Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754582Ab0D0LMh (ORCPT ); Tue, 27 Apr 2010 07:12:37 -0400 Received: from gir.skynet.ie ([193.1.99.77]:33920 "EHLO gir.skynet.ie" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751001Ab0D0LMd (ORCPT ); Tue, 27 Apr 2010 07:12:33 -0400 Date: Tue, 27 Apr 2010 12:12:14 +0100 From: Mel Gorman To: KAMEZAWA Hiroyuki Cc: Andrea Arcangeli , Minchan Kim , Christoph Lameter , Andrew Morton , Adam Litke , Avi Kivity , David Rientjes , KOSAKI Motohiro , Rik van Riel , linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH 04/14] mm,migration: Allow the migration of PageSwapCache pages Message-ID: <20100427111213.GF4895@csn.ul.ie> References: <1271947206.2100.216.camel@barrios-desktop> <20100422154443.GD30306@csn.ul.ie> <20100423183135.GT32034@random.random> <20100423192311.GC14351@csn.ul.ie> <20100423193948.GU32034@random.random> <20100423213549.GV32034@random.random> <20100424105226.GF14351@csn.ul.ie> <20100425144113.GB5789@random.random> <20100427094002.GD4895@csn.ul.ie> <20100427194139.63ef0864.kamezawa.hiroyu@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20100427194139.63ef0864.kamezawa.hiroyu@jp.fujitsu.com> User-Agent: Mutt/1.5.17+20080114 (2008-01-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 7117 Lines: 150 On Tue, Apr 27, 2010 at 07:41:39PM +0900, KAMEZAWA Hiroyuki wrote: > On Tue, 27 Apr 2010 10:40:02 +0100 > Mel Gorman wrote: > > > On Sun, Apr 25, 2010 at 04:41:13PM +0200, Andrea Arcangeli wrote: > > > On Sat, Apr 24, 2010 at 11:52:27AM +0100, Mel Gorman wrote: > > > > It should be. I expect that's why you have never seen the bugon in > > > > swapops. > > > > > > Oh I just got the very crash you're talking about with aa.git with > > > your v8 code. Weird that I never reproduced it before! I think it's > > > because I fixed gcc to be fully backed by hugepages always (without > > > khugepaged) and I was rebuilding a couple of packages, and that now > > > triggers memory compaction much more, but mixed with heavy > > > fork/execve. This is the only instability I managed to reproduce over > > > 24 hours of stress testing and it's clearly not related to transparent > > > hugepage support but it's either a bug in migrate.c (more likely) or > > > memory compaction. > > > > > > Note that I'm running with the 2.6.33 anon-vma code, so it will > > > relieve you to know it's not the anon-vma recent changes causing this > > > (well I can't rule out anon-vma bugs, but if it's anon-vma, it's a > > > longstanding one). > > > > > > kernel BUG at include/linux/swapops.h:105! > > > invalid opcode: 0000 [#1] SMP > > > last sysfs file: /sys/devices/pci0000:00/0000:00:12.0/host0/target0:0:0/0:0:0:0/block/sr0/size > > > CPU 0 > > > Modules linked in: nls_iso8859_1 loop twofish twofish_common tun bridge stp llc bnep sco rfcomm l2cap bluetooth snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss usbhid gspca_pac207 gspca_main videodev v4l1_compat v4l2_compat_ioctl32 snd_hda_codec_realtek ohci_hcd snd_hda_intel ehci_hcd usbcore snd_hda_codec snd_pcm snd_timer snd snd_page_alloc sg psmouse sr_mod pcspkr > > > > > > Pid: 13351, comm: basename Not tainted 2.6.34-rc5 #23 M2A-VM/System Product Name > > > RIP: 0010:[] [] migration_entry_wait+0x170/0x180 > > > RSP: 0000:ffff88009ab6fa58 EFLAGS: 00010246 > > > RAX: ffffea0000000000 RBX: ffffea000234eed8 RCX: ffff8800aaa95298 > > > RDX: 00000000000a168d RSI: ffff88000411ae28 RDI: ffffea00025550a8 > > > RBP: ffffea0002555098 R08: ffff88000411ae28 R09: 0000000000000000 > > > R10: 0000000000000008 R11: 0000000000000009 R12: 00000000aaa95298 > > > R13: 00007ffff8a53000 R14: ffff88000411ae28 R15: ffff88011108a7c0 > > > FS: 00002adf29469b90(0000) GS:ffff880001a00000(0000) knlGS:0000000055700d50 > > > CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b > > > CR2: 00007ffff8a53000 CR3: 0000000004f80000 CR4: 00000000000006f0 > > > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > > > DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 > > > Process basename (pid: 13351, threadinfo ffff88009ab6e000, task ffff88009ab96c70) > > > Stack: > > > ffff8800aaa95280 ffffffff810ce472 ffff8801134a7ce8 0000000000000000 > > > <0> 00000000142d1a3e ffffffff810c2e35 79f085e9c08a4db7 62d38944fd014000 > > > <0> 76b07a274b0c057a ffffea00025649f8 f8000000000a168d d19934e84d2a74f3 > > > Call Trace: > > > [] ? page_add_new_anon_rmap+0x72/0xc0 > > > [] ? handle_pte_fault+0x7a5/0x7d0 > > > [] ? do_page_fault+0x13d/0x420 > > > [] ? page_fault+0x1f/0x30 > > > [] ? strnlen_user+0x4b/0x80 > > > [] ? load_elf_binary+0x12be/0x1c80 > > > [] ? search_binary_handler+0xad/0x2c0 > > > [] ? do_execve+0x247/0x320 > > > [] ? sys_execve+0x36/0x60 > > > [] ? stub_execve+0x6a/0xc0 > > > Code: 5e ff ff ff 8d 41 01 89 4c 24 08 89 44 24 04 8b 74 24 04 8b 44 24 08 f0 0f b1 32 89 44 24 0c 8b 44 24 0c 39 c8 74 a4 89 c1 eb d1 <0f> 0b eb fe 66 66 66 2e 0f 1f 84 00 00 00 00 00 41 54 49 89 d4 > > > RIP [] migration_entry_wait+0x170/0x180 > > > RSP > > > ---[ end trace 840ce8bc6f6dc402 ]--- > > > > > > It doesn't look like a coincidence the page that had the migration PTE > > > set was the argv in the user stack during execve. The bug has to be > > > there. Or maybe it's a coincidence and it will mislead us. If you've > > > other stack traces please post them so I can have more info (I'll post > > > more stack traces if I get them again, it doesn't look easy to > > > reproduce, supposedly the bug has always been there since the first > > > time I used memory compaction, and this is the first time I reproduce > > > it). > > > > > > > The oopses I am getting look very similar. The page is encountered in > > the stack while copying the arguements in. I don't think it's a > > coincidence. > > > > Hmm. booby trap aronude here ? I think so. I have a debugging patch running at the moment that is checking for migration ptes while the page tables are being moved. > == > static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) > { > .... > /* > * cover the whole range: [new_start, old_end) > */ > if (vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL)) > return -ENOMEM; > > /* > * move the page tables downwards, on failure we rely on > * process cleanup to remove whatever mess we made. > */ > if (length != move_page_tables(vma, old_start, > vma, new_start, length)) > return -ENOMEM; > ... Specfically, I have it in move_ptes. If migration entries are being found there, it would be reasonable for exec() to wait on migration to complete but what you suggest below is more plausible. > /* > * Shrink the vma to just the new range. Always succeeds. > */ > vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL); > > > == > > I think we have wrong vma_address() -> "pte" > == > === (A) === > vma_adjust(). ---- (*) > === (B) === > move_pte(). > == > > vma_address(page, vma) > => address = vma->vm_start + ((page->index << shift) - vma->vm_pgoff) << PAGE_SHIFT); > > So, vma_address() in zone (A) and vma_address in (B) will return different address. > Yes. I was expecting that the anon_vma lock in vma_adjust would delay exec until migration completed. > When pte inludes migration_pte, this seems critical. Because an address pointed > by vma_address() in zone (B) will not contain migration_pte until > move_ptes() ends. > This is plausible considering that, like vma_adjust(), move_ptes does not appear to take the anon_vma->lock in the same fashion as i_mmap_lock is taken for files. -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/