[PATCH v8 0/8] Improve test coverage of TTM

[PATCH v8 0/8] Improve test coverage of TTM

[Karolina Stolarek]

Add tests for building blocks of the TTM subsystem, such as ttm_resource,
ttm_resource_manager, ttm_tt and ttm_buffer_object. This series covers
basic functions such as initialization, allocation and clean-up of each
struct. Testing of ttm_buffer_object also includes locking and unlocking
the object for validation, with special scenarios such as an interrupted
wait or deadlock.

Some of the test cases check the bulk move mechanism and how it interacts
with pinned buffers. This is to be seen if we want to add dedicated testing
for bulk move or not. The resource allocation subtests use ttm_sys_manager
for now. Resources that don't use system memory will be indirectly tested
via tests for ttm_bo_validate()/ttm_bo_init_validate(), using a mock
resource manager.

Use kunit_tool script to manually run all the tests:

$ ./tools/testing/kunit/kunit.py run --kunitconfig=drivers/gpu/drm/ttm/tests

To build a kernel with TTM KUnit tests, first enable CONFIG_KUNIT, and
then CONFIG_DRM_TTM_KUNIT_TEST.

Many thanks,
Karolina

v8:
  - Add Tested-by tags to commits that introduce tests
  - Improve the comment for ttm_bo_reserve_deadlock() subtest (Andi)
  - Actually clean up the resource when "error_free_blocks" is hit in
    ttm_mock_manager_alloc(). Without that change, we hit
    DEBUG_LOCKS_WARN_ON(lock->magic != lock) warning when cleaning up
    the resource manager because we try clean up an incomplete, orphaned
    resource. That's not good, and this could bite us back in the future.

v7:
 - Drop size argument from ttm_place_kunit_init(), it's no longer needed
 - Delete a TODO comment from ttm_bo_validate_tests.c
 - First evict BOs before calling ttm_resource_manager_set_used() in
   ttm_mock_manager_fini()
 - Stop calling ttm_resource_manager_cleanup() as a part of the mock manager
   fini sequence. It frees a move fence that is allocated via KUnit allocator,
   which gets freed again as a part of the test cleanup
 - Set use_tt to true in mock manager and stop passing in the flag for it
 - Make ttm_dev_empty_funcs static
   (drivers/gpu/drm/ttm/tests/ttm_tt_test.c:232:25: sparse: sparse:
   symbol 'ttm_dev_empty_funcs' was not declared. Should it be static?)
 - Cast bo->base.resv->fences to a generic pointer before it's checked by
   KUnit (drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c:98:9:
   sparse: sparse: incompatible types in comparison expression (different
   base types))
 - Clean up mock managers in ttm_bo_validate_move_fence_not_signaled subtest

v6:
  - Include tests for ttm_bo_init_reserved() and ttm_bo_validate(), with
    a mock resource manager (patches 6-8; no eviction testing)
  - Add ttm_test_devices_all_init() helper to also init ttm_device instance
  - Remove fpfn and lpfn from ttm_place_kunit_init() helper -- these fields
    are neither used nor tested

v5:
  - Actually use the page_flags parameter in ttm_tt_simple_create()

v4:
  - First unreserve the object before calling ww_acquire_fini() in
    ttm_bo_reserve_double_resv subtest
  - Silence lockdep in ttm_bo_reserve_deadlock subtest (open to suggestions
    how to fix it in a different way)
  - Use a genuine GEM object in ttm_buffer_object instead of an empty one

v3:
  - Instead of modifying the main TTM Makefile, use
    EXPORT_SYMBOL_FOR_TESTS_ONLY() macro for symbols that are tested but
    not widely exported. Thanks to this change, TTM tests can be built
    as modules, even when non-exported functions are used
  - Change the description of a patch that fixes ttm_pool_pre_populated()

v2:
  - Remove Makefile for KUnit tests and move the definitions to the
    TTM's one
  - Switch on CONFIG_DRM_TTM_KUNIT_TEST=m so the tests and TTM module
    are built as one. This allows building the tests as a module, even
    if it uses functions that are not exported
  - Fix ttm_pool_pre_populated(); a wrong flag was passed to
    ttm_tt_kunit_init() function

Karolina Stolarek (8):
  drm/ttm/tests: Add tests for ttm_resource and ttm_sys_man
  drm/ttm/tests: Add tests for ttm_tt
  drm/ttm/tests: Add tests for ttm_bo functions
  drm/ttm/tests: Fix argument in ttm_tt_kunit_init()
  drm/ttm/tests: Use an init function from the helpers lib
  drm/ttm/tests: Test simple BO creation and validation
  drm/ttm/tests: Add tests with mock resource managers
  drm/ttm/tests: Add test cases dependent on fence signaling

 drivers/gpu/drm/Kconfig                       |   1 +
 drivers/gpu/drm/ttm/tests/.kunitconfig        |   1 +
 drivers/gpu/drm/ttm/tests/Makefile            |   5 +
 drivers/gpu/drm/ttm/tests/ttm_bo_test.c       | 622 ++++++++++++++
 .../gpu/drm/ttm/tests/ttm_bo_validate_test.c  | 795 ++++++++++++++++++
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c | 109 ++-
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h |   7 +
 drivers/gpu/drm/ttm/tests/ttm_mock_manager.c  | 207 +++++
 drivers/gpu/drm/ttm/tests/ttm_mock_manager.h  |  31 +
 drivers/gpu/drm/ttm/tests/ttm_pool_test.c     |   3 +-
 drivers/gpu/drm/ttm/tests/ttm_resource_test.c | 335 ++++++++
 drivers/gpu/drm/ttm/tests/ttm_tt_test.c       | 282 +++++++
 drivers/gpu/drm/ttm/ttm_resource.c            |   3 +
 drivers/gpu/drm/ttm/ttm_tt.c                  |   3 +
 14 files changed, 2401 insertions(+), 3 deletions(-)
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_test.c
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.c
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.h
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_resource_test.c
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_tt_test.c

-- 
2.34.1

[PATCH v8 1/8] drm/ttm/tests: Add tests for ttm_resource and ttm_sys_man

[Karolina Stolarek]

Test initialization of ttm_resource using different memory domains.
Add tests for a system memory manager and functions that can be
tested without a fully-featured resource manager. Update
ttm_bo_kunit_init() to initialize BO's kref and a genuine GEM drm
object. Export ttm_resource_alloc for test purposes only.

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
Tested-by: Amaranath Somalapuram <Amaranath.Somalapuram@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
---
 drivers/gpu/drm/ttm/tests/Makefile            |   1 +
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c |  22 +-
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h |   3 +
 drivers/gpu/drm/ttm/tests/ttm_resource_test.c | 335 ++++++++++++++++++
 drivers/gpu/drm/ttm/ttm_resource.c            |   3 +
 5 files changed, 363 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_resource_test.c

diff --git a/drivers/gpu/drm/ttm/tests/Makefile b/drivers/gpu/drm/ttm/tests/Makefile
index ec87c4fc1ad5..c92fe2052ef6 100644
--- a/drivers/gpu/drm/ttm/tests/Makefile
+++ b/drivers/gpu/drm/ttm/tests/Makefile
@@ -3,4 +3,5 @@
 obj-$(CONFIG_DRM_TTM_KUNIT_TEST) += \
         ttm_device_test.o \
         ttm_pool_test.o \
+        ttm_resource_test.o \
         ttm_kunit_helpers.o
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
index 81661d8827aa..779fbc038f17 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
@@ -29,19 +29,39 @@ struct ttm_buffer_object *ttm_bo_kunit_init(struct kunit *test,
 					    struct ttm_test_devices *devs,
 					    size_t size)
 {
-	struct drm_gem_object gem_obj = { .size = size };
+	struct drm_gem_object gem_obj = { };
 	struct ttm_buffer_object *bo;
+	int err;
 
 	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
 	KUNIT_ASSERT_NOT_NULL(test, bo);
 
 	bo->base = gem_obj;
+	err = drm_gem_object_init(devs->drm, &bo->base, size);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
 	bo->bdev = devs->ttm_dev;
+	kref_init(&bo->kref);
 
 	return bo;
 }
 EXPORT_SYMBOL_GPL(ttm_bo_kunit_init);
 
+struct ttm_place *ttm_place_kunit_init(struct kunit *test,
+				       uint32_t mem_type, uint32_t flags)
+{
+	struct ttm_place *place;
+
+	place = kunit_kzalloc(test, sizeof(*place), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, place);
+
+	place->mem_type = mem_type;
+	place->flags = flags;
+
+	return place;
+}
+EXPORT_SYMBOL_GPL(ttm_place_kunit_init);
+
 struct ttm_test_devices *ttm_test_devices_basic(struct kunit *test)
 {
 	struct ttm_test_devices *devs;
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
index e261e3660d0b..2f51c833a536 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
@@ -8,6 +8,7 @@
 #include <drm/drm_drv.h>
 #include <drm/ttm/ttm_device.h>
 #include <drm/ttm/ttm_bo.h>
+#include <drm/ttm/ttm_placement.h>
 
 #include <drm/drm_kunit_helpers.h>
 #include <kunit/test.h>
@@ -28,6 +29,8 @@ int ttm_device_kunit_init(struct ttm_test_devices *priv,
 struct ttm_buffer_object *ttm_bo_kunit_init(struct kunit *test,
 					    struct ttm_test_devices *devs,
 					    size_t size);
+struct ttm_place *ttm_place_kunit_init(struct kunit *test,
+				       uint32_t mem_type, uint32_t flags);
 
 struct ttm_test_devices *ttm_test_devices_basic(struct kunit *test);
 struct ttm_test_devices *ttm_test_devices_all(struct kunit *test);
diff --git a/drivers/gpu/drm/ttm/tests/ttm_resource_test.c b/drivers/gpu/drm/ttm/tests/ttm_resource_test.c
new file mode 100644
index 000000000000..029e1f094bb0
--- /dev/null
+++ b/drivers/gpu/drm/ttm/tests/ttm_resource_test.c
@@ -0,0 +1,335 @@
+// SPDX-License-Identifier: GPL-2.0 AND MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+#include <drm/ttm/ttm_resource.h>
+
+#include "ttm_kunit_helpers.h"
+
+#define RES_SIZE		SZ_4K
+#define TTM_PRIV_DUMMY_REG	(TTM_NUM_MEM_TYPES - 1)
+
+struct ttm_resource_test_case {
+	const char *description;
+	uint32_t mem_type;
+	uint32_t flags;
+};
+
+struct ttm_resource_test_priv {
+	struct ttm_test_devices *devs;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+};
+
+static const struct ttm_resource_manager_func ttm_resource_manager_mock_funcs = { };
+
+static int ttm_resource_test_init(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv;
+
+	priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, priv);
+
+	priv->devs = ttm_test_devices_all(test);
+	KUNIT_ASSERT_NOT_NULL(test, priv->devs);
+
+	test->priv = priv;
+
+	return 0;
+}
+
+static void ttm_resource_test_fini(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+
+	ttm_test_devices_put(test, priv->devs);
+}
+
+static void ttm_init_test_mocks(struct kunit *test,
+				struct ttm_resource_test_priv *priv,
+				uint32_t mem_type, uint32_t flags)
+{
+	size_t size = RES_SIZE;
+
+	/* Make sure we have what we need for a good BO mock */
+	KUNIT_ASSERT_NOT_NULL(test, priv->devs->ttm_dev);
+
+	priv->bo = ttm_bo_kunit_init(test, priv->devs, size);
+	priv->place = ttm_place_kunit_init(test, mem_type, flags);
+}
+
+static void ttm_init_test_manager(struct kunit *test,
+				  struct ttm_resource_test_priv *priv,
+				  uint32_t mem_type)
+{
+	struct ttm_device *ttm_dev = priv->devs->ttm_dev;
+	struct ttm_resource_manager *man;
+	size_t size = SZ_16K;
+
+	man = kunit_kzalloc(test, sizeof(*man), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, man);
+
+	man->use_tt = false;
+	man->func = &ttm_resource_manager_mock_funcs;
+
+	ttm_resource_manager_init(man, ttm_dev, size);
+	ttm_set_driver_manager(ttm_dev, mem_type, man);
+	ttm_resource_manager_set_used(man, true);
+}
+
+static const struct ttm_resource_test_case ttm_resource_cases[] = {
+	{
+		.description = "Init resource in TTM_PL_SYSTEM",
+		.mem_type = TTM_PL_SYSTEM,
+	},
+	{
+		.description = "Init resource in TTM_PL_VRAM",
+		.mem_type = TTM_PL_VRAM,
+	},
+	{
+		.description = "Init resource in a private placement",
+		.mem_type = TTM_PRIV_DUMMY_REG,
+	},
+	{
+		.description = "Init resource in TTM_PL_SYSTEM, set placement flags",
+		.mem_type = TTM_PL_SYSTEM,
+		.flags = TTM_PL_FLAG_TOPDOWN,
+	},
+};
+
+static void ttm_resource_case_desc(const struct ttm_resource_test_case *t, char *desc)
+{
+	strscpy(desc, t->description, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(ttm_resource, ttm_resource_cases, ttm_resource_case_desc);
+
+static void ttm_resource_init_basic(struct kunit *test)
+{
+	const struct ttm_resource_test_case *params = test->param_value;
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource *res;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_resource_manager *man;
+	uint64_t expected_usage;
+
+	ttm_init_test_mocks(test, priv, params->mem_type, params->flags);
+	bo = priv->bo;
+	place = priv->place;
+
+	if (params->mem_type > TTM_PL_SYSTEM)
+		ttm_init_test_manager(test, priv, params->mem_type);
+
+	res = kunit_kzalloc(test, sizeof(*res), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, res);
+
+	man = ttm_manager_type(priv->devs->ttm_dev, place->mem_type);
+	expected_usage = man->usage + RES_SIZE;
+
+	KUNIT_ASSERT_TRUE(test, list_empty(&man->lru[bo->priority]));
+
+	ttm_resource_init(bo, place, res);
+
+	KUNIT_ASSERT_EQ(test, res->start, 0);
+	KUNIT_ASSERT_EQ(test, res->size, RES_SIZE);
+	KUNIT_ASSERT_EQ(test, res->mem_type, place->mem_type);
+	KUNIT_ASSERT_EQ(test, res->placement, place->flags);
+	KUNIT_ASSERT_PTR_EQ(test, res->bo, bo);
+
+	KUNIT_ASSERT_NULL(test, res->bus.addr);
+	KUNIT_ASSERT_EQ(test, res->bus.offset, 0);
+	KUNIT_ASSERT_FALSE(test, res->bus.is_iomem);
+	KUNIT_ASSERT_EQ(test, res->bus.caching, ttm_cached);
+	KUNIT_ASSERT_EQ(test, man->usage, expected_usage);
+
+	KUNIT_ASSERT_TRUE(test, list_is_singular(&man->lru[bo->priority]));
+
+	ttm_resource_fini(man, res);
+}
+
+static void ttm_resource_init_pinned(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource *res;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_resource_manager *man;
+
+	ttm_init_test_mocks(test, priv, TTM_PL_SYSTEM, 0);
+	bo = priv->bo;
+	place = priv->place;
+
+	man = ttm_manager_type(priv->devs->ttm_dev, place->mem_type);
+
+	res = kunit_kzalloc(test, sizeof(*res), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, res);
+	KUNIT_ASSERT_TRUE(test, list_empty(&bo->bdev->pinned));
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_pin(bo);
+	ttm_resource_init(bo, place, res);
+	KUNIT_ASSERT_TRUE(test, list_is_singular(&bo->bdev->pinned));
+
+	ttm_bo_unpin(bo);
+	ttm_resource_fini(man, res);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_ASSERT_TRUE(test, list_empty(&bo->bdev->pinned));
+}
+
+static void ttm_resource_fini_basic(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource *res;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_resource_manager *man;
+
+	ttm_init_test_mocks(test, priv, TTM_PL_SYSTEM, 0);
+	bo = priv->bo;
+	place = priv->place;
+
+	man = ttm_manager_type(priv->devs->ttm_dev, place->mem_type);
+
+	res = kunit_kzalloc(test, sizeof(*res), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, res);
+
+	ttm_resource_init(bo, place, res);
+	ttm_resource_fini(man, res);
+
+	KUNIT_ASSERT_TRUE(test, list_empty(&res->lru));
+	KUNIT_ASSERT_EQ(test, man->usage, 0);
+}
+
+static void ttm_resource_manager_init_basic(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource_manager *man;
+	size_t size = SZ_16K;
+
+	man = kunit_kzalloc(test, sizeof(*man), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, man);
+
+	ttm_resource_manager_init(man, priv->devs->ttm_dev, size);
+
+	KUNIT_ASSERT_PTR_EQ(test, man->bdev, priv->devs->ttm_dev);
+	KUNIT_ASSERT_EQ(test, man->size, size);
+	KUNIT_ASSERT_EQ(test, man->usage, 0);
+	KUNIT_ASSERT_NULL(test, man->move);
+	KUNIT_ASSERT_NOT_NULL(test, &man->move_lock);
+
+	for (int i = 0; i < TTM_MAX_BO_PRIORITY; ++i)
+		KUNIT_ASSERT_TRUE(test, list_empty(&man->lru[i]));
+}
+
+static void ttm_resource_manager_usage_basic(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource *res;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_resource_manager *man;
+	uint64_t actual_usage;
+
+	ttm_init_test_mocks(test, priv, TTM_PL_SYSTEM, TTM_PL_FLAG_TOPDOWN);
+	bo = priv->bo;
+	place = priv->place;
+
+	res = kunit_kzalloc(test, sizeof(*res), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, res);
+
+	man = ttm_manager_type(priv->devs->ttm_dev, place->mem_type);
+
+	ttm_resource_init(bo, place, res);
+	actual_usage = ttm_resource_manager_usage(man);
+
+	KUNIT_ASSERT_EQ(test, actual_usage, RES_SIZE);
+
+	ttm_resource_fini(man, res);
+}
+
+static void ttm_resource_manager_set_used_basic(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource_manager *man;
+
+	man = ttm_manager_type(priv->devs->ttm_dev, TTM_PL_SYSTEM);
+	KUNIT_ASSERT_TRUE(test, man->use_type);
+
+	ttm_resource_manager_set_used(man, false);
+	KUNIT_ASSERT_FALSE(test, man->use_type);
+}
+
+static void ttm_sys_man_alloc_basic(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource_manager *man;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_resource *res;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	int ret;
+
+	ttm_init_test_mocks(test, priv, mem_type, 0);
+	bo = priv->bo;
+	place = priv->place;
+
+	man = ttm_manager_type(priv->devs->ttm_dev, mem_type);
+	ret = man->func->alloc(man, bo, place, &res);
+
+	KUNIT_ASSERT_EQ(test, ret, 0);
+	KUNIT_ASSERT_EQ(test, res->size, RES_SIZE);
+	KUNIT_ASSERT_EQ(test, res->mem_type, mem_type);
+	KUNIT_ASSERT_PTR_EQ(test, res->bo, bo);
+
+	ttm_resource_fini(man, res);
+}
+
+static void ttm_sys_man_free_basic(struct kunit *test)
+{
+	struct ttm_resource_test_priv *priv = test->priv;
+	struct ttm_resource_manager *man;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_resource *res;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+
+	ttm_init_test_mocks(test, priv, mem_type, 0);
+	bo = priv->bo;
+	place = priv->place;
+
+	res = kunit_kzalloc(test, sizeof(*res), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, res);
+
+	ttm_resource_alloc(bo, place, &res);
+
+	man = ttm_manager_type(priv->devs->ttm_dev, mem_type);
+	man->func->free(man, res);
+
+	KUNIT_ASSERT_TRUE(test, list_empty(&man->lru[bo->priority]));
+	KUNIT_ASSERT_EQ(test, man->usage, 0);
+}
+
+static struct kunit_case ttm_resource_test_cases[] = {
+	KUNIT_CASE_PARAM(ttm_resource_init_basic, ttm_resource_gen_params),
+	KUNIT_CASE(ttm_resource_init_pinned),
+	KUNIT_CASE(ttm_resource_fini_basic),
+	KUNIT_CASE(ttm_resource_manager_init_basic),
+	KUNIT_CASE(ttm_resource_manager_usage_basic),
+	KUNIT_CASE(ttm_resource_manager_set_used_basic),
+	KUNIT_CASE(ttm_sys_man_alloc_basic),
+	KUNIT_CASE(ttm_sys_man_free_basic),
+	{}
+};
+
+static struct kunit_suite ttm_resource_test_suite = {
+	.name = "ttm_resource",
+	.init = ttm_resource_test_init,
+	.exit = ttm_resource_test_fini,
+	.test_cases = ttm_resource_test_cases,
+};
+
+kunit_test_suites(&ttm_resource_test_suite);
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c
index 46ff9c75bb12..02b96d23fdb9 100644
--- a/drivers/gpu/drm/ttm/ttm_resource.c
+++ b/drivers/gpu/drm/ttm/ttm_resource.c
@@ -30,6 +30,8 @@
 #include <drm/ttm/ttm_placement.h>
 #include <drm/ttm/ttm_resource.h>
 
