From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pa0-f71.google.com (mail-pa0-f71.google.com [209.85.220.71]) by kanga.kvack.org (Postfix) with ESMTP id 4497D6B0005 for ; Tue, 7 Jun 2016 02:43:26 -0400 (EDT) Received: by mail-pa0-f71.google.com with SMTP id fg1so234161181pad.1 for ; Mon, 06 Jun 2016 23:43:26 -0700 (PDT) Received: from prv-mh.provo.novell.com (prv-mh.provo.novell.com. [137.65.248.74]) by mx.google.com with ESMTPS id or6si3691548pac.233.2016.06.06.23.43.24 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 06 Jun 2016 23:43:25 -0700 (PDT) Message-Id: <5756892902000078000F26C3@prv-mh.provo.novell.com> Date: Tue, 07 Jun 2016 00:43:21 -0600 From: "Jan Beulich" Subject: Re: [Bug 119641] New: hugetlbfs: disabling because there are no supported hugepage sizes References: <20160606140123.bbc4b06d0f9d8b974f7b323f@linux-foundation.org> In-Reply-To: <20160606140123.bbc4b06d0f9d8b974f7b323f@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Sender: owner-linux-mm@kvack.org List-ID: To: jp.pozzi@izzop.net, Andrew Morton Cc: bugzilla-daemon@bugzilla.kernel.org, Ingo Molnar , linux-mm@kvack.org >>> On 06.06.16 at 23:01, wrote: > Does anyone have any theories about this? I went through the > 4.5.2->4.5.5 changelog searching for "huget" but came up blank.. Well, the original report (still quoted below) doesn't say whether that's a PV or HVM guest under Xen (not that from the .config one cannot tell). In the former case, observed behavior is the intended effect of commit 103f6112f2: There simply are no huge pages available in that environment. In the latter case I can't see where the problem would be coming from. Jan > I'm suspiciously staring at Ingo's change >=20 > commit b2eafe890d4a09bfa63ab31ff018d7d6bb8cfefc > Merge: abfb949 ea5dfb5 > Author: Ingo Molnar > AuthorDate: Fri Apr 22 10:12:19 2016 +0200 > Commit: Ingo Molnar > CommitDate: Fri Apr 22 10:13:53 2016 +0200 >=20 > Merge branch 'x86/urgent' into x86/asm, to fix semantic conflict > =20 > 'cpu_has_pse' has changed to boot_cpu_has(X86_FEATURE_PSE), fix this > up in the merge commit when merging the x86/urgent tree that = includes > the following commit: > =20 > 103f6112f253 ("x86/mm/xen: Suppress hugetlbfs in PV guests") > =20 > Signed-off-by: Ingo Molnar >=20 > --- a/arch/x86/include/asm/hugetlb.h > +++ b/arch/x86/include/asm/hugetlb.h > @@@ -4,6 -4,7 +4,7 @@@ > #include > #include > =20 > -#define hugepages_supported() cpu_has_pse > ++#define hugepages_supported() boot_cpu_has(X86_FEATURE_PSE) > =20 > static inline int is_hugepage_only_range(struct mm_struct *mm, > unsigned long addr, >=20 >=20 > Which is a followon to Jan's >=20 > y:/usr/src/git26> gitshow 103f6112f253 > commit 103f6112f253017d7062cd74d17f4a514ed4485c > Author: Jan Beulich > AuthorDate: Thu Apr 21 00:27:04 2016 -0600 > Commit: Ingo Molnar > CommitDate: Fri Apr 22 10:05:00 2016 +0200 >=20 > x86/mm/xen: Suppress hugetlbfs in PV guests > =20 > Huge pages are not normally available to PV guests. Not suppressing > hugetlbfs use results in an endless loop of page faults when user = mode > code tries to access a hugetlbfs mapped area (since the hypervisor > denies such PTEs to be created, but error indications can't be > propagated out of xen_set_pte_at(), just like for various of its > siblings), and - once killed in an oops like this: >=20 >=20 > On Sat, 04 Jun 2016 17:08:36 +0000 bugzilla-daemon@bugzilla.kernel.org = wrote: >=20 >> https://bugzilla.kernel.org/show_bug.cgi?id=3D119641=20 >>=20 >> Bug ID: 119641 >> Summary: hugetlbfs: disabling because there are no supported >> hugepage sizes >> Product: Memory Management >> Version: 2.5 >> Kernel Version: 3.6.1 >> Hardware: Intel >> OS: Linux >> Tree: Mainline >> Status: NEW >> Severity: normal >> Priority: P1 >> Component: Other >> Assignee: akpm@linux-foundation.org=20 >> Reporter: jp.pozzi@izzop.net=20 >> Regression: No >>=20 >> Created attachment 219011 >> --> https://bugzilla.kernel.org/attachment.cgi?id=3D219011&action=3Ded= it=20 >> .config for 4.6.1 kernel >>=20 >> Hello, >>=20 >> I get a message while starting the 4.6.1 kernel under Xen : >> hugetlbfs: disabling because there are no supported hugepage sizes >>=20 >> And after grepping /proc/meminfo for Huge I get only : >> grep -i huge /proc/meminfo=20 >> AnonHugePages: 0 kB >>=20 >> I get this message only when starting the kernel under Xen, when = starting >> kernel alone All is OK and I get the "normal" hugepages list. >>=20 >> I test some previous kernels versions : >> 4.5.2 OK >> 4.5.5 KO >> 4.6.0 KO >>=20 >> My system is=20 >> CPU Intel Core I7 6700 >> MEM 32Go >> Disks some ... >> System Debian unstable up to date >>=20 >> I enclose the .config file. >>=20 >> Regards >>=20 >> JP P >>=20 >> --=20 >> You are receiving this mail because: >> You are the assignee for the bug. -- 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