384304a77020 100644
@@ -414,9 +414,8 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
  * After this call, reads by the cpu to the buffer are guaranteed to see
  * whatever the device wrote there.
  */
-static void xen_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
-			     size_t size, enum dma_data_direction dir,
-			     unsigned long attrs)
+static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
+		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
 	phys_addr_t paddr = xen_bus_to_phys(dev_addr);
 
@@ -430,13 +429,6 @@ static void xen_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
 		swiotlb_tbl_unmap_single(hwdev, paddr, size, dir, attrs);
 }
 
-static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
-			    size_t size, enum dma_data_direction dir,
-			    unsigned long attrs)
-{
-	xen_unmap_single(hwdev, dev_addr, size, dir, attrs);
-}
-
 static void
 xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr,
 		size_t size, enum dma_data_direction dir)
@@ -477,7 +469,8 @@ xen_swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sgl, int nelems,
 	BUG_ON(dir == DMA_NONE);
 
 	for_each_sg(sgl, sg, nelems, i)
-		xen_unmap_single(hwdev, sg->dma_address, sg_dma_len(sg), dir, attrs);
+		xen_swiotlb_unmap_page(hwdev, sg->dma_address, sg_dma_len(sg),
+				dir, attrs);
 
 }
 
Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 10/11
#22 lore raw

Re: [PATCH 10/11] swiotlb-xen: merge xen_unmap_single into xen_swiotlb_unmap_page

Quoted 3 lines · Christoph Hellwig

No need for a no-op wrapper.

Quoted 47 lines
---
 drivers/xen/swiotlb-xen.c | 15 ++++-----------
 1 file changed, 4 insertions(+), 11 deletions(-)
drivers/xen/swiotlb-xen.c +4 -11
index 95911ff9c11c..384304a77020 100644
@@ -414,9 +414,8 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
  * After this call, reads by the cpu to the buffer are guaranteed to see
  * whatever the device wrote there.
  */
-static void xen_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
-			     size_t size, enum dma_data_direction dir,
-			     unsigned long attrs)
+static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
+		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
 	phys_addr_t paddr = xen_bus_to_phys(dev_addr);
 
@@ -430,13 +429,6 @@ static void xen_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
 		swiotlb_tbl_unmap_single(hwdev, paddr, size, dir, attrs);
 }
 
-static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
-			    size_t size, enum dma_data_direction dir,
-			    unsigned long attrs)
-{
-	xen_unmap_single(hwdev, dev_addr, size, dir, attrs);
-}
-
 static void
 xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr,
 		size_t size, enum dma_data_direction dir)
@@ -477,7 +469,8 @@ xen_swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sgl, int nelems,
 	BUG_ON(dir == DMA_NONE);
 
 	for_each_sg(sgl, sg, nelems, i)
-		xen_unmap_single(hwdev, sg->dma_address, sg_dma_len(sg), dir, attrs);
+		xen_swiotlb_unmap_page(hwdev, sg->dma_address, sg_dma_len(sg),
+				dir, attrs);
 
 }
 
Signature · 1 line
2.20.1
#23 lore raw

PATCH 11/11 arm64: use asm-generic/dma-mapping.h

In reply to Christoph Hellwig

Now that the Xen special cases are gone nothing worth mentioning is left in the arm64 <asm/dma-mapping.h> file, so switch to use the asm-generic version instead.

---
 arch/arm64/include/asm/Kbuild        |  1 +
 arch/arm64/include/asm/dma-mapping.h | 22 ----------------------
 arch/arm64/mm/dma-mapping.c          |  1 +
 3 files changed, 2 insertions(+), 22 deletions(-)
 delete mode 100644 arch/arm64/include/asm/dma-mapping.h
arch/arm64/include/asm/Kbuild +1 -0
index c52e151afab0..98a5405c8558 100644
@@ -4,6 +4,7 @@ generic-y += delay.h
 generic-y += div64.h
 generic-y += dma.h
 generic-y += dma-contiguous.h
+generic-y += dma-mapping.h
 generic-y += early_ioremap.h
 generic-y += emergency-restart.h
 generic-y += hw_irq.h
arch/arm64/include/asm/dma-mapping.h deleted +0 -22
deleted file mode 100644
index 67243255a858..000000000000
@@ -1,22 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * Copyright (C) 2012 ARM Ltd.
- */
-#ifndef __ASM_DMA_MAPPING_H
-#define __ASM_DMA_MAPPING_H
-
-#ifdef __KERNEL__
-
-#include <linux/types.h>
-#include <linux/vmalloc.h>
-
-#include <xen/xen.h>
-#include <asm/xen/hypervisor.h>
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	return NULL;
-}
-
-#endif	/* __KERNEL__ */
-#endif	/* __ASM_DMA_MAPPING_H */
arch/arm64/mm/dma-mapping.c +1 -0
index 4b244a037349..6578abcfbbc7 100644
@@ -8,6 +8,7 @@
 #include <linux/cache.h>
 #include <linux/dma-noncoherent.h>
 #include <linux/dma-iommu.h>
+#include <xen/xen.h>
 #include <xen/swiotlb-xen.h>
 
 #include <asm/cacheflush.h>
Signature · 1 line
2.20.1
maintainer Reviewed-by on v1 11/11
#24 lore raw

Re: [PATCH 11/11] arm64: use asm-generic/dma-mapping.h

Quoted 6 lines · Christoph Hellwig

Now that the Xen special cases are gone nothing worth mentioning is left in the arm64 <asm/dma-mapping.h> file, so switch to use the asm-generic version instead.

Quoted 62 lines
---
 arch/arm64/include/asm/Kbuild        |  1 +
 arch/arm64/include/asm/dma-mapping.h | 22 ----------------------
 arch/arm64/mm/dma-mapping.c          |  1 +
 3 files changed, 2 insertions(+), 22 deletions(-)
 delete mode 100644 arch/arm64/include/asm/dma-mapping.h
arch/arm64/include/asm/Kbuild +1 -0
index c52e151afab0..98a5405c8558 100644
@@ -4,6 +4,7 @@ generic-y += delay.h
 generic-y += div64.h
 generic-y += dma.h
 generic-y += dma-contiguous.h
+generic-y += dma-mapping.h
 generic-y += early_ioremap.h
 generic-y += emergency-restart.h
 generic-y += hw_irq.h
arch/arm64/include/asm/dma-mapping.h deleted +0 -22
deleted file mode 100644
index 67243255a858..000000000000
@@ -1,22 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * Copyright (C) 2012 ARM Ltd.
- */
-#ifndef __ASM_DMA_MAPPING_H
-#define __ASM_DMA_MAPPING_H
-
-#ifdef __KERNEL__
-
-#include <linux/types.h>
-#include <linux/vmalloc.h>
-
-#include <xen/xen.h>
-#include <asm/xen/hypervisor.h>
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	return NULL;
-}
-
-#endif	/* __KERNEL__ */
-#endif	/* __ASM_DMA_MAPPING_H */
arch/arm64/mm/dma-mapping.c +1 -0
index 4b244a037349..6578abcfbbc7 100644
@@ -8,6 +8,7 @@
 #include <linux/cache.h>
 #include <linux/dma-noncoherent.h>
 #include <linux/dma-iommu.h>
+#include <xen/xen.h>
 #include <xen/swiotlb-xen.h>
 
 #include <asm/cacheflush.h>
Signature · 1 line
2.20.1
#25 lore raw
In reply to Christoph Hellwig
Quoted 9 lines · Christoph Hellwig

Hi Xen maintainers and friends,

please take a look at this series that cleans up the parts of swiotlb-xen that deal with non-coherent caches.

Changes since v1:
- rewrite dma_cache_maint to be much simpler
- improve various comments and commit logs
- remove page-coherent.h entirely

Thanks for your work on this, it really makes the code better. I tested it on ARM64 with a non-coherent network device and verified it works as intended (Cadence GEM on ZynqMP).