+#include <drm/drm_util.h>
+
 /**
  * ttm_lru_bulk_move_init - initialize a bulk move structure
  * @bulk: the structure to init
@@ -240,6 +242,7 @@ int ttm_resource_alloc(struct ttm_buffer_object *bo,
 	spin_unlock(&bo->bdev->lru_lock);
 	return 0;
 }
+EXPORT_SYMBOL_FOR_TESTS_ONLY(ttm_resource_alloc);
 
 void ttm_resource_free(struct ttm_buffer_object *bo, struct ttm_resource **res)
 {
-- 
2.34.1

[PATCH v8 2/8] drm/ttm/tests: Add tests for ttm_tt

[Karolina Stolarek]

Test initialization, creation and destruction of ttm_tt instances.
Export ttm_tt_destroy and ttm_tt_create symbols for testing purposes.

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Tested-by: Amaranath Somalapuram <Amaranath.Somalapuram@amd.com>
---
 drivers/gpu/drm/ttm/tests/Makefile            |   1 +
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c |  20 ++
 drivers/gpu/drm/ttm/tests/ttm_tt_test.c       | 295 ++++++++++++++++++
 drivers/gpu/drm/ttm/ttm_tt.c                  |   3 +
 4 files changed, 319 insertions(+)
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_tt_test.c

diff --git a/drivers/gpu/drm/ttm/tests/Makefile b/drivers/gpu/drm/ttm/tests/Makefile
index c92fe2052ef6..f570530bbb60 100644
--- a/drivers/gpu/drm/ttm/tests/Makefile
+++ b/drivers/gpu/drm/ttm/tests/Makefile
@@ -4,4 +4,5 @@ obj-$(CONFIG_DRM_TTM_KUNIT_TEST) += \
         ttm_device_test.o \
         ttm_pool_test.o \
         ttm_resource_test.o \
+        ttm_tt_test.o \
         ttm_kunit_helpers.o
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
index 779fbc038f17..ba4e5c689164 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
@@ -2,9 +2,29 @@
 /*
  * Copyright © 2023 Intel Corporation
  */
+#include <drm/ttm/ttm_tt.h>
+
 #include "ttm_kunit_helpers.h"
 
+static struct ttm_tt *ttm_tt_simple_create(struct ttm_buffer_object *bo,
+					   uint32_t page_flags)
+{
+	struct ttm_tt *tt;
+
+	tt = kzalloc(sizeof(*tt), GFP_KERNEL);
+	ttm_tt_init(tt, bo, page_flags, ttm_cached, 0);
+
+	return tt;
+}
+
+static void ttm_tt_simple_destroy(struct ttm_device *bdev, struct ttm_tt *ttm)
+{
+	kfree(ttm);
+}
+
 struct ttm_device_funcs ttm_dev_funcs = {
+	.ttm_tt_create = ttm_tt_simple_create,
+	.ttm_tt_destroy = ttm_tt_simple_destroy,
 };
 EXPORT_SYMBOL_GPL(ttm_dev_funcs);
 
diff --git a/drivers/gpu/drm/ttm/tests/ttm_tt_test.c b/drivers/gpu/drm/ttm/tests/ttm_tt_test.c
new file mode 100644
index 000000000000..fd4502c18de6
--- /dev/null
+++ b/drivers/gpu/drm/ttm/tests/ttm_tt_test.c
@@ -0,0 +1,295 @@
+// SPDX-License-Identifier: GPL-2.0 AND MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+#include <linux/shmem_fs.h>
+#include <drm/ttm/ttm_tt.h>
+
+#include "ttm_kunit_helpers.h"
+
+#define BO_SIZE		SZ_4K
+
+struct ttm_tt_test_case {
+	const char *description;
+	uint32_t size;
+	uint32_t extra_pages_num;
+};
+
+static int ttm_tt_test_init(struct kunit *test)
+{
+	struct ttm_test_devices *priv;
+
+	priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, priv);
+
+	priv = ttm_test_devices_all(test);
+	test->priv = priv;
+
+	return 0;
+}
+
+static const struct ttm_tt_test_case ttm_tt_init_basic_cases[] = {
+	{
+		.description = "Page-aligned size",
+		.size = SZ_4K,
+	},
+	{
+		.description = "Extra pages requested",
+		.size = SZ_4K,
+		.extra_pages_num = 1,
+	},
+};
+
+static void ttm_tt_init_case_desc(const struct ttm_tt_test_case *t,
+				  char *desc)
+{
+	strscpy(desc, t->description, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(ttm_tt_init_basic, ttm_tt_init_basic_cases,
+		  ttm_tt_init_case_desc);
+
+static void ttm_tt_init_basic(struct kunit *test)
+{
+	const struct ttm_tt_test_case *params = test->param_value;
+	struct ttm_buffer_object *bo;
+	struct ttm_tt *tt;
+	uint32_t page_flags = TTM_TT_FLAG_ZERO_ALLOC;
+	enum ttm_caching caching = ttm_cached;
+	uint32_t extra_pages = params->extra_pages_num;
+	int num_pages = params->size >> PAGE_SHIFT;
+	int err;
+
+	tt = kunit_kzalloc(test, sizeof(*tt), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, tt);
+
+	bo = ttm_bo_kunit_init(test, test->priv, params->size);
+
+	err = ttm_tt_init(tt, bo, page_flags, caching, extra_pages);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	KUNIT_ASSERT_EQ(test, tt->num_pages, num_pages + extra_pages);
+
+	KUNIT_ASSERT_EQ(test, tt->page_flags, page_flags);
+	KUNIT_ASSERT_EQ(test, tt->caching, caching);
+
+	KUNIT_ASSERT_NULL(test, tt->dma_address);
+	KUNIT_ASSERT_NULL(test, tt->swap_storage);
+}
+
+static void ttm_tt_init_misaligned(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ttm_tt *tt;
+	enum ttm_caching caching = ttm_cached;
+	uint32_t size = SZ_8K;
+	int num_pages = (size + SZ_4K) >> PAGE_SHIFT;
+	int err;
+
+	tt = kunit_kzalloc(test, sizeof(*tt), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, tt);
+
+	bo = ttm_bo_kunit_init(test, test->priv, size);
+
+	/* Make the object size misaligned */
+	bo->base.size += 1;
+
+	err = ttm_tt_init(tt, bo, 0, caching, 0);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	KUNIT_ASSERT_EQ(test, tt->num_pages, num_pages);
+}
+
+static void ttm_tt_fini_basic(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ttm_tt *tt;
+	enum ttm_caching caching = ttm_cached;
+	int err;
+
+	tt = kunit_kzalloc(test, sizeof(*tt), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, tt);
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_tt_init(tt, bo, 0, caching, 0);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	KUNIT_ASSERT_NOT_NULL(test, tt->pages);
+
+	ttm_tt_fini(tt);
+	KUNIT_ASSERT_NULL(test, tt->pages);
+}
+
+static void ttm_tt_fini_sg(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ttm_tt *tt;
+	enum ttm_caching caching = ttm_cached;
+	int err;
+
+	tt = kunit_kzalloc(test, sizeof(*tt), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, tt);
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_sg_tt_init(tt, bo, 0, caching);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	KUNIT_ASSERT_NOT_NULL(test, tt->dma_address);
+
+	ttm_tt_fini(tt);
+	KUNIT_ASSERT_NULL(test, tt->dma_address);
+}
+
+static void ttm_tt_fini_shmem(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ttm_tt *tt;
+	struct file *shmem;
+	enum ttm_caching caching = ttm_cached;
+	int err;
+
+	tt = kunit_kzalloc(test, sizeof(*tt), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, tt);
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_tt_init(tt, bo, 0, caching, 0);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	shmem = shmem_file_setup("ttm swap", BO_SIZE, 0);
+	tt->swap_storage = shmem;
+
+	ttm_tt_fini(tt);
+	KUNIT_ASSERT_NULL(test, tt->swap_storage);
+}
+
+static void ttm_tt_create_basic(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	bo->type = ttm_bo_type_device;
+
+	dma_resv_lock(bo->base.resv, NULL);
+	err = ttm_tt_create(bo, false);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_NOT_NULL(test, bo->ttm);
+
+	/* Free manually, as it was allocated outside of KUnit */
+	kfree(bo->ttm);
+}
+
+static void ttm_tt_create_invalid_bo_type(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	bo->type = ttm_bo_type_sg + 1;
+
+	dma_resv_lock(bo->base.resv, NULL);
+	err = ttm_tt_create(bo, false);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, -EINVAL);
+	KUNIT_EXPECT_NULL(test, bo->ttm);
+}
+
+static void ttm_tt_create_ttm_exists(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ttm_tt *tt;
+	enum ttm_caching caching = ttm_cached;
+	int err;
+
+	tt = kunit_kzalloc(test, sizeof(*tt), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, tt);
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_tt_init(tt, bo, 0, caching, 0);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	bo->ttm = tt;
+
+	dma_resv_lock(bo->base.resv, NULL);
+	err = ttm_tt_create(bo, false);
+	dma_resv_unlock(bo->base.resv);
+
+	/* Expect to keep the previous TTM */
+	KUNIT_ASSERT_EQ(test, err, 0);
+	KUNIT_ASSERT_PTR_EQ(test, tt, bo->ttm);
+}
+
+static struct ttm_tt *ttm_tt_null_create(struct ttm_buffer_object *bo,
+					 uint32_t page_flags)
+{
+	return NULL;
+}
+
+static struct ttm_device_funcs ttm_dev_empty_funcs = {
+	.ttm_tt_create = ttm_tt_null_create,
+};
+
+static void ttm_tt_create_failed(struct kunit *test)
+{
+	const struct ttm_test_devices *devs = test->priv;
+	struct ttm_buffer_object *bo;
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	/* Update ttm_device_funcs so we don't alloc ttm_tt */
+	devs->ttm_dev->funcs = &ttm_dev_empty_funcs;
+
+	dma_resv_lock(bo->base.resv, NULL);
+	err = ttm_tt_create(bo, false);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_ASSERT_EQ(test, err, -ENOMEM);
+}
+
+static void ttm_tt_destroy_basic(struct kunit *test)
+{
+	const struct ttm_test_devices *devs = test->priv;
+	struct ttm_buffer_object *bo;
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	dma_resv_lock(bo->base.resv, NULL);
+	err = ttm_tt_create(bo, false);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_ASSERT_EQ(test, err, 0);
+	KUNIT_ASSERT_NOT_NULL(test, bo->ttm);
+
+	ttm_tt_destroy(devs->ttm_dev, bo->ttm);
+}
+
+static struct kunit_case ttm_tt_test_cases[] = {
+	KUNIT_CASE_PARAM(ttm_tt_init_basic, ttm_tt_init_basic_gen_params),
+	KUNIT_CASE(ttm_tt_init_misaligned),
+	KUNIT_CASE(ttm_tt_fini_basic),
+	KUNIT_CASE(ttm_tt_fini_sg),
+	KUNIT_CASE(ttm_tt_fini_shmem),
+	KUNIT_CASE(ttm_tt_create_basic),
+	KUNIT_CASE(ttm_tt_create_invalid_bo_type),
+	KUNIT_CASE(ttm_tt_create_ttm_exists),
+	KUNIT_CASE(ttm_tt_create_failed),
+	KUNIT_CASE(ttm_tt_destroy_basic),
+	{}
+};
+
+static struct kunit_suite ttm_tt_test_suite = {
+	.name = "ttm_tt",
+	.init = ttm_tt_test_init,
+	.exit = ttm_test_devices_fini,
+	.test_cases = ttm_tt_test_cases,
+};
+
+kunit_test_suites(&ttm_tt_test_suite);
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
index e0a77671edd6..d978dc539a9b 100644
--- a/drivers/gpu/drm/ttm/ttm_tt.c
+++ b/drivers/gpu/drm/ttm/ttm_tt.c
@@ -36,6 +36,7 @@
 #include <linux/file.h>
 #include <linux/module.h>
 #include <drm/drm_cache.h>
+#include <drm/drm_util.h>
 #include <drm/ttm/ttm_bo.h>
 #include <drm/ttm/ttm_tt.h>
 
