linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
* [PATCH V10 0/9] vfio, platform: add ACPI support
@ 2016-07-18 23:09 Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 1/9] vfio: platform: rename reset function Sinan Kaya
                   ` (8 more replies)
  0 siblings, 9 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

The current code only supports the device tree based platforms.
The code checks for the presence of a reset driver and calls the reset
function pointer by looking up the reset driver as a module.

ACPI defines _RST method to perform device level reset. After the _RST
method is executed, the OS can resume using the device.

The patchset is moving the device tree specific pieces out of the code
to common functions so that ACPI support is added without impacting the
rest of the code.

During probe, the ACPI HID of the object will be saved and will be used to
determine if this is an ACPI capable platform or not. If acpihid is NULL
then, device tree functions are called.

In addition to plumbing ACPI support, reset functionality is now a
requirement by default.

The code was allowing platform devices to be used without a supporting
VFIO reset driver. The hardware can be left in some inconsistent state
after a guest machine abort.

The reset driver will put the hardware back to safe state and disable
interrupts before returning the control back to the host machine.

Adding a new reset_required kernel module option to platform
VFIO drivers with a default value of true.

New requirements are:
1. A reset function needs to be implemented by the corresponding driver
via DT/ACPI.
2. The reset function needs to be discovered via DT/ACPI.

The probe of the driver will fail if any of the above conditions are
not satisfied.


Changes from V9:
1. Reorder structure to move the reset_required boolean to the end.
2. Remove the static inline from vfio_platform_acpi_has_reset
3. Place sanity checks into vfio_platform_acpi_probe for HID presence.
4. Retrieve ACPI companion device only if acpi_disabled is not set.

Sinan Kaya (9):
  vfio: platform: rename reset function
  vfio: platform: move reset call to a common function
  vfio: platform: determine reset capability
  vfio: platform: add support for ACPI probe
  vfio: platform: add extra debug info argument to call reset
  vfio: platform: call _RST method when using ACPI
  vfio, platform: make reset driver a requirement by default
  vfio: platform: check reset call return code during open
  vfio: platform: check reset call return code during release

 drivers/vfio/platform/vfio_amba.c             |   1 +
 drivers/vfio/platform/vfio_platform.c         |   5 +
 drivers/vfio/platform/vfio_platform_common.c  | 197 ++++++++++++++++++++++----
 drivers/vfio/platform/vfio_platform_private.h |   9 +-
 4 files changed, 178 insertions(+), 34 deletions(-)

-- 
1.8.2.1

^ permalink raw reply	[flat|nested] 16+ messages in thread

* [PATCH V10 1/9] vfio: platform: rename reset function
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 2/9] vfio: platform: move reset call to a common function Sinan Kaya
                   ` (7 subsequent siblings)
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

Renaming the reset function to of_reset as it is only used
by the device tree based platforms.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Reviewed-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
---
 drivers/vfio/platform/vfio_platform_common.c  | 30 +++++++++++++--------------
 drivers/vfio/platform/vfio_platform_private.h |  6 +++---
 2 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index e65b142..08fd7c2 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -41,7 +41,7 @@ static vfio_platform_reset_fn_t vfio_platform_lookup_reset(const char *compat,
 		if (!strcmp(iter->compat, compat) &&
 			try_module_get(iter->owner)) {
 			*module = iter->owner;
-			reset_fn = iter->reset;
+			reset_fn = iter->of_reset;
 			break;
 		}
 	}
@@ -51,18 +51,18 @@ static vfio_platform_reset_fn_t vfio_platform_lookup_reset(const char *compat,
 
 static void vfio_platform_get_reset(struct vfio_platform_device *vdev)
 {
-	vdev->reset = vfio_platform_lookup_reset(vdev->compat,
-						&vdev->reset_module);
-	if (!vdev->reset) {
+	vdev->of_reset = vfio_platform_lookup_reset(vdev->compat,
+						    &vdev->reset_module);
+	if (!vdev->of_reset) {
 		request_module("vfio-reset:%s", vdev->compat);
-		vdev->reset = vfio_platform_lookup_reset(vdev->compat,
-							 &vdev->reset_module);
+		vdev->of_reset = vfio_platform_lookup_reset(vdev->compat,
+							&vdev->reset_module);
 	}
 }
 
 static void vfio_platform_put_reset(struct vfio_platform_device *vdev)
 {
-	if (vdev->reset)
+	if (vdev->of_reset)
 		module_put(vdev->reset_module);
 }
 
@@ -141,9 +141,9 @@ static void vfio_platform_release(void *device_data)
 	mutex_lock(&driver_lock);
 
 	if (!(--vdev->refcnt)) {
-		if (vdev->reset) {
+		if (vdev->of_reset) {
 			dev_info(vdev->device, "reset\n");
-			vdev->reset(vdev);
+			vdev->of_reset(vdev);
 		} else {
 			dev_warn(vdev->device, "no reset function found!\n");
 		}
@@ -175,9 +175,9 @@ static int vfio_platform_open(void *device_data)
 		if (ret)
 			goto err_irq;
 
-		if (vdev->reset) {
+		if (vdev->of_reset) {
 			dev_info(vdev->device, "reset\n");
-			vdev->reset(vdev);
+			vdev->of_reset(vdev);
 		} else {
 			dev_warn(vdev->device, "no reset function found!\n");
 		}
@@ -213,7 +213,7 @@ static long vfio_platform_ioctl(void *device_data,
 		if (info.argsz < minsz)
 			return -EINVAL;
 
-		if (vdev->reset)
+		if (vdev->of_reset)
 			vdev->flags |= VFIO_DEVICE_FLAGS_RESET;
 		info.flags = vdev->flags;
 		info.num_regions = vdev->num_regions;
@@ -312,8 +312,8 @@ static long vfio_platform_ioctl(void *device_data,
 		return ret;
 
 	} else if (cmd == VFIO_DEVICE_RESET) {
-		if (vdev->reset)
-			return vdev->reset(vdev);
+		if (vdev->of_reset)
+			return vdev->of_reset(vdev);
 		else
 			return -EINVAL;
 	}
@@ -611,7 +611,7 @@ void vfio_platform_unregister_reset(const char *compat,
 
 	mutex_lock(&driver_lock);
 	list_for_each_entry_safe(iter, temp, &reset_list, link) {
-		if (!strcmp(iter->compat, compat) && (iter->reset == fn)) {
+		if (!strcmp(iter->compat, compat) && (iter->of_reset == fn)) {
 			list_del(&iter->link);
 			break;
 		}
diff --git a/drivers/vfio/platform/vfio_platform_private.h b/drivers/vfio/platform/vfio_platform_private.h
index 42816dd..71ed7d1 100644
--- a/drivers/vfio/platform/vfio_platform_private.h
+++ b/drivers/vfio/platform/vfio_platform_private.h
@@ -71,7 +71,7 @@ struct vfio_platform_device {
 	struct resource*
 		(*get_resource)(struct vfio_platform_device *vdev, int i);
 	int	(*get_irq)(struct vfio_platform_device *vdev, int i);
-	int	(*reset)(struct vfio_platform_device *vdev);
+	int	(*of_reset)(struct vfio_platform_device *vdev);
 };
 
 typedef int (*vfio_platform_reset_fn_t)(struct vfio_platform_device *vdev);
@@ -80,7 +80,7 @@ struct vfio_platform_reset_node {
 	struct list_head link;
 	char *compat;
 	struct module *owner;
-	vfio_platform_reset_fn_t reset;
+	vfio_platform_reset_fn_t of_reset;
 };
 
 extern int vfio_platform_probe_common(struct vfio_platform_device *vdev,
@@ -103,7 +103,7 @@ extern void vfio_platform_unregister_reset(const char *compat,
 static struct vfio_platform_reset_node __reset ## _node = {	\
 	.owner = THIS_MODULE,					\
 	.compat = __compat,					\
-	.reset = __reset,					\
+	.of_reset = __reset,					\
 };								\
 __vfio_platform_register_reset(&__reset ## _node)
 
