* [PATCH 28/33] dma-direct: add support for allocation from ZONE_DMA and ZONE_DMA32
From: Christoph Hellwig @ 2018-01-10 8:00 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080027.13879-1-hch@lst.de>
This allows to dip into zones for lower memory if they are available.
If one of the zones is not available the corresponding GFP_* flag
will evaluate to 0 so they won't change anything. We provide an
arch tunable for those architectures that do not use GFP_DMA for
the lowest 24-bits, given that there are a few.
Roughly based on the x86 code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
lib/dma-direct.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/lib/dma-direct.c b/lib/dma-direct.c
index f04a424f91fa..8f76032ebc3c 100644
--- a/lib/dma-direct.c
+++ b/lib/dma-direct.c
@@ -12,6 +12,14 @@
#define DIRECT_MAPPING_ERROR 0
+/*
+ * Most architectures use ZONE_DMA for the first 16 Megabytes, but
+ * some use it for entirely different regions:
+ */
+#ifndef ARCH_ZONE_DMA_BITS
+#define ARCH_ZONE_DMA_BITS 24
+#endif
+
static bool
check_addr(struct device *dev, dma_addr_t dma_addr, size_t size,
const char *caller)
@@ -34,6 +42,12 @@ static void *dma_direct_alloc(struct device *dev, size_t size,
int page_order = get_order(size);
struct page *page = NULL;
+ /* GFP_DMA32 and GFP_DMA are no ops without the corresponding zones: */
+ if (dev->coherent_dma_mask <= DMA_BIT_MASK(ARCH_ZONE_DMA_BITS))
+ gfp |= GFP_DMA;
+ if (dev->coherent_dma_mask <= DMA_BIT_MASK(32) && !(gfp & GFP_DMA))
+ gfp |= GFP_DMA32;
+
/* CMA can be used only in the context which permits sleeping */
if (gfpflags_allow_blocking(gfp))
page = dma_alloc_from_contiguous(dev, count, page_order, gfp);
--
2.14.2
^ permalink raw reply related
* [PATCH 29/33] dma-direct: retry allocations using GFP_DMA for small masks
From: Christoph Hellwig @ 2018-01-10 8:00 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080027.13879-1-hch@lst.de>
If an attempt to allocate memory succeeded, but isn't inside the
supported DMA mask, retry the allocation with GFP_DMA set as a
last resort.
Based on the x86 code, but an off by one error in what is now
dma_coherent_ok has been fixed vs the x86 code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
lib/dma-direct.c | 25 ++++++++++++++++++++++++-
1 file changed, 24 insertions(+), 1 deletion(-)
diff --git a/lib/dma-direct.c b/lib/dma-direct.c
index 8f76032ebc3c..4e43c2bb7f5f 100644
--- a/lib/dma-direct.c
+++ b/lib/dma-direct.c
@@ -35,6 +35,11 @@ check_addr(struct device *dev, dma_addr_t dma_addr, size_t size,
return true;
}
+static bool dma_coherent_ok(struct device *dev, phys_addr_t phys, size_t size)
+{
+ return phys_to_dma(dev, phys) + size - 1 <= dev->coherent_dma_mask;
+}
+
static void *dma_direct_alloc(struct device *dev, size_t size,
dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
{
@@ -48,11 +53,29 @@ static void *dma_direct_alloc(struct device *dev, size_t size,
if (dev->coherent_dma_mask <= DMA_BIT_MASK(32) && !(gfp & GFP_DMA))
gfp |= GFP_DMA32;
+again:
/* CMA can be used only in the context which permits sleeping */
- if (gfpflags_allow_blocking(gfp))
+ if (gfpflags_allow_blocking(gfp)) {
page = dma_alloc_from_contiguous(dev, count, page_order, gfp);
+ if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
+ dma_release_from_contiguous(dev, page, count);
+ page = NULL;
+ }
+ }
if (!page)
page = alloc_pages_node(dev_to_node(dev), gfp, page_order);
+
+ if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
+ __free_pages(page, page_order);
+ page = NULL;
+
+ if (dev->coherent_dma_mask < DMA_BIT_MASK(32) &&
+ !(gfp & GFP_DMA)) {
+ gfp = (gfp & ~GFP_DMA32) | GFP_DMA;
+ goto again;
+ }
+ }
+
if (!page)
return NULL;
--
2.14.2
^ permalink raw reply related
* [PATCH 30/33] dma-direct: make dma_direct_{alloc, free} available to other implementations
From: Christoph Hellwig @ 2018-01-10 8:00 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080027.13879-1-hch@lst.de>
So that they don't need to indirect through the operation vector.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
---
arch/arm/mm/dma-mapping-nommu.c | 9 +++------
include/linux/dma-direct.h | 5 +++++
lib/dma-direct.c | 6 +++---
3 files changed, 11 insertions(+), 9 deletions(-)
diff --git a/arch/arm/mm/dma-mapping-nommu.c b/arch/arm/mm/dma-mapping-nommu.c
index 4d8042521e89..619f24a42d09 100644
--- a/arch/arm/mm/dma-mapping-nommu.c
+++ b/arch/arm/mm/dma-mapping-nommu.c
@@ -11,7 +11,7 @@
#include <linux/export.h>
#include <linux/mm.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
#include <linux/scatterlist.h>
#include <asm/cachetype.h>
@@ -39,7 +39,6 @@ static void *arm_nommu_dma_alloc(struct device *dev, size_t size,
unsigned long attrs)
{
- const struct dma_map_ops *ops = &dma_direct_ops;
void *ret;
/*
@@ -48,7 +47,7 @@ static void *arm_nommu_dma_alloc(struct device *dev, size_t size,
*/
if (attrs & DMA_ATTR_NON_CONSISTENT)
- return ops->alloc(dev, size, dma_handle, gfp, attrs);
+ return dma_direct_alloc(dev, size, dma_handle, gfp, attrs);
ret = dma_alloc_from_global_coherent(size, dma_handle);
@@ -70,10 +69,8 @@ static void arm_nommu_dma_free(struct device *dev, size_t size,
void *cpu_addr, dma_addr_t dma_addr,
unsigned long attrs)
{
- const struct dma_map_ops *ops = &dma_direct_ops;
-
if (attrs & DMA_ATTR_NON_CONSISTENT) {
- ops->free(dev, size, cpu_addr, dma_addr, attrs);
+ dma_direct_free(dev, size, cpu_addr, dma_addr, attrs);
} else {
int ret = dma_release_from_global_coherent(get_order(size),
cpu_addr);
diff --git a/include/linux/dma-direct.h b/include/linux/dma-direct.h
index 10e924b7cba7..4788bf0bf683 100644
--- a/include/linux/dma-direct.h
+++ b/include/linux/dma-direct.h
@@ -38,4 +38,9 @@ static inline void dma_mark_clean(void *addr, size_t size)
}
#endif /* CONFIG_ARCH_HAS_DMA_MARK_CLEAN */
+void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
+ gfp_t gfp, unsigned long attrs);
+void dma_direct_free(struct device *dev, size_t size, void *cpu_addr,
+ dma_addr_t dma_addr, unsigned long attrs);
+
#endif /* _LINUX_DMA_DIRECT_H */
diff --git a/lib/dma-direct.c b/lib/dma-direct.c
index 4e43c2bb7f5f..784a68dfdbe3 100644
--- a/lib/dma-direct.c
+++ b/lib/dma-direct.c
@@ -40,8 +40,8 @@ static bool dma_coherent_ok(struct device *dev, phys_addr_t phys, size_t size)
return phys_to_dma(dev, phys) + size - 1 <= dev->coherent_dma_mask;
}
-static void *dma_direct_alloc(struct device *dev, size_t size,
- dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
+void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
+ gfp_t gfp, unsigned long attrs)
{
unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
int page_order = get_order(size);
@@ -84,7 +84,7 @@ static void *dma_direct_alloc(struct device *dev, size_t size,
return page_address(page);
}
-static void dma_direct_free(struct device *dev, size_t size, void *cpu_addr,
+void dma_direct_free(struct device *dev, size_t size, void *cpu_addr,
dma_addr_t dma_addr, unsigned long attrs)
{
unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
--
2.14.2
^ permalink raw reply related
* [PATCH 31/33] dma-direct: reject too small dma masks
From: Christoph Hellwig @ 2018-01-10 8:00 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080027.13879-1-hch@lst.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
include/linux/dma-direct.h | 1 +
lib/dma-direct.c | 19 +++++++++++++++++++
2 files changed, 20 insertions(+)
diff --git a/include/linux/dma-direct.h b/include/linux/dma-direct.h
index 4788bf0bf683..bcdb1a3e4b1f 100644
--- a/include/linux/dma-direct.h
+++ b/include/linux/dma-direct.h
@@ -42,5 +42,6 @@ void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
gfp_t gfp, unsigned long attrs);
void dma_direct_free(struct device *dev, size_t size, void *cpu_addr,
dma_addr_t dma_addr, unsigned long attrs);
+int dma_direct_supported(struct device *dev, u64 mask);
#endif /* _LINUX_DMA_DIRECT_H */
diff --git a/lib/dma-direct.c b/lib/dma-direct.c
index 784a68dfdbe3..40b1f92f2214 100644
--- a/lib/dma-direct.c
+++ b/lib/dma-direct.c
@@ -122,6 +122,24 @@ static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
return nents;
}
+int dma_direct_supported(struct device *dev, u64 mask)
+{
+#ifdef CONFIG_ZONE_DMA
+ if (mask < DMA_BIT_MASK(ARCH_ZONE_DMA_BITS))
+ return 0;
+#else
+ /*
+ * Because 32-bit DMA masks are so common we expect every architecture
+ * to be able to satisfy them - either by not supporting more physical
+ * memory, or by providing a ZONE_DMA32. If neither is the case, the
+ * architecture needs to use an IOMMU instead of the direct mapping.
