* [PATCH v2 0/4] Remove usage of list iterator after the loop body
@ 2022-04-02 10:33 Jakob Koschel
2022-04-02 10:33 ` [PATCH v2 1/4] tracing: " Jakob Koschel
` (3 more replies)
0 siblings, 4 replies; 9+ messages in thread
From: Jakob Koschel @ 2022-04-02 10:33 UTC (permalink / raw)
To: Steven Rostedt
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.,
Jakob Koschel
In preparation to limit the scope of a list iterator to the list
traversal loop, use a dedicated pointer as the iterator [1].
Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/
v1->v2:
- fix reverse x-mas tree (Steven Rostedt)
- rename subject line (Steven Rostedt)
- add additional tracing patches to bundle them all up in one go
Sorry, I corrupted the Subject line on the previous send, please ignore :(
Jakob Koschel (4):
tracing: Remove usage of list iterator after the loop body
tracing: Remove usage of list iterator variable after the loop
tracing: Replace usage of found with dedicated list iterator variable
tracing: Remove check of list iterator against head past the loop body
kernel/trace/ftrace.c | 20 ++++++++++++--------
kernel/trace/trace_eprobe.c | 14 ++++++++------
kernel/trace/trace_events.c | 29 ++++++++++++++---------------
kernel/trace/trace_events_hist.c | 17 ++++++++---------
kernel/trace/trace_events_trigger.c | 28 +++++++++++++---------------
kernel/trace/trace_output.c | 13 +++++++++----
6 files changed, 64 insertions(+), 57 deletions(-)
base-commit: 7a3ecddc571cc3294e5d6bb5948ff2b0cfa12735
--
2.25.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH v2 1/4] tracing: Remove usage of list iterator after the loop body
2022-04-02 10:33 [PATCH v2 0/4] Remove usage of list iterator after the loop body Jakob Koschel
@ 2022-04-02 10:33 ` Jakob Koschel
2022-04-22 22:59 ` Steven Rostedt
2022-04-02 10:33 ` [PATCH v2 2/4] tracing: Remove usage of list iterator variable after the loop Jakob Koschel
` (2 subsequent siblings)
3 siblings, 1 reply; 9+ messages in thread
From: Jakob Koschel @ 2022-04-02 10:33 UTC (permalink / raw)
To: Steven Rostedt
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.,
Jakob Koschel
In preparation to limit the scope of the list iterator variable to the
traversal loop, use a dedicated pointer to point to the found element
[1].
Before, the code implicitly used the head when no element was found
when using &pos->list. Since the new variable is only set if an
element was found, the head needs to be used explicitly if the
variable is NULL.
Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1]
Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
---
kernel/trace/trace_output.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c
index 8aa493d25c73..733a4d6c20e2 100644
--- a/kernel/trace/trace_output.c
+++ b/kernel/trace/trace_output.c
@@ -692,7 +692,7 @@ static LIST_HEAD(ftrace_event_list);
static int trace_search_list(struct list_head **list)
{
- struct trace_event *e;
+ struct trace_event *e = NULL, *iter;
int next = __TRACE_LAST_TYPE;
if (list_empty(&ftrace_event_list)) {
@@ -704,9 +704,11 @@ static int trace_search_list(struct list_head **list)
* We used up all possible max events,
* lets see if somebody freed one.
*/
- list_for_each_entry(e, &ftrace_event_list, list) {
- if (e->type != next)
+ list_for_each_entry(iter, &ftrace_event_list, list) {
+ if (iter->type != next) {
+ e = iter;
break;
+ }
next++;
}
@@ -714,7 +716,10 @@ static int trace_search_list(struct list_head **list)
if (next > TRACE_EVENT_TYPE_MAX)
return 0;
- *list = &e->list;
+ if (e)
+ *list = &e->list;
+ else
+ *list = &ftrace_event_list;
return next;
}
--
2.25.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v2 2/4] tracing: Remove usage of list iterator variable after the loop
2022-04-02 10:33 [PATCH v2 0/4] Remove usage of list iterator after the loop body Jakob Koschel
2022-04-02 10:33 ` [PATCH v2 1/4] tracing: " Jakob Koschel
@ 2022-04-02 10:33 ` Jakob Koschel
2022-04-26 14:23 ` Steven Rostedt
2022-04-02 10:33 ` [PATCH v2 3/4] tracing: Replace usage of found with dedicated list iterator variable Jakob Koschel
2022-04-02 10:33 ` [PATCH v2 4/4] tracing: Remove check of list iterator against head past the loop body Jakob Koschel
3 siblings, 1 reply; 9+ messages in thread
From: Jakob Koschel @ 2022-04-02 10:33 UTC (permalink / raw)
To: Steven Rostedt
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.,
Jakob Koschel
In preparation to limit the scope of a list iterator to the list
traversal loop, use a dedicated pointer to point to the found element
[1].
Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1]
Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
---
kernel/trace/trace_events.c | 17 ++++++++---------
1 file changed, 8 insertions(+), 9 deletions(-)
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index e11e167b7809..97c7eb2f55e5 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -1723,9 +1723,9 @@ static LIST_HEAD(event_subsystems);
static int subsystem_open(struct inode *inode, struct file *filp)
{
+ struct trace_subsystem_dir *dir = NULL, *iter_dir;
+ struct trace_array *tr = NULL, *iter_tr;
struct event_subsystem *system = NULL;
- struct trace_subsystem_dir *dir = NULL; /* Initialize for gcc */
- struct trace_array *tr;
int ret;
if (tracing_is_disabled())
@@ -1734,14 +1734,16 @@ static int subsystem_open(struct inode *inode, struct file *filp)
/* Make sure the system still exists */
mutex_lock(&event_mutex);
mutex_lock(&trace_types_lock);
- list_for_each_entry(tr, &ftrace_trace_arrays, list) {
- list_for_each_entry(dir, &tr->systems, list) {
- if (dir == inode->i_private) {
+ list_for_each_entry(iter_tr, &ftrace_trace_arrays, list) {
+ list_for_each_entry(iter_dir, &iter_tr->systems, list) {
+ if (iter_dir == inode->i_private) {
/* Don't open systems with no events */
- if (dir->nr_events) {
+ if (iter_dir->nr_events) {
__get_system_dir(dir);
system = dir->subsystem;
}
+ tr = iter_tr;
+ dir = iter_dir;
goto exit_loop;
}
}
@@ -1753,9 +1755,6 @@ static int subsystem_open(struct inode *inode, struct file *filp)
if (!system)
return -ENODEV;
- /* Some versions of gcc think dir can be uninitialized here */
- WARN_ON(!dir);
-
/* Still need to increment the ref count of the system */
if (trace_array_get(tr) < 0) {
put_system(dir);
--
2.25.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v2 3/4] tracing: Replace usage of found with dedicated list iterator variable
2022-04-02 10:33 [PATCH v2 0/4] Remove usage of list iterator after the loop body Jakob Koschel
2022-04-02 10:33 ` [PATCH v2 1/4] tracing: " Jakob Koschel
2022-04-02 10:33 ` [PATCH v2 2/4] tracing: Remove usage of list iterator variable after the loop Jakob Koschel
@ 2022-04-02 10:33 ` Jakob Koschel
2022-04-26 14:25 ` Steven Rostedt
2022-04-02 10:33 ` [PATCH v2 4/4] tracing: Remove check of list iterator against head past the loop body Jakob Koschel
3 siblings, 1 reply; 9+ messages in thread
From: Jakob Koschel @ 2022-04-02 10:33 UTC (permalink / raw)
To: Steven Rostedt
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.,
Jakob Koschel
To move the list iterator variable into the list_for_each_entry_*()
macro in the future it should be avoided to use the list iterator
variable after the loop body.
To *never* use the list iterator variable after the loop it was
concluded to use a separate iterator variable instead of a
found boolean [1].
This removes the need to use a found variable and simply checking if
the variable was set, can determine if the break/goto was hit.
Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1]
Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
---
kernel/trace/trace_events_hist.c | 17 ++++++++---------
kernel/trace/trace_events_trigger.c | 28 +++++++++++++---------------
2 files changed, 21 insertions(+), 24 deletions(-)
diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index 44db5ba9cabb..7f60d04d5b6e 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -6089,32 +6089,31 @@ static void hist_unregister_trigger(char *glob,
struct event_trigger_data *data,
struct trace_event_file *file)
{
+ struct event_trigger_data *test = NULL, *iter, *named_data = NULL;
struct hist_trigger_data *hist_data = data->private_data;
- struct event_trigger_data *test, *named_data = NULL;
- bool unregistered = false;
lockdep_assert_held(&event_mutex);
if (hist_data->attrs->name)
named_data = find_named_trigger(hist_data->attrs->name);
- list_for_each_entry(test, &file->triggers, list) {
- if (test->cmd_ops->trigger_type == ETT_EVENT_HIST) {
- if (!hist_trigger_match(data, test, named_data, false))
+ list_for_each_entry(iter, &file->triggers, list) {
+ if (iter->cmd_ops->trigger_type == ETT_EVENT_HIST) {
+ if (!hist_trigger_match(data, iter, named_data, false))
continue;
- unregistered = true;
- list_del_rcu(&test->list);
+ test = iter;
+ list_del_rcu(&iter->list);
trace_event_trigger_enable_disable(file, 0);
update_cond_flag(file);
break;
}
}
- if (unregistered && test->ops->free)
+ if (test && test->ops->free)
test->ops->free(test->ops, test);
if (hist_data->enable_timestamps) {
- if (!hist_data->remove || unregistered)
+ if (!hist_data->remove || test)
tracing_set_filter_buffering(file->tr, false);
}
}
diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c
index 7eb9d04f1c2e..1ba54a489416 100644
--- a/kernel/trace/trace_events_trigger.c
+++ b/kernel/trace/trace_events_trigger.c
@@ -614,22 +614,21 @@ static void unregister_trigger(char *glob,
struct event_trigger_data *test,
struct trace_event_file *file)
{
- struct event_trigger_data *data;
- bool unregistered = false;
+ struct event_trigger_data *data = NULL, *iter;
lockdep_assert_held(&event_mutex);
- list_for_each_entry(data, &file->triggers, list) {
- if (data->cmd_ops->trigger_type == test->cmd_ops->trigger_type) {
- unregistered = true;
- list_del_rcu(&data->list);
+ list_for_each_entry(iter, &file->triggers, list) {
+ if (iter->cmd_ops->trigger_type == test->cmd_ops->trigger_type) {
+ data = iter;
+ list_del_rcu(&iter->list);
trace_event_trigger_enable_disable(file, 0);
update_cond_flag(file);
break;
}
}
- if (unregistered && data->ops->free)
+ if (data && data->ops->free)
data->ops->free(data->ops, data);
}
@@ -1976,27 +1975,26 @@ void event_enable_unregister_trigger(char *glob,
struct trace_event_file *file)
{
struct enable_trigger_data *test_enable_data = test->private_data;
+ struct event_trigger_data *data = NULL, *iter;
struct enable_trigger_data *enable_data;
- struct event_trigger_data *data;
- bool unregistered = false;
lockdep_assert_held(&event_mutex);
- list_for_each_entry(data, &file->triggers, list) {
- enable_data = data->private_data;
+ list_for_each_entry(iter, &file->triggers, list) {
+ enable_data = iter->private_data;
if (enable_data &&
- (data->cmd_ops->trigger_type ==
+ (iter->cmd_ops->trigger_type ==
test->cmd_ops->trigger_type) &&
(enable_data->file == test_enable_data->file)) {
- unregistered = true;
- list_del_rcu(&data->list);
+ data = iter;
+ list_del_rcu(&iter->list);
trace_event_trigger_enable_disable(file, 0);
update_cond_flag(file);
break;
}
}
- if (unregistered && data->ops->free)
+ if (data && data->ops->free)
data->ops->free(data->ops, data);
}
--
2.25.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v2 4/4] tracing: Remove check of list iterator against head past the loop body
2022-04-02 10:33 [PATCH v2 0/4] Remove usage of list iterator after the loop body Jakob Koschel
` (2 preceding siblings ...)
2022-04-02 10:33 ` [PATCH v2 3/4] tracing: Replace usage of found with dedicated list iterator variable Jakob Koschel
@ 2022-04-02 10:33 ` Jakob Koschel
3 siblings, 0 replies; 9+ messages in thread
From: Jakob Koschel @ 2022-04-02 10:33 UTC (permalink / raw)
To: Steven Rostedt
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.,
Jakob Koschel
When list_for_each_entry() completes the iteration over the whole list
without breaking the loop, the iterator value will be a bogus pointer
computed based on the head element.
While it is safe to use the pointer to determine if it was computed
based on the head element, either with list_entry_is_head() or
&pos->member == head, using the iterator variable after the loop should
be avoided.
In preparation to limit the scope of a list iterator to the list
traversal loop, use a dedicated pointer to point to the found element [1].
Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/
Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
---
kernel/trace/ftrace.c | 20 ++++++++++++--------
kernel/trace/trace_eprobe.c | 14 ++++++++------
kernel/trace/trace_events.c | 12 ++++++------
3 files changed, 26 insertions(+), 20 deletions(-)
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 4f1d2f5e7263..5c465e70d146 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -4560,8 +4560,8 @@ register_ftrace_function_probe(char *glob, struct trace_array *tr,
struct ftrace_probe_ops *probe_ops,
void *data)
{
+ struct ftrace_func_probe *probe = NULL, *iter;
struct ftrace_func_entry *entry;
- struct ftrace_func_probe *probe;
struct ftrace_hash **orig_hash;
struct ftrace_hash *old_hash;
struct ftrace_hash *hash;
@@ -4580,11 +4580,13 @@ register_ftrace_function_probe(char *glob, struct trace_array *tr,
mutex_lock(&ftrace_lock);
/* Check if the probe_ops is already registered */
- list_for_each_entry(probe, &tr->func_probes, list) {
- if (probe->probe_ops == probe_ops)
+ list_for_each_entry(iter, &tr->func_probes, list) {
+ if (iter->probe_ops == probe_ops) {
+ probe = iter;
break;
+ }
}
- if (&probe->list == &tr->func_probes) {
+ if (!probe) {
probe = kzalloc(sizeof(*probe), GFP_KERNEL);
if (!probe) {
mutex_unlock(&ftrace_lock);
@@ -4702,9 +4704,9 @@ int
unregister_ftrace_function_probe_func(char *glob, struct trace_array *tr,
struct ftrace_probe_ops *probe_ops)
{
+ struct ftrace_func_probe *probe = NULL, *iter;
struct ftrace_ops_hash old_hash_ops;
struct ftrace_func_entry *entry;
- struct ftrace_func_probe *probe;
struct ftrace_glob func_g;
struct ftrace_hash **orig_hash;
struct ftrace_hash *old_hash;
@@ -4732,11 +4734,13 @@ unregister_ftrace_function_probe_func(char *glob, struct trace_array *tr,
mutex_lock(&ftrace_lock);
/* Check if the probe_ops is already registered */
- list_for_each_entry(probe, &tr->func_probes, list) {
- if (probe->probe_ops == probe_ops)
+ list_for_each_entry(iter, &tr->func_probes, list) {
+ if (iter->probe_ops == probe_ops) {
+ probe = iter;
break;
+ }
}
- if (&probe->list == &tr->func_probes)
+ if (!probe)
goto err_unlock_ftrace;
ret = -EINVAL;
diff --git a/kernel/trace/trace_eprobe.c b/kernel/trace/trace_eprobe.c
index 541aa13581b9..63e901a28425 100644
--- a/kernel/trace/trace_eprobe.c
+++ b/kernel/trace/trace_eprobe.c
@@ -650,7 +650,7 @@ static struct trace_event_functions eprobe_funcs = {
static int disable_eprobe(struct trace_eprobe *ep,
struct trace_array *tr)
{
- struct event_trigger_data *trigger;
+ struct event_trigger_data *trigger = NULL, *iter;
struct trace_event_file *file;
struct eprobe_data *edata;
@@ -658,14 +658,16 @@ static int disable_eprobe(struct trace_eprobe *ep,
if (!file)
return -ENOENT;
- list_for_each_entry(trigger, &file->triggers, list) {
- if (!(trigger->flags & EVENT_TRIGGER_FL_PROBE))
+ list_for_each_entry(iter, &file->triggers, list) {
+ if (!(iter->flags & EVENT_TRIGGER_FL_PROBE))
continue;
- edata = trigger->private_data;
- if (edata->ep == ep)
+ edata = iter->private_data;
+ if (edata->ep == ep) {
+ trigger = iter;
break;
+ }
}
- if (list_entry_is_head(trigger, &file->triggers, list))
+ if (!trigger)
return -ENODEV;
list_del_rcu(&trigger->list);
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 97c7eb2f55e5..c86cbb25879d 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -2279,8 +2279,8 @@ static struct dentry *
event_subsystem_dir(struct trace_array *tr, const char *name,
struct trace_event_file *file, struct dentry *parent)
{
+ struct event_subsystem *system, *iter;
struct trace_subsystem_dir *dir;
- struct event_subsystem *system;
struct dentry *entry;
/* First see if we did not already create this dir */
@@ -2294,13 +2294,13 @@ event_subsystem_dir(struct trace_array *tr, const char *name,
}
/* Now see if the system itself exists. */
- list_for_each_entry(system, &event_subsystems, list) {
- if (strcmp(system->name, name) == 0)
+ system = NULL;
+ list_for_each_entry(iter, &event_subsystems, list) {
+ if (strcmp(iter->name, name) == 0) {
+ system = iter;
break;
+ }
}
- /* Reset system variable when not found */
- if (&system->list == &event_subsystems)
- system = NULL;
dir = kmalloc(sizeof(*dir), GFP_KERNEL);
if (!dir)
--
2.25.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [PATCH v2 1/4] tracing: Remove usage of list iterator after the loop body
2022-04-02 10:33 ` [PATCH v2 1/4] tracing: " Jakob Koschel
@ 2022-04-22 22:59 ` Steven Rostedt
0 siblings, 0 replies; 9+ messages in thread
From: Steven Rostedt @ 2022-04-22 22:59 UTC (permalink / raw)
To: Jakob Koschel
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.