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 2/9] vfio: platform: move reset call to a common function
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 1/9] vfio: platform: rename reset function Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 3/9] vfio: platform: determine reset capability Sinan Kaya
                   ` (6 subsequent siblings)
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

The reset call sequence seems to replicate itself multiple times
across the file. Grouping them together for maintenance reasons.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Reviewed-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
---
 drivers/vfio/platform/vfio_platform_common.c | 30 +++++++++++++---------------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index 08fd7c2..3e2a7c0 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -134,6 +134,17 @@ static void vfio_platform_regions_cleanup(struct vfio_platform_device *vdev)
 	kfree(vdev->regions);
 }
 
+static int vfio_platform_call_reset(struct vfio_platform_device *vdev)
+{
+	if (vdev->of_reset) {
+		dev_info(vdev->device, "reset\n");
+		return vdev->of_reset(vdev);
+	}
+
+	dev_warn(vdev->device, "no reset function found!\n");
+	return -EINVAL;
+}
+
 static void vfio_platform_release(void *device_data)
 {
 	struct vfio_platform_device *vdev = device_data;
@@ -141,12 +152,7 @@ static void vfio_platform_release(void *device_data)
 	mutex_lock(&driver_lock);
 
 	if (!(--vdev->refcnt)) {
-		if (vdev->of_reset) {
-			dev_info(vdev->device, "reset\n");
-			vdev->of_reset(vdev);
-		} else {
-			dev_warn(vdev->device, "no reset function found!\n");
-		}
+		vfio_platform_call_reset(vdev);
 		vfio_platform_regions_cleanup(vdev);
 		vfio_platform_irq_cleanup(vdev);
 	}
@@ -175,12 +181,7 @@ static int vfio_platform_open(void *device_data)
 		if (ret)
 			goto err_irq;
 
-		if (vdev->of_reset) {
-			dev_info(vdev->device, "reset\n");
-			vdev->of_reset(vdev);
-		} else {
-			dev_warn(vdev->device, "no reset function found!\n");
-		}
+		vfio_platform_call_reset(vdev);
 	}
 
 	vdev->refcnt++;
@@ -312,10 +313,7 @@ static long vfio_platform_ioctl(void *device_data,
 		return ret;
 
 	} else if (cmd == VFIO_DEVICE_RESET) {
-		if (vdev->of_reset)
-			return vdev->of_reset(vdev);
-		else
-			return -EINVAL;
+		return vfio_platform_call_reset(vdev);
 	}
 
 	return -ENOTTY;
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 3/9] vfio: platform: determine reset capability
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 1/9] vfio: platform: rename reset function Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 2/9] vfio: platform: move reset call to a common function Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 4/9] vfio: platform: add support for ACPI probe Sinan Kaya
                   ` (5 subsequent siblings)
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

Creating a new function to determine if this driver supports reset
function or not. This is an attempt to abstract device tree calls
from the rest of the code.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Reviewed-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
---
 drivers/vfio/platform/vfio_platform_common.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index 3e2a7c0..6be92c3 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -49,6 +49,11 @@ static vfio_platform_reset_fn_t vfio_platform_lookup_reset(const char *compat,
 	return reset_fn;
 }
 
+static bool vfio_platform_has_reset(struct vfio_platform_device *vdev)
+{
+	return vdev->of_reset ? true : false;
+}
+
 static void vfio_platform_get_reset(struct vfio_platform_device *vdev)
 {
 	vdev->of_reset = vfio_platform_lookup_reset(vdev->compat,
@@ -214,7 +219,7 @@ static long vfio_platform_ioctl(void *device_data,
 		if (info.argsz < minsz)
 			return -EINVAL;
 
-		if (vdev->of_reset)
+		if (vfio_platform_has_reset(vdev))
 			vdev->flags |= VFIO_DEVICE_FLAGS_RESET;
 		info.flags = vdev->flags;
 		info.num_regions = vdev->num_regions;
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 4/9] vfio: platform: add support for ACPI probe
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
                   ` (2 preceding siblings ...)
  2016-07-18 23:09 ` [PATCH V10 3/9] vfio: platform: determine reset capability Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-19  0:00   ` Alex Williamson
  2016-07-18 23:09 ` [PATCH V10 5/9] vfio: platform: add extra debug info argument to call reset Sinan Kaya
                   ` (4 subsequent siblings)
  8 siblings, 1 reply; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

The code is using the compatible DT string to associate a reset driver
with the actual device itself. The compatible string does not exist on
ACPI based systems. HID is the unique identifier for a device driver
instead.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
---
 drivers/vfio/platform/vfio_platform_common.c  | 69 +++++++++++++++++++++++++--
 drivers/vfio/platform/vfio_platform_private.h |  1 +
 2 files changed, 65 insertions(+), 5 deletions(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index 6be92c3..a5299f6 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -13,6 +13,7 @@
  */
 
 #include <linux/device.h>
+#include <linux/acpi.h>
 #include <linux/iommu.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
@@ -49,6 +50,32 @@ static vfio_platform_reset_fn_t vfio_platform_lookup_reset(const char *compat,
 	return reset_fn;
 }
 
+static int vfio_platform_acpi_probe(struct vfio_platform_device *vdev,
+				    struct device *dev)
+{
+	struct acpi_device *adev;
+
+	if (acpi_disabled)
+		return -EPERM;
+
+	adev = ACPI_COMPANION(dev);
+	if (!adev) {
+		pr_err("VFIO: ACPI companion device not found for %s\n",
+			vdev->name);
+		return -ENODEV;
+	}
+
+#ifdef CONFIG_ACPI
+	vdev->acpihid = acpi_device_hid(adev);
+	if (!vdev->acpihid) {
+		pr_err("VFIO: cannot find ACPI HID for %s\n",
+		       vdev->name);
+		return -EINVAL;
+	}
+#endif
+	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;
+}
+
 static bool vfio_platform_has_reset(struct vfio_platform_device *vdev)
 {
 	return vdev->of_reset ? true : false;
@@ -547,6 +574,37 @@ static const struct vfio_device_ops vfio_platform_ops = {
 	.mmap		= vfio_platform_mmap,
 };
 
