This patch series is v9 of a previous posting:
https://lkml.org/lkml/2016/11/16/386
v8 -> v9
- Updated bypass flag handling in ARM SMMU v3 according to
reviews
- Removed SMMUv1/v2 configuration interrupt handling
- Rebased against v4.9-rc5
- Updated tags
v7 -> v8
- Renamed fwnode iommu_ops registration API according to review
- Minor change in ARM SMMU driver DT/ACPI split
- Added review tags
v6 -> v7
- Rebased against v4.9-rc4
- Fixed IORT probing on ACPI systems with missing IORT table
- Fixed SMMUv1/v2 global interrupt detection
- Updated iommu_ops firmware look-up
v5 -> v6
- Rebased against v4.9-rc1
- Changed FWNODE_IOMMU to FWNODE_ACPI_STATIC
- Moved platform devices creation into IORT code
- Updated fwnode handling
- Added default dma masks initialization
v4 -> v5
- Added SMMUv1/v2 support
- Rebased against v4.8-rc5 and dependencies series
- Consolidated IORT platform devices creation
v3 -> v4
- Added single mapping API (for IORT named components)
- Fixed arm_smmu_iort_xlate() return value
- Reworked fwnode registration and platform device creation
ordering to fix probe ordering dependencies
- Added code to keep device_node ref count with new iommu
fwspec API
- Added patch to make iommu_fwspec arch agnostic
- Dropped RFC status
- Rebased against v4.8-rc2
v2 -> v3
- Rebased on top of dependencies series [1][2][3](v4.7-rc3)
- Added back reliance on ACPI early probing infrastructure
- Patch[1-3] merged through other dependent series
- Added back IOMMU fwnode generalization
- Move SMMU v3 static functions configuration to IORT code
- Implemented generic IOMMU fwspec API
- Added code to implement fwnode platform device look-up
v1 -> v2:
- Rebased on top of dependencies series [1][2][3](v4.7-rc1)
- Removed IOMMU fwnode generalization
- Implemented ARM SMMU v3 ACPI probing instead of ARM SMMU v2
owing to patch series dependencies [1]
- Moved platform device creation logic to IORT code to
generalize its usage for ARM SMMU v1-v2-v3 components
- Removed reliance on ACPI early device probing
- Created IORT specific iommu_xlate() translation hook leaving
OF code unchanged according to v1 reviews
The ACPI IORT table provides information that allows instantiating
ARM SMMU devices and carrying out id mappings between components on
ARM based systems (devices, IOMMUs, interrupt controllers).
http://infocenter.arm.com/help/topic/com.arm.doc.den0049b/DEN0049B_IO_Remapping_Table.pdf
Building on basic IORT support, this patchset enables ARM SMMUs support
on ACPI systems.
Most of the code is aimed at building the required generic ACPI
infrastructure to create and enable IOMMU components and to bring
the IOMMU infrastructure for ACPI on par with DT, which is going to
make future ARM SMMU components easier to integrate.
PATCH (1) adds a FWNODE_ACPI_STATIC type to the struct fwnode_handle type.
It is required to attach a fwnode identifier to platform
devices allocated/detected through static ACPI table entries
(ie IORT tables entries).
IOMMU devices have to have an identifier to look them up
eg IOMMU core layer carrying out id translation. This can be
done through a fwnode_handle (ie IOMMU platform devices created
out of IORT tables are not ACPI devices hence they can't be
allocated as such, otherwise they would have a fwnode_handle of
type FWNODE_ACPI).
PATCH (2) makes use of the ACPI early probing API to add a linker script
section for probing devices via IORT ACPI kernel code.
PATCH (3) provides IORT support for registering IOMMU IORT node through
their fwnode handle.
PATCH (4) make of_iommu_{set/get}_ops() functions DT agnostic and
rename the registration API.
PATCH (5) convert ARM SMMU driver to use fwnode instead of of_node as
look-up and iommu_ops retrieval token.
PATCH (6) convert ARM SMMU v3 driver to use fwnode instead of of_node as
look-up and iommu_ops retrieval token.
PATCH (7) implements the of_dma_configure() API in ACPI world -
acpi_dma_configure() - and patches PCI and ACPI core code to
start making use of it.
PATCH (8) provides an IORT function to detect existence of specific type
of IORT components.
PATCH (9) creates the kernel infrastructure required to create ARM SMMU
platform devices for IORT nodes.
PATCH (10) refactors the ARM SMMU v3 driver so that the init functions are
split in a way that groups together code that probes through DT
and code that carries out HW registers FW agnostic probing, in
preparation for adding the ACPI probing path.
PATCH (11) adds ARM SMMU v3 IORT IOMMU operations to create and probe
ARM SMMU v3 components.
PATCH (12) refactors the ARM SMMU v1/v2 driver so that the init functions
are split in a way that groups together code that probes
through DT and code that carries out HW registers FW agnostic
probing, in preparation for adding the ACPI probing path.
PATCH (13) adds ARM SMMU v1/v2 IORT IOMMU operations to create and
probe ARM SMMU v1/v2 components.
PATCH (14) Extend the IORT iort_node_map_rid() to work on a type mask
instead of a single type so that the translation API can
be used on a range of components.
PATCH (15) Add IORT API to carry out id mappings for components that do
do not have an input identifier/RIDs (ie named components).
PATCH (16) provides IORT infrastructure to carry out IOMMU configuration
for devices and hook it up to the previously introduced ACPI
DMA configure API.
This patchset is provided for review/testing purposes here:
git://git.kernel.org/pub/scm/linux/kernel/git/lpieralisi/linux.git acpi/iort-smmu-v9
Tested on Juno and FVP models for ARM SMMU v1 and v3 probing path.
Lorenzo Pieralisi (16):
drivers: acpi: add FWNODE_ACPI_STATIC fwnode type
drivers: acpi: iort: introduce linker section for IORT entries probing
drivers: acpi: iort: add support for IOMMU fwnode registration
drivers: iommu: make of_iommu_set/get_ops() DT agnostic
drivers: iommu: arm-smmu: convert struct device of_node to fwnode
usage
drivers: iommu: arm-smmu-v3: convert struct device of_node to fwnode
usage
drivers: acpi: implement acpi_dma_configure
drivers: acpi: iort: add node match function
drivers: acpi: iort: add support for ARM SMMU platform devices
creation
drivers: iommu: arm-smmu-v3: split probe functions into DT/generic
portions
drivers: iommu: arm-smmu-v3: add IORT configuration
drivers: iommu: arm-smmu: split probe functions into DT/generic
portions
drivers: iommu: arm-smmu: add IORT configuration
drivers: acpi: iort: replace rid map type with type mask
drivers: acpi: iort: add single mapping function
drivers: acpi: iort: introduce iort_iommu_configure
drivers/acpi/arm64/iort.c | 585 +++++++++++++++++++++++++++++++++++++-
drivers/acpi/glue.c | 4 +-
drivers/acpi/scan.c | 45 +++
drivers/iommu/arm-smmu-v3.c | 102 +++++--
drivers/iommu/arm-smmu.c | 148 ++++++++--
drivers/iommu/iommu.c | 40 +++
drivers/iommu/of_iommu.c | 39 ---
drivers/pci/probe.c | 3 +-
include/acpi/acpi_bus.h | 2 +
include/asm-generic/vmlinux.lds.h | 1 +
include/linux/acpi.h | 26 ++
include/linux/acpi_iort.h | 14 +
include/linux/fwnode.h | 3 +-
include/linux/iommu.h | 14 +
include/linux/of_iommu.h | 12 +-
15 files changed, 934 insertions(+), 104 deletions(-)
--
2.10.0
On systems booting with a device tree, every struct device is associated
with a struct device_node, that provides its DT firmware representation.
The device node can be used in generic kernel contexts (eg IRQ
translation, IOMMU streamid mapping), to retrieve the properties
associated with the device and carry out kernel operations accordingly.
Owing to the 1:1 relationship between the device and its device_node,
the device_node can also be used as a look-up token for the device (eg
looking up a device through its device_node), to retrieve the device in
kernel paths where the device_node is available.
On systems booting with ACPI, the same abstraction provided by
the device_node is required to provide look-up functionality.
The struct acpi_device, that represents firmware objects in the
ACPI namespace already includes a struct fwnode_handle of
type FWNODE_ACPI as their member; the same abstraction is missing
though for devices that are instantiated out of static ACPI tables
entries (eg ARM SMMU devices).
Add a new fwnode_handle type to associate devices created out
of static ACPI table entries to the respective firmware components
and create a simple ACPI core layer interface to dynamically allocate
and free the corresponding firmware nodes so that kernel subsystems
can use it to instantiate the nodes and associate them with the
respective devices.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Hanjun Guo <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
include/linux/acpi.h | 21 +++++++++++++++++++++
include/linux/fwnode.h | 3 ++-
2 files changed, 23 insertions(+), 1 deletion(-)
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 61a3d90..996a29c 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -56,6 +56,27 @@ static inline acpi_handle acpi_device_handle(struct acpi_device *adev)
acpi_fwnode_handle(adev) : NULL)
#define ACPI_HANDLE(dev) acpi_device_handle(ACPI_COMPANION(dev))
+static inline struct fwnode_handle *acpi_alloc_fwnode_static(void)
+{
+ struct fwnode_handle *fwnode;
+
+ fwnode = kzalloc(sizeof(struct fwnode_handle), GFP_KERNEL);
+ if (!fwnode)
+ return NULL;
+
+ fwnode->type = FWNODE_ACPI_STATIC;
+
+ return fwnode;
+}
+
+static inline void acpi_free_fwnode_static(struct fwnode_handle *fwnode)
+{
+ if (WARN_ON(!fwnode || fwnode->type != FWNODE_ACPI_STATIC))
+ return;
+
+ kfree(fwnode);
+}
+
/**
* ACPI_DEVICE_CLASS - macro used to describe an ACPI device with
* the PCI-defined class-code information
diff --git a/include/linux/fwnode.h b/include/linux/fwnode.h
index 8516717..8bd28ce 100644
--- a/include/linux/fwnode.h
+++ b/include/linux/fwnode.h
@@ -17,8 +17,9 @@ enum fwnode_type {
FWNODE_OF,
FWNODE_ACPI,
FWNODE_ACPI_DATA,
+ FWNODE_ACPI_STATIC,
FWNODE_PDATA,
- FWNODE_IRQCHIP,
+ FWNODE_IRQCHIP
};
struct fwnode_handle {
--
2.10.0
The ACPI IORT table provide entries for IOMMU (aka SMMU in ARM world)
components that allow creating the kernel data structures required to
probe and initialize the IOMMU devices.
This patch provides support in the IORT kernel code to register IOMMU
components and their respective fwnode.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Hanjun Guo <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
drivers/acpi/arm64/iort.c | 86 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 86 insertions(+)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index 2c46ebc..1ac2720 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -20,7 +20,9 @@
#include <linux/acpi_iort.h>
#include <linux/kernel.h>
+#include <linux/list.h>
#include <linux/pci.h>
+#include <linux/slab.h>
struct iort_its_msi_chip {
struct list_head list;
@@ -28,6 +30,90 @@ struct iort_its_msi_chip {
u32 translation_id;
};
+struct iort_fwnode {
+ struct list_head list;
+ struct acpi_iort_node *iort_node;
+ struct fwnode_handle *fwnode;
+};
+static LIST_HEAD(iort_fwnode_list);
+static DEFINE_SPINLOCK(iort_fwnode_lock);
+
+/**
+ * iort_set_fwnode() - Create iort_fwnode and use it to register
+ * iommu data in the iort_fwnode_list
+ *
+ * @node: IORT table node associated with the IOMMU
+ * @fwnode: fwnode associated with the IORT node
+ *
+ * Returns: 0 on success
+ * <0 on failure
+ */
+static inline int iort_set_fwnode(struct acpi_iort_node *iort_node,
+ struct fwnode_handle *fwnode)
+{
+ struct iort_fwnode *np;
+
+ np = kzalloc(sizeof(struct iort_fwnode), GFP_ATOMIC);
+
+ if (WARN_ON(!np))
+ return -ENOMEM;
+
+ INIT_LIST_HEAD(&np->list);
+ np->iort_node = iort_node;
+ np->fwnode = fwnode;
+
+ spin_lock(&iort_fwnode_lock);
+ list_add_tail(&np->list, &iort_fwnode_list);
+ spin_unlock(&iort_fwnode_lock);
+
+ return 0;
+}
+
+/**
+ * iort_get_fwnode() - Retrieve fwnode associated with an IORT node
+ *
+ * @node: IORT table node to be looked-up
+ *
+ * Returns: fwnode_handle pointer on success, NULL on failure
+ */
+static inline
+struct fwnode_handle *iort_get_fwnode(struct acpi_iort_node *node)
+{
+ struct iort_fwnode *curr;
+ struct fwnode_handle *fwnode = NULL;
+
+ spin_lock(&iort_fwnode_lock);
+ list_for_each_entry(curr, &iort_fwnode_list, list) {
+ if (curr->iort_node == node) {
+ fwnode = curr->fwnode;
+ break;
+ }
+ }
+ spin_unlock(&iort_fwnode_lock);
+
+ return fwnode;
+}
+
+/**
+ * iort_delete_fwnode() - Delete fwnode associated with an IORT node
+ *
+ * @node: IORT table node associated with fwnode to delete
+ */
+static inline void iort_delete_fwnode(struct acpi_iort_node *node)
+{
+ struct iort_fwnode *curr, *tmp;
+
+ spin_lock(&iort_fwnode_lock);
+ list_for_each_entry_safe(curr, tmp, &iort_fwnode_list, list) {
+ if (curr->iort_node == node) {
+ list_del(&curr->list);
+ kfree(curr);
+ break;
+ }
+ }
+ spin_unlock(&iort_fwnode_lock);
+}
+
typedef acpi_status (*iort_find_node_callback)
(struct acpi_iort_node *node, void *context);
--
2.10.0
Current ARM SMMU driver rely on the struct device.of_node pointer for
device look-up and iommu_ops retrieval.
In preparation for ACPI probing enablement, convert the driver to use
the struct device.fwnode member for device and iommu_ops look-up so that
the driver infrastructure can be used also on systems that do not
associate an of_node pointer to a struct device (eg ACPI), making the
device look-up and iommu_ops retrieval firmware agnostic.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Will Deacon <[email protected]>
Reviewed-by: Robin Murphy <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Robin Murphy <[email protected]>
---
drivers/iommu/arm-smmu.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 8f72814..339a8d3 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -1379,13 +1379,14 @@ static bool arm_smmu_capable(enum iommu_cap cap)
static int arm_smmu_match_node(struct device *dev, void *data)
{
- return dev->of_node == data;
+ return dev->fwnode == data;
}
-static struct arm_smmu_device *arm_smmu_get_by_node(struct device_node *np)
+static
+struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
{
struct device *dev = driver_find_device(&arm_smmu_driver.driver, NULL,
- np, arm_smmu_match_node);
+ fwnode, arm_smmu_match_node);
put_device(dev);
return dev ? dev_get_drvdata(dev) : NULL;
}
@@ -1403,7 +1404,7 @@ static int arm_smmu_add_device(struct device *dev)
if (ret)
goto out_free;
} else if (fwspec && fwspec->ops == &arm_smmu_ops) {
- smmu = arm_smmu_get_by_node(to_of_node(fwspec->iommu_fwnode));
+ smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
} else {
return -ENODEV;
}
@@ -2007,7 +2008,7 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
}
}
- of_iommu_set_ops(dev->of_node, &arm_smmu_ops);
+ iommu_register_instance(dev->fwnode, &arm_smmu_ops);
platform_set_drvdata(pdev, smmu);
arm_smmu_device_reset(smmu);
--
2.10.0
Device drivers (eg ARM SMMU) need to know if a specific component
is part of the IORT table, so that kernel data structures are not
initialized at initcalls time if the respective component is not
part of the IORT table.
To this end, this patch adds a trivial function that allows detecting
if a given IORT node type is present or not in the ACPI table, providing
an ACPI IORT equivalent for of_find_matching_node().
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
drivers/acpi/arm64/iort.c | 15 +++++++++++++++
include/linux/acpi_iort.h | 2 ++
2 files changed, 17 insertions(+)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index 1ac2720..4bb6acb 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -227,6 +227,21 @@ static struct acpi_iort_node *iort_scan_node(enum acpi_iort_node_type type,
return NULL;
}
+static acpi_status
+iort_match_type_callback(struct acpi_iort_node *node, void *context)
+{
+ return AE_OK;
+}
+
+bool iort_node_match(u8 type)
+{
+ struct acpi_iort_node *node;
+
+ node = iort_scan_node(type, iort_match_type_callback, NULL);
+
+ return node != NULL;
+}
+
static acpi_status iort_match_node_callback(struct acpi_iort_node *node,
void *context)
{
diff --git a/include/linux/acpi_iort.h b/include/linux/acpi_iort.h
index d16fdda..17bb078 100644
--- a/include/linux/acpi_iort.h
+++ b/include/linux/acpi_iort.h
@@ -28,10 +28,12 @@ void iort_deregister_domain_token(int trans_id);
struct fwnode_handle *iort_find_domain_token(int trans_id);
#ifdef CONFIG_ACPI_IORT
void acpi_iort_init(void);
+bool iort_node_match(u8 type);
u32 iort_msi_map_rid(struct device *dev, u32 req_id);
struct irq_domain *iort_get_device_domain(struct device *dev, u32 req_id);
#else
static inline void acpi_iort_init(void) { }
+static inline bool iort_node_match(u8 type) { return false; }
static inline u32 iort_msi_map_rid(struct device *dev, u32 req_id)
{ return req_id; }
static inline struct irq_domain *iort_get_device_domain(struct device *dev,
--
2.10.0
Current ARM SMMU probe functions intermingle HW and DT probing
in the initialization functions to detect and programme the ARM SMMU
driver features. In order to allow probing the ARM SMMU with other
firmwares than DT, this patch splits the ARM SMMU init functions into
DT and HW specific portions so that other FW interfaces (ie ACPI) can
reuse the HW probing functions and skip the DT portion accordingly.
This patch implements no functional change, only code reshuffling.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Will Deacon <[email protected]>
Reviewed-by: Robin Murphy <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Robin Murphy <[email protected]>
---
drivers/iommu/arm-smmu.c | 62 +++++++++++++++++++++++++++++-------------------
1 file changed, 37 insertions(+), 25 deletions(-)
diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 339a8d3..573b2b6 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -1668,7 +1668,7 @@ static int arm_smmu_device_cfg_probe(struct arm_smmu_device *smmu)
unsigned long size;
void __iomem *gr0_base = ARM_SMMU_GR0(smmu);
u32 id;
- bool cttw_dt, cttw_reg;
+ bool cttw_reg, cttw_fw = smmu->features & ARM_SMMU_FEAT_COHERENT_WALK;
int i;
dev_notice(smmu->dev, "probing hardware configuration...\n");
@@ -1713,20 +1713,17 @@ static int arm_smmu_device_cfg_probe(struct arm_smmu_device *smmu)
/*
* In order for DMA API calls to work properly, we must defer to what
- * the DT says about coherency, regardless of what the hardware claims.
+ * the FW says about coherency, regardless of what the hardware claims.
* Fortunately, this also opens up a workaround for systems where the
* ID register value has ended up configured incorrectly.
*/
- cttw_dt = of_dma_is_coherent(smmu->dev->of_node);
cttw_reg = !!(id & ID0_CTTW);
- if (cttw_dt)
- smmu->features |= ARM_SMMU_FEAT_COHERENT_WALK;
- if (cttw_dt || cttw_reg)
+ if (cttw_fw || cttw_reg)
dev_notice(smmu->dev, "\t%scoherent table walk\n",
- cttw_dt ? "" : "non-");
- if (cttw_dt != cttw_reg)
+ cttw_fw ? "" : "non-");
+ if (cttw_fw != cttw_reg)
dev_notice(smmu->dev,
- "\t(IDR0.CTTW overridden by dma-coherent property)\n");
+ "\t(IDR0.CTTW overridden by FW configuration)\n");
/* Max. number of entries we have for stream matching/indexing */
size = 1 << ((id >> ID0_NUMSIDB_SHIFT) & ID0_NUMSIDB_MASK);
@@ -1907,15 +1904,25 @@ static const struct of_device_id arm_smmu_of_match[] = {
};
MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
-static int arm_smmu_device_dt_probe(struct platform_device *pdev)
+static int arm_smmu_device_dt_probe(struct platform_device *pdev,
+ struct arm_smmu_device *smmu)
{
const struct arm_smmu_match_data *data;
- struct resource *res;
- struct arm_smmu_device *smmu;
struct device *dev = &pdev->dev;
- int num_irqs, i, err;
bool legacy_binding;
+ if (of_property_read_u32(dev->of_node, "#global-interrupts",
+ &smmu->num_global_irqs)) {
+ dev_err(dev, "missing #global-interrupts property\n");
+ return -ENODEV;
+ }
+
+ data = of_device_get_match_data(dev);
+ smmu->version = data->version;
+ smmu->model = data->model;
+
+ parse_driver_options(smmu);
+
legacy_binding = of_find_property(dev->of_node, "mmu-masters", NULL);
if (legacy_binding && !using_generic_binding) {
if (!using_legacy_binding)
@@ -1928,6 +1935,19 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
return -ENODEV;
}
+ if (of_dma_is_coherent(dev->of_node))
+ smmu->features |= ARM_SMMU_FEAT_COHERENT_WALK;
+
+ return 0;
+}
+
+static int arm_smmu_device_probe(struct platform_device *pdev)
+{
+ struct resource *res;
+ struct arm_smmu_device *smmu;
+ struct device *dev = &pdev->dev;
+ int num_irqs, i, err;
+
smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
if (!smmu) {
dev_err(dev, "failed to allocate arm_smmu_device\n");
@@ -1935,9 +1955,9 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
}
smmu->dev = dev;
- data = of_device_get_match_data(dev);
- smmu->version = data->version;
- smmu->model = data->model;
+ err = arm_smmu_device_dt_probe(pdev, smmu);
+ if (err)
+ return err;
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
smmu->base = devm_ioremap_resource(dev, res);
@@ -1945,12 +1965,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
return PTR_ERR(smmu->base);
smmu->size = resource_size(res);
- if (of_property_read_u32(dev->of_node, "#global-interrupts",
- &smmu->num_global_irqs)) {
- dev_err(dev, "missing #global-interrupts property\n");
- return -ENODEV;
- }
-
num_irqs = 0;
while ((res = platform_get_resource(pdev, IORESOURCE_IRQ, num_irqs))) {
num_irqs++;
@@ -1985,8 +1999,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
if (err)
return err;
- parse_driver_options(smmu);
-
if (smmu->version == ARM_SMMU_V2 &&
smmu->num_context_banks != smmu->num_context_irqs) {
dev_err(dev,
@@ -2048,7 +2060,7 @@ static struct platform_driver arm_smmu_driver = {
.name = "arm-smmu",
.of_match_table = of_match_ptr(arm_smmu_of_match),
},
- .probe = arm_smmu_device_dt_probe,
+ .probe = arm_smmu_device_probe,
.remove = arm_smmu_device_remove,
};
--
2.10.0
Current ARM SMMU v3 driver rely on the struct device.of_node pointer for
device look-up and iommu_ops retrieval.
In preparation for ACPI probing enablement, convert the driver to use
the struct device.fwnode member for device and iommu_ops look-up so that
the driver infrastructure can be used also on systems that do not
associate an of_node pointer to a struct device (eg ACPI), making the
device look-up and iommu_ops retrieval firmware agnostic.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Will Deacon <[email protected]>
Reviewed-by: Robin Murphy <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Robin Murphy <[email protected]>
---
drivers/iommu/arm-smmu-v3.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c
index e6f9b2d..e6e1c87 100644
--- a/drivers/iommu/arm-smmu-v3.c
+++ b/drivers/iommu/arm-smmu-v3.c
@@ -1723,13 +1723,14 @@ static struct platform_driver arm_smmu_driver;
static int arm_smmu_match_node(struct device *dev, void *data)
{
- return dev->of_node == data;
+ return dev->fwnode == data;
}
-static struct arm_smmu_device *arm_smmu_get_by_node(struct device_node *np)
+static
+struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
{
struct device *dev = driver_find_device(&arm_smmu_driver.driver, NULL,
- np, arm_smmu_match_node);
+ fwnode, arm_smmu_match_node);
put_device(dev);
return dev ? dev_get_drvdata(dev) : NULL;
}
@@ -1765,7 +1766,7 @@ static int arm_smmu_add_device(struct device *dev)
master = fwspec->iommu_priv;
smmu = master->smmu;
} else {
- smmu = arm_smmu_get_by_node(to_of_node(fwspec->iommu_fwnode));
+ smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
if (!smmu)
return -ENODEV;
master = kzalloc(sizeof(*master), GFP_KERNEL);
@@ -2634,7 +2635,8 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
return ret;
/* And we're up. Go go go! */
- of_iommu_set_ops(dev->of_node, &arm_smmu_ops);
+ iommu_register_instance(dev->fwnode, &arm_smmu_ops);
+
#ifdef CONFIG_PCI
if (pci_bus_type.iommu_ops != &arm_smmu_ops) {
pci_request_acs();
--
2.10.0
In ACPI bases systems, in order to be able to create platform
devices and initialize them for ARM SMMU v3 components, the IORT
kernel implementation requires a set of static functions to be
used by the IORT kernel layer to configure platform devices for
ARM SMMU v3 components.
Add static configuration functions to the IORT kernel layer for
the ARM SMMU v3 components, so that the ARM SMMU v3 driver can
initialize its respective platform device by relying on the IORT
kernel infrastructure and by adding a corresponding ACPI device
early probe section entry.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Robin Murphy <[email protected]>
Cc: Joerg Roedel <[email protected]>
---
drivers/acpi/arm64/iort.c | 103 +++++++++++++++++++++++++++++++++++++++++++-
drivers/iommu/arm-smmu-v3.c | 49 ++++++++++++++++++++-
2 files changed, 150 insertions(+), 2 deletions(-)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index ddf83b5..fd52e4c 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -459,6 +459,95 @@ struct irq_domain *iort_get_device_domain(struct device *dev, u32 req_id)
return irq_find_matching_fwnode(handle, DOMAIN_BUS_PCI_MSI);
}
+static void __init acpi_iort_register_irq(int hwirq, const char *name,
+ int trigger,
+ struct resource *res)
+{
+ int irq = acpi_register_gsi(NULL, hwirq, trigger,
+ ACPI_ACTIVE_HIGH);
+
+ if (irq <= 0) {
+ pr_err("could not register gsi hwirq %d name [%s]\n", hwirq,
+ name);
+ return;
+ }
+
+ res->start = irq;
+ res->end = irq;
+ res->flags = IORESOURCE_IRQ;
+ res->name = name;
+}
+
+static int __init arm_smmu_v3_count_resources(struct acpi_iort_node *node)
+{
+ struct acpi_iort_smmu_v3 *smmu;
+ /* Always present mem resource */
+ int num_res = 1;
+
+ /* Retrieve SMMUv3 specific data */
+ smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
+
+ if (smmu->event_gsiv)
+ num_res++;
+
+ if (smmu->pri_gsiv)
+ num_res++;
+
+ if (smmu->gerr_gsiv)
+ num_res++;
+
+ if (smmu->sync_gsiv)
+ num_res++;
+
+ return num_res;
+}
+
+static void __init arm_smmu_v3_init_resources(struct resource *res,
+ struct acpi_iort_node *node)
+{
+ struct acpi_iort_smmu_v3 *smmu;
+ int num_res = 0;
+
+ /* Retrieve SMMUv3 specific data */
+ smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
+
+ res[num_res].start = smmu->base_address;
+ res[num_res].end = smmu->base_address + SZ_128K - 1;
+ res[num_res].flags = IORESOURCE_MEM;
+
+ num_res++;
+
+ if (smmu->event_gsiv)
+ acpi_iort_register_irq(smmu->event_gsiv, "eventq",
+ ACPI_EDGE_SENSITIVE,
+ &res[num_res++]);
+
+ if (smmu->pri_gsiv)
+ acpi_iort_register_irq(smmu->pri_gsiv, "priq",
+ ACPI_EDGE_SENSITIVE,
+ &res[num_res++]);
+
+ if (smmu->gerr_gsiv)
+ acpi_iort_register_irq(smmu->gerr_gsiv, "gerror",
+ ACPI_EDGE_SENSITIVE,
+ &res[num_res++]);
+
+ if (smmu->sync_gsiv)
+ acpi_iort_register_irq(smmu->sync_gsiv, "cmdq-sync",
+ ACPI_EDGE_SENSITIVE,
+ &res[num_res++]);
+}
+
+static bool __init arm_smmu_v3_is_coherent(struct acpi_iort_node *node)
+{
+ struct acpi_iort_smmu_v3 *smmu;
+
+ /* Retrieve SMMUv3 specific data */
+ smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
+
+ return smmu->flags & ACPI_IORT_SMMU_V3_COHACC_OVERRIDE;
+}
+
struct iort_iommu_config {
const char *name;
int (*iommu_init)(struct acpi_iort_node *node);
@@ -468,10 +557,22 @@ struct iort_iommu_config {
struct acpi_iort_node *node);
};
+static const struct iort_iommu_config iort_arm_smmu_v3_cfg __initconst = {
+ .name = "arm-smmu-v3",
+ .iommu_is_coherent = arm_smmu_v3_is_coherent,
+ .iommu_count_resources = arm_smmu_v3_count_resources,
+ .iommu_init_resources = arm_smmu_v3_init_resources
+};
+
static __init
const struct iort_iommu_config *iort_get_iommu_cfg(struct acpi_iort_node *node)
{
- return NULL;
+ switch (node->type) {
+ case ACPI_IORT_NODE_SMMU_V3:
+ return &iort_arm_smmu_v3_cfg;
+ default:
+ return NULL;
+ }
}
/**
diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c
index 2b3f9ac..d22c428 100644
--- a/drivers/iommu/arm-smmu-v3.c
+++ b/drivers/iommu/arm-smmu-v3.c
@@ -20,6 +20,8 @@
* This driver is powered by bad coffee and bombay mix.
*/
+#include <linux/acpi.h>
+#include <linux/acpi_iort.h>
#include <linux/delay.h>
#include <linux/dma-iommu.h>
#include <linux/err.h>
@@ -2559,6 +2561,32 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
return 0;
}
+#ifdef CONFIG_ACPI
+static int arm_smmu_device_acpi_probe(struct platform_device *pdev,
+ struct arm_smmu_device *smmu)
+{
+ struct acpi_iort_smmu_v3 *iort_smmu;
+ struct device *dev = smmu->dev;
+ struct acpi_iort_node *node;
+
+ node = *(struct acpi_iort_node **)dev_get_platdata(dev);
+
+ /* Retrieve SMMUv3 specific data */
+ iort_smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
+
+ if (iort_smmu->flags & ACPI_IORT_SMMU_V3_COHACC_OVERRIDE)
+ smmu->features |= ARM_SMMU_FEAT_COHERENCY;
+
+ return 0;
+}
+#else
+static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
+ struct arm_smmu_device *smmu)
+{
+ return -ENODEV;
+}
+#endif
+
static int arm_smmu_device_dt_probe(struct platform_device *pdev,
struct arm_smmu_device *smmu)
{
@@ -2624,8 +2652,16 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
if (irq > 0)
smmu->gerr_irq = irq;
+ if (dev->of_node) {
+ ret = arm_smmu_device_dt_probe(pdev, smmu);
+ } else {
+ ret = arm_smmu_device_acpi_probe(pdev, smmu);
+ if (ret == -ENODEV)
+ return ret;
+ }
+
/* Set bypass mode according to firmware probing result */
- bypass = !!arm_smmu_device_dt_probe(pdev, smmu);
+ bypass = !!ret;
/* Probe the h/w */
ret = arm_smmu_device_hw_probe(smmu);
@@ -2728,6 +2764,17 @@ static int __init arm_smmu_of_init(struct device_node *np)
}
IOMMU_OF_DECLARE(arm_smmuv3, "arm,smmu-v3", arm_smmu_of_init);
+#ifdef CONFIG_ACPI
+static int __init acpi_smmu_v3_init(struct acpi_table_header *table)
+{
+ if (iort_node_match(ACPI_IORT_NODE_SMMU_V3))
+ return arm_smmu_init();
+
+ return 0;
+}
+IORT_ACPI_DECLARE(arm_smmu_v3, ACPI_SIG_IORT, acpi_smmu_v3_init);
+#endif
+
MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3 implementations");
MODULE_AUTHOR("Will Deacon <[email protected]>");
MODULE_LICENSE("GPL v2");
--
2.10.0
In ACPI based systems, in order to be able to create platform
devices and initialize them for ARM SMMU components, the IORT
kernel implementation requires a set of static functions to be
used by the IORT kernel layer to configure platform devices for
ARM SMMU components.
Add static configuration functions to the IORT kernel layer for
the ARM SMMU components, so that the ARM SMMU driver can
initialize its respective platform device by relying on the IORT
kernel infrastructure and by adding a corresponding ACPI device
early probe section entry.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Robin Murphy <[email protected]>
Cc: Joerg Roedel <[email protected]>
---
drivers/acpi/arm64/iort.c | 71 +++++++++++++++++++++++++++++++++++++++++++
drivers/iommu/arm-smmu.c | 77 ++++++++++++++++++++++++++++++++++++++++++++++-
include/linux/acpi_iort.h | 3 ++
3 files changed, 150 insertions(+), 1 deletion(-)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index fd52e4c..8a8ae5e 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -548,6 +548,68 @@ static bool __init arm_smmu_v3_is_coherent(struct acpi_iort_node *node)
return smmu->flags & ACPI_IORT_SMMU_V3_COHACC_OVERRIDE;
}
+static int __init arm_smmu_count_resources(struct acpi_iort_node *node)
+{
+ struct acpi_iort_smmu *smmu;
+
+ /* Retrieve SMMU specific data */
+ smmu = (struct acpi_iort_smmu *)node->node_data;
+
+ /*
+ * Only consider the global fault interrupt and ignore the
+ * configuration access interrupt.
+ *
+ * MMIO address and global fault interrupt resources are always
+ * present so add them to the context interrupt count as a static
+ * value.
+ */
+ return smmu->context_interrupt_count + 2;
+}
+
+static void __init arm_smmu_init_resources(struct resource *res,
+ struct acpi_iort_node *node)
+{
+ struct acpi_iort_smmu *smmu;
+ int i, hw_irq, trigger, num_res = 0;
+ u64 *ctx_irq, *glb_irq;
+
+ /* Retrieve SMMU specific data */
+ smmu = (struct acpi_iort_smmu *)node->node_data;
+
+ res[num_res].start = smmu->base_address;
+ res[num_res].end = smmu->base_address + smmu->span - 1;
+ res[num_res].flags = IORESOURCE_MEM;
+ num_res++;
+
+ glb_irq = ACPI_ADD_PTR(u64, node, smmu->global_interrupt_offset);
+ /* Global IRQs */
+ hw_irq = IORT_IRQ_MASK(glb_irq[0]);
+ trigger = IORT_IRQ_TRIGGER_MASK(glb_irq[0]);
+
+ acpi_iort_register_irq(hw_irq, "arm-smmu-global", trigger,
+ &res[num_res++]);
+
+ /* Context IRQs */
+ ctx_irq = ACPI_ADD_PTR(u64, node, smmu->context_interrupt_offset);
+ for (i = 0; i < smmu->context_interrupt_count; i++) {
+ hw_irq = IORT_IRQ_MASK(ctx_irq[i]);
+ trigger = IORT_IRQ_TRIGGER_MASK(ctx_irq[i]);
+
+ acpi_iort_register_irq(hw_irq, "arm-smmu-context", trigger,
+ &res[num_res++]);
+ }
+}
+
+static bool __init arm_smmu_is_coherent(struct acpi_iort_node *node)
+{
+ struct acpi_iort_smmu *smmu;
+
+ /* Retrieve SMMU specific data */
+ smmu = (struct acpi_iort_smmu *)node->node_data;
+
+ return smmu->flags & ACPI_IORT_SMMU_COHERENT_WALK;
+}
+
struct iort_iommu_config {
const char *name;
int (*iommu_init)(struct acpi_iort_node *node);
@@ -564,12 +626,21 @@ static const struct iort_iommu_config iort_arm_smmu_v3_cfg __initconst = {
.iommu_init_resources = arm_smmu_v3_init_resources
};
+static const struct iort_iommu_config iort_arm_smmu_cfg __initconst = {
+ .name = "arm-smmu",
+ .iommu_is_coherent = arm_smmu_is_coherent,
+ .iommu_count_resources = arm_smmu_count_resources,
+ .iommu_init_resources = arm_smmu_init_resources
+};
+
static __init
const struct iort_iommu_config *iort_get_iommu_cfg(struct acpi_iort_node *node)
{
switch (node->type) {
case ACPI_IORT_NODE_SMMU_V3:
return &iort_arm_smmu_v3_cfg;
+ case ACPI_IORT_NODE_SMMU:
+ return &iort_arm_smmu_cfg;
default:
return NULL;
}
diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 573b2b6..8e66b16 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -28,6 +28,8 @@
#define pr_fmt(fmt) "arm-smmu: " fmt
+#include <linux/acpi.h>
+#include <linux/acpi_iort.h>
#include <linux/atomic.h>
#include <linux/delay.h>
#include <linux/dma-iommu.h>
@@ -1904,6 +1906,64 @@ static const struct of_device_id arm_smmu_of_match[] = {
};
MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
+#ifdef CONFIG_ACPI
+static int acpi_smmu_get_data(u32 model, struct arm_smmu_device *smmu)
+{
+ int ret = 0;
+
+ switch (model) {
+ case ACPI_IORT_SMMU_V1:
+ case ACPI_IORT_SMMU_CORELINK_MMU400:
+ smmu->version = ARM_SMMU_V1;
+ smmu->model = GENERIC_SMMU;
+ break;
+ case ACPI_IORT_SMMU_V2:
+ smmu->version = ARM_SMMU_V2;
+ smmu->model = GENERIC_SMMU;
+ break;
+ case ACPI_IORT_SMMU_CORELINK_MMU500:
+ smmu->version = ARM_SMMU_V2;
+ smmu->model = ARM_MMU500;
+ break;
+ default:
+ ret = -ENODEV;
+ }
+
+ return ret;
+}
+
+static int arm_smmu_device_acpi_probe(struct platform_device *pdev,
+ struct arm_smmu_device *smmu)
+{
+ struct device *dev = smmu->dev;
+ struct acpi_iort_node *node =
+ *(struct acpi_iort_node **)dev_get_platdata(dev);
+ struct acpi_iort_smmu *iort_smmu;
+ int ret;
+
+ /* Retrieve SMMU1/2 specific data */
+ iort_smmu = (struct acpi_iort_smmu *)node->node_data;
+
+ ret = acpi_smmu_get_data(iort_smmu->model, smmu);
+ if (ret < 0)
+ return ret;
+
+ /* Ignore the configuration access interrupt */
+ smmu->num_global_irqs = 1;
+
+ if (iort_smmu->flags & ACPI_IORT_SMMU_COHERENT_WALK)
+ smmu->features |= ARM_SMMU_FEAT_COHERENT_WALK;
+
+ return 0;
+}
+#else
+static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
+ struct arm_smmu_device *smmu)
+{
+ return -ENODEV;
+}
+#endif
+
static int arm_smmu_device_dt_probe(struct platform_device *pdev,
struct arm_smmu_device *smmu)
{
@@ -1955,7 +2015,11 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
}
smmu->dev = dev;
- err = arm_smmu_device_dt_probe(pdev, smmu);
+ if (dev->of_node)
+ err = arm_smmu_device_dt_probe(pdev, smmu);
+ else
+ err = arm_smmu_device_acpi_probe(pdev, smmu);
+
if (err)
return err;
@@ -2103,6 +2167,17 @@ IOMMU_OF_DECLARE(arm_mmu401, "arm,mmu-401", arm_smmu_of_init);
IOMMU_OF_DECLARE(arm_mmu500, "arm,mmu-500", arm_smmu_of_init);
IOMMU_OF_DECLARE(cavium_smmuv2, "cavium,smmu-v2", arm_smmu_of_init);
+#ifdef CONFIG_ACPI
+static int __init arm_smmu_acpi_init(struct acpi_table_header *table)
+{
+ if (iort_node_match(ACPI_IORT_NODE_SMMU))
+ return arm_smmu_init();
+
+ return 0;
+}
+IORT_ACPI_DECLARE(arm_smmu, ACPI_SIG_IORT, arm_smmu_acpi_init);
+#endif
+
MODULE_DESCRIPTION("IOMMU API for ARM architected SMMU implementations");
MODULE_AUTHOR("Will Deacon <[email protected]>");
MODULE_LICENSE("GPL v2");
diff --git a/include/linux/acpi_iort.h b/include/linux/acpi_iort.h
index 17bb078..79ba1bb 100644
--- a/include/linux/acpi_iort.h
+++ b/include/linux/acpi_iort.h
@@ -23,6 +23,9 @@
#include <linux/fwnode.h>
#include <linux/irqdomain.h>
+#define IORT_IRQ_MASK(irq) (irq & 0xffffffffULL)
+#define IORT_IRQ_TRIGGER_MASK(irq) ((irq >> 32) & 0xffffffffULL)
+
int iort_register_domain_token(int trans_id, struct fwnode_handle *fw_node);
void iort_deregister_domain_token(int trans_id);
struct fwnode_handle *iort_find_domain_token(int trans_id);
--
2.10.0
The current IORT id mapping API requires components to provide
an input requester ID (a Bus-Device-Function (BDF) identifier for
PCI devices) to translate an input identifier to an output
identifier through an IORT range mapping.
Named components do not have an identifiable source ID therefore
their respective input/output mapping can only be defined in
IORT tables through single mappings, that provide a translation
that does not require any input identifier.
Current IORT interface for requester id mappings (iort_node_map_rid())
is not suitable for components that do not provide a requester id,
so it cannot be used for IORT named components.
Add an interface to the IORT API to enable retrieval of id
by allowing an indexed walk of the single mappings array for
a given component, therefore completing the IORT mapping API.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
drivers/acpi/arm64/iort.c | 39 +++++++++++++++++++++++++++++++++++++++
1 file changed, 39 insertions(+)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index f3bbef8..6aae49c 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -318,6 +318,45 @@ static int iort_id_map(struct acpi_iort_id_mapping *map, u8 type, u32 rid_in,
return 0;
}
+static
+struct acpi_iort_node *iort_node_get_id(struct acpi_iort_node *node,
+ u32 *id_out, u8 type_mask,
+ int index)
+{
+ struct acpi_iort_node *parent;
+ struct acpi_iort_id_mapping *map;
+
+ if (!node->mapping_offset || !node->mapping_count ||
+ index >= node->mapping_count)
+ return NULL;
+
+ map = ACPI_ADD_PTR(struct acpi_iort_id_mapping, node,
+ node->mapping_offset);
+
+ /* Firmware bug! */
+ if (!map->output_reference) {
+ pr_err(FW_BUG "[node %p type %d] ID map has NULL parent reference\n",
+ node, node->type);
+ return NULL;
+ }
+
+ parent = ACPI_ADD_PTR(struct acpi_iort_node, iort_table,
+ map->output_reference);
+
+ if (!(IORT_TYPE_MASK(parent->type) & type_mask))
+ return NULL;
+
+ if (map[index].flags & ACPI_IORT_ID_SINGLE_MAPPING) {
+ if (node->type == ACPI_IORT_NODE_NAMED_COMPONENT ||
+ node->type == ACPI_IORT_NODE_PCI_ROOT_COMPLEX) {
+ *id_out = map[index].output_base;
+ return parent;
+ }
+ }
+
+ return NULL;
+}
+
static struct acpi_iort_node *iort_node_map_rid(struct acpi_iort_node *node,
u32 rid_in, u32 *rid_out,
u8 type_mask)
--
2.10.0
IORT tables provide data that allow the kernel to carry out
device ID mappings between endpoints and system components
(eg interrupt controllers, IOMMUs). When the mapping for a
given device ID is carried out, the translation mechanism
is done on a per-subsystem basis rather than a component
subtype (ie the IOMMU kernel layer will look for mappings
from a device to all IORT node types corresponding to IOMMU
components), therefore the corresponding mapping API should
work on a range (ie mask) of IORT node types corresponding
to a common set of components (eg IOMMUs) rather than a
specific node type.
Upgrade the IORT iort_node_map_rid() API to work with a
type mask instead of a single node type so that it can
be used for mappings that span multiple components types
(ie IOMMUs).
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
drivers/acpi/arm64/iort.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index 8a8ae5e..f3bbef8 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -26,6 +26,9 @@
#include <linux/platform_device.h>
#include <linux/slab.h>
+#define IORT_TYPE_MASK(type) (1 << (type))
+#define IORT_MSI_TYPE (1 << ACPI_IORT_NODE_ITS_GROUP)
+
struct iort_its_msi_chip {
struct list_head list;
struct fwnode_handle *fw_node;
@@ -317,7 +320,7 @@ static int iort_id_map(struct acpi_iort_id_mapping *map, u8 type, u32 rid_in,
static struct acpi_iort_node *iort_node_map_rid(struct acpi_iort_node *node,
u32 rid_in, u32 *rid_out,
- u8 type)
+ u8 type_mask)
{
u32 rid = rid_in;
@@ -326,7 +329,7 @@ static struct acpi_iort_node *iort_node_map_rid(struct acpi_iort_node *node,
struct acpi_iort_id_mapping *map;
int i;
- if (node->type == type) {
+ if (IORT_TYPE_MASK(node->type) & type_mask) {
if (rid_out)
*rid_out = rid;
return node;
@@ -399,7 +402,7 @@ u32 iort_msi_map_rid(struct device *dev, u32 req_id)
if (!node)
return req_id;
- iort_node_map_rid(node, req_id, &dev_id, ACPI_IORT_NODE_ITS_GROUP);
+ iort_node_map_rid(node, req_id, &dev_id, IORT_MSI_TYPE);
return dev_id;
}
@@ -421,7 +424,7 @@ static int iort_dev_find_its_id(struct device *dev, u32 req_id,
if (!node)
return -ENXIO;
- node = iort_node_map_rid(node, req_id, NULL, ACPI_IORT_NODE_ITS_GROUP);
+ node = iort_node_map_rid(node, req_id, NULL, IORT_MSI_TYPE);
if (!node)
return -ENXIO;
--
2.10.0
DT based systems have a generic kernel API to configure IOMMUs
for devices (ie of_iommu_configure()).
On ARM based ACPI systems, the of_iommu_configure() equivalent can
be implemented atop ACPI IORT kernel API, with the corresponding
functions to map device identifiers to IOMMUs and retrieve the
corresponding IOMMU operations necessary for DMA operations set-up.
By relying on the iommu_fwspec generic kernel infrastructure,
implement the IORT based IOMMU configuration for ARM ACPI systems
and hook it up in the ACPI kernel layer that implements DMA
configuration for a device.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Rafael J. Wysocki <[email protected]> [ACPI core]
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
drivers/acpi/arm64/iort.c | 98 +++++++++++++++++++++++++++++++++++++++++++++++
drivers/acpi/scan.c | 7 +++-
include/linux/acpi_iort.h | 6 +++
3 files changed, 110 insertions(+), 1 deletion(-)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index 6aae49c..47bace8 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -28,6 +28,8 @@
#define IORT_TYPE_MASK(type) (1 << (type))
#define IORT_MSI_TYPE (1 << ACPI_IORT_NODE_ITS_GROUP)
+#define IORT_IOMMU_TYPE ((1 << ACPI_IORT_NODE_SMMU) | \
+ (1 << ACPI_IORT_NODE_SMMU_V3))
struct iort_its_msi_chip {
struct list_head list;
@@ -501,6 +503,102 @@ struct irq_domain *iort_get_device_domain(struct device *dev, u32 req_id)
return irq_find_matching_fwnode(handle, DOMAIN_BUS_PCI_MSI);
}
+static int __get_pci_rid(struct pci_dev *pdev, u16 alias, void *data)
+{
+ u32 *rid = data;
+
+ *rid = alias;
+ return 0;
+}
+
+static int arm_smmu_iort_xlate(struct device *dev, u32 streamid,
+ struct fwnode_handle *fwnode,
+ const struct iommu_ops *ops)
+{
+ int ret = iommu_fwspec_init(dev, fwnode, ops);
+
+ if (!ret)
+ ret = iommu_fwspec_add_ids(dev, &streamid, 1);
+
+ return ret;
+}
+
+static const struct iommu_ops *iort_iommu_xlate(struct device *dev,
+ struct acpi_iort_node *node,
+ u32 streamid)
+{
+ const struct iommu_ops *ops = NULL;
+ int ret = -ENODEV;
+ struct fwnode_handle *iort_fwnode;
+
+ if (node) {
+ iort_fwnode = iort_get_fwnode(node);
+ if (!iort_fwnode)
+ return NULL;
+
+ ops = iommu_get_instance(iort_fwnode);
+ if (!ops)
+ return NULL;
+
+ ret = arm_smmu_iort_xlate(dev, streamid, iort_fwnode, ops);
+ }
+
+ return ret ? NULL : ops;
+}
+
+/**
+ * iort_iommu_configure - Set-up IOMMU configuration for a device.
+ *
+ * @dev: device to configure
+ *
+ * Returns: iommu_ops pointer on configuration success
+ * NULL on configuration failure
+ */
+const struct iommu_ops *iort_iommu_configure(struct device *dev)
+{
+ struct acpi_iort_node *node, *parent;
+ const struct iommu_ops *ops = NULL;
+ u32 streamid = 0;
+
+ if (dev_is_pci(dev)) {
+ struct pci_bus *bus = to_pci_dev(dev)->bus;
+ u32 rid;
+
+ pci_for_each_dma_alias(to_pci_dev(dev), __get_pci_rid,
+ &rid);
+
+ node = iort_scan_node(ACPI_IORT_NODE_PCI_ROOT_COMPLEX,
+ iort_match_node_callback, &bus->dev);
+ if (!node)
+ return NULL;
+
+ parent = iort_node_map_rid(node, rid, &streamid,
+ IORT_IOMMU_TYPE);
+
+ ops = iort_iommu_xlate(dev, parent, streamid);
+
+ } else {
+ int i = 0;
+
+ node = iort_scan_node(ACPI_IORT_NODE_NAMED_COMPONENT,
+ iort_match_node_callback, dev);
+ if (!node)
+ return NULL;
+
+ parent = iort_node_get_id(node, &streamid,
+ IORT_IOMMU_TYPE, i++);
+
+ while (parent) {
+ ops = iort_iommu_xlate(dev, parent, streamid);
+
+ parent = iort_node_get_id(node, &streamid,
+ IORT_IOMMU_TYPE, i++);
+ }
+ }
+
+ return ops;
+}
+
static void __init acpi_iort_register_irq(int hwirq, const char *name,
int trigger,
struct resource *res)
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 45b5710..80698d3 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -7,6 +7,7 @@
#include <linux/slab.h>
#include <linux/kernel.h>
#include <linux/acpi.h>
+#include <linux/acpi_iort.h>
#include <linux/signal.h>
#include <linux/kthread.h>
#include <linux/dmi.h>
@@ -1377,6 +1378,8 @@ enum dev_dma_attr acpi_get_dma_attr(struct acpi_device *adev)
*/
void acpi_dma_configure(struct device *dev, enum dev_dma_attr attr)
{
+ const struct iommu_ops *iommu;
+
/*
* Set default coherent_dma_mask to 32 bit. Drivers are expected to
* setup the correct supported mask.
@@ -1391,11 +1394,13 @@ void acpi_dma_configure(struct device *dev, enum dev_dma_attr attr)
if (!dev->dma_mask)
dev->dma_mask = &dev->coherent_dma_mask;
+ iommu = iort_iommu_configure(dev);
+
/*
* Assume dma valid range starts at 0 and covers the whole
* coherent_dma_mask.
*/
- arch_setup_dma_ops(dev, 0, dev->coherent_dma_mask + 1, NULL,
+ arch_setup_dma_ops(dev, 0, dev->coherent_dma_mask + 1, iommu,
attr == DEV_DMA_COHERENT);
}
EXPORT_SYMBOL_GPL(acpi_dma_configure);
diff --git a/include/linux/acpi_iort.h b/include/linux/acpi_iort.h
index 79ba1bb..dcb2b60 100644
--- a/include/linux/acpi_iort.h
+++ b/include/linux/acpi_iort.h
@@ -34,6 +34,8 @@ void acpi_iort_init(void);
bool iort_node_match(u8 type);
u32 iort_msi_map_rid(struct device *dev, u32 req_id);
struct irq_domain *iort_get_device_domain(struct device *dev, u32 req_id);
+/* IOMMU interface */
+const struct iommu_ops *iort_iommu_configure(struct device *dev);
#else
static inline void acpi_iort_init(void) { }
static inline bool iort_node_match(u8 type) { return false; }
@@ -42,6 +44,10 @@ static inline u32 iort_msi_map_rid(struct device *dev, u32 req_id)
static inline struct irq_domain *iort_get_device_domain(struct device *dev,
u32 req_id)
{ return NULL; }
+/* IOMMU interface */
+static inline
+const struct iommu_ops *iort_iommu_configure(struct device *dev)
+{ return NULL; }
#endif
#define IORT_ACPI_DECLARE(name, table_id, fn) \
--
2.10.0
In ARM ACPI systems, IOMMU components are specified through static
IORT table entries. In order to create platform devices for the
corresponding ARM SMMU components, IORT kernel code should be made
able to parse IORT table entries and create platform devices
dynamically.
This patch adds the generic IORT infrastructure required to create
platform devices for ARM SMMUs.
ARM SMMU versions have different resources requirement therefore this
patch also introduces an IORT specific structure (ie iort_iommu_config)
that contains hooks (to be defined when the corresponding ARM SMMU
driver support is added to the kernel) to be used to define the
platform devices names, init the IOMMUs, count their resources and
finally initialize them.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
drivers/acpi/arm64/iort.c | 151 ++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 151 insertions(+)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index 4bb6acb..ddf83b5 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -19,9 +19,11 @@
#define pr_fmt(fmt) "ACPI: IORT: " fmt
#include <linux/acpi_iort.h>
+#include <linux/iommu.h>
#include <linux/kernel.h>
#include <linux/list.h>
#include <linux/pci.h>
+#include <linux/platform_device.h>
#include <linux/slab.h>
struct iort_its_msi_chip {
@@ -457,6 +459,153 @@ struct irq_domain *iort_get_device_domain(struct device *dev, u32 req_id)
return irq_find_matching_fwnode(handle, DOMAIN_BUS_PCI_MSI);
}
+struct iort_iommu_config {
+ const char *name;
+ int (*iommu_init)(struct acpi_iort_node *node);
+ bool (*iommu_is_coherent)(struct acpi_iort_node *node);
+ int (*iommu_count_resources)(struct acpi_iort_node *node);
+ void (*iommu_init_resources)(struct resource *res,
+ struct acpi_iort_node *node);
+};
+
+static __init
+const struct iort_iommu_config *iort_get_iommu_cfg(struct acpi_iort_node *node)
+{
+ return NULL;
+}
+
+/**
+ * iort_add_smmu_platform_device() - Allocate a platform device for SMMU
+ * @node: Pointer to SMMU ACPI IORT node
+ *
+ * Returns: 0 on success, <0 failure
+ */
+static int __init iort_add_smmu_platform_device(struct acpi_iort_node *node)
+{
+ struct fwnode_handle *fwnode;
+ struct platform_device *pdev;
+ struct resource *r;
+ enum dev_dma_attr attr;
+ int ret, count;
+ const struct iort_iommu_config *ops = iort_get_iommu_cfg(node);
+
+ if (!ops)
+ return -ENODEV;
+
+ pdev = platform_device_alloc(ops->name, PLATFORM_DEVID_AUTO);
+ if (!pdev)
+ return PTR_ERR(pdev);
+
+ count = ops->iommu_count_resources(node);
+
+ r = kcalloc(count, sizeof(*r), GFP_KERNEL);
+ if (!r) {
+ ret = -ENOMEM;
+ goto dev_put;
+ }
+
+ ops->iommu_init_resources(r, node);
+
+ ret = platform_device_add_resources(pdev, r, count);
+ /*
+ * Resources are duplicated in platform_device_add_resources,
+ * free their allocated memory
+ */
+ kfree(r);
+
+ if (ret)
+ goto dev_put;
+
+ /*
+ * Add a copy of IORT node pointer to platform_data to
+ * be used to retrieve IORT data information.
+ */
+ ret = platform_device_add_data(pdev, &node, sizeof(node));
+ if (ret)
+ goto dev_put;
+
+ /*
+ * We expect the dma masks to be equivalent for
+ * all SMMUs set-ups
+ */
+ pdev->dev.dma_mask = &pdev->dev.coherent_dma_mask;
+
+ fwnode = iort_get_fwnode(node);
+
+ if (!fwnode) {
+ ret = -ENODEV;
+ goto dev_put;
+ }
+
+ pdev->dev.fwnode = fwnode;
+
+ attr = ops->iommu_is_coherent(node) ?
+ DEV_DMA_COHERENT : DEV_DMA_NON_COHERENT;
+
+ /* Configure DMA for the page table walker */
+ acpi_dma_configure(&pdev->dev, attr);
+
+ ret = platform_device_add(pdev);
+ if (ret)
+ goto dma_deconfigure;
+
+ return 0;
+
+dma_deconfigure:
+ acpi_dma_deconfigure(&pdev->dev);
+dev_put:
+ platform_device_put(pdev);
+
+ return ret;
+}
+
+static void __init iort_init_platform_devices(void)
+{
+ struct acpi_iort_node *iort_node, *iort_end;
+ struct acpi_table_iort *iort;
+ struct fwnode_handle *fwnode;
+ int i, ret;
+
+ /*
+ * iort_table and iort both point to the start of IORT table, but
+ * have different struct types
+ */
+ iort = (struct acpi_table_iort *)iort_table;
+
+ /* Get the first IORT node */
+ iort_node = ACPI_ADD_PTR(struct acpi_iort_node, iort,
+ iort->node_offset);
+ iort_end = ACPI_ADD_PTR(struct acpi_iort_node, iort,
+ iort_table->length);
+
+ for (i = 0; i < iort->node_count; i++) {
+ if (iort_node >= iort_end) {
+ pr_err("iort node pointer overflows, bad table\n");
+ return;
+ }
+
+ if ((iort_node->type == ACPI_IORT_NODE_SMMU) ||
+ (iort_node->type == ACPI_IORT_NODE_SMMU_V3)) {
+
+ fwnode = acpi_alloc_fwnode_static();
+ if (!fwnode)
+ return;
+
+ iort_set_fwnode(iort_node, fwnode);
+
+ ret = iort_add_smmu_platform_device(iort_node);
+ if (ret) {
+ iort_delete_fwnode(iort_node);
+ acpi_free_fwnode_static(fwnode);
+ return;
+ }
+ }
+
+ iort_node = ACPI_ADD_PTR(struct acpi_iort_node, iort_node,
+ iort_node->length);
+ }
+}
+
void __init acpi_iort_init(void)
{
acpi_status status;
@@ -472,5 +621,7 @@ void __init acpi_iort_init(void)
return;
}
+ iort_init_platform_devices();
+
acpi_probe_device_table(iort);
}
--
2.10.0
Current ARM SMMUv3 probe functions intermingle HW and DT probing in the
initialization functions to detect and programme the ARM SMMU v3 driver
features. In order to allow probing the ARM SMMUv3 with other firmwares
than DT, this patch splits the ARM SMMUv3 init functions into DT and HW
specific portions so that other FW interfaces (ie ACPI) can reuse the HW
probing functions and skip the DT portion accordingly.