+ */
+ if (mask < DMA_BIT_MASK(32))
+ return 0;
+#endif
+ return 1;
+}
+
static int dma_direct_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
return dma_addr == DIRECT_MAPPING_ERROR;
@@ -132,6 +150,7 @@ const struct dma_map_ops dma_direct_ops = {
.free = dma_direct_free,
.map_page = dma_direct_map_page,
.map_sg = dma_direct_map_sg,
+ .dma_supported = dma_direct_supported,
.mapping_error = dma_direct_mapping_error,
};
EXPORT_SYMBOL(dma_direct_ops);
--
2.14.2
^ permalink raw reply related
* [PATCH 32/33] cris: use dma-direct
From: Christoph Hellwig @ 2018-01-10 8:00 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080027.13879-1-hch@lst.de>
cris currently has an incomplete direct mapping dma_map_ops implementation
is PCI support is enabled. Replace it with the fully feature generic
dma-direct implementation.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
---
arch/cris/Kconfig | 4 ++
arch/cris/arch-v32/drivers/pci/Makefile | 2 +-
arch/cris/arch-v32/drivers/pci/dma.c | 77 ---------------------------------
arch/cris/include/asm/Kbuild | 1 +
arch/cris/include/asm/dma-mapping.h | 20 ---------
5 files changed, 6 insertions(+), 98 deletions(-)
delete mode 100644 arch/cris/arch-v32/drivers/pci/dma.c
delete mode 100644 arch/cris/include/asm/dma-mapping.h
diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig
index 54d3f426763b..cd5a0865c97f 100644
--- a/arch/cris/Kconfig
+++ b/arch/cris/Kconfig
@@ -33,6 +33,9 @@ config GENERIC_CALIBRATE_DELAY
config NO_IOPORT_MAP
def_bool y if !PCI
+config NO_DMA
+ def_bool y if !PCI
+
config FORCE_MAX_ZONEORDER
int
default 6
@@ -72,6 +75,7 @@ config CRIS
select GENERIC_SCHED_CLOCK if ETRAX_ARCH_V32
select HAVE_DEBUG_BUGVERBOSE if ETRAX_ARCH_V32
select HAVE_NMI
+ select DMA_DIRECT_OPS if PCI
config HZ
int
diff --git a/arch/cris/arch-v32/drivers/pci/Makefile b/arch/cris/arch-v32/drivers/pci/Makefile
index bff7482f2444..93c8be6170b1 100644
--- a/arch/cris/arch-v32/drivers/pci/Makefile
+++ b/arch/cris/arch-v32/drivers/pci/Makefile
@@ -2,4 +2,4 @@
# Makefile for Etrax cardbus driver
#
-obj-$(CONFIG_ETRAX_CARDBUS) += bios.o dma.o
+obj-$(CONFIG_ETRAX_CARDBUS) += bios.o
diff --git a/arch/cris/arch-v32/drivers/pci/dma.c b/arch/cris/arch-v32/drivers/pci/dma.c
deleted file mode 100644
index 8c3802244ef3..000000000000
--- a/arch/cris/arch-v32/drivers/pci/dma.c
+++ /dev/null
@@ -1,77 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Dynamic DMA mapping support.
- *
- * On cris there is no hardware dynamic DMA address translation,
- * so consistent alloc/free are merely page allocation/freeing.
- * The rest of the dynamic DMA mapping interface is implemented
- * in asm/pci.h.
- *
- * Borrowed from i386.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <linux/gfp.h>
-#include <asm/io.h>
-
-static void *v32_dma_alloc(struct device *dev, size_t size,
- dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
-{
- void *ret;
-
- if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
- gfp |= GFP_DMA;
-
- ret = (void *)__get_free_pages(gfp, get_order(size));
-
- if (ret != NULL) {
- memset(ret, 0, size);
- *dma_handle = virt_to_phys(ret);
- }
- return ret;
-}
-
-static void v32_dma_free(struct device *dev, size_t size, void *vaddr,
- dma_addr_t dma_handle, unsigned long attrs)
-{
- free_pages((unsigned long)vaddr, get_order(size));
-}
-
-static inline dma_addr_t v32_dma_map_page(struct device *dev,
- struct page *page, unsigned long offset, size_t size,
- enum dma_data_direction direction, unsigned long attrs)
-{
- return page_to_phys(page) + offset;
-}
-
-static inline int v32_dma_map_sg(struct device *dev, struct scatterlist *sg,
- int nents, enum dma_data_direction direction,
- unsigned long attrs)
-{
- printk("Map sg\n");
- return nents;
-}
-
-static inline int v32_dma_supported(struct device *dev, u64 mask)
-{
- /*
- * we fall back to GFP_DMA when the mask isn't all 1s,
- * so we can't guarantee allocations that must be
- * within a tighter range than GFP_DMA..
- */
- if (mask < 0x00ffffff)
- return 0;
- return 1;
-}
-
-const struct dma_map_ops v32_dma_ops = {
- .alloc = v32_dma_alloc,
- .free = v32_dma_free,
- .map_page = v32_dma_map_page,
- .map_sg = v32_dma_map_sg,
- .dma_supported = v32_dma_supported,
-};
-EXPORT_SYMBOL(v32_dma_ops);
diff --git a/arch/cris/include/asm/Kbuild b/arch/cris/include/asm/Kbuild
index 460349cb147f..8cf45ac30c1b 100644
--- a/arch/cris/include/asm/Kbuild
+++ b/arch/cris/include/asm/Kbuild
@@ -5,6 +5,7 @@ generic-y += cmpxchg.h
generic-y += current.h
generic-y += device.h
generic-y += div64.h
+generic-y += dma-mapping.h
generic-y += emergency-restart.h
generic-y += exec.h
generic-y += extable.h
diff --git a/arch/cris/include/asm/dma-mapping.h b/arch/cris/include/asm/dma-mapping.h
deleted file mode 100644
index 1553bdb30a0c..000000000000
--- a/arch/cris/include/asm/dma-mapping.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_CRIS_DMA_MAPPING_H
-#define _ASM_CRIS_DMA_MAPPING_H
-
-#ifdef CONFIG_PCI
-extern const struct dma_map_ops v32_dma_ops;
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
- return &v32_dma_ops;
-}
-#else
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
- BUG();
- return NULL;
-}
-#endif
-
-#endif
--
2.14.2
^ permalink raw reply related
* [PATCH 33/33] h8300: use dma-direct
From: Christoph Hellwig @ 2018-01-10 8:00 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080027.13879-1-hch@lst.de>
Replace the bare-bones h8300 direct dma mapping implementation with
the fully featured generic dma-direct one.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/h8300/Kconfig | 1 +
arch/h8300/include/asm/Kbuild | 1 +
arch/h8300/include/asm/dma-mapping.h | 12 -------
arch/h8300/kernel/Makefile | 2 +-
arch/h8300/kernel/dma.c | 66 ------------------------------------
5 files changed, 3 insertions(+), 79 deletions(-)
delete mode 100644 arch/h8300/include/asm/dma-mapping.h
delete mode 100644 arch/h8300/kernel/dma.c
diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig
index f8d3fde08190..091d6d04b5e5 100644
--- a/arch/h8300/Kconfig
+++ b/arch/h8300/Kconfig
@@ -23,6 +23,7 @@ config H8300
select HAVE_ARCH_KGDB
select HAVE_ARCH_HASH
select CPU_NO_EFFICIENT_FFS
+ select DMA_DIRECT_OPS
config CPU_BIG_ENDIAN
def_bool y
diff --git a/arch/h8300/include/asm/Kbuild b/arch/h8300/include/asm/Kbuild
index bc077491d299..642752c94306 100644
--- a/arch/h8300/include/asm/Kbuild
+++ b/arch/h8300/include/asm/Kbuild
@@ -9,6 +9,7 @@ generic-y += delay.h
generic-y += device.h
generic-y += div64.h
generic-y += dma.h
+generic-y += dma-mapping.h
generic-y += emergency-restart.h
generic-y += exec.h
generic-y += extable.h
diff --git a/arch/h8300/include/asm/dma-mapping.h b/arch/h8300/include/asm/dma-mapping.h
deleted file mode 100644
index 21bb1fc3a6f1..000000000000
--- a/arch/h8300/include/asm/dma-mapping.h
+++ /dev/null
@@ -1,12 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _H8300_DMA_MAPPING_H
-#define _H8300_DMA_MAPPING_H
-
-extern const struct dma_map_ops h8300_dma_map_ops;
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
- return &h8300_dma_map_ops;
-}
-
-#endif
diff --git a/arch/h8300/kernel/Makefile b/arch/h8300/kernel/Makefile
index b62e830525c6..307aa51576dd 100644
--- a/arch/h8300/kernel/Makefile
+++ b/arch/h8300/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y := vmlinux.lds
obj-y := process.o traps.o ptrace.o \
signal.o setup.o syscalls.o \
- irq.o entry.o dma.o
+ irq.o entry.o
obj-$(CONFIG_ROMKERNEL) += head_rom.o
obj-$(CONFIG_RAMKERNEL) += head_ram.o
diff --git a/arch/h8300/kernel/dma.c b/arch/h8300/kernel/dma.c
deleted file mode 100644
index d44ba5db4ac3..000000000000
--- a/arch/h8300/kernel/dma.c
+++ /dev/null
@@ -1,66 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License. See the file COPYING in the main directory of this archive
- * for more details.