+int vfio_platform_of_probe(struct vfio_platform_device *vdev,
+			   struct device *dev)
+{
+	int ret;
+
+	ret = device_property_read_string(dev, "compatible",
+					  &vdev->compat);
+	if (ret)
+		pr_err("VFIO: cannot retrieve compat for %s\n",
+			vdev->name);
+
+	return ret;
+}
+
+/*
+ * There can be two kernel build combinations. One build where
+ * ACPI is not selected in Kconfig and another one with the ACPI Kconfig.
+ *
+ * In the first case, vfio_platform_acpi_probe will return since
+ * acpi_disabled is 1. DT user will not see any kind of messages from
+ * ACPI.
+ *
+ * In the second case, both DT and ACPI is compiled in but the system is
+ * booting with any of these combinations.
+ *
+ * If the firmware is DT type, then acpi_disabled is 1. The ACPI probe routine
+ * terminates immediately without any messages.
+ *
+ * If the firmware is ACPI type, then acpi_disabled is 0. All other checks are
+ * valid checks. We cannot claim that this system is DT.
+ */
 int vfio_platform_probe_common(struct vfio_platform_device *vdev,
 			       struct device *dev)
 {
@@ -556,11 +614,12 @@ int vfio_platform_probe_common(struct vfio_platform_device *vdev,
 	if (!vdev)
 		return -EINVAL;
 
-	ret = device_property_read_string(dev, "compatible", &vdev->compat);
-	if (ret) {
-		pr_err("VFIO: cannot retrieve compat for %s\n", vdev->name);
-		return -EINVAL;
-	}
+	ret = vfio_platform_acpi_probe(vdev, dev);
+	if (ret)
+		ret = vfio_platform_of_probe(vdev, dev);
+
+	if (ret)
+		return ret;
 
 	vdev->device = dev;
 
diff --git a/drivers/vfio/platform/vfio_platform_private.h b/drivers/vfio/platform/vfio_platform_private.h
index 71ed7d1..ba9e4f8 100644
--- a/drivers/vfio/platform/vfio_platform_private.h
+++ b/drivers/vfio/platform/vfio_platform_private.h
@@ -58,6 +58,7 @@ struct vfio_platform_device {
 	struct mutex			igate;
 	struct module			*parent_module;
 	const char			*compat;
+	const char			*acpihid;
 	struct module			*reset_module;
 	struct device			*device;
 
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 5/9] vfio: platform: add extra debug info argument to call reset
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
                   ` (3 preceding siblings ...)
  2016-07-18 23:09 ` [PATCH V10 4/9] vfio: platform: add support for ACPI probe Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 6/9] vfio: platform: call _RST method when using ACPI Sinan Kaya
                   ` (3 subsequent siblings)
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

Getting ready to bring out extra debug information to the caller
so that more verbose information can be printed when an error is
observed.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Reviewed-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
---
 drivers/vfio/platform/vfio_platform_common.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index a5299f6..f6025c9 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -166,7 +166,8 @@ static void vfio_platform_regions_cleanup(struct vfio_platform_device *vdev)
 	kfree(vdev->regions);
 }
 
-static int vfio_platform_call_reset(struct vfio_platform_device *vdev)
+static int vfio_platform_call_reset(struct vfio_platform_device *vdev,
+				    const char **extra_dbg)
 {
 	if (vdev->of_reset) {
 		dev_info(vdev->device, "reset\n");
@@ -184,7 +185,7 @@ static void vfio_platform_release(void *device_data)
 	mutex_lock(&driver_lock);
 
 	if (!(--vdev->refcnt)) {
-		vfio_platform_call_reset(vdev);
+		vfio_platform_call_reset(vdev, NULL);
 		vfio_platform_regions_cleanup(vdev);
 		vfio_platform_irq_cleanup(vdev);
 	}
@@ -213,7 +214,7 @@ static int vfio_platform_open(void *device_data)
 		if (ret)
 			goto err_irq;
 
-		vfio_platform_call_reset(vdev);
+		vfio_platform_call_reset(vdev, NULL);
 	}
 
 	vdev->refcnt++;
@@ -345,7 +346,7 @@ static long vfio_platform_ioctl(void *device_data,
 		return ret;
 
 	} else if (cmd == VFIO_DEVICE_RESET) {
-		return vfio_platform_call_reset(vdev);
+		return vfio_platform_call_reset(vdev, NULL);
 	}
 
 	return -ENOTTY;
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 6/9] vfio: platform: call _RST method when using ACPI
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
                   ` (4 preceding siblings ...)
  2016-07-18 23:09 ` [PATCH V10 5/9] vfio: platform: add extra debug info argument to call reset Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 7/9] vfio, platform: make reset driver a requirement by default Sinan Kaya
                   ` (2 subsequent siblings)
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

The device tree code checks for the presence of a reset driver and calls
the of_reset function pointer by looking up the reset driver as a module.

ACPI defines _RST method to perform device level reset. After the _RST
method is executed, the OS can resume using the device. _RST method is
expected to stop DMA transfers and IRQs.

This patch introduces two functions as vfio_platform_acpi_has_reset and
vfio_platform_acpi_call_reset. The has reset method is used to declare
reset capability via the ioctl flag VFIO_DEVICE_FLAGS_RESET. The call
reset function is used to execute the _RST ACPI method.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
---
 drivers/vfio/platform/vfio_platform_common.c | 50 +++++++++++++++++++++++++++-
 1 file changed, 49 insertions(+), 1 deletion(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index f6025c9..b1e5801 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -28,6 +28,8 @@
 #define DRIVER_AUTHOR   "Antonios Motakis <a.motakis@virtualopensystems.com>"
 #define DRIVER_DESC     "VFIO platform base module"
 
+#define VFIO_PLATFORM_IS_ACPI(vdev) ((vdev)->acpihid != NULL)
+
 static LIST_HEAD(reset_list);
 static DEFINE_MUTEX(driver_lock);
 
@@ -76,13 +78,53 @@ static int vfio_platform_acpi_probe(struct vfio_platform_device *vdev,
 	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;
 }
 
