Em 20/11/2025 11:53, Tvrtko Ursulin escreveu:
On 20/11/2025 14:39, André Almeida wrote:
Hi Tvrtko,
Em 12/11/2025 06:27, Tvrtko Ursulin escreveu:
Using a helper shrinks the code and separating the user and kernel slabs
is more secure.
As in the commit message, the commit description should be writing in
the imperative mood.
"Use a helper to shrink the code and separate the user and kernel slabs
for better security."
This sounds good?
All good to me, thanks! With both observations addressed:
Reviewed-by: André Almeida <[email protected]>
Signed-off-by: Tvrtko Ursulin <[email protected]>
---
v2:
* Remove now unused dev local.
---
drivers/gpu/drm/drm_gem.c | 13 +++----------
1 file changed, 3 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index a1a9c828938b..68168d58a7c8 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -783,7 +783,6 @@ static int objects_lookup(struct drm_file *filp,
u32 *handle, int count,
int drm_gem_objects_lookup(struct drm_file *filp, void __user
*bo_handles,
int count, struct drm_gem_object ***objs_out)
{
- struct drm_device *dev = filp->minor->dev;
struct drm_gem_object **objs;
u32 *handles;
int ret;
@@ -798,15 +797,9 @@ int drm_gem_objects_lookup(struct drm_file
*filp, void __user *bo_handles,
*objs_out = objs;
- handles = kvmalloc_array(count, sizeof(u32), GFP_KERNEL);
- if (!handles) {
- ret = -ENOMEM;
- goto out;
- }
-
- if (copy_from_user(handles, bo_handles, count * sizeof(u32))) {
- ret = -EFAULT;
- drm_dbg_core(dev, "Failed to copy in GEM handles\n");
Why the debugged message has been removed?
I should indeed have mentioned in the commit message:
"While at it lets remove the useless debug message."
Regards,
Tvrtko
+ handles = vmemdup_array_user(bo_handles, count, sizeof(u32));
+ if (IS_ERR(handles)) {
+ ret = PTR_ERR(handles);
goto out;
}
