Hi,
This series introduce Kunit tests to the vc4 KMS driver, but unlike what we have been doing so far in KMS, it actually tests the atomic modesetting code.
In order to do so, I've had to improve a fair bit on the Kunit helpers already found in the tree in order to register a full blown and somewhat functional KMS driver.
It's of course relying on a mock so that we can test it anywhere. The mocking approach created a number of issues, the main one being that we need to create a decent mock in the first place, see patch 22. The basic idea is that I created some structures to provide a decent approximation of the actual hardware, and that would support both major architectures supported by vc4.
This is of course meant to evolve over time and support more tests, but I've focused on testing the HVS FIFO assignment code which is fairly tricky (and the tests have actually revealed one more bug with our current implementation). I used to have a userspace implementation of those tests, where I would copy and paste the kernel code and run the tests on a regular basis. It's was obviously fairly suboptimal, so it seemed like the perfect testbed for that series.
Let me know what you think, Maxime
To: David Airlie airlied@gmail.com To: Daniel Vetter daniel@ffwll.ch To: Maarten Lankhorst maarten.lankhorst@linux.intel.com To: Maxime Ripard mripard@kernel.org To: Thomas Zimmermann tzimmermann@suse.de Cc: Dave Stevenson dave.stevenson@raspberrypi.com Cc: Javier Martinez Canillas javierm@redhat.com Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org Cc: Maíra Canal mairacanal@riseup.net Cc: Brendan Higgins brendan.higgins@linux.dev Cc: David Gow davidgow@google.com Cc: linux-kselftest@vger.kernel.org Cc: kunit-dev@googlegroups.com Cc: dri-devel@lists.freedesktop.org Cc: linux-kernel@vger.kernel.org Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org Signed-off-by: Maxime Ripard maxime@cerno.tech
--- Maxime Ripard (24): drm/tests: helpers: Rename the device init helper drm/tests: helpers: Remove the name parameter drm/tests: helpers: Create the device in another function drm/tests: helpers: Switch to a platform_device drm/tests: helpers: Make sure the device is bound drm/tests: kunit: Allow for a custom device struct to be allocated drm/tests: helpers: Allow to pass a custom drm_driver drm/tests: Add a test for DRM managed actions drm/atomic: Constify the old/new state accessors drm/vc4: kms: Sort the CRTCs by output before assigning them drm/vc4: Constify container_of wrappers drm/vc4: Move HVS state to main header drm/vc4: kms: Constify the HVS old/new state helpers drm/vc4: txp: Reorder the variable assignments drm/vc4: Add TXP encoder type drm/vc4: txp: Initialise the CRTC before the encoder and connector drm/vc4: crtc: Pass the device and data in vc4_crtc_init drm/vc4: crtc: Introduce a lower-level crtc init helper drm/vc4: crtc: Make encoder lookup helper public drm/vc4: crtc: Provide a CRTC name drm/vc4: hvs: Provide a function to initialize the HVS structure drm/vc4: tests: Introduce a mocking infrastructure drm/vc4: tests: Fail the current test if we access a register drm/vc4: tests: Add unit test suite for the PV muxing
drivers/gpu/drm/drm_atomic.c | 12 +- drivers/gpu/drm/tests/Makefile | 1 + drivers/gpu/drm/tests/drm_client_modeset_test.c | 16 +- drivers/gpu/drm/tests/drm_kunit_helpers.c | 116 +++-- drivers/gpu/drm/tests/drm_kunit_helpers.h | 39 +- drivers/gpu/drm/tests/drm_managed_test.c | 68 +++ drivers/gpu/drm/vc4/Kconfig | 15 + drivers/gpu/drm/vc4/Makefile | 1 + drivers/gpu/drm/vc4/tests/.kunitconfig | 14 + drivers/gpu/drm/vc4/tests/Makefile | 8 + drivers/gpu/drm/vc4/tests/vc4_mock.c | 174 +++++++ drivers/gpu/drm/vc4/tests/vc4_mock.h | 58 +++ drivers/gpu/drm/vc4/tests/vc4_mock_crtc.c | 39 ++ drivers/gpu/drm/vc4/tests/vc4_mock_output.c | 97 ++++ drivers/gpu/drm/vc4/tests/vc4_mock_plane.c | 45 ++ drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c | 624 ++++++++++++++++++++++++ drivers/gpu/drm/vc4/vc4_crtc.c | 119 +++-- drivers/gpu/drm/vc4/vc4_dpi.c | 13 +- drivers/gpu/drm/vc4/vc4_drv.c | 4 +- drivers/gpu/drm/vc4/vc4_drv.h | 113 ++++- drivers/gpu/drm/vc4/vc4_dsi.c | 9 +- drivers/gpu/drm/vc4/vc4_hdmi_regs.h | 4 + drivers/gpu/drm/vc4/vc4_hvs.c | 81 +-- drivers/gpu/drm/vc4/vc4_kms.c | 138 +++--- drivers/gpu/drm/vc4/vc4_txp.c | 66 ++- drivers/gpu/drm/vc4/vc4_vec.c | 13 +- include/drm/drm_atomic.h | 32 +- 27 files changed, 1678 insertions(+), 241 deletions(-) --- base-commit: 35c3a2d02f0dc153a5f2f304ba33e1436b6a8d8f change-id: 20221123-rpi-kunit-tests-87a388492a73
Best regards,
The name doesn't really fit the conventions for the other helpers in DRM/KMS, so let's rename it to make it obvious that we allocate a new DRM device.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/drm_client_modeset_test.c | 2 +- drivers/gpu/drm/tests/drm_kunit_helpers.c | 4 +++- drivers/gpu/drm/tests/drm_kunit_helpers.h | 5 ++++- 3 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_client_modeset_test.c b/drivers/gpu/drm/tests/drm_client_modeset_test.c index 362a5fbd82f5..e469d1634e2d 100644 --- a/drivers/gpu/drm/tests/drm_client_modeset_test.c +++ b/drivers/gpu/drm/tests/drm_client_modeset_test.c @@ -41,7 +41,7 @@ static int drm_client_modeset_test_init(struct kunit *test)
test->priv = priv;
- priv->drm = drm_kunit_device_init(test, DRIVER_MODESET, "drm-client-modeset-test"); + priv->drm = drm_kunit_helper_alloc_drm_device(test, DRIVER_MODESET, "drm-client-modeset-test"); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->drm);
ret = drmm_connector_init(priv->drm, &priv->connector, diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index f1662091f250..a4ad030ed101 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -36,7 +36,9 @@ static void dev_free(struct kunit_resource *res) root_device_unregister(dev); }
-struct drm_device *drm_kunit_device_init(struct kunit *test, u32 features, char *name) +struct drm_device * +drm_kunit_helper_alloc_drm_device(struct kunit *test, + u32 features, char *name) { struct kunit_dev *kdev; struct drm_device *drm; diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.h b/drivers/gpu/drm/tests/drm_kunit_helpers.h index 20ab6eec4c89..e9870c7911fe 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.h +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.h @@ -6,6 +6,9 @@ struct drm_device; struct kunit;
-struct drm_device *drm_kunit_device_init(struct kunit *test, u32 features, char *name); +struct drm_device * +drm_kunit_helper_alloc_drm_device(struct kunit *test, + u32 features, + char *name);
#endif // DRM_KUNIT_HELPERS_H_
On 11/23/22 16:25, Maxime Ripard wrote:
The name doesn't really fit the conventions for the other helpers in DRM/KMS, so let's rename it to make it obvious that we allocate a new DRM device.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On 11/23/22 12:25, Maxime Ripard wrote:
The name doesn't really fit the conventions for the other helpers in DRM/KMS, so let's rename it to make it obvious that we allocate a new DRM device.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Although I believe using "drm_device" on the function name is a bit redundant (maybe drm_kunit_helper_alloc_dev or drm_kunit_helper_alloc_device would be cleaner),
Reviewed-by: Maíra Canal mairacanal@riseup.net
Best Regards, - Maíra Canal
drivers/gpu/drm/tests/drm_client_modeset_test.c | 2 +- drivers/gpu/drm/tests/drm_kunit_helpers.c | 4 +++- drivers/gpu/drm/tests/drm_kunit_helpers.h | 5 ++++- 3 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_client_modeset_test.c b/drivers/gpu/drm/tests/drm_client_modeset_test.c index 362a5fbd82f5..e469d1634e2d 100644 --- a/drivers/gpu/drm/tests/drm_client_modeset_test.c +++ b/drivers/gpu/drm/tests/drm_client_modeset_test.c @@ -41,7 +41,7 @@ static int drm_client_modeset_test_init(struct kunit *test) test->priv = priv;
- priv->drm = drm_kunit_device_init(test, DRIVER_MODESET, "drm-client-modeset-test");
- priv->drm = drm_kunit_helper_alloc_drm_device(test, DRIVER_MODESET, "drm-client-modeset-test"); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->drm);
ret = drmm_connector_init(priv->drm, &priv->connector, diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index f1662091f250..a4ad030ed101 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -36,7 +36,9 @@ static void dev_free(struct kunit_resource *res) root_device_unregister(dev); } -struct drm_device *drm_kunit_device_init(struct kunit *test, u32 features, char *name) +struct drm_device * +drm_kunit_helper_alloc_drm_device(struct kunit *test,
u32 features, char *name)diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.h b/drivers/gpu/drm/tests/drm_kunit_helpers.h index 20ab6eec4c89..e9870c7911fe 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.h +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.h @@ -6,6 +6,9 @@ struct drm_device; struct kunit; -struct drm_device *drm_kunit_device_init(struct kunit *test, u32 features, char *name); +struct drm_device * +drm_kunit_helper_alloc_drm_device(struct kunit *test,
u32 features,char *name);#endif // DRM_KUNIT_HELPERS_H_
Hi,
On Fri, Nov 25, 2022 at 11:10:02AM -0300, Maíra Canal wrote:
On 11/23/22 12:25, Maxime Ripard wrote:
The name doesn't really fit the conventions for the other helpers in DRM/KMS, so let's rename it to make it obvious that we allocate a new DRM device.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Although I believe using "drm_device" on the function name is a bit redundant (maybe drm_kunit_helper_alloc_dev or drm_kunit_helper_alloc_device would be cleaner),
Yeah, I don't quite like the name either, but we'll need to also allocate a struct device in the next few patches so we need to make the distinction between a struct drm_device and a struct device.
Maxime
The device name isn't really useful, we can just define it instead of exposing it in the API.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/drm_client_modeset_test.c | 2 +- drivers/gpu/drm/tests/drm_kunit_helpers.c | 6 ++++-- drivers/gpu/drm/tests/drm_kunit_helpers.h | 3 +-- 3 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_client_modeset_test.c b/drivers/gpu/drm/tests/drm_client_modeset_test.c index e469d1634e2d..6920c3ffdfdf 100644 --- a/drivers/gpu/drm/tests/drm_client_modeset_test.c +++ b/drivers/gpu/drm/tests/drm_client_modeset_test.c @@ -41,7 +41,7 @@ static int drm_client_modeset_test_init(struct kunit *test)
test->priv = priv;
- priv->drm = drm_kunit_helper_alloc_drm_device(test, DRIVER_MODESET, "drm-client-modeset-test"); + priv->drm = drm_kunit_helper_alloc_drm_device(test, DRIVER_MODESET); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->drm);
ret = drmm_connector_init(priv->drm, &priv->connector, diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index a4ad030ed101..7f69f56f5892 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -9,6 +9,8 @@
#include "drm_kunit_helpers.h"
+#define FAKE_DEVICE_NAME "drm-kunit-fake-device" + struct kunit_dev { struct drm_device base; }; @@ -38,7 +40,7 @@ static void dev_free(struct kunit_resource *res)
struct drm_device * drm_kunit_helper_alloc_drm_device(struct kunit *test, - u32 features, char *name) + u32 features) { struct kunit_dev *kdev; struct drm_device *drm; @@ -46,7 +48,7 @@ drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev; int ret;
- dev = kunit_alloc_resource(test, dev_init, dev_free, GFP_KERNEL, name); + dev = kunit_alloc_resource(test, dev_init, dev_free, GFP_KERNEL, FAKE_DEVICE_NAME); if (!dev) return ERR_PTR(-ENOMEM);
diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.h b/drivers/gpu/drm/tests/drm_kunit_helpers.h index e9870c7911fe..6c12b1426ba0 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.h +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.h @@ -8,7 +8,6 @@ struct kunit;
struct drm_device * drm_kunit_helper_alloc_drm_device(struct kunit *test, - u32 features, - char *name); + u32 features);
#endif // DRM_KUNIT_HELPERS_H_
On 11/23/22 16:25, Maxime Ripard wrote:
The device name isn't really useful, we can just define it instead of exposing it in the API.
Signed-off-by: Maxime Ripard maxime@cerno.tech
drivers/gpu/drm/tests/drm_client_modeset_test.c | 2 +- drivers/gpu/drm/tests/drm_kunit_helpers.c | 6 ++++-- drivers/gpu/drm/tests/drm_kunit_helpers.h | 3 +-- 3 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_client_modeset_test.c b/drivers/gpu/drm/tests/drm_client_modeset_test.c index e469d1634e2d..6920c3ffdfdf 100644 --- a/drivers/gpu/drm/tests/drm_client_modeset_test.c +++ b/drivers/gpu/drm/tests/drm_client_modeset_test.c @@ -41,7 +41,7 @@ static int drm_client_modeset_test_init(struct kunit *test) test->priv = priv;
- priv->drm = drm_kunit_helper_alloc_drm_device(test, DRIVER_MODESET, "drm-client-modeset-test");
- priv->drm = drm_kunit_helper_alloc_drm_device(test, DRIVER_MODESET); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->drm);
ret = drmm_connector_init(priv->drm, &priv->connector, diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index a4ad030ed101..7f69f56f5892 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -9,6 +9,8 @@ #include "drm_kunit_helpers.h" +#define FAKE_DEVICE_NAME "drm-kunit-fake-device"
I think that KUNIT_DEVICE_NAME and "drm-kunit-mock-device" are more descriptive names here. But I'm also OK with the patch as is.
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
We'll need in some tests to control when the device needs to be added and removed, so let's split the device creation from the DRM device creation function.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/drm_client_modeset_test.c | 14 +++++++++++++- drivers/gpu/drm/tests/drm_kunit_helpers.c | 23 ++++------------------- drivers/gpu/drm/tests/drm_kunit_helpers.h | 5 ++++- 3 files changed, 21 insertions(+), 21 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_client_modeset_test.c b/drivers/gpu/drm/tests/drm_client_modeset_test.c index 6920c3ffdfdf..fe95fda350ae 100644 --- a/drivers/gpu/drm/tests/drm_client_modeset_test.c +++ b/drivers/gpu/drm/tests/drm_client_modeset_test.c @@ -16,6 +16,7 @@
struct drm_client_modeset_test_priv { struct drm_device *drm; + struct device *dev; struct drm_connector connector; };
@@ -41,7 +42,10 @@ static int drm_client_modeset_test_init(struct kunit *test)
test->priv = priv;
- priv->drm = drm_kunit_helper_alloc_drm_device(test, DRIVER_MODESET); + priv->dev = drm_kunit_helper_alloc_device(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->dev); + + priv->drm = drm_kunit_helper_alloc_drm_device(test, priv->dev, DRIVER_MODESET); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->drm);
ret = drmm_connector_init(priv->drm, &priv->connector, @@ -55,6 +59,13 @@ static int drm_client_modeset_test_init(struct kunit *test) return 0; }
+static void drm_client_modeset_test_exit(struct kunit *test) +{ + struct drm_client_modeset_test_priv *priv = test->priv; + + drm_kunit_helper_free_device(test, priv->dev); +} + static void drm_test_pick_cmdline_res_1920_1080_60(struct kunit *test) { struct drm_client_modeset_test_priv *priv = test->priv; @@ -92,6 +103,7 @@ static struct kunit_case drm_test_pick_cmdline_tests[] = { static struct kunit_suite drm_test_pick_cmdline_test_suite = { .name = "drm_test_pick_cmdline", .init = drm_client_modeset_test_init, + .exit = drm_client_modeset_test_exit, .test_cases = drm_test_pick_cmdline_tests };
diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 7f69f56f5892..2f974f45c5b4 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -18,40 +18,25 @@ struct kunit_dev { static const struct drm_mode_config_funcs drm_mode_config_funcs = { };
-static int dev_init(struct kunit_resource *res, void *ptr) +struct device *drm_kunit_helper_alloc_device(struct kunit *test) { - char *name = ptr; - struct device *dev; - - dev = root_device_register(name); - if (IS_ERR(dev)) - return PTR_ERR(dev); - - res->data = dev; - return 0; + return root_device_register(FAKE_DEVICE_NAME); }
-static void dev_free(struct kunit_resource *res) +void drm_kunit_helper_free_device(struct kunit *test, struct device *dev) { - struct device *dev = res->data; - root_device_unregister(dev); }
struct drm_device * -drm_kunit_helper_alloc_drm_device(struct kunit *test, +drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, u32 features) { struct kunit_dev *kdev; struct drm_device *drm; struct drm_driver *driver; - struct device *dev; int ret;
- dev = kunit_alloc_resource(test, dev_init, dev_free, GFP_KERNEL, FAKE_DEVICE_NAME); - if (!dev) - return ERR_PTR(-ENOMEM); - driver = kunit_kzalloc(test, sizeof(*driver), GFP_KERNEL); if (!driver) return ERR_PTR(-ENOMEM); diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.h b/drivers/gpu/drm/tests/drm_kunit_helpers.h index 6c12b1426ba0..b4277fe92c38 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.h +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.h @@ -6,8 +6,11 @@ struct drm_device; struct kunit;
+struct device *drm_kunit_helper_alloc_device(struct kunit *test); +void drm_kunit_helper_free_device(struct kunit *test, struct device *dev); + struct drm_device * -drm_kunit_helper_alloc_drm_device(struct kunit *test, +drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, u32 features);
#endif // DRM_KUNIT_HELPERS_H_
On 11/23/22 16:25, Maxime Ripard wrote:
We'll need in some tests to control when the device needs to be added and removed, so let's split the device creation from the DRM device creation function.
Signed-off-by: Maxime Ripard maxime@cerno.tech
[...]
-static int dev_init(struct kunit_resource *res, void *ptr) +struct device *drm_kunit_helper_alloc_device(struct kunit *test) {
- char *name = ptr;
- struct device *dev;
- dev = root_device_register(name);
- if (IS_ERR(dev))
return PTR_ERR(dev);- res->data = dev;
- return 0;
- return root_device_register(FAKE_DEVICE_NAME);
I was going to ask if a platform device could be used here, to prevent the issue you mentioned the other day of root_device not having a bus set. But I see now that you are doing that change in the next patch.
