[v5,3/3] mm,page_owner: Filter out stacks by a threshold counter
Commit Message
We want to be able to filter out the output on a threshold basis,
in this way we can get rid of a lot of noise and focus only on those
stacks which have an allegedly high counter.
We can control the threshold value by a new file called
'page_owner_threshold', which is 0 by default.
Signed-off-by: Oscar Salvador <osalvador@suse.de>
---
lib/stackdepot.c | 5 ++++-
mm/page_owner.c | 21 +++++++++++++++++++++
2 files changed, 25 insertions(+), 1 deletion(-)
Comments
On Tue, May 16, 2023 at 8:25 PM Oscar Salvador <osalvador@suse.de> wrote:
>
> We want to be able to filter out the output on a threshold basis,
> in this way we can get rid of a lot of noise and focus only on those
> stacks which have an allegedly high counter.
>
> We can control the threshold value by a new file called
> 'page_owner_threshold', which is 0 by default.
>
> Signed-off-by: Oscar Salvador <osalvador@suse.de>
Nack.
We must decouple stackdepot from page_owner as much as possible, not
add extra dependencies by sharing variables between the two.
What you need here is a pair of xxx_iter_start()/xxx_iter_next()
functions in stackdepot.c that will hold the iteration state in a
struct and return the next stack trace.
The threshold value can also be part of that struct.
On 5/22/23 10:40, Alexander Potapenko wrote:
> On Tue, May 16, 2023 at 8:25 PM Oscar Salvador <osalvador@suse.de> wrote:
>> We want to be able to filter out the output on a threshold basis,
>> in this way we can get rid of a lot of noise and focus only on those
>> stacks which have an allegedly high counter.
>>
>> We can control the threshold value by a new file called
>> 'page_owner_threshold', which is 0 by default.
>>
>> Signed-off-by: Oscar Salvador <osalvador@suse.de>
> Nack.
> We must decouple stackdepot from page_owner as much as possible, not
> add extra dependencies by sharing variables between the two.
>
> What you need here is a pair of xxx_iter_start()/xxx_iter_next()
> functions in stackdepot.c that will hold the iteration state in a
> struct and return the next stack trace.
> The threshold value can also be part of that struct.
Agreed, that should be a much cleaner approach.
@@ -488,6 +488,9 @@ static struct stack_record *stack_depot_getstack(depot_stack_handle_t handle)
}
#ifdef CONFIG_PAGE_OWNER
+
+extern unsigned long page_owner_stack_threshold;
+
void *stack_start(struct seq_file *m, loff_t *ppos)
{
unsigned long *table = m->private;
@@ -543,7 +546,7 @@ int stack_print(struct seq_file *m, void *v)
if (!stack->size || stack->size < 0 ||
stack->size > PAGE_SIZE || stack->handle.valid != 1 ||
- refcount_read(&stack->count) < 1)
+ refcount_read(&stack->count) < page_owner_stack_threshold)
return 0;
buf = kzalloc(PAGE_SIZE, GFP_KERNEL);
@@ -743,6 +743,23 @@ const struct file_operations page_owner_stack_operations = {
.release = seq_release,
};
+unsigned long page_owner_stack_threshold;
+
+int page_owner_threshold_get(void *data, u64 *val)
+{
+ *val = page_owner_stack_threshold;
+ return 0;
+}
+
+int page_owner_threshold_set(void *data, u64 val)
+{
+ page_owner_stack_threshold = val;
+ return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(proc_page_owner_threshold, &page_owner_threshold_get,
+ &page_owner_threshold_set, "%llu");
+
static int __init pageowner_init(void)
{
if (!static_branch_unlikely(&page_owner_inited)) {
@@ -755,6 +772,10 @@ static int __init pageowner_init(void)
debugfs_create_file("page_owner_stacks", 0400, NULL, NULL,
&page_owner_stack_operations);
+ debugfs_create_file("page_owner_threshold", 0600, NULL, NULL,
+ &proc_page_owner_threshold);
+
+ page_owner_stack_threshold = 0;
return 0;
}