@@ -91,6 +92,7 @@ int ttm_tt_create(struct ttm_buffer_object *bo, bool zero_alloc)
 
 	return 0;
 }
+EXPORT_SYMBOL_FOR_TESTS_ONLY(ttm_tt_create);
 
 /*
  * Allocates storage for pointers to the pages that back the ttm.
@@ -129,6 +131,7 @@ void ttm_tt_destroy(struct ttm_device *bdev, struct ttm_tt *ttm)
 {
 	bdev->funcs->ttm_tt_destroy(bdev, ttm);
 }
+EXPORT_SYMBOL_FOR_TESTS_ONLY(ttm_tt_destroy);
 
 static void ttm_tt_init_fields(struct ttm_tt *ttm,
 			       struct ttm_buffer_object *bo,
-- 
2.34.1

[PATCH v8 3/8] drm/ttm/tests: Add tests for ttm_bo functions

[Karolina Stolarek]

Test reservation and release of TTM buffer objects. Add tests to check
pin and unpin operations.

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
Tested-by: Amaranath Somalapuram <Amaranath.Somalapuram@amd.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
---
 drivers/gpu/drm/ttm/tests/Makefile            |   1 +
 drivers/gpu/drm/ttm/tests/ttm_bo_test.c       | 622 ++++++++++++++++++
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c |   6 +
 3 files changed, 629 insertions(+)
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_test.c

diff --git a/drivers/gpu/drm/ttm/tests/Makefile b/drivers/gpu/drm/ttm/tests/Makefile
index f570530bbb60..468535f7eed2 100644
--- a/drivers/gpu/drm/ttm/tests/Makefile
+++ b/drivers/gpu/drm/ttm/tests/Makefile
@@ -5,4 +5,5 @@ obj-$(CONFIG_DRM_TTM_KUNIT_TEST) += \
         ttm_pool_test.o \
         ttm_resource_test.o \
         ttm_tt_test.o \
+        ttm_bo_test.o \
         ttm_kunit_helpers.o
diff --git a/drivers/gpu/drm/ttm/tests/ttm_bo_test.c b/drivers/gpu/drm/ttm/tests/ttm_bo_test.c
new file mode 100644
index 000000000000..1f8a4f8adc92
--- /dev/null
+++ b/drivers/gpu/drm/ttm/tests/ttm_bo_test.c
@@ -0,0 +1,622 @@
+// SPDX-License-Identifier: GPL-2.0 AND MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+#include <linux/dma-resv.h>
+#include <linux/kthread.h>
+#include <linux/delay.h>
+#include <linux/timer.h>
+#include <linux/jiffies.h>
+#include <linux/mutex.h>
+#include <linux/ww_mutex.h>
+
+#include <drm/ttm/ttm_resource.h>
+#include <drm/ttm/ttm_placement.h>
+#include <drm/ttm/ttm_tt.h>
+
+#include "ttm_kunit_helpers.h"
+
+#define BO_SIZE		SZ_8K
+
+struct ttm_bo_test_case {
+	const char *description;
+	bool interruptible;
+	bool no_wait;
+};
+
+static const struct ttm_bo_test_case ttm_bo_reserved_cases[] = {
+	{
+		.description = "Cannot be interrupted and sleeps",
+		.interruptible = false,
+		.no_wait = false,
+	},
+	{
+		.description = "Cannot be interrupted, locks straight away",
+		.interruptible = false,
+		.no_wait = true,
+	},
+	{
+		.description = "Can be interrupted, sleeps",
+		.interruptible = true,
+		.no_wait = false,
+	},
+};
+
+static void ttm_bo_init_case_desc(const struct ttm_bo_test_case *t,
+				  char *desc)
+{
+	strscpy(desc, t->description, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(ttm_bo_reserve, ttm_bo_reserved_cases, ttm_bo_init_case_desc);
+
+static void ttm_bo_reserve_optimistic_no_ticket(struct kunit *test)
+{
+	const struct ttm_bo_test_case *params = test->param_value;
+	struct ttm_buffer_object *bo;
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_bo_reserve(bo, params->interruptible, params->no_wait, NULL);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	dma_resv_unlock(bo->base.resv);
+}
+
+static void ttm_bo_reserve_locked_no_sleep(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	bool interruptible = false;
+	bool no_wait = true;
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	/* Let's lock it beforehand */
+	dma_resv_lock(bo->base.resv, NULL);
+
+	err = ttm_bo_reserve(bo, interruptible, no_wait, NULL);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_ASSERT_EQ(test, err, -EBUSY);
+}
+
+static void ttm_bo_reserve_no_wait_ticket(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ww_acquire_ctx ctx;
+	bool interruptible = false;
+	bool no_wait = true;
+	int err;
+
+	ww_acquire_init(&ctx, &reservation_ww_class);
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_bo_reserve(bo, interruptible, no_wait, &ctx);
+	KUNIT_ASSERT_EQ(test, err, -EBUSY);
+
+	ww_acquire_fini(&ctx);
+}
+
+static void ttm_bo_reserve_double_resv(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ww_acquire_ctx ctx;
+	bool interruptible = false;
+	bool no_wait = false;
+	int err;
+
+	ww_acquire_init(&ctx, &reservation_ww_class);
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_bo_reserve(bo, interruptible, no_wait, &ctx);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	err = ttm_bo_reserve(bo, interruptible, no_wait, &ctx);
+
+	dma_resv_unlock(bo->base.resv);
+	ww_acquire_fini(&ctx);
+
+	KUNIT_ASSERT_EQ(test, err, -EALREADY);
+}
+
+/*
+ * A test case heavily inspired by ww_test_edeadlk_normal(). It injects
+ * a deadlock by manipulating the sequence number of the context that holds
+ * dma_resv lock of bo2 so the other context is "wounded" and has to back off
+ * (indicated by -EDEADLK). The subtest checks if ttm_bo_reserve() properly
+ * propagates that error.
+ */
+static void ttm_bo_reserve_deadlock(struct kunit *test)
+{
+	struct ttm_buffer_object *bo1, *bo2;
+	struct ww_acquire_ctx ctx1, ctx2;
+	bool interruptible = false;
+	bool no_wait = false;
+	int err;
+
+	bo1 = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	bo2 = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	ww_acquire_init(&ctx1, &reservation_ww_class);
+	mutex_lock(&bo2->base.resv->lock.base);
+
+	/* The deadlock will be caught by WW mutex, don't warn about it */
+	lock_release(&bo2->base.resv->lock.base.dep_map, 1);
+
+	bo2->base.resv->lock.ctx = &ctx2;
+	ctx2 = ctx1;
+	ctx2.stamp--; /* Make the context holding the lock younger */
+
+	err = ttm_bo_reserve(bo1, interruptible, no_wait, &ctx1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	err = ttm_bo_reserve(bo2, interruptible, no_wait, &ctx1);
+	KUNIT_ASSERT_EQ(test, err, -EDEADLK);
+
+	dma_resv_unlock(bo1->base.resv);
+	ww_acquire_fini(&ctx1);
+}
+
+#if IS_BUILTIN(CONFIG_DRM_TTM_KUNIT_TEST)
+struct signal_timer {
+	struct timer_list timer;
+	struct ww_acquire_ctx *ctx;
+};
+
+static void signal_for_ttm_bo_reserve(struct timer_list *t)
+{
+	struct signal_timer *s_timer = from_timer(s_timer, t, timer);
+	struct task_struct *task = s_timer->ctx->task;
+
+	do_send_sig_info(SIGTERM, SEND_SIG_PRIV, task, PIDTYPE_PID);
+}
+
+static int threaded_ttm_bo_reserve(void *arg)
+{
+	struct ttm_buffer_object *bo = arg;
+	struct signal_timer s_timer;
+	struct ww_acquire_ctx ctx;
+	bool interruptible = true;
+	bool no_wait = false;
+	int err;
+
+	ww_acquire_init(&ctx, &reservation_ww_class);
+
+	/* Prepare a signal that will interrupt the reservation attempt */
+	timer_setup_on_stack(&s_timer.timer, &signal_for_ttm_bo_reserve, 0);
+	s_timer.ctx = &ctx;
+
+	mod_timer(&s_timer.timer, msecs_to_jiffies(100));
+
+	err = ttm_bo_reserve(bo, interruptible, no_wait, &ctx);
+
+	timer_delete_sync(&s_timer.timer);
+	destroy_timer_on_stack(&s_timer.timer);
+
+	ww_acquire_fini(&ctx);
+
+	return err;
+}
+
+static void ttm_bo_reserve_interrupted(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct task_struct *task;
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	task = kthread_create(threaded_ttm_bo_reserve, bo, "ttm-bo-reserve");
+
+	if (IS_ERR(task))
+		KUNIT_FAIL(test, "Couldn't create ttm bo reserve task\n");
+
+	/* Take a lock so the threaded reserve has to wait */
+	mutex_lock(&bo->base.resv->lock.base);
+
+	wake_up_process(task);
+	msleep(20);
+	err = kthread_stop(task);
+
+	mutex_unlock(&bo->base.resv->lock.base);
+
+	KUNIT_ASSERT_EQ(test, err, -ERESTARTSYS);
+}
+#endif /* IS_BUILTIN(CONFIG_DRM_TTM_KUNIT_TEST) */
+
+static void ttm_bo_unreserve_basic(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_device *ttm_dev;
+	struct ttm_resource *res1, *res2;
+	struct ttm_place *place;
+	struct ttm_resource_manager *man;
+	unsigned int bo_prio = TTM_MAX_BO_PRIORITY - 1;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	int err;
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	bo->priority = bo_prio;
+
+	err = ttm_resource_alloc(bo, place, &res1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+
+	bo->resource = res1;
+
+	/* Add a dummy resource to populate LRU */
+	ttm_resource_alloc(bo, place, &res2);
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_unreserve(bo);
+
+	man = ttm_manager_type(priv->ttm_dev, mem_type);
+	KUNIT_ASSERT_EQ(test,
+			list_is_last(&res1->lru, &man->lru[bo->priority]), 1);
+
+	ttm_resource_free(bo, &res2);
+	ttm_resource_free(bo, &res1);
+}
+
+static void ttm_bo_unreserve_pinned(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_device *ttm_dev;
+	struct ttm_resource *res1, *res2;
+	struct ttm_place *place;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	int err;
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	place = ttm_place_kunit_init(test, mem_type, 0);
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_pin(bo);
+
+	err = ttm_resource_alloc(bo, place, &res1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	bo->resource = res1;
+
+	/* Add a dummy resource to the pinned list */
+	err = ttm_resource_alloc(bo, place, &res2);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	KUNIT_ASSERT_EQ(test,
+			list_is_last(&res2->lru, &priv->ttm_dev->pinned), 1);
+
+	ttm_bo_unreserve(bo);
+	KUNIT_ASSERT_EQ(test,
+			list_is_last(&res1->lru, &priv->ttm_dev->pinned), 1);
+
+	ttm_resource_free(bo, &res1);
+	ttm_resource_free(bo, &res2);
+}
+
+static void ttm_bo_unreserve_bulk(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_lru_bulk_move lru_bulk_move;
+	struct ttm_lru_bulk_move_pos *pos;
+	struct ttm_buffer_object *bo1, *bo2;
+	struct ttm_resource *res1, *res2;
+	struct ttm_device *ttm_dev;
+	struct ttm_place *place;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	unsigned int bo_priority = 0;
+	int err;
+
+	ttm_lru_bulk_move_init(&lru_bulk_move);
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	bo1 = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	bo2 = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	dma_resv_lock(bo1->base.resv, NULL);
+	ttm_bo_set_bulk_move(bo1, &lru_bulk_move);
+	dma_resv_unlock(bo1->base.resv);
+
+	err = ttm_resource_alloc(bo1, place, &res1);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	bo1->resource = res1;
+
+	dma_resv_lock(bo2->base.resv, NULL);
+	ttm_bo_set_bulk_move(bo2, &lru_bulk_move);
+	dma_resv_unlock(bo2->base.resv);
+
+	err = ttm_resource_alloc(bo2, place, &res2);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	bo2->resource = res2;
+
+	ttm_bo_reserve(bo1, false, false, NULL);
+	ttm_bo_unreserve(bo1);
+
+	pos = &lru_bulk_move.pos[mem_type][bo_priority];
+	KUNIT_ASSERT_PTR_EQ(test, res1, pos->last);
+
+	ttm_resource_free(bo1, &res1);
+	ttm_resource_free(bo2, &res2);
+}
+
+static void ttm_bo_put_basic(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_resource *res;
+	struct ttm_device *ttm_dev;
+	struct ttm_place *place;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	int err;
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	bo->type = ttm_bo_type_device;
+
+	err = ttm_resource_alloc(bo, place, &res);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	bo->resource = res;
+
+	dma_resv_lock(bo->base.resv, NULL);
+	err = ttm_tt_create(bo, false);
+	dma_resv_unlock(bo->base.resv);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	ttm_bo_put(bo);
+}
+
+static const char *mock_name(struct dma_fence *f)
+{
+	return "kunit-ttm-bo-put";
+}
+
+static const struct dma_fence_ops mock_fence_ops = {
+	.get_driver_name = mock_name,
+	.get_timeline_name = mock_name,
+};
+
+static void ttm_bo_put_shared_resv(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct dma_resv *external_resv;
+	struct dma_fence *fence;
+	/* A dummy DMA fence lock */
+	spinlock_t fence_lock;
+	struct ttm_device *ttm_dev;
+	int err;
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	external_resv = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, external_resv);
+
+	dma_resv_init(external_resv);
+
+	fence = kunit_kzalloc(test, sizeof(*fence), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, fence);
+
+	spin_lock_init(&fence_lock);
+	dma_fence_init(fence, &mock_fence_ops, &fence_lock, 0, 0);
+
+	dma_resv_lock(external_resv, NULL);
+	dma_resv_reserve_fences(external_resv, 1);
+	dma_resv_add_fence(external_resv, fence, DMA_RESV_USAGE_BOOKKEEP);
+	dma_resv_unlock(external_resv);
+
+	dma_fence_signal(fence);
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+	bo->type = ttm_bo_type_device;
+	bo->base.resv = external_resv;
+
+	ttm_bo_put(bo);
+}
+
+static void ttm_bo_pin_basic(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_device *ttm_dev;
+	unsigned int no_pins = 3;
+	int err;
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	for (int i = 0; i < no_pins; i++) {
+		dma_resv_lock(bo->base.resv, NULL);
+		ttm_bo_pin(bo);
+		dma_resv_unlock(bo->base.resv);
+	}
+
+	KUNIT_ASSERT_EQ(test, bo->pin_count, no_pins);
+}
+
+static void ttm_bo_pin_unpin_resource(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_lru_bulk_move lru_bulk_move;
+	struct ttm_lru_bulk_move_pos *pos;
+	struct ttm_buffer_object *bo;
+	struct ttm_resource *res;
+	struct ttm_device *ttm_dev;
+	struct ttm_place *place;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	unsigned int bo_priority = 0;
+	int err;
+
+	ttm_lru_bulk_move_init(&lru_bulk_move);
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_resource_alloc(bo, place, &res);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	bo->resource = res;
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_set_bulk_move(bo, &lru_bulk_move);
+	ttm_bo_pin(bo);
+	dma_resv_unlock(bo->base.resv);
+
+	pos = &lru_bulk_move.pos[mem_type][bo_priority];
+
+	KUNIT_ASSERT_EQ(test, bo->pin_count, 1);
+	KUNIT_ASSERT_NULL(test, pos->first);
+	KUNIT_ASSERT_NULL(test, pos->last);
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_unpin(bo);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_ASSERT_PTR_EQ(test, res, pos->last);
+	KUNIT_ASSERT_EQ(test, bo->pin_count, 0);
+
+	ttm_resource_free(bo, &res);
+}
+
+static void ttm_bo_multiple_pin_one_unpin(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_lru_bulk_move lru_bulk_move;
+	struct ttm_lru_bulk_move_pos *pos;
+	struct ttm_buffer_object *bo;
+	struct ttm_resource *res;
+	struct ttm_device *ttm_dev;
+	struct ttm_place *place;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	unsigned int bo_priority = 0;
+	int err;
+
+	ttm_lru_bulk_move_init(&lru_bulk_move);
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+
+	ttm_dev = kunit_kzalloc(test, sizeof(*ttm_dev), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+
+	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	priv->ttm_dev = ttm_dev;
+
+	bo = ttm_bo_kunit_init(test, test->priv, BO_SIZE);
+
+	err = ttm_resource_alloc(bo, place, &res);
+	KUNIT_ASSERT_EQ(test, err, 0);
+	bo->resource = res;
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_set_bulk_move(bo, &lru_bulk_move);
+
+	/* Multiple pins */
+	ttm_bo_pin(bo);
+	ttm_bo_pin(bo);
+
+	dma_resv_unlock(bo->base.resv);
+
+	pos = &lru_bulk_move.pos[mem_type][bo_priority];
+
+	KUNIT_ASSERT_EQ(test, bo->pin_count, 2);
+	KUNIT_ASSERT_NULL(test, pos->first);
+	KUNIT_ASSERT_NULL(test, pos->last);
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_unpin(bo);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_ASSERT_EQ(test, bo->pin_count, 1);
+	KUNIT_ASSERT_NULL(test, pos->first);
+	KUNIT_ASSERT_NULL(test, pos->last);
+
+	dma_resv_lock(bo->base.resv, NULL);
+	ttm_bo_unpin(bo);
+	dma_resv_unlock(bo->base.resv);
+
+	ttm_resource_free(bo, &res);
+}
+
+static struct kunit_case ttm_bo_test_cases[] = {
+	KUNIT_CASE_PARAM(ttm_bo_reserve_optimistic_no_ticket,
+			 ttm_bo_reserve_gen_params),
+	KUNIT_CASE(ttm_bo_reserve_locked_no_sleep),
+	KUNIT_CASE(ttm_bo_reserve_no_wait_ticket),
+	KUNIT_CASE(ttm_bo_reserve_double_resv),
+#if IS_BUILTIN(CONFIG_DRM_TTM_KUNIT_TEST)
+	KUNIT_CASE(ttm_bo_reserve_interrupted),
+#endif
+	KUNIT_CASE(ttm_bo_reserve_deadlock),
+	KUNIT_CASE(ttm_bo_unreserve_basic),
+	KUNIT_CASE(ttm_bo_unreserve_pinned),
+	KUNIT_CASE(ttm_bo_unreserve_bulk),
+	KUNIT_CASE(ttm_bo_put_basic),
+	KUNIT_CASE(ttm_bo_put_shared_resv),
+	KUNIT_CASE(ttm_bo_pin_basic),
+	KUNIT_CASE(ttm_bo_pin_unpin_resource),
+	KUNIT_CASE(ttm_bo_multiple_pin_one_unpin),
+	{}
+};
+
+static struct kunit_suite ttm_bo_test_suite = {
+	.name = "ttm_bo",
+	.init = ttm_test_devices_init,
+	.exit = ttm_test_devices_fini,
+	.test_cases = ttm_bo_test_cases,
+};
+
+kunit_test_suites(&ttm_bo_test_suite);
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
index ba4e5c689164..7b7c1fa805fc 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
@@ -22,6 +22,10 @@ static void ttm_tt_simple_destroy(struct ttm_device *bdev, struct ttm_tt *ttm)
 	kfree(ttm);
 }
 