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
The device managed resources are ran if the device has bus, which is not the case of a root_device.
Let's use a platform_device instead.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/drm_kunit_helpers.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 2f974f45c5b4..651aa93a98a8 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -6,6 +6,7 @@ #include <kunit/resource.h>
#include <linux/device.h> +#include <linux/platform_device.h>
#include "drm_kunit_helpers.h"
@@ -20,12 +21,26 @@ static const struct drm_mode_config_funcs drm_mode_config_funcs = {
struct device *drm_kunit_helper_alloc_device(struct kunit *test) { - return root_device_register(FAKE_DEVICE_NAME); + struct platform_device *pdev; + int ret; + + ret = platform_driver_register(&fake_platform_driver); + KUNIT_ASSERT_EQ(test, ret, 0); + + pdev = platform_device_alloc(FAKE_DEVICE_NAME, PLATFORM_DEVID_NONE); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev); + + ret = platform_device_add(pdev); + KUNIT_ASSERT_EQ(test, ret, 0); + + return &pdev->dev; }
void drm_kunit_helper_free_device(struct kunit *test, struct device *dev) { - root_device_unregister(dev); + struct platform_device *pdev = to_platform_device(dev); + + platform_device_unregister(pdev); }
struct drm_device *
On 11/23/22 16:25, Maxime Ripard wrote:
The device managed resources are ran if the device has bus, which is not the case of a root_device.
Let's use a platform_device instead.
Signed-off-by: Maxime Ripard maxime@cerno.tech
drivers/gpu/drm/tests/drm_kunit_helpers.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 2f974f45c5b4..651aa93a98a8 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -6,6 +6,7 @@ #include <kunit/resource.h> #include <linux/device.h> +#include <linux/platform_device.h> #include "drm_kunit_helpers.h" @@ -20,12 +21,26 @@ static const struct drm_mode_config_funcs drm_mode_config_funcs = { struct device *drm_kunit_helper_alloc_device(struct kunit *test) {
- return root_device_register(FAKE_DEVICE_NAME);
- struct platform_device *pdev;
- int ret;
- ret = platform_driver_register(&fake_platform_driver);
Maybe the platform_driver_register() is something that should be done in the drm_client_modeset_test_init() callback ?
I would also as a part of this series rename the kunit_suite from "drm_test_pick_cmdline_test_suite" to "drm_test_modeset_test_suite" or something like that. Since you are already setting the .init and .exit callbacks to allocate and release the devices. So it is more than just a test suite for cmdline params testing.
- KUNIT_ASSERT_EQ(test, ret, 0);
- pdev = platform_device_alloc(FAKE_DEVICE_NAME, PLATFORM_DEVID_NONE);
- KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev);
- ret = platform_device_add(pdev);
- KUNIT_ASSERT_EQ(test, ret, 0);
- return &pdev->dev;
}
Probably won't do it as a part of this series to avoid making it cross subsystem, but I think there's room to further consolidation since this helper seems to be quite generic and something that would be useful for other subsystems. So it would make sense for kunit to have helpers to manage platform drivers and devices.
In fact, I see that drivers/base/test/test_async_driver_probe.c already has similar helpers that could be exposed to the rest of the kernel.
The patch looks good to me though and any of these changes could be done as a follow-up:
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
The device managed resources are freed when the device is detached, so it has to be bound in the first place.
Let's create a fake driver that we will bind to our fake device to benefit from the device managed cleanups in our tests.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/drm_kunit_helpers.c | 60 +++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+)
diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 651aa93a98a8..8bd7c40d0e1a 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -10,6 +10,7 @@
#include "drm_kunit_helpers.h"
+#define WQ_TIMEOUT_MS 100 #define FAKE_DEVICE_NAME "drm-kunit-fake-device"
struct kunit_dev { @@ -19,28 +20,87 @@ struct kunit_dev { static const struct drm_mode_config_funcs drm_mode_config_funcs = { };
+struct fake_pdev_sync_point { + bool done; + wait_queue_head_t wq; +}; + +struct fake_pdev_priv { + struct fake_pdev_sync_point probe; + struct fake_pdev_sync_point remove; +}; + +static int fake_probe(struct platform_device *pdev) +{ + struct fake_pdev_priv *priv = *(struct fake_pdev_priv **)dev_get_platdata(&pdev->dev); + + priv->probe.done = true; + wake_up_interruptible(&priv->probe.wq); + + return 0; +} + +static int fake_remove(struct platform_device *pdev) +{ + struct fake_pdev_priv *priv = *(struct fake_pdev_priv **)dev_get_platdata(&pdev->dev); + + priv->remove.done = true; + wake_up_interruptible(&priv->remove.wq); + + return 0; +} + +static struct platform_driver fake_platform_driver = { + .probe = fake_probe, + .remove = fake_remove, + .driver = { + .name = FAKE_DEVICE_NAME, + }, +}; + struct device *drm_kunit_helper_alloc_device(struct kunit *test) { + struct fake_pdev_priv *priv; struct platform_device *pdev; int ret;
+ priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, priv); + init_waitqueue_head(&priv->probe.wq); + init_waitqueue_head(&priv->remove.wq); + ret = platform_driver_register(&fake_platform_driver); KUNIT_ASSERT_EQ(test, ret, 0);
pdev = platform_device_alloc(FAKE_DEVICE_NAME, PLATFORM_DEVID_NONE); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev);
+ ret = platform_device_add_data(pdev, &priv, sizeof(priv)); + KUNIT_ASSERT_EQ(test, ret, 0); + ret = platform_device_add(pdev); KUNIT_ASSERT_EQ(test, ret, 0);
+ ret = wait_event_interruptible_timeout(priv->probe.wq, priv->probe.done, + msecs_to_jiffies(WQ_TIMEOUT_MS)); + KUNIT_ASSERT_GT(test, ret, 0); + return &pdev->dev; }
void drm_kunit_helper_free_device(struct kunit *test, struct device *dev) { + struct fake_pdev_priv *priv = *(struct fake_pdev_priv **)dev_get_platdata(dev); struct platform_device *pdev = to_platform_device(dev); + int ret;
platform_device_unregister(pdev); + + ret = wait_event_interruptible_timeout(priv->probe.wq, priv->remove.done, + msecs_to_jiffies(WQ_TIMEOUT_MS)); + KUNIT_ASSERT_GT(test, ret, 0); + + platform_driver_unregister(&fake_platform_driver); }
struct drm_device *
On 11/23/22 16:25, Maxime Ripard wrote:
The device managed resources are freed when the device is detached, so it has to be bound in the first place.
Let's create a fake driver that we will bind to our fake device to benefit from the device managed cleanups in our tests.
Signed-off-by: Maxime Ripard maxime@cerno.tech
drivers/gpu/drm/tests/drm_kunit_helpers.c | 60 +++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+)
If I understood the platform core code correctly, the probe is always sync in the case of platform drivers. Unless .probe_type = PROBE_PREFER_ASYNCHRONOUS is set or a module is loaded using with the "async_probe=1" parameter. So I believe the wait queue won't be needed. The only DRM driver that forces an async probe is drivers/gpu/drm/hyperv/hyperv_drm_drv.c AFAICT.
So I would drop this patch from the set for now.
The current helper to allocate a DRM device doesn't allow for any subclassing by drivers, which is going to be troublesome as we work on getting some kunit testing on atomic modesetting code.
Let's use a similar pattern to the other allocation helpers by providing the structure size and offset as arguments.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/drm_client_modeset_test.c | 4 +++- drivers/gpu/drm/tests/drm_kunit_helpers.c | 19 ++++++++----------- drivers/gpu/drm/tests/drm_kunit_helpers.h | 11 +++++++++-- 3 files changed, 20 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_client_modeset_test.c b/drivers/gpu/drm/tests/drm_client_modeset_test.c index fe95fda350ae..086469273580 100644 --- a/drivers/gpu/drm/tests/drm_client_modeset_test.c +++ b/drivers/gpu/drm/tests/drm_client_modeset_test.c @@ -45,7 +45,9 @@ static int drm_client_modeset_test_init(struct kunit *test) priv->dev = drm_kunit_helper_alloc_device(test); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->dev);
- priv->drm = drm_kunit_helper_alloc_drm_device(test, priv->dev, DRIVER_MODESET); + priv->drm = __drm_kunit_helper_alloc_drm_device(test, priv->dev, + sizeof(*priv->drm), 0, + DRIVER_MODESET); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->drm);
ret = drmm_connector_init(priv->drm, &priv->connector, diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 8bd7c40d0e1a..6d614f9e62d8 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -13,10 +13,6 @@ #define WQ_TIMEOUT_MS 100 #define FAKE_DEVICE_NAME "drm-kunit-fake-device"
-struct kunit_dev { - struct drm_device base; -}; - static const struct drm_mode_config_funcs drm_mode_config_funcs = { };
@@ -104,12 +100,13 @@ void drm_kunit_helper_free_device(struct kunit *test, struct device *dev) }
struct drm_device * -drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, - u32 features) +__drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, + size_t size, size_t offset, + u32 features) { - struct kunit_dev *kdev; struct drm_device *drm; struct drm_driver *driver; + void *container; int ret;
driver = kunit_kzalloc(test, sizeof(*driver), GFP_KERNEL); @@ -117,11 +114,11 @@ drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, return ERR_PTR(-ENOMEM);
driver->driver_features = features; - kdev = devm_drm_dev_alloc(dev, driver, struct kunit_dev, base); - if (IS_ERR(kdev)) - return ERR_CAST(kdev); + container = __devm_drm_dev_alloc(dev, driver, size, offset); + if (IS_ERR(container)) + return ERR_CAST(container);
- drm = &kdev->base; + drm = container + offset; drm->mode_config.funcs = &drm_mode_config_funcs;
ret = drmm_mode_config_init(drm); diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.h b/drivers/gpu/drm/tests/drm_kunit_helpers.h index b4277fe92c38..70abc8c806c9 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.h +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.h @@ -10,7 +10,14 @@ struct device *drm_kunit_helper_alloc_device(struct kunit *test); void drm_kunit_helper_free_device(struct kunit *test, struct device *dev);
struct drm_device * -drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, - u32 features); +__drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, + size_t size, size_t offset, + u32 features); + +#define drm_kunit_helper_alloc_drm_device(_test, _dev, _type, _member, _feat) \ + ((_type *) __drm_kunit_helper_alloc_drm_device(_test, _dev, \ + sizeof(type), \ + offsetof(type, member), \ + _feat))
#endif // DRM_KUNIT_HELPERS_H_
On 11/23/22 16:25, Maxime Ripard wrote:
The current helper to allocate a DRM device doesn't allow for any subclassing by drivers, which is going to be troublesome as we work on getting some kunit testing on atomic modesetting code.
Let's use a similar pattern to the other allocation helpers by providing the structure size and offset as arguments.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
Some tests will need to provide their own drm_driver instead of relying on the dumb one in the helpers, so let's create a helper that allows to do so.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/drm_kunit_helpers.c | 13 ++++--------- drivers/gpu/drm/tests/drm_kunit_helpers.h | 29 +++++++++++++++++++++++++++-- 2 files changed, 31 insertions(+), 11 deletions(-)
diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 6d614f9e62d8..5c758a22f526 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -100,20 +100,15 @@ void drm_kunit_helper_free_device(struct kunit *test, struct device *dev) }
struct drm_device * -__drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, - size_t size, size_t offset, - u32 features) +__drm_kunit_helper_alloc_drm_device_with_driver(struct kunit *test, + struct device *dev, + size_t size, size_t offset, + const struct drm_driver *driver) { struct drm_device *drm; - struct drm_driver *driver; void *container; int ret;
- driver = kunit_kzalloc(test, sizeof(*driver), GFP_KERNEL); - if (!driver) - return ERR_PTR(-ENOMEM); - - driver->driver_features = features; container = __devm_drm_dev_alloc(dev, driver, size, offset); if (IS_ERR(container)) return ERR_CAST(container); diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.h b/drivers/gpu/drm/tests/drm_kunit_helpers.h index 70abc8c806c9..45bc12a53cd4 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.h +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.h @@ -10,9 +10,34 @@ struct device *drm_kunit_helper_alloc_device(struct kunit *test); void drm_kunit_helper_free_device(struct kunit *test, struct device *dev);
struct drm_device * -__drm_kunit_helper_alloc_drm_device(struct kunit *test, struct device *dev, +__drm_kunit_helper_alloc_drm_device_with_driver(struct kunit *test, + struct device *dev, + size_t size, size_t offset, + const struct drm_driver *driver); + +#define drm_kunit_helper_alloc_drm_device_with_driver(_test, _dev, _type, _member, _drv) \ + ((_type *) __drm_kunit_helper_alloc_drm_device_with_driver(_test, _dev, \ + sizeof(_type), \ + offsetof(_type, _member), \ + _drv)) + +static inline struct drm_device * +__drm_kunit_helper_alloc_drm_device(struct kunit *test, + struct device *dev, size_t size, size_t offset, - u32 features); + u32 features) +{ + struct drm_driver *driver; + + driver = kunit_kzalloc(test, sizeof(*driver), GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, driver); + + driver->driver_features = features; + + return __drm_kunit_helper_alloc_drm_device_with_driver(test, dev, + size, offset, + driver); +}
#define drm_kunit_helper_alloc_drm_device(_test, _dev, _type, _member, _feat) \ ((_type *) __drm_kunit_helper_alloc_drm_device(_test, _dev, \
On 11/23/22 16:25, Maxime Ripard wrote:
Some tests will need to provide their own drm_driver instead of relying on the dumb one in the helpers, so let's create a helper that allows to do so.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
DRM-managed actions are supposed to be ran whenever the device is released. Let's introduce a basic unit test to make sure it happens.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/tests/Makefile | 1 + drivers/gpu/drm/tests/drm_managed_test.c | 68 ++++++++++++++++++++++++++++++++ 2 files changed, 69 insertions(+)
diff --git a/drivers/gpu/drm/tests/Makefile b/drivers/gpu/drm/tests/Makefile index b29ef1085cad..35962c6ef0c4 100644 --- a/drivers/gpu/drm/tests/Makefile +++ b/drivers/gpu/drm/tests/Makefile @@ -9,6 +9,7 @@ obj-$(CONFIG_DRM_KUNIT_TEST) += \ drm_format_test.o \ drm_framebuffer_test.o \ drm_kunit_helpers.o \ + drm_managed_test.o \ drm_mm_test.o \ drm_plane_helper_test.o \ drm_rect_test.o diff --git a/drivers/gpu/drm/tests/drm_managed_test.c b/drivers/gpu/drm/tests/drm_managed_test.c new file mode 100644 index 000000000000..4fc11b289d9e --- /dev/null +++ b/drivers/gpu/drm/tests/drm_managed_test.c @@ -0,0 +1,68 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <drm/drm_drv.h> +#include <drm/drm_managed.h> + +#include <kunit/resource.h> + +#include <linux/device.h> + +#include "drm_kunit_helpers.h" + +#define TEST_TIMEOUT_MS 100 + +struct managed_test_priv { + bool action_done; + wait_queue_head_t action_wq; +}; + +static void drm_action(struct drm_device *drm, void *ptr) +{ + struct managed_test_priv *priv = ptr; + + priv->action_done = true; + wake_up_interruptible(&priv->action_wq); +} + +static void drm_test_managed_run_action(struct kunit *test) +{ + struct managed_test_priv *priv; + struct drm_device *drm; + struct device *dev; + int ret; + + priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv); + init_waitqueue_head(&priv->action_wq); + + dev = drm_kunit_helper_alloc_device(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + drm = __drm_kunit_helper_alloc_drm_device(test, dev, sizeof(*drm), 0, DRIVER_MODESET); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, drm); + + ret = drmm_add_action_or_reset(drm, drm_action, priv); + KUNIT_EXPECT_EQ(test, ret, 0); + + ret = drm_dev_register(drm, 0); + KUNIT_ASSERT_EQ(test, ret, 0); + + drm_dev_unregister(drm); + drm_kunit_helper_free_device(test, dev); + + ret = wait_event_interruptible_timeout(priv->action_wq, priv->action_done, + msecs_to_jiffies(TEST_TIMEOUT_MS)); + KUNIT_EXPECT_GT(test, ret, 0); +} + +static struct kunit_case drm_managed_tests[] = { + KUNIT_CASE(drm_test_managed_run_action), + {} +}; + +static struct kunit_suite drm_managed_test_suite = { + .name = "drm-test-managed", + .test_cases = drm_managed_tests +}; + +kunit_test_suite(drm_managed_test_suite);
On 11/23/22 16:25, Maxime Ripard wrote:
DRM-managed actions are supposed to be ran whenever the device is released. Let's introduce a basic unit test to make sure it happens.
Signed-off-by: Maxime Ripard maxime@cerno.tech
drivers/gpu/drm/tests/Makefile | 1 + drivers/gpu/drm/tests/drm_managed_test.c | 68 ++++++++++++++++++++++++++++++++ 2 files changed, 69 insertions(+)
diff --git a/drivers/gpu/drm/tests/Makefile b/drivers/gpu/drm/tests/Makefile index b29ef1085cad..35962c6ef0c4 100644 --- a/drivers/gpu/drm/tests/Makefile +++ b/drivers/gpu/drm/tests/Makefile @@ -9,6 +9,7 @@ obj-$(CONFIG_DRM_KUNIT_TEST) += \ drm_format_test.o \ drm_framebuffer_test.o \ drm_kunit_helpers.o \
- drm_managed_test.o \ drm_mm_test.o \ drm_plane_helper_test.o \ drm_rect_test.o
diff --git a/drivers/gpu/drm/tests/drm_managed_test.c b/drivers/gpu/drm/tests/drm_managed_test.c new file mode 100644 index 000000000000..4fc11b289d9e --- /dev/null +++ b/drivers/gpu/drm/tests/drm_managed_test.c @@ -0,0 +1,68 @@ +// SPDX-License-Identifier: GPL-2.0
+#include <drm/drm_drv.h> +#include <drm/drm_managed.h>
+#include <kunit/resource.h>
+#include <linux/device.h>
+#include "drm_kunit_helpers.h"
+#define TEST_TIMEOUT_MS 100
Why did you choose this timeout value? Maybe we can add some comments explaining the rationale.
