dri-devel.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects
@ 2025-08-29  7:55 Danilo Krummrich
  2025-08-29  8:00 ` Nirmoy Das
                   ` (4 more replies)
  0 siblings, 5 replies; 6+ messages in thread
From: Danilo Krummrich @ 2025-08-29  7:55 UTC (permalink / raw)
  To: maarten.lankhorst, mripard, tzimmermann, airlied, simona
  Cc: dri-devel, Danilo Krummrich, Alice Ryhl, Christian König

Since commit e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
it is possible for test_prepare_array() to exceed a stack frame size of
2048 bytes depending on the exact configuration of the kernel.

  drivers/gpu/drm/tests/drm_exec_test.c: In function ‘test_prepare_array’:
  drivers/gpu/drm/tests/drm_exec_test.c:171:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
    171 | }
        | ^
  cc1: all warnings being treated as errors
  make[6]: *** [scripts/Makefile.build:287: drivers/gpu/drm/tests/drm_exec_test.o] Error 1
  make[6]: *** Waiting for unfinished jobs....

In order to fix this, allocate the GEM objects in test_prepare_array()
with kzalloc(), rather than placing them on the stack.

Cc: Alice Ryhl <aliceryhl@google.com>
Cc: Christian König <christian.koenig@amd.com>
Fixes: e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/tests/drm_exec_test.c | 26 +++++++++++++++++++-------
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/tests/drm_exec_test.c b/drivers/gpu/drm/tests/drm_exec_test.c
index d6c4dd1194a0..d59ec3baae1d 100644
--- a/drivers/gpu/drm/tests/drm_exec_test.c
+++ b/drivers/gpu/drm/tests/drm_exec_test.c
@@ -150,14 +150,22 @@ static void test_prepare(struct kunit *test)
 static void test_prepare_array(struct kunit *test)
 {
 	struct drm_exec_priv *priv = test->priv;
-	struct drm_gem_object gobj1 = { };
-	struct drm_gem_object gobj2 = { };
-	struct drm_gem_object *array[] = { &gobj1, &gobj2 };
+	struct drm_gem_object *gobj1;
+	struct drm_gem_object *gobj2;
+	struct drm_gem_object *array[] = {
+		(gobj1 = kzalloc(sizeof(*gobj1), GFP_KERNEL)),
+		(gobj2 = kzalloc(sizeof(*gobj2), GFP_KERNEL)),
+	};
 	struct drm_exec exec;
 	int ret;
 
-	drm_gem_private_object_init(priv->drm, &gobj1, PAGE_SIZE);
-	drm_gem_private_object_init(priv->drm, &gobj2, PAGE_SIZE);
+	if (!gobj1 || !gobj2) {
+		KUNIT_FAIL(test, "Failed to allocate GEM objects.\n");
+		goto out;
+	}
+
+	drm_gem_private_object_init(priv->drm, gobj1, PAGE_SIZE);
+	drm_gem_private_object_init(priv->drm, gobj2, PAGE_SIZE);
 
 	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
 	drm_exec_until_all_locked(&exec)
@@ -166,8 +174,12 @@ static void test_prepare_array(struct kunit *test)
 	KUNIT_EXPECT_EQ(test, ret, 0);
 	drm_exec_fini(&exec);
 
-	drm_gem_private_object_fini(&gobj1);
-	drm_gem_private_object_fini(&gobj2);
+	drm_gem_private_object_fini(gobj1);
+	drm_gem_private_object_fini(gobj2);
+
+out:
+	kfree(gobj1);
+	kfree(gobj2);
 }
 
 static void test_multiple_loops(struct kunit *test)

base-commit: 0b6da6d3a8d5391ff9da2358ac5afd8b7badd943
-- 
2.51.0


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects
  2025-08-29  7:55 [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects Danilo Krummrich
@ 2025-08-29  8:00 ` Nirmoy Das
  2025-08-29  8:41 ` Alice Ryhl
                   ` (3 subsequent siblings)
  4 siblings, 0 replies; 6+ messages in thread
From: Nirmoy Das @ 2025-08-29  8:00 UTC (permalink / raw)
  To: Danilo Krummrich, maarten.lankhorst, mripard, tzimmermann,
	airlied, simona
  Cc: dri-devel, Alice Ryhl, Christian König


