From: Lyude Paul <lyude@redhat.com>
To: amd-gfx@lists.freedesktop.org
Cc: "David Airlie" <airlied@linux.ie>,
linux-kernel@vger.kernel.org, stable@vger.kernel.org,
dri-devel@lists.freedesktop.org,
"Alex Deucher" <alexander.deucher@amd.com>,
"Rex Zhu" <Rex.Zhu@amd.com>,
"Christian König" <christian.koenig@amd.com>,
"Jean Delvare" <jdelvare@suse.de>
Subject: [PATCH v2 4/4] drm/amdgpu: Dynamically probe for ATIF handle
Date: Thu, 28 Jun 2018 15:14:55 -0400 [thread overview]
Message-ID: <20180628191505.10015-5-lyude@redhat.com> (raw)
In-Reply-To: <20180628191505.10015-1-lyude@redhat.com>
The other day I was testing one of the HP laptops at my office with an
i915/amdgpu hybrid setup and noticed that hotplugging was non-functional
on almost all of the display outputs. I eventually discovered that all
of the external outputs were connected to the amdgpu device instead of
i915, and that the hotplugs weren't being detected so long as the GPU
was in runtime suspend. After some talking with folks at AMD, I learned
that amdgpu is actually supposed to support hotplug detection in runtime
suspend so long as the OEM has implemented it properly in the firmware.
On this HP ZBook 15 G4 (the machine in question), amdgpu wasn't managing
to find the ATIF handle at all despite the fact that I could see acpi
events being sent in response to any hotplugging. After going through
dumps of the firmware, I discovered that this machine did in fact
support ATIF, but that it's ATIF method lived in an entirely different
namespace than this device's handle (the device handle was
\_SB_.PCI0.PEG0.PEGP, but ATIF lives in ATPX's handle at
\_SB_.PCI0.GFX0).
According to AMD, the reason for this appears to be that on laptops with
PRIME configurations, the ATIF and other related ACPI handles usually
live under the namespace of the intel GPU instead of the AMD GPU.
Machines that don't have more then a single GPU have the ATIF handle
located in the namespace of the AMD GPU.
So, fix this by probing ATPX's ACPI parent's namespace if we can't find
ATIF elsewhere, along with storing a pointer to the proper handle to use
for ATIF and using that instead of the device's handle.
This fixes HPD detection while in runtime suspend for this ZBook!
Signed-off-by: Lyude Paul <lyude@redhat.com>
Cc: stable@vger.kernel.org
---
drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 80 +++++++++++++++++-------
1 file changed, 59 insertions(+), 21 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 22c7e8ec0b9a..b0d7978c475d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -65,6 +65,8 @@ struct amdgpu_atif_functions {
};
struct amdgpu_atif {
+ acpi_handle handle;
+
struct amdgpu_atif_notifications notifications;
struct amdgpu_atif_functions functions;
struct amdgpu_atif_notification_cfg notification_cfg;
@@ -83,8 +85,9 @@ struct amdgpu_atif {
* Executes the requested ATIF function (all asics).
* Returns a pointer to the acpi output buffer.
*/
-static union acpi_object *amdgpu_atif_call(acpi_handle handle, int function,
- struct acpi_buffer *params)
+static union acpi_object *amdgpu_atif_call(struct amdgpu_atif *atif,
+ int function,
+ struct acpi_buffer *params)
{
acpi_status status;
union acpi_object atif_arg_elements[2];
@@ -107,7 +110,8 @@ static union acpi_object *amdgpu_atif_call(acpi_handle handle, int function,
atif_arg_elements[1].integer.value = 0;
}
- status = acpi_evaluate_object(handle, "ATIF", &atif_arg, &buffer);
+ status = acpi_evaluate_object(atif->handle, NULL, &atif_arg,
+ &buffer);
/* Fail only if calling the method fails and ATIF is supported */
if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
@@ -178,15 +182,14 @@ static void amdgpu_atif_parse_functions(struct amdgpu_atif_functions *f, u32 mas
* (all asics).
* returns 0 on success, error on failure.
*/
-static int amdgpu_atif_verify_interface(acpi_handle handle,
- struct amdgpu_atif *atif)
+static int amdgpu_atif_verify_interface(struct amdgpu_atif *atif)
{
union acpi_object *info;
struct atif_verify_interface output;
size_t size;
int err = 0;
- info = amdgpu_atif_call(handle, ATIF_FUNCTION_VERIFY_INTERFACE, NULL);
+ info = amdgpu_atif_call(atif, ATIF_FUNCTION_VERIFY_INTERFACE, NULL);
if (!info)
return -EIO;
@@ -213,6 +216,36 @@ static int amdgpu_atif_verify_interface(acpi_handle handle,
return err;
}
+static acpi_handle amdgpu_atif_probe_handle(acpi_handle dhandle)
+{
+ acpi_handle handle = NULL;
+ char acpi_method_name[255] = { 0 };
+ struct acpi_buffer buffer = { sizeof(acpi_method_name), acpi_method_name };
+ acpi_status status;
+
+ /* First check the device handle */
+ status = acpi_get_handle(dhandle, "ATIF", &handle);
+ if (ACPI_SUCCESS(status))
+ goto out;
+
+ /* Some vendors stick the ATIF methods in the same namespace as the
+ * ATPX methods
+ */
+ if (amdgpu_has_atpx()) {
+ status = acpi_get_handle(amdgpu_atpx_get_dhandle(), "ATIF",
+ &handle);
+ if (ACPI_SUCCESS(status))
+ goto out;
+ }
+
+ DRM_DEBUG_DRIVER("No ATIF handle found\n");
+ return NULL;
+out:
+ acpi_get_name(handle, ACPI_FULL_PATHNAME, &buffer);
+ DRM_DEBUG_DRIVER("Found ATIF handle %s\n", acpi_method_name);
+ return handle;
+}
+
/**
* amdgpu_atif_get_notification_params - determine notify configuration
*
@@ -225,15 +258,16 @@ static int amdgpu_atif_verify_interface(acpi_handle handle,
* where n is specified in the result if a notifier is used.
* Returns 0 on success, error on failure.
*/
-static int amdgpu_atif_get_notification_params(acpi_handle handle,
- struct amdgpu_atif_notification_cfg *n)
+static int amdgpu_atif_get_notification_params(struct amdgpu_atif *atif)
{
union acpi_object *info;
+ struct amdgpu_atif_notification_cfg *n = &atif->notification_cfg;
struct atif_system_params params;
size_t size;
int err = 0;
- info = amdgpu_atif_call(handle, ATIF_FUNCTION_GET_SYSTEM_PARAMETERS, NULL);
+ info = amdgpu_atif_call(atif, ATIF_FUNCTION_GET_SYSTEM_PARAMETERS,
+ NULL);
if (!info) {
err = -EIO;
goto out;
@@ -287,14 +321,15 @@ static int amdgpu_atif_get_notification_params(acpi_handle handle,
* (all asics).
* Returns 0 on success, error on failure.
*/
-static int amdgpu_atif_get_sbios_requests(acpi_handle handle,
- struct atif_sbios_requests *req)
+static int amdgpu_atif_get_sbios_requests(struct amdgpu_atif *atif,
+ struct atif_sbios_requests *req)
{
union acpi_object *info;
size_t size;
int count = 0;
- info = amdgpu_atif_call(handle, ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS, NULL);
+ info = amdgpu_atif_call(atif, ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS,
+ NULL);
if (!info)
return -EIO;
@@ -327,11 +362,10 @@ static int amdgpu_atif_get_sbios_requests(acpi_handle handle,
* Returns NOTIFY code
*/
static int amdgpu_atif_handler(struct amdgpu_device *adev,
- struct acpi_bus_event *event)
+ struct acpi_bus_event *event)
{
struct amdgpu_atif *atif = adev->atif;
struct atif_sbios_requests req;
- acpi_handle handle;
int count;
DRM_DEBUG_DRIVER("event, device_class = %s, type = %#x\n",
@@ -347,8 +381,7 @@ static int amdgpu_atif_handler(struct amdgpu_device *adev,
return NOTIFY_DONE;
/* Check pending SBIOS requests */
- handle = ACPI_HANDLE(&adev->pdev->dev);
- count = amdgpu_atif_get_sbios_requests(handle, &req);
+ count = amdgpu_atif_get_sbios_requests(atif, &req);
if (count <= 0)
return NOTIFY_DONE;
@@ -679,7 +712,7 @@ static int amdgpu_acpi_event(struct notifier_block *nb,
*/
int amdgpu_acpi_init(struct amdgpu_device *adev)
{
- acpi_handle handle;
+ acpi_handle handle, atif_handle;
struct amdgpu_atif *atif;
struct amdgpu_atcs *atcs = &adev->atcs;
int ret;
@@ -696,14 +729,20 @@ int amdgpu_acpi_init(struct amdgpu_device *adev)
DRM_DEBUG_DRIVER("Call to ATCS verify_interface failed: %d\n", ret);
}
- /* Call the ATIF method */
+ /* Probe for ATIF, and initialize it if found */
+ atif_handle = amdgpu_atif_probe_handle(handle);
+ if (!atif_handle)
+ goto out;
+
atif = kzalloc(sizeof(*atif), GFP_KERNEL);
if (!atif) {
DRM_WARN("Not enough memory to initialize ATIF\n");
goto out;
}
+ atif->handle = atif_handle;
- ret = amdgpu_atif_verify_interface(handle, atif);
+ /* Call the ATIF method */
+ ret = amdgpu_atif_verify_interface(atif);
if (ret) {
DRM_DEBUG_DRIVER("Call to ATIF verify_interface failed: %d\n", ret);
kfree(atif);
@@ -739,8 +778,7 @@ int amdgpu_acpi_init(struct amdgpu_device *adev)
}
if (atif->functions.system_params) {
- ret = amdgpu_atif_get_notification_params(handle,
- &atif->notification_cfg);
+ ret = amdgpu_atif_get_notification_params(atif);
if (ret) {
DRM_DEBUG_DRIVER("Call to GET_SYSTEM_PARAMS failed: %d\n",
ret);
--
2.17.1
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
prev parent reply other threads:[~2018-06-28 19:14 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-28 19:14 [PATCH v2 0/4] Fix HPD detection in runtime suspend Lyude Paul
[not found] ` <20180628191505.10015-1-lyude-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-06-28 19:14 ` [PATCH v2 1/4] drm/amdgpu: Make struct amdgpu_atif private to amdgpu_acpi.c Lyude Paul
2018-06-28 19:14 ` [PATCH v2 2/4] drm/amdgpu: s/disp_detetion_ports/disp_detection_ports/ Lyude Paul
2018-06-28 19:14 ` [PATCH v2 3/4] drm/amdgpu: Add amdgpu_atpx_get_dhandle() Lyude Paul
2018-06-28 19:14 ` Lyude Paul [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180628191505.10015-5-lyude@redhat.com \
--to=lyude@redhat.com \
--cc=Rex.Zhu@amd.com \
--cc=airlied@linux.ie \
--cc=alexander.deucher@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=jdelvare@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox