2009-09-16 03:50:15 -06:00
|
|
|
/*
|
|
|
|
* Copyright (C) 2008, 2009 Intel Corporation
|
|
|
|
* Authors: Andi Kleen, Fengguang Wu
|
|
|
|
*
|
|
|
|
* This software may be redistributed and/or modified under the terms of
|
|
|
|
* the GNU General Public License ("GPL") version 2 only as published by the
|
|
|
|
* Free Software Foundation.
|
|
|
|
*
|
|
|
|
* High level machine check handler. Handles pages reported by the
|
2010-09-27 15:09:51 -06:00
|
|
|
* hardware as being corrupted usually due to a multi-bit ECC memory or cache
|
2009-09-16 03:50:15 -06:00
|
|
|
* failure.
|
2010-09-27 15:09:51 -06:00
|
|
|
*
|
|
|
|
* In addition there is a "soft offline" entry point that allows stop using
|
|
|
|
* not-yet-corrupted-by-suspicious pages without killing anything.
|
2009-09-16 03:50:15 -06:00
|
|
|
*
|
|
|
|
* Handles page cache pages in various states. The tricky part
|
2010-09-27 15:09:51 -06:00
|
|
|
* here is that we can access any page asynchronously in respect to
|
|
|
|
* other VM users, because memory failures could happen anytime and
|
|
|
|
* anywhere. This could violate some of their assumptions. This is why
|
|
|
|
* this code has to be extremely careful. Generally it tries to use
|
|
|
|
* normal locking rules, as in get the standard locks, even if that means
|
|
|
|
* the error handling takes potentially a long time.
|
|
|
|
*
|
|
|
|
* There are several operations here with exponential complexity because
|
|
|
|
* of unsuitable VM data structures. For example the operation to map back
|
|
|
|
* from RMAP chains to processes has to walk the complete process list and
|
|
|
|
* has non linear complexity with the number. But since memory corruptions
|
|
|
|
* are rare we hope to get away with this. This avoids impacting the core
|
|
|
|
* VM.
|
2009-09-16 03:50:15 -06:00
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Notebook:
|
|
|
|
* - hugetlb needs more code
|
|
|
|
* - kcore/oldmem/vmcore/mem/kmem check for hwpoison pages
|
|
|
|
* - pass bad pages to kdump next kernel
|
|
|
|
*/
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/page-flags.h>
|
2009-12-16 04:19:59 -07:00
|
|
|
#include <linux/kernel-page-flags.h>
|
2009-09-16 03:50:15 -06:00
|
|
|
#include <linux/sched.h>
|
2009-10-13 08:02:11 -06:00
|
|
|
#include <linux/ksm.h>
|
2009-09-16 03:50:15 -06:00
|
|
|
#include <linux/rmap.h>
|
2011-05-26 14:00:52 -06:00
|
|
|
#include <linux/export.h>
|
2009-09-16 03:50:15 -06:00
|
|
|
#include <linux/pagemap.h>
|
|
|
|
#include <linux/swap.h>
|
|
|
|
#include <linux/backing-dev.h>
|
2009-12-16 04:20:00 -07:00
|
|
|
#include <linux/migrate.h>
|
|
|
|
#include <linux/page-isolation.h>
|
|
|
|
#include <linux/suspend.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 02:04:11 -06:00
|
|
|
#include <linux/slab.h>
|
2010-05-31 00:28:19 -06:00
|
|
|
#include <linux/swapops.h>
|
2010-05-27 18:29:17 -06:00
|
|
|
#include <linux/hugetlb.h>
|
2010-12-02 15:31:19 -07:00
|
|
|
#include <linux/memory_hotplug.h>
|
2011-06-15 16:08:48 -06:00
|
|
|
#include <linux/mm_inline.h>
|
2011-07-12 23:14:27 -06:00
|
|
|
#include <linux/kfifo.h>
|
2009-09-16 03:50:15 -06:00
|
|
|
#include "internal.h"
|
|
|
|
|
|
|
|
int sysctl_memory_failure_early_kill __read_mostly = 0;
|
|
|
|
|
|
|
|
int sysctl_memory_failure_recovery __read_mostly = 1;
|
|
|
|
|
2013-02-22 17:34:02 -07:00
|
|
|
atomic_long_t num_poisoned_pages __read_mostly = ATOMIC_LONG_INIT(0);
|
2009-09-16 03:50:15 -06:00
|
|
|
|
2009-12-21 11:56:42 -07:00
|
|
|
#if defined(CONFIG_HWPOISON_INJECT) || defined(CONFIG_HWPOISON_INJECT_MODULE)
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
u32 hwpoison_filter_enable = 0;
|
2009-12-16 04:19:59 -07:00
|
|
|
u32 hwpoison_filter_dev_major = ~0U;
|
|
|
|
u32 hwpoison_filter_dev_minor = ~0U;
|
2009-12-16 04:19:59 -07:00
|
|
|
u64 hwpoison_filter_flags_mask;
|
|
|
|
u64 hwpoison_filter_flags_value;
|
2009-12-16 04:19:59 -07:00
|
|
|
EXPORT_SYMBOL_GPL(hwpoison_filter_enable);
|
2009-12-16 04:19:59 -07:00
|
|
|
EXPORT_SYMBOL_GPL(hwpoison_filter_dev_major);
|
|
|
|
EXPORT_SYMBOL_GPL(hwpoison_filter_dev_minor);
|
2009-12-16 04:19:59 -07:00
|
|
|
EXPORT_SYMBOL_GPL(hwpoison_filter_flags_mask);
|
|
|
|
EXPORT_SYMBOL_GPL(hwpoison_filter_flags_value);
|
2009-12-16 04:19:59 -07:00
|
|
|
|
|
|
|
static int hwpoison_filter_dev(struct page *p)
|
|
|
|
{
|
|
|
|
struct address_space *mapping;
|
|
|
|
dev_t dev;
|
|
|
|
|
|
|
|
if (hwpoison_filter_dev_major == ~0U &&
|
|
|
|
hwpoison_filter_dev_minor == ~0U)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/*
|
2010-09-27 15:09:51 -06:00
|
|
|
* page_mapping() does not accept slab pages.
|
2009-12-16 04:19:59 -07:00
|
|
|
*/
|
|
|
|
if (PageSlab(p))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
mapping = page_mapping(p);
|
|
|
|
if (mapping == NULL || mapping->host == NULL)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
dev = mapping->host->i_sb->s_dev;
|
|
|
|
if (hwpoison_filter_dev_major != ~0U &&
|
|
|
|
hwpoison_filter_dev_major != MAJOR(dev))
|
|
|
|
return -EINVAL;
|
|
|
|
if (hwpoison_filter_dev_minor != ~0U &&
|
|
|
|
hwpoison_filter_dev_minor != MINOR(dev))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
static int hwpoison_filter_flags(struct page *p)
|
|
|
|
{
|
|
|
|
if (!hwpoison_filter_flags_mask)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if ((stable_page_flags(p) & hwpoison_filter_flags_mask) ==
|
|
|
|
hwpoison_filter_flags_value)
|
|
|
|
return 0;
|
|
|
|
else
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
/*
|
|
|
|
* This allows stress tests to limit test scope to a collection of tasks
|
|
|
|
* by putting them under some memcg. This prevents killing unrelated/important
|
|
|
|
* processes such as /sbin/init. Note that the target task may share clean
|
|
|
|
* pages with init (eg. libc text), which is harmless. If the target task
|
|
|
|
* share _dirty_ pages with another task B, the test scheme must make sure B
|
|
|
|
* is also included in the memcg. At last, due to race conditions this filter
|
|
|
|
* can only guarantee that the page either belongs to the memcg tasks, or is
|
|
|
|
* a freed page.
|
|
|
|
*/
|
2012-07-31 17:43:02 -06:00
|
|
|
#ifdef CONFIG_MEMCG_SWAP
|
2009-12-16 04:19:59 -07:00
|
|
|
u64 hwpoison_filter_memcg;
|
|
|
|
EXPORT_SYMBOL_GPL(hwpoison_filter_memcg);
|
|
|
|
static int hwpoison_filter_task(struct page *p)
|
|
|
|
{
|
|
|
|
struct mem_cgroup *mem;
|
|
|
|
struct cgroup_subsys_state *css;
|
|
|
|
unsigned long ino;
|
|
|
|
|
|
|
|
if (!hwpoison_filter_memcg)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
mem = try_get_mem_cgroup_from_page(p);
|
|
|
|
if (!mem)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
css = mem_cgroup_css(mem);
|
2014-02-11 09:52:49 -07:00
|
|
|
ino = cgroup_ino(css->cgroup);
|
2009-12-16 04:19:59 -07:00
|
|
|
css_put(css);
|
|
|
|
|
2014-09-19 02:29:31 -06:00
|
|
|
if (ino != hwpoison_filter_memcg)
|
2009-12-16 04:19:59 -07:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
static int hwpoison_filter_task(struct page *p) { return 0; }
|
|
|
|
#endif
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
int hwpoison_filter(struct page *p)
|
|
|
|
{
|
2009-12-16 04:19:59 -07:00
|
|
|
if (!hwpoison_filter_enable)
|
|
|
|
return 0;
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
if (hwpoison_filter_dev(p))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
if (hwpoison_filter_flags(p))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
if (hwpoison_filter_task(p))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
return 0;
|
|
|
|
}
|
2009-12-21 11:56:42 -07:00
|
|
|
#else
|
|
|
|
int hwpoison_filter(struct page *p)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2009-12-16 04:19:59 -07:00
|
|
|
EXPORT_SYMBOL_GPL(hwpoison_filter);
|
|
|
|
|
2009-09-16 03:50:15 -06:00
|
|
|
/*
|
2011-12-13 10:27:58 -07:00
|
|
|
* Send all the processes who have the page mapped a signal.
|
|
|
|
* ``action optional'' if they are not immediately affected by the error
|
|
|
|
* ``action required'' if error happened in current execution context
|
2009-09-16 03:50:15 -06:00
|
|
|
*/
|
2011-12-13 10:27:58 -07:00
|
|
|
static int kill_proc(struct task_struct *t, unsigned long addr, int trapno,
|
|
|
|
unsigned long pfn, struct page *page, int flags)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
struct siginfo si;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
printk(KERN_ERR
|
2011-12-13 10:27:58 -07:00
|
|
|
"MCE %#lx: Killing %s:%d due to hardware memory corruption\n",
|
2009-09-16 03:50:15 -06:00
|
|
|
pfn, t->comm, t->pid);
|
|
|
|
si.si_signo = SIGBUS;
|
|
|
|
si.si_errno = 0;
|
|
|
|
si.si_addr = (void *)addr;
|
|
|
|
#ifdef __ARCH_SI_TRAPNO
|
|
|
|
si.si_trapno = trapno;
|
|
|
|
#endif
|
2013-09-11 15:22:52 -06:00
|
|
|
si.si_addr_lsb = compound_order(compound_head(page)) + PAGE_SHIFT;
|
2011-12-13 10:27:58 -07:00
|
|
|
|
2014-06-04 17:10:59 -06:00
|
|
|
if ((flags & MF_ACTION_REQUIRED) && t->mm == current->mm) {
|
2011-12-13 10:27:58 -07:00
|
|
|
si.si_code = BUS_MCEERR_AR;
|
2014-06-04 17:10:59 -06:00
|
|
|
ret = force_sig_info(SIGBUS, &si, current);
|
2011-12-13 10:27:58 -07:00
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* Don't use force here, it's convenient if the signal
|
|
|
|
* can be temporarily blocked.
|
|
|
|
* This could cause a loop when the user sets SIGBUS
|
|
|
|
* to SIG_IGN, but hopefully no one will do that?
|
|
|
|
*/
|
|
|
|
si.si_code = BUS_MCEERR_AO;
|
|
|
|
ret = send_sig_info(SIGBUS, &si, t); /* synchronous? */
|
|
|
|
}
|
2009-09-16 03:50:15 -06:00
|
|
|
if (ret < 0)
|
|
|
|
printk(KERN_INFO "MCE: Error sending signal to %s:%d: %d\n",
|
|
|
|
t->comm, t->pid, ret);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2009-12-16 04:19:57 -07:00
|
|
|
/*
|
|
|
|
* When a unknown page type is encountered drain as many buffers as possible
|
|
|
|
* in the hope to turn the page into a LRU or free page, which we can handle.
|
|
|
|
*/
|
2009-12-16 04:20:00 -07:00
|
|
|
void shake_page(struct page *p, int access)
|
2009-12-16 04:19:57 -07:00
|
|
|
{
|
|
|
|
if (!PageSlab(p)) {
|
|
|
|
lru_add_drain_all();
|
|
|
|
if (PageLRU(p))
|
|
|
|
return;
|
|
|
|
drain_all_pages();
|
|
|
|
if (PageLRU(p) || is_free_buddy_page(p))
|
|
|
|
return;
|
|
|
|
}
|
2009-12-16 04:20:00 -07:00
|
|
|
|
2009-12-16 04:19:57 -07:00
|
|
|
/*
|
2011-02-01 16:52:41 -07:00
|
|
|
* Only call shrink_slab here (which would also shrink other caches) if
|
|
|
|
* access is not potentially fatal.