On 29.08.25 09:55, Danilo Krummrich wrote:
> Since commit e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> it is possible for test_prepare_array() to exceed a stack frame size of
> 2048 bytes depending on the exact configuration of the kernel.
>
>    drivers/gpu/drm/tests/drm_exec_test.c: In function ‘test_prepare_array’:
>    drivers/gpu/drm/tests/drm_exec_test.c:171:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
>      171 | }
>          | ^
>    cc1: all warnings being treated as errors
>    make[6]: *** [scripts/Makefile.build:287: drivers/gpu/drm/tests/drm_exec_test.o] Error 1
>    make[6]: *** Waiting for unfinished jobs....
>
> In order to fix this, allocate the GEM objects in test_prepare_array()
> with kzalloc(), rather than placing them on the stack.
>
> Cc: Alice Ryhl <aliceryhl@google.com>
> Cc: Christian König <christian.koenig@amd.com>
> Fixes: e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Nirmoy Das <nirmoyd@nvidia.com>
> ---
>   drivers/gpu/drm/tests/drm_exec_test.c | 26 +++++++++++++++++++-------
>   1 file changed, 19 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/tests/drm_exec_test.c b/drivers/gpu/drm/tests/drm_exec_test.c
> index d6c4dd1194a0..d59ec3baae1d 100644
> --- a/drivers/gpu/drm/tests/drm_exec_test.c
> +++ b/drivers/gpu/drm/tests/drm_exec_test.c
> @@ -150,14 +150,22 @@ static void test_prepare(struct kunit *test)
>   static void test_prepare_array(struct kunit *test)
>   {
>   	struct drm_exec_priv *priv = test->priv;
> -	struct drm_gem_object gobj1 = { };
> -	struct drm_gem_object gobj2 = { };
> -	struct drm_gem_object *array[] = { &gobj1, &gobj2 };
> +	struct drm_gem_object *gobj1;
> +	struct drm_gem_object *gobj2;
> +	struct drm_gem_object *array[] = {
> +		(gobj1 = kzalloc(sizeof(*gobj1), GFP_KERNEL)),
> +		(gobj2 = kzalloc(sizeof(*gobj2), GFP_KERNEL)),
> +	};
>   	struct drm_exec exec;
>   	int ret;
>   
> -	drm_gem_private_object_init(priv->drm, &gobj1, PAGE_SIZE);
> -	drm_gem_private_object_init(priv->drm, &gobj2, PAGE_SIZE);
> +	if (!gobj1 || !gobj2) {
> +		KUNIT_FAIL(test, "Failed to allocate GEM objects.\n");
> +		goto out;
> +	}
> +
> +	drm_gem_private_object_init(priv->drm, gobj1, PAGE_SIZE);
> +	drm_gem_private_object_init(priv->drm, gobj2, PAGE_SIZE);
>   
>   	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
>   	drm_exec_until_all_locked(&exec)
> @@ -166,8 +174,12 @@ static void test_prepare_array(struct kunit *test)
>   	KUNIT_EXPECT_EQ(test, ret, 0);
>   	drm_exec_fini(&exec);
>   
> -	drm_gem_private_object_fini(&gobj1);
> -	drm_gem_private_object_fini(&gobj2);
> +	drm_gem_private_object_fini(gobj1);
> +	drm_gem_private_object_fini(gobj2);
> +
> +out:
> +	kfree(gobj1);
> +	kfree(gobj2);
>   }
>   
>   static void test_multiple_loops(struct kunit *test)
>
> base-commit: 0b6da6d3a8d5391ff9da2358ac5afd8b7badd943

