linux-kernel - Re: [patch][v2] swap: virtual swap readahead

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20090603132751.GA1813@cmpxchg.org>
Date:	Wed, 3 Jun 2009 15:27:52 +0200
From:	Johannes Weiner <hannes@...xchg.org>
To:	Andi Kleen <andi@...stfloor.org>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Rik van Riel <riel@...hat.com>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Hugh Dickins <hugh.dickins@...cali.co.uk>, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org
Subject: Re: [patch][v2] swap: virtual swap readahead

On Wed, Jun 03, 2009 at 01:34:57AM +0200, Andi Kleen wrote:
> On Wed, Jun 03, 2009 at 12:37:39AM +0200, Johannes Weiner wrote:
> > + *
> > + * Caller must hold down_read on the vma->vm_mm if vma is not NULL.
> > + */
> > +struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask,
> > +			struct vm_area_struct *vma, unsigned long addr)
> > +{
> > +	unsigned long start, pos, end;
> > +	unsigned long pmin, pmax;
> > +	int cluster, window;
> > +
> > +	if (!vma || !vma->vm_mm)	/* XXX: shmem case */
> > +		return swapin_readahead_phys(entry, gfp_mask, vma, addr);
> > +
> > +	cluster = 1 << page_cluster;
> > +	window = cluster << PAGE_SHIFT;
> > +
> > +	/* Physical range to read from */
> > +	pmin = swp_offset(entry) & ~(cluster - 1);
> 
> Is cluster really properly sign extended on 64bit? Looks a little
> dubious. long from the start would be safer

Fixed.

> > +	/* Virtual range to read from */
> > +	start = addr & ~(window - 1);
> 
> Same.

Fixed.

> > +		pgd = pgd_offset(vma->vm_mm, pos);
> > +		if (!pgd_present(*pgd))
> > +			continue;
> > +		pud = pud_offset(pgd, pos);
> > +		if (!pud_present(*pud))
> > +			continue;
> > +		pmd = pmd_offset(pud, pos);
> > +		if (!pmd_present(*pmd))
> > +			continue;
> > +		pte = pte_offset_map_lock(vma->vm_mm, pmd, pos, &ptl);
> 
> You could be more efficient here by using the standard mm/* nested loop
> pattern that avoids relookup of everything in each iteration. I suppose
> it would mainly make a difference with 32bit highpte where mapping a pte
> can be somewhat costly. And you would take less locks this way.

I ran into weird problems here.  The above version is actually faster
in the benchmarks than writing a nested level walker or using
walk_page_range().  Still digging but it can take some time.  Busy
week :(

> > +		page = read_swap_cache_async(swp, gfp_mask, vma, pos);
> > +		if (!page)
> > +			continue;
> 
> That's out of memory, break would be better here because prefetch
> while oom is usually harmful.

It can also happen due to a race with something releasing the swap
slot (i.e. swap_duplicate() fails).  But the old version did a break
too and this patch shouldn't do it differently.  Fixed.

> > +		page_cache_release(page);
> > +	}
> > +	lru_add_drain();	/* Push any new pages onto the LRU now */
> > +	return read_swap_cache_async(entry, gfp_mask, vma, addr);
> 
> Shouldn't that page be already handled in the loop earlier? Why doing that
> again? It would be better to remember it from there.

When doing the nested page table level walker, communicating even more
state back and forth gets pretty ugly.  I see what I can do.

Thanks for your input Andi,

	Hannes
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/