From: Chris Wright <chrisw@sous-sol.org>
To: linux-kernel@vger.kernel.org, stable@kernel.org, jejb@kernel.org
Cc: Justin Forbes <jmforbes@linuxtx.org>,
Zwane Mwaikambo <zwane@arm.linux.org.uk>,
"Theodore Ts'o" <tytso@mit.edu>,
Randy Dunlap <rdunlap@xenotime.net>,
Dave Jones <davej@redhat.com>,
Chuck Wolber <chuckw@quantumlinux.com>,
Chris Wedgwood <reviews@ml.cw.f00f.org>,
Michael Krufky <mkrufky@linuxtv.org>,
Chuck Ebbert <cebbert@redhat.com>,
Domenico Andreoli <cavokz@gmail.com>,
torvalds@linux-foundation.org, akpm@linux-foundation.org,
alan@lxorguk.ukuu.org.uk,
Heiko Carstens <heiko.carstens@de.ibm.com>,
Andy Whitcroft <apw@shadowen.org>,
Dave Hansen <haveblue@us.ibm.com>,
Gerald Schaefer <gerald.schaefer@de.ibm.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Yasunori Goto <y-goto@jp.fujitsu.com>,
Greg Kroah-Hartman <gregkh@suse.de>
Subject: [patch 11/50] memory_hotplug: always initialize pageblock bitmap
Date: Fri, 06 Jun 2008 18:02:26 -0700 [thread overview]
Message-ID: <20080607010556.884011697@sous-sol.org> (raw)
In-Reply-To: <20080607010215.358296706@sous-sol.org>
[-- Attachment #1: memory_hotplug-always-initialize-pageblock-bitmap.patch --]
[-- Type: text/plain, Size: 6025 bytes --]
-stable review patch. If anyone has any objections, please let us know.
---------------------
From: Heiko Carstens <heiko.carstens@de.ibm.com>
commit 76cdd58e558669366adfaded436fda01b30cce3e upstream
Trying to online a new memory section that was added via memory hotplug
sometimes results in crashes when the new pages are added via __free_page.
Reason for that is that the pageblock bitmap isn't initialized and hence
contains random stuff. That means that get_pageblock_migratetype()
returns also random stuff and therefore
list_add(&page->lru,
&zone->free_area[order].free_list[migratetype]);
in __free_one_page() tries to do a list_add to something that isn't even
necessarily a list.
This happens since 86051ca5eaf5e560113ec7673462804c54284456 ("mm: fix
usemap initialization") which makes sure that the pageblock bitmap gets
only initialized for pages present in a zone. Unfortunately for hot-added
memory the zones "grow" after the memmap and the pageblock memmap have
been initialized. Which means that the new pages have an unitialized
bitmap. To solve this the calls to grow_zone_span() and grow_pgdat_span()
are moved to __add_zone() just before the initialization happens.
The patch also moves the two functions since __add_zone() is the only
caller and I didn't want to add a forward declaration.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
---
mm/memory_hotplug.c | 78 ++++++++++++++++++++++++++--------------------------
mm/page_alloc.c | 3 --
2 files changed, 40 insertions(+), 41 deletions(-)
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -58,23 +58,59 @@ static void release_memory_resource(stru
return;
}
-
#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
+static void grow_zone_span(struct zone *zone, unsigned long start_pfn,
+ unsigned long end_pfn)
+{
+ unsigned long old_zone_end_pfn;
+
+ zone_span_writelock(zone);