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects
  2025-08-29  7:55 [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects Danilo Krummrich
  2025-08-29  8:00 ` Nirmoy Das
@ 2025-08-29  8:41 ` Alice Ryhl
  2025-08-29 10:54 ` Christian König
                   ` (2 subsequent siblings)
  4 siblings, 0 replies; 6+ messages in thread
From: Alice Ryhl @ 2025-08-29  8:41 UTC (permalink / raw)
  To: Danilo Krummrich
  Cc: maarten.lankhorst, mripard, tzimmermann, airlied, simona,
	dri-devel, Christian König

On Fri, Aug 29, 2025 at 09:55:39AM +0200, Danilo Krummrich wrote:
> Since commit e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> it is possible for test_prepare_array() to exceed a stack frame size of
> 2048 bytes depending on the exact configuration of the kernel.
> 
>   drivers/gpu/drm/tests/drm_exec_test.c: In function ‘test_prepare_array’:
>   drivers/gpu/drm/tests/drm_exec_test.c:171:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
>     171 | }
>         | ^
>   cc1: all warnings being treated as errors
>   make[6]: *** [scripts/Makefile.build:287: drivers/gpu/drm/tests/drm_exec_test.o] Error 1
>   make[6]: *** Waiting for unfinished jobs....
> 
> In order to fix this, allocate the GEM objects in test_prepare_array()
> with kzalloc(), rather than placing them on the stack.
> 
> Cc: Alice Ryhl <aliceryhl@google.com>
> Cc: Christian König <christian.koenig@amd.com>
> Fixes: e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> Signed-off-by: Danilo Krummrich <dakr@kernel.org>

Reviewed-by: Alice Ryhl <aliceryhl@google.com>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects
  2025-08-29  7:55 [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects Danilo Krummrich
  2025-08-29  8:00 ` Nirmoy Das
  2025-08-29  8:41 ` Alice Ryhl
@ 2025-08-29 10:54 ` Christian König
  2025-08-29 11:03 ` Danilo Krummrich
  2025-08-29 16:03 ` Danilo Krummrich
  4 siblings, 0 replies; 6+ messages in thread
From: Christian König @ 2025-08-29 10:54 UTC (permalink / raw)
  To: Danilo Krummrich, maarten.lankhorst, mripard, tzimmermann,
	airlied, simona
  Cc: dri-devel, Alice Ryhl



On 29.08.25 09:55, Danilo Krummrich wrote:
> Since commit e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> it is possible for test_prepare_array() to exceed a stack frame size of
> 2048 bytes depending on the exact configuration of the kernel.
> 
>   drivers/gpu/drm/tests/drm_exec_test.c: In function ‘test_prepare_array’:
>   drivers/gpu/drm/tests/drm_exec_test.c:171:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
>     171 | }
>         | ^
>   cc1: all warnings being treated as errors
>   make[6]: *** [scripts/Makefile.build:287: drivers/gpu/drm/tests/drm_exec_test.o] Error 1
>   make[6]: *** Waiting for unfinished jobs....
> 
> In order to fix this, allocate the GEM objects in test_prepare_array()
> with kzalloc(), rather than placing them on the stack.
> 
> Cc: Alice Ryhl <aliceryhl@google.com>
> Cc: Christian König <christian.koenig@amd.com>
> Fixes: e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> Signed-off-by: Danilo Krummrich <dakr@kernel.org>