|
2009-12-16 04:19:57 -07:00
|
|
|
*/
|
2009-12-16 04:20:00 -07:00
|
|
|
if (access) {
|
|
|
|
int nr;
|
2013-08-27 18:18:03 -06:00
|
|
|
int nid = page_to_nid(p);
|
2009-12-16 04:20:00 -07:00
|
|
|
do {
|
2011-05-24 18:12:26 -06:00
|
|
|
struct shrink_control shrink = {
|
|
|
|
.gfp_mask = GFP_KERNEL,
|
|
|
|
};
|
2013-08-27 18:18:03 -06:00
|
|
|
node_set(nid, shrink.nodes_to_scan);
|
2011-05-24 18:12:26 -06:00
|
|
|
|
2011-05-24 18:12:27 -06:00
|
|
|
nr = shrink_slab(&shrink, 1000, 1000);
|
2010-09-27 23:37:55 -06:00
|
|
|
if (page_count(p) == 1)
|
2009-12-16 04:20:00 -07:00
|
|
|
break;
|
|
|
|
} while (nr > 10);
|
|
|
|
}
|
2009-12-16 04:19:57 -07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(shake_page);
|
|
|
|
|
2009-09-16 03:50:15 -06:00
|
|
|
/*
|
|
|
|
* Kill all processes that have a poisoned page mapped and then isolate
|
|
|
|
* the page.
|
|
|
|
*
|
|
|
|
* General strategy:
|
|
|
|
* Find all processes having the page mapped and kill them.
|
|
|
|
* But we keep a page reference around so that the page is not
|
|
|
|
* actually freed yet.
|
|
|
|
* Then stash the page away
|
|
|
|
*
|
|
|
|
* There's no convenient way to get back to mapped processes
|
|
|
|
* from the VMAs. So do a brute-force search over all
|
|
|
|
* running processes.
|
|
|
|
*
|
|
|
|
* Remember that machine checks are not common (or rather
|
|
|
|
* if they are common you have other problems), so this shouldn't
|
|
|
|
* be a performance issue.
|
|
|
|
*
|
|
|
|
* Also there are some races possible while we get from the
|
|
|
|
* error detection to actually handle it.
|
|
|
|
*/
|
|
|
|
|
|
|
|
struct to_kill {
|
|
|
|
struct list_head nd;
|
|
|
|
struct task_struct *tsk;
|
|
|
|
unsigned long addr;
|
2010-09-27 15:36:05 -06:00
|
|
|
char addr_valid;
|
2009-09-16 03:50:15 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Failure handling: if we can't find or can't kill a process there's
|
|
|
|
* not much we can do. We just print a message and ignore otherwise.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Schedule a process for later kill.
|
|
|
|
* Uses GFP_ATOMIC allocations to avoid potential recursions in the VM.
|
|
|
|
* TBD would GFP_NOIO be enough?
|
|
|
|
*/
|
|
|
|
static void add_to_kill(struct task_struct *tsk, struct page *p,
|
|
|
|
struct vm_area_struct *vma,
|
|
|
|
struct list_head *to_kill,
|
|
|
|
struct to_kill **tkc)
|
|
|
|
{
|
|
|
|
struct to_kill *tk;
|
|
|
|
|
|
|
|
if (*tkc) {
|
|
|
|
tk = *tkc;
|
|
|
|
*tkc = NULL;
|
|
|
|
} else {
|
|
|
|
tk = kmalloc(sizeof(struct to_kill), GFP_ATOMIC);
|
|
|
|
if (!tk) {
|
|
|
|
printk(KERN_ERR
|
|
|
|
"MCE: Out of memory while machine check handling\n");
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
tk->addr = page_address_in_vma(p, vma);
|
|
|
|
tk->addr_valid = 1;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In theory we don't have to kill when the page was
|
|
|
|
* munmaped. But it could be also a mremap. Since that's
|
|
|
|
* likely very rare kill anyways just out of paranoia, but use
|
|
|
|
* a SIGKILL because the error is not contained anymore.
|
|
|
|
*/
|
|
|
|
if (tk->addr == -EFAULT) {
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("MCE: Unable to find user space address %lx in %s\n",
|
2009-09-16 03:50:15 -06:00
|
|
|
page_to_pfn(p), tsk->comm);
|
|
|
|
tk->addr_valid = 0;
|
|
|
|
}
|
|
|
|
get_task_struct(tsk);
|
|
|
|
tk->tsk = tsk;
|
|
|
|
list_add_tail(&tk->nd, to_kill);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Kill the processes that have been collected earlier.
|
|
|
|
*
|
|
|
|
* Only do anything when DOIT is set, otherwise just free the list
|
|
|
|
* (this is used for clean pages which do not need killing)
|
|
|
|
* Also when FAIL is set do a force kill because something went
|
|
|
|
* wrong earlier.
|
|
|
|
*/
|
2012-07-11 11:20:47 -06:00
|
|
|
static void kill_procs(struct list_head *to_kill, int forcekill, int trapno,
|
2011-12-13 10:27:58 -07:00
|
|
|
int fail, struct page *page, unsigned long pfn,
|
|
|
|
int flags)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
struct to_kill *tk, *next;
|
|
|
|
|
|
|
|
list_for_each_entry_safe (tk, next, to_kill, nd) {
|
2012-07-11 11:20:47 -06:00
|
|
|
if (forcekill) {
|
2009-09-16 03:50:15 -06:00
|
|
|
/*
|
tree-wide: fix assorted typos all over the place
That is "success", "unknown", "through", "performance", "[re|un]mapping"
, "access", "default", "reasonable", "[con]currently", "temperature"
, "channel", "[un]used", "application", "example","hierarchy", "therefore"
, "[over|under]flow", "contiguous", "threshold", "enough" and others.
Signed-off-by: André Goddard Rosa <andre.goddard@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2009-11-14 08:09:05 -07:00
|
|
|
* In case something went wrong with munmapping
|
2009-09-16 03:50:15 -06:00
|
|
|
* make sure the process doesn't catch the
|
|
|
|
* signal and then access the memory. Just kill it.
|
|
|
|
*/
|
|
|
|
if (fail || tk->addr_valid == 0) {
|
|
|
|
printk(KERN_ERR
|
|
|
|
"MCE %#lx: forcibly killing %s:%d because of failure to unmap corrupted page\n",
|
|
|
|
pfn, tk->tsk->comm, tk->tsk->pid);
|
|
|
|
force_sig(SIGKILL, tk->tsk);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In theory the process could have mapped
|
|
|
|
* something else on the address in-between. We could
|
|
|
|
* check for that, but we need to tell the
|
|
|
|
* process anyways.
|
|
|
|
*/
|
2011-12-13 10:27:58 -07:00
|
|
|
else if (kill_proc(tk->tsk, tk->addr, trapno,
|
|
|
|
pfn, page, flags) < 0)
|
2009-09-16 03:50:15 -06:00
|
|
|
printk(KERN_ERR
|
|
|
|
"MCE %#lx: Cannot send advisory machine check signal to %s:%d\n",
|
|
|
|
pfn, tk->tsk->comm, tk->tsk->pid);
|
|
|
|
}
|
|
|
|
put_task_struct(tk->tsk);
|
|
|
|
kfree(tk);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-06-04 17:11:02 -06:00
|
|
|
/*
|
|
|
|
* Find a dedicated thread which is supposed to handle SIGBUS(BUS_MCEERR_AO)
|
|
|
|
* on behalf of the thread group. Return task_struct of the (first found)
|
|
|
|
* dedicated thread if found, and return NULL otherwise.
|
|
|
|
*
|
|
|
|
* We already hold read_lock(&tasklist_lock) in the caller, so we don't
|
|
|
|
* have to call rcu_read_lock/unlock() in this function.
|
|
|
|
*/
|
|
|
|
static struct task_struct *find_early_kill_thread(struct task_struct *tsk)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
2014-06-04 17:11:02 -06:00
|
|
|
struct task_struct *t;
|
|
|
|
|
|
|
|
for_each_thread(tsk, t)
|
|
|
|
if ((t->flags & PF_MCE_PROCESS) && (t->flags & PF_MCE_EARLY))
|
|
|
|
return t;
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Determine whether a given process is "early kill" process which expects
|
|
|
|
* to be signaled when some page under the process is hwpoisoned.
|
|
|
|
* Return task_struct of the dedicated thread (main thread unless explicitly
|
|
|
|
* specified) if the process is "early kill," and otherwise returns NULL.
|
|
|
|
*/
|
|
|
|
static struct task_struct *task_early_kill(struct task_struct *tsk,
|
|
|
|
int force_early)
|
|
|
|
{
|
|
|
|
struct task_struct *t;
|
2009-09-16 03:50:15 -06:00
|
|
|
if (!tsk->mm)
|
2014-06-04 17:11:02 -06:00
|
|
|
return NULL;
|
2014-06-04 17:11:01 -06:00
|
|
|
if (force_early)
|
2014-06-04 17:11:02 -06:00
|
|
|
return tsk;
|
|
|
|
t = find_early_kill_thread(tsk);
|
|
|
|
if (t)
|
|
|
|
return t;
|
|
|
|
if (sysctl_memory_failure_early_kill)
|
|
|
|
return tsk;
|
|
|
|
return NULL;
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Collect processes when the error hit an anonymous page.
|
|
|
|
*/
|
|
|
|
static void collect_procs_anon(struct page *page, struct list_head *to_kill,
|
2014-06-04 17:11:01 -06:00
|
|
|
struct to_kill **tkc, int force_early)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
struct vm_area_struct *vma;
|
|
|
|
struct task_struct *tsk;
|
|
|
|
struct anon_vma *av;
|
mm anon rmap: replace same_anon_vma linked list with an interval tree.
When a large VMA (anon or private file mapping) is first touched, which
will populate its anon_vma field, and then split into many regions through
the use of mprotect(), the original anon_vma ends up linking all of the
vmas on a linked list. This can cause rmap to become inefficient, as we
have to walk potentially thousands of irrelevent vmas before finding the
one a given anon page might fall into.
By replacing the same_anon_vma linked list with an interval tree (where
each avc's interval is determined by its vma's start and last pgoffs), we
can make rmap efficient for this use case again.
While the change is large, all of its pieces are fairly simple.
Most places that were walking the same_anon_vma list were looking for a
known pgoff, so they can just use the anon_vma_interval_tree_foreach()
interval tree iterator instead. The exception here is ksm, where the
page's index is not known. It would probably be possible to rework ksm so
that the index would be known, but for now I have decided to keep things
simple and just walk the entirety of the interval tree there.
When updating vma's that already have an anon_vma assigned, we must take
care to re-index the corresponding avc's on their interval tree. This is
done through the use of anon_vma_interval_tree_pre_update_vma() and
anon_vma_interval_tree_post_update_vma(), which remove the avc's from
their interval tree before the update and re-insert them after the update.
The anon_vma stays locked during the update, so there is no chance that
rmap would miss the vmas that are being updated.
Signed-off-by: Michel Lespinasse <walken@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Daniel Santos <daniel.santos@pobox.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-08 17:31:39 -06:00
|
|
|
pgoff_t pgoff;
|
2009-09-16 03:50:15 -06:00
|
|
|
|
2012-12-02 12:56:50 -07:00
|
|
|
av = page_lock_anon_vma_read(page);
|
2009-09-16 03:50:15 -06:00
|
|
|
if (av == NULL) /* Not actually mapped anymore */
|
2011-06-27 17:18:09 -06:00
|
|
|
return;
|
|
|
|
|
2014-07-23 15:00:01 -06:00
|
|
|
pgoff = page_to_pgoff(page);
|
2011-06-27 17:18:09 -06:00
|
|
|
read_lock(&tasklist_lock);
|
2009-09-16 03:50:15 -06:00
|
|
|
for_each_process (tsk) {
|
mm: change anon_vma linking to fix multi-process server scalability issue
The old anon_vma code can lead to scalability issues with heavily forking
workloads. Specifically, each anon_vma will be shared between the parent
process and all its child processes.
In a workload with 1000 child processes and a VMA with 1000 anonymous
pages per process that get COWed, this leads to a system with a million
anonymous pages in the same anon_vma, each of which is mapped in just one
of the 1000 processes. However, the current rmap code needs to walk them
all, leading to O(N) scanning complexity for each page.
This can result in systems where one CPU is walking the page tables of
1000 processes in page_referenced_one, while all other CPUs are stuck on
the anon_vma lock. This leads to catastrophic failure for a benchmark
like AIM7, where the total number of processes can reach in the tens of
thousands. Real workloads are still a factor 10 less process intensive
than AIM7, but they are catching up.
This patch changes the way anon_vmas and VMAs are linked, which allows us
to associate multiple anon_vmas with a VMA. At fork time, each child
process gets its own anon_vmas, in which its COWed pages will be
instantiated. The parents' anon_vma is also linked to the VMA, because
non-COWed pages could be present in any of the children.
This reduces rmap scanning complexity to O(1) for the pages of the 1000
child processes, with O(N) complexity for at most 1/N pages in the system.