This patch implements no functional change, only code reshuffling.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Robin Murphy <[email protected]>
Cc: Joerg Roedel <[email protected]>
---
drivers/iommu/arm-smmu-v3.c | 43 +++++++++++++++++++++++++++----------------
1 file changed, 27 insertions(+), 16 deletions(-)
diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c
index e6e1c87..2b3f9ac 100644
--- a/drivers/iommu/arm-smmu-v3.c
+++ b/drivers/iommu/arm-smmu-v3.c
@@ -2381,10 +2381,10 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
return 0;
}
-static int arm_smmu_device_probe(struct arm_smmu_device *smmu)
+static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
{
u32 reg;
- bool coherent;
+ bool coherent = smmu->features & ARM_SMMU_FEAT_COHERENCY;
/* IDR0 */
reg = readl_relaxed(smmu->base + ARM_SMMU_IDR0);
@@ -2436,13 +2436,9 @@ static int arm_smmu_device_probe(struct arm_smmu_device *smmu)
smmu->features |= ARM_SMMU_FEAT_HYP;
/*
- * The dma-coherent property is used in preference to the ID
+ * The coherency feature as set by FW is used in preference to the ID
* register, but warn on mismatch.
*/
- coherent = of_dma_is_coherent(smmu->dev->of_node);
- if (coherent)
- smmu->features |= ARM_SMMU_FEAT_COHERENCY;
-
if (!!(reg & IDR0_COHACC) != coherent)
dev_warn(smmu->dev, "IDR0.COHACC overridden by dma-coherent property (%s)\n",
coherent ? "true" : "false");
@@ -2563,21 +2559,35 @@ static int arm_smmu_device_probe(struct arm_smmu_device *smmu)
return 0;
}
-static int arm_smmu_device_dt_probe(struct platform_device *pdev)
+static int arm_smmu_device_dt_probe(struct platform_device *pdev,
+ struct arm_smmu_device *smmu)
{
- int irq, ret;
- struct resource *res;
- struct arm_smmu_device *smmu;
struct device *dev = &pdev->dev;
- bool bypass = true;
u32 cells;
+ int ret = -EINVAL;
if (of_property_read_u32(dev->of_node, "#iommu-cells", &cells))
dev_err(dev, "missing #iommu-cells property\n");
else if (cells != 1)
dev_err(dev, "invalid #iommu-cells value (%d)\n", cells);
else
- bypass = false;
+ ret = 0;
+
+ parse_driver_options(smmu);
+
+ if (of_dma_is_coherent(dev->of_node))
+ smmu->features |= ARM_SMMU_FEAT_COHERENCY;
+
+ return ret;
+}
+
+static int arm_smmu_device_probe(struct platform_device *pdev)
+{
+ int irq, ret;
+ struct resource *res;
+ struct arm_smmu_device *smmu;
+ struct device *dev = &pdev->dev;
+ bool bypass;
smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
if (!smmu) {
@@ -2614,10 +2624,11 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
if (irq > 0)
smmu->gerr_irq = irq;
- parse_driver_options(smmu);
+ /* Set bypass mode according to firmware probing result */
+ bypass = !!arm_smmu_device_dt_probe(pdev, smmu);
/* Probe the h/w */
- ret = arm_smmu_device_probe(smmu);
+ ret = arm_smmu_device_hw_probe(smmu);
if (ret)
return ret;
@@ -2679,7 +2690,7 @@ static struct platform_driver arm_smmu_driver = {
.name = "arm-smmu-v3",
.of_match_table = of_match_ptr(arm_smmu_of_match),
},
- .probe = arm_smmu_device_dt_probe,
+ .probe = arm_smmu_device_probe,
.remove = arm_smmu_device_remove,
};
--
2.10.0
On DT based systems, the of_dma_configure() API implements DMA
configuration for a given device. On ACPI systems an API equivalent to
of_dma_configure() is missing which implies that it is currently not
possible to set-up DMA operations for devices through the ACPI generic
kernel layer.
This patch fills the gap by introducing acpi_dma_configure/deconfigure()
calls that for now are just wrappers around arch_setup_dma_ops() and
arch_teardown_dma_ops() and also updates ACPI and PCI core code to use
the newly introduced acpi_dma_configure/acpi_dma_deconfigure functions.
Since acpi_dma_configure() is used to configure DMA operations, the
function initializes the dma/coherent_dma masks to sane default values
if the current masks are uninitialized (also to keep the default values
consistent with DT systems) to make sure the device has a complete
default DMA set-up.
The DMA range size passed to arch_setup_dma_ops() is sized according
to the device coherent_dma_mask (starting at address 0x0), mirroring the
DT probing path behaviour when a dma-ranges property is not provided
for the device being probed; this changes the current arch_setup_dma_ops()
call parameters in the ACPI probing case, but since arch_setup_dma_ops()
is a NOP on all architectures but ARM/ARM64 this patch does not change
the current kernel behaviour on them.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Bjorn Helgaas <[email protected]> [pci]
Acked-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Robin Murphy <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: Joerg Roedel <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
---
drivers/acpi/glue.c | 4 ++--
drivers/acpi/scan.c | 40 ++++++++++++++++++++++++++++++++++++++++
drivers/pci/probe.c | 3 +--
include/acpi/acpi_bus.h | 2 ++
include/linux/acpi.h | 5 +++++
5 files changed, 50 insertions(+), 4 deletions(-)
diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 5ea5dc2..f8d6564 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -227,8 +227,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev)
attr = acpi_get_dma_attr(acpi_dev);
if (attr != DEV_DMA_NOT_SUPPORTED)
- arch_setup_dma_ops(dev, 0, 0, NULL,
- attr == DEV_DMA_COHERENT);
+ acpi_dma_configure(dev, attr);
acpi_physnode_link_name(physical_node_name, node_id);
retval = sysfs_create_link(&acpi_dev->dev.kobj, &dev->kobj,
@@ -251,6 +250,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev)
return 0;
err:
+ acpi_dma_deconfigure(dev);
ACPI_COMPANION_SET(dev, NULL);
put_device(dev);
put_device(&acpi_dev->dev);
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 3d1856f..45b5710 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -1370,6 +1370,46 @@ enum dev_dma_attr acpi_get_dma_attr(struct acpi_device *adev)
return DEV_DMA_NON_COHERENT;
}
+/**
+ * acpi_dma_configure - Set-up DMA configuration for the device.
+ * @dev: The pointer to the device
+ * @attr: device dma attributes
+ */
+void acpi_dma_configure(struct device *dev, enum dev_dma_attr attr)
+{
+ /*
+ * Set default coherent_dma_mask to 32 bit. Drivers are expected to
+ * setup the correct supported mask.
+ */
+ if (!dev->coherent_dma_mask)
+ dev->coherent_dma_mask = DMA_BIT_MASK(32);
+
+ /*
+ * Set it to coherent_dma_mask by default if the architecture
+ * code has not set it.
+ */
+ if (!dev->dma_mask)
+ dev->dma_mask = &dev->coherent_dma_mask;
+
+ /*
+ * Assume dma valid range starts at 0 and covers the whole
+ * coherent_dma_mask.
+ */
+ arch_setup_dma_ops(dev, 0, dev->coherent_dma_mask + 1, NULL,
+ attr == DEV_DMA_COHERENT);
+}
+EXPORT_SYMBOL_GPL(acpi_dma_configure);
+
+/**
+ * acpi_dma_deconfigure - Tear-down DMA configuration for the device.
+ * @dev: The pointer to the device
+ */
+void acpi_dma_deconfigure(struct device *dev)
+{
+ arch_teardown_dma_ops(dev);
+}
+EXPORT_SYMBOL_GPL(acpi_dma_deconfigure);
+
static void acpi_init_coherency(struct acpi_device *adev)
{
unsigned long long cca = 0;
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index ab00267..c29e07a 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -1738,8 +1738,7 @@ static void pci_dma_configure(struct pci_dev *dev)
if (attr == DEV_DMA_NOT_SUPPORTED)
dev_warn(&dev->dev, "DMA not supported.\n");
else
- arch_setup_dma_ops(&dev->dev, 0, 0, NULL,
- attr == DEV_DMA_COHERENT);
+ acpi_dma_configure(&dev->dev, attr);
}
pci_put_host_bridge_device(bridge);
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index c1a524d..4242c31 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -573,6 +573,8 @@ struct acpi_pci_root {
bool acpi_dma_supported(struct acpi_device *adev);
enum dev_dma_attr acpi_get_dma_attr(struct acpi_device *adev);
+void acpi_dma_configure(struct device *dev, enum dev_dma_attr attr);
+void acpi_dma_deconfigure(struct device *dev);
struct acpi_device *acpi_find_child_device(struct acpi_device *parent,
u64 address, bool check_children);
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 996a29c..8d15fc5 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -765,6 +765,11 @@ static inline enum dev_dma_attr acpi_get_dma_attr(struct acpi_device *adev)
return DEV_DMA_NOT_SUPPORTED;
}
+static inline void acpi_dma_configure(struct device *dev,
+ enum dev_dma_attr attr) { }
+
+static inline void acpi_dma_deconfigure(struct device *dev) { }
+
#define ACPI_PTR(_ptr) (NULL)
static inline void acpi_device_set_enumerated(struct acpi_device *adev)
--
2.10.0
The of_iommu_{set/get}_ops() API is used to associate a device
tree node with a specific set of IOMMU operations. The same
kernel interface is required on systems booting with ACPI, where
devices are not associated with a device tree node, therefore
the interface requires generalization.
The struct device fwnode member represents the fwnode token associated
with the device and the struct it points at is firmware specific;
regardless, it is initialized on both ACPI and DT systems and makes an
ideal candidate to use it to associate a set of IOMMU operations to a
given device, through its struct device.fwnode member pointer, paving
the way for representing per-device iommu_ops (ie an iommu instance
associated with a device).
Convert the DT specific of_iommu_{set/get}_ops() interface to
use struct device.fwnode as a look-up token, making the interface
usable on ACPI systems and rename the data structures and the
registration API so that they are made to represent their usage
more clearly.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Will Deacon <[email protected]>
Reviewed-by: Robin Murphy <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Robin Murphy <[email protected]>
Cc: Joerg Roedel <[email protected]>
---
drivers/iommu/iommu.c | 40 ++++++++++++++++++++++++++++++++++++++++
drivers/iommu/of_iommu.c | 39 ---------------------------------------
include/linux/iommu.h | 14 ++++++++++++++
include/linux/of_iommu.h | 12 ++++++++++--
4 files changed, 64 insertions(+), 41 deletions(-)
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 9a2f196..8d3e847 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -1615,6 +1615,46 @@ int iommu_request_dm_for_dev(struct device *dev)
return ret;
}
+struct iommu_instance {
+ struct list_head list;
+ struct fwnode_handle *fwnode;
+ const struct iommu_ops *ops;
+};
+static LIST_HEAD(iommu_instance_list);
+static DEFINE_SPINLOCK(iommu_instance_lock);
+
+void iommu_register_instance(struct fwnode_handle *fwnode,
+ const struct iommu_ops *ops)
+{
+ struct iommu_instance *iommu = kzalloc(sizeof(*iommu), GFP_KERNEL);
+
+ if (WARN_ON(!iommu))
+ return;
+
+ of_node_get(to_of_node(fwnode));
+ INIT_LIST_HEAD(&iommu->list);
+ iommu->fwnode = fwnode;
+ iommu->ops = ops;
+ spin_lock(&iommu_instance_lock);
+ list_add_tail(&iommu->list, &iommu_instance_list);
+ spin_unlock(&iommu_instance_lock);
+}
+
+const struct iommu_ops *iommu_get_instance(struct fwnode_handle *fwnode)
+{
+ struct iommu_instance *instance;
+ const struct iommu_ops *ops = NULL;
+
+ spin_lock(&iommu_instance_lock);
+ list_for_each_entry(instance, &iommu_instance_list, list)
+ if (instance->fwnode == fwnode) {
+ ops = instance->ops;
+ break;
+ }
+ spin_unlock(&iommu_instance_lock);
+ return ops;
+}
+
int iommu_fwspec_init(struct device *dev, struct fwnode_handle *iommu_fwnode,
const struct iommu_ops *ops)
{
diff --git a/drivers/iommu/of_iommu.c b/drivers/iommu/of_iommu.c
index 5b82862..0f57ddc 100644
--- a/drivers/iommu/of_iommu.c
+++ b/drivers/iommu/of_iommu.c
@@ -96,45 +96,6 @@ int of_get_dma_window(struct device_node *dn, const char *prefix, int index,
}
EXPORT_SYMBOL_GPL(of_get_dma_window);
-struct of_iommu_node {
- struct list_head list;
- struct device_node *np;
- const struct iommu_ops *ops;
-};
-static LIST_HEAD(of_iommu_list);
-static DEFINE_SPINLOCK(of_iommu_lock);
-
-void of_iommu_set_ops(struct device_node *np, const struct iommu_ops *ops)
-{
- struct of_iommu_node *iommu = kzalloc(sizeof(*iommu), GFP_KERNEL);
-
- if (WARN_ON(!iommu))
- return;
-
- of_node_get(np);
- INIT_LIST_HEAD(&iommu->list);
- iommu->np = np;
- iommu->ops = ops;
- spin_lock(&of_iommu_lock);
- list_add_tail(&iommu->list, &of_iommu_list);
- spin_unlock(&of_iommu_lock);
-}
-
-const struct iommu_ops *of_iommu_get_ops(struct device_node *np)
-{
- struct of_iommu_node *node;
- const struct iommu_ops *ops = NULL;
-
- spin_lock(&of_iommu_lock);
- list_for_each_entry(node, &of_iommu_list, list)
- if (node->np == np) {
- ops = node->ops;
- break;
- }
- spin_unlock(&of_iommu_lock);
- return ops;
-}
-
static int __get_pci_rid(struct pci_dev *pdev, u16 alias, void *data)
{
struct of_phandle_args *iommu_spec = data;
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index 436dc21..f2960e4 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -351,6 +351,9 @@ int iommu_fwspec_init(struct device *dev, struct fwnode_handle *iommu_fwnode,
const struct iommu_ops *ops);
void iommu_fwspec_free(struct device *dev);
int iommu_fwspec_add_ids(struct device *dev, u32 *ids, int num_ids);
+void iommu_register_instance(struct fwnode_handle *fwnode,
+ const struct iommu_ops *ops);
+const struct iommu_ops *iommu_get_instance(struct fwnode_handle *fwnode);
#else /* CONFIG_IOMMU_API */
@@ -580,6 +583,17 @@ static inline int iommu_fwspec_add_ids(struct device *dev, u32 *ids,
return -ENODEV;
}
+static inline void iommu_register_instance(struct fwnode_handle *fwnode,
+ const struct iommu_ops *ops)
+{
+}
+
+static inline
+const struct iommu_ops *iommu_get_instance(struct fwnode_handle *fwnode)
+{
+ return NULL;
+}
+
#endif /* CONFIG_IOMMU_API */
#endif /* __LINUX_IOMMU_H */
diff --git a/include/linux/of_iommu.h b/include/linux/of_iommu.h
index e80b9c7..6a7fc50 100644
--- a/include/linux/of_iommu.h
+++ b/include/linux/of_iommu.h
@@ -31,8 +31,16 @@ static inline const struct iommu_ops *of_iommu_configure(struct device *dev,
#endif /* CONFIG_OF_IOMMU */
-void of_iommu_set_ops(struct device_node *np, const struct iommu_ops *ops);
-const struct iommu_ops *of_iommu_get_ops(struct device_node *np);
+static inline void of_iommu_set_ops(struct device_node *np,
+ const struct iommu_ops *ops)
+{
+ iommu_register_instance(&np->fwnode, ops);
+}
+
+static inline const struct iommu_ops *of_iommu_get_ops(struct device_node *np)
+{
+ return iommu_get_instance(&np->fwnode);
+}
extern struct of_device_id __iommu_of_table;
--
2.10.0
Since commit e647b532275b ("ACPI: Add early device probing
infrastructure") the kernel has gained the infrastructure that allows
adding linker script section entries to execute ACPI driver callbacks
(ie probe routines) for all subsystems that register a table entry
in the respective kernel section (eg clocksource, irqchip).