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
The drm_atomic_get_(old|new)_*_state don't modify the passed drm_atomic_state, so we can make it const.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/drm_atomic.c | 12 ++++++------ include/drm/drm_atomic.h | 32 ++++++++++++++++---------------- 2 files changed, 22 insertions(+), 22 deletions(-)
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index f197f59f6d99..e666799a46d5 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -880,7 +880,7 @@ EXPORT_SYMBOL(drm_atomic_get_private_obj_state); * or NULL if the private_obj is not part of the global atomic state. */ struct drm_private_state * -drm_atomic_get_old_private_obj_state(struct drm_atomic_state *state, +drm_atomic_get_old_private_obj_state(const struct drm_atomic_state *state, struct drm_private_obj *obj) { int i; @@ -902,7 +902,7 @@ EXPORT_SYMBOL(drm_atomic_get_old_private_obj_state); * or NULL if the private_obj is not part of the global atomic state. */ struct drm_private_state * -drm_atomic_get_new_private_obj_state(struct drm_atomic_state *state, +drm_atomic_get_new_private_obj_state(const struct drm_atomic_state *state, struct drm_private_obj *obj) { int i; @@ -934,7 +934,7 @@ EXPORT_SYMBOL(drm_atomic_get_new_private_obj_state); * not connected. */ struct drm_connector * -drm_atomic_get_old_connector_for_encoder(struct drm_atomic_state *state, +drm_atomic_get_old_connector_for_encoder(const struct drm_atomic_state *state, struct drm_encoder *encoder) { struct drm_connector_state *conn_state; @@ -968,7 +968,7 @@ EXPORT_SYMBOL(drm_atomic_get_old_connector_for_encoder); * not connected. */ struct drm_connector * -drm_atomic_get_new_connector_for_encoder(struct drm_atomic_state *state, +drm_atomic_get_new_connector_for_encoder(const struct drm_atomic_state *state, struct drm_encoder *encoder) { struct drm_connector_state *conn_state; @@ -1117,7 +1117,7 @@ EXPORT_SYMBOL(drm_atomic_get_bridge_state); * the bridge is not part of the global atomic state. */ struct drm_bridge_state * -drm_atomic_get_old_bridge_state(struct drm_atomic_state *state, +drm_atomic_get_old_bridge_state(const struct drm_atomic_state *state, struct drm_bridge *bridge) { struct drm_private_state *obj_state; @@ -1139,7 +1139,7 @@ EXPORT_SYMBOL(drm_atomic_get_old_bridge_state); * the bridge is not part of the global atomic state. */ struct drm_bridge_state * -drm_atomic_get_new_bridge_state(struct drm_atomic_state *state, +drm_atomic_get_new_bridge_state(const struct drm_atomic_state *state, struct drm_bridge *bridge) { struct drm_private_state *obj_state; diff --git a/include/drm/drm_atomic.h b/include/drm/drm_atomic.h index 10b1990bc1f6..92586ab55ef5 100644 --- a/include/drm/drm_atomic.h +++ b/include/drm/drm_atomic.h @@ -515,17 +515,17 @@ struct drm_private_state * __must_check drm_atomic_get_private_obj_state(struct drm_atomic_state *state, struct drm_private_obj *obj); struct drm_private_state * -drm_atomic_get_old_private_obj_state(struct drm_atomic_state *state, +drm_atomic_get_old_private_obj_state(const struct drm_atomic_state *state, struct drm_private_obj *obj); struct drm_private_state * -drm_atomic_get_new_private_obj_state(struct drm_atomic_state *state, +drm_atomic_get_new_private_obj_state(const struct drm_atomic_state *state, struct drm_private_obj *obj);
struct drm_connector * -drm_atomic_get_old_connector_for_encoder(struct drm_atomic_state *state, +drm_atomic_get_old_connector_for_encoder(const struct drm_atomic_state *state, struct drm_encoder *encoder); struct drm_connector * -drm_atomic_get_new_connector_for_encoder(struct drm_atomic_state *state, +drm_atomic_get_new_connector_for_encoder(const struct drm_atomic_state *state, struct drm_encoder *encoder);
/** @@ -540,7 +540,7 @@ drm_atomic_get_new_connector_for_encoder(struct drm_atomic_state *state, * @drm_atomic_get_new_crtc_state should be used instead. */ static inline struct drm_crtc_state * -drm_atomic_get_existing_crtc_state(struct drm_atomic_state *state, +drm_atomic_get_existing_crtc_state(const struct drm_atomic_state *state, struct drm_crtc *crtc) { return state->crtcs[drm_crtc_index(crtc)].state; @@ -555,7 +555,7 @@ drm_atomic_get_existing_crtc_state(struct drm_atomic_state *state, * NULL if the CRTC is not part of the global atomic state. */ static inline struct drm_crtc_state * -drm_atomic_get_old_crtc_state(struct drm_atomic_state *state, +drm_atomic_get_old_crtc_state(const struct drm_atomic_state *state, struct drm_crtc *crtc) { return state->crtcs[drm_crtc_index(crtc)].old_state; @@ -569,7 +569,7 @@ drm_atomic_get_old_crtc_state(struct drm_atomic_state *state, * NULL if the CRTC is not part of the global atomic state. */ static inline struct drm_crtc_state * -drm_atomic_get_new_crtc_state(struct drm_atomic_state *state, +drm_atomic_get_new_crtc_state(const struct drm_atomic_state *state, struct drm_crtc *crtc) { return state->crtcs[drm_crtc_index(crtc)].new_state; @@ -587,7 +587,7 @@ drm_atomic_get_new_crtc_state(struct drm_atomic_state *state, * @drm_atomic_get_new_plane_state should be used instead. */ static inline struct drm_plane_state * -drm_atomic_get_existing_plane_state(struct drm_atomic_state *state, +drm_atomic_get_existing_plane_state(const struct drm_atomic_state *state, struct drm_plane *plane) { return state->planes[drm_plane_index(plane)].state; @@ -602,7 +602,7 @@ drm_atomic_get_existing_plane_state(struct drm_atomic_state *state, * NULL if the plane is not part of the global atomic state. */ static inline struct drm_plane_state * -drm_atomic_get_old_plane_state(struct drm_atomic_state *state, +drm_atomic_get_old_plane_state(const struct drm_atomic_state *state, struct drm_plane *plane) { return state->planes[drm_plane_index(plane)].old_state; @@ -617,7 +617,7 @@ drm_atomic_get_old_plane_state(struct drm_atomic_state *state, * NULL if the plane is not part of the global atomic state. */ static inline struct drm_plane_state * -drm_atomic_get_new_plane_state(struct drm_atomic_state *state, +drm_atomic_get_new_plane_state(const struct drm_atomic_state *state, struct drm_plane *plane) { return state->planes[drm_plane_index(plane)].new_state; @@ -635,7 +635,7 @@ drm_atomic_get_new_plane_state(struct drm_atomic_state *state, * @drm_atomic_get_new_connector_state should be used instead. */ static inline struct drm_connector_state * -drm_atomic_get_existing_connector_state(struct drm_atomic_state *state, +drm_atomic_get_existing_connector_state(const struct drm_atomic_state *state, struct drm_connector *connector) { int index = drm_connector_index(connector); @@ -655,7 +655,7 @@ drm_atomic_get_existing_connector_state(struct drm_atomic_state *state, * or NULL if the connector is not part of the global atomic state. */ static inline struct drm_connector_state * -drm_atomic_get_old_connector_state(struct drm_atomic_state *state, +drm_atomic_get_old_connector_state(const struct drm_atomic_state *state, struct drm_connector *connector) { int index = drm_connector_index(connector); @@ -675,7 +675,7 @@ drm_atomic_get_old_connector_state(struct drm_atomic_state *state, * or NULL if the connector is not part of the global atomic state. */ static inline struct drm_connector_state * -drm_atomic_get_new_connector_state(struct drm_atomic_state *state, +drm_atomic_get_new_connector_state(const struct drm_atomic_state *state, struct drm_connector *connector) { int index = drm_connector_index(connector); @@ -713,7 +713,7 @@ drm_atomic_get_new_connector_state(struct drm_atomic_state *state, * Read-only pointer to the current plane state. */ static inline const struct drm_plane_state * -__drm_atomic_get_current_plane_state(struct drm_atomic_state *state, +__drm_atomic_get_current_plane_state(const struct drm_atomic_state *state, struct drm_plane *plane) { if (state->planes[drm_plane_index(plane)].state) @@ -1134,10 +1134,10 @@ struct drm_bridge_state * drm_atomic_get_bridge_state(struct drm_atomic_state *state, struct drm_bridge *bridge); struct drm_bridge_state * -drm_atomic_get_old_bridge_state(struct drm_atomic_state *state, +drm_atomic_get_old_bridge_state(const struct drm_atomic_state *state, struct drm_bridge *bridge); struct drm_bridge_state * -drm_atomic_get_new_bridge_state(struct drm_atomic_state *state, +drm_atomic_get_new_bridge_state(const struct drm_atomic_state *state, struct drm_bridge *bridge);
#endif /* DRM_ATOMIC_H_ */
On 11/23/22 16:25, Maxime Ripard wrote:
The drm_atomic_get_(old|new)_*_state don't modify the passed drm_atomic_state, so we can make it const.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:51 +0100, Maxime Ripard wrote:
The drm_atomic_get_(old|new)_*_state don't modify the passed drm_atomic_state, so we can make it const.
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
On the vc4 devices (and later), the blending is done by a single device called the HVS. The HVS has three FIFO that can operate in parallel, and route their output to 6 CRTCs and 7 encoders on the BCM2711.
Each of these CRTCs and encoders have some contraints on which FIFO they can feed from, so we need some code to take all those constraints into account and assign FIFOs to CRTCs.
The problem can be simplified by assigning those FIFOs to CRTCs by ascending output index number. We had a comment mentioning it already, but we were never actually enforcing it.
It was working still in most situations because the probe order is roughly equivalent, except for the (optional, and fairly rarely used on the Pi4) VEC which was last in the probe order sequence, but one of the earliest device to assign.
This resulted in configurations that were rejected by our code but were still valid with a different assignment.
We can fix this by making sure we assign CRTCs to FIFOs by ordering them by ascending HVS output index.
Fixes: 87ebcd42fb7b ("drm/vc4: crtc: Assign output to channel automatically") Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_kms.c | 109 +++++++++++++++++++++++++++++------------- 1 file changed, 76 insertions(+), 33 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c index 5c97642ed66a..17a28f9a2f31 100644 --- a/drivers/gpu/drm/vc4/vc4_kms.c +++ b/drivers/gpu/drm/vc4/vc4_kms.c @@ -12,6 +12,7 @@ */
#include <linux/clk.h> +#include <linux/sort.h>
#include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> @@ -776,6 +777,20 @@ static int vc4_hvs_channels_obj_init(struct vc4_dev *vc4) return drmm_add_action_or_reset(&vc4->base, vc4_hvs_channels_obj_fini, NULL); }
+static int cmp_vc4_crtc_hvs_output(const void *a, const void *b) +{ + const struct vc4_crtc *crtc_a = + to_vc4_crtc(*(const struct drm_crtc **)a); + const struct vc4_crtc_data *data_a = + vc4_crtc_to_vc4_crtc_data(crtc_a); + const struct vc4_crtc *crtc_b = + to_vc4_crtc(*(const struct drm_crtc **)b); + const struct vc4_crtc_data *data_b = + vc4_crtc_to_vc4_crtc_data(crtc_b); + + return data_a->hvs_output - data_b->hvs_output; +} + /* * The BCM2711 HVS has up to 7 outputs connected to the pixelvalves and * the TXP (and therefore all the CRTCs found on that platform). @@ -810,10 +825,11 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev, struct drm_atomic_state *state) { struct vc4_hvs_state *hvs_new_state; - struct drm_crtc_state *old_crtc_state, *new_crtc_state; + struct drm_crtc **sorted_crtcs; struct drm_crtc *crtc; unsigned int unassigned_channels = 0; unsigned int i; + int ret;
hvs_new_state = vc4_hvs_get_global_state(state); if (IS_ERR(hvs_new_state)) @@ -823,15 +839,59 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev, if (!hvs_new_state->fifo_state[i].in_use) unassigned_channels |= BIT(i);
- for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { - struct vc4_crtc_state *old_vc4_crtc_state = - to_vc4_crtc_state(old_crtc_state); - struct vc4_crtc_state *new_vc4_crtc_state = - to_vc4_crtc_state(new_crtc_state); - struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc); + /* + * The problem we have to solve here is that we have up to 7 + * encoders, connected to up to 6 CRTCs. + * + * Those CRTCs, depending on the instance, can be routed to 1, 2 + * or 3 HVS FIFOs, and we need to set the change the muxing + * between FIFOs and outputs in the HVS accordingly. + * + * It would be pretty hard to come up with an algorithm that + * would generically solve this. However, the current routing + * trees we support allow us to simplify a bit the problem. + * + * Indeed, with the current supported layouts, if we try to + * assign in the ascending crtc index order the FIFOs, we can't + * fall into the situation where an earlier CRTC that had + * multiple routes is assigned one that was the only option for + * a later CRTC. + * + * If the layout changes and doesn't give us that in the future, + * we will need to have something smarter, but it works so far. + */ + sorted_crtcs = kmalloc_array(dev->num_crtcs, sizeof(*sorted_crtcs), GFP_KERNEL); + if (!sorted_crtcs) + return -ENOMEM; + + i = 0; + drm_for_each_crtc(crtc, dev) + sorted_crtcs[i++] = crtc; + + sort(sorted_crtcs, i, sizeof(*sorted_crtcs), cmp_vc4_crtc_hvs_output, NULL); + + for (i = 0; i < dev->num_crtcs; i++) { + struct vc4_crtc_state *old_vc4_crtc_state, *new_vc4_crtc_state; + struct drm_crtc_state *old_crtc_state, *new_crtc_state; + struct vc4_crtc *vc4_crtc; unsigned int matching_channels; unsigned int channel;
+ crtc = sorted_crtcs[i]; + if (!crtc) + continue; + vc4_crtc = to_vc4_crtc(crtc); + + old_crtc_state = drm_atomic_get_old_crtc_state(state, crtc); + if (!old_crtc_state) + continue; + old_vc4_crtc_state = to_vc4_crtc_state(old_crtc_state); + + new_crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + if (!new_crtc_state) + continue; + new_vc4_crtc_state = to_vc4_crtc_state(new_crtc_state); + drm_dbg(dev, "%s: Trying to find a channel.\n", crtc->name);
/* Nothing to do here, let's skip it */ @@ -860,33 +920,11 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev, continue; }
- /* - * The problem we have to solve here is that we have - * up to 7 encoders, connected to up to 6 CRTCs. - * - * Those CRTCs, depending on the instance, can be - * routed to 1, 2 or 3 HVS FIFOs, and we need to set - * the change the muxing between FIFOs and outputs in - * the HVS accordingly. - * - * It would be pretty hard to come up with an - * algorithm that would generically solve - * this. However, the current routing trees we support - * allow us to simplify a bit the problem. - * - * Indeed, with the current supported layouts, if we - * try to assign in the ascending crtc index order the - * FIFOs, we can't fall into the situation where an - * earlier CRTC that had multiple routes is assigned - * one that was the only option for a later CRTC. - * - * If the layout changes and doesn't give us that in - * the future, we will need to have something smarter, - * but it works so far. - */ matching_channels = unassigned_channels & vc4_crtc->data->hvs_available_channels; - if (!matching_channels) - return -EINVAL; + if (!matching_channels) { + ret = -EINVAL; + goto err_free_crtc_array; + }
channel = ffs(matching_channels) - 1;
@@ -896,7 +934,12 @@ static int vc4_pv_muxing_atomic_check(struct drm_device *dev, hvs_new_state->fifo_state[channel].in_use = true; }
+ kfree(sorted_crtcs); return 0; + +err_free_crtc_array: + kfree(sorted_crtcs); + return ret; }
static int
On 11/23/22 16:25, Maxime Ripard wrote:
On the vc4 devices (and later), the blending is done by a single device called the HVS. The HVS has three FIFO that can operate in parallel, and route their output to 6 CRTCs and 7 encoders on the BCM2711.
Each of these CRTCs and encoders have some contraints on which FIFO they
constraints.
can feed from, so we need some code to take all those constraints into account and assign FIFOs to CRTCs.
The problem can be simplified by assigning those FIFOs to CRTCs by ascending output index number. We had a comment mentioning it already, but we were never actually enforcing it.
It was working still in most situations because the probe order is roughly equivalent, except for the (optional, and fairly rarely used on the Pi4) VEC which was last in the probe order sequence, but one of the earliest device to assign.
This resulted in configurations that were rejected by our code but were still valid with a different assignment.
We can fix this by making sure we assign CRTCs to FIFOs by ordering them by ascending HVS output index.
Fixes: 87ebcd42fb7b ("drm/vc4: crtc: Assign output to channel automatically") Signed-off-by: Maxime Ripard maxime@cerno.tech
[...]
- for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
struct vc4_crtc_state *old_vc4_crtc_state =to_vc4_crtc_state(old_crtc_state);struct vc4_crtc_state *new_vc4_crtc_state =to_vc4_crtc_state(new_crtc_state);struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
- /*
* The problem we have to solve here is that we have up to 7* encoders, connected to up to 6 CRTCs.** Those CRTCs, depending on the instance, can be routed to 1, 2* or 3 HVS FIFOs, and we need to set the change the muxing
This sentence sounds a little bit off to me. Did you mean:
"we need to set the muxing between" or "we need to change the muxing" ?
I'm not familiar with VC4 but the patch seems to do what the commit message says, so the changes look good to me.
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:52 +0100, Maxime Ripard wrote:
On the vc4 devices (and later), the blending is done by a single device called the HVS. The HVS has three FIFO that can operate in parallel, and route their output to 6 CRTCs and 7 encoders on the BCM2711.
Each of these CRTCs and encoders have some contraints on which FIFO they can feed from, so we need some code to take all those constraints into account and assign FIFOs to CRTCs.
[...]
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
None of our wrappers around container_of to access our objects from the DRM object pointer actually modify the latter.
