Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1420677 > unrolled thread

Re: [PATCH v1 3/3] mm: per-process reclaim

Started by"Hillf Danton" <hillf.zj@alibaba-inc.com>
First post2016-06-13 12:10 +0200
Last post2016-06-15 02:50 +0200
Articles 2 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH v1 3/3] mm: per-process reclaim "Hillf Danton" <hillf.zj@alibaba-inc.com> - 2016-06-13 12:10 +0200
    Re: [PATCH v1 3/3] mm: per-process reclaim Minchan Kim <minchan@kernel.org> - 2016-06-15 02:50 +0200

#1420677 — Re: [PATCH v1 3/3] mm: per-process reclaim

From"Hillf Danton" <hillf.zj@alibaba-inc.com>
Date2016-06-13 12:10 +0200
SubjectRe: [PATCH v1 3/3] mm: per-process reclaim
Message-ID<rJwYF-4R0-7@gated-at.bofh.it>
> +static ssize_t reclaim_write(struct file *file, const char __user *buf,
> +				size_t count, loff_t *ppos)
> +{
> +	struct task_struct *task;
> +	char buffer[PROC_NUMBUF];
> +	struct mm_struct *mm;
> +	struct vm_area_struct *vma;
> +	int itype;
> +	int rv;
> +	enum reclaim_type type;
> +
> +	memset(buffer, 0, sizeof(buffer));
> +	if (count > sizeof(buffer) - 1)
> +		count = sizeof(buffer) - 1;
> +	if (copy_from_user(buffer, buf, count))
> +		return -EFAULT;
> +	rv = kstrtoint(strstrip(buffer), 10, &itype);
> +	if (rv < 0)
> +		return rv;
> +	type = (enum reclaim_type)itype;
> +	if (type < RECLAIM_FILE || type > RECLAIM_ALL)
> +		return -EINVAL;
> +
> +	task = get_proc_task(file->f_path.dentry->d_inode);
> +	if (!task)
> +		return -ESRCH;
> +
> +	mm = get_task_mm(task);
> +	if (mm) {
> +		struct mm_walk reclaim_walk = {
> +			.pmd_entry = reclaim_pte_range,
> +			.mm = mm,
> +		};
> +
> +		down_read(&mm->mmap_sem);
> +		for (vma = mm->mmap; vma; vma = vma->vm_next) {
> +			reclaim_walk.private = vma;
> +
> +			if (is_vm_hugetlb_page(vma))
> +				continue;
> +
> +			if (!vma_is_anonymous(vma) && !(type & RECLAIM_FILE))
> +				continue;
> +
> +			if (vma_is_anonymous(vma) && !(type & RECLAIM_ANON))
> +				continue;
> +
> +			walk_page_range(vma->vm_start, vma->vm_end,
> +					&reclaim_walk);

Check fatal signal after reclaiming a mapping?

> +		}
> +		flush_tlb_mm(mm);
> +		up_read(&mm->mmap_sem);
> +		mmput(mm);
> +	}
> +	put_task_struct(task);
> +
> +	return count;
> +}

[toc] | [next] | [standalone]


#1422458

FromMinchan Kim <minchan@kernel.org>
Date2016-06-15 02:50 +0200
Message-ID<rK7bP-3Jm-7@gated-at.bofh.it>
In reply to#1420677
On Mon, Jun 13, 2016 at 06:07:09PM +0800, Hillf Danton wrote:
> > +static ssize_t reclaim_write(struct file *file, const char __user *buf,
> > +				size_t count, loff_t *ppos)
> > +{
> > +	struct task_struct *task;
> > +	char buffer[PROC_NUMBUF];
> > +	struct mm_struct *mm;
> > +	struct vm_area_struct *vma;
> > +	int itype;
> > +	int rv;
> > +	enum reclaim_type type;
> > +
> > +	memset(buffer, 0, sizeof(buffer));
> > +	if (count > sizeof(buffer) - 1)
> > +		count = sizeof(buffer) - 1;
> > +	if (copy_from_user(buffer, buf, count))
> > +		return -EFAULT;
> > +	rv = kstrtoint(strstrip(buffer), 10, &itype);
> > +	if (rv < 0)
> > +		return rv;
> > +	type = (enum reclaim_type)itype;
> > +	if (type < RECLAIM_FILE || type > RECLAIM_ALL)
> > +		return -EINVAL;
> > +
> > +	task = get_proc_task(file->f_path.dentry->d_inode);
> > +	if (!task)
> > +		return -ESRCH;
> > +
> > +	mm = get_task_mm(task);
> > +	if (mm) {
> > +		struct mm_walk reclaim_walk = {
> > +			.pmd_entry = reclaim_pte_range,
> > +			.mm = mm,
> > +		};
> > +
> > +		down_read(&mm->mmap_sem);
> > +		for (vma = mm->mmap; vma; vma = vma->vm_next) {
> > +			reclaim_walk.private = vma;
> > +
> > +			if (is_vm_hugetlb_page(vma))
> > +				continue;
> > +
> > +			if (!vma_is_anonymous(vma) && !(type & RECLAIM_FILE))
> > +				continue;
> > +
> > +			if (vma_is_anonymous(vma) && !(type & RECLAIM_ANON))
> > +				continue;
> > +
> > +			walk_page_range(vma->vm_start, vma->vm_end,
> > +					&reclaim_walk);
> 
> Check fatal signal after reclaiming a mapping?

Yeb, We might need it in page_walker.

Thanks.

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web