Linux Trace Kernel
 help / color / mirror / Atom feed
From: Ryan Roberts <ryan.roberts@arm.com>
To: Steven Rostedt <rostedt@goodmis.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Linux trace kernel <linux-trace-kernel@vger.kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Subject: Re: [PATCH] fgraph: Give ret_stack its own kmem cache
Date: Mon, 21 Oct 2024 08:58:11 +0100	[thread overview]
Message-ID: <bf425884-b355-4da9-8e2b-6da693f2760b@arm.com> (raw)
In-Reply-To: <20241019152719.321772eb@rorschach.local.home>

On 19/10/2024 20:27, Steven Rostedt wrote:
> From: Steven Rostedt <rostedt@goodmis.org>
> 
> The ret_stack (shadow stack used by function graph infrastructure) is
> created for every task on the system when function graph is enabled. Give
> it its own kmem_cache. This will make it easier to see how much memory is
> being used specifically for function graph shadow stacks.
> 
> In the future, this size may change and may not be a power of two. Having
> its own cache can also keep it from fragmenting memory.
> 
> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
> ---
>  kernel/trace/fgraph.c | 35 ++++++++++++++++++++++++++++++-----
>  1 file changed, 30 insertions(+), 5 deletions(-)
> 
> diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c
> index 51e81b299a0d..51849bc42d3a 100644
> --- a/kernel/trace/fgraph.c
> +++ b/kernel/trace/fgraph.c
> @@ -172,6 +172,8 @@ enum {
>  DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph);
>  int ftrace_graph_active;
>  
> +static struct kmem_cache *fgraph_stack_cachep;
> +
>  static struct fgraph_ops *fgraph_array[FGRAPH_ARRAY_SIZE];
>  static unsigned long fgraph_array_bitmask;
>  
> @@ -1022,8 +1024,11 @@ static int alloc_retstack_tasklist(unsigned long **ret_stack_list)
>  	int start = 0, end = FTRACE_RETSTACK_ALLOC_SIZE;
>  	struct task_struct *g, *t;
>  
> +	if (WARN_ON_ONCE(!fgraph_stack_cachep))
> +		return -ENOMEM;
> +
>  	for (i = 0; i < FTRACE_RETSTACK_ALLOC_SIZE; i++) {
> -		ret_stack_list[i] = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL);
> +		ret_stack_list[i] = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL);
>  		if (!ret_stack_list[i]) {
>  			start = 0;
>  			end = i;
> @@ -1054,7 +1059,7 @@ static int alloc_retstack_tasklist(unsigned long **ret_stack_list)
>  	rcu_read_unlock();
>  free:
>  	for (i = start; i < end; i++)
> -		kfree(ret_stack_list[i]);
> +		kmem_cache_free(fgraph_stack_cachep, ret_stack_list[i]);
>  	return ret;
>  }
>  
> @@ -1117,9 +1122,12 @@ void ftrace_graph_init_idle_task(struct task_struct *t, int cpu)
>  	if (ftrace_graph_active) {
>  		unsigned long *ret_stack;
>  
> +		if (WARN_ON_ONCE(!fgraph_stack_cachep))
> +			return;
> +
>  		ret_stack = per_cpu(idle_ret_stack, cpu);
>  		if (!ret_stack) {
> -			ret_stack = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL);
> +			ret_stack = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL);
>  			if (!ret_stack)
>  				return;
>  			per_cpu(idle_ret_stack, cpu) = ret_stack;
> @@ -1139,7 +1147,10 @@ void ftrace_graph_init_task(struct task_struct *t)
>  	if (ftrace_graph_active) {
>  		unsigned long *ret_stack;
>  
> -		ret_stack = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL);
> +		if (WARN_ON_ONCE(!fgraph_stack_cachep))
> +			return;
> +
> +		ret_stack = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL);
>  		if (!ret_stack)
>  			return;
>  		graph_init_task(t, ret_stack);
> @@ -1154,7 +1165,11 @@ void ftrace_graph_exit_task(struct task_struct *t)
>  	/* NULL must become visible to IRQs before we free it: */
>  	barrier();
>  
> -	kfree(ret_stack);
> +	if (ret_stack) {
> +		if (WARN_ON_ONCE(!fgraph_stack_cachep))
> +			return;
> +		kmem_cache_free(fgraph_stack_cachep, ret_stack);
> +	}
>  }
>  
>  #ifdef CONFIG_DYNAMIC_FTRACE
> @@ -1290,6 +1305,16 @@ int register_ftrace_graph(struct fgraph_ops *gops)
>  
>  	mutex_lock(&ftrace_lock);
>  
> +	if (!fgraph_stack_cachep) {
> +		fgraph_stack_cachep = kmem_cache_create("fgraph_stack",
> +							SHADOW_STACK_SIZE,
> +							SHADOW_STACK_SIZE, 0, NULL);

(I don't have any experience with this code, but...) is there any value/need to
destroy the cache in unregister_ftrace_graph()? I guess you would need to
refcount it, so its created on the first call to register and destroyed on the
last call to unregister?

Thanks,
Ryan

> +		if (!fgraph_stack_cachep) {
> +			ret = -ENOMEM;
> +			goto out;
> +		}
> +	}
> +
>  	if (!fgraph_array[0]) {
>  		/* The array must always have real data on it */
>  		for (i = 0; i < FGRAPH_ARRAY_SIZE; i++)


  reply	other threads:[~2024-10-21  7:58 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-19 19:27 [PATCH] fgraph: Give ret_stack its own kmem cache Steven Rostedt
2024-10-21  7:58 ` Ryan Roberts [this message]
2024-10-21  8:37   ` Steven Rostedt
2024-10-21  9:29     ` Ryan Roberts

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=bf425884-b355-4da9-8e2b-6da693f2760b@arm.com \
    --to=ryan.roberts@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mhiramat@kernel.org \
    --cc=rostedt@goodmis.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox