kernel/trace/ftrace.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-)
The global notrace hash should be jointly decided by the intersection of
each subops's notrace hash, but not the filter hash.
Fixes: 5fccc7552ccb ("ftrace: Add subops logic to allow one ops to manage many")
Signed-off-by: Andy Chiu <andybnac@gmail.com>
---
kernel/trace/ftrace.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 1a48aedb5255..ee662f380b61 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -3526,18 +3526,17 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int
ftrace_hash_empty(subops->func_hash->notrace_hash)) {
notrace_hash = EMPTY_HASH;
} else {
- size_bits = max(ops->func_hash->filter_hash->size_bits,
- subops->func_hash->filter_hash->size_bits);
+ size_bits = max(ops->func_hash->notrace_hash->size_bits,
+ subops->func_hash->notrace_hash->size_bits);
notrace_hash = alloc_ftrace_hash(size_bits);
if (!notrace_hash) {
- free_ftrace_hash(filter_hash);
+ free_ftrace_hash(notrace_hash);
return -ENOMEM;
}
- ret = intersect_hash(¬race_hash, ops->func_hash->filter_hash,
- subops->func_hash->filter_hash);
+ ret = intersect_hash(¬race_hash, ops->func_hash->notrace_hash,
+ subops->func_hash->notrace_hash);
if (ret < 0) {
- free_ftrace_hash(filter_hash);
free_ftrace_hash(notrace_hash);
return ret;
}
--
2.39.3 (Apple Git-145)
On Tue, 8 Apr 2025 02:07:44 +0800
Andy Chiu <andybnac@gmail.com> wrote:
> The global notrace hash should be jointly decided by the intersection of
> each subops's notrace hash, but not the filter hash.
>
> Fixes: 5fccc7552ccb ("ftrace: Add subops logic to allow one ops to manage many")
> Signed-off-by: Andy Chiu <andybnac@gmail.com>
> ---
> kernel/trace/ftrace.c | 11 +++++------
> 1 file changed, 5 insertions(+), 6 deletions(-)
>
> diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
> index 1a48aedb5255..ee662f380b61 100644
> --- a/kernel/trace/ftrace.c
> +++ b/kernel/trace/ftrace.c
> @@ -3526,18 +3526,17 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int
> ftrace_hash_empty(subops->func_hash->notrace_hash)) {
> notrace_hash = EMPTY_HASH;
> } else {
> - size_bits = max(ops->func_hash->filter_hash->size_bits,
> - subops->func_hash->filter_hash->size_bits);
> + size_bits = max(ops->func_hash->notrace_hash->size_bits,
> + subops->func_hash->notrace_hash->size_bits);
> notrace_hash = alloc_ftrace_hash(size_bits);
> if (!notrace_hash) {
> - free_ftrace_hash(filter_hash);
> + free_ftrace_hash(notrace_hash);
> return -ENOMEM;
> }
>
> - ret = intersect_hash(¬race_hash, ops->func_hash->filter_hash,
> - subops->func_hash->filter_hash);
> + ret = intersect_hash(¬race_hash, ops->func_hash->notrace_hash,
> + subops->func_hash->notrace_hash);
Thanks for catching this.
> if (ret < 0) {
> - free_ftrace_hash(filter_hash);
The filter_hash still needs to be freed, as it could have been allocated in
the previous if statement and never used (both the filter_hash and
notrace_hash get used at the end of the function via ftrace_update_ops().
Care to send a v2?
-- Steve
> free_ftrace_hash(notrace_hash);
> return ret;
> }
Steven Rostedt <rostedt@goodmis.org> 於 2025年4月8日 週二 上午4:08寫道:
>
> On Tue, 8 Apr 2025 02:07:44 +0800
> Andy Chiu <andybnac@gmail.com> wrote:
>
> > The global notrace hash should be jointly decided by the intersection of
> > each subops's notrace hash, but not the filter hash.
> >
> > Fixes: 5fccc7552ccb ("ftrace: Add subops logic to allow one ops to manage many")
> > Signed-off-by: Andy Chiu <andybnac@gmail.com>
> > ---
> > kernel/trace/ftrace.c | 11 +++++------
> > 1 file changed, 5 insertions(+), 6 deletions(-)
> >
> > diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
> > index 1a48aedb5255..ee662f380b61 100644
> > --- a/kernel/trace/ftrace.c
> > +++ b/kernel/trace/ftrace.c
> > @@ -3526,18 +3526,17 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int
> > ftrace_hash_empty(subops->func_hash->notrace_hash)) {
> > notrace_hash = EMPTY_HASH;
> > } else {
> > - size_bits = max(ops->func_hash->filter_hash->size_bits,
> > - subops->func_hash->filter_hash->size_bits);
> > + size_bits = max(ops->func_hash->notrace_hash->size_bits,
> > + subops->func_hash->notrace_hash->size_bits);
> > notrace_hash = alloc_ftrace_hash(size_bits);
> > if (!notrace_hash) {
> > - free_ftrace_hash(filter_hash);
> > + free_ftrace_hash(notrace_hash);
> > return -ENOMEM;
> > }
> >
> > - ret = intersect_hash(¬race_hash, ops->func_hash->filter_hash,
> > - subops->func_hash->filter_hash);
> > + ret = intersect_hash(¬race_hash, ops->func_hash->notrace_hash,
> > + subops->func_hash->notrace_hash);
>
> Thanks for catching this.
>
>
> > if (ret < 0) {
> > - free_ftrace_hash(filter_hash);
>
> The filter_hash still needs to be freed, as it could have been allocated in
> the previous if statement and never used (both the filter_hash and
> notrace_hash get used at the end of the function via ftrace_update_ops().
>
> Care to send a v2?
Yes, thanks for reminding! Let me send a v2 with filter_hash freed on
this condition.
Regards,
Andy
© 2016 - 2026 Red Hat, Inc.