linux-kernel - Re: [RFC][PATCH] memcg: avoid THP split in task migration

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 for Android: free password hash cracker in your pocket

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

Date:	Thu,  1 Mar 2012 17:01:38 -0500
From:	Naoya Horiguchi <n-horiguchi@...jp.nec.com>
To:	Naoya Horiguchi <n-horiguchi@...jp.nec.com>
Cc:	Andrea Arcangeli <aarcange@...hat.com>, linux-mm@...ck.org,
	Andrew Morton <akpm@...ux-foundation.org>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	Daisuke Nishimura <nishimura@....nes.nec.co.jp>,
	Hillf Danton <dhillf@...il.com>, linux-kernel@...r.kernel.org
Subject: Re: [RFC][PATCH] memcg: avoid THP split in task migration

On Thu, Mar 01, 2012 at 03:22:16PM -0500, Naoya Horiguchi wrote:
> > > @@ -5219,7 +5255,13 @@ static int mem_cgroup_count_precharge_pte_range(pmd_t *pmd,
> > >  	pte_t *pte;
> > >  	spinlock_t *ptl;
> > >
> > > -	split_huge_page_pmd(walk->mm, pmd);
> > > +	if (pmd_trans_huge_lock(pmd, vma) == 1) {
> > > +		if (is_target_huge_pmd_for_mc(vma, addr, *pmd, NULL))
> > > +			mc.precharge += HPAGE_PMD_NR;
> >
> > Your use of HPAGE_PMD_NR looks fine, that path will be eliminated at
> > build time if THP is off. This is the nice way to write code that is
> > already optimal for THP=off without making special cases or #ifdefs.
> >
> > Other review suggests changing HPAGE_PMD_NR as BUILD_BUG, that sounds
> > good idea too, but in this (correct) usage of HPAGE_PMD_NR it wouldn't
> > make a difference because of the whole branch is correctly eliminated
> > at build time. In short changing it to BUILD_BUG will simply make sure
> > the whole pmd_trans_huge_lock == 1 branch is eliminated at build
> > time. It looks good change too but it's orthogonal so I'd leave it for
> > a separate patch.
> 
> In my trial, without changing HPAGE_PMD_NR as BUILD_BUG a build did not
> pass with !CONFIG_TRANSPARENT_HUGEPAGE as Hillf said.
> Evaluating HPAGE_PMD_NR seems to be prior to eliminating whole
> pmd_trans_huge_lock == 1 branch, so I think this change is necessary.

I said the wrong thing.
The error I experienced was just "HPAGE_PMD_NR is undefined."
This is not related to changeing BUG() to BUILD_BUG() in already defined
HPAGE_PMD_(SHIFT|MASK|SIZE).
And using BUILD_BUG() to confirm elimination is good for me. 

Sorry for confusion.
Naoya
--
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/