+
+ old_zone_end_pfn = zone->zone_start_pfn + zone->spanned_pages;
+ if (start_pfn < zone->zone_start_pfn)
+ zone->zone_start_pfn = start_pfn;
+
+ zone->spanned_pages = max(old_zone_end_pfn, end_pfn) -
+ zone->zone_start_pfn;
+
+ zone_span_writeunlock(zone);
+}
+
+static void grow_pgdat_span(struct pglist_data *pgdat, unsigned long start_pfn,
+ unsigned long end_pfn)
+{
+ unsigned long old_pgdat_end_pfn =
+ pgdat->node_start_pfn + pgdat->node_spanned_pages;
+
+ if (start_pfn < pgdat->node_start_pfn)
+ pgdat->node_start_pfn = start_pfn;
+
+ pgdat->node_spanned_pages = max(old_pgdat_end_pfn, end_pfn) -
+ pgdat->node_start_pfn;
+}
+
static int __add_zone(struct zone *zone, unsigned long phys_start_pfn)
{
struct pglist_data *pgdat = zone->zone_pgdat;
int nr_pages = PAGES_PER_SECTION;
int nid = pgdat->node_id;
int zone_type;
+ unsigned long flags;
zone_type = zone - pgdat->node_zones;
if (!zone->wait_table) {
- int ret = 0;
+ int ret;
+
ret = init_currently_empty_zone(zone, phys_start_pfn,
nr_pages, MEMMAP_HOTPLUG);
- if (ret < 0)
+ if (ret)
return ret;
}
+ pgdat_resize_lock(zone->zone_pgdat, &flags);
+ grow_zone_span(zone, phys_start_pfn, phys_start_pfn + nr_pages);
+ grow_pgdat_span(zone->zone_pgdat, phys_start_pfn,
+ phys_start_pfn + nr_pages);
+ pgdat_resize_unlock(zone->zone_pgdat, &flags);
memmap_init_zone(nr_pages, nid, zone_type,
phys_start_pfn, MEMMAP_HOTPLUG);
return 0;
@@ -134,36 +170,6 @@ int __add_pages(struct zone *zone, unsig
}
EXPORT_SYMBOL_GPL(__add_pages);
-static void grow_zone_span(struct zone *zone,
- unsigned long start_pfn, unsigned long end_pfn)
-{
- unsigned long old_zone_end_pfn;
-
- zone_span_writelock(zone);
-
- old_zone_end_pfn = zone->zone_start_pfn + zone->spanned_pages;
- if (start_pfn < zone->zone_start_pfn)
- zone->zone_start_pfn = start_pfn;
-
- zone->spanned_pages = max(old_zone_end_pfn, end_pfn) -
- zone->zone_start_pfn;
-
- zone_span_writeunlock(zone);
-}
-
-static void grow_pgdat_span(struct pglist_data *pgdat,
- unsigned long start_pfn, unsigned long end_pfn)
-{
- unsigned long old_pgdat_end_pfn =
- pgdat->node_start_pfn + pgdat->node_spanned_pages;
-
- if (start_pfn < pgdat->node_start_pfn)
- pgdat->node_start_pfn = start_pfn;
-
- pgdat->node_spanned_pages = max(old_pgdat_end_pfn, end_pfn) -
- pgdat->node_start_pfn;
-}
-
static int online_pages_range(unsigned long start_pfn, unsigned long nr_pages,
void *arg)
{
@@ -183,7 +189,6 @@ static int online_pages_range(unsigned l
int online_pages(unsigned long pfn, unsigned long nr_pages)
{
- unsigned long flags;
unsigned long onlined_pages = 0;
struct zone *zone;
int need_zonelists_rebuild = 0;
@@ -211,11 +216,6 @@ int online_pages(unsigned long pfn, unsi
* memory_block->state_sem.
*/
zone = page_zone(pfn_to_page(pfn));
- pgdat_resize_lock(zone->zone_pgdat, &flags);
- grow_zone_span(zone, pfn, pfn + nr_pages);
- grow_pgdat_span(zone->zone_pgdat, pfn, pfn + nr_pages);
- pgdat_resize_unlock(zone->zone_pgdat, &flags);
-
/*
* If this zone is not populated, then it is not in zonelist.
* This means the page allocator ignores this zone.
