| Message ID | 20250319133523.641009-1-memxor@gmail.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | BPF |
| Headers | show |
| Series | [for-next,v1] locking: Add __percpu tag to decode_tail qnode argument | expand |
| Context | Check | Description |
|---|---|---|
| netdev/tree_selection | success | Not a local patch |
On Wed, Mar 19, 2025 at 6:35 AM Kumar Kartikeya Dwivedi <memxor@gmail.com> wrote: > > The decode_tail function now takes a qnode argument to use distinct > qnodes for qspinlock and rqspinlock. Add a __percpu tag to avoid > warnings and errors and reflect that the qnodes are per-cpu. There > is no actual bug here as the value was not dereferenced directly and > passed to per_cpu_ptr, so this is just to suppress the warning/errors > at compile time. > > Fixes: 06988910ee2d ("locking: Move common qspinlock helpers to a private header") > Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> > --- > kernel/locking/qspinlock.h | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/kernel/locking/qspinlock.h b/kernel/locking/qspinlock.h > index d4ceb9490365..f63b6e04ddd4 100644 > --- a/kernel/locking/qspinlock.h > +++ b/kernel/locking/qspinlock.h > @@ -59,7 +59,7 @@ static inline __pure u32 encode_tail(int cpu, int idx) > return tail; > } > > -static inline __pure struct mcs_spinlock *decode_tail(u32 tail, struct qnode *qnodes) > +static inline __pure struct mcs_spinlock *decode_tail(u32 tail, struct qnode __percpu *qnodes) Applied. Decided to squash it in and adjusted to new line to make it shorter.
diff --git a/kernel/locking/qspinlock.h b/kernel/locking/qspinlock.h index d4ceb9490365..f63b6e04ddd4 100644 --- a/kernel/locking/qspinlock.h +++ b/kernel/locking/qspinlock.h @@ -59,7 +59,7 @@ static inline __pure u32 encode_tail(int cpu, int idx) return tail; } -static inline __pure struct mcs_spinlock *decode_tail(u32 tail, struct qnode *qnodes) +static inline __pure struct mcs_spinlock *decode_tail(u32 tail, struct qnode __percpu *qnodes) { int cpu = (tail >> _Q_TAIL_CPU_OFFSET) - 1; int idx = (tail & _Q_TAIL_IDX_MASK) >> _Q_TAIL_IDX_OFFSET;
The decode_tail function now takes a qnode argument to use distinct qnodes for qspinlock and rqspinlock. Add a __percpu tag to avoid warnings and errors and reflect that the qnodes are per-cpu. There is no actual bug here as the value was not dereferenced directly and passed to per_cpu_ptr, so this is just to suppress the warning/errors at compile time. Fixes: 06988910ee2d ("locking: Move common qspinlock helpers to a private header") Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> --- kernel/locking/qspinlock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) -- 2.47.1