+static void dummy_ttm_bo_destroy(struct ttm_buffer_object *bo)
+{
+}
+
 struct ttm_device_funcs ttm_dev_funcs = {
 	.ttm_tt_create = ttm_tt_simple_create,
 	.ttm_tt_destroy = ttm_tt_simple_destroy,
@@ -61,6 +65,8 @@ struct ttm_buffer_object *ttm_bo_kunit_init(struct kunit *test,
 	KUNIT_ASSERT_EQ(test, err, 0);
 
 	bo->bdev = devs->ttm_dev;
+	bo->destroy = dummy_ttm_bo_destroy;
+
 	kref_init(&bo->kref);
 
 	return bo;
-- 
2.34.1

[PATCH v8 4/8] drm/ttm/tests: Fix argument in ttm_tt_kunit_init()

[Karolina Stolarek]

Remove a leftover definition of page order and pass an empty flag value
in ttm_pool_pre_populated().

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
Tested-by: Amaranath Somalapuram <Amaranath.Somalapuram@amd.com>
Reviewed-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@intel.com>
Acked-by: Christian König <christian.koenig@amd.com>
---
 drivers/gpu/drm/ttm/tests/ttm_pool_test.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/ttm/tests/ttm_pool_test.c b/drivers/gpu/drm/ttm/tests/ttm_pool_test.c
index 2d9cae8cd984..b97f7b6daf5b 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_pool_test.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_pool_test.c
@@ -78,10 +78,9 @@ static struct ttm_pool *ttm_pool_pre_populated(struct kunit *test,
 	struct ttm_test_devices *devs = priv->devs;
 	struct ttm_pool *pool;
 	struct ttm_tt *tt;
-	unsigned long order = __fls(size / PAGE_SIZE);
 	int err;
 
-	tt = ttm_tt_kunit_init(test, order, caching, size);
+	tt = ttm_tt_kunit_init(test, 0, caching, size);
 	KUNIT_ASSERT_NOT_NULL(test, tt);
 
 	pool = kunit_kzalloc(test, sizeof(*pool), GFP_KERNEL);
-- 
2.34.1

[PATCH v8 5/8] drm/ttm/tests: Use an init function from the helpers lib

[Karolina Stolarek]

Add a new helper function that also initializes the device. Use it in
ttm_tt test suite and delete the local definition.

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c | 14 ++++++++++++++
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h |  1 +
 drivers/gpu/drm/ttm/tests/ttm_tt_test.c       | 15 +--------------
 3 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
index 7b7c1fa805fc..899a54dbe443 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
@@ -150,6 +150,20 @@ int ttm_test_devices_init(struct kunit *test)
 }
 EXPORT_SYMBOL_GPL(ttm_test_devices_init);
 
+int ttm_test_devices_all_init(struct kunit *test)
+{
+	struct ttm_test_devices *priv;
+
+	priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, priv);
+
+	priv = ttm_test_devices_all(test);
+	test->priv = priv;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ttm_test_devices_all_init);
+
 void ttm_test_devices_fini(struct kunit *test)
 {
 	ttm_test_devices_put(test, test->priv);
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
index 2f51c833a536..53bb5834939f 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
@@ -39,6 +39,7 @@ void ttm_test_devices_put(struct kunit *test, struct ttm_test_devices *devs);
 
 /* Generic init/fini for tests that only need DRM/TTM devices */
 int ttm_test_devices_init(struct kunit *test);
+int ttm_test_devices_all_init(struct kunit *test);
 void ttm_test_devices_fini(struct kunit *test);
 
 #endif // TTM_KUNIT_HELPERS_H
diff --git a/drivers/gpu/drm/ttm/tests/ttm_tt_test.c b/drivers/gpu/drm/ttm/tests/ttm_tt_test.c
index fd4502c18de6..a33a426a814d 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_tt_test.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_tt_test.c
@@ -15,19 +15,6 @@ struct ttm_tt_test_case {
 	uint32_t extra_pages_num;
 };
 
-static int ttm_tt_test_init(struct kunit *test)
-{
-	struct ttm_test_devices *priv;
-
-	priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL);
-	KUNIT_ASSERT_NOT_NULL(test, priv);
-
-	priv = ttm_test_devices_all(test);
-	test->priv = priv;
-
-	return 0;
-}
-
 static const struct ttm_tt_test_case ttm_tt_init_basic_cases[] = {
 	{
 		.description = "Page-aligned size",
@@ -285,7 +272,7 @@ static struct kunit_case ttm_tt_test_cases[] = {
 
 static struct kunit_suite ttm_tt_test_suite = {
 	.name = "ttm_tt",
-	.init = ttm_tt_test_init,
+	.init = ttm_test_devices_all_init,
 	.exit = ttm_test_devices_fini,
 	.test_cases = ttm_tt_test_cases,
 };
-- 
2.34.1

[PATCH v8 6/8] drm/ttm/tests: Test simple BO creation and validation

[Karolina Stolarek]

Add tests for ttm_bo_init_reserved() and ttm_bo_validate() that use
sys manager. Define a simple move function in ttm_device_funcs. Expose
destroy callback of the buffer object to make testing of
ttm_bo_init_reserved() behaviour easier.

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Tested-by: Amaranath Somalapuram <Amaranath.Somalapuram@amd.com>
---
 drivers/gpu/drm/ttm/tests/Makefile            |   1 +
 .../gpu/drm/ttm/tests/ttm_bo_validate_test.c  | 211 ++++++++++++++++++
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c |  14 +-
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h |   1 +
 4 files changed, 226 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c

diff --git a/drivers/gpu/drm/ttm/tests/Makefile b/drivers/gpu/drm/ttm/tests/Makefile
index 468535f7eed2..2e5ed63fb414 100644
--- a/drivers/gpu/drm/ttm/tests/Makefile
+++ b/drivers/gpu/drm/ttm/tests/Makefile
@@ -6,4 +6,5 @@ obj-$(CONFIG_DRM_TTM_KUNIT_TEST) += \
         ttm_resource_test.o \
         ttm_tt_test.o \
         ttm_bo_test.o \
+        ttm_bo_validate_test.o \
         ttm_kunit_helpers.o
diff --git a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
new file mode 100644
index 000000000000..1d50e4ba9775
--- /dev/null
+++ b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
@@ -0,0 +1,211 @@
+// SPDX-License-Identifier: GPL-2.0 AND MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <drm/ttm/ttm_resource.h>
+#include <drm/ttm/ttm_placement.h>
+#include <drm/ttm/ttm_tt.h>
+
+#include "ttm_kunit_helpers.h"
+
+#define BO_SIZE		SZ_4K
+
+struct ttm_bo_validate_test_case {
+	const char *description;
+	enum ttm_bo_type bo_type;
+	bool with_ttm;
+};
+
+static struct ttm_placement *ttm_placement_kunit_init(struct kunit *test,
+						      struct ttm_place *places,
+						      unsigned int num_places,
+						      struct ttm_place *busy_places,
+						      unsigned int num_busy_places)
+{
+	struct ttm_placement *placement;
+
+	placement = kunit_kzalloc(test, sizeof(*placement), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, placement);
+
+	placement->num_placement = num_places;
+	placement->placement = places;
+	placement->num_busy_placement = num_busy_places;
+	placement->busy_placement = busy_places;
+
+	return placement;
+}
+
+static void ttm_bo_validate_case_desc(const struct ttm_bo_validate_test_case *t,
+				      char *desc)
+{
+	strscpy(desc, t->description, KUNIT_PARAM_DESC_SIZE);
+}
+
+static const struct ttm_bo_validate_test_case ttm_bo_type_cases[] = {
+	{
+		.description = "Buffer object for userspace",
+		.bo_type = ttm_bo_type_device,
+	},
+	{
+		.description = "Kernel buffer object",
+		.bo_type = ttm_bo_type_kernel,
+	},
+	{
+		.description = "Shared buffer object",
+		.bo_type = ttm_bo_type_sg,
+	},
+};
+
+KUNIT_ARRAY_PARAM(ttm_bo_types, ttm_bo_type_cases,
+		  ttm_bo_validate_case_desc);
+
+static void ttm_bo_init_reserved_sys_man(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	enum ttm_bo_type bo_type = params->bo_type;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	place = ttm_place_kunit_init(test, TTM_PL_SYSTEM, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, bo_type, placement,
+				   PAGE_SIZE, &ctx, NULL, NULL,
+				   &dummy_ttm_bo_destroy);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, kref_read(&bo->kref), 1);
+	KUNIT_EXPECT_PTR_EQ(test, bo->bdev, priv->ttm_dev);
+	KUNIT_EXPECT_EQ(test, bo->type, bo_type);
+	KUNIT_EXPECT_EQ(test, bo->page_alignment, PAGE_SIZE);
+	KUNIT_EXPECT_PTR_EQ(test, bo->destroy, &dummy_ttm_bo_destroy);
+	KUNIT_EXPECT_EQ(test, bo->pin_count, 0);
+	KUNIT_EXPECT_NULL(test, bo->bulk_move);
+	KUNIT_EXPECT_NOT_NULL(test, bo->ttm);
+	KUNIT_EXPECT_FALSE(test, ttm_tt_is_populated(bo->ttm));
+	KUNIT_EXPECT_NOT_NULL(test, (void *)bo->base.resv->fences);
+	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, size);
+
+	if (bo_type != ttm_bo_type_kernel)
+		KUNIT_EXPECT_TRUE(test,
+				  drm_mm_node_allocated(&bo->base.vma_node.vm_node));
+
+	ttm_resource_free(bo, &bo->resource);
+	ttm_bo_put(bo);
+}
+
+static void ttm_bo_init_reserved_resv(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	struct dma_resv resv;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	place = ttm_place_kunit_init(test, TTM_PL_SYSTEM, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+	dma_resv_init(&resv);
+	dma_resv_lock(&resv, NULL);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, bo_type, placement,
+				   PAGE_SIZE, &ctx, NULL, &resv,
+				   &dummy_ttm_bo_destroy);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_PTR_EQ(test, bo->base.resv, &resv);
+
+	ttm_resource_free(bo, &bo->resource);
+	ttm_bo_put(bo);
+}
+
+static void ttm_bo_validate_invalid_placement(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	uint32_t unknown_mem_type = TTM_PL_PRIV + 1;
+	int err;
+
+	place = ttm_place_kunit_init(test, unknown_mem_type, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	bo = ttm_bo_kunit_init(test, test->priv, size);
+	bo->type = bo_type;
+
+	ttm_bo_reserve(bo, false, false, NULL);
+	err = ttm_bo_validate(bo, placement, &ctx);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, -EINVAL);
+
+	ttm_bo_put(bo);
+}
+
+static void ttm_bo_validate_pinned(struct kunit *test)
+{
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	bo = ttm_bo_kunit_init(test, test->priv, size);
+	bo->type = bo_type;
+
+	ttm_bo_reserve(bo, false, false, NULL);
+	ttm_bo_pin(bo);
+	err = ttm_bo_validate(bo, placement, &ctx);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, -EINVAL);
+}
+
+static struct kunit_case ttm_bo_validate_test_cases[] = {
+	KUNIT_CASE_PARAM(ttm_bo_init_reserved_sys_man, ttm_bo_types_gen_params),
+	KUNIT_CASE(ttm_bo_init_reserved_resv),
+	KUNIT_CASE(ttm_bo_validate_invalid_placement),
+	KUNIT_CASE(ttm_bo_validate_pinned),
+	{}
+};
+
+static struct kunit_suite ttm_bo_validate_test_suite = {
+	.name = "ttm_bo_validate",
+	.init = ttm_test_devices_all_init,
+	.exit = ttm_test_devices_fini,
+	.test_cases = ttm_bo_validate_test_cases,
+};
+
+kunit_test_suites(&ttm_bo_validate_test_suite);
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
index 899a54dbe443..b07ea760d819 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
@@ -22,13 +22,19 @@ static void ttm_tt_simple_destroy(struct ttm_device *bdev, struct ttm_tt *ttm)
 	kfree(ttm);
 }
 
