[PATCH v3 0/4] drm/tests: Fix some memory leaks

[PATCH v3 0/4] drm/tests: Fix some memory leaks

[Jinjie Ruan]

Fix some memory leaks in drm tests.

Changes in v3:
- Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
- Drop the "helper" in the helper name.
- s/fllowing/following/
- Add Acked-by.

Changes in v2:
- Fix it with new introduced helper instead of drm_mode_destroy().
- Update the commit message.
- Add Reviewed-by.

Jinjie Ruan (4):
  drm/tests: helpers: Add helper for drm_display_mode_from_cea_vic()
  drm/connector: hdmi: Fix memory leak in
    drm_display_mode_from_cea_vic()
  drm/ttm/tests: Fix memory leak in ttm_tt_simple_create()
  drm/tests: hdmi: Fix memory leaks in drm_display_mode_from_cea_vic()

 drivers/gpu/drm/tests/drm_connector_test.c    | 24 +++++------
 .../drm/tests/drm_hdmi_state_helper_test.c    |  8 ++--
 drivers/gpu/drm/tests/drm_kunit_helpers.c     | 40 +++++++++++++++++++
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c |  1 +
 include/drm/drm_kunit_helpers.h               |  4 ++
 5 files changed, 61 insertions(+), 16 deletions(-)

-- 
2.34.1

[PATCH v3 1/4] drm/tests: helpers: Add helper for drm_display_mode_from_cea_vic()

[Jinjie Ruan]

As Maxime suggested, add a new helper
drm_kunit_display_mode_from_cea_vic(), it can replace the direct call
of drm_display_mode_from_cea_vic(), and it will help solving
the `mode` memory leaks.

Acked-by: Maxime Ripard <mripard@kernel.org>
Suggested-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
---
v3:
- Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
- Drop the "helper" in the helper name.
- Add Acked-by.
---
 drivers/gpu/drm/tests/drm_kunit_helpers.c | 40 +++++++++++++++++++++++
 include/drm/drm_kunit_helpers.h           |  4 +++
 2 files changed, 44 insertions(+)

diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c
index aa62719dab0e..565172990044 100644
--- a/drivers/gpu/drm/tests/drm_kunit_helpers.c
+++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c
@@ -3,6 +3,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_drv.h>
+#include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_kunit_helpers.h>
 #include <drm/drm_managed.h>
@@ -311,6 +312,45 @@ drm_kunit_helper_create_crtc(struct kunit *test,
 }
 EXPORT_SYMBOL_GPL(drm_kunit_helper_create_crtc);
 
+static void kunit_action_drm_mode_destroy(void *ptr)
+{
+	struct drm_display_mode *mode = ptr;
+
+	drm_mode_destroy(NULL, mode);
+}
+
+/**
+ * drm_kunit_display_mode_from_cea_vic() - return a mode for CEA VIC
+					   for a KUnit test
+ * @test: The test context object
+ * @dev: DRM device
+ * @video_code: CEA VIC of the mode
+ *
+ * Creates a new mode matching the specified CEA VIC for a KUnit test.
+ *
+ * Resources will be cleaned up automatically.
+ *
+ * Returns: A new drm_display_mode on success or NULL on failure
+ */
+struct drm_display_mode *
+drm_kunit_display_mode_from_cea_vic(struct kunit *test, struct drm_device *dev,
+				    u8 video_code)
+{
+	struct drm_display_mode *mode;
+	int ret;
+
+	mode = drm_display_mode_from_cea_vic(dev, video_code);
+
+	ret = kunit_add_action_or_reset(test,
+					kunit_action_drm_mode_destroy,
+					mode);
+	if (ret)
+		return NULL;
+
+	return mode;
+}
+EXPORT_SYMBOL_GPL(drm_kunit_display_mode_from_cea_vic);
+
 MODULE_AUTHOR("Maxime Ripard <maxime@cerno.tech>");
 MODULE_DESCRIPTION("KUnit test suite helper functions");
 MODULE_LICENSE("GPL");
diff --git a/include/drm/drm_kunit_helpers.h b/include/drm/drm_kunit_helpers.h
index e7cc17ee4934..afdd46ef04f7 100644
--- a/include/drm/drm_kunit_helpers.h
+++ b/include/drm/drm_kunit_helpers.h
@@ -120,4 +120,8 @@ drm_kunit_helper_create_crtc(struct kunit *test,
 			     const struct drm_crtc_funcs *funcs,
 			     const struct drm_crtc_helper_funcs *helper_funcs);
 
+struct drm_display_mode *
+drm_kunit_display_mode_from_cea_vic(struct kunit *test, struct drm_device *dev,
+				    u8 video_code);
+
 #endif // DRM_KUNIT_HELPERS_H_
-- 
2.34.1

Re: [PATCH v3 1/4] drm/tests: helpers: Add helper for drm_display_mode_from_cea_vic()

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 2583 bytes --]

On Thu, Oct 17, 2024 at 02:31:22PM +0800, Jinjie Ruan wrote:
> As Maxime suggested, add a new helper
> drm_kunit_display_mode_from_cea_vic(), it can replace the direct call
> of drm_display_mode_from_cea_vic(), and it will help solving
> the `mode` memory leaks.
> 
> Acked-by: Maxime Ripard <mripard@kernel.org>
> Suggested-by: Maxime Ripard <mripard@kernel.org>
> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
> ---
> v3:
> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
> - Drop the "helper" in the helper name.
> - Add Acked-by.
> ---
>  drivers/gpu/drm/tests/drm_kunit_helpers.c | 40 +++++++++++++++++++++++
>  include/drm/drm_kunit_helpers.h           |  4 +++
>  2 files changed, 44 insertions(+)
> 
> diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c
> index aa62719dab0e..565172990044 100644
> --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c
> +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c
> @@ -3,6 +3,7 @@
>  #include <drm/drm_atomic.h>
>  #include <drm/drm_atomic_helper.h>
>  #include <drm/drm_drv.h>
> +#include <drm/drm_edid.h>
>  #include <drm/drm_fourcc.h>
>  #include <drm/drm_kunit_helpers.h>
>  #include <drm/drm_managed.h>
> @@ -311,6 +312,45 @@ drm_kunit_helper_create_crtc(struct kunit *test,
>  }
>  EXPORT_SYMBOL_GPL(drm_kunit_helper_create_crtc);
>  
> +static void kunit_action_drm_mode_destroy(void *ptr)
> +{
> +	struct drm_display_mode *mode = ptr;
> +
> +	drm_mode_destroy(NULL, mode);
> +}
> +
> +/**
> + * drm_kunit_display_mode_from_cea_vic() - return a mode for CEA VIC
> +					   for a KUnit test
> + * @test: The test context object
> + * @dev: DRM device
> + * @video_code: CEA VIC of the mode
> + *
> + * Creates a new mode matching the specified CEA VIC for a KUnit test.
> + *
> + * Resources will be cleaned up automatically.
> + *
> + * Returns: A new drm_display_mode on success or NULL on failure
> + */
> +struct drm_display_mode *
> +drm_kunit_display_mode_from_cea_vic(struct kunit *test, struct drm_device *dev,
> +				    u8 video_code)
> +{
> +	struct drm_display_mode *mode;
> +	int ret;
> +
> +	mode = drm_display_mode_from_cea_vic(dev, video_code);

I'd rather return directly if mode is NULL here...

> +	ret = kunit_add_action_or_reset(test,
> +					kunit_action_drm_mode_destroy,
> +					mode);
> +	if (ret)
> +		return NULL;

Because it doesn't really make much sense to register a cleanup action
if we know that it's going to be useless, and possibly be executed right
away.

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 273 bytes --]

Re: [PATCH v3 1/4] drm/tests: helpers: Add helper for drm_display_mode_from_cea_vic()

[Jinjie Ruan]

On 2024/10/29 16:16, Maxime Ripard wrote:
> On Thu, Oct 17, 2024 at 02:31:22PM +0800, Jinjie Ruan wrote:
>> As Maxime suggested, add a new helper
>> drm_kunit_display_mode_from_cea_vic(), it can replace the direct call
>> of drm_display_mode_from_cea_vic(), and it will help solving
>> the `mode` memory leaks.
>>
>> Acked-by: Maxime Ripard <mripard@kernel.org>
>> Suggested-by: Maxime Ripard <mripard@kernel.org>
>> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
>> ---
>> v3:
>> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
>> - Drop the "helper" in the helper name.
>> - Add Acked-by.
>> ---
>>  drivers/gpu/drm/tests/drm_kunit_helpers.c | 40 +++++++++++++++++++++++
>>  include/drm/drm_kunit_helpers.h           |  4 +++
>>  2 files changed, 44 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c
>> index aa62719dab0e..565172990044 100644
>> --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c
>> +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c
>> @@ -3,6 +3,7 @@
>>  #include <drm/drm_atomic.h>
>>  #include <drm/drm_atomic_helper.h>
>>  #include <drm/drm_drv.h>
>> +#include <drm/drm_edid.h>
>>  #include <drm/drm_fourcc.h>
>>  #include <drm/drm_kunit_helpers.h>
>>  #include <drm/drm_managed.h>
>> @@ -311,6 +312,45 @@ drm_kunit_helper_create_crtc(struct kunit *test,
>>  }
>>  EXPORT_SYMBOL_GPL(drm_kunit_helper_create_crtc);
>>  
>> +static void kunit_action_drm_mode_destroy(void *ptr)
>> +{
>> +	struct drm_display_mode *mode = ptr;
>> +
>> +	drm_mode_destroy(NULL, mode);
>> +}
>> +
>> +/**
>> + * drm_kunit_display_mode_from_cea_vic() - return a mode for CEA VIC
>> +					   for a KUnit test
>> + * @test: The test context object
>> + * @dev: DRM device
>> + * @video_code: CEA VIC of the mode
>> + *
>> + * Creates a new mode matching the specified CEA VIC for a KUnit test.
>> + *
>> + * Resources will be cleaned up automatically.
>> + *
>> + * Returns: A new drm_display_mode on success or NULL on failure
>> + */
>> +struct drm_display_mode *
>> +drm_kunit_display_mode_from_cea_vic(struct kunit *test, struct drm_device *dev,
>> +				    u8 video_code)
>> +{
>> +	struct drm_display_mode *mode;
>> +	int ret;
>> +
>> +	mode = drm_display_mode_from_cea_vic(dev, video_code);
> 
> I'd rather return directly if mode is NULL here...
> 
>> +	ret = kunit_add_action_or_reset(test,
>> +					kunit_action_drm_mode_destroy,
>> +					mode);
>> +	if (ret)
>> +		return NULL;
> 
> Because it doesn't really make much sense to register a cleanup action
> if we know that it's going to be useless, and possibly be executed right
> away.

Sure, will udpate it.

> 
> Maxime

[PATCH v3 2/4] drm/connector: hdmi: Fix memory leak in drm_display_mode_from_cea_vic()

[Jinjie Ruan]

modprobe drm_connector_test and then rmmod drm_connector_test,
the following memory leak occurs.

The `mode` allocated in drm_mode_duplicate() called by
drm_display_mode_from_cea_vic() is not freed, which cause the memory leak:

	unreferenced object 0xffffff80cb0ee400 (size 128):
	  comm "kunit_try_catch", pid 1948, jiffies 4294950339
	  hex dump (first 32 bytes):
	    14 44 02 00 80 07 d8 07 04 08 98 08 00 00 38 04  .D............8.
	    3c 04 41 04 65 04 00 00 05 00 00 00 00 00 00 00  <.A.e...........
	  backtrace (crc 90e9585c):
	    [<00000000ec42e3d7>] kmemleak_alloc+0x34/0x40
	    [<00000000d0ef055a>] __kmalloc_cache_noprof+0x26c/0x2f4
	    [<00000000c2062161>] drm_mode_duplicate+0x44/0x19c
	    [<00000000f96c74aa>] drm_display_mode_from_cea_vic+0x88/0x98
	    [<00000000d8f2c8b4>] 0xffffffdc982a4868
	    [<000000005d164dbc>] kunit_try_run_case+0x13c/0x3ac
	    [<000000006fb23398>] kunit_generic_run_threadfn_adapter+0x80/0xec
	    [<000000006ea56ca0>] kthread+0x2e8/0x374
	    [<000000000676063f>] ret_from_fork+0x10/0x20
	......

Free `mode` by using drm_kunit_display_mode_from_cea_vic()
to fix it.

