From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx120.postini.com [74.125.245.120]) by kanga.kvack.org (Postfix) with SMTP id B9D686B005D for ; Mon, 10 Sep 2012 10:44:15 -0400 (EDT) Date: Mon, 10 Sep 2012 17:44:38 +0300 From: "Kirill A. Shutemov" Subject: Re: [PATCH v2 10/10] thp: implement refcounting for huge zero page Message-ID: <20120910144438.GA31697@otc-wbsnb-06> References: <1347282813-21935-1-git-send-email-kirill.shutemov@linux.intel.com> <1347282813-21935-11-git-send-email-kirill.shutemov@linux.intel.com> <1347285759.1234.1645.camel@edumazet-glaptop> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="3MwIy2ne0vdjdPXF" Content-Disposition: inline In-Reply-To: <1347285759.1234.1645.camel@edumazet-glaptop> Sender: owner-linux-mm@kvack.org List-ID: To: Eric Dumazet Cc: Andrew Morton , Andrea Arcangeli , linux-mm@kvack.org, Andi Kleen , "H. Peter Anvin" , linux-kernel@vger.kernel.org, "Kirill A. Shutemov" --3MwIy2ne0vdjdPXF Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Sep 10, 2012 at 04:02:39PM +0200, Eric Dumazet wrote: > On Mon, 2012-09-10 at 16:13 +0300, Kirill A. Shutemov wrote: > > From: "Kirill A. Shutemov" > >=20 > > H. Peter Anvin doesn't like huge zero page which sticks in memory forev= er > > after the first allocation. Here's implementation of lockless refcounti= ng > > for huge zero page. > >=20 > ... >=20 > > +static unsigned long get_huge_zero_page(void) > > +{ > > + struct page *zero_page; > > +retry: > > + if (likely(atomic_inc_not_zero(&huge_zero_refcount))) > > + return ACCESS_ONCE(huge_zero_pfn); > > + > > + zero_page =3D alloc_pages(GFP_TRANSHUGE | __GFP_ZERO, HPAGE_PMD_ORDER= ); > > + if (!zero_page) > > + return 0; > > + if (cmpxchg(&huge_zero_pfn, 0, page_to_pfn(zero_page))) { > > + __free_page(zero_page); > > + goto retry; > > + } >=20 > This might break if preemption can happen here ? >=20 > The second thread might loop forever because huge_zero_refcount is 0, > and huge_zero_pfn not zero. I fail to see why the second thread might loop forever. Long time yes, but forever? Yes, disabling preemption before alloc_pages() and enabling after atomic_set() looks reasonable. Thanks. >=20 > If preemption already disabled, a comment would be nice. >=20 >=20 > > + > > + /* We take additional reference here. It will be put back by shinker = */ >=20 > typo : shrinker Thx. > > + atomic_set(&huge_zero_refcount, 2); > > + return ACCESS_ONCE(huge_zero_pfn); > > +} > > + >=20 >=20 >=20 --=20 Kirill A. Shutemov --3MwIy2ne0vdjdPXF Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBAgAGBQJQTfzWAAoJEAd+omnVudOMnMoQAK+JYB2wxNHp9jOcWZl8Tqhn fOB9vDh2PpJca5q7gYDlCHYt8y0JOC/nflDZS/CdaKpztlyszqrkFcAeH0jbhQEc 2rDntfM+riieZPQFvyIRxruQu65P4vooXhzTB/1PqBzOq9F2PblHLxtyVYg9N9V3 P0T/4HicTRbaJS4bwKno9Nm69LbV/Qjqspel2EiH9tpTQa+8JD7EelB/ziqzs/Hk /5SVVqzwdElHOEd2eqvUPRPvy7/SyG6q7Ar1QRu3Oq5KbmJsN6xWM9XhiUyinNGA sm6GKfEMZ3LQDZL3I71ud18GgyPiPJdl44K2j4DoRvLwouxpRtKF3RjyDMUaQKHy XDVqfiDdZtF3hix8SppCoTYntLHxLtXXduBp1y4DRt/Mg6hr1zDGort/VaOLB/iq 3WNGJSyPpC0SZ7Ev0dCa4aUIutfx8SuIK8HZYszkPJqzWJgAkrfripCrIHqKM2au 0YXXwkJvcDZ/mNhOX2Pt1dBqzfXv73rU6MCb5XxO3MczSU9lzpeEblob7HHVpSt1 /AX6u3WIIt/EUJE+8M+DyLuh2MUaAvD+g7PUap/4BPlbsiHLToHFnrwU/1+oC4h3 AP4rOKHdjJFQJjOXFS0/AHjZi4RZIC44M/NHHfPF5cbaaSF/54JBiLi/uS+hZLTh QKX9Z5uVrW41+6FkKH1h =pLyw -----END PGP SIGNATURE----- --3MwIy2ne0vdjdPXF-- -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org