- */
-
-#include <linux/dma-mapping.h>
-#include <linux/kernel.h>
-#include <linux/scatterlist.h>
-#include <linux/module.h>
-#include <asm/pgalloc.h>
-
-static void *dma_alloc(struct device *dev, size_t size,
- dma_addr_t *dma_handle, gfp_t gfp,
- unsigned long attrs)
-{
- void *ret;
-
- if (dev == NULL || (*dev->dma_mask < 0xffffffff))
- gfp |= GFP_DMA;
- ret = (void *)__get_free_pages(gfp, get_order(size));
-
- if (ret != NULL) {
- memset(ret, 0, size);
- *dma_handle = virt_to_phys(ret);
- }
- return ret;
-}
-
-static void dma_free(struct device *dev, size_t size,
- void *vaddr, dma_addr_t dma_handle,
- unsigned long attrs)
-
-{
- free_pages((unsigned long)vaddr, get_order(size));
-}
-
-static dma_addr_t map_page(struct device *dev, struct page *page,
- unsigned long offset, size_t size,
- enum dma_data_direction direction,
- unsigned long attrs)
-{
- return page_to_phys(page) + offset;
-}
-
-static int map_sg(struct device *dev, struct scatterlist *sgl,
- int nents, enum dma_data_direction direction,
- unsigned long attrs)
-{
- struct scatterlist *sg;
- int i;
-
- for_each_sg(sgl, sg, nents, i) {
- sg->dma_address = sg_phys(sg);
- }
-
- return nents;
-}
-
-const struct dma_map_ops h8300_dma_map_ops = {
- .alloc = dma_alloc,
- .free = dma_free,
- .map_page = map_page,
- .map_sg = map_sg,
-};
-EXPORT_SYMBOL(h8300_dma_map_ops);
--
2.14.2
^ permalink raw reply related
* [RFC PATCH] of_pci_irq: add a check to fallback to device tree parsing
From: Ryder Lee @ 2018-01-10 8:04 UTC (permalink / raw)
To: linux-arm-kernel
A root complex normally consist of a host/PCI bridge and multiple P2P bridges.
It will get mismatched results if we implement the bindings in the form of
a root node with multiple subnodes (P2P bridge) and list all interrupt-map
properties for each slot in the parent - It maens that we want to propagate
IRQs from a root port to the devices in the hierarchy below it.
If we have a PCIe device which is connected to slot 1, and will get something
like this:
pcieport 0000:00:01.0: assign IRQ: got 213
--> igb 0000:01:00.0: assign IRQ: got 212
The reason is that we use the subordinate 'devfn' but didn't obtain the actual
slot numbers from device tree, this patch add a check to fallback to use device
tree parsing if needed.
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
---
Discussion thread: https://patchwork.ozlabs.org/patch/829108/
---
drivers/of/of_pci_irq.c | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/drivers/of/of_pci_irq.c b/drivers/of/of_pci_irq.c
index 3a05568..e445866 100644
--- a/drivers/of/of_pci_irq.c
+++ b/drivers/of/of_pci_irq.c
@@ -86,8 +86,18 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq
out_irq->np = ppnode;
out_irq->args_count = 1;
out_irq->args[0] = pin;
- laddr[0] = cpu_to_be32((pdev->bus->number << 16) | (pdev->devfn << 8));
- laddr[1] = laddr[2] = cpu_to_be32(0);
+
+ if (!dn && ppnode) {
+ const __be32 *addr;
+
+ addr = of_get_property(ppnode, "reg", NULL);
+ if (addr)
+ memcpy(laddr, addr, 3);
+ } else {
+ laddr[0] = cpu_to_be32((pdev->bus->number << 16) | (pdev->devfn << 8));
+ laddr[1] = laddr[2] = cpu_to_be32(0);
+ }
+
rc = of_irq_parse_raw(laddr, out_irq);
if (rc)
goto err;
--
1.9.1
^ permalink raw reply related
* [PATCH V5 00/13] drivers: Boot Constraint core
From: Greg Kroah-Hartman @ 2018-01-10 8:08 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110034440.GA3335@vireshk-i7>
On Wed, Jan 10, 2018 at 09:14:40AM +0530, Viresh Kumar wrote:
> On 09-01-18, 19:47, Greg Kroah-Hartman wrote:
> > Can you resend this? As you can tell, I've been a bit busy for the past
> > month or so :(
>
> Sure.
>
> > Also, why is there no signed-off-by on the OF core patches?
>
> A bit confused, sorry. Are you looking for my signed-off ? They are already
> there.
>
> Or are you looking for Rob's (OF maintainer) signed-off ?
Looking for Rob's, why hasn't he reviewed this yet?
thanks,
greg k-h
^ permalink raw reply
* consolidate swiotlb dma_map implementations
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
A lot of architectures have essentially identical dma_map_ops
implementations to use swiotlb. This series adds new generic
swiotlb_alloc/free helpers that take the attrs argument exposed
in dma_map_ops, and which do an enhanced direct allocation
modelled after x86 and reused from the dma-direct code, and
then switches most architectures over to it. The only exceptions
are mips, which requires additional cache flushing which will
need a new abstraction, and x86 itself which will be handled in
a later series with other x86 dma mapping changes.
To support the generic code a few architectures that currently
use ZONE_DMA/GFP_DMA for <= 32-bit allocations are switched to
implement ZONE_DMA32 instead.
This series is based on the previously sent series to consolidate
the direct dma mapping implementation. A git tree with this
series as well as the prerequisites is available here:
git://git.infradead.org/users/hch/misc.git swiotlb
Gitweb:
http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/swiotlb
^ permalink raw reply
* [PATCH 01/22] swiotlb: suppress warning when __GFP_NOWARN is set
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
From: Christian K?nig <ckoenig.leichtzumerken@gmail.com>
TTM tries to allocate coherent memory in chunks of 2MB first to improve
TLB efficiency and falls back to allocating 4K pages if that fails.
Suppress the warning when the 2MB allocations fails since there is a
valid fall back path.
Signed-off-by: Christian K?nig <christian.koenig@amd.com>
Reported-by: Mike Galbraith <efault@gmx.de>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=104082
CC: stable at vger.kernel.org
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
lib/swiotlb.c | 15 +++++++++------
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 6583f3512386..125c1062119f 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -586,7 +586,7 @@ phys_addr_t swiotlb_tbl_map_single(struct device *hwdev,
not_found:
spin_unlock_irqrestore(&io_tlb_lock, flags);
- if (printk_ratelimit())
+ if (!(attrs & DMA_ATTR_NO_WARN) && printk_ratelimit())
dev_warn(hwdev, "swiotlb buffer is full (sz: %zd bytes)\n", size);
return SWIOTLB_MAP_ERROR;
found:
@@ -713,6 +713,7 @@ void *
swiotlb_alloc_coherent(struct device *hwdev, size_t size,
dma_addr_t *dma_handle, gfp_t flags)
{
+ bool warn = !(flags & __GFP_NOWARN);
dma_addr_t dev_addr;
void *ret;
int order = get_order(size);
@@ -738,8 +739,8 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size,
* GFP_DMA memory; fall back on map_single(), which
* will grab memory from the lowest available address range.
*/
- phys_addr_t paddr = map_single(hwdev, 0, size,
- DMA_FROM_DEVICE, 0);
+ phys_addr_t paddr = map_single(hwdev, 0, size, DMA_FROM_DEVICE,
+ warn ? 0 : DMA_ATTR_NO_WARN);
if (paddr == SWIOTLB_MAP_ERROR)
goto err_warn;
@@ -769,9 +770,11 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size,
return ret;
err_warn:
- pr_warn("swiotlb: coherent allocation failed for device %s size=%zu\n",
- dev_name(hwdev), size);
- dump_stack();
+ if (warn && printk_ratelimit()) {
+ pr_warn("swiotlb: coherent allocation failed for device %s size=%zu\n",
+ dev_name(hwdev), size);
+ dump_stack();
+ }
return NULL;
}
--
2.14.2
^ permalink raw reply related
* [PATCH 02/22] arm64: rename swiotlb_dma_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
We'll need that name for a generic implementation soon.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/arm64/mm/dma-mapping.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index f3a637b98487..6840426bbe77 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c
@@ -368,7 +368,7 @@ static int __swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t addr)
return 0;
}
-static const struct dma_map_ops swiotlb_dma_ops = {
+static const struct dma_map_ops arm64_swiotlb_dma_ops = {
.alloc = __dma_alloc,
.free = __dma_free,
.mmap = __swiotlb_mmap,
@@ -923,7 +923,7 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
const struct iommu_ops *iommu, bool coherent)
{
if (!dev->dma_ops)
- dev->dma_ops = &swiotlb_dma_ops;
+ dev->dma_ops = &arm64_swiotlb_dma_ops;
dev->archdata.dma_coherent = coherent;
__iommu_setup_dma_ops(dev, dma_base, size, iommu);
--
2.14.2
^ permalink raw reply related
* [PATCH 03/22] ia64: rename swiotlb_dma_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
We'll need that name for a generic implementation soon.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/ia64/hp/common/hwsw_iommu.c | 4 ++--
arch/ia64/hp/common/sba_iommu.c | 6 +++---
arch/ia64/kernel/pci-swiotlb.c | 6 +++---
3 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/arch/ia64/hp/common/hwsw_iommu.c b/arch/ia64/hp/common/hwsw_iommu.c
index 63d8e1d2477f..41279f0442bd 100644
--- a/arch/ia64/hp/common/hwsw_iommu.c
+++ b/arch/ia64/hp/common/hwsw_iommu.c
@@ -19,7 +19,7 @@
#include <linux/export.h>
#include <asm/machvec.h>
-extern const struct dma_map_ops sba_dma_ops, swiotlb_dma_ops;
+extern const struct dma_map_ops sba_dma_ops, ia64_swiotlb_dma_ops;
/* swiotlb declarations & definitions: */
extern int swiotlb_late_init_with_default_size (size_t size);
@@ -38,7 +38,7 @@ static inline int use_swiotlb(struct device *dev)
const struct dma_map_ops *hwsw_dma_get_ops(struct device *dev)
{
if (use_swiotlb(dev))
- return &swiotlb_dma_ops;
+ return &ia64_swiotlb_dma_ops;
return &sba_dma_ops;
}
EXPORT_SYMBOL(hwsw_dma_get_ops);
diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index aec4a3354abe..8c0a9ae6afec 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -2096,7 +2096,7 @@ static int __init acpi_sba_ioc_init_acpi(void)
/* This has to run before acpi_scan_init(). */
arch_initcall(acpi_sba_ioc_init_acpi);
-extern const struct dma_map_ops swiotlb_dma_ops;
+extern const struct dma_map_ops ia64_swiotlb_dma_ops;
static int __init
sba_init(void)
@@ -2111,7 +2111,7 @@ sba_init(void)
* a successful kdump kernel boot is to use the swiotlb.
