diff mbox series

[v6,7/9] tracing: Replace strncpy() with strscpy()

Message ID 20240812022933.69850-8-laoar.shao@gmail.com (mailing list archive)
State New
Headers show
Series Improve the copy of task comm | expand

Commit Message

Yafang Shao Aug. 12, 2024, 2:29 a.m. UTC
Using strscpy() to read the task comm ensures that the name is
always NUL-terminated, regardless of the source string. This approach also
facilitates future extensions to the task comm.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
---
 kernel/trace/trace.c             | 2 +-
 kernel/trace/trace_events_hist.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

Comments

Justin Stitt Aug. 13, 2024, 10:19 p.m. UTC | #1
Hi,

On Mon, Aug 12, 2024 at 10:29:31AM GMT, Yafang Shao wrote:
> Using strscpy() to read the task comm ensures that the name is
> always NUL-terminated, regardless of the source string. This approach also
> facilitates future extensions to the task comm.

Thanks for sending patches replacing str{n}cpy's!

I believe there's at least two more instances of strncpy in trace.c as
well as in trace_events_hist.c (for a grand total of 6 instances in the
files you've touched in this specific patch).

It'd be great if you could replace those instances in this patch as well :>)

This would help greatly with [1].

> 
> Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
> Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
> ---
>  kernel/trace/trace.c             | 2 +-
>  kernel/trace/trace_events_hist.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
> index 578a49ff5c32..1b2577f9d734 100644
> --- a/kernel/trace/trace.c
> +++ b/kernel/trace/trace.c
> @@ -1907,7 +1907,7 @@ __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
>  	max_data->critical_start = data->critical_start;
>  	max_data->critical_end = data->critical_end;
>  
> -	strncpy(max_data->comm, tsk->comm, TASK_COMM_LEN);
> +	strscpy(max_data->comm, tsk->comm, TASK_COMM_LEN);

If max_data->comm wants to be NUL-terminated then this is the right
replacement. Without knowing how the trace stack works at all, it's hard
for me to tell if that is the case.

There's a length-supplied format specifier for which this comm field is
used with; Either this is just another safeguard against spilling over
the buffer or this field really doesn't care about NUL-termination.
| seq_printf(m, "#    | task: %.16s-%d "
|       "(uid:%d nice:%ld policy:%ld rt_prio:%ld)\n",
|       data->comm, data->pid,

In the event this field doesn't need to be NUL-terminated then we are
introducing an off-by-one error where we are copying one less useful
byte with strscpy -- Linus pointed out earlier [2] that these things all
just want to be c-strings so this is probably the right change :>)

>  	max_data->pid = tsk->pid;
>  	/*
>  	 * If tsk == current, then use current_uid(), as that does not use
> diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
> index 6ece1308d36a..4cd24c25ce05 100644
> --- a/kernel/trace/trace_events_hist.c
> +++ b/kernel/trace/trace_events_hist.c
> @@ -1599,7 +1599,7 @@ static inline void save_comm(char *comm, struct task_struct *task)
>  		return;
>  	}
>  
> -	strncpy(comm, task->comm, TASK_COMM_LEN);
> +	strscpy(comm, task->comm, TASK_COMM_LEN);
>  }
>  
>  static void hist_elt_data_free(struct hist_elt_data *elt_data)
> -- 
> 2.43.5
> 

Link: https://github.com/KSPP/linux/issues/90 [1]
Link: https://lore.kernel.org/all/CAHk-=whWtUC-AjmGJveAETKOMeMFSTwKwu99v7+b6AyHMmaDFA@mail.gmail.com/ [2]

Thanks
Justin
Justin Stitt Aug. 13, 2024, 10:31 p.m. UTC | #2
On Tue, Aug 13, 2024 at 3:19 PM Justin Stitt <justinstitt@google.com> wrote:
>
> Hi,
>
> On Mon, Aug 12, 2024 at 10:29:31AM GMT, Yafang Shao wrote:
> > Using strscpy() to read the task comm ensures that the name is
> > always NUL-terminated, regardless of the source string. This approach also
> > facilitates future extensions to the task comm.
>
> Thanks for sending patches replacing str{n}cpy's!
>
> I believe there's at least two more instances of strncpy in trace.c as
> well as in trace_events_hist.c (for a grand total of 6 instances in the
> files you've touched in this specific patch).
>
> It'd be great if you could replace those instances in this patch as well :>)
>
> This would help greatly with [1].
>

I just saw that Jinjie Ruan sent replacements for these strncpy's too
and tracked down and replaced an instance of strscpy() that was
present in trace.c but was moved to trace_sched_switch.c during a
refactor.

They even used the new 2-argument strscpy which is pretty neat.

See their patch here:
https://lore.kernel.org/all/20240731075058.617588-1-ruanjinjie@huawei.com/

> Link: https://github.com/KSPP/linux/issues/90 [1]
> Link: https://lore.kernel.org/all/CAHk-=whWtUC-AjmGJveAETKOMeMFSTwKwu99v7+b6AyHMmaDFA@mail.gmail.com/ [2]
>
> Thanks
> Justin
Yafang Shao Aug. 14, 2024, 2:34 a.m. UTC | #3
On Wed, Aug 14, 2024 at 6:31 AM Justin Stitt <justinstitt@google.com> wrote:
>
> On Tue, Aug 13, 2024 at 3:19 PM Justin Stitt <justinstitt@google.com> wrote:
> >
> > Hi,
> >
> > On Mon, Aug 12, 2024 at 10:29:31AM GMT, Yafang Shao wrote:
> > > Using strscpy() to read the task comm ensures that the name is
> > > always NUL-terminated, regardless of the source string. This approach also
> > > facilitates future extensions to the task comm.
> >
> > Thanks for sending patches replacing str{n}cpy's!
> >
> > I believe there's at least two more instances of strncpy in trace.c as
> > well as in trace_events_hist.c (for a grand total of 6 instances in the
> > files you've touched in this specific patch).
> >
> > It'd be great if you could replace those instances in this patch as well :>)
> >
> > This would help greatly with [1].
> >
>
> I just saw that Jinjie Ruan sent replacements for these strncpy's too
> and tracked down and replaced an instance of strscpy() that was
> present in trace.c but was moved to trace_sched_switch.c during a
> refactor.
>
> They even used the new 2-argument strscpy which is pretty neat.
>
> See their patch here:
> https://lore.kernel.org/all/20240731075058.617588-1-ruanjinjie@huawei.com/

+ Jinjie

That sounds good. Since this change can be handled as a separate
patch, I will drop it from the next version and leave it to Jinjie.
Please note that Steven might have a better solution for handling
task->comm in trace events, so it’s probably best to leave any changes
related to trace events to him [0].

[0] https://lore.kernel.org/all/20240603184016.3374559f@gandalf.local.home/#t

--
Regards
Yafang
diff mbox series

Patch

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 578a49ff5c32..1b2577f9d734 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -1907,7 +1907,7 @@  __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
 	max_data->critical_start = data->critical_start;
 	max_data->critical_end = data->critical_end;
 
-	strncpy(max_data->comm, tsk->comm, TASK_COMM_LEN);
+	strscpy(max_data->comm, tsk->comm, TASK_COMM_LEN);
 	max_data->pid = tsk->pid;
 	/*
 	 * If tsk == current, then use current_uid(), as that does not use
diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index 6ece1308d36a..4cd24c25ce05 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -1599,7 +1599,7 @@  static inline void save_comm(char *comm, struct task_struct *task)
 		return;
 	}
 
-	strncpy(comm, task->comm, TASK_COMM_LEN);
+	strscpy(comm, task->comm, TASK_COMM_LEN);
 }
 
 static void hist_elt_data_free(struct hist_elt_data *elt_data)