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2837,8 +2837,6 @@ __meminit int init_currently_empty_zone(
zone->zone_start_pfn = zone_start_pfn;
- memmap_init(size, pgdat->node_id, zone_idx(zone), zone_start_pfn);
-
zone_init_free_lists(zone);
return 0;
@@ -3408,6 +3406,7 @@ static void __paginginit free_area_init_
ret = init_currently_empty_zone(zone, zone_start_pfn,
size, MEMMAP_EARLY);
BUG_ON(ret);
+ memmap_init(size, nid, j, zone_start_pfn);
zone_start_pfn += size;
}
}
--
next prev parent reply other threads:[~2008-06-07 1:30 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-06-07 1:02 [patch 00/50] 2.6.25.6 -stable review Chris Wright
2008-06-07 1:02 ` [patch 01/50] block: do_mounts - accept root=<non-existant partition> Chris Wright
2008-06-09 13:38 ` Adrian Bunk
2008-06-09 16:48 ` Chris Wright
2008-06-07 1:02 ` [patch 02/50] POWERPC Bolt in SLB entry for kernel stack on secondary cpus Chris Wright
2008-06-07 1:02 ` [patch 03/50] USB: remove PICDEM FS USB demo (04d8:000c) device from ldusb Chris Wright
2008-06-07 1:02 ` [patch 04/50] USB: unusual_devs: Add support for GI 0401 SD-Card interface Chris Wright
2008-06-07 1:02 ` [patch 05/50] USB: option: add new Dell 5520 HSDPA variant Chris Wright
2008-06-07 1:02 ` [patch 06/50] USB: add Zoom Telephonics Model 3095F V.92 USB Mini External modem to cdc-acm Chris Wright
2008-06-07 1:02 ` [patch 07/50] USB: do not handle device 1410:5010 in option driver Chris Wright
2008-06-07 1:02 ` [patch 08/50] USB: add Telstra NextG CDMA id to " Chris Wright
2008-06-07 1:02 ` [patch 09/50] ext3/4: fix uninitialized bs in ext3/4_xattr_set_handle() Chris Wright
2008-06-07 1:02 ` [patch 10/50] libata: force hardreset if link is in powersave mode Chris Wright
2008-06-07 1:27 ` Linus Torvalds
2008-06-07 1:57 ` [stable] " Greg KH
2008-06-07 16:51 ` Linus Torvalds
2008-06-07 3:14 ` Jeff Garzik
2008-06-07 1:02 ` Chris Wright [this message]
2008-06-07 1:02 ` [patch 12/50] USB: fix build errors in ohci-omap.c and ohci-sm501.c Chris Wright
2008-06-07 1:02 ` [patch 13/50] i2c-nforce2: Disable the second SMBus channel on the DFI Lanparty NF4 Expert Chris Wright
2008-06-07 1:02 ` [patch 14/50] i2c/max6875: Really prevent 24RF08 corruption Chris Wright
2008-06-07 1:02 ` [patch 15/50] x86: user_regset_view table fix for ia32 on 64-bit Chris Wright
2008-06-07 1:02 ` [patch 16/50] i386: fix asm constraint in do_IRQ() Chris Wright
2008-06-07 1:02 ` [patch 17/50] CIFS: Fix UNC path prefix on QueryUnixPathInfo to have correct slash Chris Wright
2008-06-07 1:02 ` [patch 18/50] USB: add TELIT HDSPA UC864-E modem to option driver Chris Wright
2008-06-07 1:02 ` [patch 19/50] usb-serial: Use ftdi_sio driver for RATOC REX-USB60F Chris Wright
2008-06-07 1:02 ` [patch 20/50] cgroups: remove node_ prefix_from ns subsystem Chris Wright
2008-06-07 1:02 ` [patch 21/50] brd: dont show ramdisks in /proc/partitions Chris Wright
2008-06-07 21:51 ` H. Peter Anvin
2008-06-09 17:08 ` [stable] " Chris Wright
2008-06-09 19:49 ` Willy Tarreau
2008-06-09 20:28 ` H. Peter Anvin
2008-06-07 1:02 ` [patch 22/50] types.h: dont expose struct ustat to userspace Chris Wright
2008-06-07 1:02 ` [patch 23/50] x86: dont read maxlvt before checking if APIC is mapped Chris Wright
2008-06-07 1:02 ` [patch 24/50] IPoIB: Test for NULL broadcast object in ipiob_mcast_join_finish() Chris Wright
2008-06-07 1:02 ` [patch 25/50] x86: fix setup of cyc2ns in tsc_64.c Chris Wright
2008-06-07 1:02 ` [patch 26/50] x86: if we cannot calibrate the TSC, we panic Chris Wright
2008-06-07 1:02 ` [patch 27/50] x86: distangle user disabled TSC from unstable Chris Wright
2008-06-07 1:02 ` [patch 28/50] x86: disable TSC for sched_clock() when calibration failed Chris Wright
2008-06-07 1:02 ` [patch 29/50] XFS: Fix memory corruption with small buffer reads Chris Wright
2008-06-07 1:02 ` [patch 30/50] x86: prevent PGE flush from interruption/preemption Chris Wright
2008-06-07 1:02 ` [patch 31/50] Revert "PCI: remove default PCI expansion ROM memory allocation" Chris Wright
2008-06-07 1:02 ` [patch 32/50] PS3: gelic: fix memory leak Chris Wright
2008-06-07 1:02 ` [patch 33/50] netfilter: xt_iprange: module aliases for xt_iprange Chris Wright
2008-06-07 1:02 ` [patch 34/50] HID: split Numlock emulation quirk from HID_QUIRK_APPLE_HAS_FN Chris Wright
2008-06-07 1:02 ` [patch 35/50] CPUFREQ: Make acpi-cpufreq more robust against BIOS freq changes behind our back Chris Wright
2008-06-07 1:02 ` [patch 36/50] x86, fpu: fix CONFIG_PREEMPT=y corruption of applications FPU stack Chris Wright
2008-06-07 1:02 ` [patch 37/50] netfilter: nf_conntrack_expect: fix error path unwind in nf_conntrack_expect_init() Chris Wright
2008-06-07 1:02 ` [patch 38/50] netfilter: xt_connlimit: fix accouning when receive RST packet in ESTABLISHED state Chris Wright
2008-06-07 1:02 ` [patch 39/50] netfilter: nf_conntrack_ipv6: fix inconsistent lock state in nf_ct_frag6_gather() Chris Wright
2008-06-07 1:02 ` [patch 40/50] atl1: fix 4G memory corruption bug Chris Wright
2008-06-07 1:02 ` [patch 41/50] Smack: fuse mount hang fix Chris Wright
2008-06-07 1:02 ` [patch 42/50] proc: calculate the correct /proc/<pid> link count Chris Wright
2008-06-07 1:02 ` [patch 43/50] pagemap: fix bug in add_to_pagemap, require aligned-length reads of /proc/pid/pagemap Chris Wright
2008-06-07 1:02 ` [patch 44/50] brk: make sys_brk() honor COMPAT_BRK when computing lower bound Chris Wright
2008-06-07 1:03 ` [patch 45/50] eCryptfs: remove unnecessary page decrypt call Chris Wright
2008-06-07 1:03 ` [patch 46/50] md: do not compute parity unless it is on a failed drive Chris Wright
2008-06-07 1:03 ` [patch 47/50] md: fix uninitialized use of mddev->recovery_wait Chris Wright
2008-06-07 1:03 ` [patch 48/50] md: fix prexor vs sync_request race Chris Wright
2008-06-07 1:03 ` [patch 49/50] capabilities: remain source compatible with 32-bit raw legacy capability support Chris Wright
2008-06-07 1:03 ` [patch 50/50] cpufreq: fix null object access on Transmeta CPU Chris Wright
2008-06-07 3:49 ` [patch 00/50] 2.6.25.6 -stable review Hugh Dickins
2008-06-07 4:24 ` [stable] " Chris Wright
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20080607010556.884011697@sous-sol.org \
--to=chrisw@sous-sol.org \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=apw@shadowen.org \
--cc=cavokz@gmail.com \
--cc=cebbert@redhat.com \
--cc=chuckw@quantumlinux.com \
--cc=davej@redhat.com \
--cc=gerald.schaefer@de.ibm.com \
--cc=gregkh@suse.de \
--cc=haveblue@us.ibm.com \
--cc=heiko.carstens@de.ibm.com \
--cc=jejb@kernel.org \
--cc=jmforbes@linuxtx.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mkrufky@linuxtv.org \
--cc=rdunlap@xenotime.net \
--cc=reviews@ml.cw.f00f.org \
--cc=stable@kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=tytso@mit.edu \
--cc=y-goto@jp.fujitsu.com \
--cc=zwane@arm.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®