+int vfio_platform_acpi_call_reset(struct vfio_platform_device *vdev,
+				  const char **extra_dbg)
+{
+#ifdef CONFIG_ACPI
+	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
+	struct device *dev = vdev->device;
+	acpi_handle handle = ACPI_HANDLE(dev);
+	acpi_status acpi_ret;
+
+	acpi_ret = acpi_evaluate_object(handle, "_RST", NULL, &buffer);
+	if (ACPI_FAILURE(acpi_ret)) {
+		if (extra_dbg)
+			*extra_dbg = acpi_format_exception(acpi_ret);
+		return -EINVAL;
+	}
+
+	return 0;
+#else
+	return -ENOENT;
+#endif
+}
+
+bool vfio_platform_acpi_has_reset(struct vfio_platform_device *vdev)
+{
+#ifdef CONFIG_ACPI
+	struct device *dev = vdev->device;
+	acpi_handle handle = ACPI_HANDLE(dev);
+
+	return acpi_has_method(handle, "_RST");
+#else
+	return false;
+#endif
+}
+
 static bool vfio_platform_has_reset(struct vfio_platform_device *vdev)
 {
+	if (VFIO_PLATFORM_IS_ACPI(vdev))
+		return vfio_platform_acpi_has_reset(vdev);
+
 	return vdev->of_reset ? true : false;
 }
 
 static void vfio_platform_get_reset(struct vfio_platform_device *vdev)
 {
+	if (VFIO_PLATFORM_IS_ACPI(vdev))
+		return;
+
 	vdev->of_reset = vfio_platform_lookup_reset(vdev->compat,
 						    &vdev->reset_module);
 	if (!vdev->of_reset) {
@@ -94,6 +136,9 @@ static void vfio_platform_get_reset(struct vfio_platform_device *vdev)
 
 static void vfio_platform_put_reset(struct vfio_platform_device *vdev)
 {
+	if (VFIO_PLATFORM_IS_ACPI(vdev))
+		return;
+
 	if (vdev->of_reset)
 		module_put(vdev->reset_module);
 }
@@ -169,7 +214,10 @@ static void vfio_platform_regions_cleanup(struct vfio_platform_device *vdev)
 static int vfio_platform_call_reset(struct vfio_platform_device *vdev,
 				    const char **extra_dbg)
 {
-	if (vdev->of_reset) {
+	if (VFIO_PLATFORM_IS_ACPI(vdev)) {
+		dev_info(vdev->device, "reset\n");
+		return vfio_platform_acpi_call_reset(vdev, extra_dbg);
+	} else if (vdev->of_reset) {
 		dev_info(vdev->device, "reset\n");
 		return vdev->of_reset(vdev);
 	}
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 7/9] vfio, platform: make reset driver a requirement by default
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
                   ` (5 preceding siblings ...)
  2016-07-18 23:09 ` [PATCH V10 6/9] vfio: platform: call _RST method when using ACPI Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 8/9] vfio: platform: check reset call return code during open Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 9/9] vfio: platform: check reset call return code during release Sinan Kaya
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

The code was allowing platform devices to be used without a supporting
VFIO reset driver. The hardware can be left in some inconsistent state
after a guest machine abort.

The reset driver will put the hardware back to safe state and disable
interrupts before returning the control back to the host machine.

Adding a new reset_required kernel module option to platform VFIO drivers.
The default value is true for the DT and ACPI based drivers.
The reset requirement value for AMBA drivers is set to false and is
unchangeable to maintain the existing functionality.

New requirements are:
1. A reset function needs to be implemented by the corresponding driver
via DT/ACPI.
2. The reset function needs to be discovered via DT/ACPI.

The probe of the driver will fail if any of the above conditions are
not satisfied.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
---
 drivers/vfio/platform/vfio_amba.c             |  1 +
 drivers/vfio/platform/vfio_platform.c         |  5 +++++
 drivers/vfio/platform/vfio_platform_common.c  | 15 +++++++++++----
 drivers/vfio/platform/vfio_platform_private.h |  2 ++
 4 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/drivers/vfio/platform/vfio_amba.c b/drivers/vfio/platform/vfio_amba.c
index a66479b..31372fb 100644
--- a/drivers/vfio/platform/vfio_amba.c
+++ b/drivers/vfio/platform/vfio_amba.c
@@ -68,6 +68,7 @@ static int vfio_amba_probe(struct amba_device *adev, const struct amba_id *id)
 	vdev->get_resource = get_amba_resource;
 	vdev->get_irq = get_amba_irq;
 	vdev->parent_module = THIS_MODULE;
+	vdev->reset_required = false;
 
 	ret = vfio_platform_probe_common(vdev, &adev->dev);
 	if (ret) {
diff --git a/drivers/vfio/platform/vfio_platform.c b/drivers/vfio/platform/vfio_platform.c
index b1cc3a7..6561751 100644
--- a/drivers/vfio/platform/vfio_platform.c
+++ b/drivers/vfio/platform/vfio_platform.c
@@ -23,6 +23,10 @@
 #define DRIVER_AUTHOR   "Antonios Motakis <a.motakis@virtualopensystems.com>"
 #define DRIVER_DESC     "VFIO for platform devices - User Level meta-driver"
 
+static bool reset_required = true;
+module_param(reset_required, bool, 0444);
+MODULE_PARM_DESC(reset_required, "override reset requirement (default: 1)");
+
 /* probing devices from the linux platform bus */
 
 static struct resource *get_platform_resource(struct vfio_platform_device *vdev,
@@ -66,6 +70,7 @@ static int vfio_platform_probe(struct platform_device *pdev)
 	vdev->get_resource = get_platform_resource;
 	vdev->get_irq = get_platform_irq;
 	vdev->parent_module = THIS_MODULE;
+	vdev->reset_required = reset_required;
 
 	ret = vfio_platform_probe_common(vdev, &pdev->dev);
 	if (ret)
diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index b1e5801..16709f3 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -120,10 +120,10 @@ static bool vfio_platform_has_reset(struct vfio_platform_device *vdev)
 	return vdev->of_reset ? true : false;
 }
 
-static void vfio_platform_get_reset(struct vfio_platform_device *vdev)
+static int vfio_platform_get_reset(struct vfio_platform_device *vdev)
 {
 	if (VFIO_PLATFORM_IS_ACPI(vdev))
-		return;
+		return vfio_platform_acpi_has_reset(vdev) ? 0 : -ENOENT;
 
 	vdev->of_reset = vfio_platform_lookup_reset(vdev->compat,
 						    &vdev->reset_module);
@@ -132,6 +132,8 @@ static void vfio_platform_get_reset(struct vfio_platform_device *vdev)
 		vdev->of_reset = vfio_platform_lookup_reset(vdev->compat,
 							&vdev->reset_module);
 	}
+
+	return vdev->of_reset ? 0 : -ENOENT;
 }
 
 static void vfio_platform_put_reset(struct vfio_platform_device *vdev)
@@ -672,6 +674,13 @@ int vfio_platform_probe_common(struct vfio_platform_device *vdev,
 
 	vdev->device = dev;
 
+	ret = vfio_platform_get_reset(vdev);
+	if (ret && vdev->reset_required) {
+		pr_err("vfio: no reset function found for device %s\n",
+		       vdev->name);
+		return ret;
+	}
+
 	group = iommu_group_get(dev);
 	if (!group) {
 		pr_err("VFIO: No IOMMU group for device %s\n", vdev->name);
@@ -684,8 +693,6 @@ int vfio_platform_probe_common(struct vfio_platform_device *vdev,
 		return ret;
 	}
 
-	vfio_platform_get_reset(vdev);
-
 	mutex_init(&vdev->igate);
 
 	return 0;
diff --git a/drivers/vfio/platform/vfio_platform_private.h b/drivers/vfio/platform/vfio_platform_private.h
index ba9e4f8..85ffe5d 100644
--- a/drivers/vfio/platform/vfio_platform_private.h
+++ b/drivers/vfio/platform/vfio_platform_private.h
@@ -73,6 +73,8 @@ struct vfio_platform_device {
 		(*get_resource)(struct vfio_platform_device *vdev, int i);
 	int	(*get_irq)(struct vfio_platform_device *vdev, int i);
 	int	(*of_reset)(struct vfio_platform_device *vdev);
+
+	bool				reset_required;
 };
 
 typedef int (*vfio_platform_reset_fn_t)(struct vfio_platform_device *vdev);
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 8/9] vfio: platform: check reset call return code during open
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
                   ` (6 preceding siblings ...)
  2016-07-18 23:09 ` [PATCH V10 7/9] vfio, platform: make reset driver a requirement by default Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  2016-07-18 23:09 ` [PATCH V10 9/9] vfio: platform: check reset call return code during release Sinan Kaya
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

Open call is ignoring the return code from reset call and can
potentially continue even though reset call failed.