Cc: stable@vger.kernel.org
Fixes: abb6f74973e2 ("drm/tests: Add HDMI TDMS character rate tests")
Acked-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
---
v3:
- Update the commit message.
- Add Acked-by.
v2:
- Fix it with new introduced helper instead of drm_mode_destroy().
- Update the commit message.
---
 drivers/gpu/drm/tests/drm_connector_test.c | 24 +++++++++++-----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/tests/drm_connector_test.c b/drivers/gpu/drm/tests/drm_connector_test.c
index 15e36a8db685..6bba97d0be88 100644
--- a/drivers/gpu/drm/tests/drm_connector_test.c
+++ b/drivers/gpu/drm/tests/drm_connector_test.c
@@ -996,7 +996,7 @@ static void drm_test_drm_hdmi_compute_mode_clock_rgb(struct kunit *test)
 	unsigned long long rate;
 	struct drm_device *drm = &priv->drm;
 
-	mode = drm_display_mode_from_cea_vic(drm, 16);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 16);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1017,7 +1017,7 @@ static void drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc(struct kunit *test)
 	unsigned long long rate;
 	struct drm_device *drm = &priv->drm;
 
-	mode = drm_display_mode_from_cea_vic(drm, 16);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 16);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1038,7 +1038,7 @@ static void drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1(struct kunit *t
 	unsigned long long rate;
 	struct drm_device *drm = &priv->drm;
 
-	mode = drm_display_mode_from_cea_vic(drm, 1);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	rate = drm_hdmi_compute_mode_clock(mode, 10, HDMI_COLORSPACE_RGB);
@@ -1056,7 +1056,7 @@ static void drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc(struct kunit *test)
 	unsigned long long rate;
 	struct drm_device *drm = &priv->drm;
 
-	mode = drm_display_mode_from_cea_vic(drm, 16);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 16);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1077,7 +1077,7 @@ static void drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1(struct kunit *t
 	unsigned long long rate;
 	struct drm_device *drm = &priv->drm;
 
-	mode = drm_display_mode_from_cea_vic(drm, 1);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	rate = drm_hdmi_compute_mode_clock(mode, 12, HDMI_COLORSPACE_RGB);
@@ -1095,7 +1095,7 @@ static void drm_test_drm_hdmi_compute_mode_clock_rgb_double(struct kunit *test)
 	unsigned long long rate;
 	struct drm_device *drm = &priv->drm;
 
-	mode = drm_display_mode_from_cea_vic(drm, 6);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 6);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_TRUE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1118,7 +1118,7 @@ static void drm_test_connector_hdmi_compute_mode_clock_yuv420_valid(struct kunit
 	unsigned long long rate;
 	unsigned int vic = *(unsigned int *)test->param_value;
 
-	mode = drm_display_mode_from_cea_vic(drm, vic);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, vic);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1155,7 +1155,7 @@ static void drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc(struct kuni
 		drm_hdmi_compute_mode_clock_yuv420_vic_valid_tests[0];
 	unsigned long long rate;
 
-	mode = drm_display_mode_from_cea_vic(drm, vic);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, vic);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1180,7 +1180,7 @@ static void drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc(struct kuni
 		drm_hdmi_compute_mode_clock_yuv420_vic_valid_tests[0];
 	unsigned long long rate;
 
-	mode = drm_display_mode_from_cea_vic(drm, vic);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, vic);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1203,7 +1203,7 @@ static void drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc(struct kunit
 	struct drm_device *drm = &priv->drm;
 	unsigned long long rate;
 
-	mode = drm_display_mode_from_cea_vic(drm, 16);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 16);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1225,7 +1225,7 @@ static void drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc(struct kuni
 	struct drm_device *drm = &priv->drm;
 	unsigned long long rate;
 
-	mode = drm_display_mode_from_cea_vic(drm, 16);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 16);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
@@ -1247,7 +1247,7 @@ static void drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc(struct kuni
 	struct drm_device *drm = &priv->drm;
 	unsigned long long rate;
 
-	mode = drm_display_mode_from_cea_vic(drm, 16);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 16);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	KUNIT_ASSERT_FALSE(test, mode->flags & DRM_MODE_FLAG_DBLCLK);
-- 
2.34.1

[PATCH v3 3/4] drm/ttm/tests: Fix memory leak in ttm_tt_simple_create()

[Jinjie Ruan]

modprobe ttm_device_test and then rmmod ttm_device_test, the following
memory leaks occurs:

The ttm->pages allocated in ttm_tt_init() is not freed after calling
ttm_tt_simple_create(), which cause the memory leak:

	unreferenced object 0xffffff80caf27750 (size 8):
	  comm "kunit_try_catch", pid 2242, jiffies 4295055735
	  hex dump (first 8 bytes):
	    c0 1e 3d c3 fe ff ff ff                          ..=.....
	  backtrace (crc 3d11615a):
	    [<000000007f57312a>] kmemleak_alloc+0x34/0x40
	    [<000000008c6c4c7e>] __kmalloc_node_noprof+0x304/0x3e4
	    [<00000000679c1182>] __kvmalloc_node_noprof+0x1c/0x144
	    [<000000006aed0a3d>] ttm_tt_init+0x138/0x28c [ttm]
	    [<000000005c331998>] drm_gem_shmem_free+0x60/0x534 [drm_shmem_helper]
	    [<0000000022b4f375>] kunit_try_run_case+0x13c/0x3ac
	    [<00000000c525d725>] kunit_generic_run_threadfn_adapter+0x80/0xec
	    [<000000002db94a1f>] kthread+0x2e8/0x374
	    [<000000002c457ad7>] ret_from_fork+0x10/0x20
	......

Fix it by calling ttm_tt_fini() in the exit function.

Cc: stable@vger.kernel.org
Fixes: e6f7c641fae3 ("drm/ttm/tests: Add tests for ttm_tt")
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
---
v3:
- s/fllowing/following/
v2:
- Add Reviewed-by.
---
 drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
index b91c13f46225..9ff216ec58ef 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c
@@ -54,6 +54,7 @@ static struct ttm_tt *ttm_tt_simple_create(struct ttm_buffer_object *bo, u32 pag
 
 static void ttm_tt_simple_destroy(struct ttm_device *bdev, struct ttm_tt *ttm)
 {
+	ttm_tt_fini(ttm);
 	kfree(ttm);
 }
 
-- 
2.34.1

[PATCH v3 4/4] drm/tests: hdmi: Fix memory leaks in drm_display_mode_from_cea_vic()

[Jinjie Ruan]

modprobe drm_hdmi_state_helper_test and then rmmod it, the following
memory leak occurs.

The `mode` allocated in drm_mode_duplicate() called by
drm_display_mode_from_cea_vic() is not freed, which cause the memory leak:

	unreferenced object 0xffffff80ccd18100 (size 128):
	  comm "kunit_try_catch", pid 1851, jiffies 4295059695
	  hex dump (first 32 bytes):
	    57 62 00 00 80 02 90 02 f0 02 20 03 00 00 e0 01  Wb........ .....
	    ea 01 ec 01 0d 02 00 00 0a 00 00 00 00 00 00 00  ................
	  backtrace (crc c2f1aa95):
	    [<000000000f10b11b>] kmemleak_alloc+0x34/0x40
	    [<000000001cd4cf73>] __kmalloc_cache_noprof+0x26c/0x2f4
	    [<00000000f1f3cffa>] drm_mode_duplicate+0x44/0x19c
	    [<000000008cbeef13>] drm_display_mode_from_cea_vic+0x88/0x98
	    [<0000000019daaacf>] 0xffffffedc11ae69c
	    [<000000000aad0f85>] kunit_try_run_case+0x13c/0x3ac
	    [<00000000a9210bac>] kunit_generic_run_threadfn_adapter+0x80/0xec
	    [<000000000a0b2e9e>] kthread+0x2e8/0x374
	    [<00000000bd668858>] ret_from_fork+0x10/0x20
	......

Free `mode` by using drm_kunit_display_mode_from_cea_vic()
to fix it.

Cc: stable@vger.kernel.org
Fixes: 4af70f19e559 ("drm/tests: Add RGB Quantization tests")
Acked-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
---
v3:
- Update the commit message.
- Add Acked-by.
v2:
- Fix it with new introduced helper instead of drm_mode_destroy().
- Update the commit message.
---
 drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c
index 34ee95d41f29..294773342e71 100644
--- a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c
+++ b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c
@@ -441,7 +441,7 @@ static void drm_test_check_broadcast_rgb_auto_cea_mode_vic_1(struct kunit *test)
 	ctx = drm_kunit_helper_acquire_ctx_alloc(test);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx);
 
-	mode = drm_display_mode_from_cea_vic(drm, 1);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	drm = &priv->drm;
@@ -555,7 +555,7 @@ static void drm_test_check_broadcast_rgb_full_cea_mode_vic_1(struct kunit *test)
 	ctx = drm_kunit_helper_acquire_ctx_alloc(test);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx);
 
-	mode = drm_display_mode_from_cea_vic(drm, 1);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	drm = &priv->drm;
@@ -671,7 +671,7 @@ static void drm_test_check_broadcast_rgb_limited_cea_mode_vic_1(struct kunit *te
 	ctx = drm_kunit_helper_acquire_ctx_alloc(test);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx);
 
-	mode = drm_display_mode_from_cea_vic(drm, 1);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	drm = &priv->drm;
@@ -1263,7 +1263,7 @@ static void drm_test_check_output_bpc_format_vic_1(struct kunit *test)
 	ctx = drm_kunit_helper_acquire_ctx_alloc(test);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx);
 
-	mode = drm_display_mode_from_cea_vic(drm, 1);
+	mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	/*
-- 
2.34.1

Re: [PATCH v3 0/4] drm/tests: Fix some memory leaks

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 378 bytes --]

Hi,

On Thu, Oct 17, 2024 at 02:31:21PM GMT, Jinjie Ruan wrote:
> Fix some memory leaks in drm tests.
> 
> Changes in v3:
> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
> - Drop the "helper" in the helper name.
> - s/fllowing/following/
> - Add Acked-by.

This creates build failures since drm_display_mode were const before,
and can't anymore.

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 273 bytes --]

Re: [PATCH v3 0/4] drm/tests: Fix some memory leaks

[Jinjie Ruan]

On 2024/10/18 15:55, Maxime Ripard wrote:
> Hi,
> 
> On Thu, Oct 17, 2024 at 02:31:21PM GMT, Jinjie Ruan wrote:
>> Fix some memory leaks in drm tests.
>>
>> Changes in v3:
>> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
>> - Drop the "helper" in the helper name.
>> - s/fllowing/following/
>> - Add Acked-by.
> 
> This creates build failures since drm_display_mode were const before,
> and can't anymore.

Let me fix it, thank you!

> 
> Maxime

Re: [PATCH v3 0/4] drm/tests: Fix some memory leaks

[Jinjie Ruan]

On 2024/10/18 15:55, Maxime Ripard wrote:
> Hi,
> 
> On Thu, Oct 17, 2024 at 02:31:21PM GMT, Jinjie Ruan wrote:
>> Fix some memory leaks in drm tests.
>>
>> Changes in v3:
>> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
>> - Drop the "helper" in the helper name.
>> - s/fllowing/following/
>> - Add Acked-by.
> 
> This creates build failures since drm_display_mode were const before,
> and can't anymore.

It seems it came from bellowing v1, and this v3 has not reported the
issue yet.

https://lore.kernel.org/all/202410180830.oitxTsOv-lkp@intel.com/

> 
> Maxime

Re: [PATCH v3 0/4] drm/tests: Fix some memory leaks

[Jinjie Ruan]

On 2024/10/18 16:12, Jinjie Ruan wrote:
> 
> 
> On 2024/10/18 15:55, Maxime Ripard wrote:
>> Hi,
>>
>> On Thu, Oct 17, 2024 at 02:31:21PM GMT, Jinjie Ruan wrote:
>>> Fix some memory leaks in drm tests.
>>>
>>> Changes in v3:
>>> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
>>> - Drop the "helper" in the helper name.
>>> - s/fllowing/following/
>>> - Add Acked-by.
>>
>> This creates build failures since drm_display_mode were const before,
>> and can't anymore.
> 
> It seems it came from bellowing v1, and this v3 has not reported the
> issue yet.
> 
> https://lore.kernel.org/all/202410180830.oitxTsOv-lkp@intel.com/

Hi, Maxime,

Should this series send again? The issue seems not related to this version.

> 
>>
>> Maxime
> 
> 

Re: [PATCH v3 0/4] drm/tests: Fix some memory leaks

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 946 bytes --]

On Wed, Oct 23, 2024 at 09:35:59AM +0800, Jinjie Ruan wrote:
> 
> 
> On 2024/10/18 16:12, Jinjie Ruan wrote:
> > 
> > 
> > On 2024/10/18 15:55, Maxime Ripard wrote:
> >> Hi,
> >>
> >> On Thu, Oct 17, 2024 at 02:31:21PM GMT, Jinjie Ruan wrote:
> >>> Fix some memory leaks in drm tests.
> >>>
> >>> Changes in v3:
> >>> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
> >>> - Drop the "helper" in the helper name.
> >>> - s/fllowing/following/
> >>> - Add Acked-by.
> >>
> >> This creates build failures since drm_display_mode were const before,
> >> and can't anymore.
> > 
> > It seems it came from bellowing v1, and this v3 has not reported the
> > issue yet.
> > 
> > https://lore.kernel.org/all/202410180830.oitxTsOv-lkp@intel.com/
> 
> Hi, Maxime,
> 
> Should this series send again? The issue seems not related to this version.

As far as I know, the issues reported still apply there, so yes

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 273 bytes --]

Re: [PATCH v3 0/4] drm/tests: Fix some memory leaks

[Jinjie Ruan]

On 2024/10/25 22:33, Maxime Ripard wrote:
> On Wed, Oct 23, 2024 at 09:35:59AM +0800, Jinjie Ruan wrote:
>>
>>
>> On 2024/10/18 16:12, Jinjie Ruan wrote:
>>>
>>>
>>> On 2024/10/18 15:55, Maxime Ripard wrote:
>>>> Hi,
>>>>
>>>> On Thu, Oct 17, 2024 at 02:31:21PM GMT, Jinjie Ruan wrote:
>>>>> Fix some memory leaks in drm tests.
>>>>>
>>>>> Changes in v3:
>>>>> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
>>>>> - Drop the "helper" in the helper name.
>>>>> - s/fllowing/following/
>>>>> - Add Acked-by.
>>>>
>>>> This creates build failures since drm_display_mode were const before,
>>>> and can't anymore.
>>>
>>> It seems it came from bellowing v1, and this v3 has not reported the
>>> issue yet.
>>>
>>> https://lore.kernel.org/all/202410180830.oitxTsOv-lkp@intel.com/
>>
>> Hi, Maxime,
>>
>> Should this series send again? The issue seems not related to this version.
> 
> As far as I know, the issues reported still apply there, so yes

Hi, Maxime,

I make this version code with "C=2", there is no these build failures.

> 
> Maxime

Re: [PATCH v3 0/4] drm/tests: Fix some memory leaks

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 1271 bytes --]

On Sat, Oct 26, 2024 at 10:02:51AM +0800, Jinjie Ruan wrote:
> 
> 
> On 2024/10/25 22:33, Maxime Ripard wrote:
> > On Wed, Oct 23, 2024 at 09:35:59AM +0800, Jinjie Ruan wrote:
> >>
> >>
> >> On 2024/10/18 16:12, Jinjie Ruan wrote:
> >>>
> >>>
> >>> On 2024/10/18 15:55, Maxime Ripard wrote:
> >>>> Hi,
> >>>>
> >>>> On Thu, Oct 17, 2024 at 02:31:21PM GMT, Jinjie Ruan wrote:
> >>>>> Fix some memory leaks in drm tests.
> >>>>>
> >>>>> Changes in v3:
> >>>>> - Adjust drm/drm_edid.h header to drm_kunit_helpers.c.
> >>>>> - Drop the "helper" in the helper name.
> >>>>> - s/fllowing/following/
> >>>>> - Add Acked-by.
> >>>>
> >>>> This creates build failures since drm_display_mode were const before,
> >>>> and can't anymore.
> >>>
> >>> It seems it came from bellowing v1, and this v3 has not reported the
> >>> issue yet.
> >>>
> >>> https://lore.kernel.org/all/202410180830.oitxTsOv-lkp@intel.com/
> >>
> >> Hi, Maxime,
> >>
> >> Should this series send again? The issue seems not related to this version.
> > 
> > As far as I know, the issues reported still apply there, so yes
>
> I make this version code with "C=2", there is no these build failures.

Sorry, you're right. I still have a comment on the first patch

Maxime

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 273 bytes --]