-static void dummy_ttm_bo_destroy(struct ttm_buffer_object *bo)
+static int mock_move(struct ttm_buffer_object *bo, bool evict,
+		     struct ttm_operation_ctx *ctx,
+		     struct ttm_resource *new_mem,
+		     struct ttm_place *hop)
 {
+	bo->resource = new_mem;
+	return 0;
 }
 
 struct ttm_device_funcs ttm_dev_funcs = {
 	.ttm_tt_create = ttm_tt_simple_create,
 	.ttm_tt_destroy = ttm_tt_simple_destroy,
+	.move = mock_move,
 };
 EXPORT_SYMBOL_GPL(ttm_dev_funcs);
 
@@ -88,6 +94,12 @@ struct ttm_place *ttm_place_kunit_init(struct kunit *test,
 }
 EXPORT_SYMBOL_GPL(ttm_place_kunit_init);
 
+void dummy_ttm_bo_destroy(struct ttm_buffer_object *bo)
+{
+	drm_gem_object_release(&bo->base);
+}
+EXPORT_SYMBOL_GPL(dummy_ttm_bo_destroy);
+
 struct ttm_test_devices *ttm_test_devices_basic(struct kunit *test)
 {
 	struct ttm_test_devices *devs;
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
index 53bb5834939f..3e5fe272c54c 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
@@ -31,6 +31,7 @@ struct ttm_buffer_object *ttm_bo_kunit_init(struct kunit *test,
 					    size_t size);
 struct ttm_place *ttm_place_kunit_init(struct kunit *test,
 				       uint32_t mem_type, uint32_t flags);
+void dummy_ttm_bo_destroy(struct ttm_buffer_object *bo);
 
 struct ttm_test_devices *ttm_test_devices_basic(struct kunit *test);
 struct ttm_test_devices *ttm_test_devices_all(struct kunit *test);
-- 
2.34.1

[PATCH v8 7/8] drm/ttm/tests: Add tests with mock resource managers

[Karolina Stolarek]

Add mock resource manager to test ttm_bo_validate() with non-system
placements. Update KConfig entry to enable DRM Buddy allocator, used
by the mock manager. Update move function to do more than just assign
a resource.

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Tested-by: Amaranath Somalapuram <Amaranath.Somalapuram@amd.com>
---
 drivers/gpu/drm/Kconfig                       |   1 +
 drivers/gpu/drm/ttm/tests/.kunitconfig        |   1 +
 drivers/gpu/drm/ttm/tests/Makefile            |   1 +
 .../gpu/drm/ttm/tests/ttm_bo_validate_test.c  | 276 ++++++++++++++++++
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c |  39 ++-
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h |   2 +
 drivers/gpu/drm/ttm/tests/ttm_mock_manager.c  | 207 +++++++++++++
 drivers/gpu/drm/ttm/tests/ttm_mock_manager.h  |  31 ++
 8 files changed, 556 insertions(+), 2 deletions(-)
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.c
 create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.h

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 740c1c0bd068..3be7afa6df3e 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -200,6 +200,7 @@ config DRM_TTM_KUNIT_TEST
         default n
         depends on DRM && KUNIT && MMU
         select DRM_TTM
+        select DRM_BUDDY
         select DRM_EXPORT_FOR_TESTS if m
         select DRM_KUNIT_TEST_HELPERS
         default KUNIT_ALL_TESTS
diff --git a/drivers/gpu/drm/ttm/tests/.kunitconfig b/drivers/gpu/drm/ttm/tests/.kunitconfig
index 75fdce0cd98e..9228ce9b913c 100644
--- a/drivers/gpu/drm/ttm/tests/.kunitconfig
+++ b/drivers/gpu/drm/ttm/tests/.kunitconfig
@@ -2,3 +2,4 @@ CONFIG_KUNIT=y
 CONFIG_DRM=y
 CONFIG_DRM_KUNIT_TEST_HELPERS=y
 CONFIG_DRM_TTM_KUNIT_TEST=y
+CONFIG_DRM_BUDDY=y
diff --git a/drivers/gpu/drm/ttm/tests/Makefile b/drivers/gpu/drm/ttm/tests/Makefile
index 2e5ed63fb414..f3149de77541 100644
--- a/drivers/gpu/drm/ttm/tests/Makefile
+++ b/drivers/gpu/drm/ttm/tests/Makefile
@@ -7,4 +7,5 @@ obj-$(CONFIG_DRM_TTM_KUNIT_TEST) += \
         ttm_tt_test.o \
         ttm_bo_test.o \
         ttm_bo_validate_test.o \
+        ttm_mock_manager.o \
         ttm_kunit_helpers.o
diff --git a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
index 1d50e4ba9775..5f6c24979f83 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
@@ -8,12 +8,15 @@
 #include <drm/ttm/ttm_tt.h>
 
 #include "ttm_kunit_helpers.h"
+#include "ttm_mock_manager.h"
 
 #define BO_SIZE		SZ_4K
+#define MANAGER_SIZE	SZ_1M
 
 struct ttm_bo_validate_test_case {
 	const char *description;
 	enum ttm_bo_type bo_type;
+	uint32_t mem_type;
 	bool with_ttm;
 };
 
@@ -106,6 +109,49 @@ static void ttm_bo_init_reserved_sys_man(struct kunit *test)
 	ttm_bo_put(bo);
 }
 
+static void ttm_bo_init_reserved_mock_man(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	enum ttm_bo_type bo_type = params->bo_type;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t mem_type = TTM_PL_VRAM;
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	ttm_mock_manager_init(priv->ttm_dev, mem_type, MANAGER_SIZE);
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, bo_type, placement,
+				   PAGE_SIZE, &ctx, NULL, NULL,
+				   &dummy_ttm_bo_destroy);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, kref_read(&bo->kref), 1);
+	KUNIT_EXPECT_PTR_EQ(test, bo->bdev, priv->ttm_dev);
+	KUNIT_EXPECT_EQ(test, bo->type, bo_type);
+	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, size);
+
+	if (bo_type != ttm_bo_type_kernel)
+		KUNIT_EXPECT_TRUE(test,
+				  drm_mm_node_allocated(&bo->base.vma_node.vm_node));
+
+	ttm_resource_free(bo, &bo->resource);
+	ttm_bo_put(bo);
+	ttm_mock_manager_fini(priv->ttm_dev, mem_type);
+}
+
 static void ttm_bo_init_reserved_resv(struct kunit *test)
 {
 	struct ttm_test_devices *priv = test->priv;
@@ -140,6 +186,51 @@ static void ttm_bo_init_reserved_resv(struct kunit *test)
 	ttm_bo_put(bo);
 }
 
+static void ttm_bo_validate_basic(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *fst_place, *snd_place;
+	struct ttm_placement *fst_placement, *snd_placement;
+	struct ttm_operation_ctx ctx_init = { }, ctx_val = { };
+	uint32_t size = ALIGN(SZ_8K, PAGE_SIZE);
+	uint32_t fst_mem = TTM_PL_SYSTEM, snd_mem = TTM_PL_VRAM;
+	int err;
+
+	ttm_mock_manager_init(priv->ttm_dev, snd_mem, MANAGER_SIZE);
+
+	fst_place = ttm_place_kunit_init(test, fst_mem, 0);
+	fst_placement = ttm_placement_kunit_init(test, fst_place, 1, NULL, 0);
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, params->bo_type,
+				   fst_placement, PAGE_SIZE, &ctx_init, NULL,
+				   NULL, &dummy_ttm_bo_destroy);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	snd_place = ttm_place_kunit_init(test, snd_mem, DRM_BUDDY_TOPDOWN_ALLOCATION);
+	snd_placement = ttm_placement_kunit_init(test, snd_place, 1, NULL, 0);
+
+	err = ttm_bo_validate(bo, snd_placement, &ctx_val);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, ctx_val.bytes_moved, bo->base.size);
+	KUNIT_EXPECT_NOT_NULL(test, bo->ttm);
+	KUNIT_EXPECT_TRUE(test, ttm_tt_is_populated(bo->ttm));
+	KUNIT_EXPECT_EQ(test, bo->resource->mem_type, snd_mem);
+	KUNIT_EXPECT_EQ(test, bo->resource->placement,
+			DRM_BUDDY_TOPDOWN_ALLOCATION);
+
+	ttm_bo_put(bo);
+	ttm_mock_manager_fini(priv->ttm_dev, snd_mem);
+}
+
 static void ttm_bo_validate_invalid_placement(struct kunit *test)
 {
 	struct ttm_buffer_object *bo;
@@ -166,6 +257,36 @@ static void ttm_bo_validate_invalid_placement(struct kunit *test)
 	ttm_bo_put(bo);
 }
 
+static void ttm_bo_validate_failed_alloc(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	uint32_t mem_type = TTM_PL_VRAM;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	bo = ttm_bo_kunit_init(test, test->priv, size);
+	bo->type = bo_type;
+
+	ttm_bad_manager_init(priv->ttm_dev, mem_type, MANAGER_SIZE);
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	ttm_bo_reserve(bo, false, false, NULL);
+	err = ttm_bo_validate(bo, placement, &ctx);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, -ENOMEM);
+
+	ttm_bo_put(bo);
+	ttm_bad_manager_fini(priv->ttm_dev, mem_type);
+}
+
 static void ttm_bo_validate_pinned(struct kunit *test)
 {
 	struct ttm_buffer_object *bo;
@@ -191,11 +312,166 @@ static void ttm_bo_validate_pinned(struct kunit *test)
 	KUNIT_EXPECT_EQ(test, err, -EINVAL);
 }
 
+static const struct ttm_bo_validate_test_case ttm_mem_type_cases[] = {
+	{
+		.description = "System manager",
+		.mem_type = TTM_PL_SYSTEM,
+	},
+	{
+		.description = "VRAM manager",
+		.mem_type = TTM_PL_SYSTEM,
+	},
+};
+
+KUNIT_ARRAY_PARAM(ttm_bo_validate_mem, ttm_mem_type_cases,
+		  ttm_bo_validate_case_desc);
+
+static void ttm_bo_validate_same_placement(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	struct ttm_operation_ctx ctx_init = { }, ctx_val = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	place = ttm_place_kunit_init(test, params->mem_type, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	if (params->mem_type != TTM_PL_SYSTEM)
+		ttm_mock_manager_init(priv->ttm_dev, params->mem_type, MANAGER_SIZE);
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, params->bo_type,
+				   placement, PAGE_SIZE, &ctx_init, NULL,
+				   NULL, &dummy_ttm_bo_destroy);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	err = ttm_bo_validate(bo, placement, &ctx_val);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, ctx_val.bytes_moved, 0);
+
+	if (params->mem_type != TTM_PL_SYSTEM)
+		ttm_mock_manager_fini(priv->ttm_dev, params->mem_type);
+
+	ttm_bo_put(bo);
+}
+
+static void ttm_bo_validate_busy_placement(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *init_place, *place_bad, *place_busy;
+	struct ttm_placement *placement_init, *placement_val;
+	struct ttm_resource_manager *man;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	struct ttm_operation_ctx ctx_init = { }, ctx_val = { };
+	uint32_t fst_mem = TTM_PL_VRAM, snd_mem = TTM_PL_VRAM + 1;
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	ttm_bad_manager_init(priv->ttm_dev, fst_mem, MANAGER_SIZE);
+	ttm_mock_manager_init(priv->ttm_dev, snd_mem, MANAGER_SIZE);
+
+	init_place = ttm_place_kunit_init(test, TTM_PL_SYSTEM, 0);
+	placement_init = ttm_placement_kunit_init(test, init_place, 1, NULL, 0);
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, bo_type, placement_init,
+				   PAGE_SIZE, &ctx_init, NULL, NULL,
+				   &dummy_ttm_bo_destroy);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	place_bad = ttm_place_kunit_init(test, fst_mem, 0);
+	place_busy = ttm_place_kunit_init(test, snd_mem, 0);
+	placement_val = ttm_placement_kunit_init(test, place_bad, 1, place_busy, 1);
+
+	err = ttm_bo_validate(bo, placement_val, &ctx_val);
+	dma_resv_unlock(bo->base.resv);
+
+	man = ttm_manager_type(priv->ttm_dev, place_busy->mem_type);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, ctx_val.bytes_moved, bo->base.size);
+	KUNIT_EXPECT_EQ(test, bo->resource->mem_type, snd_mem);
+	KUNIT_ASSERT_TRUE(test, list_is_singular(&man->lru[bo->priority]));
+
+	ttm_bo_put(bo);
+	ttm_bad_manager_fini(priv->ttm_dev, fst_mem);
+	ttm_mock_manager_fini(priv->ttm_dev, snd_mem);
+}
+
+static void ttm_bo_validate_multihop(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *fst_place, *final_place;
+	struct ttm_placement *placement_init, *placement_val;
+	struct ttm_operation_ctx ctx_init = { }, ctx_val = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	uint32_t fst_mem = TTM_PL_VRAM,  tmp_mem = TTM_PL_TT,
+		 final_mem = TTM_PL_MULTIHOP;
+	int err;
+
+	ttm_mock_manager_init(priv->ttm_dev, fst_mem, MANAGER_SIZE);
+	ttm_mock_manager_init(priv->ttm_dev, tmp_mem, MANAGER_SIZE);
+	ttm_mock_manager_init(priv->ttm_dev, final_mem, MANAGER_SIZE);
+
+	fst_place = ttm_place_kunit_init(test, fst_mem, 0);
+	placement_init = ttm_placement_kunit_init(test, fst_place, 1, NULL, 0);
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, params->bo_type,
+				   placement_init, PAGE_SIZE, &ctx_init, NULL,
+				   NULL, &dummy_ttm_bo_destroy);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	final_place = ttm_place_kunit_init(test, final_mem, 0);
+	placement_val = ttm_placement_kunit_init(test, final_place, 1, NULL, 0);
+
+	err = ttm_bo_validate(bo, placement_val, &ctx_val);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, ctx_val.bytes_moved, size * 2);
+	KUNIT_EXPECT_EQ(test, bo->resource->mem_type, final_mem);
+
+	ttm_bo_put(bo);
+
+	ttm_mock_manager_fini(priv->ttm_dev, fst_mem);
+	ttm_mock_manager_fini(priv->ttm_dev, tmp_mem);
+	ttm_mock_manager_fini(priv->ttm_dev, final_mem);
+}
+
 static struct kunit_case ttm_bo_validate_test_cases[] = {
 	KUNIT_CASE_PARAM(ttm_bo_init_reserved_sys_man, ttm_bo_types_gen_params),
+	KUNIT_CASE_PARAM(ttm_bo_init_reserved_mock_man, ttm_bo_types_gen_params),
 	KUNIT_CASE(ttm_bo_init_reserved_resv),
+	KUNIT_CASE_PARAM(ttm_bo_validate_basic, ttm_bo_types_gen_params),
 	KUNIT_CASE(ttm_bo_validate_invalid_placement),
+	KUNIT_CASE_PARAM(ttm_bo_validate_same_placement,
+			 ttm_bo_validate_mem_gen_params),
+	KUNIT_CASE(ttm_bo_validate_failed_alloc),
 	KUNIT_CASE(ttm_bo_validate_pinned),
+	KUNIT_CASE(ttm_bo_validate_busy_placement),
+	KUNIT_CASE_PARAM(ttm_bo_validate_multihop, ttm_bo_types_gen_params),
 	{}
 };
 