If reset_required module parameter is set, this patch is going
to validate the return code and will abort open if reset fails.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Reviewed-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
---
 drivers/vfio/platform/vfio_platform_common.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index 16709f3..cd785a9 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -256,6 +256,8 @@ static int vfio_platform_open(void *device_data)
 	mutex_lock(&driver_lock);
 
 	if (!vdev->refcnt) {
+		const char *extra_dbg = NULL;
+
 		ret = vfio_platform_regions_init(vdev);
 		if (ret)
 			goto err_reg;
@@ -264,7 +266,12 @@ static int vfio_platform_open(void *device_data)
 		if (ret)
 			goto err_irq;
 
-		vfio_platform_call_reset(vdev, NULL);
+		ret = vfio_platform_call_reset(vdev, &extra_dbg);
+		if (ret && vdev->reset_required) {
+			dev_warn(vdev->device, "reset driver is required and reset call failed in open (%d) %s\n",
+				 ret, extra_dbg ? extra_dbg : "");
+			goto err_rst;
+		}
 	}
 
 	vdev->refcnt++;
@@ -272,6 +279,8 @@ static int vfio_platform_open(void *device_data)
 	mutex_unlock(&driver_lock);
 	return 0;
 
+err_rst:
+	vfio_platform_irq_cleanup(vdev);
 err_irq:
 	vfio_platform_regions_cleanup(vdev);
 err_reg:
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 9/9] vfio: platform: check reset call return code during release
  2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
                   ` (7 preceding siblings ...)
  2016-07-18 23:09 ` [PATCH V10 8/9] vfio: platform: check reset call return code during open Sinan Kaya
@ 2016-07-18 23:09 ` Sinan Kaya
  8 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-18 23:09 UTC (permalink / raw)
  To: linux-arm-kernel

Release call is ignoring the return code from reset call and can
potentially continue even though reset call failed.

If reset_required module parameter is set, this patch is going
to validate the return code and will cause stack dump with
WARN_ON and warn the user of failure.

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Baptiste Reynal <b.reynal@virtualopensystems.com>
---
 drivers/vfio/platform/vfio_platform_common.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index cd785a9..cc32339 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -235,7 +235,15 @@ static void vfio_platform_release(void *device_data)
 	mutex_lock(&driver_lock);
 
 	if (!(--vdev->refcnt)) {
-		vfio_platform_call_reset(vdev, NULL);
+		const char *extra_dbg = NULL;
+		int ret;
+
+		ret = vfio_platform_call_reset(vdev, &extra_dbg);
+		if (ret && vdev->reset_required) {
+			dev_warn(vdev->device, "reset driver is required and reset call failed in release (%d) %s\n",
+				 ret, extra_dbg ? extra_dbg : "");
+			WARN_ON(1);
+		}
 		vfio_platform_regions_cleanup(vdev);
 		vfio_platform_irq_cleanup(vdev);
 	}
-- 
1.8.2.1

^ permalink raw reply related	[flat|nested] 16+ messages in thread

* [PATCH V10 4/9] vfio: platform: add support for ACPI probe
  2016-07-18 23:09 ` [PATCH V10 4/9] vfio: platform: add support for ACPI probe Sinan Kaya
@ 2016-07-19  0:00   ` Alex Williamson
  2016-07-19  0:16     ` okaya at codeaurora.org
  0 siblings, 1 reply; 16+ messages in thread
From: Alex Williamson @ 2016-07-19  0:00 UTC (permalink / raw)
  To: linux-arm-kernel

On Mon, 18 Jul 2016 19:09:22 -0400
Sinan Kaya <okaya@codeaurora.org> wrote:

> The code is using the compatible DT string to associate a reset driver
> with the actual device itself. The compatible string does not exist on
> ACPI based systems. HID is the unique identifier for a device driver
> instead.
> 
> Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
> ---
>  drivers/vfio/platform/vfio_platform_common.c  | 69 +++++++++++++++++++++++++--
>  drivers/vfio/platform/vfio_platform_private.h |  1 +
>  2 files changed, 65 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
> index 6be92c3..a5299f6 100644
> --- a/drivers/vfio/platform/vfio_platform_common.c
> +++ b/drivers/vfio/platform/vfio_platform_common.c
> @@ -13,6 +13,7 @@
>   */
>  
>  #include <linux/device.h>
> +#include <linux/acpi.h>
>  #include <linux/iommu.h>
>  #include <linux/module.h>
>  #include <linux/mutex.h>
> @@ -49,6 +50,32 @@ static vfio_platform_reset_fn_t vfio_platform_lookup_reset(const char *compat,
>  	return reset_fn;
>  }
>  
> +static int vfio_platform_acpi_probe(struct vfio_platform_device *vdev,
> +				    struct device *dev)
> +{
> +	struct acpi_device *adev;
> +
> +	if (acpi_disabled)
> +		return -EPERM;
> +
> +	adev = ACPI_COMPANION(dev);

I didn't necessarily have a problem with this being set in the
declaration.

> +	if (!adev) {
> +		pr_err("VFIO: ACPI companion device not found for %s\n",
> +			vdev->name);
> +		return -ENODEV;
> +	}
> +
> +#ifdef CONFIG_ACPI
> +	vdev->acpihid = acpi_device_hid(adev);
> +	if (!vdev->acpihid) {
> +		pr_err("VFIO: cannot find ACPI HID for %s\n",
> +		       vdev->name);
> +		return -EINVAL;
> +	}
> +#endif
> +	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;

?!?!  The point was that that entire if{} branch is unnecessary.  The
WARN_ON handles the (impossible) case of !vdev->acpihid.  We just need:

#ifdef CONFIG_ACPI
	vdev->acpihid = acpi_device_hid(adev);
#endif
	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;

nit, might make sense to replace EPERM with ENOENT and use EINVAL here.