*/
if (is_kdump_kernel()) {
- dma_ops = &swiotlb_dma_ops;
+ dma_ops = &ia64_swiotlb_dma_ops;
if (swiotlb_late_init_with_default_size(64 * (1<<20)) != 0)
panic("Unable to initialize software I/O TLB:"
" Try machvec=dig boot option");
@@ -2133,7 +2133,7 @@ sba_init(void)
* If we didn't find something sba_iommu can claim, we
* need to setup the swiotlb and switch to the dig machvec.
*/
- dma_ops = &swiotlb_dma_ops;
+ dma_ops = &ia64_swiotlb_dma_ops;
if (swiotlb_late_init_with_default_size(64 * (1<<20)) != 0)
panic("Unable to find SBA IOMMU or initialize "
"software I/O TLB: Try machvec=dig boot option");
diff --git a/arch/ia64/kernel/pci-swiotlb.c b/arch/ia64/kernel/pci-swiotlb.c
index 5e50939aa03e..f1ae873a8c35 100644
--- a/arch/ia64/kernel/pci-swiotlb.c
+++ b/arch/ia64/kernel/pci-swiotlb.c
@@ -31,7 +31,7 @@ static void ia64_swiotlb_free_coherent(struct device *dev, size_t size,
swiotlb_free_coherent(dev, size, vaddr, dma_addr);
}
-const struct dma_map_ops swiotlb_dma_ops = {
+const struct dma_map_ops ia64_swiotlb_dma_ops = {
.alloc = ia64_swiotlb_alloc_coherent,
.free = ia64_swiotlb_free_coherent,
.map_page = swiotlb_map_page,
@@ -48,7 +48,7 @@ const struct dma_map_ops swiotlb_dma_ops = {
void __init swiotlb_dma_init(void)
{
- dma_ops = &swiotlb_dma_ops;
+ dma_ops = &ia64_swiotlb_dma_ops;
swiotlb_init(1);
}
@@ -60,7 +60,7 @@ void __init pci_swiotlb_init(void)
printk(KERN_INFO "PCI-DMA: Re-initialize machine vector.\n");
machvec_init("dig");
swiotlb_init(1);
- dma_ops = &swiotlb_dma_ops;
+ dma_ops = &ia64_swiotlb_dma_ops;
#else
panic("Unable to find Intel IOMMU");
#endif
--
2.14.2
^ permalink raw reply related
* [PATCH 04/22] powerpc: rename swiotlb_dma_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
We'll need that name for a generic implementation soon.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/powerpc/include/asm/swiotlb.h | 2 +-
arch/powerpc/kernel/dma-swiotlb.c | 4 ++--
arch/powerpc/kernel/dma.c | 2 +-
arch/powerpc/sysdev/fsl_pci.c | 2 +-
4 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/swiotlb.h b/arch/powerpc/include/asm/swiotlb.h
index 9341ee804d19..f65ecf57b66c 100644
--- a/arch/powerpc/include/asm/swiotlb.h
+++ b/arch/powerpc/include/asm/swiotlb.h
@@ -13,7 +13,7 @@
#include <linux/swiotlb.h>
-extern const struct dma_map_ops swiotlb_dma_ops;
+extern const struct dma_map_ops powerpc_swiotlb_dma_ops;
extern unsigned int ppc_swiotlb_enable;
int __init swiotlb_setup_bus_notifier(void);
diff --git a/arch/powerpc/kernel/dma-swiotlb.c b/arch/powerpc/kernel/dma-swiotlb.c
index f1e99b9cee97..506ac4fafac5 100644
--- a/arch/powerpc/kernel/dma-swiotlb.c
+++ b/arch/powerpc/kernel/dma-swiotlb.c
@@ -46,7 +46,7 @@ static u64 swiotlb_powerpc_get_required(struct device *dev)
* map_page, and unmap_page on highmem, use normal dma_ops
* for everything else.
*/
-const struct dma_map_ops swiotlb_dma_ops = {
+const struct dma_map_ops powerpc_swiotlb_dma_ops = {
.alloc = __dma_nommu_alloc_coherent,
.free = __dma_nommu_free_coherent,
.mmap = dma_nommu_mmap_coherent,
@@ -89,7 +89,7 @@ static int ppc_swiotlb_bus_notify(struct notifier_block *nb,
/* May need to bounce if the device can't address all of DRAM */
if ((dma_get_mask(dev) + 1) < memblock_end_of_DRAM())
- set_dma_ops(dev, &swiotlb_dma_ops);
+ set_dma_ops(dev, &powerpc_swiotlb_dma_ops);
return NOTIFY_DONE;
}
diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c
index 76079841d3d0..da20569de9d4 100644
--- a/arch/powerpc/kernel/dma.c
+++ b/arch/powerpc/kernel/dma.c
@@ -33,7 +33,7 @@ static u64 __maybe_unused get_pfn_limit(struct device *dev)
struct dev_archdata __maybe_unused *sd = &dev->archdata;
#ifdef CONFIG_SWIOTLB
- if (sd->max_direct_dma_addr && dev->dma_ops == &swiotlb_dma_ops)
+ if (sd->max_direct_dma_addr && dev->dma_ops == &powerpc_swiotlb_dma_ops)
pfn = min_t(u64, pfn, sd->max_direct_dma_addr >> PAGE_SHIFT);
#endif
diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c
index e4d0133bbeeb..61e07c78d64f 100644
--- a/arch/powerpc/sysdev/fsl_pci.c
+++ b/arch/powerpc/sysdev/fsl_pci.c
@@ -118,7 +118,7 @@ static void setup_swiotlb_ops(struct pci_controller *hose)
{
if (ppc_swiotlb_enable) {
hose->controller_ops.dma_dev_setup = pci_dma_dev_setup_swiotlb;
- set_pci_dma_ops(&swiotlb_dma_ops);
+ set_pci_dma_ops(&powerpc_swiotlb_dma_ops);
}
}
#else
--
2.14.2
^ permalink raw reply related
* [PATCH 05/22] x86: rename swiotlb_dma_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
We'll need that name for a generic implementation soon.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/x86/kernel/pci-swiotlb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 9d3e35c33d94..0d77603c2f50 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -48,7 +48,7 @@ void x86_swiotlb_free_coherent(struct device *dev, size_t size,
dma_generic_free_coherent(dev, size, vaddr, dma_addr, attrs);
}
-static const struct dma_map_ops swiotlb_dma_ops = {
+static const struct dma_map_ops x86_swiotlb_dma_ops = {
.mapping_error = swiotlb_dma_mapping_error,
.alloc = x86_swiotlb_alloc_coherent,
.free = x86_swiotlb_free_coherent,
@@ -112,7 +112,7 @@ void __init pci_swiotlb_init(void)
{
if (swiotlb) {
swiotlb_init(0);
- dma_ops = &swiotlb_dma_ops;
+ dma_ops = &x86_swiotlb_dma_ops;
}
}
--
2.14.2
^ permalink raw reply related
* [PATCH 06/22] swiotlb: rename swiotlb_free to swiotlb_exit
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/powerpc/kernel/dma-swiotlb.c | 2 +-
arch/x86/kernel/pci-swiotlb.c | 2 +-
include/linux/swiotlb.h | 4 ++--
lib/swiotlb.c | 2 +-
4 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/kernel/dma-swiotlb.c b/arch/powerpc/kernel/dma-swiotlb.c
index 506ac4fafac5..88f3963ca30f 100644
--- a/arch/powerpc/kernel/dma-swiotlb.c
+++ b/arch/powerpc/kernel/dma-swiotlb.c
@@ -121,7 +121,7 @@ static int __init check_swiotlb_enabled(void)
if (ppc_swiotlb_enable)
swiotlb_print_info();
else
- swiotlb_free();
+ swiotlb_exit();
return 0;
}
diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 0d77603c2f50..0ee0f8f34251 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -120,7 +120,7 @@ void __init pci_swiotlb_late_init(void)
{
/* An IOMMU turned us off. */
if (!swiotlb)
- swiotlb_free();
+ swiotlb_exit();
else {
printk(KERN_INFO "PCI-DMA: "
"Using software bounce buffering for IO (SWIOTLB)\n");
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index 24ed817082ee..606375e35d87 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -115,10 +115,10 @@ extern int
swiotlb_dma_supported(struct device *hwdev, u64 mask);
#ifdef CONFIG_SWIOTLB
-extern void __init swiotlb_free(void);
+extern void __init swiotlb_exit(void);
unsigned int swiotlb_max_segment(void);
#else
-static inline void swiotlb_free(void) { }
+static inline void swiotlb_exit(void) { }
static inline unsigned int swiotlb_max_segment(void) { return 0; }
#endif
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 125c1062119f..cf5311908fa9 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -417,7 +417,7 @@ swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs)
return -ENOMEM;
}
-void __init swiotlb_free(void)
+void __init swiotlb_exit(void)
{
if (!io_tlb_orig_addr)
return;
--
2.14.2
^ permalink raw reply related
* [PATCH 07/22] swiotlb: add common swiotlb_map_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
Currently all architectures that want to use swiotlb have to implement
their own dma_map_ops instances. Provide a generic one based on the
x86 implementation which first calls into dma_direct to try a full blown
direct mapping implementation (including e.g. CMA) before falling back
allocating from the swiotlb buffer.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
include/linux/swiotlb.h | 8 ++++++++
lib/swiotlb.c | 43 +++++++++++++++++++++++++++++++++++++++++++
2 files changed, 51 insertions(+)
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index 606375e35d87..5b1f2a00491c 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -66,6 +66,12 @@ extern void swiotlb_tbl_sync_single(struct device *hwdev,
enum dma_sync_target target);
/* Accessory functions. */
+
+void *swiotlb_alloc(struct device *hwdev, size_t size, dma_addr_t *dma_handle,
+ gfp_t flags, unsigned long attrs);
+void swiotlb_free(struct device *dev, size_t size, void *vaddr,
+ dma_addr_t dma_addr, unsigned long attrs);
+
extern void
*swiotlb_alloc_coherent(struct device *hwdev, size_t size,
dma_addr_t *dma_handle, gfp_t flags);
@@ -126,4 +132,6 @@ extern void swiotlb_print_info(void);
extern int is_swiotlb_buffer(phys_addr_t paddr);
extern void swiotlb_set_max_segment(unsigned int);
+extern const struct dma_map_ops swiotlb_dma_ops;
+
#endif /* __LINUX_SWIOTLB_H */
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index cf5311908fa9..0fae2f45c3c0 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -1087,3 +1087,46 @@ swiotlb_dma_supported(struct device *hwdev, u64 mask)
return swiotlb_phys_to_dma(hwdev, io_tlb_end - 1) <= mask;
}
EXPORT_SYMBOL(swiotlb_dma_supported);
+
+#ifdef CONFIG_DMA_DIRECT_OPS
+void *swiotlb_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
+ gfp_t gfp, unsigned long attrs)
+{
+ void *vaddr;
+
+ /*
+ * Don't print a warning when the first allocation attempt fails.
+ * swiotlb_alloc_coherent() will print a warning when the DMA memory
+ * allocation ultimately failed.
+ */
+ gfp |= __GFP_NOWARN;
+
+ vaddr = dma_direct_alloc(dev, size, dma_handle, gfp, attrs);
+ if (!vaddr)
+ vaddr = swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
+ return vaddr;
+}
+
+void swiotlb_free(struct device *dev, size_t size, void *vaddr,
+ dma_addr_t dma_addr, unsigned long attrs)
+{
+ if (is_swiotlb_buffer(dma_to_phys(dev, dma_addr)))
+ swiotlb_free_coherent(dev, size, vaddr, dma_addr);
+ else
+ dma_direct_free(dev, size, vaddr, dma_addr, attrs);
+}
+
+const struct dma_map_ops swiotlb_dma_ops = {
+ .mapping_error = swiotlb_dma_mapping_error,
+ .alloc = swiotlb_alloc,
+ .free = swiotlb_free,
+ .sync_single_for_cpu = swiotlb_sync_single_for_cpu,
+ .sync_single_for_device = swiotlb_sync_single_for_device,
+ .sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
+ .sync_sg_for_device = swiotlb_sync_sg_for_device,
+ .map_sg = swiotlb_map_sg_attrs,
+ .unmap_sg = swiotlb_unmap_sg_attrs,
+ .map_page = swiotlb_map_page,
+ .unmap_page = swiotlb_unmap_page,
+};
+#endif /* CONFIG_DMA_DIRECT_OPS */
--
2.14.2
^ permalink raw reply related
* [PATCH 08/22] swiotlb: wire up ->dma_supported in swiotlb_dma_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
To properly reject too small DMA masks based on the addressability of the
bounce buffer.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
lib/swiotlb.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 0fae2f45c3c0..539fd1099ba9 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -1128,5 +1128,6 @@ const struct dma_map_ops swiotlb_dma_ops = {
.unmap_sg = swiotlb_unmap_sg_attrs,
.map_page = swiotlb_map_page,
.unmap_page = swiotlb_unmap_page,
+ .dma_supported = swiotlb_dma_supported,
};
#endif /* CONFIG_DMA_DIRECT_OPS */
--
2.14.2
^ permalink raw reply related
* [PATCH 09/22] swiotlb: refactor coherent buffer freeing
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
Factor out a new swiotlb_free_buffer helper that checks if an address
is allocated from the swiotlb bounce buffer, and if yes frees it.
This allows to simplify the swiotlb_free implemenation that uses
dma_direct_free to free the non-bounce buffer allocations.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
lib/swiotlb.c | 35 +++++++++++++++++++++--------------
1 file changed, 21 insertions(+), 14 deletions(-)
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 539fd1099ba9..1a147f1354a1 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -780,22 +780,31 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size,
}
EXPORT_SYMBOL(swiotlb_alloc_coherent);
+static bool swiotlb_free_buffer(struct device *dev, size_t size,
+ dma_addr_t dma_addr)
+{
+ phys_addr_t phys_addr = dma_to_phys(dev, dma_addr);
+
+ WARN_ON_ONCE(irqs_disabled());
+
+ if (!is_swiotlb_buffer(phys_addr))
+ return false;
+
+ /*
+ * DMA_TO_DEVICE to avoid memcpy in swiotlb_tbl_unmap_single.
+ * DMA_ATTR_SKIP_CPU_SYNC is optional.
+ */
+ swiotlb_tbl_unmap_single(dev, phys_addr, size, DMA_TO_DEVICE,
+ DMA_ATTR_SKIP_CPU_SYNC);
+ return true;
+}
+
void
swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr,
dma_addr_t dev_addr)
{
- phys_addr_t paddr = dma_to_phys(hwdev, dev_addr);
-
- WARN_ON(irqs_disabled());
- if (!is_swiotlb_buffer(paddr))
+ if (!swiotlb_free_buffer(hwdev, size, dev_addr))
free_pages((unsigned long)vaddr, get_order(size));
- else
- /*
- * DMA_TO_DEVICE to avoid memcpy in swiotlb_tbl_unmap_single.
- * DMA_ATTR_SKIP_CPU_SYNC is optional.
- */
- swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE,
- DMA_ATTR_SKIP_CPU_SYNC);
}
EXPORT_SYMBOL(swiotlb_free_coherent);
@@ -1110,9 +1119,7 @@ void *swiotlb_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
void swiotlb_free(struct device *dev, size_t size, void *vaddr,
dma_addr_t dma_addr, unsigned long attrs)
{
- if (is_swiotlb_buffer(dma_to_phys(dev, dma_addr)))
- swiotlb_free_coherent(dev, size, vaddr, dma_addr);
- else
+ if (!swiotlb_free_buffer(dev, size, dma_addr))
dma_direct_free(dev, size, vaddr, dma_addr, attrs);
}
--
2.14.2
^ permalink raw reply related
* [PATCH 10/22] swiotlb: refactor coherent buffer allocation
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
Factor out a new swiotlb_alloc_buffer helper that allocates DMA coherent
memory from the swiotlb bounce buffer.
This allows to simplify the swiotlb_alloc implemenation that uses
dma_direct_alloc to try to allocate a reachable buffer first.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
lib/swiotlb.c | 122 +++++++++++++++++++++++++++++++---------------------------
1 file changed, 65 insertions(+), 57 deletions(-)
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 1a147f1354a1..bf2d19ee91c1 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -709,75 +709,79 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
}
EXPORT_SYMBOL_GPL(swiotlb_tbl_sync_single);
-void *
-swiotlb_alloc_coherent(struct device *hwdev, size_t size,
- dma_addr_t *dma_handle, gfp_t flags)
+static inline bool dma_coherent_ok(struct device *dev, dma_addr_t addr,
+ size_t size)
{
- bool warn = !(flags & __GFP_NOWARN);
- dma_addr_t dev_addr;
- void *ret;
- int order = get_order(size);
- u64 dma_mask = DMA_BIT_MASK(32);
+ u64 mask = DMA_BIT_MASK(32);
- if (hwdev && hwdev->coherent_dma_mask)
- dma_mask = hwdev->coherent_dma_mask;
+ if (dev && dev->coherent_dma_mask)
+ mask = dev->coherent_dma_mask;
+ return addr + size - 1 <= mask;
+}
- ret = (void *)__get_free_pages(flags, order);
- if (ret) {
- dev_addr = swiotlb_virt_to_bus(hwdev, ret);
- if (dev_addr + size - 1 > dma_mask) {
- /*
- * The allocated memory isn't reachable by the device.
- */
- free_pages((unsigned long) ret, order);
- ret = NULL;
- }
- }
- if (!ret) {
- /*
- * We are either out of memory or the device can't DMA to
- * GFP_DMA memory; fall back on map_single(), which
- * will grab memory from the lowest available address range.