On Sat, 2 Apr 2022 12:33:38 +0200
Jakob Koschel <jakobkoschel@gmail.com> wrote:
> In preparation to limit the scope of the list iterator variable to the
> traversal loop, use a dedicated pointer to point to the found element
> [1].
>
> Before, the code implicitly used the head when no element was found
> when using &pos->list. Since the new variable is only set if an
> element was found, the head needs to be used explicitly if the
> variable is NULL.
>
> Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1]
> Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
> ---
> kernel/trace/trace_output.c | 13 +++++++++----
> 1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c
> index 8aa493d25c73..733a4d6c20e2 100644
> --- a/kernel/trace/trace_output.c
> +++ b/kernel/trace/trace_output.c
> @@ -692,7 +692,7 @@ static LIST_HEAD(ftrace_event_list);
>
> static int trace_search_list(struct list_head **list)
> {
> - struct trace_event *e;
> + struct trace_event *e = NULL, *iter;
> int next = __TRACE_LAST_TYPE;
>
> if (list_empty(&ftrace_event_list)) {
> @@ -704,9 +704,11 @@ static int trace_search_list(struct list_head **list)
> * We used up all possible max events,
> * lets see if somebody freed one.
> */
> - list_for_each_entry(e, &ftrace_event_list, list) {
> - if (e->type != next)
> + list_for_each_entry(iter, &ftrace_event_list, list) {
> + if (iter->type != next) {
> + e = iter;
> break;
> + }
> next++;
> }
>
> @@ -714,7 +716,10 @@ static int trace_search_list(struct list_head **list)
> if (next > TRACE_EVENT_TYPE_MAX)
> return 0;
>
> - *list = &e->list;
> + if (e)
Technically, if e was NULL, then next would be too big. But as that's very
subtle with the algorithm, I'm fine adding this.
-- Steve
> + *list = &e->list;
> + else
> + *list = &ftrace_event_list;
> return next;
> }
>
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH v2 2/4] tracing: Remove usage of list iterator variable after the loop
2022-04-02 10:33 ` [PATCH v2 2/4] tracing: Remove usage of list iterator variable after the loop Jakob Koschel
@ 2022-04-26 14:23 ` Steven Rostedt
0 siblings, 0 replies; 9+ messages in thread
From: Steven Rostedt @ 2022-04-26 14:23 UTC (permalink / raw)
To: Jakob Koschel
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.
On Sat, 2 Apr 2022 12:33:39 +0200
Jakob Koschel <jakobkoschel@gmail.com> wrote:
This patch crashed in my testing.