> +}
> +
>  static bool vfio_platform_has_reset(struct vfio_platform_device *vdev)
>  {
>  	return vdev->of_reset ? true : false;
> @@ -547,6 +574,37 @@ static const struct vfio_device_ops vfio_platform_ops = {
>  	.mmap		= vfio_platform_mmap,
>  };
>  
> +int vfio_platform_of_probe(struct vfio_platform_device *vdev,
> +			   struct device *dev)
> +{
> +	int ret;
> +
> +	ret = device_property_read_string(dev, "compatible",
> +					  &vdev->compat);
> +	if (ret)
> +		pr_err("VFIO: cannot retrieve compat for %s\n",
> +			vdev->name);
> +
> +	return ret;
> +}
> +
> +/*
> + * There can be two kernel build combinations. One build where
> + * ACPI is not selected in Kconfig and another one with the ACPI Kconfig.
> + *
> + * In the first case, vfio_platform_acpi_probe will return since
> + * acpi_disabled is 1. DT user will not see any kind of messages from
> + * ACPI.
> + *
> + * In the second case, both DT and ACPI is compiled in but the system is
> + * booting with any of these combinations.
> + *
> + * If the firmware is DT type, then acpi_disabled is 1. The ACPI probe routine
> + * terminates immediately without any messages.
> + *
> + * If the firmware is ACPI type, then acpi_disabled is 0. All other checks are
> + * valid checks. We cannot claim that this system is DT.
> + */
>  int vfio_platform_probe_common(struct vfio_platform_device *vdev,
>  			       struct device *dev)
>  {
> @@ -556,11 +614,12 @@ int vfio_platform_probe_common(struct vfio_platform_device *vdev,
>  	if (!vdev)
>  		return -EINVAL;
>  
> -	ret = device_property_read_string(dev, "compatible", &vdev->compat);
> -	if (ret) {
> -		pr_err("VFIO: cannot retrieve compat for %s\n", vdev->name);
> -		return -EINVAL;
> -	}
> +	ret = vfio_platform_acpi_probe(vdev, dev);
> +	if (ret)
> +		ret = vfio_platform_of_probe(vdev, dev);
> +
> +	if (ret)
> +		return ret;
>  
>  	vdev->device = dev;
>  
> diff --git a/drivers/vfio/platform/vfio_platform_private.h b/drivers/vfio/platform/vfio_platform_private.h
> index 71ed7d1..ba9e4f8 100644
> --- a/drivers/vfio/platform/vfio_platform_private.h
> +++ b/drivers/vfio/platform/vfio_platform_private.h
> @@ -58,6 +58,7 @@ struct vfio_platform_device {
>  	struct mutex			igate;
>  	struct module			*parent_module;
>  	const char			*compat;
> +	const char			*acpihid;
>  	struct module			*reset_module;
>  	struct device			*device;
>  

^ permalink raw reply	[flat|nested] 16+ messages in thread

* [PATCH V10 4/9] vfio: platform: add support for ACPI probe
  2016-07-19  0:00   ` Alex Williamson
@ 2016-07-19  0:16     ` okaya at codeaurora.org
  2016-07-19  0:24       ` Alex Williamson
  0 siblings, 1 reply; 16+ messages in thread
From: okaya at codeaurora.org @ 2016-07-19  0:16 UTC (permalink / raw)
  To: linux-arm-kernel

On 2016-07-18 20:00, Alex Williamson wrote:
> On Mon, 18 Jul 2016 19:09:22 -0400
> Sinan Kaya <okaya@codeaurora.org> wrote:
> 
>> The code is using the compatible DT string to associate a reset driver
>> with the actual device itself. The compatible string does not exist on
>> ACPI based systems. HID is the unique identifier for a device driver
>> instead.
>> 
>> Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
>> ---
>>  drivers/vfio/platform/vfio_platform_common.c  | 69 
>> +++++++++++++++++++++++++--
>>  drivers/vfio/platform/vfio_platform_private.h |  1 +
>>  2 files changed, 65 insertions(+), 5 deletions(-)
>> 
>> diff --git a/drivers/vfio/platform/vfio_platform_common.c 
>> b/drivers/vfio/platform/vfio_platform_common.c
>> index 6be92c3..a5299f6 100644
>> --- a/drivers/vfio/platform/vfio_platform_common.c
>> +++ b/drivers/vfio/platform/vfio_platform_common.c
>> @@ -13,6 +13,7 @@
>>   */
>> 
>>  #include <linux/device.h>
>> +#include <linux/acpi.h>
>>  #include <linux/iommu.h>
>>  #include <linux/module.h>
>>  #include <linux/mutex.h>
>> @@ -49,6 +50,32 @@ static vfio_platform_reset_fn_t 
>> vfio_platform_lookup_reset(const char *compat,
>>  	return reset_fn;
>>  }
>> 
>> +static int vfio_platform_acpi_probe(struct vfio_platform_device 
>> *vdev,
>> +				    struct device *dev)
>> +{
>> +	struct acpi_device *adev;
>> +
>> +	if (acpi_disabled)
>> +		return -EPERM;
>> +
>> +	adev = ACPI_COMPANION(dev);
> 
> I didn't necessarily have a problem with this being set in the
> declaration.

I think this is better. If ACPI is disabled, it is dangerous to call an 
ACPI API.


> 
>> +	if (!adev) {
>> +		pr_err("VFIO: ACPI companion device not found for %s\n",
>> +			vdev->name);
>> +		return -ENODEV;
>> +	}
>> +
>> +#ifdef CONFIG_ACPI
>> +	vdev->acpihid = acpi_device_hid(adev);
>> +	if (!vdev->acpihid) {
>> +		pr_err("VFIO: cannot find ACPI HID for %s\n",
>> +		       vdev->name);
>> +		return -EINVAL;
>> +	}
>> +#endif
>> +	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;
> 
> ?!?!  The point was that that entire if{} branch is unnecessary.  The
> WARN_ON handles the (impossible) case of !vdev->acpihid.  We just need:
> 
> #ifdef CONFIG_ACPI
> 	vdev->acpihid = acpi_device_hid(adev);
> #endif
> 	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;
> 

OK, got it now. I thought you were trying to get rid of #else

> nit, might make sense to replace EPERM with ENOENT and use EINVAL here.
> 

Sure, will take carr of it.

Anything else I need to take care of?

>> +}
>> +
>>  static bool vfio_platform_has_reset(struct vfio_platform_device 
>> *vdev)
>>  {
>>  	return vdev->of_reset ? true : false;
>> @@ -547,6 +574,37 @@ static const struct vfio_device_ops 
>> vfio_platform_ops = {
>>  	.mmap		= vfio_platform_mmap,
>>  };
>> 
>> +int vfio_platform_of_probe(struct vfio_platform_device *vdev,
>> +			   struct device *dev)
>> +{
>> +	int ret;
>> +
>> +	ret = device_property_read_string(dev, "compatible",
>> +					  &vdev->compat);
>> +	if (ret)
>> +		pr_err("VFIO: cannot retrieve compat for %s\n",
>> +			vdev->name);
>> +
>> +	return ret;
>> +}
>> +
>> +/*
>> + * There can be two kernel build combinations. One build where
>> + * ACPI is not selected in Kconfig and another one with the ACPI 
>> Kconfig.
>> + *
>> + * In the first case, vfio_platform_acpi_probe will return since
>> + * acpi_disabled is 1. DT user will not see any kind of messages from
>> + * ACPI.
>> + *
>> + * In the second case, both DT and ACPI is compiled in but the system 
>> is
>> + * booting with any of these combinations.
>> + *
>> + * If the firmware is DT type, then acpi_disabled is 1. The ACPI 
>> probe routine
>> + * terminates immediately without any messages.
>> + *
>> + * If the firmware is ACPI type, then acpi_disabled is 0. All other 
>> checks are
>> + * valid checks. We cannot claim that this system is DT.
>> + */
>>  int vfio_platform_probe_common(struct vfio_platform_device *vdev,
>>  			       struct device *dev)
>>  {
>> @@ -556,11 +614,12 @@ int vfio_platform_probe_common(struct 
>> vfio_platform_device *vdev,
>>  	if (!vdev)
>>  		return -EINVAL;
>> 
>> -	ret = device_property_read_string(dev, "compatible", &vdev->compat);
>> -	if (ret) {
>> -		pr_err("VFIO: cannot retrieve compat for %s\n", vdev->name);
>> -		return -EINVAL;
>> -	}
>> +	ret = vfio_platform_acpi_probe(vdev, dev);
>> +	if (ret)
>> +		ret = vfio_platform_of_probe(vdev, dev);
>> +
>> +	if (ret)
>> +		return ret;
>> 
>>  	vdev->device = dev;
>> 
>> diff --git a/drivers/vfio/platform/vfio_platform_private.h 
>> b/drivers/vfio/platform/vfio_platform_private.h
>> index 71ed7d1..ba9e4f8 100644
>> --- a/drivers/vfio/platform/vfio_platform_private.h
>> +++ b/drivers/vfio/platform/vfio_platform_private.h
>> @@ -58,6 +58,7 @@ struct vfio_platform_device {
>>  	struct mutex			igate;
>>  	struct module			*parent_module;
>>  	const char			*compat;
>> +	const char			*acpihid;
>>  	struct module			*reset_module;
>>  	struct device			*device;
>> 

^ permalink raw reply	[flat|nested] 16+ messages in thread

* [PATCH V10 4/9] vfio: platform: add support for ACPI probe
  2016-07-19  0:16     ` okaya at codeaurora.org
@ 2016-07-19  0:24       ` Alex Williamson
  2016-07-19 14:49         ` Sinan Kaya
  0 siblings, 1 reply; 16+ messages in thread
From: Alex Williamson @ 2016-07-19  0:24 UTC (permalink / raw)
  To: linux-arm-kernel

On Mon, 18 Jul 2016 20:16:50 -0400
okaya at codeaurora.org wrote:

> On 2016-07-18 20:00, Alex Williamson wrote:
> > On Mon, 18 Jul 2016 19:09:22 -0400
> > Sinan Kaya <okaya@codeaurora.org> wrote:
> >   
> >> The code is using the compatible DT string to associate a reset driver
> >> with the actual device itself. The compatible string does not exist on
> >> ACPI based systems. HID is the unique identifier for a device driver
> >> instead.
> >> 
> >> Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
> >> ---
> >>  drivers/vfio/platform/vfio_platform_common.c  | 69 
> >> +++++++++++++++++++++++++--
> >>  drivers/vfio/platform/vfio_platform_private.h |  1 +
> >>  2 files changed, 65 insertions(+), 5 deletions(-)
> >> 
> >> diff --git a/drivers/vfio/platform/vfio_platform_common.c 
> >> b/drivers/vfio/platform/vfio_platform_common.c
> >> index 6be92c3..a5299f6 100644
> >> --- a/drivers/vfio/platform/vfio_platform_common.c
> >> +++ b/drivers/vfio/platform/vfio_platform_common.c
> >> @@ -13,6 +13,7 @@
> >>   */
> >> 
> >>  #include <linux/device.h>
> >> +#include <linux/acpi.h>
> >>  #include <linux/iommu.h>
> >>  #include <linux/module.h>
> >>  #include <linux/mutex.h>
> >> @@ -49,6 +50,32 @@ static vfio_platform_reset_fn_t 
> >> vfio_platform_lookup_reset(const char *compat,
> >>  	return reset_fn;
> >>  }
> >> 
> >> +static int vfio_platform_acpi_probe(struct vfio_platform_device 
> >> *vdev,
> >> +				    struct device *dev)
> >> +{
> >> +	struct acpi_device *adev;
> >> +
> >> +	if (acpi_disabled)
> >> +		return -EPERM;
> >> +
> >> +	adev = ACPI_COMPANION(dev);  
> > 
> > I didn't necessarily have a problem with this being set in the
> > declaration.  
> 
> I think this is better. If ACPI is disabled, it is dangerous to call an 
> ACPI API.

Ok, fair enough.

> >   
> >> +	if (!adev) {
> >> +		pr_err("VFIO: ACPI companion device not found for %s\n",
> >> +			vdev->name);
> >> +		return -ENODEV;
> >> +	}
> >> +
> >> +#ifdef CONFIG_ACPI
> >> +	vdev->acpihid = acpi_device_hid(adev);
> >> +	if (!vdev->acpihid) {
> >> +		pr_err("VFIO: cannot find ACPI HID for %s\n",
> >> +		       vdev->name);
> >> +		return -EINVAL;
> >> +	}
> >> +#endif
> >> +	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;  
> > 
> > ?!?!  The point was that that entire if{} branch is unnecessary.  The
> > WARN_ON handles the (impossible) case of !vdev->acpihid.  We just need:
> > 
> > #ifdef CONFIG_ACPI
> > 	vdev->acpihid = acpi_device_hid(adev);
> > #endif
> > 	return WARN_ON(!vdev->acpihid) ? -ENOENT : 0;
> >   
> 
> OK, got it now. I thought you were trying to get rid of #else
> 
> > nit, might make sense to replace EPERM with ENOENT and use EINVAL here.
> >   
> 
> Sure, will take carr of it.
> 
> Anything else I need to take care of?

Not that I see, maybe just send a new version of this patch if the
changes don't trickle through too much.  Thanks,

Alex

 
> >> +}
> >> +
> >>  static bool vfio_platform_has_reset(struct vfio_platform_device 
> >> *vdev)
> >>  {
> >>  	return vdev->of_reset ? true : false;
> >> @@ -547,6 +574,37 @@ static const struct vfio_device_ops 
> >> vfio_platform_ops = {
> >>  	.mmap		= vfio_platform_mmap,
> >>  };
> >> 
> >> +int vfio_platform_of_probe(struct vfio_platform_device *vdev,
> >> +			   struct device *dev)
> >> +{
> >> +	int ret;
> >> +
> >> +	ret = device_property_read_string(dev, "compatible",
> >> +					  &vdev->compat);
> >> +	if (ret)
> >> +		pr_err("VFIO: cannot retrieve compat for %s\n",
> >> +			vdev->name);
> >> +
> >> +	return ret;
> >> +}
> >> +
> >> +/*
> >> + * There can be two kernel build combinations. One build where
> >> + * ACPI is not selected in Kconfig and another one with the ACPI 
> >> Kconfig.
> >> + *
> >> + * In the first case, vfio_platform_acpi_probe will return since
> >> + * acpi_disabled is 1. DT user will not see any kind of messages from
> >> + * ACPI.
> >> + *
> >> + * In the second case, both DT and ACPI is compiled in but the system 
> >> is
> >> + * booting with any of these combinations.
> >> + *
> >> + * If the firmware is DT type, then acpi_disabled is 1. The ACPI 
> >> probe routine
> >> + * terminates immediately without any messages.
> >> + *
> >> + * If the firmware is ACPI type, then acpi_disabled is 0. All other 
> >> checks are
> >> + * valid checks. We cannot claim that this system is DT.
> >> + */
> >>  int vfio_platform_probe_common(struct vfio_platform_device *vdev,
> >>  			       struct device *dev)
> >>  {
> >> @@ -556,11 +614,12 @@ int vfio_platform_probe_common(struct 
> >> vfio_platform_device *vdev,
> >>  	if (!vdev)
> >>  		return -EINVAL;
> >> 
> >> -	ret = device_property_read_string(dev, "compatible", &vdev->compat);
> >> -	if (ret) {
> >> -		pr_err("VFIO: cannot retrieve compat for %s\n", vdev->name);
> >> -		return -EINVAL;
> >> -	}
> >> +	ret = vfio_platform_acpi_probe(vdev, dev);
> >> +	if (ret)
> >> +		ret = vfio_platform_of_probe(vdev, dev);
> >> +
> >> +	if (ret)
> >> +		return ret;
> >> 
> >>  	vdev->device = dev;
> >> 
> >> diff --git a/drivers/vfio/platform/vfio_platform_private.h 
> >> b/drivers/vfio/platform/vfio_platform_private.h
> >> index 71ed7d1..ba9e4f8 100644
> >> --- a/drivers/vfio/platform/vfio_platform_private.h
> >> +++ b/drivers/vfio/platform/vfio_platform_private.h
> >> @@ -58,6 +58,7 @@ struct vfio_platform_device {
> >>  	struct mutex			igate;
> >>  	struct module			*parent_module;
> >>  	const char			*compat;
> >> +	const char			*acpihid;
> >>  	struct module			*reset_module;
> >>  	struct device			*device;
> >>   

^ permalink raw reply	[flat|nested] 16+ messages in thread

* [PATCH V10 4/9] vfio: platform: add support for ACPI probe
  2016-07-19  0:24       ` Alex Williamson
@ 2016-07-19 14:49         ` Sinan Kaya
  2016-07-19 15:28           ` Alex Williamson
  0 siblings, 1 reply; 16+ messages in thread
From: Sinan Kaya @ 2016-07-19 14:49 UTC (permalink / raw)
  To: linux-arm-kernel

On 7/18/2016 8:24 PM, Alex Williamson wrote:
>>> nit, might make sense to replace EPERM with ENOENT and use EINVAL here.
>>> > >   
>> > 
>> > Sure, will take carr of it.
>> > 
>> > Anything else I need to take care of?
> Not that I see, maybe just send a new version of this patch if the
> changes don't trickle through too much.  Thanks,

I posted V11 with the discussed change. 

[PATCH V11 0/9] vfio, platform: add ACPI support

Do you see this making to 4.8 kernel?

-- 
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.

^ permalink raw reply	[flat|nested] 16+ messages in thread

* [PATCH V10 4/9] vfio: platform: add support for ACPI probe
  2016-07-19 14:49         ` Sinan Kaya
@ 2016-07-19 15:28           ` Alex Williamson
  2016-07-19 15:29             ` Sinan Kaya
  0 siblings, 1 reply; 16+ messages in thread
From: Alex Williamson @ 2016-07-19 15:28 UTC (permalink / raw)
  To: linux-arm-kernel

On Tue, 19 Jul 2016 10:49:54 -0400
Sinan Kaya <okaya@codeaurora.org> wrote:

> On 7/18/2016 8:24 PM, Alex Williamson wrote:
> >>> nit, might make sense to replace EPERM with ENOENT and use EINVAL here.  
> >>> > >     
> >> > 
> >> > Sure, will take carr of it.
> >> > 
> >> > Anything else I need to take care of?  
> > Not that I see, maybe just send a new version of this patch if the
> > changes don't trickle through too much.  Thanks,  
> 
> I posted V11 with the discussed change. 
> 
> [PATCH V11 0/9] vfio, platform: add ACPI support
> 
> Do you see this making to 4.8 kernel?


I hope so, I'll try to get it into my next branch within the next day
or so.  Thanks,

Alex

^ permalink raw reply	[flat|nested] 16+ messages in thread

* [PATCH V10 4/9] vfio: platform: add support for ACPI probe
  2016-07-19 15:28           ` Alex Williamson
@ 2016-07-19 15:29             ` Sinan Kaya
  0 siblings, 0 replies; 16+ messages in thread
From: Sinan Kaya @ 2016-07-19 15:29 UTC (permalink / raw)
  To: linux-arm-kernel

On 7/19/2016 11:28 AM, Alex Williamson wrote:
> On Tue, 19 Jul 2016 10:49:54 -0400
> Sinan Kaya <okaya@codeaurora.org> wrote:
> 
>> On 7/18/2016 8:24 PM, Alex Williamson wrote:
>>>>> nit, might make sense to replace EPERM with ENOENT and use EINVAL here.  
>>>>>>>     
>>>>>
>>>>> Sure, will take carr of it.
>>>>>
>>>>> Anything else I need to take care of?  
>>> Not that I see, maybe just send a new version of this patch if the
>>> changes don't trickle through too much.  Thanks,  
>>
>> I posted V11 with the discussed change. 
>>
>> [PATCH V11 0/9] vfio, platform: add ACPI support
>>
>> Do you see this making to 4.8 kernel?
> 
> 
> I hope so, I'll try to get it into my next branch within the next day
> or so.  Thanks,
> 
> Alex
> 

Thank you.

-- 
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.

^ permalink raw reply	[flat|nested] 16+ messages in thread

end of thread, other threads:[~2016-07-19 15:29 UTC | newest]

Thread overview: 16+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-07-18 23:09 [PATCH V10 0/9] vfio, platform: add ACPI support Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 1/9] vfio: platform: rename reset function Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 2/9] vfio: platform: move reset call to a common function Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 3/9] vfio: platform: determine reset capability Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 4/9] vfio: platform: add support for ACPI probe Sinan Kaya
2016-07-19  0:00   ` Alex Williamson
2016-07-19  0:16     ` okaya at codeaurora.org
2016-07-19  0:24       ` Alex Williamson
2016-07-19 14:49         ` Sinan Kaya
2016-07-19 15:28           ` Alex Williamson
2016-07-19 15:29             ` Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 5/9] vfio: platform: add extra debug info argument to call reset Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 6/9] vfio: platform: call _RST method when using ACPI Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 7/9] vfio, platform: make reset driver a requirement by default Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 8/9] vfio: platform: check reset call return code during open Sinan Kaya
2016-07-18 23:09 ` [PATCH V10 9/9] vfio: platform: check reset call return code during release Sinan Kaya

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).