- */
- phys_addr_t paddr = map_single(hwdev, 0, size, DMA_FROM_DEVICE,
- warn ? 0 : DMA_ATTR_NO_WARN);
- if (paddr == SWIOTLB_MAP_ERROR)
- goto err_warn;
+static void *
+swiotlb_alloc_buffer(struct device *dev, size_t size, dma_addr_t *dma_handle,
+ unsigned long attrs)
+{
+ phys_addr_t phys_addr;
+
+ if (swiotlb_force == SWIOTLB_NO_FORCE)
+ goto out_warn;
- ret = phys_to_virt(paddr);
- dev_addr = swiotlb_phys_to_dma(hwdev, paddr);
+ phys_addr = swiotlb_tbl_map_single(dev,
+ swiotlb_phys_to_dma(dev, io_tlb_start),
+ 0, size, DMA_FROM_DEVICE, 0);
+ if (phys_addr == SWIOTLB_MAP_ERROR)
+ goto out_warn;
- /* Confirm address can be DMA'd by device */
- if (dev_addr + size - 1 > dma_mask) {
- printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n",
- (unsigned long long)dma_mask,
- (unsigned long long)dev_addr);
+ *dma_handle = swiotlb_phys_to_dma(dev, phys_addr);
- /*
- * DMA_TO_DEVICE to avoid memcpy in unmap_single.
- * The DMA_ATTR_SKIP_CPU_SYNC is optional.
- */
- swiotlb_tbl_unmap_single(hwdev, paddr,
- size, DMA_TO_DEVICE,
- DMA_ATTR_SKIP_CPU_SYNC);
- goto err_warn;
- }
- }
+ if (dma_coherent_ok(dev, *dma_handle, size))
+ goto out_unmap;
- *dma_handle = dev_addr;
- memset(ret, 0, size);
+ memset(phys_to_virt(phys_addr), 0, size);
+ return phys_to_virt(phys_addr);
- return ret;
+out_unmap:
+ dev_warn(dev, "hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n",
+ (unsigned long long)(dev ? dev->coherent_dma_mask : 0),
+ (unsigned long long)*dma_handle);
-err_warn:
- if (warn && printk_ratelimit()) {
- pr_warn("swiotlb: coherent allocation failed for device %s size=%zu\n",
- dev_name(hwdev), size);
+ /*
+ * DMA_TO_DEVICE to avoid memcpy in unmap_single.
+ * DMA_ATTR_SKIP_CPU_SYNC is optional.
+ */
+ swiotlb_tbl_unmap_single(dev, phys_addr, size, DMA_TO_DEVICE,
+ DMA_ATTR_SKIP_CPU_SYNC);
+out_warn:
+ if ((attrs & DMA_ATTR_NO_WARN) && printk_ratelimit()) {
+ dev_warn(dev,
+ "swiotlb: coherent allocation failed, size=%zu\n",
+ size);
dump_stack();
}
-
return NULL;
}
+
+void *
+swiotlb_alloc_coherent(struct device *hwdev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flags)
+{
+ int order = get_order(size);
+ unsigned long attrs = (flags & __GFP_NOWARN) ? DMA_ATTR_NO_WARN : 0;
+ void *ret;
+
+ ret = (void *)__get_free_pages(flags, order);
+ if (ret) {
+ *dma_handle = swiotlb_virt_to_bus(hwdev, ret);
+ if (dma_coherent_ok(hwdev, *dma_handle, size)) {
+ memset(ret, 0, size);
+ return ret;
+ }
+ }
+
+ return swiotlb_alloc_buffer(hwdev, size, dma_handle, attrs);
+}
EXPORT_SYMBOL(swiotlb_alloc_coherent);
static bool swiotlb_free_buffer(struct device *dev, size_t size,
@@ -1103,6 +1107,10 @@ void *swiotlb_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
{
void *vaddr;
+ /* temporary workaround: */
+ if (gfp & __GFP_NOWARN)
+ attrs |= DMA_ATTR_NO_WARN;
+
/*
* Don't print a warning when the first allocation attempt fails.
* swiotlb_alloc_coherent() will print a warning when the DMA memory
@@ -1112,7 +1120,7 @@ void *swiotlb_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
vaddr = dma_direct_alloc(dev, size, dma_handle, gfp, attrs);
if (!vaddr)
- vaddr = swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
+ vaddr = swiotlb_alloc_buffer(dev, size, dma_handle, attrs);
return vaddr;
}
--
2.14.2
^ permalink raw reply related
* [PATCH 11/22] swiotlb: remove various exports
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
All these symbols are only used by arch dma_ops implementations or
xen-swiotlb. None of which can be modular.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
lib/swiotlb.c | 13 -------------
1 file changed, 13 deletions(-)
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index bf2d19ee91c1..1eac51ff77a4 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -605,7 +605,6 @@ phys_addr_t swiotlb_tbl_map_single(struct device *hwdev,
return tlb_addr;
}
-EXPORT_SYMBOL_GPL(swiotlb_tbl_map_single);
/*
* Allocates bounce buffer and returns its kernel virtual address.
@@ -675,7 +674,6 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr,
}
spin_unlock_irqrestore(&io_tlb_lock, flags);
}
-EXPORT_SYMBOL_GPL(swiotlb_tbl_unmap_single);
void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
size_t size, enum dma_data_direction dir,
@@ -707,7 +705,6 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
BUG();
}
}
-EXPORT_SYMBOL_GPL(swiotlb_tbl_sync_single);
static inline bool dma_coherent_ok(struct device *dev, dma_addr_t addr,
size_t size)
@@ -884,7 +881,6 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
return swiotlb_phys_to_dma(dev, io_tlb_overflow_buffer);
}
-EXPORT_SYMBOL_GPL(swiotlb_map_page);
/*
* Unmap a single streaming mode DMA translation. The dma_addr and size must
@@ -925,7 +921,6 @@ void swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
{
unmap_single(hwdev, dev_addr, size, dir, attrs);
}
-EXPORT_SYMBOL_GPL(swiotlb_unmap_page);
/*
* Make physical memory consistent for a single streaming mode DMA translation
@@ -963,7 +958,6 @@ swiotlb_sync_single_for_cpu(struct device *hwdev, dma_addr_t dev_addr,
{
swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_CPU);
}
-EXPORT_SYMBOL(swiotlb_sync_single_for_cpu);
void
swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
@@ -971,7 +965,6 @@ swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
{
swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_DEVICE);
}
-EXPORT_SYMBOL(swiotlb_sync_single_for_device);
/*
* Map a set of buffers described by scatterlist in streaming mode for DMA.
@@ -1023,7 +1016,6 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems,
}
return nelems;
}
-EXPORT_SYMBOL(swiotlb_map_sg_attrs);
/*
* Unmap a set of streaming mode DMA translations. Again, cpu read rules
@@ -1043,7 +1035,6 @@ swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
unmap_single(hwdev, sg->dma_address, sg_dma_len(sg), dir,
attrs);
}
-EXPORT_SYMBOL(swiotlb_unmap_sg_attrs);
/*
* Make physical memory consistent for a set of streaming mode DMA translations
@@ -1071,7 +1062,6 @@ swiotlb_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
{
swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_CPU);
}
-EXPORT_SYMBOL(swiotlb_sync_sg_for_cpu);
void
swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
@@ -1079,14 +1069,12 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
{
swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_DEVICE);
}
-EXPORT_SYMBOL(swiotlb_sync_sg_for_device);
int
swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
{
return (dma_addr == swiotlb_phys_to_dma(hwdev, io_tlb_overflow_buffer));
}
-EXPORT_SYMBOL(swiotlb_dma_mapping_error);
/*
* Return whether the given device DMA address mask can be supported
@@ -1099,7 +1087,6 @@ swiotlb_dma_supported(struct device *hwdev, u64 mask)
{
return swiotlb_phys_to_dma(hwdev, io_tlb_end - 1) <= mask;
}
-EXPORT_SYMBOL(swiotlb_dma_supported);
#ifdef CONFIG_DMA_DIRECT_OPS
void *swiotlb_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
--
2.14.2
^ permalink raw reply related
* [PATCH 12/22] ia64: replace ZONE_DMA with ZONE_DMA32
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
ia64 uses ZONE_DMA for allocations below 32-bits. These days we
name the zone for that ZONE_DMA32, which will allow to use the
dma-direct and generic swiotlb code as-is, so rename it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/ia64/Kconfig | 2 +-
arch/ia64/kernel/pci-swiotlb.c | 2 +-
arch/ia64/mm/contig.c | 4 ++--
arch/ia64/mm/discontig.c | 8 ++++----
4 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 4d18fca885ee..888acdb163cb 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -66,7 +66,7 @@ config 64BIT
select ATA_NONSTANDARD if ATA
default y
-config ZONE_DMA
+config ZONE_DMA32
def_bool y
depends on !IA64_SGI_SN2
diff --git a/arch/ia64/kernel/pci-swiotlb.c b/arch/ia64/kernel/pci-swiotlb.c
index f1ae873a8c35..4a9a6e58ad6a 100644
--- a/arch/ia64/kernel/pci-swiotlb.c