This reduces the average scanning cost in heavily forking workloads from
O(N) to 2.
The only real complexity in this patch stems from the fact that linking a
VMA to anon_vmas now involves memory allocations. This means vma_adjust
can fail, if it needs to attach a VMA to anon_vma structures. This in
turn means error handling needs to be added to the calling functions.
A second source of complexity is that, because there can be multiple
anon_vmas, the anon_vma linking in vma_adjust can no longer be done under
"the" anon_vma lock. To prevent the rmap code from walking up an
incomplete VMA, this patch introduces the VM_LOCK_RMAP VMA flag. This bit
flag uses the same slot as the NOMMU VM_MAPPED_COPY, with an ifdef in mm.h
to make sure it is impossible to compile a kernel that needs both symbolic
values for the same bitflag.
Some test results:
Without the anon_vma changes, when AIM7 hits around 9.7k users (on a test
box with 16GB RAM and not quite enough IO), the system ends up running
>99% in system time, with every CPU on the same anon_vma lock in the
pageout code.
With these changes, AIM7 hits the cross-over point around 29.7k users.
This happens with ~99% IO wait time, there never seems to be any spike in
system time. The anon_vma lock contention appears to be resolved.
[akpm@linux-foundation.org: cleanups]
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-05 14:42:07 -07:00
|
|
|
struct anon_vma_chain *vmac;
|
2014-06-04 17:11:02 -06:00
|
|
|
struct task_struct *t = task_early_kill(tsk, force_early);
|
mm: change anon_vma linking to fix multi-process server scalability issue
The old anon_vma code can lead to scalability issues with heavily forking
workloads. Specifically, each anon_vma will be shared between the parent
process and all its child processes.
In a workload with 1000 child processes and a VMA with 1000 anonymous
pages per process that get COWed, this leads to a system with a million
anonymous pages in the same anon_vma, each of which is mapped in just one
of the 1000 processes. However, the current rmap code needs to walk them
all, leading to O(N) scanning complexity for each page.
This can result in systems where one CPU is walking the page tables of
1000 processes in page_referenced_one, while all other CPUs are stuck on
the anon_vma lock. This leads to catastrophic failure for a benchmark
like AIM7, where the total number of processes can reach in the tens of
thousands. Real workloads are still a factor 10 less process intensive
than AIM7, but they are catching up.
This patch changes the way anon_vmas and VMAs are linked, which allows us
to associate multiple anon_vmas with a VMA. At fork time, each child
process gets its own anon_vmas, in which its COWed pages will be
instantiated. The parents' anon_vma is also linked to the VMA, because
non-COWed pages could be present in any of the children.
This reduces rmap scanning complexity to O(1) for the pages of the 1000
child processes, with O(N) complexity for at most 1/N pages in the system.
This reduces the average scanning cost in heavily forking workloads from
O(N) to 2.
The only real complexity in this patch stems from the fact that linking a
VMA to anon_vmas now involves memory allocations. This means vma_adjust
can fail, if it needs to attach a VMA to anon_vma structures. This in
turn means error handling needs to be added to the calling functions.
A second source of complexity is that, because there can be multiple
anon_vmas, the anon_vma linking in vma_adjust can no longer be done under
"the" anon_vma lock. To prevent the rmap code from walking up an
incomplete VMA, this patch introduces the VM_LOCK_RMAP VMA flag. This bit
flag uses the same slot as the NOMMU VM_MAPPED_COPY, with an ifdef in mm.h
to make sure it is impossible to compile a kernel that needs both symbolic
values for the same bitflag.
Some test results:
Without the anon_vma changes, when AIM7 hits around 9.7k users (on a test
box with 16GB RAM and not quite enough IO), the system ends up running
>99% in system time, with every CPU on the same anon_vma lock in the
pageout code.
With these changes, AIM7 hits the cross-over point around 29.7k users.
This happens with ~99% IO wait time, there never seems to be any spike in
system time. The anon_vma lock contention appears to be resolved.
[akpm@linux-foundation.org: cleanups]
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-05 14:42:07 -07:00
|
|
|
|
2014-06-04 17:11:02 -06:00
|
|
|
if (!t)
|
2009-09-16 03:50:15 -06:00
|
|
|
continue;
|
mm anon rmap: replace same_anon_vma linked list with an interval tree.
When a large VMA (anon or private file mapping) is first touched, which
will populate its anon_vma field, and then split into many regions through
the use of mprotect(), the original anon_vma ends up linking all of the
vmas on a linked list. This can cause rmap to become inefficient, as we
have to walk potentially thousands of irrelevent vmas before finding the
one a given anon page might fall into.
By replacing the same_anon_vma linked list with an interval tree (where
each avc's interval is determined by its vma's start and last pgoffs), we
can make rmap efficient for this use case again.
While the change is large, all of its pieces are fairly simple.
Most places that were walking the same_anon_vma list were looking for a
known pgoff, so they can just use the anon_vma_interval_tree_foreach()
interval tree iterator instead. The exception here is ksm, where the
page's index is not known. It would probably be possible to rework ksm so
that the index would be known, but for now I have decided to keep things
simple and just walk the entirety of the interval tree there.
When updating vma's that already have an anon_vma assigned, we must take
care to re-index the corresponding avc's on their interval tree. This is
done through the use of anon_vma_interval_tree_pre_update_vma() and
anon_vma_interval_tree_post_update_vma(), which remove the avc's from
their interval tree before the update and re-insert them after the update.
The anon_vma stays locked during the update, so there is no chance that
rmap would miss the vmas that are being updated.
Signed-off-by: Michel Lespinasse <walken@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Daniel Santos <daniel.santos@pobox.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-10-08 17:31:39 -06:00
|
|
|
anon_vma_interval_tree_foreach(vmac, &av->rb_root,
|
|
|
|
pgoff, pgoff) {
|
mm: change anon_vma linking to fix multi-process server scalability issue
The old anon_vma code can lead to scalability issues with heavily forking
workloads. Specifically, each anon_vma will be shared between the parent
process and all its child processes.
In a workload with 1000 child processes and a VMA with 1000 anonymous
pages per process that get COWed, this leads to a system with a million
anonymous pages in the same anon_vma, each of which is mapped in just one
of the 1000 processes. However, the current rmap code needs to walk them
all, leading to O(N) scanning complexity for each page.
This can result in systems where one CPU is walking the page tables of
1000 processes in page_referenced_one, while all other CPUs are stuck on
the anon_vma lock. This leads to catastrophic failure for a benchmark
like AIM7, where the total number of processes can reach in the tens of
thousands. Real workloads are still a factor 10 less process intensive
than AIM7, but they are catching up.
This patch changes the way anon_vmas and VMAs are linked, which allows us
to associate multiple anon_vmas with a VMA. At fork time, each child
process gets its own anon_vmas, in which its COWed pages will be
instantiated. The parents' anon_vma is also linked to the VMA, because
non-COWed pages could be present in any of the children.
This reduces rmap scanning complexity to O(1) for the pages of the 1000
child processes, with O(N) complexity for at most 1/N pages in the system.
This reduces the average scanning cost in heavily forking workloads from
O(N) to 2.
The only real complexity in this patch stems from the fact that linking a
VMA to anon_vmas now involves memory allocations. This means vma_adjust
can fail, if it needs to attach a VMA to anon_vma structures. This in
turn means error handling needs to be added to the calling functions.
A second source of complexity is that, because there can be multiple
anon_vmas, the anon_vma linking in vma_adjust can no longer be done under
"the" anon_vma lock. To prevent the rmap code from walking up an
incomplete VMA, this patch introduces the VM_LOCK_RMAP VMA flag. This bit
flag uses the same slot as the NOMMU VM_MAPPED_COPY, with an ifdef in mm.h
to make sure it is impossible to compile a kernel that needs both symbolic
values for the same bitflag.
Some test results:
Without the anon_vma changes, when AIM7 hits around 9.7k users (on a test
box with 16GB RAM and not quite enough IO), the system ends up running
>99% in system time, with every CPU on the same anon_vma lock in the
pageout code.
With these changes, AIM7 hits the cross-over point around 29.7k users.
This happens with ~99% IO wait time, there never seems to be any spike in
system time. The anon_vma lock contention appears to be resolved.
[akpm@linux-foundation.org: cleanups]
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-05 14:42:07 -07:00
|
|
|
vma = vmac->vma;
|
2009-09-16 03:50:15 -06:00
|
|
|
if (!page_mapped_in_vma(page, vma))
|
|
|
|
continue;
|
2014-06-04 17:11:02 -06:00
|
|
|
if (vma->vm_mm == t->mm)
|
|
|
|
add_to_kill(t, page, vma, to_kill, tkc);
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
read_unlock(&tasklist_lock);
|
2012-12-02 12:56:50 -07:00
|
|
|
page_unlock_anon_vma_read(av);
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Collect processes when the error hit a file mapped page.
|
|
|
|
*/
|
|
|
|
static void collect_procs_file(struct page *page, struct list_head *to_kill,
|
2014-06-04 17:11:01 -06:00
|
|
|
struct to_kill **tkc, int force_early)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
struct vm_area_struct *vma;
|
|
|
|
struct task_struct *tsk;
|
|
|
|
struct address_space *mapping = page->mapping;
|
|
|
|
|
2011-05-24 18:12:06 -06:00
|
|
|
mutex_lock(&mapping->i_mmap_mutex);
|
2011-06-27 17:18:09 -06:00
|
|
|
read_lock(&tasklist_lock);
|
2009-09-16 03:50:15 -06:00
|
|
|
for_each_process(tsk) {
|
2014-07-23 15:00:01 -06:00
|
|
|
pgoff_t pgoff = page_to_pgoff(page);
|
2014-06-04 17:11:02 -06:00
|
|
|
struct task_struct *t = task_early_kill(tsk, force_early);
|
2009-09-16 03:50:15 -06:00
|
|
|
|
2014-06-04 17:11:02 -06:00
|
|
|
if (!t)
|
2009-09-16 03:50:15 -06:00
|
|
|
continue;
|
2012-10-08 17:31:25 -06:00
|
|
|
vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff,
|
2009-09-16 03:50:15 -06:00
|
|
|
pgoff) {
|
|
|
|
/*
|
|
|
|
* Send early kill signal to tasks where a vma covers
|
|
|
|
* the page but the corrupted page is not necessarily
|
|
|
|
* mapped it in its pte.
|
|
|
|
* Assume applications who requested early kill want
|
|
|
|
* to be informed of all such data corruptions.
|
|
|
|
*/
|
2014-06-04 17:11:02 -06:00
|
|
|
if (vma->vm_mm == t->mm)
|
|
|
|
add_to_kill(t, page, vma, to_kill, tkc);
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
read_unlock(&tasklist_lock);
|
2011-06-27 17:18:09 -06:00
|
|
|
mutex_unlock(&mapping->i_mmap_mutex);
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Collect the processes who have the corrupted page mapped to kill.
|
|
|
|
* This is done in two steps for locking reasons.
|
|
|
|
* First preallocate one tokill structure outside the spin locks,
|
|
|
|
* so that we can kill at least one process reasonably reliable.
|
|
|
|
*/
|
2014-06-04 17:11:01 -06:00
|
|
|
static void collect_procs(struct page *page, struct list_head *tokill,
|
|
|
|
int force_early)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
struct to_kill *tk;
|
|
|
|
|
|
|
|
if (!page->mapping)
|
|
|
|
return;
|
|
|
|
|
|
|
|
tk = kmalloc(sizeof(struct to_kill), GFP_NOIO);
|
|
|
|
if (!tk)
|
|
|
|
return;
|
|
|
|
if (PageAnon(page))
|
2014-06-04 17:11:01 -06:00
|
|
|
collect_procs_anon(page, tokill, &tk, force_early);
|
2009-09-16 03:50:15 -06:00
|
|
|
else
|
2014-06-04 17:11:01 -06:00
|
|
|
collect_procs_file(page, tokill, &tk, force_early);
|
2009-09-16 03:50:15 -06:00
|
|
|
kfree(tk);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Error handlers for various types of pages.
|
|
|
|
*/
|
|
|
|
|
|
|
|
enum outcome {
|
2009-12-16 04:19:58 -07:00
|
|
|
IGNORED, /* Error: cannot be handled */
|
|
|
|
FAILED, /* Error: handling failed */
|
2009-09-16 03:50:15 -06:00
|
|
|
DELAYED, /* Will be handled later */
|
|
|
|
RECOVERED, /* Successfully recovered */
|
|
|
|
};
|
|
|
|
|
|
|
|
static const char *action_name[] = {
|
2009-12-16 04:19:58 -07:00
|
|
|
[IGNORED] = "Ignored",
|
2009-09-16 03:50:15 -06:00
|
|
|
[FAILED] = "Failed",
|
|
|
|
[DELAYED] = "Delayed",
|
|
|
|
[RECOVERED] = "Recovered",
|
|
|
|
};
|
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
/*
|
|
|
|
* XXX: It is possible that a page is isolated from LRU cache,
|
|
|
|
* and then kept in swap cache or failed to remove from page cache.