> @@ -1734,14 +1734,16 @@ static int subsystem_open(struct inode *inode, struct file *filp)
> /* Make sure the system still exists */
> mutex_lock(&event_mutex);
> mutex_lock(&trace_types_lock);
> - list_for_each_entry(tr, &ftrace_trace_arrays, list) {
> - list_for_each_entry(dir, &tr->systems, list) {
> - if (dir == inode->i_private) {
> + list_for_each_entry(iter_tr, &ftrace_trace_arrays, list) {
> + list_for_each_entry(iter_dir, &iter_tr->systems, list) {
> + if (iter_dir == inode->i_private) {
> /* Don't open systems with no events */
> - if (dir->nr_events) {
> + if (iter_dir->nr_events) {
> __get_system_dir(dir);
> system = dir->subsystem;
system = NULL->subsystem
> }
> + tr = iter_tr;
> + dir = iter_dir;
But do not change that dir, move the setting above it. That is:
tr = iter_tr;
dir = iter_dir;
if (iter_dir->nr_events) {
__get_system_dir(dir);
system = dir->subsystem;
}
-- Steve
> goto exit_loop;
> }
> }
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH v2 3/4] tracing: Replace usage of found with dedicated list iterator variable
2022-04-02 10:33 ` [PATCH v2 3/4] tracing: Replace usage of found with dedicated list iterator variable Jakob Koschel
@ 2022-04-26 14:25 ` Steven Rostedt
0 siblings, 0 replies; 9+ messages in thread
From: Steven Rostedt @ 2022-04-26 14:25 UTC (permalink / raw)
To: Jakob Koschel
Cc: Ingo Molnar, linux-kernel, Mike Rapoport, Brian Johannesmeyer,
Cristiano Giuffrida, Bos, H.J.
On Sat, 2 Apr 2022 12:33:40 +0200
Jakob Koschel <jakobkoschel@gmail.com> wrote:
> To move the list iterator variable into the list_for_each_entry_*()
> macro in the future it should be avoided to use the list iterator
> variable after the loop body.
>
> To *never* use the list iterator variable after the loop it was
> concluded to use a separate iterator variable instead of a
> found boolean [1].
>
> This removes the need to use a found variable and simply checking if
> the variable was set, can determine if the break/goto was hit.
>
> Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1]
> Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
> ---
> kernel/trace/trace_events_hist.c | 17 ++++++++---------
> kernel/trace/trace_events_trigger.c | 28 +++++++++++++---------------
> 2 files changed, 21 insertions(+), 24 deletions(-)
>
> diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
> index 44db5ba9cabb..7f60d04d5b6e 100644
> --- a/kernel/trace/trace_events_hist.c
> +++ b/kernel/trace/trace_events_hist.c
> @@ -6089,32 +6089,31 @@ static void hist_unregister_trigger(char *glob,
> struct event_trigger_data *data,
> struct trace_event_file *file)
> {
> + struct event_trigger_data *test = NULL, *iter, *named_data = NULL;
> struct hist_trigger_data *hist_data = data->private_data;
> - struct event_trigger_data *test, *named_data = NULL;
> - bool unregistered = false;
>
> lockdep_assert_held(&event_mutex);
>
> if (hist_data->attrs->name)
> named_data = find_named_trigger(hist_data->attrs->name);
>
> - list_for_each_entry(test, &file->triggers, list) {
> - if (test->cmd_ops->trigger_type == ETT_EVENT_HIST) {
> - if (!hist_trigger_match(data, test, named_data, false))
> + list_for_each_entry(iter, &file->triggers, list) {
> + if (iter->cmd_ops->trigger_type == ETT_EVENT_HIST) {
> + if (!hist_trigger_match(data, iter, named_data, false))
> continue;
> - unregistered = true;
> - list_del_rcu(&test->list);
> + test = iter;
> + list_del_rcu(&iter->list);
> trace_event_trigger_enable_disable(file, 0);
> update_cond_flag(file);
> break;
> }
> }
>
> - if (unregistered && test->ops->free)
> + if (test && test->ops->free)
> test->ops->free(test->ops, test);
>
> if (hist_data->enable_timestamps) {
> - if (!hist_data->remove || unregistered)
> + if (!hist_data->remove || test)
> tracing_set_filter_buffering(file->tr, false);
> }
> }
> diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c
> index 7eb9d04f1c2e..1ba54a489416 100644
> --- a/kernel/trace/trace_events_trigger.c
> +++ b/kernel/trace/trace_events_trigger.c
> @@ -614,22 +614,21 @@ static void unregister_trigger(char *glob,
> struct event_trigger_data *test,
> struct trace_event_file *file)
> {
> - struct event_trigger_data *data;
> - bool unregistered = false;
> + struct event_trigger_data *data = NULL, *iter;
>
> lockdep_assert_held(&event_mutex);
>
> - list_for_each_entry(data, &file->triggers, list) {
> - if (data->cmd_ops->trigger_type == test->cmd_ops->trigger_type) {
> - unregistered = true;
> - list_del_rcu(&data->list);
> + list_for_each_entry(iter, &file->triggers, list) {
> + if (iter->cmd_ops->trigger_type == test->cmd_ops->trigger_type) {
> + data = iter;
> + list_del_rcu(&iter->list);
Don't use the iterator in the logic.
data = iter;
list_del_rcu(&data->list);
is just fine.
> trace_event_trigger_enable_disable(file, 0);
> update_cond_flag(file);
> break;
> }
> }
>
> - if (unregistered && data->ops->free)
> + if (data && data->ops->free)
> data->ops->free(data->ops, data);
> }
>
> @@ -1976,27 +1975,26 @@ void event_enable_unregister_trigger(char *glob,
> struct trace_event_file *file)
> {
> struct enable_trigger_data *test_enable_data = test->private_data;
> + struct event_trigger_data *data = NULL, *iter;
> struct enable_trigger_data *enable_data;
> - struct event_trigger_data *data;
> - bool unregistered = false;
>
> lockdep_assert_held(&event_mutex);
>
> - list_for_each_entry(data, &file->triggers, list) {
> - enable_data = data->private_data;
> + list_for_each_entry(iter, &file->triggers, list) {
> + enable_data = iter->private_data;
> if (enable_data &&
> - (data->cmd_ops->trigger_type ==
> + (iter->cmd_ops->trigger_type ==
> test->cmd_ops->trigger_type) &&
> (enable_data->file == test_enable_data->file)) {
> - unregistered = true;
> - list_del_rcu(&data->list);
> + data = iter;
> + list_del_rcu(&iter->list);
Same here.
And please rebase on top of Linus's lastest tree.
-- Steve
> trace_event_trigger_enable_disable(file, 0);
> update_cond_flag(file);
> break;
> }
> }
>
> - if (unregistered && data->ops->free)
> + if (data && data->ops->free)
> data->ops->free(data->ops, data);
> }
>
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH v2 0/4] Remove usage of list iterator after the loop body
@ 2022-04-27 17:07 Jakob Koschel
0 siblings, 0 replies; 9+ messages in thread
From: Jakob Koschel @ 2022-04-27 17:07 UTC (permalink / raw)
To: Steven Rostedt; +Cc: Ingo Molnar, linux-kernel, Jakob Koschel
In preparation to limit the scope of the list iterator variable to the
traversal loop, use a dedicated pointer to point to the found element
[1].
Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1]
v1->v2:
- fix NULL ptr dereference in subsystem_open() (Steven Rostedt)
- don't use the iterator in PATCH 3/4 when not necessary (Steven Rostedt)
Jakob Koschel (4):
tracing: Remove usage of list iterator after the loop body
tracing: Remove usage of list iterator variable after the loop
tracing: Replace usage of found with dedicated list iterator variable
tracing: Remove check of list iterator against head past the loop body
kernel/trace/ftrace.c | 20 ++++++++++++--------
kernel/trace/trace_eprobe.c | 14 ++++++++------
kernel/trace/trace_events.c | 27 +++++++++++++--------------
kernel/trace/trace_events_hist.c | 15 +++++++--------
kernel/trace/trace_events_trigger.c | 24 +++++++++++-------------
kernel/trace/trace_output.c | 13 +++++++++----
6 files changed, 60 insertions(+), 53 deletions(-)
base-commit: 46cf2c613f4b10eb12f749207b0fd2c1bfae3088
--
2.25.1
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2022-04-27 17:08 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-04-02 10:33 [PATCH v2 0/4] Remove usage of list iterator after the loop body Jakob Koschel
2022-04-02 10:33 ` [PATCH v2 1/4] tracing: " Jakob Koschel
2022-04-22 22:59 ` Steven Rostedt
2022-04-02 10:33 ` [PATCH v2 2/4] tracing: Remove usage of list iterator variable after the loop Jakob Koschel
2022-04-26 14:23 ` Steven Rostedt
2022-04-02 10:33 ` [PATCH v2 3/4] tracing: Replace usage of found with dedicated list iterator variable Jakob Koschel
2022-04-26 14:25 ` Steven Rostedt
2022-04-02 10:33 ` [PATCH v2 4/4] tracing: Remove check of list iterator against head past the loop body Jakob Koschel
2022-04-27 17:07 [PATCH v2 0/4] Remove usage of list iterator after " Jakob Koschel
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).