Let's make them const.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_drv.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 515228682e8e..3ff56c1821ef 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -233,7 +233,7 @@ struct vc4_dev { };
static inline struct vc4_dev * -to_vc4_dev(struct drm_device *dev) +to_vc4_dev(const struct drm_device *dev) { return container_of(dev, struct vc4_dev, base); } @@ -286,7 +286,7 @@ struct vc4_bo { };
static inline struct vc4_bo * -to_vc4_bo(struct drm_gem_object *bo) +to_vc4_bo(const struct drm_gem_object *bo) { return container_of(to_drm_gem_dma_obj(bo), struct vc4_bo, base); } @@ -299,7 +299,7 @@ struct vc4_fence { };
static inline struct vc4_fence * -to_vc4_fence(struct dma_fence *fence) +to_vc4_fence(const struct dma_fence *fence) { return container_of(fence, struct vc4_fence, base); } @@ -360,7 +360,7 @@ struct vc4_plane { };
static inline struct vc4_plane * -to_vc4_plane(struct drm_plane *plane) +to_vc4_plane(const struct drm_plane *plane) { return container_of(plane, struct vc4_plane, base); } @@ -436,7 +436,7 @@ struct vc4_plane_state { };
static inline struct vc4_plane_state * -to_vc4_plane_state(struct drm_plane_state *state) +to_vc4_plane_state(const struct drm_plane_state *state) { return container_of(state, struct vc4_plane_state, base); } @@ -466,7 +466,7 @@ struct vc4_encoder { };
static inline struct vc4_encoder * -to_vc4_encoder(struct drm_encoder *encoder) +to_vc4_encoder(const struct drm_encoder *encoder) { return container_of(encoder, struct vc4_encoder, base); } @@ -539,7 +539,7 @@ struct vc4_crtc { };
static inline struct vc4_crtc * -to_vc4_crtc(struct drm_crtc *crtc) +to_vc4_crtc(const struct drm_crtc *crtc) { return container_of(crtc, struct vc4_crtc, base); } @@ -584,7 +584,7 @@ struct vc4_crtc_state { #define VC4_HVS_CHANNEL_DISABLED ((unsigned int)-1)
static inline struct vc4_crtc_state * -to_vc4_crtc_state(struct drm_crtc_state *crtc_state) +to_vc4_crtc_state(const struct drm_crtc_state *crtc_state) { return container_of(crtc_state, struct vc4_crtc_state, base); }
On 11/23/22 16:25, Maxime Ripard wrote:
None of our wrappers around container_of to access our objects from the DRM object pointer actually modify the latter.
Let's make them const.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:53 +0100, Maxime Ripard wrote:
None of our wrappers around container_of to access our objects from the DRM object pointer actually modify the latter.
Let's make them const.
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
In order to introduce unit tests for the HVS state computation, we'll need access to the vc4_hvs_state struct definition and its associated helpers.
Let's move them in our driver header.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_drv.h | 23 +++++++++++++++++++++++ drivers/gpu/drm/vc4/vc4_kms.c | 25 +++---------------------- 2 files changed, 26 insertions(+), 22 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 3ff56c1821ef..6ff63036915c 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -355,6 +355,29 @@ struct vc4_hvs { bool vc5_hdmi_enable_4096by2160; };
+#define HVS_NUM_CHANNELS 3 + +struct vc4_hvs_state { + struct drm_private_state base; + unsigned long core_clock_rate; + + struct { + unsigned in_use: 1; + unsigned long fifo_load; + struct drm_crtc_commit *pending_commit; + } fifo_state[HVS_NUM_CHANNELS]; +}; + +static inline struct vc4_hvs_state * +to_vc4_hvs_state(const struct drm_private_state *priv) +{ + return container_of(priv, struct vc4_hvs_state, base); +} + +struct vc4_hvs_state *vc4_hvs_get_global_state(struct drm_atomic_state *state); +struct vc4_hvs_state *vc4_hvs_get_old_global_state(struct drm_atomic_state *state); +struct vc4_hvs_state *vc4_hvs_get_new_global_state(struct drm_atomic_state *state); + struct vc4_plane { struct drm_plane base; }; diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c index 17a28f9a2f31..cb2197622b2b 100644 --- a/drivers/gpu/drm/vc4/vc4_kms.c +++ b/drivers/gpu/drm/vc4/vc4_kms.c @@ -25,8 +25,6 @@ #include "vc4_drv.h" #include "vc4_regs.h"
-#define HVS_NUM_CHANNELS 3 - struct vc4_ctm_state { struct drm_private_state base; struct drm_color_ctm *ctm; @@ -39,23 +37,6 @@ to_vc4_ctm_state(const struct drm_private_state *priv) return container_of(priv, struct vc4_ctm_state, base); }
-struct vc4_hvs_state { - struct drm_private_state base; - unsigned long core_clock_rate; - - struct { - unsigned in_use: 1; - unsigned long fifo_load; - struct drm_crtc_commit *pending_commit; - } fifo_state[HVS_NUM_CHANNELS]; -}; - -static struct vc4_hvs_state * -to_vc4_hvs_state(const struct drm_private_state *priv) -{ - return container_of(priv, struct vc4_hvs_state, base); -} - struct vc4_load_tracker_state { struct drm_private_state base; u64 hvs_load; @@ -191,7 +172,7 @@ vc4_ctm_commit(struct vc4_dev *vc4, struct drm_atomic_state *state) VC4_SET_FIELD(ctm_state->fifo, SCALER_OLEDOFFS_DISPFIFO)); }
-static struct vc4_hvs_state * +struct vc4_hvs_state * vc4_hvs_get_new_global_state(struct drm_atomic_state *state) { struct vc4_dev *vc4 = to_vc4_dev(state->dev); @@ -204,7 +185,7 @@ vc4_hvs_get_new_global_state(struct drm_atomic_state *state) return to_vc4_hvs_state(priv_state); }
-static struct vc4_hvs_state * +struct vc4_hvs_state * vc4_hvs_get_old_global_state(struct drm_atomic_state *state) { struct vc4_dev *vc4 = to_vc4_dev(state->dev); @@ -217,7 +198,7 @@ vc4_hvs_get_old_global_state(struct drm_atomic_state *state) return to_vc4_hvs_state(priv_state); }
-static struct vc4_hvs_state * +struct vc4_hvs_state * vc4_hvs_get_global_state(struct drm_atomic_state *state) { struct vc4_dev *vc4 = to_vc4_dev(state->dev);
On 11/23/22 16:25, Maxime Ripard wrote:
In order to introduce unit tests for the HVS state computation, we'll need access to the vc4_hvs_state struct definition and its associated helpers.
Let's move them in our driver header.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
The vc4_hvs_get_(old|new)_global_state functions don't modify the drm_atomic_state passed as an argument, so let's make it const.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_drv.h | 4 ++-- drivers/gpu/drm/vc4/vc4_kms.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 6ff63036915c..855207c07e2e 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -375,8 +375,8 @@ to_vc4_hvs_state(const struct drm_private_state *priv) }
struct vc4_hvs_state *vc4_hvs_get_global_state(struct drm_atomic_state *state); -struct vc4_hvs_state *vc4_hvs_get_old_global_state(struct drm_atomic_state *state); -struct vc4_hvs_state *vc4_hvs_get_new_global_state(struct drm_atomic_state *state); +struct vc4_hvs_state *vc4_hvs_get_old_global_state(const struct drm_atomic_state *state); +struct vc4_hvs_state *vc4_hvs_get_new_global_state(const struct drm_atomic_state *state);
struct vc4_plane { struct drm_plane base; diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c index cb2197622b2b..6c9949399310 100644 --- a/drivers/gpu/drm/vc4/vc4_kms.c +++ b/drivers/gpu/drm/vc4/vc4_kms.c @@ -173,7 +173,7 @@ vc4_ctm_commit(struct vc4_dev *vc4, struct drm_atomic_state *state) }
struct vc4_hvs_state * -vc4_hvs_get_new_global_state(struct drm_atomic_state *state) +vc4_hvs_get_new_global_state(const struct drm_atomic_state *state) { struct vc4_dev *vc4 = to_vc4_dev(state->dev); struct drm_private_state *priv_state; @@ -186,7 +186,7 @@ vc4_hvs_get_new_global_state(struct drm_atomic_state *state) }
struct vc4_hvs_state * -vc4_hvs_get_old_global_state(struct drm_atomic_state *state) +vc4_hvs_get_old_global_state(const struct drm_atomic_state *state) { struct vc4_dev *vc4 = to_vc4_dev(state->dev); struct drm_private_state *priv_state;
On 11/23/22 16:25, Maxime Ripard wrote:
The vc4_hvs_get_(old|new)_global_state functions don't modify the drm_atomic_state passed as an argument, so let's make it const.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:55 +0100, Maxime Ripard wrote:
The vc4_hvs_get_(old|new)_global_state functions don't modify the drm_atomic_state passed as an argument, so let's make it const.
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
The current order of variable assignments is unneccessarily complex, let's make it simpler.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_txp.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index bd181b5a7b52..b00c6fae972f 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -490,7 +490,6 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) struct drm_device *drm = dev_get_drvdata(master); struct vc4_crtc *vc4_crtc; struct vc4_txp *txp; - struct drm_crtc *crtc; struct drm_encoder *encoder; int ret, irq;
@@ -501,18 +500,16 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) txp = drmm_kzalloc(drm, sizeof(*txp), GFP_KERNEL); if (!txp) return -ENOMEM; - vc4_crtc = &txp->base; - crtc = &vc4_crtc->base; - - vc4_crtc->pdev = pdev; - vc4_crtc->data = &vc4_txp_crtc_data; - vc4_crtc->feeds_txp = true;
txp->pdev = pdev; - txp->regs = vc4_ioremap_regs(pdev, 0); if (IS_ERR(txp->regs)) return PTR_ERR(txp->regs); + + vc4_crtc = &txp->base; + vc4_crtc->pdev = pdev; + vc4_crtc->data = &vc4_txp_crtc_data; + vc4_crtc->feeds_txp = true; vc4_crtc->regset.base = txp->regs; vc4_crtc->regset.regs = txp_regs; vc4_crtc->regset.nregs = ARRAY_SIZE(txp_regs); @@ -533,7 +530,7 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) return ret;
encoder = &txp->connector.encoder; - encoder->possible_crtcs = drm_crtc_mask(crtc); + encoder->possible_crtcs = drm_crtc_mask(&vc4_crtc->base);
ret = devm_request_irq(dev, irq, vc4_txp_interrupt, 0, dev_name(dev), txp);
On 11/23/22 16:25, Maxime Ripard wrote:
The current order of variable assignments is unneccessarily complex, let's make it simpler.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Much easier to follow indeed.
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:56 +0100, Maxime Ripard wrote:
The current order of variable assignments is unneccessarily complex, let's make it simpler.
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
The TXP is integrated as a separate CRTC/Encoder/Connector combo, but for some reason doesn't rely on the vc4_encoder type and it's associated type.
Let's create a type to make it consistent with the other encoders.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_drv.h | 1 + drivers/gpu/drm/vc4/vc4_txp.c | 30 ++++++++++++++++++++---------- 2 files changed, 21 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 855207c07e2e..31228d4cf080 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -473,6 +473,7 @@ enum vc4_encoder_type { VC4_ENCODER_TYPE_DSI1, VC4_ENCODER_TYPE_SMI, VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, };
struct vc4_encoder { diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index b00c6fae972f..0bb8e97d7891 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -153,6 +153,7 @@ struct vc4_txp {
struct platform_device *pdev;
+ struct vc4_encoder encoder; struct drm_writeback_connector connector;
void __iomem *regs; @@ -160,7 +161,7 @@ struct vc4_txp {
static inline struct vc4_txp *encoder_to_vc4_txp(struct drm_encoder *encoder) { - return container_of(encoder, struct vc4_txp, connector.encoder); + return container_of(encoder, struct vc4_txp, encoder.base); }
static inline struct vc4_txp *connector_to_vc4_txp(struct drm_connector *conn) @@ -488,9 +489,10 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) { struct platform_device *pdev = to_platform_device(dev); struct drm_device *drm = dev_get_drvdata(master); + struct vc4_encoder *vc4_encoder; + struct drm_encoder *encoder; struct vc4_crtc *vc4_crtc; struct vc4_txp *txp; - struct drm_encoder *encoder; int ret, irq;
irq = platform_get_irq(pdev, 0); @@ -514,13 +516,24 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) vc4_crtc->regset.regs = txp_regs; vc4_crtc->regset.nregs = ARRAY_SIZE(txp_regs);
+ vc4_encoder = &txp->encoder; + txp->encoder.type = VC4_ENCODER_TYPE_TXP; + + encoder = &vc4_encoder->base; + encoder->possible_crtcs = drm_crtc_mask(&vc4_crtc->base); + + drm_encoder_helper_add(encoder, &vc4_txp_encoder_helper_funcs); + + ret = drmm_encoder_init(drm, encoder, NULL, DRM_MODE_ENCODER_VIRTUAL, NULL); + if (ret) + return ret; + drm_connector_helper_add(&txp->connector.base, &vc4_txp_connector_helper_funcs); - ret = drm_writeback_connector_init(drm, &txp->connector, - &vc4_txp_connector_funcs, - &vc4_txp_encoder_helper_funcs, - drm_fmts, ARRAY_SIZE(drm_fmts), - 0); + ret = drm_writeback_connector_init_with_encoder(drm, &txp->connector, + encoder, + &vc4_txp_connector_funcs, + drm_fmts, ARRAY_SIZE(drm_fmts)); if (ret) return ret;
@@ -529,9 +542,6 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) if (ret) return ret;
- encoder = &txp->connector.encoder; - encoder->possible_crtcs = drm_crtc_mask(&vc4_crtc->base); - ret = devm_request_irq(dev, irq, vc4_txp_interrupt, 0, dev_name(dev), txp); if (ret)
On 11/23/22 16:25, Maxime Ripard wrote:
The TXP is integrated as a separate CRTC/Encoder/Connector combo, but for some reason doesn't rely on the vc4_encoder type and it's associated type.
Let's create a type to make it consistent with the other encoders.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:57 +0100, Maxime Ripard wrote:
The TXP is integrated as a separate CRTC/Encoder/Connector combo, but for some reason doesn't rely on the vc4_encoder type and it's associated type.
Let's create a type to make it consistent with the other encoders.
[...]
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
It makes more sense to register the CRTC before the encoder and connectors, so let's move our call around.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_txp.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index 0bb8e97d7891..c2a6bea6fd96 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -516,6 +516,11 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) vc4_crtc->regset.regs = txp_regs; vc4_crtc->regset.nregs = ARRAY_SIZE(txp_regs);
+ ret = vc4_crtc_init(drm, vc4_crtc, + &vc4_txp_crtc_funcs, &vc4_txp_crtc_helper_funcs); + if (ret) + return ret; + vc4_encoder = &txp->encoder; txp->encoder.type = VC4_ENCODER_TYPE_TXP;
@@ -537,11 +542,6 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) if (ret) return ret;
- ret = vc4_crtc_init(drm, vc4_crtc, - &vc4_txp_crtc_funcs, &vc4_txp_crtc_helper_funcs); - if (ret) - return ret; - ret = devm_request_irq(dev, irq, vc4_txp_interrupt, 0, dev_name(dev), txp); if (ret)
On 11/23/22 16:25, Maxime Ripard wrote:
It makes more sense to register the CRTC before the encoder and connectors, so let's move our call around.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:58 +0100, Maxime Ripard wrote:
It makes more sense to register the CRTC before the encoder and connectors, so let's move our call around.
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
Both users of vc4_crtc_init need the same extra initialization to set the pointer to the platform_device and the CRTC data. Since it's mandatory, let's make them both arguments of vc4_crtc_init().
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_crtc.c | 17 +++++++++++------ drivers/gpu/drm/vc4/vc4_drv.h | 6 ++++-- drivers/gpu/drm/vc4/vc4_txp.c | 7 ++----- 3 files changed, 17 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 0108613e79d5..1ea190dffe87 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -1278,9 +1278,12 @@ static void vc4_set_crtc_possible_masks(struct drm_device *drm, } }
-int vc4_crtc_init(struct drm_device *drm, struct vc4_crtc *vc4_crtc, +int vc4_crtc_init(struct drm_device *drm, struct platform_device *pdev, + struct vc4_crtc *vc4_crtc, + const struct vc4_crtc_data *data, const struct drm_crtc_funcs *crtc_funcs, - const struct drm_crtc_helper_funcs *crtc_helper_funcs) + const struct drm_crtc_helper_funcs *crtc_helper_funcs, + bool feeds_txp) { struct vc4_dev *vc4 = to_vc4_dev(drm); struct drm_crtc *crtc = &vc4_crtc->base; @@ -1300,6 +1303,9 @@ int vc4_crtc_init(struct drm_device *drm, struct vc4_crtc *vc4_crtc, return PTR_ERR(primary_plane); }
+ vc4_crtc->data = data; + vc4_crtc->pdev = pdev; + vc4_crtc->feeds_txp = feeds_txp; spin_lock_init(&vc4_crtc->irq_lock); ret = drmm_crtc_init_with_planes(drm, crtc, primary_plane, NULL, crtc_funcs, NULL); @@ -1345,8 +1351,6 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data) pv_data = of_device_get_match_data(dev); if (!pv_data) return -ENODEV; - vc4_crtc->data = &pv_data->base; - vc4_crtc->pdev = pdev;
vc4_crtc->regs = vc4_ioremap_regs(pdev, 0); if (IS_ERR(vc4_crtc->regs)) @@ -1356,8 +1360,9 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data) vc4_crtc->regset.regs = crtc_regs; vc4_crtc->regset.nregs = ARRAY_SIZE(crtc_regs);
- ret = vc4_crtc_init(drm, vc4_crtc, - &vc4_crtc_funcs, &vc4_crtc_helper_funcs); + ret = vc4_crtc_init(drm, pdev, vc4_crtc, &pv_data->base, + &vc4_crtc_funcs, &vc4_crtc_helper_funcs, + false); if (ret) return ret; vc4_set_crtc_possible_masks(drm, crtc); diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 31228d4cf080..599d24f1f55a 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -886,9 +886,11 @@ int vc4_bo_debugfs_init(struct drm_minor *minor); /* vc4_crtc.c */ extern struct platform_driver vc4_crtc_driver; int vc4_crtc_disable_at_boot(struct drm_crtc *crtc); -int vc4_crtc_init(struct drm_device *drm, struct vc4_crtc *vc4_crtc, +int vc4_crtc_init(struct drm_device *drm, struct platform_device *pdev, + struct vc4_crtc *vc4_crtc, const struct vc4_crtc_data *data, const struct drm_crtc_funcs *crtc_funcs, - const struct drm_crtc_helper_funcs *crtc_helper_funcs); + const struct drm_crtc_helper_funcs *crtc_helper_funcs, + bool feeds_txp); int vc4_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb, struct drm_pending_vblank_event *event, diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index c2a6bea6fd96..841da240d93a 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -509,15 +509,12 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data) return PTR_ERR(txp->regs);
vc4_crtc = &txp->base; - vc4_crtc->pdev = pdev; - vc4_crtc->data = &vc4_txp_crtc_data; - vc4_crtc->feeds_txp = true; vc4_crtc->regset.base = txp->regs; vc4_crtc->regset.regs = txp_regs; vc4_crtc->regset.nregs = ARRAY_SIZE(txp_regs);
- ret = vc4_crtc_init(drm, vc4_crtc, - &vc4_txp_crtc_funcs, &vc4_txp_crtc_helper_funcs); + ret = vc4_crtc_init(drm, pdev, vc4_crtc, &vc4_txp_crtc_data, + &vc4_txp_crtc_funcs, &vc4_txp_crtc_helper_funcs, true); if (ret) return ret;
On 11/23/22 16:25, Maxime Ripard wrote:
Both users of vc4_crtc_init need the same extra initialization to set the pointer to the platform_device and the CRTC data. Since it's mandatory, let's make them both arguments of vc4_crtc_init().
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:25:59 +0100, Maxime Ripard wrote:
Both users of vc4_crtc_init need the same extra initialization to set the pointer to the platform_device and the CRTC data. Since it's mandatory, let's make them both arguments of vc4_crtc_init().
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
The current vc4_crtc_init() helper assumes that we will be using hardware planes and calls vc4_plane_init().
While it's a reasonable assumption, we'll want to mock the plane and thus provide our own. Let's create a helper that will take the plane as an argument.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_crtc.c | 52 +++++++++++++++++++++++++++--------------- drivers/gpu/drm/vc4/vc4_drv.h | 6 +++++ 2 files changed, 39 insertions(+), 19 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 1ea190dffe87..340c39921bce 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -1278,31 +1278,20 @@ static void vc4_set_crtc_possible_masks(struct drm_device *drm, } }
-int vc4_crtc_init(struct drm_device *drm, struct platform_device *pdev, - struct vc4_crtc *vc4_crtc, - const struct vc4_crtc_data *data, - const struct drm_crtc_funcs *crtc_funcs, - const struct drm_crtc_helper_funcs *crtc_helper_funcs, - bool feeds_txp) +int __vc4_crtc_init(struct drm_device *drm, + struct platform_device *pdev, + struct vc4_crtc *vc4_crtc, + const struct vc4_crtc_data *data, + struct drm_plane *primary_plane, + const struct drm_crtc_funcs *crtc_funcs, + const struct drm_crtc_helper_funcs *crtc_helper_funcs, + bool feeds_txp) { struct vc4_dev *vc4 = to_vc4_dev(drm); struct drm_crtc *crtc = &vc4_crtc->base; - struct drm_plane *primary_plane; unsigned int i; int ret;
- /* For now, we create just the primary and the legacy cursor - * planes. We should be able to stack more planes on easily, - * but to do that we would need to compute the bandwidth - * requirement of the plane configuration, and reject ones - * that will take too much. - */ - primary_plane = vc4_plane_init(drm, DRM_PLANE_TYPE_PRIMARY, 0); - if (IS_ERR(primary_plane)) { - dev_err(drm->dev, "failed to construct primary plane\n"); - return PTR_ERR(primary_plane); - } - vc4_crtc->data = data; vc4_crtc->pdev = pdev; vc4_crtc->feeds_txp = feeds_txp; @@ -1334,6 +1323,31 @@ int vc4_crtc_init(struct drm_device *drm, struct platform_device *pdev, return 0; }
+int vc4_crtc_init(struct drm_device *drm, struct platform_device *pdev, + struct vc4_crtc *vc4_crtc, + const struct vc4_crtc_data *data, + const struct drm_crtc_funcs *crtc_funcs, + const struct drm_crtc_helper_funcs *crtc_helper_funcs, + bool feeds_txp) +{ + struct drm_plane *primary_plane; + + /* For now, we create just the primary and the legacy cursor + * planes. We should be able to stack more planes on easily, + * but to do that we would need to compute the bandwidth + * requirement of the plane configuration, and reject ones + * that will take too much. + */ + primary_plane = vc4_plane_init(drm, DRM_PLANE_TYPE_PRIMARY, 0); + if (IS_ERR(primary_plane)) { + dev_err(drm->dev, "failed to construct primary plane\n"); + return PTR_ERR(primary_plane); + } + + return __vc4_crtc_init(drm, pdev, vc4_crtc, data, primary_plane, + crtc_funcs, crtc_helper_funcs, feeds_txp); +} + static int vc4_crtc_bind(struct device *dev, struct device *master, void *data) { struct platform_device *pdev = to_platform_device(dev); diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 599d24f1f55a..01ca2b25d2e3 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -886,6 +886,12 @@ int vc4_bo_debugfs_init(struct drm_minor *minor); /* vc4_crtc.c */ extern struct platform_driver vc4_crtc_driver; int vc4_crtc_disable_at_boot(struct drm_crtc *crtc); +int __vc4_crtc_init(struct drm_device *drm, struct platform_device *pdev, + struct vc4_crtc *vc4_crtc, const struct vc4_crtc_data *data, + struct drm_plane *primary_plane, + const struct drm_crtc_funcs *crtc_funcs, + const struct drm_crtc_helper_funcs *crtc_helper_funcs, + bool feeds_txp); int vc4_crtc_init(struct drm_device *drm, struct platform_device *pdev, struct vc4_crtc *vc4_crtc, const struct vc4_crtc_data *data, const struct drm_crtc_funcs *crtc_funcs,
On 11/23/22 16:26, Maxime Ripard wrote:
The current vc4_crtc_init() helper assumes that we will be using hardware planes and calls vc4_plane_init().
While it's a reasonable assumption, we'll want to mock the plane and thus provide our own. Let's create a helper that will take the plane as an argument.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
We'll need a function that looks up an encoder by its vc4_encoder_type. Such a function is already present in the CRTC code, so let's make it public so that we can reuse it in the unit tests.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_crtc.c | 17 +---------------- drivers/gpu/drm/vc4/vc4_drv.h | 16 ++++++++++++++++ 2 files changed, 17 insertions(+), 16 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 340c39921bce..8bc30ad0904b 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -486,21 +486,6 @@ static int vc4_crtc_disable(struct drm_crtc *crtc, return 0; }
-static struct drm_encoder *vc4_crtc_get_encoder_by_type(struct drm_crtc *crtc, - enum vc4_encoder_type type) -{ - struct drm_encoder *encoder; - - drm_for_each_encoder(encoder, crtc->dev) { - struct vc4_encoder *vc4_encoder = to_vc4_encoder(encoder); - - if (vc4_encoder->type == type) - return encoder; - } - - return NULL; -} - int vc4_crtc_disable_at_boot(struct drm_crtc *crtc) { struct drm_device *drm = crtc->dev; @@ -536,7 +521,7 @@ int vc4_crtc_disable_at_boot(struct drm_crtc *crtc)
pv_data = vc4_crtc_to_vc4_pv_data(vc4_crtc); encoder_type = pv_data->encoder_types[encoder_sel]; - encoder = vc4_crtc_get_encoder_by_type(crtc, encoder_type); + encoder = vc4_find_encoder_by_type(drm, encoder_type); if (WARN_ON(!encoder)) return 0;
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 01ca2b25d2e3..dca7be3fccb5 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -495,6 +495,22 @@ to_vc4_encoder(const struct drm_encoder *encoder) return container_of(encoder, struct vc4_encoder, base); }
+static inline +struct drm_encoder *vc4_find_encoder_by_type(struct drm_device *drm, + enum vc4_encoder_type type) +{ + struct drm_encoder *encoder; + + drm_for_each_encoder(encoder, drm) { + struct vc4_encoder *vc4_encoder = to_vc4_encoder(encoder); + + if (vc4_encoder->type == type) + return encoder; + } + + return NULL; +} + struct vc4_crtc_data { const char *debugfs_name;
On 11/23/22 16:26, Maxime Ripard wrote:
We'll need a function that looks up an encoder by its vc4_encoder_type. Such a function is already present in the CRTC code, so let's make it public so that we can reuse it in the unit tests.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
It's fairly hard to figure out the instance of the CRTC affected by an atomic change using the default name.
Since we can provide our own to the CRTC initialization functions, let's do so to make the debugging sessions easier.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_crtc.c | 10 +++++++++- drivers/gpu/drm/vc4/vc4_drv.h | 2 ++ drivers/gpu/drm/vc4/vc4_txp.c | 1 + 3 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 8bc30ad0904b..59e473059fa2 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -1118,6 +1118,7 @@ static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = {
static const struct vc4_pv_data bcm2835_pv0_data = { .base = { + .name = "pixelvalve-0", .debugfs_name = "crtc0_regs", .hvs_available_channels = BIT(0), .hvs_output = 0, @@ -1132,6 +1133,7 @@ static const struct vc4_pv_data bcm2835_pv0_data = {
static const struct vc4_pv_data bcm2835_pv1_data = { .base = { + .name = "pixelvalve-1", .debugfs_name = "crtc1_regs", .hvs_available_channels = BIT(2), .hvs_output = 2, @@ -1146,6 +1148,7 @@ static const struct vc4_pv_data bcm2835_pv1_data = {
static const struct vc4_pv_data bcm2835_pv2_data = { .base = { + .name = "pixelvalve-2", .debugfs_name = "crtc2_regs", .hvs_available_channels = BIT(1), .hvs_output = 1, @@ -1160,6 +1163,7 @@ static const struct vc4_pv_data bcm2835_pv2_data = {
static const struct vc4_pv_data bcm2711_pv0_data = { .base = { + .name = "pixelvalve-0", .debugfs_name = "crtc0_regs", .hvs_available_channels = BIT(0), .hvs_output = 0, @@ -1174,6 +1178,7 @@ static const struct vc4_pv_data bcm2711_pv0_data = {
static const struct vc4_pv_data bcm2711_pv1_data = { .base = { + .name = "pixelvalve-1", .debugfs_name = "crtc1_regs", .hvs_available_channels = BIT(0) | BIT(1) | BIT(2), .hvs_output = 3, @@ -1188,6 +1193,7 @@ static const struct vc4_pv_data bcm2711_pv1_data = {
static const struct vc4_pv_data bcm2711_pv2_data = { .base = { + .name = "pixelvalve-2", .debugfs_name = "crtc2_regs", .hvs_available_channels = BIT(0) | BIT(1) | BIT(2), .hvs_output = 4, @@ -1201,6 +1207,7 @@ static const struct vc4_pv_data bcm2711_pv2_data = {
static const struct vc4_pv_data bcm2711_pv3_data = { .base = { + .name = "pixelvalve-3", .debugfs_name = "crtc3_regs", .hvs_available_channels = BIT(1), .hvs_output = 1, @@ -1214,6 +1221,7 @@ static const struct vc4_pv_data bcm2711_pv3_data = {
static const struct vc4_pv_data bcm2711_pv4_data = { .base = { + .name = "pixelvalve-4", .debugfs_name = "crtc4_regs", .hvs_available_channels = BIT(0) | BIT(1) | BIT(2), .hvs_output = 5, @@ -1282,7 +1290,7 @@ int __vc4_crtc_init(struct drm_device *drm, vc4_crtc->feeds_txp = feeds_txp; spin_lock_init(&vc4_crtc->irq_lock); ret = drmm_crtc_init_with_planes(drm, crtc, primary_plane, NULL, - crtc_funcs, NULL); + crtc_funcs, data->name); if (ret) return ret;
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index dca7be3fccb5..54352db48476 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -512,6 +512,8 @@ struct drm_encoder *vc4_find_encoder_by_type(struct drm_device *drm, }
struct vc4_crtc_data { + const char *name; + const char *debugfs_name;
/* Bitmask of channels (FIFOs) of the HVS that the output can source from */ diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index 841da240d93a..4f7ce5d3e8ad 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -480,6 +480,7 @@ static irqreturn_t vc4_txp_interrupt(int irq, void *data) }
static const struct vc4_crtc_data vc4_txp_crtc_data = { + .name = "txp", .debugfs_name = "txp_regs", .hvs_available_channels = BIT(2), .hvs_output = 2,
On 11/23/22 16:26, Maxime Ripard wrote:
It's fairly hard to figure out the instance of the CRTC affected by an atomic change using the default name.
Since we can provide our own to the CRTC initialization functions, let's do so to make the debugging sessions easier.
Signed-off-by: Maxime Ripard maxime@cerno.tech
drivers/gpu/drm/vc4/vc4_crtc.c | 10 +++++++++- drivers/gpu/drm/vc4/vc4_drv.h | 2 ++ drivers/gpu/drm/vc4/vc4_txp.c | 1 + 3 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 8bc30ad0904b..59e473059fa2 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -1118,6 +1118,7 @@ static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = { static const struct vc4_pv_data bcm2835_pv0_data = { .base = {
.name = "pixelvalve-0",
I wonder if would make sense to add the SoC name too, but either way:
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
On Wed, 23 Nov 2022 16:26:02 +0100, Maxime Ripard wrote:
It's fairly hard to figure out the instance of the CRTC affected by an atomic change using the default name.
Since we can provide our own to the CRTC initialization functions, let's do so to make the debugging sessions easier.
[...]
Applied to drm/drm-misc (drm-misc-next).
Thanks! Maxime
We'll need to initialize the HVS structure without a backing device to create a mock we'll use for testing.
Split the structure initialization part into a separate function.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_drv.h | 1 + drivers/gpu/drm/vc4/vc4_hvs.c | 81 +++++++++++++++++++++++++------------------ 2 files changed, 48 insertions(+), 34 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 54352db48476..e0be7a81a24a 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -1009,6 +1009,7 @@ void vc4_irq_reset(struct drm_device *dev);
/* vc4_hvs.c */ extern struct platform_driver vc4_hvs_driver; +struct vc4_hvs *__vc4_hvs_alloc(struct vc4_dev *vc4, struct platform_device *pdev); void vc4_hvs_stop_channel(struct vc4_hvs *hvs, unsigned int output); int vc4_hvs_get_fifo_from_output(struct vc4_hvs *hvs, unsigned int output); u8 vc4_hvs_get_fifo_frame_count(struct vc4_hvs *hvs, unsigned int fifo); diff --git a/drivers/gpu/drm/vc4/vc4_hvs.c b/drivers/gpu/drm/vc4/vc4_hvs.c index c4453a5ae163..94c29f8547bb 100644 --- a/drivers/gpu/drm/vc4/vc4_hvs.c +++ b/drivers/gpu/drm/vc4/vc4_hvs.c @@ -768,22 +768,60 @@ int vc4_hvs_debugfs_init(struct drm_minor *minor) return 0; }
-static int vc4_hvs_bind(struct device *dev, struct device *master, void *data) +struct vc4_hvs *__vc4_hvs_alloc(struct vc4_dev *vc4, struct platform_device *pdev) { - struct platform_device *pdev = to_platform_device(dev); - struct drm_device *drm = dev_get_drvdata(master); - struct vc4_dev *vc4 = to_vc4_dev(drm); - struct vc4_hvs *hvs = NULL; - int ret; - u32 dispctrl; - u32 reg; + struct drm_device *drm = &vc4->base; + struct vc4_hvs *hvs;
hvs = drmm_kzalloc(drm, sizeof(*hvs), GFP_KERNEL); if (!hvs) - return -ENOMEM; + return ERR_PTR(-ENOMEM); + hvs->vc4 = vc4; hvs->pdev = pdev;
+ spin_lock_init(&hvs->mm_lock); + + /* Set up the HVS display list memory manager. We never + * overwrite the setup from the bootloader (just 128b out of + * our 16K), since we don't want to scramble the screen when + * transitioning from the firmware's boot setup to runtime. + */ + drm_mm_init(&hvs->dlist_mm, + HVS_BOOTLOADER_DLIST_END, + (SCALER_DLIST_SIZE >> 2) - HVS_BOOTLOADER_DLIST_END); + + /* Set up the HVS LBM memory manager. We could have some more + * complicated data structure that allowed reuse of LBM areas + * between planes when they don't overlap on the screen, but + * for now we just allocate globally. + */ + if (!vc4->is_vc5) + /* 48k words of 2x12-bit pixels */ + drm_mm_init(&hvs->lbm_mm, 0, 48 * 1024); + else + /* 60k words of 4x12-bit pixels */ + drm_mm_init(&hvs->lbm_mm, 0, 60 * 1024); + + vc4->hvs = hvs; + + return hvs; +} + +static int vc4_hvs_bind(struct device *dev, struct device *master, void *data) +{ + struct platform_device *pdev = to_platform_device(dev); + struct drm_device *drm = dev_get_drvdata(master); + struct vc4_dev *vc4 = to_vc4_dev(drm); + struct vc4_hvs *hvs = NULL; + int ret; + u32 dispctrl; + u32 reg; + + hvs = __vc4_hvs_alloc(vc4, NULL); + if (IS_ERR(hvs)) + return PTR_ERR(hvs); + hvs->regs = vc4_ioremap_regs(pdev, 0); if (IS_ERR(hvs->regs)) return PTR_ERR(hvs->regs); @@ -835,29 +873,6 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data) else hvs->dlist = hvs->regs + SCALER5_DLIST_START;
- spin_lock_init(&hvs->mm_lock); - - /* Set up the HVS display list memory manager. We never - * overwrite the setup from the bootloader (just 128b out of - * our 16K), since we don't want to scramble the screen when - * transitioning from the firmware's boot setup to runtime. - */ - drm_mm_init(&hvs->dlist_mm, - HVS_BOOTLOADER_DLIST_END, - (SCALER_DLIST_SIZE >> 2) - HVS_BOOTLOADER_DLIST_END); - - /* Set up the HVS LBM memory manager. We could have some more - * complicated data structure that allowed reuse of LBM areas - * between planes when they don't overlap on the screen, but - * for now we just allocate globally. - */ - if (!vc4->is_vc5) - /* 48k words of 2x12-bit pixels */ - drm_mm_init(&hvs->lbm_mm, 0, 48 * 1024); - else - /* 60k words of 4x12-bit pixels */ - drm_mm_init(&hvs->lbm_mm, 0, 60 * 1024); - /* Upload filter kernels. We only have the one for now, so we * keep it around for the lifetime of the driver. */ @@ -867,8 +882,6 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data) if (ret) return ret;
- vc4->hvs = hvs; - reg = HVS_READ(SCALER_DISPECTRL); reg &= ~SCALER_DISPECTRL_DSP2_MUX_MASK; HVS_WRITE(SCALER_DISPECTRL,
On 11/23/22 16:26, Maxime Ripard wrote:
We'll need to initialize the HVS structure without a backing device to create a mock we'll use for testing.
Split the structure initialization part into a separate function.
Signed-off-by: Maxime Ripard maxime@cerno.tech
Reviewed-by: Javier Martinez Canillas javierm@redhat.com
In order to test the current atomic_check hooks we need to have a DRM device that has roughly the same capabilities and layout that the actual hardware. We'll also need a bunch of functions to create arbitrary atomic states.
Let's create some helpers to create a device that behaves like the real one, and some helpers to maintain the atomic state we want to check.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/Kconfig | 15 +++ drivers/gpu/drm/vc4/Makefile | 1 + drivers/gpu/drm/vc4/tests/.kunitconfig | 14 +++ drivers/gpu/drm/vc4/tests/Makefile | 7 ++ drivers/gpu/drm/vc4/tests/vc4_mock.c | 174 ++++++++++++++++++++++++++++ drivers/gpu/drm/vc4/tests/vc4_mock.h | 58 ++++++++++ drivers/gpu/drm/vc4/tests/vc4_mock_crtc.c | 39 +++++++ drivers/gpu/drm/vc4/tests/vc4_mock_output.c | 97 ++++++++++++++++ drivers/gpu/drm/vc4/tests/vc4_mock_plane.c | 45 +++++++ drivers/gpu/drm/vc4/vc4_crtc.c | 20 ++-- drivers/gpu/drm/vc4/vc4_drv.c | 4 +- drivers/gpu/drm/vc4/vc4_drv.h | 13 +++ drivers/gpu/drm/vc4/vc4_txp.c | 2 +- 13 files changed, 476 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/vc4/Kconfig b/drivers/gpu/drm/vc4/Kconfig index 246305d17a52..eb5e8367da67 100644 --- a/drivers/gpu/drm/vc4/Kconfig +++ b/drivers/gpu/drm/vc4/Kconfig @@ -34,3 +34,18 @@ config DRM_VC4_HDMI_CEC help Choose this option if you have a Broadcom VC4 GPU and want to use CEC. + +config DRM_VC4_KUNIT_TEST + tristate "KUnit tests for VC4" if !KUNIT_ALL_TESTS + depends on DRM_VC4 && KUNIT + default KUNIT_ALL_TESTS + help + This builds unit tests for the VC4 DRM/KMS driver. This option is + not useful for distributions or general kernels, but only for kernel + developers working on the VC4 driver. + + For more information on KUnit and unit tests in general, + please refer to the KUnit documentation in + Documentation/dev-tools/kunit/. + + If in doubt, say "N". diff --git a/drivers/gpu/drm/vc4/Makefile b/drivers/gpu/drm/vc4/Makefile index d0163e18e9ca..8984b2b5da77 100644 --- a/drivers/gpu/drm/vc4/Makefile +++ b/drivers/gpu/drm/vc4/Makefile @@ -28,3 +28,4 @@ vc4-y := \ vc4-$(CONFIG_DEBUG_FS) += vc4_debugfs.o
obj-$(CONFIG_DRM_VC4) += vc4.o +obj-$(CONFIG_DRM_VC4_KUNIT_TEST) += tests/ diff --git a/drivers/gpu/drm/vc4/tests/.kunitconfig b/drivers/gpu/drm/vc4/tests/.kunitconfig new file mode 100644 index 000000000000..bfd9899ed6e0 --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/.kunitconfig @@ -0,0 +1,14 @@ +CONFIG_ARCH_BCM=y +CONFIG_ARCH_BCM2835=y +CONFIG_BCM2835_MBOX=y +CONFIG_KUNIT=y +CONFIG_DRM=y +CONFIG_DRM_KUNIT_TEST=y +CONFIG_DRM_VC4=y +CONFIG_DRM_VC4_KUNIT_TEST=y +CONFIG_MAILBOX=y +CONFIG_RASPBERRYPI_FIRMWARE=y +CONFIG_SND=y +CONFIG_SND_SOC=y +CONFIG_SOUND=y +CONFIG_COMMON_CLK=y diff --git a/drivers/gpu/drm/vc4/tests/Makefile b/drivers/gpu/drm/vc4/tests/Makefile new file mode 100644 index 000000000000..3e8a7b4f9017 --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/Makefile @@ -0,0 +1,7 @@ +# SPDX-License-Identifier: GPL-2.0 + +obj-$(CONFIG_DRM_VC4_KUNIT_TEST) += \ + vc4_mock.o \ + vc4_mock_crtc.o \ + vc4_mock_output.o \ + vc4_mock_plane.o diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock.c b/drivers/gpu/drm/vc4/tests/vc4_mock.c new file mode 100644 index 000000000000..f4c2213ea0e2 --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/vc4_mock.c @@ -0,0 +1,174 @@ +#include <drm/drm_drv.h> + +#include <kunit/test.h> + +#include "../../tests/drm_kunit_helpers.h" + +#include "vc4_mock.h" + +struct vc4_mock_output_desc { + enum vc4_encoder_type vc4_encoder_type; + unsigned int encoder_type; + unsigned int connector_type; +}; + +#define VC4_MOCK_OUTPUT_DESC(_vc4_type, _etype, _ctype) \ + { \ + .vc4_encoder_type = _vc4_type, \ + .encoder_type = _etype, \ + .connector_type = _ctype, \ + } + +struct vc4_mock_pipe_desc { + const struct vc4_crtc_data *data; + const struct vc4_mock_output_desc *outputs; + unsigned int noutputs; +}; + +#define VC4_MOCK_CRTC_DESC(_data, ...) \ + { \ + .data = _data, \ + .outputs = (struct vc4_mock_output_desc[]) { __VA_ARGS__ }, \ + .noutputs = sizeof((struct vc4_mock_output_desc[]) { __VA_ARGS__ }) / \ + sizeof(struct vc4_mock_output_desc), \ + } + +#define VC4_MOCK_PIXELVALVE_DESC(_data, ...) \ + VC4_MOCK_CRTC_DESC(&(_data)->base, __VA_ARGS__) + +struct vc4_mock_desc { + const struct vc4_mock_pipe_desc *pipes; + unsigned int npipes; +}; + +#define VC4_MOCK_DESC(...) \ + { \ + .pipes = (struct vc4_mock_pipe_desc[]) { __VA_ARGS__ }, \ + .npipes = sizeof((struct vc4_mock_pipe_desc[]) { __VA_ARGS__ }) / \ + sizeof(struct vc4_mock_pipe_desc), \ + } + +static const struct vc4_mock_desc vc5_mock = + VC4_MOCK_DESC( + VC4_MOCK_CRTC_DESC(&vc4_txp_crtc_data, + VC4_MOCK_OUTPUT_DESC(VC4_ENCODER_TYPE_TXP, + DRM_MODE_ENCODER_VIRTUAL, + DRM_MODE_CONNECTOR_WRITEBACK)), + VC4_MOCK_PIXELVALVE_DESC(&bcm2711_pv0_data, + VC4_MOCK_OUTPUT_DESC(VC4_ENCODER_TYPE_DSI0, + DRM_MODE_ENCODER_DSI, + DRM_MODE_CONNECTOR_DSI), + VC4_MOCK_OUTPUT_DESC(VC4_ENCODER_TYPE_DPI, + DRM_MODE_ENCODER_DPI, + DRM_MODE_CONNECTOR_DPI)), + VC4_MOCK_PIXELVALVE_DESC(&bcm2711_pv1_data, + VC4_MOCK_OUTPUT_DESC(VC4_ENCODER_TYPE_DSI1, + DRM_MODE_ENCODER_DSI, + DRM_MODE_CONNECTOR_DSI)), + VC4_MOCK_PIXELVALVE_DESC(&bcm2711_pv2_data, + VC4_MOCK_OUTPUT_DESC(VC4_ENCODER_TYPE_HDMI0, + DRM_MODE_ENCODER_TMDS, + DRM_MODE_CONNECTOR_HDMIA)), + VC4_MOCK_PIXELVALVE_DESC(&bcm2711_pv3_data, + VC4_MOCK_OUTPUT_DESC(VC4_ENCODER_TYPE_VEC, + DRM_MODE_ENCODER_TVDAC, + DRM_MODE_CONNECTOR_Composite)), + VC4_MOCK_PIXELVALVE_DESC(&bcm2711_pv4_data, + VC4_MOCK_OUTPUT_DESC(VC4_ENCODER_TYPE_HDMI1, + DRM_MODE_ENCODER_TMDS, + DRM_MODE_CONNECTOR_HDMIA)), +); + +static int __build_one_pipe(struct kunit *test, struct drm_device *drm, + const struct vc4_mock_pipe_desc *pipe) +{ + struct vc4_dummy_plane *dummy_plane; + struct drm_plane *plane; + struct vc4_dummy_crtc *dummy_crtc; + struct drm_crtc *crtc; + unsigned int i; + + dummy_plane = vc4_dummy_plane(test, drm, DRM_PLANE_TYPE_PRIMARY); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dummy_plane); + + plane = &dummy_plane->plane.base; + dummy_crtc = vc4_mock_pv(test, drm, plane, pipe->data); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dummy_crtc); + + crtc = &dummy_crtc->crtc.base; + for (i = 0; i < pipe->noutputs; i++) { + const struct vc4_mock_output_desc *mock_output = &pipe->outputs[i]; + struct vc4_dummy_output *dummy_output; + + dummy_output = vc4_dummy_output(test, drm, crtc, + mock_output->vc4_encoder_type, + mock_output->encoder_type, + mock_output->connector_type); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dummy_output); + } + + return 0; +} + +static int __build_mock(struct kunit *test, struct drm_device *drm, + const struct vc4_mock_desc *mock) +{ + + unsigned int i; + + for (i = 0; i < mock->npipes; i++) { + const struct vc4_mock_pipe_desc *pipe = &mock->pipes[i]; + int ret; + + ret = __build_one_pipe(test, drm, pipe); + KUNIT_ASSERT_EQ(test, ret, 0); + } + + return 0; +} + +static struct vc4_dev *__mock_device(struct kunit *test, bool is_vc5) +{ + struct drm_device *drm; + const struct drm_driver *drv = is_vc5 ? &vc5_drm_driver : &vc4_drm_driver; + const struct vc4_mock_desc *desc = is_vc5 ? &vc5_mock : NULL; + struct vc4_dev *vc4; + struct device *dev; + int ret; + + dev = drm_kunit_helper_alloc_device(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + vc4 = drm_kunit_helper_alloc_drm_device_with_driver(test, dev, + struct vc4_dev, base, + drv); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vc4); + + vc4->dev = dev; + vc4->is_vc5 = is_vc5; + + vc4->hvs = __vc4_hvs_alloc(vc4, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vc4->hvs); + + drm = &vc4->base; + ret = __build_mock(test, drm, desc); + KUNIT_ASSERT_EQ(test, ret, 0); + + ret = vc4_kms_load(drm); + KUNIT_ASSERT_EQ(test, ret, 0); + + ret = drm_dev_register(drm, 0); + KUNIT_ASSERT_EQ(test, ret, 0); + + return vc4; +} + +struct vc4_dev *vc4_mock_device(struct kunit *test) +{ + return __mock_device(test, false); +} + +struct vc4_dev *vc5_mock_device(struct kunit *test) +{ + return __mock_device(test, true); +} diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock.h b/drivers/gpu/drm/vc4/tests/vc4_mock.h new file mode 100644 index 000000000000..144af2ed6c40 --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/vc4_mock.h @@ -0,0 +1,58 @@ +#ifndef VC4_MOCK_H_ +#define VC4_MOCK_H_ + +#include "../vc4_drv.h" + +static inline +struct drm_crtc *vc4_find_crtc_for_encoder(struct kunit *test, + struct drm_device *drm, + struct drm_encoder *encoder) +{ + struct drm_crtc *crtc; + + KUNIT_ASSERT_EQ(test, hweight32(encoder->possible_crtcs), 1); + + drm_for_each_crtc(crtc, drm) + if (encoder->possible_crtcs & drm_crtc_mask(crtc)) + return crtc; + + return NULL; +} + +struct vc4_dummy_plane { + struct vc4_plane plane; +}; + +struct vc4_dummy_plane *vc4_dummy_plane(struct kunit *test, + struct drm_device *drm, + enum drm_plane_type type); + +struct vc4_dummy_crtc { + struct vc4_crtc crtc; +}; + +struct vc4_dummy_crtc *vc4_mock_pv(struct kunit *test, + struct drm_device *drm, + struct drm_plane *plane, + const struct vc4_crtc_data *data); + +struct vc4_dummy_output { + struct vc4_encoder encoder; + struct drm_connector connector; +}; + +struct vc4_dummy_output *vc4_dummy_output(struct kunit *test, + struct drm_device *drm, + struct drm_crtc *crtc, + enum vc4_encoder_type vc4_encoder_type, + unsigned int kms_encoder_type, + unsigned int connector_type); + +struct vc4_dev *vc4_mock_device(struct kunit *test); +struct vc4_dev *vc5_mock_device(struct kunit *test); + +int vc4_mock_atomic_add_output(struct kunit *test, struct drm_device *drm, + enum vc4_encoder_type type, + struct drm_atomic_state *state); + +#endif // VC4_MOCK_H_ diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock_crtc.c b/drivers/gpu/drm/vc4/tests/vc4_mock_crtc.c new file mode 100644 index 000000000000..213622f9fba2 --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/vc4_mock_crtc.c @@ -0,0 +1,39 @@ +#include <drm/drm_atomic_state_helper.h> +#include <drm/drm_modeset_helper_vtables.h> + +#include <kunit/test.h> + +#include "vc4_mock.h" + +static const struct drm_crtc_helper_funcs vc4_dummy_crtc_helper_funcs = { + .atomic_check = vc4_crtc_atomic_check, +}; + +static const struct drm_crtc_funcs vc4_dummy_crtc_funcs = { + .atomic_destroy_state = vc4_crtc_destroy_state, + .atomic_duplicate_state = vc4_crtc_duplicate_state, + .reset = vc4_crtc_reset, +}; + +struct vc4_dummy_crtc *vc4_mock_pv(struct kunit *test, + struct drm_device *drm, + struct drm_plane *plane, + const struct vc4_crtc_data *data) +{ + struct vc4_dummy_crtc *dummy_crtc; + struct vc4_crtc *vc4_crtc; + int ret; + + dummy_crtc = kunit_kzalloc(test, sizeof(*dummy_crtc), GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, dummy_crtc); + + vc4_crtc = &dummy_crtc->crtc; + ret = __vc4_crtc_init(drm, NULL, + vc4_crtc, data, plane, + &vc4_dummy_crtc_funcs, + &vc4_dummy_crtc_helper_funcs, + false); + KUNIT_ASSERT_EQ(test, ret, 0); + + return dummy_crtc; +} diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock_output.c b/drivers/gpu/drm/vc4/tests/vc4_mock_output.c new file mode 100644 index 000000000000..18bd088530fe --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/vc4_mock_output.c @@ -0,0 +1,97 @@ +#include <drm/drm_atomic_state_helper.h> +#include <drm/drm_atomic_uapi.h> +#include <drm/drm_connector.h> +#include <drm/drm_crtc.h> +#include <drm/drm_encoder.h> +#include <drm/drm_modeset_helper_vtables.h> + +#include <kunit/test.h> + +#include "vc4_mock.h" + +static const struct drm_connector_helper_funcs vc4_dummy_connector_helper_funcs = { +}; + +static const struct drm_connector_funcs vc4_dummy_connector_funcs = { + .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, + .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, + .reset = drm_atomic_helper_connector_reset, +}; + +struct vc4_dummy_output *vc4_dummy_output(struct kunit *test, + struct drm_device *drm, + struct drm_crtc *crtc, + enum vc4_encoder_type vc4_encoder_type, + unsigned int kms_encoder_type, + unsigned int connector_type) +{ + struct vc4_dummy_output *dummy_output; + struct drm_connector *conn; + struct drm_encoder *enc; + int ret; + + dummy_output = kunit_kzalloc(test, sizeof(*dummy_output), GFP_KERNEL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dummy_output); + dummy_output->encoder.type = vc4_encoder_type; + + enc = &dummy_output->encoder.base; + ret = drmm_encoder_init(drm, enc, + NULL, + kms_encoder_type, + NULL); + KUNIT_ASSERT_EQ(test, ret, 0); + enc->possible_crtcs = drm_crtc_mask(crtc); + + conn = &dummy_output->connector; + ret = drmm_connector_init(drm, conn, + &vc4_dummy_connector_funcs, + connector_type, + NULL); + KUNIT_ASSERT_EQ(test, ret, 0); + + drm_connector_helper_add(conn, &vc4_dummy_connector_helper_funcs); + drm_connector_attach_encoder(conn, enc); + + return dummy_output; +} + +static const struct drm_display_mode default_mode = { + DRM_SIMPLE_MODE(640, 480, 64, 48) +}; + +int vc4_mock_atomic_add_output(struct kunit *test, struct drm_device *drm, + enum vc4_encoder_type type, + struct drm_atomic_state *state) +{ + struct vc4_dummy_output *output; + struct drm_connector *conn; + struct drm_connector_state *conn_state; + struct drm_encoder *encoder; + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; + int ret; + + encoder = vc4_find_encoder_by_type(drm, type); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder); + + crtc = vc4_find_crtc_for_encoder(test, drm, encoder); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc); + + output = container_of(encoder, struct vc4_dummy_output, encoder.base); + conn = &output->connector; + conn_state = drm_atomic_get_connector_state(state, conn); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state); + + ret = drm_atomic_set_crtc_for_connector(conn_state, crtc); + KUNIT_EXPECT_EQ(test, ret, 0); + + crtc_state = drm_atomic_get_crtc_state(state, crtc); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_state); + + ret = drm_atomic_set_mode_for_crtc(crtc_state, &default_mode); + KUNIT_EXPECT_EQ(test, ret, 0); + + crtc_state->active = true; + + return 0; +} diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock_plane.c b/drivers/gpu/drm/vc4/tests/vc4_mock_plane.c new file mode 100644 index 000000000000..106d3bd53aa5 --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/vc4_mock_plane.c @@ -0,0 +1,45 @@ +#include <drm/drm_atomic_state_helper.h> +#include <drm/drm_fourcc.h> +#include <drm/drm_modeset_helper_vtables.h> +#include <drm/drm_plane.h> + +#include <kunit/test.h> + +#include "vc4_mock.h" + +static const struct drm_plane_helper_funcs vc4_dummy_plane_helper_funcs = { +}; + +static const struct drm_plane_funcs vc4_dummy_plane_funcs = { + .atomic_destroy_state = drm_atomic_helper_plane_destroy_state, + .atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state, + .reset = drm_atomic_helper_plane_reset, +}; + +static const uint32_t vc4_dummy_plane_formats[] = { + DRM_FORMAT_XRGB8888, +}; + +struct vc4_dummy_plane *vc4_dummy_plane(struct kunit *test, + struct drm_device *drm, + enum drm_plane_type type) +{ + struct vc4_dummy_plane *dummy_plane; + struct drm_plane *plane; + + dummy_plane = drmm_universal_plane_alloc(drm, + struct vc4_dummy_plane, plane.base, + 0, + &vc4_dummy_plane_funcs, + vc4_dummy_plane_formats, + ARRAY_SIZE(vc4_dummy_plane_formats), + NULL, + DRM_PLANE_TYPE_PRIMARY, + NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dummy_plane); + + plane = &dummy_plane->plane.base; + drm_plane_helper_add(plane, &vc4_dummy_plane_helper_funcs); + + return dummy_plane; +} diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 59e473059fa2..21b40d644ace 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -675,8 +675,8 @@ void vc4_crtc_get_margins(struct drm_crtc_state *state, } }
-static int vc4_crtc_atomic_check(struct drm_crtc *crtc, - struct drm_atomic_state *state) +int vc4_crtc_atomic_check(struct drm_crtc *crtc, + struct drm_atomic_state *state) { struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, crtc); @@ -1116,7 +1116,7 @@ static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = { .get_scanout_position = vc4_crtc_get_scanout_position, };
-static const struct vc4_pv_data bcm2835_pv0_data = { +const struct vc4_pv_data bcm2835_pv0_data = { .base = { .name = "pixelvalve-0", .debugfs_name = "crtc0_regs", @@ -1131,7 +1131,7 @@ static const struct vc4_pv_data bcm2835_pv0_data = { }, };
-static const struct vc4_pv_data bcm2835_pv1_data = { +const struct vc4_pv_data bcm2835_pv1_data = { .base = { .name = "pixelvalve-1", .debugfs_name = "crtc1_regs", @@ -1146,7 +1146,7 @@ static const struct vc4_pv_data bcm2835_pv1_data = { }, };
-static const struct vc4_pv_data bcm2835_pv2_data = { +const struct vc4_pv_data bcm2835_pv2_data = { .base = { .name = "pixelvalve-2", .debugfs_name = "crtc2_regs", @@ -1161,7 +1161,7 @@ static const struct vc4_pv_data bcm2835_pv2_data = { }, };
-static const struct vc4_pv_data bcm2711_pv0_data = { +const struct vc4_pv_data bcm2711_pv0_data = { .base = { .name = "pixelvalve-0", .debugfs_name = "crtc0_regs", @@ -1176,7 +1176,7 @@ static const struct vc4_pv_data bcm2711_pv0_data = { }, };
-static const struct vc4_pv_data bcm2711_pv1_data = { +const struct vc4_pv_data bcm2711_pv1_data = { .base = { .name = "pixelvalve-1", .debugfs_name = "crtc1_regs", @@ -1191,7 +1191,7 @@ static const struct vc4_pv_data bcm2711_pv1_data = { }, };
-static const struct vc4_pv_data bcm2711_pv2_data = { +const struct vc4_pv_data bcm2711_pv2_data = { .base = { .name = "pixelvalve-2", .debugfs_name = "crtc2_regs", @@ -1205,7 +1205,7 @@ static const struct vc4_pv_data bcm2711_pv2_data = { }, };
-static const struct vc4_pv_data bcm2711_pv3_data = { +const struct vc4_pv_data bcm2711_pv3_data = { .base = { .name = "pixelvalve-3", .debugfs_name = "crtc3_regs", @@ -1219,7 +1219,7 @@ static const struct vc4_pv_data bcm2711_pv3_data = { }, };
-static const struct vc4_pv_data bcm2711_pv4_data = { +const struct vc4_pv_data bcm2711_pv4_data = { .base = { .name = "pixelvalve-4", .debugfs_name = "crtc4_regs", diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index b66bf7aea632..af6a83ac8526 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -196,7 +196,7 @@ static const struct drm_ioctl_desc vc4_drm_ioctls[] = { DRM_IOCTL_DEF_DRV(VC4_PERFMON_GET_VALUES, vc4_perfmon_get_values_ioctl, DRM_RENDER_ALLOW), };
-static const struct drm_driver vc4_drm_driver = { +const struct drm_driver vc4_drm_driver = { .driver_features = (DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_GEM | @@ -225,7 +225,7 @@ static const struct drm_driver vc4_drm_driver = { .patchlevel = DRIVER_PATCHLEVEL, };
-static const struct drm_driver vc5_drm_driver = { +const struct drm_driver vc5_drm_driver = { .driver_features = (DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_GEM), diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index e0be7a81a24a..e2a56c32a124 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -24,6 +24,9 @@ struct drm_device; struct drm_gem_object;
+extern const struct drm_driver vc4_drm_driver; +extern const struct drm_driver vc5_drm_driver; + /* Don't forget to update vc4_bo.c: bo_type_names[] when adding to * this. */ @@ -523,6 +526,8 @@ struct vc4_crtc_data { int hvs_output; };
+extern const struct vc4_crtc_data vc4_txp_crtc_data; + struct vc4_pv_data { struct vc4_crtc_data base;
@@ -535,6 +540,12 @@ struct vc4_pv_data { enum vc4_encoder_type encoder_types[4]; };
+extern const struct vc4_pv_data bcm2711_pv0_data; +extern const struct vc4_pv_data bcm2711_pv1_data; +extern const struct vc4_pv_data bcm2711_pv2_data; +extern const struct vc4_pv_data bcm2711_pv3_data; +extern const struct vc4_pv_data bcm2711_pv4_data; + struct vc4_crtc { struct drm_crtc base; struct platform_device *pdev; @@ -920,6 +931,8 @@ int vc4_page_flip(struct drm_crtc *crtc, struct drm_pending_vblank_event *event, uint32_t flags, struct drm_modeset_acquire_ctx *ctx); +int vc4_crtc_atomic_check(struct drm_crtc *crtc, + struct drm_atomic_state *state); struct drm_crtc_state *vc4_crtc_duplicate_state(struct drm_crtc *crtc); void vc4_crtc_destroy_state(struct drm_crtc *crtc, struct drm_crtc_state *state); diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index 4f7ce5d3e8ad..2b69454b8534 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -479,7 +479,7 @@ static irqreturn_t vc4_txp_interrupt(int irq, void *data) return IRQ_HANDLED; }
-static const struct vc4_crtc_data vc4_txp_crtc_data = { +const struct vc4_crtc_data vc4_txp_crtc_data = { .name = "txp", .debugfs_name = "txp_regs", .hvs_available_channels = BIT(2),
Hi Maxime,
I love your patch! Yet something to improve:
[auto build test ERROR on 35c3a2d02f0dc153a5f2f304ba33e1436b6a8d8f]
url: https://github.com/intel-lab-lkp/linux/commits/Maxime-Ripard/drm-Introduce-K... base: 35c3a2d02f0dc153a5f2f304ba33e1436b6a8d8f patch link: https://lore.kernel.org/r/20221123-rpi-kunit-tests-v1-22-051a0bb60a16%40cern... patch subject: [PATCH 22/24] drm/vc4: tests: Introduce a mocking infrastructure config: arc-randconfig-c024-20221123 compiler: arceb-elf-gcc (GCC) 12.1.0 reproduce (this is a W=1 build): wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross # https://github.com/intel-lab-lkp/linux/commit/b7d27608f26b831b79b9e4e25bce03... git remote add linux-review https://github.com/intel-lab-lkp/linux git fetch --no-tags linux-review Maxime-Ripard/drm-Introduce-Kunit-Tests-to-VC4/20221123-233528 git checkout b7d27608f26b831b79b9e4e25bce033d5284eb4c # save the config file mkdir build_dir && cp config build_dir/.config COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=arc SHELL=/bin/bash
If you fix the issue, kindly add following tag where applicable | Reported-by: kernel test robot lkp@intel.com
All errors (new ones prefixed by >>, old ones prefixed by <<):
ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/tests/drm_managed_test.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_crtc.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_output.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_plane.o ERROR: modpost: "drm_kunit_helper_alloc_device" [drivers/gpu/drm/tests/drm_managed_test.ko] undefined! ERROR: modpost: "__drm_kunit_helper_alloc_drm_device_with_driver" [drivers/gpu/drm/tests/drm_managed_test.ko] undefined! ERROR: modpost: "drm_kunit_helper_free_device" [drivers/gpu/drm/tests/drm_managed_test.ko] undefined!
ERROR: modpost: "vc4_dummy_plane" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_mock_pv" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_dummy_output" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_drm_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc5_drm_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "drm_kunit_helper_alloc_device" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "__drm_kunit_helper_alloc_drm_device_with_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined!
WARNING: modpost: suppressed 13 unresolved symbol warnings because there were too many)
Accessing a register when running under kunit is a bad idea since our device is completely mocked.
Fail the current test if we ever access any of our hardware registers.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_crtc.c | 13 +++++++++++-- drivers/gpu/drm/vc4/vc4_dpi.c | 13 +++++++++++-- drivers/gpu/drm/vc4/vc4_drv.h | 29 +++++++++++++++++++++++++---- drivers/gpu/drm/vc4/vc4_dsi.c | 9 ++++++++- drivers/gpu/drm/vc4/vc4_hdmi_regs.h | 4 ++++ drivers/gpu/drm/vc4/vc4_txp.c | 13 +++++++++++-- drivers/gpu/drm/vc4/vc4_vec.c | 13 +++++++++++-- 7 files changed, 81 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 21b40d644ace..ad868c24672a 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -50,8 +50,17 @@
#define HVS_FIFO_LATENCY_PIX 6
-#define CRTC_WRITE(offset, val) writel(val, vc4_crtc->regs + (offset)) -#define CRTC_READ(offset) readl(vc4_crtc->regs + (offset)) +#define CRTC_WRITE(offset, val) \ + do { \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + writel(val, vc4_crtc->regs + (offset)); \ + } while (0) + +#define CRTC_READ(offset) \ + ({ \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + readl(vc4_crtc->regs + (offset)); \ + })
static const struct debugfs_reg32 crtc_regs[] = { VC4_REG32(PV_CONTROL), diff --git a/drivers/gpu/drm/vc4/vc4_dpi.c b/drivers/gpu/drm/vc4/vc4_dpi.c index 1f8f44b7b5a5..0edf3c4c98c8 100644 --- a/drivers/gpu/drm/vc4/vc4_dpi.c +++ b/drivers/gpu/drm/vc4/vc4_dpi.c @@ -103,8 +103,17 @@ to_vc4_dpi(struct drm_encoder *encoder) return container_of(encoder, struct vc4_dpi, encoder.base); }
-#define DPI_READ(offset) readl(dpi->regs + (offset)) -#define DPI_WRITE(offset, val) writel(val, dpi->regs + (offset)) +#define DPI_READ(offset) \ + ({ \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + readl(dpi->regs + (offset)); \ + }) + +#define DPI_WRITE(offset, val) \ + do { \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + writel(val, dpi->regs + (offset)); \ + } while (0)
static const struct debugfs_reg32 dpi_regs[] = { VC4_REG32(DPI_C), diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index e2a56c32a124..d13033ed8848 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -19,6 +19,8 @@ #include <drm/drm_mm.h> #include <drm/drm_modeset_lock.h>
+#include <kunit/test-bug.h> + #include "uapi/drm/vc4_drm.h"
struct drm_device; @@ -642,10 +644,29 @@ to_vc4_crtc_state(const struct drm_crtc_state *crtc_state) return container_of(crtc_state, struct vc4_crtc_state, base); }
-#define V3D_READ(offset) readl(vc4->v3d->regs + offset) -#define V3D_WRITE(offset, val) writel(val, vc4->v3d->regs + offset) -#define HVS_READ(offset) readl(hvs->regs + offset) -#define HVS_WRITE(offset, val) writel(val, hvs->regs + offset) +#define V3D_READ(offset) \ + ({ \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + readl(vc4->v3d->regs + (offset)); \ + }) + +#define V3D_WRITE(offset, val) \ + do { \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + writel(val, vc4->v3d->regs + (offset)); \ + } while (0) + +#define HVS_READ(offset) \ + ({ \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + readl(hvs->regs + (offset)); \ + }) + +#define HVS_WRITE(offset, val) \ + do { \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + writel(val, hvs->regs + (offset)); \ + } while (0)
#define VC4_REG32(reg) { .name = #reg, .offset = reg }
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c index 878e05d79e81..2c9cb27903a0 100644 --- a/drivers/gpu/drm/vc4/vc4_dsi.c +++ b/drivers/gpu/drm/vc4/vc4_dsi.c @@ -617,6 +617,8 @@ dsi_dma_workaround_write(struct vc4_dsi *dsi, u32 offset, u32 val) dma_cookie_t cookie; int ret;
+ kunit_fail_current_test("Accessing a register in a unit test!\n"); + /* DSI0 should be able to write normally. */ if (!chan) { writel(val, dsi->regs + offset); @@ -645,7 +647,12 @@ dsi_dma_workaround_write(struct vc4_dsi *dsi, u32 offset, u32 val) DRM_ERROR("Failed to wait for DMA: %d\n", ret); }
-#define DSI_READ(offset) readl(dsi->regs + (offset)) +#define DSI_READ(offset) \ + ({ \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + readl(dsi->regs + (offset)); \ + }) + #define DSI_WRITE(offset, val) dsi_dma_workaround_write(dsi, offset, val) #define DSI_PORT_READ(offset) \ DSI_READ(dsi->variant->port ? DSI1_##offset : DSI0_##offset) diff --git a/drivers/gpu/drm/vc4/vc4_hdmi_regs.h b/drivers/gpu/drm/vc4/vc4_hdmi_regs.h index 48db438550b1..b04b2fc8d831 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi_regs.h +++ b/drivers/gpu/drm/vc4/vc4_hdmi_regs.h @@ -456,6 +456,8 @@ static inline u32 vc4_hdmi_read(struct vc4_hdmi *hdmi,
WARN_ON(pm_runtime_status_suspended(&hdmi->pdev->dev));
+ kunit_fail_current_test("Accessing an HDMI register in a unit test!\n"); + if (reg >= variant->num_registers) { dev_warn(&hdmi->pdev->dev, "Invalid register ID %u\n", reg); @@ -486,6 +488,8 @@ static inline void vc4_hdmi_write(struct vc4_hdmi *hdmi,
WARN_ON(pm_runtime_status_suspended(&hdmi->pdev->dev));
+ kunit_fail_current_test("Accessing an HDMI register in a unit test!\n"); + if (reg >= variant->num_registers) { dev_warn(&hdmi->pdev->dev, "Invalid register ID %u\n", reg); diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index 2b69454b8534..ef5cab2a3aa9 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -145,8 +145,17 @@ /* Number of lines received and committed to memory. */ #define TXP_PROGRESS 0x10
-#define TXP_READ(offset) readl(txp->regs + (offset)) -#define TXP_WRITE(offset, val) writel(val, txp->regs + (offset)) +#define TXP_READ(offset) \ + ({ \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + readl(txp->regs + (offset)); \ + }) + +#define TXP_WRITE(offset, val) \ + do { \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + writel(val, txp->regs + (offset)); \ + } while (0)
struct vc4_txp { struct vc4_crtc base; diff --git a/drivers/gpu/drm/vc4/vc4_vec.c b/drivers/gpu/drm/vc4/vc4_vec.c index 92c07e31d632..79bdef35db76 100644 --- a/drivers/gpu/drm/vc4/vc4_vec.c +++ b/drivers/gpu/drm/vc4/vc4_vec.c @@ -175,8 +175,17 @@ struct vc4_vec { struct debugfs_regset32 regset; };
-#define VEC_READ(offset) readl(vec->regs + (offset)) -#define VEC_WRITE(offset, val) writel(val, vec->regs + (offset)) +#define VEC_READ(offset) \ + ({ \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + readl(vec->regs + (offset)); \ + }) + +#define VEC_WRITE(offset, val) \ + do { \ + kunit_fail_current_test("Accessing a register in a unit test!\n"); \ + writel(val, vec->regs + (offset)); \ + } while (0)
static inline struct vc4_vec * encoder_to_vc4_vec(struct drm_encoder *encoder)
The HVS to PixelValve muxing code is fairly error prone and has a bunch of arbitrary constraints due to the hardware setup.
Let's create a test suite that makes sure that the possible combinations work and the invalid ones don't.
Signed-off-by: Maxime Ripard maxime@cerno.tech --- drivers/gpu/drm/vc4/tests/Makefile | 3 +- drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c | 624 +++++++++++++++++++++++++ 2 files changed, 626 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/vc4/tests/Makefile b/drivers/gpu/drm/vc4/tests/Makefile index 3e8a7b4f9017..1ae143c415a4 100644 --- a/drivers/gpu/drm/vc4/tests/Makefile +++ b/drivers/gpu/drm/vc4/tests/Makefile @@ -4,4 +4,5 @@ obj-$(CONFIG_DRM_VC4_KUNIT_TEST) += \ vc4_mock.o \ vc4_mock_crtc.o \ vc4_mock_output.o \ - vc4_mock_plane.o + vc4_mock_plane.o \ + vc4_test_pv_muxing.o diff --git a/drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c b/drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c new file mode 100644 index 000000000000..3047fc768c09 --- /dev/null +++ b/drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c @@ -0,0 +1,624 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <drm/drm_atomic.h> +#include <drm/drm_atomic_state_helper.h> +#include <drm/drm_atomic_uapi.h> +#include <drm/drm_crtc.h> +#include <drm/drm_drv.h> +#include <drm/drm_fourcc.h> +#include <drm/drm_mode.h> +#include <drm/drm_modeset_helper_vtables.h> +#include <drm/drm_plane.h> + +#include <kunit/test.h> + +#include "../../tests/drm_kunit_helpers.h" + +#include "../vc4_drv.h" + +#include "vc4_mock.h" + +struct pv_muxing_priv { + struct vc4_dev *vc4; + struct drm_modeset_acquire_ctx ctx; + struct drm_atomic_state *state; +}; + +static bool check_fifo_conflict(struct kunit *test, + const struct drm_atomic_state *state) +{ + struct vc4_hvs_state *hvs_state; + unsigned int used_fifos = 0; + unsigned int i; + + hvs_state = vc4_hvs_get_new_global_state(state); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hvs_state); + + for (i = 0; i < HVS_NUM_CHANNELS; i++) { + if (!hvs_state->fifo_state[i].in_use) + continue; + + KUNIT_EXPECT_FALSE(test, used_fifos & BIT(i)); + used_fifos |= BIT(i); + } + + return true; +} + +struct encoder_constraint { + enum vc4_encoder_type type; + unsigned int *channels; + size_t nchannels; +}; + +#define ENCODER_CONSTRAINT(_type, ...) \ + { \ + .type = _type, \ + .channels = (unsigned int[]) { __VA_ARGS__ }, \ + .nchannels = sizeof((unsigned int[]) { __VA_ARGS__ }) / \ + sizeof(unsigned int), \ + } + +static bool __check_encoder_constraints(const struct encoder_constraint *constraints, + size_t nconstraints, + enum vc4_encoder_type type, + unsigned int channel) +{ + unsigned int i; + + for (i = 0; i < nconstraints; i++) { + const struct encoder_constraint *constraint = &constraints[i]; + unsigned int j; + + if (constraint->type != type) + continue; + + for (j = 0; j < constraint->nchannels; j++) { + unsigned int _channel = constraint->channels[j]; + + if (channel != _channel) + continue; + + return true; + } + } + + return false; +} + +static const struct encoder_constraint vc5_encoder_constraints[] = { + ENCODER_CONSTRAINT(VC4_ENCODER_TYPE_DPI, 0), + ENCODER_CONSTRAINT(VC4_ENCODER_TYPE_DSI0, 0), + ENCODER_CONSTRAINT(VC4_ENCODER_TYPE_VEC, 1), + ENCODER_CONSTRAINT(VC4_ENCODER_TYPE_TXP, 0, 2), + ENCODER_CONSTRAINT(VC4_ENCODER_TYPE_DSI1, 0, 1, 2), + ENCODER_CONSTRAINT(VC4_ENCODER_TYPE_HDMI0, 0, 1, 2), + ENCODER_CONSTRAINT(VC4_ENCODER_TYPE_HDMI1, 0, 1, 2), +}; + +static bool check_vc5_encoder_constraints(enum vc4_encoder_type type, unsigned int channel) +{ + return __check_encoder_constraints(vc5_encoder_constraints, + ARRAY_SIZE(vc5_encoder_constraints), + type, channel); +} + +static bool check_channel_for_encoder(struct kunit *test, + const struct drm_atomic_state *state, + enum vc4_encoder_type type) +{ + struct drm_device *drm = state->dev; + struct drm_encoder *encoder; + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; + struct vc4_crtc_state *new_vc4_crtc_state; + struct vc4_hvs_state *new_hvs_state; + unsigned int channel; + + new_hvs_state = vc4_hvs_get_new_global_state(state); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, new_hvs_state); + + encoder = vc4_find_encoder_by_type(drm, type); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder); + + crtc = vc4_find_crtc_for_encoder(test, drm, encoder); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc); + + crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_state); + + new_vc4_crtc_state = to_vc4_crtc_state(crtc_state); + channel = new_vc4_crtc_state->assigned_channel; + KUNIT_EXPECT_NE(test, channel, VC4_HVS_CHANNEL_DISABLED); + + KUNIT_EXPECT_TRUE(test, new_hvs_state->fifo_state[channel].in_use); + + KUNIT_EXPECT_TRUE(test, + check_vc5_encoder_constraints(type, + channel)); + + return true; +} + +struct pv_muxing_param { + const char *name; + enum vc4_encoder_type *encoders; + size_t nencoders; +}; + +#define PV_MUXING_TEST(_name, ...) \ + { \ + .name = _name, \ + .encoders = (enum vc4_encoder_type[]) { __VA_ARGS__ }, \ + .nencoders = sizeof((enum vc4_encoder_type[]) { __VA_ARGS__ }) / \ + sizeof(enum vc4_encoder_type), \ + } + +static const struct pv_muxing_param vc5_test_pv_muxing_params[] = { + PV_MUXING_TEST("1 output: DPI", + VC4_ENCODER_TYPE_DPI), + PV_MUXING_TEST("1 output: DSI0", + VC4_ENCODER_TYPE_DSI0), + PV_MUXING_TEST("1 output: DSI1", + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("1 output: HDMI0", + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("1 output: HDMI1", + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("1 output: VEC", + VC4_ENCODER_TYPE_VEC), + PV_MUXING_TEST("2 output: DPI, DSI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("2 output: DPI, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("2 output: DPI, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("2 output: DPI, TXP", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP), + PV_MUXING_TEST("2 output: DPI, VEC", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC), + PV_MUXING_TEST("2 output: DPI, DSI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("2 output: DSI0, DSI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("2 output: DSI0, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("2 output: DSI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("2 output: DSI0, TXP", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP), + PV_MUXING_TEST("2 output: DSI0, VEC", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC), + PV_MUXING_TEST("2 output: DSI0, DSI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("2 output: DSI1, VEC", + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_VEC), + PV_MUXING_TEST("2 output: DSI1, TXP", + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_TXP), + PV_MUXING_TEST("2 output: DSI1, HDMI0", + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("2 output: DSI1, HDMI1", + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("2 output: HDMI0, VEC", + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_VEC), + PV_MUXING_TEST("2 output: HDMI0, TXP", + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_TXP), + PV_MUXING_TEST("2 output: HDMI0, HDMI1", + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("2 output: HDMI1, VEC", + VC4_ENCODER_TYPE_HDMI1, + VC4_ENCODER_TYPE_VEC), + PV_MUXING_TEST("2 output: HDMI1, TXP", + VC4_ENCODER_TYPE_HDMI1, + VC4_ENCODER_TYPE_TXP), + PV_MUXING_TEST("2 output: TXP, VEC", + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_VEC), + PV_MUXING_TEST("3 output: DPI, VEC, TXP", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP), + PV_MUXING_TEST("3 output: DPI, VEC, DSI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("3 output: DPI, VEC, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("3 output: DPI, VEC, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("3 output: DPI, TXP, DSI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("3 output: DPI, TXP, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("3 output: DPI, TXP, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("3 output: DPI, DSI1, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("3 output: DPI, DSI1, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("3 output: DPI, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("3 output: DSI0, VEC, TXP", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP), + PV_MUXING_TEST("3 output: DSI0, VEC, DSI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("3 output: DSI0, VEC, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("3 output: DSI0, VEC, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("3 output: DSI0, TXP, DSI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("3 output: DSI0, TXP, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("3 output: DSI0, TXP, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("3 output: DSI0, DSI1, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("3 output: DSI0, DSI1, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("3 output: DSI0, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), +}; + +static void vc4_test_pv_muxing_desc(const struct pv_muxing_param *t, char *desc) +{ + strscpy(desc, t->name, KUNIT_PARAM_DESC_SIZE); +} + +KUNIT_ARRAY_PARAM(vc5_test_pv_muxing, + vc5_test_pv_muxing_params, + vc4_test_pv_muxing_desc); + +static const struct pv_muxing_param vc5_test_pv_muxing_invalid_params[] = { + PV_MUXING_TEST("DPI/DSI0 Conflict", + VC4_ENCODER_TYPE_DPI, VC4_ENCODER_TYPE_DSI0), + PV_MUXING_TEST("More than 3 output: DPI, VEC, TXP, DSI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("More than 3 output: DPI, VEC, TXP, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DPI, VEC, TXP, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, VEC, DSI1, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DPI, VEC, DSI1, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, VEC, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, TXP, DSI1, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DPI, TXP, DSI1, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, TXP, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, VEC, TXP, DSI1, HDMI0", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DPI, VEC, TXP, DSI1, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, VEC, TXP, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, VEC, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, TXP, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, TXP, DSI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, TXP, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, TXP, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, DSI1, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, DSI1, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, TXP, DSI1, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DSI0, TXP, DSI1, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, TXP, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, TXP, DSI1, HDMI0", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, TXP, DSI1, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, TXP, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, TXP, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: VEC, TXP, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DPI, VEC, TXP, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DPI, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), + PV_MUXING_TEST("More than 3 output: DSI0, VEC, TXP, DSI1, HDMI0, HDMI1", + VC4_ENCODER_TYPE_DSI0, + VC4_ENCODER_TYPE_VEC, + VC4_ENCODER_TYPE_TXP, + VC4_ENCODER_TYPE_DSI1, + VC4_ENCODER_TYPE_HDMI0, + VC4_ENCODER_TYPE_HDMI1), +}; + +KUNIT_ARRAY_PARAM(vc5_test_pv_muxing_invalid, + vc5_test_pv_muxing_invalid_params, + vc4_test_pv_muxing_desc); + +static void drm_vc5_test_pv_muxing(struct kunit *test) +{ + const struct pv_muxing_priv *priv = test->priv; + const struct pv_muxing_param *params = test->param_value; + struct vc4_dev *vc4 = priv->vc4; + struct drm_device *drm = &vc4->base; + struct drm_atomic_state *state = priv->state; + unsigned int i; + int ret; + + for (i = 0; i < params->nencoders; i++) { + enum vc4_encoder_type enc_type = params->encoders[i]; + + ret = vc4_mock_atomic_add_output(test, drm, enc_type, state); + KUNIT_ASSERT_EQ(test, ret, 0); + } + + ret = drm_atomic_check_only(state); + KUNIT_EXPECT_EQ(test, ret, 0); + + KUNIT_EXPECT_TRUE(test, + check_fifo_conflict(test, state)); + + for (i = 0; i < params->nencoders; i++) { + enum vc4_encoder_type enc_type = params->encoders[i]; + + KUNIT_EXPECT_TRUE(test, check_channel_for_encoder(test, state, enc_type)); + } +} + +static void drm_vc4_test_pv_muxing_invalid(struct kunit *test) +{ + const struct pv_muxing_priv *priv = test->priv; + const struct pv_muxing_param *params = test->param_value; + struct vc4_dev *vc4 = priv->vc4; + struct drm_device *drm = &vc4->base; + struct drm_atomic_state *state = priv->state; + unsigned int i; + int ret; + + for (i = 0; i < params->nencoders; i++) { + enum vc4_encoder_type enc_type = params->encoders[i]; + + ret = vc4_mock_atomic_add_output(test, drm, enc_type, state); + KUNIT_ASSERT_EQ(test, ret, 0); + } + + ret = drm_atomic_check_only(state); + KUNIT_EXPECT_LT(test, ret, 0); +} + +static int vc5_pv_muxing_test_init(struct kunit *test) +{ + struct drm_atomic_state *state; + struct pv_muxing_priv *priv; + struct drm_device *drm; + struct vc4_dev *vc4; + + priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, priv); + test->priv = priv; + + vc4 = vc5_mock_device(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vc4); + priv->vc4 = vc4; + + drm_modeset_acquire_init(&priv->ctx, 0); + + drm = &vc4->base; + state = drm_atomic_state_alloc(drm); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state); + + state->acquire_ctx = &priv->ctx; + + priv->state = state; + + return 0; +} + +static void vc4_pv_muxing_test_exit(struct kunit *test) +{ + struct pv_muxing_priv *priv = test->priv; + struct vc4_dev *vc4 = priv->vc4; + struct drm_device *drm = &vc4->base; + struct drm_atomic_state *state = priv->state; + + drm_atomic_state_put(state); + drm_modeset_drop_locks(&priv->ctx); + drm_modeset_acquire_fini(&priv->ctx); + drm_dev_unregister(drm); + drm_kunit_helper_free_device(test, vc4->dev); +} + +static struct kunit_case vc5_pv_muxing_tests[] = { + KUNIT_CASE_PARAM(drm_vc5_test_pv_muxing, + vc5_test_pv_muxing_gen_params), + KUNIT_CASE_PARAM(drm_vc4_test_pv_muxing_invalid, + vc5_test_pv_muxing_invalid_gen_params), + {} +}; + +static struct kunit_suite vc5_pv_muxing_test_suite = { + .name = "vc5-pv-muxing", + .init = vc5_pv_muxing_test_init, + .exit = vc4_pv_muxing_test_exit, + .test_cases = vc5_pv_muxing_tests, +}; + +kunit_test_suite(vc5_pv_muxing_test_suite);
On Wed, Nov 23, 2022 at 11:28 PM Maxime Ripard maxime@cerno.tech wrote:
Hi,
This series introduce Kunit tests to the vc4 KMS driver, but unlike what we have been doing so far in KMS, it actually tests the atomic modesetting code.
In order to do so, I've had to improve a fair bit on the Kunit helpers already found in the tree in order to register a full blown and somewhat functional KMS driver.
It's of course relying on a mock so that we can test it anywhere. The mocking approach created a number of issues, the main one being that we need to create a decent mock in the first place, see patch 22. The basic idea is that I created some structures to provide a decent approximation of the actual hardware, and that would support both major architectures supported by vc4.
This is of course meant to evolve over time and support more tests, but I've focused on testing the HVS FIFO assignment code which is fairly tricky (and the tests have actually revealed one more bug with our current implementation). I used to have a userspace implementation of those tests, where I would copy and paste the kernel code and run the tests on a regular basis. It's was obviously fairly suboptimal, so it seemed like the perfect testbed for that series.
Let me know what you think, Maxime
To: David Airlie airlied@gmail.com To: Daniel Vetter daniel@ffwll.ch To: Maarten Lankhorst maarten.lankhorst@linux.intel.com To: Maxime Ripard mripard@kernel.org To: Thomas Zimmermann tzimmermann@suse.de Cc: Dave Stevenson dave.stevenson@raspberrypi.com Cc: Javier Martinez Canillas javierm@redhat.com Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org Cc: Maíra Canal mairacanal@riseup.net Cc: Brendan Higgins brendan.higgins@linux.dev Cc: David Gow davidgow@google.com Cc: linux-kselftest@vger.kernel.org Cc: kunit-dev@googlegroups.com Cc: dri-devel@lists.freedesktop.org Cc: linux-kernel@vger.kernel.org Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org Signed-off-by: Maxime Ripard maxime@cerno.tech
Hi Maxime,
Thanks very much for this! I'm really excited to see these sorts of tests being written.
I was able to successfully run these under qemu with: ./tools/testing/kunit/kunit.py run --kunitconfig drivers/gpu/drm/vc4/tests --arch arm64 --cross_compile=aarch64-linux-gnu- (and also with clang, using --make_options LLVM=1 instead of the --cross_compile flag)
On the other hand, they don't compile as a module: ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_crtc.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_output.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_plane.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/tests/drm_managed_test.o ERROR: modpost: "vc4_drm_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc5_drm_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "drm_kunit_helper_alloc_device" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "__drm_kunit_helper_alloc_drm_device_with_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "__vc4_hvs_alloc" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_dummy_plane" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_mock_pv" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_dummy_output" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_kms_load" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_txp_crtc_data" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! WARNING: modpost: suppressed 17 unresolved symbol warnings because there were too many)
Most of those are just the need to export some symbols. There's some work underway to support conditionally exporting symbols only if KUnit is enabled, which may help: https://lore.kernel.org/linux-kselftest/20221102175959.2921063-1-rmoar@googl...
Otherwise, I suspect the better short-term solution would just be to require that the tests are built-in (or at least compiled into whatever of the drm/vc4 modules makes most sense).
The only other thing which has me a little confused is the naming of some of the functions, specifically with the __ prefix. Is it just for internal functions (many of them aren't static, but maybe they could use the VISIBLE_IF_KUNIT macro if that makes sense), or for versions of functions which accept extra arguments? Not a big deal (and maybe it's a DRM naming convention I'm ignorant of), but I couldn't quite find a pattern on my first read through.
But on the whole, these look good from a KUnit point-of-view. It's really to see some solid mocking and driver testing, too. There would be ways to avoid passing the 'struct kunit' around in more places (or to store extra data as a kunit_resource), but I think it's better overall to pass it around like you have in this case -- it's certainly more compatible with things which might span threads (e.g. the workqueues).
Thanks a bunch, -- David
Hi David,
On Thu, Nov 24, 2022 at 04:31:14PM +0800, David Gow wrote:
On Wed, Nov 23, 2022 at 11:28 PM Maxime Ripard maxime@cerno.tech wrote:
Hi,
This series introduce Kunit tests to the vc4 KMS driver, but unlike what we have been doing so far in KMS, it actually tests the atomic modesetting code.
In order to do so, I've had to improve a fair bit on the Kunit helpers already found in the tree in order to register a full blown and somewhat functional KMS driver.
It's of course relying on a mock so that we can test it anywhere. The mocking approach created a number of issues, the main one being that we need to create a decent mock in the first place, see patch 22. The basic idea is that I created some structures to provide a decent approximation of the actual hardware, and that would support both major architectures supported by vc4.
This is of course meant to evolve over time and support more tests, but I've focused on testing the HVS FIFO assignment code which is fairly tricky (and the tests have actually revealed one more bug with our current implementation). I used to have a userspace implementation of those tests, where I would copy and paste the kernel code and run the tests on a regular basis. It's was obviously fairly suboptimal, so it seemed like the perfect testbed for that series.
Thanks very much for this! I'm really excited to see these sorts of tests being written.
I was able to successfully run these under qemu with: ./tools/testing/kunit/kunit.py run --kunitconfig drivers/gpu/drm/vc4/tests --arch arm64 --cross_compile=aarch64-linux-gnu- (and also with clang, using --make_options LLVM=1 instead of the --cross_compile flag)
On the other hand, they don't compile as a module: ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_crtc.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_output.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_mock_plane.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.o ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/drm/tests/drm_managed_test.o ERROR: modpost: "vc4_drm_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc5_drm_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "drm_kunit_helper_alloc_device" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "__drm_kunit_helper_alloc_drm_device_with_driver" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "__vc4_hvs_alloc" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_dummy_plane" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_mock_pv" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_dummy_output" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_kms_load" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! ERROR: modpost: "vc4_txp_crtc_data" [drivers/gpu/drm/vc4/tests/vc4_mock.ko] undefined! WARNING: modpost: suppressed 17 unresolved symbol warnings because there were too many)
Thanks I'll fix it
Most of those are just the need to export some symbols. There's some work underway to support conditionally exporting symbols only if KUnit is enabled, which may help: https://lore.kernel.org/linux-kselftest/20221102175959.2921063-1-rmoar@googl...
That's awesome :)
The current solution to include the test implementation is not ideal, so it's great to see a nicer solution being worked on.
Otherwise, I suspect the better short-term solution would just be to require that the tests are built-in (or at least compiled into whatever of the drm/vc4 modules makes most sense).
The only other thing which has me a little confused is the naming of some of the functions, specifically with the __ prefix. Is it just for internal functions (many of them aren't static, but maybe they could use the VISIBLE_IF_KUNIT macro if that makes sense), or for versions of functions which accept extra arguments?
It was for internal functions that would definitely benefit from VISIBLE_IF_KUNIT indeed
Not a big deal (and maybe it's a DRM naming convention I'm ignorant of), but I couldn't quite find a pattern on my first read through.
But on the whole, these look good from a KUnit point-of-view. It's really to see some solid mocking and driver testing, too. There would be ways to avoid passing the 'struct kunit' around in more places (or to store extra data as a kunit_resource), but I think it's better overall to pass it around like you have in this case -- it's certainly more compatible with things which might span threads (e.g. the workqueues).
One thing I'm really unsure about and would like your input on is basically the entire device instantiation code in drm_kunit_helpers.c
It's a little fishy since it will allocate a platform_device while the driver might expect some other bus device. And the code to bind the driver based around probe and workqueues seems like a hack.
This is something that would benefit from having proper functions in kunit to allocate a proper device for a given test. This is already something that other unit test suites seems to get wrong, and I'm sure there's some bugs somewhere in the helpers I did for DRM. What do you think?
Maxime
linaro-mm-sig@lists.linaro.org
-
David Gow -
Javier Martinez Canillas -
kernel test robot -
Maxime Ripard -
Maíra Canal