Revert "percpu: add __percpu to SHIFT_PERCPU_PTR"
This reverts commit 825dbc6ff7
.
I mistakenly applied this and only now have thought about it a little
more and had time to evaluate a kbuild error for dmaengine.
Once we're calling RELOC_HIDE, we're moving back into the __kernel
address space and letting users interact with the actual memory address
rather than in __percpu which is before adding the offsets.
Signed-off-by: Dennis Zhou <dennis@kernel.org>
This commit is contained in:
parent
9e8d42a0f7
commit
ba30e27405
|
@ -229,7 +229,7 @@ do { \
|
|||
* pointer value. The weird cast keeps both GCC and sparse happy.
|
||||
*/
|
||||
#define SHIFT_PERCPU_PTR(__p, __offset) \
|
||||
RELOC_HIDE((typeof(*(__p)) __kernel __percpu __force *)(__p), (__offset))
|
||||
RELOC_HIDE((typeof(*(__p)) __kernel __force *)(__p), (__offset))
|
||||
|
||||
#define per_cpu_ptr(ptr, cpu) \
|
||||
({ \
|
||||
|
|
Loading…
Reference in New Issue