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
index b07ea760d819..719c1d56388e 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
@@ -27,8 +27,43 @@ static int mock_move(struct ttm_buffer_object *bo, bool evict,
 		     struct ttm_resource *new_mem,
 		     struct ttm_place *hop)
 {
-	bo->resource = new_mem;
-	return 0;
+	struct ttm_resource *old_mem = bo->resource;
+	int ret;
+
+	if (!old_mem || (old_mem->mem_type == TTM_PL_SYSTEM && !bo->ttm)) {
+		ttm_bo_move_null(bo, new_mem);
+		return 0;
+	}
+
+	/* Dummy multihop */
+	if (bo->resource->mem_type == TTM_PL_VRAM &&
+	    new_mem->mem_type == TTM_PL_MULTIHOP) {
+		hop->mem_type = TTM_PL_TT;
+		hop->flags = TTM_PL_FLAG_TEMPORARY;
+		hop->fpfn = 0;
+		hop->lpfn = 0;
+		return -EMULTIHOP;
+	}
+
+	if (old_mem->mem_type == TTM_PL_SYSTEM &&
+	    new_mem->mem_type == TTM_PL_TT) {
+		ttm_bo_move_null(bo, new_mem);
+		return 0;
+	}
+
+	if (old_mem->mem_type == TTM_PL_TT &&
+	    new_mem->mem_type == TTM_PL_SYSTEM) {
+		ret = ttm_bo_wait_ctx(bo, ctx);
+
+		if (ret)
+			return ret;
+
+		ttm_resource_free(bo, &bo->resource);
+		ttm_bo_assign_mem(bo, new_mem);
+		return 0;
+	}
+
+	return ttm_bo_move_memcpy(bo, ctx, new_mem);
 }
 
 struct ttm_device_funcs ttm_dev_funcs = {
diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
index 3e5fe272c54c..a8977eb606e5 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h
@@ -13,6 +13,8 @@
 #include <drm/drm_kunit_helpers.h>
 #include <kunit/test.h>
 
+#define TTM_PL_MULTIHOP (TTM_PL_PRIV + 1)
+
 extern struct ttm_device_funcs ttm_dev_funcs;
 
 struct ttm_test_devices {
diff --git a/drivers/gpu/drm/ttm/tests/ttm_mock_manager.c b/drivers/gpu/drm/ttm/tests/ttm_mock_manager.c
new file mode 100644
index 000000000000..21636b96ac38
--- /dev/null
+++ b/drivers/gpu/drm/ttm/tests/ttm_mock_manager.c
@@ -0,0 +1,207 @@
+// SPDX-License-Identifier: GPL-2.0 AND MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+#include <drm/ttm/ttm_resource.h>
+#include <drm/ttm/ttm_device.h>
+#include <drm/ttm/ttm_placement.h>
+
+#include "ttm_mock_manager.h"
+
+static inline struct ttm_mock_manager *
+to_mock_mgr(struct ttm_resource_manager *man)
+{
+	return container_of(man, struct ttm_mock_manager, man);
+}
+
+static inline struct ttm_mock_resource *
+to_mock_mgr_resource(struct ttm_resource *res)
+{
+	return container_of(res, struct ttm_mock_resource, base);
+}
+
+static int ttm_mock_manager_alloc(struct ttm_resource_manager *man,
+				  struct ttm_buffer_object *bo,
+				  const struct ttm_place *place,
+				  struct ttm_resource **res)
+{
+	struct ttm_mock_manager *manager = to_mock_mgr(man);
+	struct ttm_mock_resource *mock_res;
+	struct drm_buddy *mm = &manager->mm;
+	uint64_t lpfn, fpfn, alloc_size;
+	int err;
+
+	mock_res = kzalloc(sizeof(*mock_res), GFP_KERNEL);
+
+	if (!mock_res)
+		return -ENOMEM;
+
+	fpfn = 0;
+	lpfn = man->size;
+
+	ttm_resource_init(bo, place, &mock_res->base);
+	INIT_LIST_HEAD(&mock_res->blocks);
+
+	if (place->flags & TTM_PL_FLAG_TOPDOWN)
+		mock_res->flags |= DRM_BUDDY_TOPDOWN_ALLOCATION;
+
+	if (place->flags & TTM_PL_FLAG_CONTIGUOUS)
+		mock_res->flags |= DRM_BUDDY_CONTIGUOUS_ALLOCATION;
+
+	alloc_size = (uint64_t)mock_res->base.size;
+	mutex_lock(&manager->lock);
+	err = drm_buddy_alloc_blocks(mm, fpfn, lpfn, alloc_size,
+				     manager->default_page_size,
+				     &mock_res->blocks,
+				     mock_res->flags);
+
+	if (err)
+		goto error_free_blocks;
+	mutex_unlock(&manager->lock);
+
+	*res = &mock_res->base;
+	return 0;
+
+error_free_blocks:
+	drm_buddy_free_list(mm, &mock_res->blocks);
+	ttm_resource_fini(man, &mock_res->base);
+	mutex_unlock(&manager->lock);
+
+	return err;
+}
+
+static void ttm_mock_manager_free(struct ttm_resource_manager *man,
+				  struct ttm_resource *res)
+{
+	struct ttm_mock_manager *manager = to_mock_mgr(man);
+	struct ttm_mock_resource *mock_res = to_mock_mgr_resource(res);
+	struct drm_buddy *mm = &manager->mm;
+
+	mutex_lock(&manager->lock);
+	drm_buddy_free_list(mm, &mock_res->blocks);
+	mutex_unlock(&manager->lock);
+
+	ttm_resource_fini(man, res);
+	kfree(mock_res);
+}
+
+static const struct ttm_resource_manager_func ttm_mock_manager_funcs = {
+	.alloc = ttm_mock_manager_alloc,
+	.free = ttm_mock_manager_free,
+};
+
+int ttm_mock_manager_init(struct ttm_device *bdev, uint32_t mem_type, uint32_t size)
+{
+	struct ttm_mock_manager *manager;
+	struct ttm_resource_manager *base;
+	int err;
+
+	manager = kzalloc(sizeof(*manager), GFP_KERNEL);
+	if (!manager)
+		return -ENOMEM;
+
+	mutex_init(&manager->lock);
+
+	err = drm_buddy_init(&manager->mm, size, PAGE_SIZE);
+
+	if (err) {
+		kfree(manager);
+		return err;
+	}
+
+	manager->default_page_size = PAGE_SIZE;
+	base = &manager->man;
+	base->func = &ttm_mock_manager_funcs;
+	base->use_tt = true;
+
+	ttm_resource_manager_init(base, bdev, size);
+	ttm_set_driver_manager(bdev, mem_type, base);
+	ttm_resource_manager_set_used(base, true);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ttm_mock_manager_init);
+
+void ttm_mock_manager_fini(struct ttm_device *bdev, uint32_t mem_type)
+{
+	struct ttm_resource_manager *man;
+	struct ttm_mock_manager *mock_man;
+	int err;
+
+	man = ttm_manager_type(bdev, mem_type);
+	mock_man = to_mock_mgr(man);
+
+	err = ttm_resource_manager_evict_all(bdev, man);
+	if (err)
+		return;
+
+	ttm_resource_manager_set_used(man, false);
+
+	mutex_lock(&mock_man->lock);
+	drm_buddy_fini(&mock_man->mm);
+	mutex_unlock(&mock_man->lock);
+
+	ttm_set_driver_manager(bdev, mem_type, NULL);
+}
+EXPORT_SYMBOL_GPL(ttm_mock_manager_fini);
+
+static int ttm_bad_manager_alloc(struct ttm_resource_manager *man,
+				 struct ttm_buffer_object *bo,
+				 const struct ttm_place *place,
+				 struct ttm_resource **res)
+{
+	return -ENOSPC;
+}
+
+static void ttm_bad_manager_free(struct ttm_resource_manager *man,
+				 struct ttm_resource *res)
+{
+}
+
+static bool ttm_bad_manager_compatible(struct ttm_resource_manager *man,
+				       struct ttm_resource *res,
+				       const struct ttm_place *place,
+				       size_t size)
+{
+	return true;
+}
+
+static const struct ttm_resource_manager_func ttm_bad_manager_funcs = {
+	.alloc = ttm_bad_manager_alloc,
+	.free = ttm_bad_manager_free,
+	.compatible = ttm_bad_manager_compatible
+};
+
+int ttm_bad_manager_init(struct ttm_device *bdev, uint32_t mem_type,
+			 uint32_t size)
+{
+	struct ttm_resource_manager *man;
+
+	man = kzalloc(sizeof(*man), GFP_KERNEL);
+	if (!man)
+		return -ENOMEM;
+
+	man->func = &ttm_bad_manager_funcs;
+
+	ttm_resource_manager_init(man, bdev, size);
+	ttm_set_driver_manager(bdev, mem_type, man);
+	ttm_resource_manager_set_used(man, true);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ttm_bad_manager_init);
+
+void ttm_bad_manager_fini(struct ttm_device *bdev, uint32_t mem_type)
+{
+	struct ttm_resource_manager *man;
+
+	man = ttm_manager_type(bdev, mem_type);
+
+	ttm_resource_manager_set_used(man, false);
+	ttm_set_driver_manager(bdev, mem_type, NULL);
+
+	kfree(man);
+}
+EXPORT_SYMBOL_GPL(ttm_bad_manager_fini);
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/ttm/tests/ttm_mock_manager.h b/drivers/gpu/drm/ttm/tests/ttm_mock_manager.h
new file mode 100644
index 000000000000..d2db9de9d876
--- /dev/null
+++ b/drivers/gpu/drm/ttm/tests/ttm_mock_manager.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0 AND MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+#ifndef TTM_MOCK_MANAGER_H
+#define TTM_MOCK_MANAGER_H
+
+#include <drm/drm_buddy.h>
+
+struct ttm_mock_manager {
+	struct ttm_resource_manager man;
+	struct drm_buddy mm;
+	uint64_t default_page_size;
+	/* protects allocations of mock buffer objects */
+	struct mutex lock;
+};
+
+struct ttm_mock_resource {
+	struct ttm_resource base;
+	struct list_head blocks;
+	unsigned long flags;
+};
+
+int ttm_mock_manager_init(struct ttm_device *bdev, uint32_t mem_type,
+			  uint32_t size);
+int ttm_bad_manager_init(struct ttm_device *bdev, uint32_t mem_type,
+			 uint32_t size);
+void ttm_mock_manager_fini(struct ttm_device *bdev, uint32_t mem_type);
+void ttm_bad_manager_fini(struct ttm_device *bdev, uint32_t mem_type);
+
+#endif // TTM_MOCK_MANAGER_H
-- 
2.34.1

[PATCH v8 8/8] drm/ttm/tests: Add test cases dependent on fence signaling

[Karolina Stolarek]

Add test cases that check how the state of dma fences in BO's
reservation object influence the ttm_bo_validation() flow. Do similar
tests for resource manager's move fence.

Signed-off-by: Karolina Stolarek <karolina.stolarek@intel.com>
Tested-by: Amaranath Somalapuram <Amaranath.Somalapuram@amd.com>
---
 .../gpu/drm/ttm/tests/ttm_bo_validate_test.c  | 308 ++++++++++++++++++
 1 file changed, 308 insertions(+)

diff --git a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
index 5f6c24979f83..a659fc837186 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
@@ -2,6 +2,8 @@
 /*
  * Copyright © 2023 Intel Corporation
  */
+#include <linux/delay.h>
+#include <linux/kthread.h>
 
 #include <drm/ttm/ttm_resource.h>
 #include <drm/ttm/ttm_placement.h>
@@ -13,11 +15,14 @@
 #define BO_SIZE		SZ_4K
 #define MANAGER_SIZE	SZ_1M
 
+static struct spinlock fence_lock;
+
 struct ttm_bo_validate_test_case {
 	const char *description;
 	enum ttm_bo_type bo_type;
 	uint32_t mem_type;
 	bool with_ttm;
+	bool no_gpu_wait;
 };
 
 static struct ttm_placement *ttm_placement_kunit_init(struct kunit *test,
@@ -39,6 +44,43 @@ static struct ttm_placement *ttm_placement_kunit_init(struct kunit *test,
 	return placement;
 }
 
+static const char *fence_name(struct dma_fence *f)
+{
+	return "ttm-bo-validate-fence";
+}
+
+static const struct dma_fence_ops fence_ops = {
+	.get_driver_name = fence_name,
+	.get_timeline_name = fence_name,
+};
+
+static struct dma_fence *alloc_mock_fence(struct kunit *test)
+{
+	struct dma_fence *fence;
+
+	fence = kunit_kzalloc(test, sizeof(*fence), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, fence);
+
+	dma_fence_init(fence, &fence_ops, &fence_lock, 0, 0);
+
+	return fence;
+}
+
+static void dma_resv_kunit_active_fence_init(struct kunit *test,
+					     struct dma_resv *resv,
+					     enum dma_resv_usage usage)
+{
+	struct dma_fence *fence;
+
+	fence = alloc_mock_fence(test);
+	dma_fence_enable_sw_signaling(fence);
+
+	dma_resv_lock(resv, NULL);
+	dma_resv_reserve_fences(resv, 1);
+	dma_resv_add_fence(resv, fence, usage);
+	dma_resv_unlock(resv);
+}
+
 static void ttm_bo_validate_case_desc(const struct ttm_bo_validate_test_case *t,
 				      char *desc)
 {
@@ -460,6 +502,265 @@ static void ttm_bo_validate_multihop(struct kunit *test)
 	ttm_mock_manager_fini(priv->ttm_dev, final_mem);
 }
 