Since ARM IOMMU devices data is described through IORT tables when
booting with ACPI, the ARM IOMMU drivers must be made able to hook ACPI
callback routines that are called to probe IORT entries and initialize
the respective IOMMU devices.
To avoid adding driver specific hooks into IORT table initialization
code (breaking therefore code modularity - ie ACPI IORT code must be made
aware of ARM SMMU drivers ACPI init callbacks), this patch adds code
that allows ARM SMMU drivers to take advantage of the ACPI early probing
infrastructure, so that they can add linker script section entries
containing drivers callback to be executed on IORT tables detection.
Since IORT nodes are differentiated by a type, the callback routines
can easily parse the IORT table entries, check the IORT nodes and
carry out some actions whenever the IORT node type associated with
the driver specific callback is matched.
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Hanjun Guo <[email protected]>
Reviewed-by: Tomasz Nowicki <[email protected]>
Tested-by: Hanjun Guo <[email protected]>
Tested-by: Tomasz Nowicki <[email protected]>
Cc: Tomasz Nowicki <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
Cc: Marc Zyngier <[email protected]>
---
drivers/acpi/arm64/iort.c | 13 ++++++++++---
include/asm-generic/vmlinux.lds.h | 1 +
include/linux/acpi_iort.h | 3 +++
3 files changed, 14 insertions(+), 3 deletions(-)
diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
index 6b81746..2c46ebc 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -361,8 +361,15 @@ void __init acpi_iort_init(void)
acpi_status status;
status = acpi_get_table(ACPI_SIG_IORT, 0, &iort_table);
- if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
- const char *msg = acpi_format_exception(status);
- pr_err("Failed to get table, %s\n", msg);
+ if (ACPI_FAILURE(status)) {
+ if (status != AE_NOT_FOUND) {
+ const char *msg = acpi_format_exception(status);
+
+ pr_err("Failed to get table, %s\n", msg);
+ }
+
+ return;
}
+
+ acpi_probe_device_table(iort);
}
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 31e1d63..9e3aa34 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -566,6 +566,7 @@
IRQCHIP_OF_MATCH_TABLE() \
ACPI_PROBE_TABLE(irqchip) \
ACPI_PROBE_TABLE(clksrc) \
+ ACPI_PROBE_TABLE(iort) \
EARLYCON_TABLE()
#define INIT_TEXT \
diff --git a/include/linux/acpi_iort.h b/include/linux/acpi_iort.h
index 0e32dac..d16fdda 100644
--- a/include/linux/acpi_iort.h
+++ b/include/linux/acpi_iort.h
@@ -39,4 +39,7 @@ static inline struct irq_domain *iort_get_device_domain(struct device *dev,
{ return NULL; }
#endif
+#define IORT_ACPI_DECLARE(name, table_id, fn) \
+ ACPI_DECLARE_PROBE_ENTRY(iort, name, table_id, 0, NULL, 0, fn)
+
#endif /* __ACPI_IORT_H__ */
--
2.10.0
On 2016/11/21 18:01, Lorenzo Pieralisi wrote:
> This patch series is v9 of a previous posting:
>
> https://lkml.org/lkml/2016/11/16/386
>
> v8 -> v9
> - Updated bypass flag handling in ARM SMMU v3 according to
> reviews
> - Removed SMMUv1/v2 configuration interrupt handling
> - Rebased against v4.9-rc5
> - Updated tags
I rebased on top of latest 4.9-rc7 (can apply cleanly), and tested on
Hisilicon D03, the SMMUv3 works fine as previous version.
Thanks
Hanjun
On 2016/11/21 18:01, Lorenzo Pieralisi wrote:
> Device drivers (eg ARM SMMU) need to know if a specific component
> is part of the IORT table, so that kernel data structures are not
> initialized at initcalls time if the respective component is not
> part of the IORT table.
>
> To this end, this patch adds a trivial function that allows detecting
> if a given IORT node type is present or not in the ACPI table, providing
> an ACPI IORT equivalent for of_find_matching_node().
>
> Signed-off-by: Lorenzo Pieralisi <[email protected]>
> Reviewed-by: Tomasz Nowicki <[email protected]>
> Tested-by: Hanjun Guo <[email protected]>
> Tested-by: Tomasz Nowicki <[email protected]>
> Cc: Hanjun Guo <[email protected]>
Acked-by: Hanjun Guo <[email protected]>
Thanks
Hanjun
On 2016/11/21 18:01, Lorenzo Pieralisi wrote:
> In ARM ACPI systems, IOMMU components are specified through static
> IORT table entries. In order to create platform devices for the
> corresponding ARM SMMU components, IORT kernel code should be made
> able to parse IORT table entries and create platform devices
> dynamically.
>
> This patch adds the generic IORT infrastructure required to create
> platform devices for ARM SMMUs.
>
> ARM SMMU versions have different resources requirement therefore this
> patch also introduces an IORT specific structure (ie iort_iommu_config)
> that contains hooks (to be defined when the corresponding ARM SMMU
> driver support is added to the kernel) to be used to define the
> platform devices names, init the IOMMUs, count their resources and
> finally initialize them.
>
> Signed-off-by: Lorenzo Pieralisi <[email protected]>
> Reviewed-by: Tomasz Nowicki <[email protected]>
> Tested-by: Hanjun Guo <[email protected]>
> Tested-by: Tomasz Nowicki <[email protected]>
> Cc: Hanjun Guo <[email protected]>
> Cc: Tomasz Nowicki <[email protected]>
> Cc: "Rafael J. Wysocki" <[email protected]>
> ---
> drivers/acpi/arm64/iort.c | 151 ++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 151 insertions(+)
Acked-by: Hanjun Guo <[email protected]>
Thanks
Hanjun
On 2016/11/21 18:01, Lorenzo Pieralisi wrote:
> IORT tables provide data that allow the kernel to carry out
> device ID mappings between endpoints and system components
> (eg interrupt controllers, IOMMUs). When the mapping for a
> given device ID is carried out, the translation mechanism
> is done on a per-subsystem basis rather than a component
> subtype (ie the IOMMU kernel layer will look for mappings
> from a device to all IORT node types corresponding to IOMMU
> components), therefore the corresponding mapping API should
> work on a range (ie mask) of IORT node types corresponding
> to a common set of components (eg IOMMUs) rather than a
> specific node type.
>
> Upgrade the IORT iort_node_map_rid() API to work with a
> type mask instead of a single node type so that it can
> be used for mappings that span multiple components types
> (ie IOMMUs).
>
> Signed-off-by: Lorenzo Pieralisi <[email protected]>
> Reviewed-by: Tomasz Nowicki <[email protected]>
> Tested-by: Hanjun Guo <[email protected]>
> Tested-by: Tomasz Nowicki <[email protected]>
> Cc: Hanjun Guo <[email protected]>
> Cc: Tomasz Nowicki <[email protected]>
> Cc: "Rafael J. Wysocki" <[email protected]>
> ---
> drivers/acpi/arm64/iort.c | 11 +++++++----
> 1 file changed, 7 insertions(+), 4 deletions(-)
Acked-by: Hanjun Guo <[email protected]>
Thanks
Hanjun
On 2016/11/21 18:01, Lorenzo Pieralisi wrote:
> The current IORT id mapping API requires components to provide
> an input requester ID (a Bus-Device-Function (BDF) identifier for
> PCI devices) to translate an input identifier to an output
> identifier through an IORT range mapping.
>
> Named components do not have an identifiable source ID therefore
> their respective input/output mapping can only be defined in
> IORT tables through single mappings, that provide a translation
> that does not require any input identifier.
>
> Current IORT interface for requester id mappings (iort_node_map_rid())
> is not suitable for components that do not provide a requester id,
> so it cannot be used for IORT named components.
>
> Add an interface to the IORT API to enable retrieval of id
> by allowing an indexed walk of the single mappings array for
> a given component, therefore completing the IORT mapping API.
>
> Signed-off-by: Lorenzo Pieralisi <[email protected]>
> Reviewed-by: Tomasz Nowicki <[email protected]>
> Tested-by: Hanjun Guo <[email protected]>
> Tested-by: Tomasz Nowicki <[email protected]>
> Cc: Hanjun Guo <[email protected]>
> Cc: Tomasz Nowicki <[email protected]>
> Cc: "Rafael J. Wysocki" <[email protected]>
> ---
> drivers/acpi/arm64/iort.c | 39 +++++++++++++++++++++++++++++++++++++++
> 1 file changed, 39 insertions(+)
Acked-by: Hanjun Guo <[email protected]>
Thanks
Hanjun
On 2016/11/21 18:01, Lorenzo Pieralisi wrote:
> DT based systems have a generic kernel API to configure IOMMUs
> for devices (ie of_iommu_configure()).
>
> On ARM based ACPI systems, the of_iommu_configure() equivalent can
> be implemented atop ACPI IORT kernel API, with the corresponding
> functions to map device identifiers to IOMMUs and retrieve the
> corresponding IOMMU operations necessary for DMA operations set-up.
>
> By relying on the iommu_fwspec generic kernel infrastructure,
> implement the IORT based IOMMU configuration for ARM ACPI systems
> and hook it up in the ACPI kernel layer that implements DMA
> configuration for a device.
>
> Signed-off-by: Lorenzo Pieralisi <[email protected]>
> Acked-by: Rafael J. Wysocki <[email protected]> [ACPI core]
> Reviewed-by: Tomasz Nowicki <[email protected]>
> Tested-by: Hanjun Guo <[email protected]>
> Tested-by: Tomasz Nowicki <[email protected]>
> Cc: Hanjun Guo <[email protected]>
Acked-by: Hanjun Guo <[email protected]>
Thanks
Hanjun
Rafael, Mark, Suravee,
On Mon, Nov 21, 2016 at 10:01:39AM +0000, Lorenzo Pieralisi wrote:
> On DT based systems, the of_dma_configure() API implements DMA
> configuration for a given device. On ACPI systems an API equivalent to
> of_dma_configure() is missing which implies that it is currently not
> possible to set-up DMA operations for devices through the ACPI generic
> kernel layer.
>
> This patch fills the gap by introducing acpi_dma_configure/deconfigure()
> calls that for now are just wrappers around arch_setup_dma_ops() and
> arch_teardown_dma_ops() and also updates ACPI and PCI core code to use
> the newly introduced acpi_dma_configure/acpi_dma_deconfigure functions.
>
> Since acpi_dma_configure() is used to configure DMA operations, the
> function initializes the dma/coherent_dma masks to sane default values
> if the current masks are uninitialized (also to keep the default values
> consistent with DT systems) to make sure the device has a complete
> default DMA set-up.
I spotted a niggle that unfortunately was hard to spot (and should not
be a problem per se but better safe than sorry) and I am not comfortable
with it.
Following commit d0562674838c ("ACPI / scan: Parse _CCA and setup
device coherency") in acpi_bind_one() we check if the acpi_device
associated with a device just added supports DMA, first it was
done with acpi_check_dma() and then commit 1831eff876bd ("device
property: ACPI: Make use of the new DMA Attribute APIs") changed
it to acpi_get_dma_attr().
The subsequent check (attr != DEV_DMA_NOT_SUPPORTED) is always true
on _any_ acpi device we pass to acpi_bind_one() on x86, which was
fine because we used it to call arch_setup_dma_ops(), which is a nop
on x86. On ARM64 a _CCA method is required to define if a device
supports DMA so (attr != DEV_DMA_NOT_SUPPORTED) may well be false.
Now, acpi_bind_one() is used to bind an acpi_device to its physical
node also for pseudo-devices like cpus and memory nodes. For those
objects, on x86, attr will always be != DEV_DMA_NOT_SUPPORTED.
So far so good, because on x86 arch_setup_dma_ops() is empty code.
With this patch, I use the (attr != DEV_DMA_NOT_SUPPORTED) check
to call acpi_dma_configure() which is basically a nop on x86 except
that it sets up the dma_mask/coherent_dma_mask to a sane default value
(after all we are setting up DMA for the device so it makes sense to
initialize the masks there if they were unset since we are configuring
DMA for the device in question) for the given device.
Problem is, as per the explanation above, we are also setting the
default dma masks for pseudo-devices (eg CPUs) that were previously
untouched, it should not be a problem per-se but I am not comfortable
with that, honestly it does not make much sense.
An easy "fix" would be to move the default dma masks initialization out
of acpi_dma_configure() (as it was in previous patch versions of this
series - I moved it to acpi_dma_configure() just a consolidation point
for initializing the masks instead of scattering them in every
acpi_dma_configure caller) I can send this as a fix-up patch to Joerg if
we think that's the right thing to do (or I can send it to Rafael later
when the code is in the merged depending on the timing) just let me
know please.
Thanks !
Lorenzo
> The DMA range size passed to arch_setup_dma_ops() is sized according
> to the device coherent_dma_mask (starting at address 0x0), mirroring the
> DT probing path behaviour when a dma-ranges property is not provided
> for the device being probed; this changes the current arch_setup_dma_ops()
> call parameters in the ACPI probing case, but since arch_setup_dma_ops()
> is a NOP on all architectures but ARM/ARM64 this patch does not change
> the current kernel behaviour on them.
>
> Signed-off-by: Lorenzo Pieralisi <[email protected]>
> Acked-by: Bjorn Helgaas <[email protected]> [pci]
> Acked-by: Rafael J. Wysocki <[email protected]>
> Reviewed-by: Tomasz Nowicki <[email protected]>
> Tested-by: Hanjun Guo <[email protected]>
> Tested-by: Tomasz Nowicki <[email protected]>
> Cc: Bjorn Helgaas <[email protected]>
> Cc: Robin Murphy <[email protected]>
> Cc: Tomasz Nowicki <[email protected]>
> Cc: Joerg Roedel <[email protected]>
> Cc: "Rafael J. Wysocki" <[email protected]>
> ---
> drivers/acpi/glue.c | 4 ++--
> drivers/acpi/scan.c | 40 ++++++++++++++++++++++++++++++++++++++++
> drivers/pci/probe.c | 3 +--
> include/acpi/acpi_bus.h | 2 ++
> include/linux/acpi.h | 5 +++++
> 5 files changed, 50 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
> index 5ea5dc2..f8d6564 100644
> --- a/drivers/acpi/glue.c
> +++ b/drivers/acpi/glue.c
> @@ -227,8 +227,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev)
>
> attr = acpi_get_dma_attr(acpi_dev);
> if (attr != DEV_DMA_NOT_SUPPORTED)
> - arch_setup_dma_ops(dev, 0, 0, NULL,
> - attr == DEV_DMA_COHERENT);
> + acpi_dma_configure(dev, attr);
>
> acpi_physnode_link_name(physical_node_name, node_id);
> retval = sysfs_create_link(&acpi_dev->dev.kobj, &dev->kobj,
> @@ -251,6 +250,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev)
> return 0;
>
> err:
> + acpi_dma_deconfigure(dev);
> ACPI_COMPANION_SET(dev, NULL);
> put_device(dev);
> put_device(&acpi_dev->dev);
> diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
> index 3d1856f..45b5710 100644
> --- a/drivers/acpi/scan.c
> +++ b/drivers/acpi/scan.c
> @@ -1370,6 +1370,46 @@ enum dev_dma_attr acpi_get_dma_attr(struct acpi_device *adev)
> return DEV_DMA_NON_COHERENT;
> }
>
> +/**
> + * acpi_dma_configure - Set-up DMA configuration for the device.
> + * @dev: The pointer to the device
> + * @attr: device dma attributes
> + */
> +void acpi_dma_configure(struct device *dev, enum dev_dma_attr attr)
> +{
> + /*
> + * Set default coherent_dma_mask to 32 bit. Drivers are expected to
> + * setup the correct supported mask.
> + */
> + if (!dev->coherent_dma_mask)
> + dev->coherent_dma_mask = DMA_BIT_MASK(32);
> +
> + /*
> + * Set it to coherent_dma_mask by default if the architecture
> + * code has not set it.
> + */
> + if (!dev->dma_mask)
> + dev->dma_mask = &dev->coherent_dma_mask;
> +
> + /*
> + * Assume dma valid range starts at 0 and covers the whole
> + * coherent_dma_mask.
> + */
> + arch_setup_dma_ops(dev, 0, dev->coherent_dma_mask + 1, NULL,
> + attr == DEV_DMA_COHERENT);
> +}
> +EXPORT_SYMBOL_GPL(acpi_dma_configure);
> +
> +/**
> + * acpi_dma_deconfigure - Tear-down DMA configuration for the device.
> + * @dev: The pointer to the device
> + */
> +void acpi_dma_deconfigure(struct device *dev)
> +{
> + arch_teardown_dma_ops(dev);
> +}
> +EXPORT_SYMBOL_GPL(acpi_dma_deconfigure);
> +
> static void acpi_init_coherency(struct acpi_device *adev)
> {
> unsigned long long cca = 0;
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index ab00267..c29e07a 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -1738,8 +1738,7 @@ static void pci_dma_configure(struct pci_dev *dev)
> if (attr == DEV_DMA_NOT_SUPPORTED)
> dev_warn(&dev->dev, "DMA not supported.\n");
> else
> - arch_setup_dma_ops(&dev->dev, 0, 0, NULL,
> - attr == DEV_DMA_COHERENT);
> + acpi_dma_configure(&dev->dev, attr);
> }
>
> pci_put_host_bridge_device(bridge);
> diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
> index c1a524d..4242c31 100644
> --- a/include/acpi/acpi_bus.h
> +++ b/include/acpi/acpi_bus.h
> @@ -573,6 +573,8 @@ struct acpi_pci_root {
>
> bool acpi_dma_supported(struct acpi_device *adev);
> enum dev_dma_attr acpi_get_dma_attr(struct acpi_device *adev);
> +void acpi_dma_configure(struct device *dev, enum dev_dma_attr attr);
> +void acpi_dma_deconfigure(struct device *dev);
>
> struct acpi_device *acpi_find_child_device(struct acpi_device *parent,
> u64 address, bool check_children);
> diff --git a/include/linux/acpi.h b/include/linux/acpi.h
> index 996a29c..8d15fc5 100644
> --- a/include/linux/acpi.h
> +++ b/include/linux/acpi.h
> @@ -765,6 +765,11 @@ static inline enum dev_dma_attr acpi_get_dma_attr(struct acpi_device *adev)
> return DEV_DMA_NOT_SUPPORTED;
> }
>
> +static inline void acpi_dma_configure(struct device *dev,
> + enum dev_dma_attr attr) { }
> +
> +static inline void acpi_dma_deconfigure(struct device *dev) { }
> +
> #define ACPI_PTR(_ptr) (NULL)
>
> static inline void acpi_device_set_enumerated(struct acpi_device *adev)
> --
> 2.10.0
>
On Fri, Dec 2, 2016 at 4:38 PM, Lorenzo Pieralisi
<[email protected]> wrote:
> Rafael, Mark, Suravee,
>
> On Mon, Nov 21, 2016 at 10:01:39AM +0000, Lorenzo Pieralisi wrote:
>> On DT based systems, the of_dma_configure() API implements DMA
>> configuration for a given device. On ACPI systems an API equivalent to
>> of_dma_configure() is missing which implies that it is currently not
>> possible to set-up DMA operations for devices through the ACPI generic
>> kernel layer.
>>
>> This patch fills the gap by introducing acpi_dma_configure/deconfigure()
>> calls that for now are just wrappers around arch_setup_dma_ops() and
>> arch_teardown_dma_ops() and also updates ACPI and PCI core code to use
>> the newly introduced acpi_dma_configure/acpi_dma_deconfigure functions.
>>
>> Since acpi_dma_configure() is used to configure DMA operations, the
>> function initializes the dma/coherent_dma masks to sane default values
>> if the current masks are uninitialized (also to keep the default values
>> consistent with DT systems) to make sure the device has a complete
>> default DMA set-up.
>
> I spotted a niggle that unfortunately was hard to spot (and should not
> be a problem per se but better safe than sorry) and I am not comfortable
> with it.
>
> Following commit d0562674838c ("ACPI / scan: Parse _CCA and setup
> device coherency") in acpi_bind_one() we check if the acpi_device
> associated with a device just added supports DMA, first it was
> done with acpi_check_dma() and then commit 1831eff876bd ("device
> property: ACPI: Make use of the new DMA Attribute APIs") changed
> it to acpi_get_dma_attr().
>
> The subsequent check (attr != DEV_DMA_NOT_SUPPORTED) is always true
> on _any_ acpi device we pass to acpi_bind_one() on x86, which was
> fine because we used it to call arch_setup_dma_ops(), which is a nop
> on x86. On ARM64 a _CCA method is required to define if a device
> supports DMA so (attr != DEV_DMA_NOT_SUPPORTED) may well be false.
>
> Now, acpi_bind_one() is used to bind an acpi_device to its physical
> node also for pseudo-devices like cpus and memory nodes. For those
> objects, on x86, attr will always be != DEV_DMA_NOT_SUPPORTED.
>
> So far so good, because on x86 arch_setup_dma_ops() is empty code.
>
> With this patch, I use the (attr != DEV_DMA_NOT_SUPPORTED) check
> to call acpi_dma_configure() which is basically a nop on x86 except
> that it sets up the dma_mask/coherent_dma_mask to a sane default value
> (after all we are setting up DMA for the device so it makes sense to
> initialize the masks there if they were unset since we are configuring
> DMA for the device in question) for the given device.
>
> Problem is, as per the explanation above, we are also setting the
> default dma masks for pseudo-devices (eg CPUs) that were previously
> untouched, it should not be a problem per-se but I am not comfortable
> with that, honestly it does not make much sense.
>
> An easy "fix" would be to move the default dma masks initialization out
> of acpi_dma_configure() (as it was in previous patch versions of this
> series - I moved it to acpi_dma_configure() just a consolidation point
> for initializing the masks instead of scattering them in every
> acpi_dma_configure caller) I can send this as a fix-up patch to Joerg if
> we think that's the right thing to do (or I can send it to Rafael later
> when the code is in the merged depending on the timing) just let me
> know please.
Why can't arch_setup_dma_ops() set those masks too?
Thanks,
Rafael
On Sat, Dec 03, 2016 at 03:11:09AM +0100, Rafael J. Wysocki wrote:
> On Fri, Dec 2, 2016 at 4:38 PM, Lorenzo Pieralisi
> <[email protected]> wrote:
> > Rafael, Mark, Suravee,
> >
> > On Mon, Nov 21, 2016 at 10:01:39AM +0000, Lorenzo Pieralisi wrote:
> >> On DT based systems, the of_dma_configure() API implements DMA
> >> configuration for a given device. On ACPI systems an API equivalent to
> >> of_dma_configure() is missing which implies that it is currently not
> >> possible to set-up DMA operations for devices through the ACPI generic
> >> kernel layer.
> >>
> >> This patch fills the gap by introducing acpi_dma_configure/deconfigure()
> >> calls that for now are just wrappers around arch_setup_dma_ops() and
> >> arch_teardown_dma_ops() and also updates ACPI and PCI core code to use
> >> the newly introduced acpi_dma_configure/acpi_dma_deconfigure functions.
> >>
> >> Since acpi_dma_configure() is used to configure DMA operations, the
> >> function initializes the dma/coherent_dma masks to sane default values
> >> if the current masks are uninitialized (also to keep the default values
> >> consistent with DT systems) to make sure the device has a complete
> >> default DMA set-up.
> >
> > I spotted a niggle that unfortunately was hard to spot (and should not
> > be a problem per se but better safe than sorry) and I am not comfortable
> > with it.
> >
> > Following commit d0562674838c ("ACPI / scan: Parse _CCA and setup
> > device coherency") in acpi_bind_one() we check if the acpi_device
> > associated with a device just added supports DMA, first it was
> > done with acpi_check_dma() and then commit 1831eff876bd ("device
> > property: ACPI: Make use of the new DMA Attribute APIs") changed
> > it to acpi_get_dma_attr().
> >
> > The subsequent check (attr != DEV_DMA_NOT_SUPPORTED) is always true
> > on _any_ acpi device we pass to acpi_bind_one() on x86, which was
> > fine because we used it to call arch_setup_dma_ops(), which is a nop
> > on x86. On ARM64 a _CCA method is required to define if a device
> > supports DMA so (attr != DEV_DMA_NOT_SUPPORTED) may well be false.
> >
> > Now, acpi_bind_one() is used to bind an acpi_device to its physical
> > node also for pseudo-devices like cpus and memory nodes. For those
> > objects, on x86, attr will always be != DEV_DMA_NOT_SUPPORTED.
> >
> > So far so good, because on x86 arch_setup_dma_ops() is empty code.
> >
> > With this patch, I use the (attr != DEV_DMA_NOT_SUPPORTED) check
> > to call acpi_dma_configure() which is basically a nop on x86 except
> > that it sets up the dma_mask/coherent_dma_mask to a sane default value
> > (after all we are setting up DMA for the device so it makes sense to
> > initialize the masks there if they were unset since we are configuring
> > DMA for the device in question) for the given device.
> >
> > Problem is, as per the explanation above, we are also setting the
> > default dma masks for pseudo-devices (eg CPUs) that were previously
> > untouched, it should not be a problem per-se but I am not comfortable
> > with that, honestly it does not make much sense.
> >
> > An easy "fix" would be to move the default dma masks initialization out
> > of acpi_dma_configure() (as it was in previous patch versions of this
> > series - I moved it to acpi_dma_configure() just a consolidation point
> > for initializing the masks instead of scattering them in every
> > acpi_dma_configure caller) I can send this as a fix-up patch to Joerg if
> > we think that's the right thing to do (or I can send it to Rafael later
> > when the code is in the merged depending on the timing) just let me
> > know please.
>
> Why can't arch_setup_dma_ops() set those masks too?
Because the dma masks set-up is done by the caller (see
of_dma_configure()) according to firmware configuration or
platform data knowledge. I wanted to replicate the of_dma_configure()
interface on ACPI for obvious reasons (on ARM systems), I stopped
short of adding ACPI code to mirror of_dma_get_range() equivalent
(through the _DMA object) but I am really really nervous about changing
the code path on x86 because in theory all is fine, in practice even
just setting the masks to sane values can have unexpected consequences,
I just can't know (that's why I wasn't doing it in the first iterations
of this series).
Side note: DT with of_dma_configure() and ACPI with
acpi_create_platform_device() set the default dma mask for all
platform devices already _regardless_ of what they really are, though
arguably acpi_bind_one() touches ways more devices.
I really think that removing the default dma masks settings from
acpi_dma_configure() is the safer thing to do for the time being (or
moving acpi_dma_configure() to acpi_create_platform_device(), where the
DMA masks are set-up by default by core ACPI. Mark, Suravee, what was
the rationale behind calling arch_setup_dma_ops() in acpi_bind_one() ?)
Please let me know, fix-up is trivial however we decide to proceed.
Thank you !
Lorenzo
On Sat, Dec 3, 2016 at 11:39 AM, Lorenzo Pieralisi
<[email protected]> wrote:
> On Sat, Dec 03, 2016 at 03:11:09AM +0100, Rafael J. Wysocki wrote:
>> On Fri, Dec 2, 2016 at 4:38 PM, Lorenzo Pieralisi
>> <[email protected]> wrote:
>> > Rafael, Mark, Suravee,
>> >
>> > On Mon, Nov 21, 2016 at 10:01:39AM +0000, Lorenzo Pieralisi wrote:
>> >> On DT based systems, the of_dma_configure() API implements DMA
>> >> configuration for a given device. On ACPI systems an API equivalent to
>> >> of_dma_configure() is missing which implies that it is currently not
>> >> possible to set-up DMA operations for devices through the ACPI generic
>> >> kernel layer.
>> >>
>> >> This patch fills the gap by introducing acpi_dma_configure/deconfigure()
>> >> calls that for now are just wrappers around arch_setup_dma_ops() and
>> >> arch_teardown_dma_ops() and also updates ACPI and PCI core code to use
>> >> the newly introduced acpi_dma_configure/acpi_dma_deconfigure functions.
>> >>
>> >> Since acpi_dma_configure() is used to configure DMA operations, the
>> >> function initializes the dma/coherent_dma masks to sane default values
>> >> if the current masks are uninitialized (also to keep the default values
>> >> consistent with DT systems) to make sure the device has a complete
>> >> default DMA set-up.
>> >
>> > I spotted a niggle that unfortunately was hard to spot (and should not
>> > be a problem per se but better safe than sorry) and I am not comfortable
>> > with it.
>> >
>> > Following commit d0562674838c ("ACPI / scan: Parse _CCA and setup
>> > device coherency") in acpi_bind_one() we check if the acpi_device
>> > associated with a device just added supports DMA, first it was
>> > done with acpi_check_dma() and then commit 1831eff876bd ("device
>> > property: ACPI: Make use of the new DMA Attribute APIs") changed
>> > it to acpi_get_dma_attr().
>> >
>> > The subsequent check (attr != DEV_DMA_NOT_SUPPORTED) is always true
>> > on _any_ acpi device we pass to acpi_bind_one() on x86, which was
>> > fine because we used it to call arch_setup_dma_ops(), which is a nop
>> > on x86. On ARM64 a _CCA method is required to define if a device
>> > supports DMA so (attr != DEV_DMA_NOT_SUPPORTED) may well be false.
>> >
>> > Now, acpi_bind_one() is used to bind an acpi_device to its physical
>> > node also for pseudo-devices like cpus and memory nodes. For those
>> > objects, on x86, attr will always be != DEV_DMA_NOT_SUPPORTED.
