linux-kernel - Re: [RFC] mm: a question about high-order check in __zone_watermark

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]

Date:   Wed, 28 Sep 2016 14:52:03 +0900
From:   Joonsoo Kim <iamjoonsoo.kim@....com>
To:     Michal Hocko <mhocko@...nel.org>
Cc:     Xishi Qiu <qiuxishi@...wei.com>,
        Mel Gorman <mgorman@...hsingularity.net>,
        Johannes Weiner <hannes@...xchg.org>,
        Vlastimil Babka <vbabka@...e.cz>,
        LKML <linux-kernel@...r.kernel.org>,
        Linux MM <linux-mm@...ck.org>,
        Yisheng Xie <xieyisheng1@...wei.com>
Subject: Re: [RFC] mm: a question about high-order check in
 __zone_watermark_ok()

On Mon, Sep 26, 2016 at 01:02:31PM +0200, Michal Hocko wrote:
> On Mon 26-09-16 18:17:50, Xishi Qiu wrote:
> > On 2016/9/26 17:43, Michal Hocko wrote:
> > 
> > > On Mon 26-09-16 17:16:54, Xishi Qiu wrote:
> > >> On 2016/9/26 16:58, Michal Hocko wrote:
> > >>
> > >>> On Mon 26-09-16 16:47:57, Xishi Qiu wrote:
> > >>>> commit 97a16fc82a7c5b0cfce95c05dfb9561e306ca1b1
> > >>>> (mm, page_alloc: only enforce watermarks for order-0 allocations)
> > >>>> rewrite the high-order check in __zone_watermark_ok(), but I think it
> > >>>> quietly fix a bug. Please see the following.
> > >>>>
> > >>>> Before this patch, the high-order check is this:
> > >>>> __zone_watermark_ok()
> > >>>> 	...
> > >>>> 	for (o = 0; o < order; o++) {
> > >>>> 		/* At the next order, this order's pages become unavailable */
> > >>>> 		free_pages -= z->free_area[o].nr_free << o;
> > >>>>
> > >>>> 		/* Require fewer higher order pages to be free */
> > >>>> 		min >>= 1;
> > >>>>
> > >>>> 		if (free_pages <= min)
> > >>>> 			return false;
> > >>>> 	}
> > >>>> 	...
> > >>>>
> > >>>> If we have cma memory, and we alloc a high-order movable page, then it's right.
> > >>>>
> > >>>> But if we alloc a high-order unmovable page(e.g. alloc kernel stack in dup_task_struct()),
> > >>>> and there are a lot of high-order cma pages, but little high-order unmovable
> > >>>> pages, the it is still return *true*, but we will alloc *failed* finally, because
> > >>>> we cannot fallback from migrate_unmovable to migrate_cma, right?
> > >>>
> > >>> AFAIR CMA wmark check was always tricky and the above commit has made
> > >>> the situation at least a bit more clear. Anyway IIRC 
> > >>>
> > >>> #ifdef CONFIG_CMA
> > >>> 	/* If allocation can't use CMA areas don't use free CMA pages */
> > >>> 	if (!(alloc_flags & ALLOC_CMA))
> > >>> 		free_cma = zone_page_state(z, NR_FREE_CMA_PAGES);
> > >>> #endif
> > >>>
> > >>> 	if (free_pages - free_cma <= min + z->lowmem_reserve[classzone_idx])
> > >>> 		return false;
> > >>>
> > >>> should reduce the prioblem because a lot of CMA pages should just get us
> > >>> below the wmark + reserve boundary.
> > >>
> > >> Hi Michal,
> > >>
> > >> If we have many high-order cma pages, and the left pages (unmovable/movable/reclaimable)
> > >> are also enough, but they are fragment, then it will triger the problem.
> > >> If we alloc a high-order unmovable page, water mark check return *true*, but we
> > >> will alloc *failed*, right?
> > > 
> > > As Vlastimil has written. There were known issues with the wmark checks
> > > and high order requests.
> > 
> > Shall we backport to stable?
> 
> I dunno, it was a part of a larger series with high atomic reserves and
> changes which sound a bit intrusive for the stable kernel. Considering
> that CMA was known to be problematic and there are still some issues
> left I do not think this is worth the trouble/risk.

CMA problem is known one. I mentioned it on my ZONE_CMA series v1 but
removed due to Mel's high atomic reserve series.

That series is rather large and has some problems so I think that it
is not suitable for stable tree.

Thanks.