Reviewed-by: Christian König <christian.koenig@amd.com>

> ---
>  drivers/gpu/drm/tests/drm_exec_test.c | 26 +++++++++++++++++++-------
>  1 file changed, 19 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/tests/drm_exec_test.c b/drivers/gpu/drm/tests/drm_exec_test.c
> index d6c4dd1194a0..d59ec3baae1d 100644
> --- a/drivers/gpu/drm/tests/drm_exec_test.c
> +++ b/drivers/gpu/drm/tests/drm_exec_test.c
> @@ -150,14 +150,22 @@ static void test_prepare(struct kunit *test)
>  static void test_prepare_array(struct kunit *test)
>  {
>  	struct drm_exec_priv *priv = test->priv;
> -	struct drm_gem_object gobj1 = { };
> -	struct drm_gem_object gobj2 = { };
> -	struct drm_gem_object *array[] = { &gobj1, &gobj2 };
> +	struct drm_gem_object *gobj1;
> +	struct drm_gem_object *gobj2;
> +	struct drm_gem_object *array[] = {
> +		(gobj1 = kzalloc(sizeof(*gobj1), GFP_KERNEL)),
> +		(gobj2 = kzalloc(sizeof(*gobj2), GFP_KERNEL)),
> +	};
>  	struct drm_exec exec;
>  	int ret;
>  
> -	drm_gem_private_object_init(priv->drm, &gobj1, PAGE_SIZE);
> -	drm_gem_private_object_init(priv->drm, &gobj2, PAGE_SIZE);
> +	if (!gobj1 || !gobj2) {
> +		KUNIT_FAIL(test, "Failed to allocate GEM objects.\n");
> +		goto out;
> +	}
> +
> +	drm_gem_private_object_init(priv->drm, gobj1, PAGE_SIZE);
> +	drm_gem_private_object_init(priv->drm, gobj2, PAGE_SIZE);
>  
>  	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
>  	drm_exec_until_all_locked(&exec)
> @@ -166,8 +174,12 @@ static void test_prepare_array(struct kunit *test)
>  	KUNIT_EXPECT_EQ(test, ret, 0);
>  	drm_exec_fini(&exec);
>  
> -	drm_gem_private_object_fini(&gobj1);
> -	drm_gem_private_object_fini(&gobj2);
> +	drm_gem_private_object_fini(gobj1);
> +	drm_gem_private_object_fini(gobj2);
> +
> +out:
> +	kfree(gobj1);
> +	kfree(gobj2);
>  }
>  
>  static void test_multiple_loops(struct kunit *test)
> 
> base-commit: 0b6da6d3a8d5391ff9da2358ac5afd8b7badd943


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects
  2025-08-29  7:55 [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects Danilo Krummrich
                   ` (2 preceding siblings ...)
  2025-08-29 10:54 ` Christian König
@ 2025-08-29 11:03 ` Danilo Krummrich
  2025-08-29 16:03 ` Danilo Krummrich
  4 siblings, 0 replies; 6+ messages in thread
From: Danilo Krummrich @ 2025-08-29 11:03 UTC (permalink / raw)
  To: maarten.lankhorst, mripard, tzimmermann, airlied, simona
  Cc: dri-devel, Alice Ryhl, Christian König

On Fri Aug 29, 2025 at 9:55 AM CEST, Danilo Krummrich wrote:
> diff --git a/drivers/gpu/drm/tests/drm_exec_test.c b/drivers/gpu/drm/tests/drm_exec_test.c
> index d6c4dd1194a0..d59ec3baae1d 100644
> --- a/drivers/gpu/drm/tests/drm_exec_test.c
> +++ b/drivers/gpu/drm/tests/drm_exec_test.c
> @@ -150,14 +150,22 @@ static void test_prepare(struct kunit *test)
>  static void test_prepare_array(struct kunit *test)
>  {
>  	struct drm_exec_priv *priv = test->priv;
> -	struct drm_gem_object gobj1 = { };
> -	struct drm_gem_object gobj2 = { };
> -	struct drm_gem_object *array[] = { &gobj1, &gobj2 };
> +	struct drm_gem_object *gobj1;
> +	struct drm_gem_object *gobj2;
> +	struct drm_gem_object *array[] = {
> +		(gobj1 = kzalloc(sizeof(*gobj1), GFP_KERNEL)),
> +		(gobj2 = kzalloc(sizeof(*gobj2), GFP_KERNEL)),
> +	};

Actually, I think this should use kunit_kzmalloc() instead. Unless anyone
disagrees, I'd apply the following hunk when applying the patch.

diff --git a/drivers/gpu/drm/tests/drm_exec_test.c b/drivers/gpu/drm/tests/drm_exec_test.c
index d59ec3baae1d..3a20c788c51f 100644
--- a/drivers/gpu/drm/tests/drm_exec_test.c
+++ b/drivers/gpu/drm/tests/drm_exec_test.c
@@ -153,15 +153,15 @@ static void test_prepare_array(struct kunit *test)
        struct drm_gem_object *gobj1;
        struct drm_gem_object *gobj2;
        struct drm_gem_object *array[] = {
-               (gobj1 = kzalloc(sizeof(*gobj1), GFP_KERNEL)),
-               (gobj2 = kzalloc(sizeof(*gobj2), GFP_KERNEL)),
+               (gobj1 = kunit_kzalloc(test, sizeof(*gobj1), GFP_KERNEL)),
+               (gobj2 = kunit_kzalloc(test, sizeof(*gobj2), GFP_KERNEL)),
        };
        struct drm_exec exec;
        int ret;

        if (!gobj1 || !gobj2) {
                KUNIT_FAIL(test, "Failed to allocate GEM objects.\n");
-               goto out;
+               return;
        }

        drm_gem_private_object_init(priv->drm, gobj1, PAGE_SIZE);
@@ -176,10 +176,6 @@ static void test_prepare_array(struct kunit *test)

        drm_gem_private_object_fini(gobj1);
        drm_gem_private_object_fini(gobj2);
-
-out:
-       kfree(gobj1);
-       kfree(gobj2);
 }

 static void test_multiple_loops(struct kunit *test)

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects
  2025-08-29  7:55 [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects Danilo Krummrich
                   ` (3 preceding siblings ...)
  2025-08-29 11:03 ` Danilo Krummrich
@ 2025-08-29 16:03 ` Danilo Krummrich
  4 siblings, 0 replies; 6+ messages in thread
From: Danilo Krummrich @ 2025-08-29 16:03 UTC (permalink / raw)
  To: maarten.lankhorst, mripard, tzimmermann, airlied, simona
  Cc: dri-devel, Alice Ryhl, Christian König

On Fri Aug 29, 2025 at 9:55 AM CEST, Danilo Krummrich wrote:
> Since commit e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> it is possible for test_prepare_array() to exceed a stack frame size of
> 2048 bytes depending on the exact configuration of the kernel.
>
>   drivers/gpu/drm/tests/drm_exec_test.c: In function ‘test_prepare_array’:
>   drivers/gpu/drm/tests/drm_exec_test.c:171:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
>     171 | }
>         | ^
>   cc1: all warnings being treated as errors
>   make[6]: *** [scripts/Makefile.build:287: drivers/gpu/drm/tests/drm_exec_test.o] Error 1
>   make[6]: *** Waiting for unfinished jobs....
>
> In order to fix this, allocate the GEM objects in test_prepare_array()
> with kzalloc(), rather than placing them on the stack.
>
> Cc: Alice Ryhl <aliceryhl@google.com>
> Cc: Christian König <christian.koenig@amd.com>
> Fixes: e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
> Signed-off-by: Danilo Krummrich <dakr@kernel.org>

Applied to drm-misc-next, thanks!

    [ Use kunit_kzalloc() instead of kzalloc(). - Danilo ]

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2025-08-29 16:04 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-08-29  7:55 [PATCH] drm/test: drm_exec: use kzalloc() to allocate GEM objects Danilo Krummrich
2025-08-29  8:00 ` Nirmoy Das
2025-08-29  8:41 ` Alice Ryhl
2025-08-29 10:54 ` Christian König
2025-08-29 11:03 ` Danilo Krummrich
2025-08-29 16:03 ` Danilo Krummrich

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