Re: [PATCH 3/7] revoke: core code

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Fri, 2007-03-09 at 10:15 +0200, Pekka J Enberg wrote:

> +static int revoke_vma(struct vm_area_struct *vma, struct zap_details *details)
> +{
> +	unsigned long restart_addr, start_addr, end_addr;
> +	int need_break;
> +
> +	start_addr = vma->vm_start;
> +	end_addr = vma->vm_end;
> +
> +	/*
> + 	 * Not holding ->mmap_sem here.
> +	 */
> +	vma->vm_flags |= VM_REVOKED;
> +	smp_mb();

Hmm, i_mmap_lock pins the vma and excludes modifications, but doesn't
exclude concurrent faults.

I guess its save.

> +  again:
> +	restart_addr = zap_page_range(vma, start_addr, end_addr - start_addr,
> +				      details);
> +
> +	need_break = need_resched() || need_lockbreak(details->i_mmap_lock);
> +	if (need_break)
> +		goto out_need_break;
> +
> +	if (restart_addr < end_addr) {
> +		start_addr = restart_addr;
> +		goto again;
> +	}
> +	return 0;
> +
> +  out_need_break:
> +	spin_unlock(details->i_mmap_lock);
> +	cond_resched();
> +	spin_lock(details->i_mmap_lock);
> +	return -EINTR;

I'm not sure this scheme works, given a sufficiently loaded machine,
this might never complete.

> +}
> +
> +static int revoke_mapping(struct address_space *mapping, struct file *to_exclude)
> +{
> +	struct vm_area_struct *vma;
> +	struct prio_tree_iter iter;
> +	struct zap_details details;
> +	int err = 0;
> +
> +	details.i_mmap_lock = &mapping->i_mmap_lock;
> +
> +	spin_lock(&mapping->i_mmap_lock);
> +	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, 0, ULONG_MAX) {
> +		if (vma->vm_flags & VM_SHARED && vma->vm_file != to_exclude) {

I'm never sure of operator precedence and prefer:

 (vma->vm_flags & VM_SHARED) && ...

which leaves no room for error.

> +			err = revoke_vma(vma, &details);
> +			if (err)
> +				goto out;
> +		}
> +	}
> +
> +	list_for_each_entry(vma, &mapping->i_mmap_nonlinear, shared.vm_set.list) {
> +		if (vma->vm_flags & VM_SHARED && vma->vm_file != to_exclude) {

Idem.

> +			err = revoke_vma(vma, &details);
> +			if (err)
> +				goto out;
> +		}
> +	}
> +  out:
> +	spin_unlock(&mapping->i_mmap_lock);
> +	return err;
> +}