+static const struct ttm_bo_validate_test_case ttm_bo_no_placement_cases[] = {
+	{
+		.description = "Buffer object in system domain, no page vector",
+	},
+	{
+		.description = "Buffer object in system domain with an existing page vector",
+		.with_ttm = true,
+	},
+};
+
+KUNIT_ARRAY_PARAM(ttm_bo_no_placement, ttm_bo_no_placement_cases,
+		  ttm_bo_validate_case_desc);
+
+static void ttm_bo_validate_no_placement_signaled(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_tt *old_tt;
+	struct ttm_placement *placement;
+	struct ttm_resource_manager *man;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	uint32_t flags;
+	int err;
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+	man = ttm_manager_type(priv->ttm_dev, mem_type);
+
+	bo = ttm_bo_kunit_init(test, test->priv, size);
+	bo->type = bo_type;
+
+	if (params->with_ttm) {
+		old_tt = priv->ttm_dev->funcs->ttm_tt_create(bo, 0);
+		ttm_pool_alloc(&priv->ttm_dev->pool, old_tt, &ctx);
+		bo->ttm = old_tt;
+	}
+
+	err = ttm_resource_alloc(bo, place, &bo->resource);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_ASSERT_EQ(test, man->usage, size);
+
+	placement = kunit_kzalloc(test, sizeof(*placement), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, placement);
+
+	ttm_bo_reserve(bo, false, false, NULL);
+	err = ttm_bo_validate(bo, placement, &ctx);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_ASSERT_EQ(test, man->usage, 0);
+	KUNIT_ASSERT_NOT_NULL(test, bo->ttm);
+	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, 0);
+
+	if (params->with_ttm) {
+		flags = bo->ttm->page_flags;
+
+		KUNIT_ASSERT_PTR_EQ(test, bo->ttm, old_tt);
+		KUNIT_ASSERT_FALSE(test, flags & TTM_TT_FLAG_PRIV_POPULATED);
+		KUNIT_ASSERT_TRUE(test, flags & TTM_TT_FLAG_ZERO_ALLOC);
+	}
+
+	ttm_bo_put(bo);
+}
+
+static int threaded_dma_resv_signal(void *arg)
+{
+	struct ttm_buffer_object *bo = arg;
+	struct dma_resv *resv = bo->base.resv;
+	struct dma_resv_iter cursor;
+	struct dma_fence *fence;
+
+	dma_resv_iter_begin(&cursor, resv, DMA_RESV_USAGE_BOOKKEEP);
+	dma_resv_for_each_fence_unlocked(&cursor, fence) {
+		dma_fence_signal(fence);
+	}
+	dma_resv_iter_end(&cursor);
+
+	return 0;
+}
+
+static void ttm_bo_validate_no_placement_not_signaled(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	struct task_struct *task;
+	enum dma_resv_usage usage = DMA_RESV_USAGE_BOOKKEEP;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+
+	bo = ttm_bo_kunit_init(test, test->priv, size);
+	bo->type = params->bo_type;
+
+	err = ttm_resource_alloc(bo, place, &bo->resource);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	placement = kunit_kzalloc(test, sizeof(*placement), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, placement);
+
+	/* Create an active fence to simulate a non-idle resv object */
+	spin_lock_init(&fence_lock);
+	dma_resv_kunit_active_fence_init(test, bo->base.resv, usage);
+
+	task = kthread_create(threaded_dma_resv_signal, bo, "dma-resv-signal");
+	if (IS_ERR(task))
+		KUNIT_FAIL(test, "Couldn't create dma resv signal task\n");
+
+	wake_up_process(task);
+	ttm_bo_reserve(bo, false, false, NULL);
+	err = ttm_bo_validate(bo, placement, &ctx);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_ASSERT_NOT_NULL(test, bo->ttm);
+	KUNIT_ASSERT_NULL(test, bo->resource);
+	KUNIT_ASSERT_NULL(test, bo->bulk_move);
+	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, 0);
+
+	if (bo->type != ttm_bo_type_sg)
+		KUNIT_ASSERT_PTR_EQ(test, bo->base.resv, &bo->base._resv);
+
+	/* Make sure we have an idle object at this point */
+	dma_resv_wait_timeout(bo->base.resv, usage, false, MAX_SCHEDULE_TIMEOUT);
+
+	ttm_bo_put(bo);
+}
+
+static void ttm_bo_validate_move_fence_signaled(struct kunit *test)
+{
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *place;
+	struct ttm_placement *placement;
+	struct ttm_resource_manager *man;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	uint32_t mem_type = TTM_PL_SYSTEM;
+	struct ttm_operation_ctx ctx = { };
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	int err;
+
+	man = ttm_manager_type(priv->ttm_dev, mem_type);
+	spin_lock_init(&fence_lock);
+	man->move = dma_fence_get_stub();
+
+	bo = ttm_bo_kunit_init(test, test->priv, size);
+	bo->type = bo_type;
+
+	place = ttm_place_kunit_init(test, mem_type, 0);
+	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
+
+	ttm_bo_reserve(bo, false, false, NULL);
+	err = ttm_bo_validate(bo, placement, &ctx);
+	dma_resv_unlock(bo->base.resv);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, bo->resource->mem_type, mem_type);
+	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, size);
+
+	ttm_bo_put(bo);
+	dma_fence_put(man->move);
+}
+
+static const struct ttm_bo_validate_test_case ttm_bo_validate_wait_cases[] = {
+	{
+		.description = "Waits for GPU",
+		.no_gpu_wait = false,
+	},
+	{
+		.description = "Tries to lock straight away",
+		.no_gpu_wait = true,
+	},
+};
+
+KUNIT_ARRAY_PARAM(ttm_bo_validate_wait, ttm_bo_validate_wait_cases,
+		  ttm_bo_validate_case_desc);
+
+static int threaded_fence_signal(void *arg)
+{
+	struct dma_fence *fence = arg;
+	int err;
+
+	msleep(20);
+	err = dma_fence_signal(fence);
+
+	return err;
+}
+
+static void ttm_bo_validate_move_fence_not_signaled(struct kunit *test)
+{
+	const struct ttm_bo_validate_test_case *params = test->param_value;
+	struct ttm_test_devices *priv = test->priv;
+	struct ttm_buffer_object *bo;
+	struct ttm_place *init_place, *fst_place, *snd_place;
+	struct ttm_placement *placement_init, *placement_val;
+	struct task_struct *task;
+	struct ttm_resource_manager *man;
+	enum ttm_bo_type bo_type = ttm_bo_type_device;
+	uint32_t fst_mem = TTM_PL_VRAM, snd_mem = TTM_PL_VRAM + 1;
+	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
+	struct ttm_operation_ctx ctx_init = { },
+				 ctx_val  = { .no_wait_gpu = params->no_gpu_wait };
+	int err;
+
+	init_place = ttm_place_kunit_init(test, TTM_PL_SYSTEM, 0);
+	placement_init = ttm_placement_kunit_init(test, init_place, 1, NULL, 0);
+
+	bo = kunit_kzalloc(test, sizeof(*bo), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, bo);
+
+	drm_gem_private_object_init(priv->drm, &bo->base, size);
+
+	err = ttm_bo_init_reserved(priv->ttm_dev, bo, bo_type, placement_init,
+				   PAGE_SIZE, &ctx_init, NULL, NULL,
+				   &dummy_ttm_bo_destroy);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	ttm_mock_manager_init(priv->ttm_dev, fst_mem, MANAGER_SIZE);
+	ttm_mock_manager_init(priv->ttm_dev, snd_mem, MANAGER_SIZE);
+
+	fst_place = ttm_place_kunit_init(test, fst_mem, 0);
+	snd_place = ttm_place_kunit_init(test, snd_mem, 0);
+	placement_val = ttm_placement_kunit_init(test, fst_place, 1, snd_place, 1);
+
+	spin_lock_init(&fence_lock);
+	man = ttm_manager_type(priv->ttm_dev, fst_place->mem_type);
+	man->move = alloc_mock_fence(test);
+
+	task = kthread_create(threaded_fence_signal, man->move, "move-fence-signal");
+	if (IS_ERR(task))
+		KUNIT_FAIL(test, "Couldn't create move fence signal task\n");
+
+	wake_up_process(task);
+	err = ttm_bo_validate(bo, placement_val, &ctx_val);
+	dma_resv_unlock(bo->base.resv);
+
+	dma_fence_wait_timeout(man->move, false, MAX_SCHEDULE_TIMEOUT);
+
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_EQ(test, ctx_val.bytes_moved, size);
+
+	if (params->no_gpu_wait)
+		KUNIT_EXPECT_EQ(test, bo->resource->mem_type, snd_mem);
+	else
+		KUNIT_EXPECT_EQ(test, bo->resource->mem_type, fst_mem);
+
+	ttm_bo_put(bo);
+	ttm_mock_manager_fini(priv->ttm_dev, fst_mem);
+	ttm_mock_manager_fini(priv->ttm_dev, snd_mem);
+}
+
 static struct kunit_case ttm_bo_validate_test_cases[] = {
 	KUNIT_CASE_PARAM(ttm_bo_init_reserved_sys_man, ttm_bo_types_gen_params),
 	KUNIT_CASE_PARAM(ttm_bo_init_reserved_mock_man, ttm_bo_types_gen_params),
@@ -472,6 +773,13 @@ static struct kunit_case ttm_bo_validate_test_cases[] = {
 	KUNIT_CASE(ttm_bo_validate_pinned),
 	KUNIT_CASE(ttm_bo_validate_busy_placement),
 	KUNIT_CASE_PARAM(ttm_bo_validate_multihop, ttm_bo_types_gen_params),
+	KUNIT_CASE_PARAM(ttm_bo_validate_no_placement_signaled,
+			 ttm_bo_no_placement_gen_params),
+	KUNIT_CASE_PARAM(ttm_bo_validate_no_placement_not_signaled,
+			 ttm_bo_types_gen_params),
+	KUNIT_CASE(ttm_bo_validate_move_fence_signaled),
+	KUNIT_CASE_PARAM(ttm_bo_validate_move_fence_not_signaled,
+			 ttm_bo_validate_wait_gen_params),
 	{}
 };
 
-- 
2.34.1

Re: [PATCH v8 8/8] drm/ttm/tests: Add test cases dependent on fence signaling

[Andi Shyti]

Hi Karolina,

...

> +static void ttm_bo_validate_no_placement_signaled(struct kunit *test)
> +{
> +	const struct ttm_bo_validate_test_case *params = test->param_value;
> +	struct ttm_test_devices *priv = test->priv;
> +	struct ttm_buffer_object *bo;
> +	struct ttm_place *place;
> +	struct ttm_tt *old_tt;
> +	struct ttm_placement *placement;
> +	struct ttm_resource_manager *man;
> +	uint32_t mem_type = TTM_PL_SYSTEM;
> +	enum ttm_bo_type bo_type = ttm_bo_type_device;
> +	struct ttm_operation_ctx ctx = { };
> +	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
> +	uint32_t flags;
> +	int err;
> +
> +	place = ttm_place_kunit_init(test, mem_type, 0);
> +	man = ttm_manager_type(priv->ttm_dev, mem_type);

For next time, I find it difficult to follow all these variables,
it's easier to read

	man = ttm_manager_type(priv->ttm_dev, TTM_PL_SYSTEM);

than

	mem_type = TTM_PL_SYSTEM;
	...
	...
	...
	man = ttm_manager_type(priv->ttm_dev, mem_type);


> +	bo = ttm_bo_kunit_init(test, test->priv, size);
> +	bo->type = bo_type;

same here... the bo_type variable is not giving any value.

	bo->type = ttm_bo_type_device;

is way more readable. You keep doing this all the way and I need
to check everytime what's the value in the declaration :-)

I'm not going to comment on this anymore.

> +	if (params->with_ttm) {
> +		old_tt = priv->ttm_dev->funcs->ttm_tt_create(bo, 0);
> +		ttm_pool_alloc(&priv->ttm_dev->pool, old_tt, &ctx);
> +		bo->ttm = old_tt;
> +	}
> +
> +	err = ttm_resource_alloc(bo, place, &bo->resource);
> +	KUNIT_EXPECT_EQ(test, err, 0);
> +	KUNIT_ASSERT_EQ(test, man->usage, size);
> +
> +	placement = kunit_kzalloc(test, sizeof(*placement), GFP_KERNEL);
> +	KUNIT_ASSERT_NOT_NULL(test, placement);
> +
> +	ttm_bo_reserve(bo, false, false, NULL);
> +	err = ttm_bo_validate(bo, placement, &ctx);
> +	dma_resv_unlock(bo->base.resv);

why don't you use here ttm_bo_unreserve()?

> +	KUNIT_EXPECT_EQ(test, err, 0);
> +	KUNIT_ASSERT_EQ(test, man->usage, 0);
> +	KUNIT_ASSERT_NOT_NULL(test, bo->ttm);
> +	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, 0);
> +
> +	if (params->with_ttm) {
> +		flags = bo->ttm->page_flags;
> +
> +		KUNIT_ASSERT_PTR_EQ(test, bo->ttm, old_tt);
> +		KUNIT_ASSERT_FALSE(test, flags & TTM_TT_FLAG_PRIV_POPULATED);
> +		KUNIT_ASSERT_TRUE(test, flags & TTM_TT_FLAG_ZERO_ALLOC);
> +	}
> +
> +	ttm_bo_put(bo);
> +}

...

> +static void ttm_bo_validate_move_fence_signaled(struct kunit *test)
> +{
> +	struct ttm_test_devices *priv = test->priv;
> +	struct ttm_buffer_object *bo;
> +	struct ttm_place *place;
> +	struct ttm_placement *placement;
> +	struct ttm_resource_manager *man;
> +	enum ttm_bo_type bo_type = ttm_bo_type_device;
> +	uint32_t mem_type = TTM_PL_SYSTEM;
> +	struct ttm_operation_ctx ctx = { };
> +	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
> +	int err;
> +
> +	man = ttm_manager_type(priv->ttm_dev, mem_type);
> +	spin_lock_init(&fence_lock);

where are we using the fence_lock here?

> +	man->move = dma_fence_get_stub();
> +
> +	bo = ttm_bo_kunit_init(test, test->priv, size);
> +	bo->type = bo_type;
> +	place = ttm_place_kunit_init(test, mem_type, 0);
> +	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
> +
> +	ttm_bo_reserve(bo, false, false, NULL);
> +	err = ttm_bo_validate(bo, placement, &ctx);
> +	dma_resv_unlock(bo->base.resv);
> +
> +	KUNIT_EXPECT_EQ(test, err, 0);
> +	KUNIT_EXPECT_EQ(test, bo->resource->mem_type, mem_type);
> +	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, size);

Do we want to check also bo->ttm here?

> +	ttm_bo_put(bo);
> +	dma_fence_put(man->move);
> +}
> +
> +static const struct ttm_bo_validate_test_case ttm_bo_validate_wait_cases[] = {
> +	{
> +		.description = "Waits for GPU",
> +		.no_gpu_wait = false,
> +	},
> +	{
> +		.description = "Tries to lock straight away",
> +		.no_gpu_wait = true,
> +	},
> +};
> +
> +KUNIT_ARRAY_PARAM(ttm_bo_validate_wait, ttm_bo_validate_wait_cases,
> +		  ttm_bo_validate_case_desc);
> +
> +static int threaded_fence_signal(void *arg)
> +{
> +	struct dma_fence *fence = arg;
> +	int err;
> +
> +	msleep(20);
> +	err = dma_fence_signal(fence);
> +
> +	return err;

if you do here "return dma_fence_signal(fence);" you don't need
for err.

Not a binding review, though, your choice.

Andi

...

Re: [PATCH v8 0/8] Improve test coverage of TTM

[Andi Shyti]

Hi Karolina and Christian,

> Karolina Stolarek (8):
>   drm/ttm/tests: Add tests for ttm_resource and ttm_sys_man
>   drm/ttm/tests: Add tests for ttm_tt
>   drm/ttm/tests: Add tests for ttm_bo functions
>   drm/ttm/tests: Fix argument in ttm_tt_kunit_init()
>   drm/ttm/tests: Use an init function from the helpers lib
>   drm/ttm/tests: Test simple BO creation and validation
>   drm/ttm/tests: Add tests with mock resource managers
>   drm/ttm/tests: Add test cases dependent on fence signaling

I see that all these files (and previous patches, as well) don't
have a consistent prefix system, like kunit_ttm_*. But they all
start with ttm_*, thread_*, drm_*, etc, which makes it a bit
difficult to follow and grep through.

Can we please maintain a consistent prefix system?

I know it looks bad to come out with it after this series (and
previous work too) has been out for several months already. I
need to say my "mea culpa" as well as I've been in the review
loop, as well.

Thanks,
Andi

Re: [PATCH v8 0/8] Improve test coverage of TTM

[Karolina Stolarek]

Hi Andi,

On 5.12.2023 16:39, Andi Shyti wrote:
> Hi Karolina and Christian,
> 
>> Karolina Stolarek (8): drm/ttm/tests: Add tests for ttm_resource 
>> and ttm_sys_man drm/ttm/tests: Add tests for ttm_tt drm/ttm/tests:
>>  Add tests for ttm_bo functions drm/ttm/tests: Fix argument in 
>> ttm_tt_kunit_init() drm/ttm/tests: Use an init function from the 
>> helpers lib drm/ttm/tests: Test simple BO creation and validation 
>> drm/ttm/tests: Add tests with mock resource managers drm/ttm/tests:
>> Add test cases dependent on fence signaling
> 
> I see that all these files (and previous patches, as well) don't have
> a consistent prefix system, like kunit_ttm_*. But they all start with
> ttm_*, thread_*, drm_*, etc, which makes it a bit difficult to follow
> and grep through.

I see what you mean. As for ttm_* part, I decided to keep it as it is
based on what was in DRM KUnit tests and helpers lib. Almost all
functions and subtests start with drm_*, and as I'm working on TTM
tests, ttm_* prefix seemed like a natural choice. With thread_*, I could
add ttm_kunit_* in front of it, but not sure what would be the benefit
of doing this.

To be honest, I haven't considered using kunit_* prefix here. In the
code context, it's obvious these are kunit tests, and repeating that
information would make already long function/struct names even longer.

I had a quick glance at other KUnit tests in the kernel and this seems
to be the case, no kunit_* prefixes are used.

> Can we please maintain a consistent prefix system?
> 
> I know it looks bad to come out with it after this series (and 
> previous work too) has been out for several months already. I need to
> say my "mea culpa" as well as I've been in the review loop, as well.

After this series, I plan to send a small one to wrap up my work in this
field. How about adding a TODO to clean these up? I mean, that's
just how I see it, I'd like to hear Christian's thoughts on this.

All the best,
Karolina

Re: [PATCH v8 0/8] Improve test coverage of TTM

[Andi Shyti]

Hi Karolina,

On Wed, Dec 06, 2023 at 12:28:14PM +0100, Karolina Stolarek wrote:
> On 5.12.2023 16:39, Andi Shyti wrote:
...
> > Hi Karolina and Christian,
...
> > > Karolina Stolarek (8): drm/ttm/tests: Add tests for ttm_resource and
> > > ttm_sys_man drm/ttm/tests: Add tests for ttm_tt drm/ttm/tests:
> > >  Add tests for ttm_bo functions drm/ttm/tests: Fix argument in
> > > ttm_tt_kunit_init() drm/ttm/tests: Use an init function from the
> > > helpers lib drm/ttm/tests: Test simple BO creation and validation
> > > drm/ttm/tests: Add tests with mock resource managers drm/ttm/tests:
> > > Add test cases dependent on fence signaling
> > 
> > I see that all these files (and previous patches, as well) don't have
> > a consistent prefix system, like kunit_ttm_*. But they all start with
> > ttm_*, thread_*, drm_*, etc, which makes it a bit difficult to follow
> > and grep through.
> 
> I see what you mean. As for ttm_* part, I decided to keep it as it is
> based on what was in DRM KUnit tests and helpers lib. Almost all
> functions and subtests start with drm_*, and as I'm working on TTM
> tests, ttm_* prefix seemed like a natural choice. With thread_*, I could
> add ttm_kunit_* in front of it, but not sure what would be the benefit
> of doing this.

I do not agree here, ttm_* are related to ttm and not "testing
ttm" and we need to be consistent.

I'd be more inclined to a kunit_ttm_* or ttm_kunit_*.

Anyway, let's hear also what Christian thinks.

> To be honest, I haven't considered using kunit_* prefix here. In the
> code context, it's obvious these are kunit tests, and repeating that
> information would make already long function/struct names even longer.
> 
> I had a quick glance at other KUnit tests in the kernel and this seems
> to be the case, no kunit_* prefixes are used.
> 
> > Can we please maintain a consistent prefix system?
> > 
> > I know it looks bad to come out with it after this series (and previous
> > work too) has been out for several months already. I need to
> > say my "mea culpa" as well as I've been in the review loop, as well.
> 
> After this series, I plan to send a small one to wrap up my work in this
> field. How about adding a TODO to clean these up? I mean, that's
> just how I see it, I'd like to hear Christian's thoughts on this.

nah... a TODO note would be too much... your promise is enough
for me.

Andi

Re: [PATCH v8 8/8] drm/ttm/tests: Add test cases dependent on fence signaling

[Karolina Stolarek]

Hi Andi,

> For next time, I find it difficult to follow all these variables,
> it's easier to read
> 
> 	man = ttm_manager_type(priv->ttm_dev, TTM_PL_SYSTEM);
> 
> than
> 
> 	mem_type = TTM_PL_SYSTEM;
> 	...
> 	...
> 	...
> 	man = ttm_manager_type(priv->ttm_dev, mem_type);
> 
> 
>> +	bo = ttm_bo_kunit_init(test, test->priv, size);
>> +	bo->type = bo_type;
> 
> same here... the bo_type variable is not giving any value.
> 
> 	bo->type = ttm_bo_type_device;
> 
> is way more readable. You keep doing this all the way and I need
> to check everytime what's the value in the declaration :-)

The idea was that I'd provide these as parameters, but I limited the
scope of my tests and stick to set values. Also, in some cases, I
defined them because I keep checking for them in assertions, and didn't
want to repeat myself.

> 
> I'm not going to comment on this anymore.
> 
>> +	if (params->with_ttm) {
>> +		old_tt = priv->ttm_dev->funcs->ttm_tt_create(bo, 0);
>> +		ttm_pool_alloc(&priv->ttm_dev->pool, old_tt, &ctx);
>> +		bo->ttm = old_tt;
>> +	}
>> +
>> +	err = ttm_resource_alloc(bo, place, &bo->resource);
>> +	KUNIT_EXPECT_EQ(test, err, 0);
>> +	KUNIT_ASSERT_EQ(test, man->usage, size);
>> +
>> +	placement = kunit_kzalloc(test, sizeof(*placement), GFP_KERNEL);
>> +	KUNIT_ASSERT_NOT_NULL(test, placement);
>> +
>> +	ttm_bo_reserve(bo, false, false, NULL);
>> +	err = ttm_bo_validate(bo, placement, &ctx);
>> +	dma_resv_unlock(bo->base.resv);
> 
> why don't you use here ttm_bo_unreserve()?

That's a good question! I think that unreserve would work here as well.

(...)

>> +static void ttm_bo_validate_move_fence_signaled(struct kunit *test)
>> +{
>> +	struct ttm_test_devices *priv = test->priv;
>> +	struct ttm_buffer_object *bo;
>> +	struct ttm_place *place;
>> +	struct ttm_placement *placement;
>> +	struct ttm_resource_manager *man;
>> +	enum ttm_bo_type bo_type = ttm_bo_type_device;
>> +	uint32_t mem_type = TTM_PL_SYSTEM;
>> +	struct ttm_operation_ctx ctx = { };
>> +	uint32_t size = ALIGN(BO_SIZE, PAGE_SIZE);
>> +	int err;
>> +
>> +	man = ttm_manager_type(priv->ttm_dev, mem_type);
>> +	spin_lock_init(&fence_lock);
> 
> where are we using the fence_lock here?

Argh, it's a copy-paste mistake, sorry. We don't need it, as we use a
stub fence in man->move. I will delete that.

> 
>> +	man->move = dma_fence_get_stub();
>> +
>> +	bo = ttm_bo_kunit_init(test, test->priv, size);
>> +	bo->type = bo_type;
>> +	place = ttm_place_kunit_init(test, mem_type, 0);
>> +	placement = ttm_placement_kunit_init(test, place, 1, NULL, 0);
>> +
>> +	ttm_bo_reserve(bo, false, false, NULL);
>> +	err = ttm_bo_validate(bo, placement, &ctx);
>> +	dma_resv_unlock(bo->base.resv);
>> +
>> +	KUNIT_EXPECT_EQ(test, err, 0);
>> +	KUNIT_EXPECT_EQ(test, bo->resource->mem_type, mem_type);
>> +	KUNIT_EXPECT_EQ(test, ctx.bytes_moved, size);
> 
> Do we want to check also bo->ttm here?

Hmm, I think that I covered that case with "normal" testing of
ttm_bo_validate, and kept this one minimal. Here, I'm just making sure
that a signaled move fence doesn't get into a way of BO validation.

> 
>> +	ttm_bo_put(bo);
>> +	dma_fence_put(man->move);
>> +}
>> +
>> +static const struct ttm_bo_validate_test_case ttm_bo_validate_wait_cases[] = {
>> +	{
>> +		.description = "Waits for GPU",
>> +		.no_gpu_wait = false,
>> +	},
>> +	{
>> +		.description = "Tries to lock straight away",
>> +		.no_gpu_wait = true,
>> +	},
>> +};
>> +
>> +KUNIT_ARRAY_PARAM(ttm_bo_validate_wait, ttm_bo_validate_wait_cases,
>> +		  ttm_bo_validate_case_desc);
>> +
>> +static int threaded_fence_signal(void *arg)
>> +{
>> +	struct dma_fence *fence = arg;
>> +	int err;
>> +
>> +	msleep(20);
>> +	err = dma_fence_signal(fence);
>> +
>> +	return err;
> 
> if you do here "return dma_fence_signal(fence);" you don't need
> for err.

That is true!

> Not a binding review, though, your choice.

You spotted a couple of things I can improve, so I plan to include them 
in the next version, after getting more review comments.

All the best,
Karolina

Re: [PATCH v8 0/8] Improve test coverage of TTM

[Karolina Stolarek]

Hi Christian,

On 29.11.2023 13:02, Karolina Stolarek wrote:
> Karolina Stolarek (8):
>    drm/ttm/tests: Add tests for ttm_resource and ttm_sys_man
>    drm/ttm/tests: Add tests for ttm_tt
>    drm/ttm/tests: Add tests for ttm_bo functions
 >    drm/ttm/tests: Fix argument in ttm_tt_kunit_init()

Would it be possible to get these patches merged? They already have
r-b's. There are a couple more, but these don't depend on any other
patch that is pending review, and should be easy to get in.

I'm asking for this because I have three final patches ready for
submission (eviction, ttm_tt_populate, TODO file) and wondered if I
should submit them as v9, or if I could create a new series. I'd prefer
creating a new one, this patchset is already big and intimidating :)
Having said that, thank you for taking your time to review the tests, I
really appreciate it.

All the best,
Karolina

>    drm/ttm/tests: Use an init function from the helpers lib
>    drm/ttm/tests: Test simple BO creation and validation
>    drm/ttm/tests: Add tests with mock resource managers
>    drm/ttm/tests: Add test cases dependent on fence signaling
> 
>   drivers/gpu/drm/Kconfig                       |   1 +
>   drivers/gpu/drm/ttm/tests/.kunitconfig        |   1 +
>   drivers/gpu/drm/ttm/tests/Makefile            |   5 +
>   drivers/gpu/drm/ttm/tests/ttm_bo_test.c       | 622 ++++++++++++++
>   .../gpu/drm/ttm/tests/ttm_bo_validate_test.c  | 795 ++++++++++++++++++
>   drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c | 109 ++-
>   drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h |   7 +
>   drivers/gpu/drm/ttm/tests/ttm_mock_manager.c  | 207 +++++
>   drivers/gpu/drm/ttm/tests/ttm_mock_manager.h  |  31 +
>   drivers/gpu/drm/ttm/tests/ttm_pool_test.c     |   3 +-
>   drivers/gpu/drm/ttm/tests/ttm_resource_test.c | 335 ++++++++
>   drivers/gpu/drm/ttm/tests/ttm_tt_test.c       | 282 +++++++
>   drivers/gpu/drm/ttm/ttm_resource.c            |   3 +
>   drivers/gpu/drm/ttm/ttm_tt.c                  |   3 +
>   14 files changed, 2401 insertions(+), 3 deletions(-)
>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_test.c
>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.c
>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.h
>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_resource_test.c
>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_tt_test.c
> 

Re: [PATCH v8 0/8] Improve test coverage of TTM

[Christian König]

Am 14.12.23 um 09:20 schrieb Karolina Stolarek:
>
> Hi Christian,
>
> On 29.11.2023 13:02, Karolina Stolarek wrote:
>> Karolina Stolarek (8):
>>    drm/ttm/tests: Add tests for ttm_resource and ttm_sys_man
>>    drm/ttm/tests: Add tests for ttm_tt
>>    drm/ttm/tests: Add tests for ttm_bo functions
> >    drm/ttm/tests: Fix argument in ttm_tt_kunit_init()
>
> Would it be possible to get these patches merged? They already have
> r-b's. There are a couple more, but these don't depend on any other
> patch that is pending review, and should be easy to get in.

Sure, just pushed those three to drm-misc-next.

>
> I'm asking for this because I have three final patches ready for
> submission (eviction, ttm_tt_populate, TODO file) and wondered if I
> should submit them as v9, or if I could create a new series. I'd prefer
> creating a new one, this patchset is already big and intimidating :)
> Having said that, thank you for taking your time to review the tests, I
> really appreciate it.

No problem, you are actually helping getting stuff from my way to long 
TODO list.

Regards,
Christian.

>
> All the best,
> Karolina
>
>>    drm/ttm/tests: Use an init function from the helpers lib
>>    drm/ttm/tests: Test simple BO creation and validation
>>    drm/ttm/tests: Add tests with mock resource managers
>>    drm/ttm/tests: Add test cases dependent on fence signaling
>>
>>   drivers/gpu/drm/Kconfig                       |   1 +
>>   drivers/gpu/drm/ttm/tests/.kunitconfig        |   1 +
>>   drivers/gpu/drm/ttm/tests/Makefile            |   5 +
>>   drivers/gpu/drm/ttm/tests/ttm_bo_test.c       | 622 ++++++++++++++
>>   .../gpu/drm/ttm/tests/ttm_bo_validate_test.c  | 795 ++++++++++++++++++
>>   drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c | 109 ++-
>>   drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.h |   7 +
>>   drivers/gpu/drm/ttm/tests/ttm_mock_manager.c  | 207 +++++
>>   drivers/gpu/drm/ttm/tests/ttm_mock_manager.h  |  31 +
>>   drivers/gpu/drm/ttm/tests/ttm_pool_test.c     |   3 +-
>>   drivers/gpu/drm/ttm/tests/ttm_resource_test.c | 335 ++++++++
>>   drivers/gpu/drm/ttm/tests/ttm_tt_test.c       | 282 +++++++
>>   drivers/gpu/drm/ttm/ttm_resource.c            |   3 +
>>   drivers/gpu/drm/ttm/ttm_tt.c                  |   3 +
>>   14 files changed, 2401 insertions(+), 3 deletions(-)
>>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_test.c
>>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c
>>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.c
>>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_mock_manager.h
>>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_resource_test.c
>>   create mode 100644 drivers/gpu/drm/ttm/tests/ttm_tt_test.c
>>

Re: [PATCH v8 0/8] Improve test coverage of TTM

[Karolina Stolarek]

On 14.12.2023 11:22, Christian König wrote:
> 
> 
> Am 14.12.23 um 09:20 schrieb Karolina Stolarek:
>>
>> Hi Christian,
>>
>> On 29.11.2023 13:02, Karolina Stolarek wrote:
>>> Karolina Stolarek (8):
>>>    drm/ttm/tests: Add tests for ttm_resource and ttm_sys_man
>>>    drm/ttm/tests: Add tests for ttm_tt
>>>    drm/ttm/tests: Add tests for ttm_bo functions
>> >    drm/ttm/tests: Fix argument in ttm_tt_kunit_init()
>>
>> Would it be possible to get these patches merged? They already have
>> r-b's. There are a couple more, but these don't depend on any other
>> patch that is pending review, and should be easy to get in.
> 
> Sure, just pushed those three to drm-misc-next.

Awesome, thanks a lot. I see they went into drm-tip already, so I 
rebased my series on the top of that.

>> I'm asking for this because I have three final patches ready for
>> submission (eviction, ttm_tt_populate, TODO file) and wondered if I
>> should submit them as v9, or if I could create a new series. I'd prefer
>> creating a new one, this patchset is already big and intimidating :)
>> Having said that, thank you for taking your time to review the tests, I
>> really appreciate it.
> 
> No problem, you are actually helping getting stuff from my way to long 
> TODO list.

By producing 4k+ lines of code for you to look at? ;-)

All the best,
Karolina

> 
> Regards,
> Christian.