>> >
>> > So far so good, because on x86 arch_setup_dma_ops() is empty code.
>> >
>> > With this patch, I use the (attr != DEV_DMA_NOT_SUPPORTED) check
>> > to call acpi_dma_configure() which is basically a nop on x86 except
>> > that it sets up the dma_mask/coherent_dma_mask to a sane default value
>> > (after all we are setting up DMA for the device so it makes sense to
>> > initialize the masks there if they were unset since we are configuring
>> > DMA for the device in question) for the given device.
>> >
>> > Problem is, as per the explanation above, we are also setting the
>> > default dma masks for pseudo-devices (eg CPUs) that were previously
>> > untouched, it should not be a problem per-se but I am not comfortable
>> > with that, honestly it does not make much sense.
>> >
>> > An easy "fix" would be to move the default dma masks initialization out
>> > of acpi_dma_configure() (as it was in previous patch versions of this
>> > series - I moved it to acpi_dma_configure() just a consolidation point
>> > for initializing the masks instead of scattering them in every
>> > acpi_dma_configure caller) I can send this as a fix-up patch to Joerg if
>> > we think that's the right thing to do (or I can send it to Rafael later
>> > when the code is in the merged depending on the timing) just let me
>> > know please.
>>
>> Why can't arch_setup_dma_ops() set those masks too?
>
> Because the dma masks set-up is done by the caller (see
> of_dma_configure()) according to firmware configuration or
> platform data knowledge. I wanted to replicate the of_dma_configure()
> interface on ACPI for obvious reasons (on ARM systems), I stopped
> short of adding ACPI code to mirror of_dma_get_range() equivalent
> (through the _DMA object) but I am really really nervous about changing
> the code path on x86 because in theory all is fine, in practice even
> just setting the masks to sane values can have unexpected consequences,
> I just can't know (that's why I wasn't doing it in the first iterations
> of this series).
>
> Side note: DT with of_dma_configure() and ACPI with
> acpi_create_platform_device() set the default dma mask for all
> platform devices already _regardless_ of what they really are, though
> arguably acpi_bind_one() touches ways more devices.
>
> I really think that removing the default dma masks settings from
> acpi_dma_configure() is the safer thing to do for the time being (or
> moving acpi_dma_configure() to acpi_create_platform_device(), where the
> DMA masks are set-up by default by core ACPI. Mark, Suravee, what was
> the rationale behind calling arch_setup_dma_ops() in acpi_bind_one() ?)
Alternatively, you can add one more arch wrapper that will be a no-op
on x86 and that will set up the default masks and call
arch_setup_dma_ops() on ARM. Then, you can invoke that from
acpi_dma_configure().
Or make the definition of acpi_dma_configure() itself depend on the
architecture.
Thanks,
Rafael
Hi Lorenzo,
>
>On Sat, Dec 3, 2016 at 11:39 AM, Lorenzo Pieralisi
><[email protected]> wrote:
>> On Sat, Dec 03, 2016 at 03:11:09AM +0100, Rafael J. Wysocki wrote:
>>> On Fri, Dec 2, 2016 at 4:38 PM, Lorenzo Pieralisi
>>> <[email protected]> wrote:
>>> > Rafael, Mark, Suravee,
>>> >
>>> > On Mon, Nov 21, 2016 at 10:01:39AM +0000, Lorenzo Pieralisi wrote:
>>> >> On DT based systems, the of_dma_configure() API implements DMA
>>> >> configuration for a given device. On ACPI systems an API equivalent to
>>> >> of_dma_configure() is missing which implies that it is currently not
>>> >> possible to set-up DMA operations for devices through the ACPI generic
>>> >> kernel layer.
>>> >>
>>> >> This patch fills the gap by introducing acpi_dma_configure/deconfigure()
>>> >> calls that for now are just wrappers around arch_setup_dma_ops() and
>>> >> arch_teardown_dma_ops() and also updates ACPI and PCI core code to use
>>> >> the newly introduced acpi_dma_configure/acpi_dma_deconfigure functions.
>>> >>
>>> >> Since acpi_dma_configure() is used to configure DMA operations, the
>>> >> function initializes the dma/coherent_dma masks to sane default values
>>> >> if the current masks are uninitialized (also to keep the default values
>>> >> consistent with DT systems) to make sure the device has a complete
>>> >> default DMA set-up.
>>> >
>>> > I spotted a niggle that unfortunately was hard to spot (and should not
>>> > be a problem per se but better safe than sorry) and I am not comfortable
>>> > with it.
>>> >
>>> > Following commit d0562674838c ("ACPI / scan: Parse _CCA and setup
>>> > device coherency") in acpi_bind_one() we check if the acpi_device
>>> > associated with a device just added supports DMA, first it was
>>> > done with acpi_check_dma() and then commit 1831eff876bd ("device
>>> > property: ACPI: Make use of the new DMA Attribute APIs") changed
>>> > it to acpi_get_dma_attr().
>>> >
>>> > The subsequent check (attr != DEV_DMA_NOT_SUPPORTED) is always true
>>> > on _any_ acpi device we pass to acpi_bind_one() on x86, which was
>>> > fine because we used it to call arch_setup_dma_ops(), which is a nop
>>> > on x86. On ARM64 a _CCA method is required to define if a device
>>> > supports DMA so (attr != DEV_DMA_NOT_SUPPORTED) may well be false.
>>> >
>>> > Now, acpi_bind_one() is used to bind an acpi_device to its physical
>>> > node also for pseudo-devices like cpus and memory nodes. For those
>>> > objects, on x86, attr will always be != DEV_DMA_NOT_SUPPORTED.
>>> >
>>> > So far so good, because on x86 arch_setup_dma_ops() is empty code.
>>> >
>>> > With this patch, I use the (attr != DEV_DMA_NOT_SUPPORTED) check
>>> > to call acpi_dma_configure() which is basically a nop on x86 except
>>> > that it sets up the dma_mask/coherent_dma_mask to a sane default value
>>> > (after all we are setting up DMA for the device so it makes sense to
>>> > initialize the masks there if they were unset since we are configuring
>>> > DMA for the device in question) for the given device.
>>> >
>>> > Problem is, as per the explanation above, we are also setting the
>>> > default dma masks for pseudo-devices (eg CPUs) that were previously
>>> > untouched, it should not be a problem per-se but I am not comfortable
>>> > with that, honestly it does not make much sense.
>>> >
>>> > An easy "fix" would be to move the default dma masks initialization out
>>> > of acpi_dma_configure() (as it was in previous patch versions of this
>>> > series - I moved it to acpi_dma_configure() just a consolidation point
>>> > for initializing the masks instead of scattering them in every
>>> > acpi_dma_configure caller) I can send this as a fix-up patch to Joerg if
>>> > we think that's the right thing to do (or I can send it to Rafael later
>>> > when the code is in the merged depending on the timing) just let me
>>> > know please.
>>>
>>> Why can't arch_setup_dma_ops() set those masks too?
>>
>> Because the dma masks set-up is done by the caller (see
>> of_dma_configure()) according to firmware configuration or
>> platform data knowledge. I wanted to replicate the of_dma_configure()
>> interface on ACPI for obvious reasons (on ARM systems), I stopped
>> short of adding ACPI code to mirror of_dma_get_range() equivalent
>> (through the _DMA object) but I am really really nervous about changing
>> the code path on x86 because in theory all is fine, in practice even
>> just setting the masks to sane values can have unexpected consequences,
>> I just can't know (that's why I wasn't doing it in the first iterations
>> of this series).
>>
>> Side note: DT with of_dma_configure() and ACPI with
>> acpi_create_platform_device() set the default dma mask for all
>> platform devices already _regardless_ of what they really are, though
>> arguably acpi_bind_one() touches ways more devices.
>>
>> I really think that removing the default dma masks settings from
>> acpi_dma_configure() is the safer thing to do for the time being (or
>> moving acpi_dma_configure() to acpi_create_platform_device(), where the
>> DMA masks are set-up by default by core ACPI. Mark, Suravee, what was
>> the rationale behind calling arch_setup_dma_ops() in acpi_bind_one() ?)
>
>Alternatively, you can add one more arch wrapper that will be a no-op
>on x86 and that will set up the default masks and call
>arch_setup_dma_ops() on ARM. Then, you can invoke that from
>acpi_dma_configure().
>
>Or make the definition of acpi_dma_configure() itself depend on the
>architecture.
>
So is it better that either removing the masks from acpi_dma_configure (or)
creating the wrapper as Rafael mentioned, than moving
acpi_dma_configure itself , because with something like iommu probe
deferral that is tried, acpi_dma_configure is getting invoked from a device's
really_probe, a different path again ?
Regards,
Sricharan
On Mon, Dec 05, 2016 at 03:22:02PM +0530, Sricharan wrote:
> Hi Lorenzo,
>
> >
> >On Sat, Dec 3, 2016 at 11:39 AM, Lorenzo Pieralisi
> ><[email protected]> wrote:
> >> On Sat, Dec 03, 2016 at 03:11:09AM +0100, Rafael J. Wysocki wrote:
> >>> On Fri, Dec 2, 2016 at 4:38 PM, Lorenzo Pieralisi
> >>> <[email protected]> wrote:
> >>> > Rafael, Mark, Suravee,
> >>> >
> >>> > On Mon, Nov 21, 2016 at 10:01:39AM +0000, Lorenzo Pieralisi wrote:
> >>> >> On DT based systems, the of_dma_configure() API implements DMA
> >>> >> configuration for a given device. On ACPI systems an API equivalent to
> >>> >> of_dma_configure() is missing which implies that it is currently not
> >>> >> possible to set-up DMA operations for devices through the ACPI generic
> >>> >> kernel layer.
> >>> >>
> >>> >> This patch fills the gap by introducing acpi_dma_configure/deconfigure()
> >>> >> calls that for now are just wrappers around arch_setup_dma_ops() and
> >>> >> arch_teardown_dma_ops() and also updates ACPI and PCI core code to use
> >>> >> the newly introduced acpi_dma_configure/acpi_dma_deconfigure functions.
> >>> >>
> >>> >> Since acpi_dma_configure() is used to configure DMA operations, the
> >>> >> function initializes the dma/coherent_dma masks to sane default values
> >>> >> if the current masks are uninitialized (also to keep the default values
> >>> >> consistent with DT systems) to make sure the device has a complete
> >>> >> default DMA set-up.
> >>> >
> >>> > I spotted a niggle that unfortunately was hard to spot (and should not
> >>> > be a problem per se but better safe than sorry) and I am not comfortable
> >>> > with it.
> >>> >
> >>> > Following commit d0562674838c ("ACPI / scan: Parse _CCA and setup
> >>> > device coherency") in acpi_bind_one() we check if the acpi_device
> >>> > associated with a device just added supports DMA, first it was
> >>> > done with acpi_check_dma() and then commit 1831eff876bd ("device
> >>> > property: ACPI: Make use of the new DMA Attribute APIs") changed
> >>> > it to acpi_get_dma_attr().
> >>> >
> >>> > The subsequent check (attr != DEV_DMA_NOT_SUPPORTED) is always true
> >>> > on _any_ acpi device we pass to acpi_bind_one() on x86, which was
> >>> > fine because we used it to call arch_setup_dma_ops(), which is a nop
> >>> > on x86. On ARM64 a _CCA method is required to define if a device
> >>> > supports DMA so (attr != DEV_DMA_NOT_SUPPORTED) may well be false.
> >>> >
> >>> > Now, acpi_bind_one() is used to bind an acpi_device to its physical
> >>> > node also for pseudo-devices like cpus and memory nodes. For those
> >>> > objects, on x86, attr will always be != DEV_DMA_NOT_SUPPORTED.
> >>> >
> >>> > So far so good, because on x86 arch_setup_dma_ops() is empty code.
> >>> >
> >>> > With this patch, I use the (attr != DEV_DMA_NOT_SUPPORTED) check
> >>> > to call acpi_dma_configure() which is basically a nop on x86 except
> >>> > that it sets up the dma_mask/coherent_dma_mask to a sane default value
> >>> > (after all we are setting up DMA for the device so it makes sense to
> >>> > initialize the masks there if they were unset since we are configuring
> >>> > DMA for the device in question) for the given device.
> >>> >
> >>> > Problem is, as per the explanation above, we are also setting the
> >>> > default dma masks for pseudo-devices (eg CPUs) that were previously
> >>> > untouched, it should not be a problem per-se but I am not comfortable
> >>> > with that, honestly it does not make much sense.
> >>> >
> >>> > An easy "fix" would be to move the default dma masks initialization out
> >>> > of acpi_dma_configure() (as it was in previous patch versions of this
> >>> > series - I moved it to acpi_dma_configure() just a consolidation point
> >>> > for initializing the masks instead of scattering them in every
> >>> > acpi_dma_configure caller) I can send this as a fix-up patch to Joerg if
> >>> > we think that's the right thing to do (or I can send it to Rafael later
> >>> > when the code is in the merged depending on the timing) just let me
> >>> > know please.
> >>>
> >>> Why can't arch_setup_dma_ops() set those masks too?
> >>
> >> Because the dma masks set-up is done by the caller (see
> >> of_dma_configure()) according to firmware configuration or
> >> platform data knowledge. I wanted to replicate the of_dma_configure()
> >> interface on ACPI for obvious reasons (on ARM systems), I stopped
> >> short of adding ACPI code to mirror of_dma_get_range() equivalent
> >> (through the _DMA object) but I am really really nervous about changing
> >> the code path on x86 because in theory all is fine, in practice even
> >> just setting the masks to sane values can have unexpected consequences,
> >> I just can't know (that's why I wasn't doing it in the first iterations
> >> of this series).
> >>
> >> Side note: DT with of_dma_configure() and ACPI with
> >> acpi_create_platform_device() set the default dma mask for all
> >> platform devices already _regardless_ of what they really are, though
> >> arguably acpi_bind_one() touches ways more devices.
> >>
> >> I really think that removing the default dma masks settings from
> >> acpi_dma_configure() is the safer thing to do for the time being (or
> >> moving acpi_dma_configure() to acpi_create_platform_device(), where the
> >> DMA masks are set-up by default by core ACPI. Mark, Suravee, what was
> >> the rationale behind calling arch_setup_dma_ops() in acpi_bind_one() ?)
> >
> >Alternatively, you can add one more arch wrapper that will be a no-op
> >on x86 and that will set up the default masks and call
> >arch_setup_dma_ops() on ARM. Then, you can invoke that from
> >acpi_dma_configure().
> >
> >Or make the definition of acpi_dma_configure() itself depend on the
> >architecture.
> >
>
> So is it better that either removing the masks from acpi_dma_configure
> (or) creating the wrapper as Rafael mentioned, than moving
> acpi_dma_configure itself , because with something like iommu probe
> deferral that is tried, acpi_dma_configure is getting invoked from a
> device's really_probe, a different path again ?
Yes, I thought about that too. Given what I said above (ie I would like
to extend the mask set-up with _DMA object - that is generic ACPI but
can affect legacy x86 - and if that does not work through IORT specific
bindings), as per Rafael suggestion I added an iort_set_dma_mask wrapper
that is a NOP on x86, leaving acpi_dma_configure() unchanged for ARM64.
Patch coming, thanks everyone.
Lorenzo