+++ b/arch/ia64/kernel/pci-swiotlb.c
@@ -20,7 +20,7 @@ static void *ia64_swiotlb_alloc_coherent(struct device *dev, size_t size,
unsigned long attrs)
{
if (dev->coherent_dma_mask != DMA_BIT_MASK(64))
- gfp |= GFP_DMA;
+ gfp |= GFP_DMA32;
return swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
}
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index 52715a71aede..7d64b30913d1 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -237,9 +237,9 @@ paging_init (void)
unsigned long max_zone_pfns[MAX_NR_ZONES];
memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
- max_zone_pfns[ZONE_DMA] = max_dma;
+ max_zone_pfns[ZONE_DMA32] = max_dma;
#endif
max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 9b2d994cddf6..ac46f0d60b66 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -38,7 +38,7 @@ struct early_node_data {
struct ia64_node_data *node_data;
unsigned long pernode_addr;
unsigned long pernode_size;
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
unsigned long num_dma_physpages;
#endif
unsigned long min_pfn;
@@ -669,7 +669,7 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n
{
unsigned long end = start + len;
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
if (start <= __pa(MAX_DMA_ADDRESS))
mem_data[node].num_dma_physpages +=
(min(end, __pa(MAX_DMA_ADDRESS)) - start) >>PAGE_SHIFT;
@@ -724,8 +724,8 @@ void __init paging_init(void)
}
memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
-#ifdef CONFIG_ZONE_DMA
- max_zone_pfns[ZONE_DMA] = max_dma;
+#ifdef CONFIG_ZONE_DMA32
+ max_zone_pfns[ZONE_DMA32] = max_dma;
#endif
max_zone_pfns[ZONE_NORMAL] = max_pfn;
free_area_init_nodes(max_zone_pfns);
--
2.14.2
^ permalink raw reply related
* [PATCH 13/22] ia64: use generic swiotlb_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
These are identical to the ia64 ops, and would also support CMA
if enabled on ia64.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/ia64/Kconfig | 5 +++++
arch/ia64/hp/common/hwsw_iommu.c | 4 ++--
arch/ia64/hp/common/sba_iommu.c | 6 +++---
arch/ia64/kernel/pci-swiotlb.c | 38 +++-----------------------------------
4 files changed, 13 insertions(+), 40 deletions(-)
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 888acdb163cb..29148fe4bf5a 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -146,6 +146,7 @@ config IA64_GENERIC
bool "generic"
select NUMA
select ACPI_NUMA
+ select DMA_DIRECT_OPS
select SWIOTLB
select PCI_MSI
help
@@ -166,6 +167,7 @@ config IA64_GENERIC
config IA64_DIG
bool "DIG-compliant"
+ select DMA_DIRECT_OPS
select SWIOTLB
config IA64_DIG_VTD
@@ -181,6 +183,7 @@ config IA64_HP_ZX1
config IA64_HP_ZX1_SWIOTLB
bool "HP-zx1/sx1000 with software I/O TLB"
+ select DMA_DIRECT_OPS
select SWIOTLB
help
Build a kernel that runs on HP zx1 and sx1000 systems even when they
@@ -204,6 +207,7 @@ config IA64_SGI_UV
bool "SGI-UV"
select NUMA
select ACPI_NUMA
+ select DMA_DIRECT_OPS
select SWIOTLB
help
Selecting this option will optimize the kernel for use on UV based
@@ -214,6 +218,7 @@ config IA64_SGI_UV
config IA64_HP_SIM
bool "Ski-simulator"
+ select DMA_DIRECT_OPS
select SWIOTLB
depends on !PM
diff --git a/arch/ia64/hp/common/hwsw_iommu.c b/arch/ia64/hp/common/hwsw_iommu.c
index 41279f0442bd..58969039bed2 100644
--- a/arch/ia64/hp/common/hwsw_iommu.c
+++ b/arch/ia64/hp/common/hwsw_iommu.c
@@ -19,7 +19,7 @@
#include <linux/export.h>
#include <asm/machvec.h>
-extern const struct dma_map_ops sba_dma_ops, ia64_swiotlb_dma_ops;
+extern const struct dma_map_ops sba_dma_ops;
/* swiotlb declarations & definitions: */
extern int swiotlb_late_init_with_default_size (size_t size);
@@ -38,7 +38,7 @@ static inline int use_swiotlb(struct device *dev)
const struct dma_map_ops *hwsw_dma_get_ops(struct device *dev)
{
if (use_swiotlb(dev))
- return &ia64_swiotlb_dma_ops;
+ return &swiotlb_dma_ops;
return &sba_dma_ops;
}
EXPORT_SYMBOL(hwsw_dma_get_ops);
diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index 8c0a9ae6afec..aec4a3354abe 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -2096,7 +2096,7 @@ static int __init acpi_sba_ioc_init_acpi(void)
/* This has to run before acpi_scan_init(). */
arch_initcall(acpi_sba_ioc_init_acpi);
-extern const struct dma_map_ops ia64_swiotlb_dma_ops;
+extern const struct dma_map_ops swiotlb_dma_ops;
static int __init
sba_init(void)
@@ -2111,7 +2111,7 @@ sba_init(void)
* a successful kdump kernel boot is to use the swiotlb.
*/
if (is_kdump_kernel()) {
- dma_ops = &ia64_swiotlb_dma_ops;
+ dma_ops = &swiotlb_dma_ops;
if (swiotlb_late_init_with_default_size(64 * (1<<20)) != 0)
panic("Unable to initialize software I/O TLB:"
" Try machvec=dig boot option");
@@ -2133,7 +2133,7 @@ sba_init(void)
* If we didn't find something sba_iommu can claim, we
* need to setup the swiotlb and switch to the dig machvec.
*/
- dma_ops = &ia64_swiotlb_dma_ops;
+ dma_ops = &swiotlb_dma_ops;
if (swiotlb_late_init_with_default_size(64 * (1<<20)) != 0)
panic("Unable to find SBA IOMMU or initialize "
"software I/O TLB: Try machvec=dig boot option");
diff --git a/arch/ia64/kernel/pci-swiotlb.c b/arch/ia64/kernel/pci-swiotlb.c
index 4a9a6e58ad6a..0f8d5fbd86bd 100644
--- a/arch/ia64/kernel/pci-swiotlb.c
+++ b/arch/ia64/kernel/pci-swiotlb.c
@@ -6,8 +6,7 @@
#include <linux/cache.h>
#include <linux/module.h>
#include <linux/dma-mapping.h>
-
-#include <asm/swiotlb.h>
+#include <linux/swiotlb.h>
#include <asm/dma.h>
#include <asm/iommu.h>
#include <asm/machvec.h>
@@ -15,40 +14,9 @@
int swiotlb __read_mostly;
EXPORT_SYMBOL(swiotlb);
-static void *ia64_swiotlb_alloc_coherent(struct device *dev, size_t size,
- dma_addr_t *dma_handle, gfp_t gfp,
- unsigned long attrs)
-{
- if (dev->coherent_dma_mask != DMA_BIT_MASK(64))
- gfp |= GFP_DMA32;
- return swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
-}
-
-static void ia64_swiotlb_free_coherent(struct device *dev, size_t size,
- void *vaddr, dma_addr_t dma_addr,
- unsigned long attrs)
-{
- swiotlb_free_coherent(dev, size, vaddr, dma_addr);
-}
-
-const struct dma_map_ops ia64_swiotlb_dma_ops = {
- .alloc = ia64_swiotlb_alloc_coherent,
- .free = ia64_swiotlb_free_coherent,
- .map_page = swiotlb_map_page,
- .unmap_page = swiotlb_unmap_page,
- .map_sg = swiotlb_map_sg_attrs,
- .unmap_sg = swiotlb_unmap_sg_attrs,
- .sync_single_for_cpu = swiotlb_sync_single_for_cpu,
- .sync_single_for_device = swiotlb_sync_single_for_device,
- .sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
- .sync_sg_for_device = swiotlb_sync_sg_for_device,
- .dma_supported = swiotlb_dma_supported,
- .mapping_error = swiotlb_dma_mapping_error,
-};
-
void __init swiotlb_dma_init(void)
{
- dma_ops = &ia64_swiotlb_dma_ops;
+ dma_ops = &swiotlb_dma_ops;
swiotlb_init(1);
}
@@ -60,7 +28,7 @@ void __init pci_swiotlb_init(void)
printk(KERN_INFO "PCI-DMA: Re-initialize machine vector.\n");
machvec_init("dig");
swiotlb_init(1);
- dma_ops = &ia64_swiotlb_dma_ops;
+ dma_ops = &swiotlb_dma_ops;
#else
panic("Unable to find Intel IOMMU");
#endif
--
2.14.2
^ permalink raw reply related
* [PATCH 14/22] ia64: clean up swiotlb support
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
Move the few remaining bits of swiotlb glue towards their callers,
and remove the pointless on ia64 swiotlb variable.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/ia64/include/asm/dma-mapping.h | 1 -
arch/ia64/include/asm/swiotlb.h | 18 ------------------
arch/ia64/kernel/dma-mapping.c | 9 +++++++++
arch/ia64/kernel/pci-dma.c | 12 ++++++++++--
arch/ia64/kernel/pci-swiotlb.c | 36 ------------------------------------
drivers/iommu/intel-iommu.c | 2 +-
6 files changed, 20 insertions(+), 58 deletions(-)
delete mode 100644 arch/ia64/include/asm/swiotlb.h
delete mode 100644 arch/ia64/kernel/pci-swiotlb.c
diff --git a/arch/ia64/include/asm/dma-mapping.h b/arch/ia64/include/asm/dma-mapping.h
index eabee56d995c..76e4d6632d68 100644
--- a/arch/ia64/include/asm/dma-mapping.h
+++ b/arch/ia64/include/asm/dma-mapping.h
@@ -8,7 +8,6 @@
*/
#include <asm/machvec.h>
#include <linux/scatterlist.h>
-#include <asm/swiotlb.h>
#include <linux/dma-debug.h>
#define ARCH_HAS_DMA_GET_REQUIRED_MASK
diff --git a/arch/ia64/include/asm/swiotlb.h b/arch/ia64/include/asm/swiotlb.h
deleted file mode 100644
index 841e2c7d0b21..000000000000
--- a/arch/ia64/include/asm/swiotlb.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef ASM_IA64__SWIOTLB_H
-#define ASM_IA64__SWIOTLB_H
-
-#include <linux/dma-mapping.h>
-#include <linux/swiotlb.h>
-
-#ifdef CONFIG_SWIOTLB
-extern int swiotlb;
-extern void pci_swiotlb_init(void);
-#else
-#define swiotlb 0
-static inline void pci_swiotlb_init(void)
-{
-}
-#endif
-
-#endif /* ASM_IA64__SWIOTLB_H */
diff --git a/arch/ia64/kernel/dma-mapping.c b/arch/ia64/kernel/dma-mapping.c
index 7a82c9259609..f2d57e66fd86 100644
--- a/arch/ia64/kernel/dma-mapping.c
+++ b/arch/ia64/kernel/dma-mapping.c
@@ -1,5 +1,6 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/dma-mapping.h>
+#include <linux/swiotlb.h>
#include <linux/export.h>
/* Set this to 1 if there is a HW IOMMU in the system */
@@ -23,3 +24,11 @@ const struct dma_map_ops *dma_get_ops(struct device *dev)
return dma_ops;
}
EXPORT_SYMBOL(dma_get_ops);
+
+#ifdef CONFIG_SWIOTLB
+void __init swiotlb_dma_init(void)
+{
+ dma_ops = &swiotlb_dma_ops;
+ swiotlb_init(1);
+}
+#endif
diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c
index 3ba87c22dfbc..35e0cad33b7d 100644
--- a/arch/ia64/kernel/pci-dma.c
+++ b/arch/ia64/kernel/pci-dma.c
@@ -104,8 +104,16 @@ void __init pci_iommu_alloc(void)
detect_intel_iommu();
#ifdef CONFIG_SWIOTLB
- pci_swiotlb_init();
-#endif
+ if (!iommu_detected) {
+#ifdef CONFIG_IA64_GENERIC
+ printk(KERN_INFO "PCI-DMA: Re-initialize machine vector.\n");
+ machvec_init("dig");
+ swiotlb_dma_init();
+#else
+ panic("Unable to find Intel IOMMU");
+#endif /* CONFIG_IA64_GENERIC */
+ }
+#endif /* CONFIG_SWIOTLB */
}
#endif
diff --git a/arch/ia64/kernel/pci-swiotlb.c b/arch/ia64/kernel/pci-swiotlb.c
deleted file mode 100644
index 0f8d5fbd86bd..000000000000
--- a/arch/ia64/kernel/pci-swiotlb.c
+++ /dev/null
@@ -1,36 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Glue code to lib/swiotlb.c */
-
-#include <linux/pci.h>
-#include <linux/gfp.h>
-#include <linux/cache.h>
-#include <linux/module.h>
-#include <linux/dma-mapping.h>
-#include <linux/swiotlb.h>
-#include <asm/dma.h>
-#include <asm/iommu.h>
-#include <asm/machvec.h>
-
-int swiotlb __read_mostly;
-EXPORT_SYMBOL(swiotlb);
-
-void __init swiotlb_dma_init(void)
-{
- dma_ops = &swiotlb_dma_ops;
- swiotlb_init(1);
-}
-
-void __init pci_swiotlb_init(void)
-{
- if (!iommu_detected) {
-#ifdef CONFIG_IA64_GENERIC
- swiotlb = 1;
- printk(KERN_INFO "PCI-DMA: Re-initialize machine vector.\n");
- machvec_init("dig");
- swiotlb_init(1);
- dma_ops = &swiotlb_dma_ops;
-#else
- panic("Unable to find Intel IOMMU");
-#endif
- }
-}
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 4a2de34895ec..a1373cf34326 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4808,7 +4808,7 @@ int __init intel_iommu_init(void)
up_write(&dmar_global_lock);
pr_info("Intel(R) Virtualization Technology for Directed I/O\n");
-#ifdef CONFIG_SWIOTLB
+#if defined(CONFIG_X86) && defined(CONFIG_SWIOTLB)
swiotlb = 0;
#endif
dma_ops = &intel_dma_ops;
--
2.14.2
^ permalink raw reply related
* [PATCH 15/22] ia64: remove an ifdef around the content of pci-dma.c
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
The file is only compiled if CONFIG_INTEL_IOMMU is set to start with.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/ia64/kernel/pci-dma.c | 7 -------
1 file changed, 7 deletions(-)
diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c
index 35e0cad33b7d..b5df084c0af4 100644
--- a/arch/ia64/kernel/pci-dma.c
+++ b/arch/ia64/kernel/pci-dma.c
@@ -12,12 +12,7 @@
#include <asm/iommu.h>
#include <asm/machvec.h>
#include <linux/dma-mapping.h>
-
-
-#ifdef CONFIG_INTEL_IOMMU
-
#include <linux/kernel.h>
-
#include <asm/page.h>
dma_addr_t bad_dma_address __read_mostly;
@@ -115,5 +110,3 @@ void __init pci_iommu_alloc(void)
}
#endif /* CONFIG_SWIOTLB */
}
-
-#endif
--
2.14.2
^ permalink raw reply related
* [PATCH 16/22] unicore32: use generic swiotlb_ops
From: Christoph Hellwig @ 2018-01-10 8:09 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20180110080932.14157-1-hch@lst.de>
These are identical to the unicore32 ops, and would also support CMA
if enabled on unicore32.
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
arch/unicore32/include/asm/dma-mapping.h | 9 +-----
arch/unicore32/mm/Kconfig | 1 +
arch/unicore32/mm/Makefile | 2 --
arch/unicore32/mm/dma-swiotlb.c | 48 --------------------------------
4 files changed, 2 insertions(+), 58 deletions(-)
delete mode 100644 arch/unicore32/mm/dma-swiotlb.c
diff --git a/arch/unicore32/include/asm/dma-mapping.h b/arch/unicore32/include/asm/dma-mapping.h
index f2bfec273aa7..790bc2ef4af2 100644
--- a/arch/unicore32/include/asm/dma-mapping.h
+++ b/arch/unicore32/include/asm/dma-mapping.h
@@ -12,18 +12,11 @@
#ifndef __UNICORE_DMA_MAPPING_H__
#define __UNICORE_DMA_MAPPING_H__
-#ifdef __KERNEL__
-
-#include <linux/mm_types.h>
-#include <linux/scatterlist.h>
#include <linux/swiotlb.h>
-extern const struct dma_map_ops swiotlb_dma_map_ops;
-
static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
{
- return &swiotlb_dma_map_ops;
+ return &swiotlb_dma_ops;
}
-#endif /* __KERNEL__ */
#endif
diff --git a/arch/unicore32/mm/Kconfig b/arch/unicore32/mm/Kconfig
index c256460cd363..e9154a59d561 100644
--- a/arch/unicore32/mm/Kconfig
+++ b/arch/unicore32/mm/Kconfig
@@ -42,6 +42,7 @@ config CPU_TLB_SINGLE_ENTRY_DISABLE
config SWIOTLB
def_bool y
+ select DMA_DIRECT_OPS
config IOMMU_HELPER
def_bool SWIOTLB
diff --git a/arch/unicore32/mm/Makefile b/arch/unicore32/mm/Makefile
index 681c0ef5ec9e..8106260583ab 100644
--- a/arch/unicore32/mm/Makefile
+++ b/arch/unicore32/mm/Makefile
@@ -6,8 +6,6 @@
obj-y := extable.o fault.o init.o pgd.o mmu.o
obj-y += flush.o ioremap.o
-obj-$(CONFIG_SWIOTLB) += dma-swiotlb.o
-
obj-$(CONFIG_MODULES) += proc-syms.o
obj-$(CONFIG_ALIGNMENT_TRAP) += alignment.o
diff --git a/arch/unicore32/mm/dma-swiotlb.c b/arch/unicore32/mm/dma-swiotlb.c
deleted file mode 100644
index 525413d6690e..000000000000
--- a/arch/unicore32/mm/dma-swiotlb.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * Contains routines needed to support swiotlb for UniCore32.
- *
- * Copyright (C) 2010 Guan Xuetao
- *
- * 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.
- */
-#include <linux/pci.h>
-#include <linux/cache.h>
-#include <linux/module.h>
-#include <linux/dma-mapping.h>
-#include <linux/swiotlb.h>
-#include <linux/bootmem.h>
-
-#include <asm/dma.h>
-
-static void *unicore_swiotlb_alloc_coherent(struct device *dev, size_t size,
- dma_addr_t *dma_handle, gfp_t flags,
- unsigned long attrs)
-{
- return swiotlb_alloc_coherent(dev, size, dma_handle, flags);
-}
-
-static void unicore_swiotlb_free_coherent(struct device *dev, size_t size,
- void *vaddr, dma_addr_t dma_addr,
- unsigned long attrs)
-{
- swiotlb_free_coherent(dev, size, vaddr, dma_addr);
-}
-
-const struct dma_map_ops swiotlb_dma_map_ops = {
- .alloc = unicore_swiotlb_alloc_coherent,
- .free = unicore_swiotlb_free_coherent,
- .map_sg = swiotlb_map_sg_attrs,
- .unmap_sg = swiotlb_unmap_sg_attrs,
- .dma_supported = swiotlb_dma_supported,
- .map_page = swiotlb_map_page,
- .unmap_page = swiotlb_unmap_page,
- .sync_single_for_cpu = swiotlb_sync_single_for_cpu,
- .sync_single_for_device = swiotlb_sync_single_for_device,
- .sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
- .sync_sg_for_device = swiotlb_sync_sg_for_device,
- .mapping_error = swiotlb_dma_mapping_error,
-};
-EXPORT_SYMBOL(swiotlb_dma_map_ops);
--
2.14.2
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox