All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stefani Seibold <stefani@seibold.net>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [patch] procfs: provide stack information for threads
Date: Fri, 05 Jun 2009 21:12:56 +0200	[thread overview]
Message-ID: <1244229176.31924.13.camel@wall-e> (raw)
In-Reply-To: <20090604142358.95af81d5.akpm@linux-foundation.org>


> > +static inline unsigned long get_stack_usage_in_bytes(struct vm_area_struct *vma,
> > +					struct task_struct *p)
> > +{
> > +	unsigned long	i;
> > +	struct page	*page;
> > +	unsigned long	stkpage;
> > +
> > +	stkpage = KSTK_ESP(p) & PAGE_MASK;
> > +
> > +#ifdef CONFIG_STACK_GROWSUP
> > +	for (i = vma->vm_end; i-PAGE_SIZE > stkpage; i -= PAGE_SIZE) {
> > +
> > +		page = follow_page(vma, i-PAGE_SIZE, 0);
> > +
> > +		if (!IS_ERR(page) && page)
> > +			break;
> > +	}
> > +	return i - (p->stack_start & PAGE_MASK);
> > +#else
> > +	for (i = vma->vm_start; i+PAGE_SIZE <= stkpage; i += PAGE_SIZE) {
> > +
> > +		page = follow_page(vma, i, 0);
> > +
> > +		if (!IS_ERR(page) && page)
> > +			break;
> > +	}
> > +	return (p->stack_start & PAGE_MASK) - i + PAGE_SIZE;
> > +#endif
> > +}
> > +
> > +static inline void task_show_stack_usage(struct seq_file *m,
> > +						struct task_struct *p)
> > +{
> > +	struct vm_area_struct	*vma;
> > +	struct mm_struct	*mm;
> > +
> > +	mm = get_task_mm(p);
> > +
> > +	if (mm) {
> > +		vma = find_vma(mm, p->stack_start);
> > +
> > +		if (vma)
> > +			seq_printf(m, "Stack usage:\t%lu kB\n",
> > +				get_stack_usage_in_bytes(vma, p) >> 10);
> > +
> > +		mmput(mm);
> > +	}
> > +}
> 
> Both follow_page() and find_vma() require locking: down_read(mmap_sem)
> or down_write(mmap_sem).  down_read() is appropriate here.
> 

Didn't know that this require a lock, but i will fix this.

> > +				} else {
> > +					unsigned long stack_start;
> > +					struct proc_maps_private *pmp;
> > +
> > +					pmp = m->private;
> > +					stack_start = pmp->task->stack_start;
> > +
> > +					if (vma->vm_start <= stack_start &&
> > +					    vma->vm_end >= stack_start) {
> > +						pad_len_spaces(m, len);
> > +						seq_printf(m,
> > +						 "[thread stack: %08lx]",
> > +						 stack_start);
> > +					}
> 
> You're making changes to the user interface but it's not terribly clear
> what they look like.  Please include sample output from the affected
> procfs files so that we can review the proposed changes.
> 

This change to the user interface was suggest by ingo molnar. I will add
the sample output in the next version.

> Please update the userspace documentation to reflect the changes. 
> Documentation/filesystems/proc.txt documents VmStk, so probably that is
> the appropriate place.
> 

Good idea, but the proc.txt documentation about /proc/<pid>/status (line
140) and the contents of the /proc/<pid>/stat file (line 192) is
complete outdated. And there is also no documentation
about /proc/<pid>/*maps. 

So i want to try to fix the missing documentation in proc.txt in a
separate patch. Will this okay by you?

Greetings



WARNING: multiple messages have this Message-ID (diff)
From: Stefani Seibold <stefani@seibold.net>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [patch] procfs: provide stack information for threads
Date: Fri, 05 Jun 2009 21:12:56 +0200	[thread overview]
Message-ID: <1244229176.31924.13.camel@wall-e> (raw)
In-Reply-To: <20090604142358.95af81d5.akpm@linux-foundation.org>


> > +static inline unsigned long get_stack_usage_in_bytes(struct vm_area_struct *vma,
> > +					struct task_struct *p)
> > +{
> > +	unsigned long	i;
> > +	struct page	*page;
> > +	unsigned long	stkpage;
> > +
> > +	stkpage = KSTK_ESP(p) & PAGE_MASK;
> > +
> > +#ifdef CONFIG_STACK_GROWSUP
> > +	for (i = vma->vm_end; i-PAGE_SIZE > stkpage; i -= PAGE_SIZE) {
> > +
> > +		page = follow_page(vma, i-PAGE_SIZE, 0);
> > +
> > +		if (!IS_ERR(page) && page)
> > +			break;
> > +	}
> > +	return i - (p->stack_start & PAGE_MASK);
> > +#else
> > +	for (i = vma->vm_start; i+PAGE_SIZE <= stkpage; i += PAGE_SIZE) {
> > +
> > +		page = follow_page(vma, i, 0);
> > +
> > +		if (!IS_ERR(page) && page)
> > +			break;
> > +	}
> > +	return (p->stack_start & PAGE_MASK) - i + PAGE_SIZE;
> > +#endif
> > +}
> > +
> > +static inline void task_show_stack_usage(struct seq_file *m,
> > +						struct task_struct *p)
> > +{
> > +	struct vm_area_struct	*vma;
> > +	struct mm_struct	*mm;
> > +
> > +	mm = get_task_mm(p);
> > +
> > +	if (mm) {
> > +		vma = find_vma(mm, p->stack_start);
> > +
> > +		if (vma)
> > +			seq_printf(m, "Stack usage:\t%lu kB\n",
> > +				get_stack_usage_in_bytes(vma, p) >> 10);
> > +
> > +		mmput(mm);
> > +	}
> > +}
> 
> Both follow_page() and find_vma() require locking: down_read(mmap_sem)
> or down_write(mmap_sem).  down_read() is appropriate here.
> 

Didn't know that this require a lock, but i will fix this.

> > +				} else {
> > +					unsigned long stack_start;
> > +					struct proc_maps_private *pmp;
> > +
> > +					pmp = m->private;
> > +					stack_start = pmp->task->stack_start;
> > +
> > +					if (vma->vm_start <= stack_start &&
> > +					    vma->vm_end >= stack_start) {
> > +						pad_len_spaces(m, len);
> > +						seq_printf(m,
> > +						 "[thread stack: %08lx]",
> > +						 stack_start);
> > +					}
> 
> You're making changes to the user interface but it's not terribly clear
> what they look like.  Please include sample output from the affected
> procfs files so that we can review the proposed changes.
> 

This change to the user interface was suggest by ingo molnar. I will add
the sample output in the next version.

> Please update the userspace documentation to reflect the changes. 
> Documentation/filesystems/proc.txt documents VmStk, so probably that is
> the appropriate place.
> 

Good idea, but the proc.txt documentation about /proc/<pid>/status (line
140) and the contents of the /proc/<pid>/stat file (line 192) is
complete outdated. And there is also no documentation
about /proc/<pid>/*maps. 

So i want to try to fix the missing documentation in proc.txt in a
separate patch. Will this okay by you?

Greetings


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2009-06-05 19:13 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-03-31 14:58 Detailed Stack Information Patch [1/3] Stefani Seibold
2009-03-31 14:58 ` Stefani Seibold
2009-04-01 19:31 ` Ingo Molnar
2009-04-01 19:31   ` Ingo Molnar
2009-04-02 21:26   ` Stefani Seibold
2009-04-02 21:26     ` Stefani Seibold
2009-06-03 20:34   ` Detailed Stack Information Patch Next Generation Stefani Seibold
2009-06-03 20:34     ` Stefani Seibold
2009-06-03 21:06     ` Andrew Morton
2009-06-03 21:06       ` Andrew Morton
2009-06-04 11:23   ` [patch] procfs: provide stack information for threads Stefani Seibold
2009-06-04 11:23     ` Stefani Seibold
2009-06-04 11:37     ` Andrew Morton
2009-06-04 11:37       ` Andrew Morton
2009-06-04 11:56       ` Stefani Seibold
2009-06-04 11:56         ` Stefani Seibold
2009-06-04 17:57         ` Andrew Morton
2009-06-04 17:57           ` Andrew Morton
2009-06-04 20:21   ` Stefani Seibold
2009-06-04 20:21     ` Stefani Seibold
2009-06-04 21:23     ` Andrew Morton
2009-06-04 21:23       ` Andrew Morton
2009-06-05 19:12       ` Stefani Seibold [this message]
2009-06-05 19:12         ` Stefani Seibold
2009-06-05 19:19         ` Andrew Morton
2009-06-05 19:19           ` Andrew Morton
2009-10-02 21:17     ` Andreas Schwab
2009-10-02 21:17       ` Andreas Schwab
2009-10-02 21:44       ` Andreas Schwab
2009-10-02 21:44         ` Andreas Schwab
2009-10-03  6:47         ` Andreas Schwab
2009-10-03  6:47           ` Andreas Schwab
2009-10-03  7:40           ` Stefani Seibold
2009-10-03  7:40             ` Stefani Seibold
2009-10-03 11:33           ` Stefani Seibold
2009-10-03 11:33             ` Stefani Seibold
2009-06-06 10:01   ` [patch] procfs: provide stack information for threads V0.6 Stefani Seibold
2009-06-06 10:01     ` Stefani Seibold
2009-06-09 10:35   ` [patch] proc.txt: Update kernel filesystem/proc.txt documentation Stefani Seibold
2009-06-09 10:35     ` Stefani Seibold
2009-06-09 19:36     ` Andrew Morton
2009-06-09 19:36       ` Andrew Morton
2009-06-09 20:53       ` Stefani Seibold
2009-06-09 20:53         ` Stefani Seibold
2009-06-09 21:13         ` Andrew Morton
2009-06-09 21:13           ` Andrew Morton
2009-06-10  6:46   ` [patch 1/2] " Stefani Seibold
2009-06-10  6:46     ` Stefani Seibold
2009-06-10  6:46   ` [patch 2/2] procfs: provide stack information for threads V0.7 Stefani Seibold
2009-06-10  6:46     ` Stefani Seibold
2009-06-10  7:20   ` [patch 2/2] procfs: provide stack information for threads V0.8 Stefani Seibold
2009-06-10  7:20     ` Stefani Seibold
2009-06-15 22:01     ` Andrew Morton
2009-06-15 22:01       ` Andrew Morton
2009-06-16  7:14       ` Stefani Seibold
2009-06-16  7:14         ` Stefani Seibold

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=1244229176.31924.13.camel@wall-e \
    --to=stefani@seibold.net \
    --cc=akpm@linux-foundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.