|
|
|
|
* The page count will stop it from being freed by unpoison.
|
|
|
|
* Stress tests should be aware of this memory leak problem.
|
|
|
|
*/
|
|
|
|
static int delete_from_lru_cache(struct page *p)
|
|
|
|
{
|
|
|
|
if (!isolate_lru_page(p)) {
|
|
|
|
/*
|
|
|
|
* Clear sensible page flags, so that the buddy system won't
|
|
|
|
* complain when the page is unpoison-and-freed.
|
|
|
|
*/
|
|
|
|
ClearPageActive(p);
|
|
|
|
ClearPageUnevictable(p);
|
|
|
|
/*
|
|
|
|
* drop the page count elevated by isolate_lru_page()
|
|
|
|
*/
|
|
|
|
page_cache_release(p);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
|
2009-09-16 03:50:15 -06:00
|
|
|
/*
|
|
|
|
* Error hit kernel page.
|
|
|
|
* Do nothing, try to be lucky and not touch this instead. For a few cases we
|
|
|
|
* could be more sophisticated.
|
|
|
|
*/
|
|
|
|
static int me_kernel(struct page *p, unsigned long pfn)
|
|
|
|
{
|
|
|
|
return IGNORED;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Page in unknown state. Do nothing.
|
|
|
|
*/
|
|
|
|
static int me_unknown(struct page *p, unsigned long pfn)
|
|
|
|
{
|
|
|
|
printk(KERN_ERR "MCE %#lx: Unknown page state\n", pfn);
|
|
|
|
return FAILED;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Clean (or cleaned) page cache page.
|
|
|
|
*/
|
|
|
|
static int me_pagecache_clean(struct page *p, unsigned long pfn)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
int ret = FAILED;
|
|
|
|
struct address_space *mapping;
|
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
delete_from_lru_cache(p);
|
|
|
|
|
2009-09-16 03:50:15 -06:00
|
|
|
/*
|
|
|
|
* For anonymous pages we're done the only reference left
|
|
|
|
* should be the one m_f() holds.
|
|
|
|
*/
|
|
|
|
if (PageAnon(p))
|
|
|
|
return RECOVERED;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Now truncate the page in the page cache. This is really
|
|
|
|
* more like a "temporary hole punch"
|
|
|
|
* Don't do this for block devices when someone else
|
|
|
|
* has a reference, because it could be file system metadata
|
|
|
|
* and that's not safe to truncate.
|
|
|
|
*/
|
|
|
|
mapping = page_mapping(p);
|
|
|
|
if (!mapping) {
|
|
|
|
/*
|
|
|
|
* Page has been teared down in the meanwhile
|
|
|
|
*/
|
|
|
|
return FAILED;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Truncation is a bit tricky. Enable it per file system for now.
|
|
|
|
*
|
|
|
|
* Open: to take i_mutex or not for this? Right now we don't.
|
|
|
|
*/
|
|
|
|
if (mapping->a_ops->error_remove_page) {
|
|
|
|
err = mapping->a_ops->error_remove_page(mapping, p);
|
|
|
|
if (err != 0) {
|
|
|
|
printk(KERN_INFO "MCE %#lx: Failed to punch page: %d\n",
|
|
|
|
pfn, err);
|
|
|
|
} else if (page_has_private(p) &&
|
|
|
|
!try_to_release_page(p, GFP_NOIO)) {
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("MCE %#lx: failed to release buffers\n", pfn);
|
2009-09-16 03:50:15 -06:00
|
|
|
} else {
|
|
|
|
ret = RECOVERED;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* If the file system doesn't support it just invalidate
|
|
|
|
* This fails on dirty or anything with private pages
|
|
|
|
*/
|
|
|
|
if (invalidate_inode_page(p))
|
|
|
|
ret = RECOVERED;
|
|
|
|
else
|
|
|
|
printk(KERN_INFO "MCE %#lx: Failed to invalidate\n",
|
|
|
|
pfn);
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2014-01-21 16:49:08 -07:00
|
|
|
* Dirty pagecache page
|
2009-09-16 03:50:15 -06:00
|
|
|
* Issues: when the error hit a hole page the error is not properly
|
|
|
|
* propagated.
|
|
|
|
*/
|
|
|
|
static int me_pagecache_dirty(struct page *p, unsigned long pfn)
|
|
|
|
{
|
|
|
|
struct address_space *mapping = page_mapping(p);
|
|
|
|
|
|
|
|
SetPageError(p);
|
|
|
|
/* TBD: print more information about the file. */
|
|
|
|
if (mapping) {
|
|
|
|
/*
|
|
|
|
* IO error will be reported by write(), fsync(), etc.
|
|
|
|
* who check the mapping.
|
|
|
|
* This way the application knows that something went
|
|
|
|
* wrong with its dirty file data.
|
|
|
|
*
|
|
|
|
* There's one open issue:
|
|
|
|
*
|
|
|
|
* The EIO will be only reported on the next IO
|
|
|
|
* operation and then cleared through the IO map.
|
|
|
|
* Normally Linux has two mechanisms to pass IO error
|
|
|
|
* first through the AS_EIO flag in the address space
|
|
|
|
* and then through the PageError flag in the page.
|
|
|
|
* Since we drop pages on memory failure handling the
|
|
|
|
* only mechanism open to use is through AS_AIO.
|
|
|
|
*
|
|
|
|
* This has the disadvantage that it gets cleared on
|
|
|
|
* the first operation that returns an error, while
|
|
|
|
* the PageError bit is more sticky and only cleared
|
|
|
|
* when the page is reread or dropped. If an
|
|
|
|
* application assumes it will always get error on
|
|
|
|
* fsync, but does other operations on the fd before
|
2011-03-30 19:57:33 -06:00
|
|
|
* and the page is dropped between then the error
|
2009-09-16 03:50:15 -06:00
|
|
|
* will not be properly reported.
|
|
|
|
*
|
|
|
|
* This can already happen even without hwpoisoned
|
|
|
|
* pages: first on metadata IO errors (which only
|
|
|
|
* report through AS_EIO) or when the page is dropped
|
|
|
|
* at the wrong time.
|
|
|
|
*
|
|
|
|
* So right now we assume that the application DTRT on
|
|
|
|
* the first EIO, but we're not worse than other parts
|
|
|
|
* of the kernel.
|
|
|
|
*/
|
|
|
|
mapping_set_error(mapping, EIO);
|
|
|
|
}
|
|
|
|
|
|
|
|
return me_pagecache_clean(p, pfn);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Clean and dirty swap cache.
|
|
|
|
*
|
|
|
|
* Dirty swap cache page is tricky to handle. The page could live both in page
|
|
|
|
* cache and swap cache(ie. page is freshly swapped in). So it could be
|
|
|
|
* referenced concurrently by 2 types of PTEs:
|
|
|
|
* normal PTEs and swap PTEs. We try to handle them consistently by calling
|
|
|
|
* try_to_unmap(TTU_IGNORE_HWPOISON) to convert the normal PTEs to swap PTEs,
|
|
|
|
* and then
|
|
|
|
* - clear dirty bit to prevent IO
|
|
|
|
* - remove from LRU
|
|
|
|
* - but keep in the swap cache, so that when we return to it on
|
|
|
|
* a later page fault, we know the application is accessing
|
|
|
|
* corrupted data and shall be killed (we installed simple
|
|
|
|
* interception code in do_swap_page to catch it).
|
|
|
|
*
|
|
|
|
* Clean swap cache pages can be directly isolated. A later page fault will
|
|
|
|
* bring in the known good data from disk.
|
|
|
|
*/
|
|
|
|
static int me_swapcache_dirty(struct page *p, unsigned long pfn)
|
|
|
|
{
|
|
|
|
ClearPageDirty(p);
|
|
|
|
/* Trigger EIO in shmem: */
|
|
|
|
ClearPageUptodate(p);
|
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
if (!delete_from_lru_cache(p))
|
|
|
|
return DELAYED;
|
|
|
|
else
|
|
|
|
return FAILED;
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static int me_swapcache_clean(struct page *p, unsigned long pfn)
|
|
|
|
{
|
|
|
|
delete_from_swap_cache(p);
|
2009-09-28 23:16:20 -06:00
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
if (!delete_from_lru_cache(p))
|
|
|
|
return RECOVERED;
|
|
|
|
else
|
|
|
|
return FAILED;
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Huge pages. Needs work.
|
|
|
|
* Issues:
|
2010-05-27 18:29:20 -06:00
|
|
|
* - Error on hugepage is contained in hugepage unit (not in raw page unit.)
|
|
|
|
* To narrow down kill region to one page, we need to break up pmd.
|
2009-09-16 03:50:15 -06:00
|
|
|
*/
|
|
|
|
static int me_huge_page(struct page *p, unsigned long pfn)
|
|
|
|
{
|
2010-09-07 19:19:36 -06:00
|
|
|
int res = 0;
|
2010-05-27 18:29:20 -06:00
|
|
|
struct page *hpage = compound_head(p);
|
|
|
|
/*
|
|
|
|
* We can safely recover from error on free or reserved (i.e.
|
|
|
|
* not in-use) hugepage by dequeuing it from freelist.
|
|
|
|
* To check whether a hugepage is in-use or not, we can't use
|
|
|
|
* page->lru because it can be used in other hugepage operations,
|
|
|
|
* such as __unmap_hugepage_range() and gather_surplus_pages().
|
|
|
|
* So instead we use page_mapping() and PageAnon().
|
|
|
|
* We assume that this function is called with page lock held,
|
|
|
|
* so there is no race between isolation and mapping/unmapping.
|
|
|
|
*/
|
|
|
|
if (!(page_mapping(hpage) || PageAnon(hpage))) {
|
2010-09-07 19:19:36 -06:00
|
|
|
res = dequeue_hwpoisoned_huge_page(hpage);
|
|
|
|
if (!res)
|
|
|
|
return RECOVERED;
|
2010-05-27 18:29:20 -06:00
|
|
|
}
|
|
|
|
return DELAYED;
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Various page states we can handle.
|
|
|
|
*
|
|
|
|
* A page state is defined by its current page->flags bits.
|
|
|
|
* The table matches them in order and calls the right handler.
|
|
|
|
*
|
|
|
|
* This is quite tricky because we can access page at any time
|
2011-03-30 19:57:33 -06:00
|
|
|
* in its live cycle, so all accesses have to be extremely careful.
|
2009-09-16 03:50:15 -06:00
|
|
|
*
|
|
|
|
* This is not complete. More states could be added.
|
|
|
|
* For any missing state don't attempt recovery.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#define dirty (1UL << PG_dirty)
|
|
|
|
#define sc (1UL << PG_swapcache)
|
|
|
|
#define unevict (1UL << PG_unevictable)
|
|
|
|
#define mlock (1UL << PG_mlocked)
|
|
|
|
#define writeback (1UL << PG_writeback)
|
|
|
|
#define lru (1UL << PG_lru)
|
|
|
|
#define swapbacked (1UL << PG_swapbacked)
|
|
|
|
#define head (1UL << PG_head)
|
|
|
|
#define tail (1UL << PG_tail)
|
|
|
|
#define compound (1UL << PG_compound)
|
|
|
|
#define slab (1UL << PG_slab)
|
|
|
|
#define reserved (1UL << PG_reserved)
|
|
|
|
|
|
|
|
static struct page_state {
|
|
|
|
unsigned long mask;
|
|
|
|
unsigned long res;
|
|
|
|
char *msg;
|
|
|
|
int (*action)(struct page *p, unsigned long pfn);
|
|
|
|
} error_states[] = {
|
2009-12-16 04:19:58 -07:00
|
|
|
{ reserved, reserved, "reserved kernel", me_kernel },
|
2009-12-16 04:19:58 -07:00
|
|
|
/*
|
|
|
|
* free pages are specially detected outside this table:
|
|
|
|
* PG_buddy pages only make a small fraction of all free pages.
|
|
|
|
*/
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Could in theory check if slab page is free or if we can drop
|
|
|
|
* currently unused objects without touching them. But just
|
|
|
|
* treat it as standard kernel for now.
|
|
|
|
*/
|
|
|
|
{ slab, slab, "kernel slab", me_kernel },
|
|
|
|
|
|
|
|
#ifdef CONFIG_PAGEFLAGS_EXTENDED
|
|
|
|
{ head, head, "huge", me_huge_page },
|
|
|
|
{ tail, tail, "huge", me_huge_page },
|
|
|
|
#else
|
|
|
|
{ compound, compound, "huge", me_huge_page },
|
|
|
|
#endif
|
|
|
|
|
2012-12-11 17:01:32 -07:00
|
|
|
{ sc|dirty, sc|dirty, "dirty swapcache", me_swapcache_dirty },
|
|
|
|
{ sc|dirty, sc, "clean swapcache", me_swapcache_clean },
|
2009-09-16 03:50:15 -06:00
|
|
|
|
2012-12-11 17:01:32 -07:00
|
|
|
{ mlock|dirty, mlock|dirty, "dirty mlocked LRU", me_pagecache_dirty },
|
2013-04-29 16:06:08 -06:00
|
|
|
{ mlock|dirty, mlock, "clean mlocked LRU", me_pagecache_clean },
|
2009-09-16 03:50:15 -06:00
|
|
|
|
2013-02-22 17:35:53 -07:00
|
|
|
{ unevict|dirty, unevict|dirty, "dirty unevictable LRU", me_pagecache_dirty },
|
2013-04-29 16:06:08 -06:00
|
|
|
{ unevict|dirty, unevict, "clean unevictable LRU", me_pagecache_clean },
|
2013-02-22 17:35:53 -07:00
|
|
|
|
2012-12-11 17:01:32 -07:00
|
|
|
{ lru|dirty, lru|dirty, "dirty LRU", me_pagecache_dirty },
|
2009-09-16 03:50:15 -06:00
|
|
|
{ lru|dirty, lru, "clean LRU", me_pagecache_clean },
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Catchall entry: must be at end.
|
|
|
|
*/
|
|
|
|
{ 0, 0, "unknown page state", me_unknown },
|
|
|
|
};
|
|
|
|
|
2009-12-16 04:20:00 -07:00
|
|
|
#undef dirty
|
|
|
|
#undef sc
|
|
|
|
#undef unevict
|
|
|
|
#undef mlock
|
|
|
|
#undef writeback
|
|
|
|
#undef lru
|
|
|
|
#undef swapbacked
|
|
|
|
#undef head
|
|
|
|
#undef tail
|
|
|
|
#undef compound
|
|
|
|
#undef slab
|
|
|
|
#undef reserved
|
|
|
|
|
2012-12-11 17:01:32 -07:00
|
|
|
/*
|
|
|
|
* "Dirty/Clean" indication is not 100% accurate due to the possibility of
|
|
|
|
* setting PG_dirty outside page lock. See also comment above set_page_dirty().
|
|
|
|
*/
|
2009-09-16 03:50:15 -06:00
|
|
|
static void action_result(unsigned long pfn, char *msg, int result)
|
|
|
|
{
|
2012-12-11 17:01:32 -07:00
|
|
|
pr_err("MCE %#lx: %s page recovery: %s\n",
|
|
|
|
pfn, msg, action_name[result]);
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
static int page_action(struct page_state *ps, struct page *p,
|
2009-12-16 04:19:57 -07:00
|
|
|
unsigned long pfn)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
int result;
|
2009-10-19 00:15:01 -06:00
|
|
|
int count;
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
result = ps->action(p, pfn);
|
|
|
|
action_result(pfn, ps->msg, result);
|
2009-10-19 00:15:01 -06:00
|
|
|
|
2009-12-16 04:19:57 -07:00
|
|
|
count = page_count(p) - 1;
|
2009-12-16 04:19:58 -07:00
|
|
|
if (ps->action == me_swapcache_dirty && result == DELAYED)
|
|
|
|
count--;
|
|
|
|
if (count != 0) {
|
2009-09-16 03:50:15 -06:00
|
|
|
printk(KERN_ERR
|
|
|
|
"MCE %#lx: %s page still referenced by %d users\n",
|
2009-10-19 00:15:01 -06:00
|
|
|
pfn, ps->msg, count);
|
2009-12-16 04:19:58 -07:00
|
|
|
result = FAILED;
|
|
|
|
}
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
/* Could do more checks here if page looks ok */
|
|
|
|
/*
|
|
|
|
* Could adjust zone counters here to correct for the missing page.
|
|
|
|
*/
|
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
return (result == RECOVERED || result == DELAYED) ? 0 : -EBUSY;
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Do all that is necessary to remove user space mappings. Unmap
|
|
|
|
* the pages and send SIGBUS to the processes if the data was dirty.
|
|
|
|
*/
|
2009-12-16 04:19:58 -07:00
|
|
|
static int hwpoison_user_mappings(struct page *p, unsigned long pfn,
|
2014-01-23 16:53:14 -07:00
|
|
|
int trapno, int flags, struct page **hpagep)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
enum ttu_flags ttu = TTU_UNMAP | TTU_IGNORE_MLOCK | TTU_IGNORE_ACCESS;
|
|
|
|
struct address_space *mapping;
|
|
|
|
LIST_HEAD(tokill);
|
|
|
|
int ret;
|
2012-07-11 11:20:47 -06:00
|
|
|
int kill = 1, forcekill;
|
2014-01-23 16:53:14 -07:00
|
|
|
struct page *hpage = *hpagep;
|
2011-02-01 16:52:40 -07:00
|
|
|
struct page *ppage;
|
2009-09-16 03:50:15 -06:00
|
|
|
|
2014-07-30 17:08:28 -06:00
|
|
|
/*
|
|
|
|
* Here we are interested only in user-mapped pages, so skip any
|
|
|
|
* other types of pages.
|
|
|
|
*/
|
|
|
|
if (PageReserved(p) || PageSlab(p))
|
|
|
|
return SWAP_SUCCESS;
|
|
|
|
if (!(PageLRU(hpage) || PageHuge(p)))
|
2009-12-16 04:19:58 -07:00
|
|
|
return SWAP_SUCCESS;
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* This check implies we don't kill processes if their pages
|
|
|
|
* are in the swap cache early. Those are always late kills.
|
|
|
|
*/
|
2010-05-27 18:29:17 -06:00
|
|
|
if (!page_mapped(hpage))
|
2009-12-16 04:19:58 -07:00
|
|
|
return SWAP_SUCCESS;
|
|
|
|
|
2014-07-30 17:08:30 -06:00
|
|
|
if (PageKsm(p)) {
|
|
|
|
pr_err("MCE %#lx: can't handle KSM pages.\n", pfn);
|
2009-12-16 04:19:58 -07:00
|
|
|
return SWAP_FAIL;
|
2014-07-30 17:08:30 -06:00
|
|
|
}
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
if (PageSwapCache(p)) {
|
|
|
|
printk(KERN_ERR
|
|
|
|
"MCE %#lx: keeping poisoned page in swap cache\n", pfn);
|
|
|
|
ttu |= TTU_IGNORE_HWPOISON;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Propagate the dirty bit from PTEs to struct page first, because we
|
|
|
|
* need this to decide if we should kill or just drop the page.
|
2009-12-16 04:19:58 -07:00
|
|
|
* XXX: the dirty test could be racy: set_page_dirty() may not always
|
|
|
|
* be called inside page lock (it's recommended but not enforced).
|
2009-09-16 03:50:15 -06:00
|
|
|
*/
|
2010-05-27 18:29:17 -06:00
|
|
|
mapping = page_mapping(hpage);
|
2012-07-11 11:20:47 -06:00
|
|
|
if (!(flags & MF_MUST_KILL) && !PageDirty(hpage) && mapping &&
|
2010-05-27 18:29:17 -06:00
|
|
|
mapping_cap_writeback_dirty(mapping)) {
|
|
|
|
if (page_mkclean(hpage)) {
|
|
|
|
SetPageDirty(hpage);
|
2009-09-16 03:50:15 -06:00
|
|
|
} else {
|
|
|
|
kill = 0;
|
|
|
|
ttu |= TTU_IGNORE_HWPOISON;
|
|
|
|
printk(KERN_INFO
|
|
|
|
"MCE %#lx: corrupted page was clean: dropped without side effects\n",
|
|
|
|
pfn);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-02-01 16:52:40 -07:00
|
|
|
/*
|
|
|
|
* ppage: poisoned page
|
|
|
|
* if p is regular page(4k page)
|
|
|
|
* ppage == real poisoned page;
|
|
|
|
* else p is hugetlb or THP, ppage == head page.
|
|
|
|
*/
|
|
|
|
ppage = hpage;
|
|
|
|
|
2011-02-01 16:52:39 -07:00
|
|
|
if (PageTransHuge(hpage)) {
|
|
|
|
/*
|
|
|
|
* Verify that this isn't a hugetlbfs head page, the check for
|
|
|
|
* PageAnon is just for avoid tripping a split_huge_page
|
|
|
|
* internal debug check, as split_huge_page refuses to deal with
|
|
|
|
* anything that isn't an anon page. PageAnon can't go away fro
|
|
|
|
* under us because we hold a refcount on the hpage, without a
|
|
|
|
* refcount on the hpage. split_huge_page can't be safely called
|
|
|
|
* in the first place, having a refcount on the tail isn't
|
|
|
|
* enough * to be safe.
|
|
|
|
*/
|
|
|
|
if (!PageHuge(hpage) && PageAnon(hpage)) {
|
|
|
|
if (unlikely(split_huge_page(hpage))) {
|
|
|
|
/*
|
|
|
|
* FIXME: if splitting THP is failed, it is
|
|
|
|
* better to stop the following operation rather
|
|
|
|
* than causing panic by unmapping. System might
|
|
|
|
* survive if the page is freed later.
|
|
|
|
*/
|
|
|
|
printk(KERN_INFO
|
|
|
|
"MCE %#lx: failed to split THP\n", pfn);
|
|
|
|
|
|
|
|
BUG_ON(!PageHWPoison(p));
|
|
|
|
return SWAP_FAIL;
|
|
|
|
}
|
2014-01-02 13:58:51 -07:00
|
|
|
/*
|
|
|
|
* We pinned the head page for hwpoison handling,
|
|
|
|
* now we split the thp and we are interested in
|
|
|
|
* the hwpoisoned raw page, so move the refcount
|
2014-01-23 16:53:14 -07:00
|
|
|
* to it. Similarly, page lock is shifted.
|
2014-01-02 13:58:51 -07:00
|
|
|
*/
|
|
|
|
if (hpage != p) {
|
2014-02-10 15:25:50 -07:00
|
|
|
if (!(flags & MF_COUNT_INCREASED)) {
|
|
|
|
put_page(hpage);
|
|
|
|
get_page(p);
|
|
|
|
}
|
2014-01-23 16:53:14 -07:00
|
|
|
lock_page(p);
|
|
|
|
unlock_page(hpage);
|
|
|
|
*hpagep = p;
|
2014-01-02 13:58:51 -07:00
|
|
|
}
|
2011-02-01 16:52:40 -07:00
|
|
|
/* THP is split, so ppage should be the real poisoned page. */
|
|
|
|
ppage = p;
|
2011-02-01 16:52:39 -07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-09-16 03:50:15 -06:00
|
|
|
/*
|
|
|
|
* First collect all the processes that have the page
|
|
|
|
* mapped in dirty form. This has to be done before try_to_unmap,
|
|
|
|
* because ttu takes the rmap data structures down.
|
|
|
|
*
|
|
|
|
* Error handling: We ignore errors here because
|
|
|
|
* there's nothing that can be done.
|
|
|
|
*/
|
|
|
|
if (kill)
|
2014-06-04 17:11:01 -06:00
|
|
|
collect_procs(ppage, &tokill, flags & MF_ACTION_REQUIRED);
|
2009-09-16 03:50:15 -06:00
|
|
|
|
2011-02-01 16:52:40 -07:00
|
|
|
ret = try_to_unmap(ppage, ttu);
|
2009-09-16 03:50:15 -06:00
|
|
|
if (ret != SWAP_SUCCESS)
|
|
|
|
printk(KERN_ERR "MCE %#lx: failed to unmap page (mapcount=%d)\n",
|
2011-02-01 16:52:40 -07:00
|
|
|
pfn, page_mapcount(ppage));
|
|
|
|
|
2009-09-16 03:50:15 -06:00
|
|
|
/*
|
|
|
|
* Now that the dirty bit has been propagated to the
|
|
|
|
* struct page and all unmaps done we can decide if
|
|
|
|
* killing is needed or not. Only kill when the page
|
2012-07-11 11:20:47 -06:00
|
|
|
* was dirty or the process is not restartable,
|
|
|
|
* otherwise the tokill list is merely
|
2009-09-16 03:50:15 -06:00
|
|
|
* freed. When there was a problem unmapping earlier
|
|
|
|
* use a more force-full uncatchable kill to prevent
|
|
|
|
* any accesses to the poisoned memory.
|
|
|
|
*/
|
2012-07-11 11:20:47 -06:00
|
|
|
forcekill = PageDirty(ppage) || (flags & MF_MUST_KILL);
|
|
|
|
kill_procs(&tokill, forcekill, trapno,
|
2011-12-13 10:27:58 -07:00
|
|
|
ret != SWAP_SUCCESS, p, pfn, flags);
|
2009-12-16 04:19:58 -07:00
|
|
|
|
|
|
|
return ret;
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
2010-05-27 18:29:18 -06:00
|
|
|
static void set_page_hwpoison_huge_page(struct page *hpage)
|
|
|
|
{
|
|
|
|
int i;
|
2013-09-11 15:22:52 -06:00
|
|
|
int nr_pages = 1 << compound_order(hpage);
|
2010-05-27 18:29:18 -06:00
|
|
|
for (i = 0; i < nr_pages; i++)
|
|
|
|
SetPageHWPoison(hpage + i);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void clear_page_hwpoison_huge_page(struct page *hpage)
|
|
|
|
{
|
|
|
|
int i;
|
2013-09-11 15:22:52 -06:00
|
|
|
int nr_pages = 1 << compound_order(hpage);
|
2010-05-27 18:29:18 -06:00
|
|
|
for (i = 0; i < nr_pages; i++)
|
|
|
|
ClearPageHWPoison(hpage + i);
|
|
|
|
}
|
|
|
|
|
2011-12-15 11:48:12 -07:00
|
|
|
/**
|
|
|
|
* memory_failure - Handle memory failure of a page.
|
|
|
|
* @pfn: Page Number of the corrupted page
|
|
|
|
* @trapno: Trap number reported in the signal to user space.
|
|
|
|
* @flags: fine tune action taken
|
|
|
|
*
|
|
|
|
* This function is called by the low level machine check code
|
|
|
|
* of an architecture when it detects hardware memory corruption
|
|
|
|
* of a page. It tries its best to recover, which includes
|
|
|
|
* dropping pages, killing processes etc.
|
|
|
|
*
|
|
|
|
* The function is primarily of use for corruptions that
|
|
|
|
* happen outside the current execution context (e.g. when
|
|
|
|
* detected by a background scrubber)
|
|
|
|
*
|
|
|
|
* Must run in process context (e.g. a work queue) with interrupts
|
|
|
|
* enabled and no spinlocks hold.
|
|
|
|
*/
|
|
|
|
int memory_failure(unsigned long pfn, int trapno, int flags)
|
2009-09-16 03:50:15 -06:00
|
|
|
{
|
|
|
|
struct page_state *ps;
|
|
|
|
struct page *p;
|
2010-05-27 18:29:17 -06:00
|
|
|
struct page *hpage;
|
2009-09-16 03:50:15 -06:00
|
|
|
int res;
|
2010-05-27 18:29:19 -06:00
|
|
|
unsigned int nr_pages;
|
2013-02-22 17:35:51 -07:00
|
|
|
unsigned long page_flags;
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
if (!sysctl_memory_failure_recovery)
|
|
|
|
panic("Memory failure from trap %d on page %lx", trapno, pfn);
|
|
|
|
|
|
|
|
if (!pfn_valid(pfn)) {
|
2009-12-16 04:19:57 -07:00
|
|
|
printk(KERN_ERR
|
|
|
|
"MCE %#lx: memory outside kernel control\n",
|
|
|
|
pfn);
|
|
|
|
return -ENXIO;
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
p = pfn_to_page(pfn);
|
2010-05-27 18:29:17 -06:00
|
|
|
hpage = compound_head(p);
|
2009-09-16 03:50:15 -06:00
|
|
|
if (TestSetPageHWPoison(p)) {
|
2009-12-16 04:19:58 -07:00
|
|
|
printk(KERN_ERR "MCE %#lx: already hardware poisoned\n", pfn);
|
2009-09-16 03:50:15 -06:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-02-22 17:34:05 -07:00
|
|
|
/*
|
|
|
|
* Currently errors on hugetlbfs pages are measured in hugepage units,
|
|
|
|
* so nr_pages should be 1 << compound_order. OTOH when errors are on
|
|
|
|
* transparent hugepages, they are supposed to be split and error
|
|
|
|
* measurement is done in normal page units. So nr_pages should be one
|
|
|
|
* in this case.
|
|
|
|
*/
|
|
|
|
if (PageHuge(p))
|
|
|
|
nr_pages = 1 << compound_order(hpage);
|
|
|
|
else /* normal page or thp */
|
|
|
|
nr_pages = 1;
|
2013-02-22 17:34:02 -07:00
|
|
|
atomic_long_add(nr_pages, &num_poisoned_pages);
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* We need/can do nothing about count=0 pages.
|
|
|
|
* 1) it's a free page, and therefore in safe hand:
|
|
|
|
* prep_new_page() will be the gate keeper.
|
2010-09-07 19:19:38 -06:00
|
|
|
* 2) it's a free hugepage, which is also safe:
|
|
|
|
* an affected hugepage will be dequeued from hugepage freelist,
|
|
|
|
* so there's no concern about reusing it ever after.
|
|
|
|
* 3) it's part of a non-compound high order page.
|
2009-09-16 03:50:15 -06:00
|
|
|
* Implies some kernel user: cannot stop them from
|
|
|
|
* R/W the page; let's pray that the page has been
|
|
|
|
* used and will be freed some time later.
|
|
|
|
* In fact it's dangerous to directly bump up page count from 0,
|
|
|
|
* that may make page_freeze_refs()/page_unfreeze_refs() mismatch.
|
|
|
|
*/
|
2009-12-16 04:19:57 -07:00
|
|
|
if (!(flags & MF_COUNT_INCREASED) &&
|
2010-05-27 18:29:17 -06:00
|
|
|
!get_page_unless_zero(hpage)) {
|
2009-12-16 04:19:58 -07:00
|
|
|
if (is_free_buddy_page(p)) {
|
|
|
|
action_result(pfn, "free buddy", DELAYED);
|
|
|
|
return 0;
|
2010-09-07 19:19:38 -06:00
|
|
|
} else if (PageHuge(hpage)) {
|
|
|
|
/*
|
2014-05-22 12:54:15 -06:00
|
|
|
* Check "filter hit" and "race with other subpage."
|
2010-09-07 19:19:38 -06:00
|
|
|
*/
|
2011-03-10 00:52:07 -07:00
|
|
|
lock_page(hpage);
|
2014-05-22 12:54:15 -06:00
|
|
|
if (PageHWPoison(hpage)) {
|
|
|
|
if ((hwpoison_filter(p) && TestClearPageHWPoison(p))
|
|
|
|
|| (p != hpage && TestSetPageHWPoison(hpage))) {
|
|
|
|
atomic_long_sub(nr_pages, &num_poisoned_pages);
|
|
|
|
unlock_page(hpage);
|
|
|
|
return 0;
|
|
|
|
}
|
2010-09-07 19:19:38 -06:00
|
|
|
}
|
|
|
|
set_page_hwpoison_huge_page(hpage);
|
|
|
|
res = dequeue_hwpoisoned_huge_page(hpage);
|
|
|
|
action_result(pfn, "free huge",
|
|
|
|
res ? IGNORED : DELAYED);
|
|
|
|
unlock_page(hpage);
|
|
|
|
return res;
|
2009-12-16 04:19:58 -07:00
|
|
|
} else {
|
|
|
|
action_result(pfn, "high order kernel", IGNORED);
|
|
|
|
return -EBUSY;
|
|
|
|
}
|
2009-09-16 03:50:15 -06:00
|
|
|
}
|
|
|
|
|
2009-09-28 23:16:20 -06:00
|
|
|
/*
|
|
|
|
* We ignore non-LRU pages for good reasons.
|
|
|
|
* - PG_locked is only well defined for LRU pages and a few others
|
|
|
|
* - to avoid races with __set_page_locked()
|
|
|
|
* - to avoid races with __SetPageSlab*() (and more non-atomic ops)
|
|
|
|
* The check (unnecessarily) ignores LRU pages being isolated and
|
|
|
|
* walked by the page reclaim code, however that's not a big loss.
|
|
|
|
*/
|
2012-03-21 17:34:05 -06:00
|
|
|
if (!PageHuge(p) && !PageTransTail(p)) {
|
2011-02-01 16:52:41 -07:00
|
|
|
if (!PageLRU(p))
|
|
|
|
shake_page(p, 0);
|
|
|
|
if (!PageLRU(p)) {
|
|
|
|
/*
|
|
|
|
* shake_page could have turned it free.
|
|
|
|
*/
|
|
|
|
if (is_free_buddy_page(p)) {
|
2013-09-30 14:45:23 -06:00
|
|
|
if (flags & MF_COUNT_INCREASED)
|
|
|
|
action_result(pfn, "free buddy", DELAYED);
|
|
|
|
else
|
|
|
|
action_result(pfn, "free buddy, 2nd try", DELAYED);
|
2011-02-01 16:52:41 -07:00
|
|
|
return 0;
|
|
|
|
}
|
2009-12-16 04:20:00 -07:00
|
|
|
}
|
2009-09-28 23:16:20 -06:00
|
|
|
}
|
|
|
|
|
2011-03-10 00:52:07 -07:00
|
|
|
lock_page(hpage);
|
2009-12-16 04:19:58 -07:00
|
|
|
|
2014-08-06 17:06:49 -06:00
|
|
|
/*
|
|
|
|
* The page could have changed compound pages during the locking.
|
|
|
|
* If this happens just bail out.
|
|
|
|
*/
|
|
|
|
if (compound_head(p) != hpage) {
|
|
|
|
action_result(pfn, "different compound page after locking", IGNORED);
|
|
|
|
res = -EBUSY;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2013-02-22 17:35:51 -07:00
|
|
|
/*
|
|
|
|
* We use page flags to determine what action should be taken, but
|
|
|
|
* the flags can be modified by the error containment action. One
|
|
|
|
* example is an mlocked page, where PG_mlocked is cleared by
|
|
|
|
* page_remove_rmap() in try_to_unmap_one(). So to determine page status
|
|
|
|
* correctly, we save a copy of the page flags at this time.
|
|
|
|
*/
|
|
|
|
page_flags = p->flags;
|
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
/*
|
|
|
|
* unpoison always clear PG_hwpoison inside page lock
|
|
|
|
*/
|
|
|
|
if (!PageHWPoison(p)) {
|
2009-12-16 04:19:58 -07:00
|
|
|
printk(KERN_ERR "MCE %#lx: just unpoisoned\n", pfn);
|
2014-05-22 12:54:21 -06:00
|
|
|
atomic_long_sub(nr_pages, &num_poisoned_pages);
|
|
|
|
put_page(hpage);
|
2009-12-16 04:19:58 -07:00
|
|
|
res = 0;
|
|
|
|
goto out;
|
|
|
|
}
|
2009-12-16 04:19:59 -07:00
|
|
|
if (hwpoison_filter(p)) {
|
|
|
|
if (TestClearPageHWPoison(p))
|
2013-02-22 17:34:02 -07:00
|
|
|
atomic_long_sub(nr_pages, &num_poisoned_pages);
|
2010-05-27 18:29:17 -06:00
|
|
|
unlock_page(hpage);
|
|
|
|
put_page(hpage);
|
2009-12-16 04:19:59 -07:00
|
|
|
return 0;
|
|
|
|
}
|
2009-12-16 04:19:58 -07:00
|
|
|
|
hwpoison: fix the handling path of the victimized page frame that belong to non-LRU
Until now, the kernel has the same policy to handle victimized page
frames that belong to kernel-space(reserved/slab-subsystem) or
non-LRU(unknown page state). In other word, the result of handling
either of these victimized page frames is (IGNORED | FAILED), and the
return value of memory_failure() is -EBUSY.
This patch is to avoid that memory_failure() returns very soon due to
the "true" value of (!PageLRU(p)), and it also ensures that
action_result() can report more precise information("reserved kernel",
"kernel slab", and "unknown page state") instead of "non LRU",
especially for memory errors which are detected by memory-scrubbing.
Andi said:
: While running the mcelog test suite on 3.14 I hit the following VM_BUG_ON:
:
: soft_offline: 0x56d4: unknown non LRU page type 3ffff800008000
: page:ffffea000015b400 count:3 mapcount:2097169 mapping: (null) index:0xffff8800056d7000
: page flags: 0x3ffff800004081(locked|slab|head)
: ------------[ cut here ]------------
: kernel BUG at mm/rmap.c:1495!
:
: I think what happened is that a LRU page turned into a slab page in
: parallel with offlining. memory_failure initially tests for this case,
: but doesn't retest later after the page has been locked.
:
: ...
:
: I ran this patch in a loop over night with some stress plus
: the mcelog test suite running in a loop. I cannot guarantee it hit it,
: but it should have given it a good beating.
:
: The kernel survived with no messages, although the mcelog test suite
: got killed at some point because it couldn't fork anymore. Probably
: some unrelated problem.
:
: So the patch is ok for me for .16.
Signed-off-by: Chen Yucong <slaoub@gmail.com>
Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reported-by: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-07-02 16:22:37 -06:00
|
|
|
if (!PageHuge(p) && !PageTransTail(p) && !PageLRU(p))
|
|
|
|
goto identify_page_state;
|
|
|
|
|
2010-05-27 18:29:18 -06:00
|
|
|
/*
|
|
|
|
* For error on the tail page, we should set PG_hwpoison
|
|
|
|
* on the head page to show that the hugepage is hwpoisoned
|
|
|
|
*/
|
2011-02-01 16:52:40 -07:00
|
|
|
if (PageHuge(p) && PageTail(p) && TestSetPageHWPoison(hpage)) {
|
2010-05-27 18:29:18 -06:00
|
|
|
action_result(pfn, "hugepage already hardware poisoned",
|
|
|
|
IGNORED);
|
|
|
|
unlock_page(hpage);
|
|
|
|
put_page(hpage);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* Set PG_hwpoison on all pages in an error hugepage,
|
|
|
|
* because containment is done in hugepage unit for now.
|
|
|
|
* Since we have done TestSetPageHWPoison() for the head page with
|
|
|
|
* page lock held, we can safely set PG_hwpoison bits on tail pages.
|
|
|
|
*/
|
|
|
|
if (PageHuge(p))
|
|
|
|
set_page_hwpoison_huge_page(hpage);
|
|
|
|
|
2014-06-04 17:10:35 -06:00
|
|
|
/*
|
|
|
|
* It's very difficult to mess with pages currently under IO
|
|
|
|
* and in many cases impossible, so we just avoid it here.
|
|
|
|
*/
|
2009-09-16 03:50:15 -06:00
|
|
|
wait_on_page_writeback(p);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Now take care of user space mappings.
|
2011-03-22 17:32:44 -06:00
|
|
|
* Abort on fail: __delete_from_page_cache() assumes unmapped page.
|
2014-01-23 16:53:14 -07:00
|
|
|
*
|
|
|
|
* When the raw error page is thp tail page, hpage points to the raw
|
|
|
|
* page after thp split.
|
2009-09-16 03:50:15 -06:00
|
|
|
*/
|
2014-01-23 16:53:14 -07:00
|
|
|
if (hwpoison_user_mappings(p, pfn, trapno, flags, &hpage)
|
|
|
|
!= SWAP_SUCCESS) {
|
2014-07-30 17:08:30 -06:00
|
|
|
action_result(pfn, "unmapping failed", IGNORED);
|
2009-12-16 04:19:58 -07:00
|
|
|
res = -EBUSY;
|
|
|
|
goto out;
|
|
|
|
}
|
2009-09-16 03:50:15 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Torn down by someone else?
|
|
|
|
*/
|
2009-12-16 04:19:58 -07:00
|
|
|
if (PageLRU(p) && !PageSwapCache(p) && p->mapping == NULL) {
|
2009-09-16 03:50:15 -06:00
|
|
|
action_result(pfn, "already truncated LRU", IGNORED);
|
2009-12-16 04:19:58 -07:00
|
|
|
res = -EBUSY;
|
2009-09-16 03:50:15 -06:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
hwpoison: fix the handling path of the victimized page frame that belong to non-LRU
Until now, the kernel has the same policy to handle victimized page
frames that belong to kernel-space(reserved/slab-subsystem) or
non-LRU(unknown page state). In other word, the result of handling
either of these victimized page frames is (IGNORED | FAILED), and the
return value of memory_failure() is -EBUSY.
This patch is to avoid that memory_failure() returns very soon due to
the "true" value of (!PageLRU(p)), and it also ensures that
action_result() can report more precise information("reserved kernel",
"kernel slab", and "unknown page state") instead of "non LRU",
especially for memory errors which are detected by memory-scrubbing.
Andi said:
: While running the mcelog test suite on 3.14 I hit the following VM_BUG_ON:
:
: soft_offline: 0x56d4: unknown non LRU page type 3ffff800008000
: page:ffffea000015b400 count:3 mapcount:2097169 mapping: (null) index:0xffff8800056d7000
: page flags: 0x3ffff800004081(locked|slab|head)
: ------------[ cut here ]------------
: kernel BUG at mm/rmap.c:1495!
:
: I think what happened is that a LRU page turned into a slab page in
: parallel with offlining. memory_failure initially tests for this case,
: but doesn't retest later after the page has been locked.
:
: ...
:
: I ran this patch in a loop over night with some stress plus
: the mcelog test suite running in a loop. I cannot guarantee it hit it,
: but it should have given it a good beating.
:
: The kernel survived with no messages, although the mcelog test suite
: got killed at some point because it couldn't fork anymore. Probably
: some unrelated problem.
:
: So the patch is ok for me for .16.
Signed-off-by: Chen Yucong <slaoub@gmail.com>
Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reported-by: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-07-02 16:22:37 -06:00
|
|
|
identify_page_state:
|
2009-09-16 03:50:15 -06:00
|
|
|
res = -EBUSY;
|
2013-02-22 17:35:51 -07:00
|
|
|
/*
|
|
|
|
* The first check uses the current page flags which may not have any
|
|
|
|
* relevant information. The second check with the saved page flagss is
|
|
|
|
* carried out only if the first check can't determine the page status.
|
|
|
|
*/
|
|
|
|
for (ps = error_states;; ps++)
|
|
|
|
if ((p->flags & ps->mask) == ps->res)
|
2009-09-16 03:50:15 -06:00
|
|
|
break;
|
mm/hwpoison: fix loss of PG_dirty for errors on mlocked pages
memory_failure() store the page flag of the error page before doing unmap,
and (only) if the first check with page flags at the time decided the
error page is unknown, it do the second check with the stored page flag
since memory_failure() does unmapping of the error pages before doing
page_action(). This unmapping changes the page state, especially
page_remove_rmap() (called from try_to_unmap_one()) clears PG_mlocked, so
page_action() can't catch mlocked pages after that.
However, memory_failure() can't handle memory errors on dirty mlocked
pages correctly. try_to_unmap_one will move the dirty bit from pte to the
physical page, the second check lose it since it check the stored page
flag. This patch fix it by restore PG_dirty flag to stored page flag if
the page is dirty.
Testcase:
#define _GNU_SOURCE
#include <stdlib.h>
#include <stdio.h>
#include <sys/mman.h>
#include <sys/types.h>
#include <errno.h>
#define PAGES_TO_TEST 2
#define PAGE_SIZE 4096
int main(void)
{
char *mem;
int i;
mem = mmap(NULL, PAGES_TO_TEST * PAGE_SIZE,
PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS | MAP_LOCKED, 0, 0);
for (i = 0; i < PAGES_TO_TEST; i++)
mem[i * PAGE_SIZE] = 'a';
if (madvise(mem, PAGES_TO_TEST * PAGE_SIZE, MADV_HWPOISON) == -1)
return -1;
return 0;
}
Before patch:
[ 912.839247] Injecting memory failure for page 7dfb8 at 7f6b4e37b000
[ 912.839257] MCE 0x7dfb8: clean mlocked LRU page recovery: Recovered
[ 912.845550] MCE 0x7dfb8: clean mlocked LRU page still referenced by 1 users
[ 912.852586] Injecting memory failure for page 7e6aa at 7f6b4e37c000
[ 912.852594] MCE 0x7e6aa: clean mlocked LRU page recovery: Recovered
[ 912.858936] MCE 0x7e6aa: clean mlocked LRU page still referenced by 1 users
After patch:
[ 163.590225] Injecting memory failure for page 91bc2f at 7f9f5b0e5000
[ 163.590264] MCE 0x91bc2f: dirty mlocked LRU page recovery: Recovered
[ 163.596680] MCE 0x91bc2f: dirty mlocked LRU page still referenced by 1 users
[ 163.603831] Injecting memory failure for page 91cdd3 at 7f9f5b0e6000
[ 163.603852] MCE 0x91cdd3: dirty mlocked LRU page recovery: Recovered
[ 163.610305] MCE 0x91cdd3: dirty mlocked LRU page still referenced by 1 users
Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-09-11 15:22:50 -06:00
|
|
|
|
|
|
|
page_flags |= (p->flags & (1UL << PG_dirty));
|
|
|
|
|
2013-02-22 17:35:51 -07:00
|
|
|
if (!ps->mask)
|
|
|
|
for (ps = error_states;; ps++)
|
|
|
|
if ((page_flags & ps->mask) == ps->res)
|
|
|
|
break;
|
|
|
|
res = page_action(ps, p, pfn);
|
2009-09-16 03:50:15 -06:00
|
|
|
out:
|
2010-05-27 18:29:17 -06:00
|
|
|
unlock_page(hpage);
|
2009-09-16 03:50:15 -06:00
|
|
|
return res;
|
|
|
|
}
|
2011-12-15 11:48:12 -07:00
|
|
|
EXPORT_SYMBOL_GPL(memory_failure);
|
2009-12-16 04:19:58 -07:00
|
|
|
|
2011-07-12 23:14:27 -06:00
|
|
|
#define MEMORY_FAILURE_FIFO_ORDER 4
|
|
|
|
#define MEMORY_FAILURE_FIFO_SIZE (1 << MEMORY_FAILURE_FIFO_ORDER)
|
|
|
|
|
|
|
|
struct memory_failure_entry {
|
|
|
|
unsigned long pfn;
|
|
|
|
int trapno;
|
|
|
|
int flags;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct memory_failure_cpu {
|
|
|
|
DECLARE_KFIFO(fifo, struct memory_failure_entry,
|
|
|
|
MEMORY_FAILURE_FIFO_SIZE);
|
|
|
|
spinlock_t lock;
|
|
|
|
struct work_struct work;
|
|
|
|
};
|
|
|
|
|
|
|
|
static DEFINE_PER_CPU(struct memory_failure_cpu, memory_failure_cpu);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* memory_failure_queue - Schedule handling memory failure of a page.
|
|
|
|
* @pfn: Page Number of the corrupted page
|
|
|
|
* @trapno: Trap number reported in the signal to user space.
|
|
|
|
* @flags: Flags for memory failure handling
|
|
|
|
*
|
|
|
|
* This function is called by the low level hardware error handler
|
|
|
|
* when it detects hardware memory corruption of a page. It schedules
|
|
|
|
* the recovering of error page, including dropping pages, killing
|
|
|
|
* processes etc.
|
|
|
|
*
|
|
|
|
* The function is primarily of use for corruptions that
|
|
|
|
* happen outside the current execution context (e.g. when
|
|
|
|
* detected by a background scrubber)
|
|
|
|
*
|
|
|
|
* Can run in IRQ context.
|
|
|
|
*/
|
|
|
|
void memory_failure_queue(unsigned long pfn, int trapno, int flags)
|
|
|
|
{
|
|
|
|
struct memory_failure_cpu *mf_cpu;
|
|
|
|
unsigned long proc_flags;
|
|
|
|
struct memory_failure_entry entry = {
|
|
|
|
.pfn = pfn,
|
|
|
|
.trapno = trapno,
|
|
|
|
.flags = flags,
|
|
|
|
};
|
|
|
|
|
|
|
|
mf_cpu = &get_cpu_var(memory_failure_cpu);
|
|
|
|
spin_lock_irqsave(&mf_cpu->lock, proc_flags);
|
2013-11-14 15:32:17 -07:00
|
|
|
if (kfifo_put(&mf_cpu->fifo, entry))
|
2011-07-12 23:14:27 -06:00
|
|
|
schedule_work_on(smp_processor_id(), &mf_cpu->work);
|
|
|
|
else
|
2013-07-25 12:53:25 -06:00
|
|
|
pr_err("Memory failure: buffer overflow when queuing memory failure at %#lx\n",
|
2011-07-12 23:14:27 -06:00
|
|
|
pfn);
|
|
|
|
spin_unlock_irqrestore(&mf_cpu->lock, proc_flags);
|
|
|
|
put_cpu_var(memory_failure_cpu);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(memory_failure_queue);
|
|
|
|
|
|
|
|
static void memory_failure_work_func(struct work_struct *work)
|
|
|
|
{
|
|
|
|
struct memory_failure_cpu *mf_cpu;
|
|
|
|
struct memory_failure_entry entry = { 0, };
|
|
|
|
unsigned long proc_flags;
|
|
|
|
int gotten;
|
|
|
|
|
2014-06-04 17:07:56 -06:00
|
|
|
mf_cpu = this_cpu_ptr(&memory_failure_cpu);
|
2011-07-12 23:14:27 -06:00
|
|
|
for (;;) {
|
|
|
|
spin_lock_irqsave(&mf_cpu->lock, proc_flags);
|
|
|
|
gotten = kfifo_get(&mf_cpu->fifo, &entry);
|
|
|
|
spin_unlock_irqrestore(&mf_cpu->lock, proc_flags);
|
|
|
|
if (!gotten)
|
|
|
|
break;
|
2013-07-10 03:27:01 -06:00
|
|
|
if (entry.flags & MF_SOFT_OFFLINE)
|
|
|
|
soft_offline_page(pfn_to_page(entry.pfn), entry.flags);
|
|
|
|
else
|
|
|
|
memory_failure(entry.pfn, entry.trapno, entry.flags);
|
2011-07-12 23:14:27 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __init memory_failure_init(void)
|
|
|
|
{
|
|
|
|
struct memory_failure_cpu *mf_cpu;
|
|
|
|
int cpu;
|
|
|
|
|
|
|
|
for_each_possible_cpu(cpu) {
|
|
|
|
mf_cpu = &per_cpu(memory_failure_cpu, cpu);
|
|
|
|
spin_lock_init(&mf_cpu->lock);
|
|
|
|
INIT_KFIFO(mf_cpu->fifo);
|
|
|
|
INIT_WORK(&mf_cpu->work, memory_failure_work_func);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
core_initcall(memory_failure_init);
|
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
/**
|
|
|
|
* unpoison_memory - Unpoison a previously poisoned page
|
|
|
|
* @pfn: Page number of the to be unpoisoned page
|
|
|
|
*
|
|
|
|
* Software-unpoison a page that has been poisoned by
|
|
|
|
* memory_failure() earlier.
|
|
|
|
*
|
|
|
|
* This is only done on the software-level, so it only works
|
|
|
|
* for linux injected failures, not real hardware failures
|
|
|
|
*
|
|
|
|
* Returns 0 for success, otherwise -errno.
|
|
|
|
*/
|
|
|
|
int unpoison_memory(unsigned long pfn)
|
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
struct page *p;
|
|
|
|
int freeit = 0;
|
2010-05-27 18:29:19 -06:00
|
|
|
unsigned int nr_pages;
|
2009-12-16 04:19:58 -07:00
|
|
|
|
|
|
|
if (!pfn_valid(pfn))
|
|
|
|
return -ENXIO;
|
|
|
|
|
|
|
|
p = pfn_to_page(pfn);
|
|
|
|
page = compound_head(p);
|
|
|
|
|
|
|
|
if (!PageHWPoison(p)) {
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("MCE: Page was already unpoisoned %#lx\n", pfn);
|
2009-12-16 04:19:58 -07:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-09-11 15:22:53 -06:00
|
|
|
/*
|
|
|
|
* unpoison_memory() can encounter thp only when the thp is being
|
|
|
|
* worked by memory_failure() and the page lock is not held yet.
|
|
|
|
* In such case, we yield to memory_failure() and make unpoison fail.
|
|
|
|
*/
|
2013-09-30 14:45:22 -06:00
|
|
|
if (!PageHuge(page) && PageTransHuge(page)) {
|
2013-09-11 15:22:53 -06:00
|
|
|
pr_info("MCE: Memory failure is now running on %#lx\n", pfn);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-09-11 15:22:52 -06:00
|
|
|
nr_pages = 1 << compound_order(page);
|
2010-05-27 18:29:19 -06:00
|
|
|
|
2009-12-16 04:19:58 -07:00
|
|
|
if (!get_page_unless_zero(page)) {
|
2010-09-07 19:19:38 -06:00
|
|
|
/*
|
|
|
|
* Since HWPoisoned hugepage should have non-zero refcount,
|
|
|
|
* race between memory failure and unpoison seems to happen.
|
|
|
|
* In such case unpoison fails and memory failure runs
|
|
|
|
* to the end.
|
|
|
|
*/
|
|
|
|
if (PageHuge(page)) {
|
2011-10-31 18:09:04 -06:00
|
|
|
pr_info("MCE: Memory failure is now running on free hugepage %#lx\n", pfn);
|
2010-09-07 19:19:38 -06:00
|
|
|
return 0;
|
|
|
|
}
|
2009-12-16 04:19:58 -07:00
|
|
|
if (TestClearPageHWPoison(p))
|
2013-09-11 15:22:54 -06:00
|
|
|
atomic_long_dec(&num_poisoned_pages);
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("MCE: Software-unpoisoned free page %#lx\n", pfn);
|
2009-12-16 04:19:58 -07:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-03-10 00:52:07 -07:00
|
|
|
lock_page(page);
|
2009-12-16 04:19:58 -07:00
|
|
|
/*
|
|
|
|
* This test is racy because PG_hwpoison is set outside of page lock.
|
|
|
|
* That's acceptable because that won't trigger kernel panic. Instead,
|
|
|
|
* the PG_hwpoison page will be caught and isolated on the entrance to
|
|
|
|
* the free buddy page pool.
|
|
|
|
*/
|
2010-05-27 18:29:19 -06:00
|
|
|
if (TestClearPageHWPoison(page)) {
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("MCE: Software-unpoisoned page %#lx\n", pfn);
|
2013-02-22 17:34:02 -07:00
|
|
|
atomic_long_sub(nr_pages, &num_poisoned_pages);
|
2009-12-16 04:19:58 -07:00
|
|
|
freeit = 1;
|
2010-09-07 19:19:40 -06:00
|
|
|
if (PageHuge(page))
|
|
|
|
clear_page_hwpoison_huge_page(page);
|
2009-12-16 04:19:58 -07:00
|
|
|
}
|
|
|
|
unlock_page(page);
|
|
|
|
|
|
|
|
put_page(page);
|
mm/memory-failure.c: fix bug triggered by unpoisoning empty zero page
Injecting memory failure for page 0x19d0 at 0xb77d2000
MCE 0x19d0: non LRU page recovery: Ignored
MCE: Software-unpoisoned page 0x19d0
BUG: Bad page state in process bash pfn:019d0
page:f3461a00 count:0 mapcount:0 mapping: (null) index:0x0
page flags: 0x40000404(referenced|reserved)
Modules linked in: nfsd auth_rpcgss i915 nfs_acl nfs lockd video drm_kms_helper drm bnep rfcomm sunrpc bluetooth psmouse parport_pc ppdev lp serio_raw fscache parport gpio_ich lpc_ich mac_hid i2c_algo_bit tpm_tis wmi usb_storage hid_generic usbhid hid e1000e firewire_ohci firewire_core ahci ptp libahci pps_core crc_itu_t
CPU: 3 PID: 2123 Comm: bash Not tainted 3.11.0-rc6+ #12
Hardware name: LENOVO 7034DD7/ , BIOS 9HKT47AUS 01//2012
00000000 00000000 e9625ea0 c15ec49b f3461a00 e9625eb8 c15ea119 c17cbf18
ef084314 000019d0 f3461a00 e9625ed8 c110dc8a f3461a00 00000001 00000000
f3461a00 40000404 00000000 e9625ef8 c110dcc1 f3461a00 f3461a00 000019d0
Call Trace:
dump_stack+0x41/0x52
bad_page+0xcf/0xeb
free_pages_prepare+0x12a/0x140
free_hot_cold_page+0x21/0x110
__put_single_page+0x21/0x30
put_page+0x25/0x40
unpoison_memory+0x107/0x200
hwpoison_unpoison+0x20/0x30
simple_attr_write+0xb6/0xd0
vfs_write+0xa0/0x1b0
SyS_write+0x4f/0x90
sysenter_do_call+0x12/0x22
Disabling lock debugging due to kernel taint
Testcase:
#define _GNU_SOURCE
#include <stdlib.h>
#include <stdio.h>
#include <sys/mman.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/types.h>
#include <errno.h>
#define PAGES_TO_TEST 1
#define PAGE_SIZE 4096
int main(void)
{
char *mem;
mem = mmap(NULL, PAGES_TO_TEST * PAGE_SIZE,
PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, 0, 0);
if (madvise(mem, PAGES_TO_TEST * PAGE_SIZE, MADV_HWPOISON) == -1)
return -1;
munmap(mem, PAGES_TO_TEST * PAGE_SIZE);
return 0;
}
There is one page reference count for default empty zero page,
madvise_hwpoison add another one by get_user_pages_fast. memory_hwpoison
reduce one page reference count since it's a non LRU page.
unpoison_memory release the last page reference count and free empty zero
page to buddy system which is not correct since empty zero page has
PG_reserved flag. This patch fix it by don't reduce the page reference
count under 1 against empty zero page.
Signed-off-by: Wanpeng Li <liwanp@linux.vnet.ibm.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-09-11 15:23:01 -06:00
|
|
|
if (freeit && !(pfn == my_zero_pfn(0) && page_count(p) == 1))
|
2009-12-16 04:19:58 -07:00
|
|
|
put_page(page);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(unpoison_memory);
|
2009-12-16 04:20:00 -07:00
|
|
|
|
|
|
|
static struct page *new_page(struct page *p, unsigned long private, int **x)
|
|
|
|
{
|
2009-12-16 04:20:01 -07:00
|
|
|
int nid = page_to_nid(p);
|
2010-09-07 19:19:39 -06:00
|
|
|
if (PageHuge(p))
|
|
|
|
return alloc_huge_page_node(page_hstate(compound_head(p)),
|
|
|
|
nid);
|
|
|
|
else
|
|
|
|
return alloc_pages_exact_node(nid, GFP_HIGHUSER_MOVABLE, 0);
|
2009-12-16 04:20:00 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Safely get reference count of an arbitrary page.
|
|
|
|
* Returns 0 for a free page, -EIO for a zero refcount page
|
|
|
|
* that is not free, and 1 for any other page type.
|
|
|
|
* For 1 the page is returned with increased page count, otherwise not.
|
|
|
|
*/
|
2013-02-22 17:34:03 -07:00
|
|
|
static int __get_any_page(struct page *p, unsigned long pfn, int flags)
|
2009-12-16 04:20:00 -07:00
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (flags & MF_COUNT_INCREASED)
|
|
|
|
return 1;
|
|
|
|
|
2010-09-07 19:19:39 -06:00
|
|
|
/*
|
|
|
|
* When the target page is a free hugepage, just remove it
|
|
|
|
* from free hugepage list.
|
|
|
|
*/
|
2009-12-16 04:20:00 -07:00
|
|
|
if (!get_page_unless_zero(compound_head(p))) {
|
2010-09-07 19:19:39 -06:00
|
|
|
if (PageHuge(p)) {
|
2012-05-29 16:06:16 -06:00
|
|
|
pr_info("%s: %#lx free huge page\n", __func__, pfn);
|
2013-02-22 17:34:03 -07:00
|
|
|
ret = 0;
|
2010-09-07 19:19:39 -06:00
|
|
|
} else if (is_free_buddy_page(p)) {
|
2012-05-29 16:06:16 -06:00
|
|
|
pr_info("%s: %#lx free buddy page\n", __func__, pfn);
|
2009-12-16 04:20:00 -07:00
|
|
|
ret = 0;
|
|
|
|
} else {
|
2012-05-29 16:06:16 -06:00
|
|
|
pr_info("%s: %#lx: unknown zero refcount page type %lx\n",
|
|
|
|
__func__, pfn, p->flags);
|
2009-12-16 04:20:00 -07:00
|
|
|
ret = -EIO;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
/* Not a free page */
|
|
|
|
ret = 1;
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2013-02-22 17:34:03 -07:00
|
|
|
static int get_any_page(struct page *page, unsigned long pfn, int flags)
|
|
|
|
{
|
|
|
|
int ret = __get_any_page(page, pfn, flags);
|
|
|
|
|
|
|
|
if (ret == 1 && !PageHuge(page) && !PageLRU(page)) {
|
|
|
|
/*
|
|
|
|
* Try to free it.
|
|
|
|
*/
|
|
|
|
put_page(page);
|
|
|
|
shake_page(page, 1);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Did it turn free?
|
|
|
|
*/
|
|
|
|
ret = __get_any_page(page, pfn, 0);
|
|
|
|
if (!PageLRU(page)) {
|
|
|
|
pr_info("soft_offline: %#lx: unknown non LRU page type %lx\n",
|
|
|
|
pfn, page->flags);
|
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2010-09-07 19:19:39 -06:00
|
|
|
static int soft_offline_huge_page(struct page *page, int flags)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
unsigned long pfn = page_to_pfn(page);
|
|
|
|
struct page *hpage = compound_head(page);
|
2013-09-11 15:22:01 -06:00
|
|
|
LIST_HEAD(pagelist);
|
2010-09-07 19:19:39 -06:00
|
|
|
|
2013-02-22 17:34:03 -07:00
|
|
|
/*
|
|
|
|
* This double-check of PageHWPoison is to avoid the race with
|
|
|
|
* memory_failure(). See also comment in __soft_offline_page().
|
|
|
|
*/
|
|
|
|
lock_page(hpage);
|
2013-02-22 17:33:59 -07:00
|
|
|
if (PageHWPoison(hpage)) {
|
2013-02-22 17:34:03 -07:00
|
|
|
unlock_page(hpage);
|
|
|
|
put_page(hpage);
|
2013-02-22 17:33:59 -07:00
|
|
|
pr_info("soft offline: %#lx hugepage already poisoned\n", pfn);
|
2013-02-22 17:34:03 -07:00
|
|
|
return -EBUSY;
|
2013-02-22 17:33:59 -07:00
|
|
|
}
|
2013-02-22 17:34:03 -07:00
|
|
|
unlock_page(hpage);
|
2010-09-07 19:19:39 -06:00
|
|
|
|
|
|
|
/* Keep page count to indicate a given hugepage is isolated. */
|
2013-09-11 15:22:01 -06:00
|
|
|
list_move(&hpage->lru, &pagelist);
|
2014-06-04 17:08:25 -06:00
|
|
|
ret = migrate_pages(&pagelist, new_page, NULL, MPOL_MF_MOVE_ALL,
|
2013-09-11 15:22:01 -06:00
|
|
|
MIGRATE_SYNC, MR_MEMORY_FAILURE);
|
2010-09-07 19:19:39 -06:00
|
|
|
if (ret) {
|
2011-10-31 18:09:04 -06:00
|
|
|
pr_info("soft offline: %#lx: migration failed %d, type %lx\n",
|
|
|
|
pfn, ret, page->flags);
|
2013-09-11 15:22:01 -06:00
|
|
|
/*
|
|
|
|
* We know that soft_offline_huge_page() tries to migrate
|
|
|
|
* only one hugepage pointed to by hpage, so we need not
|
|
|
|
* run through the pagelist here.
|
|
|
|
*/
|
|
|
|
putback_active_hugepage(hpage);
|
|
|
|
if (ret > 0)
|
|
|
|
ret = -EIO;
|
2013-02-22 17:34:03 -07:00
|
|
|
} else {
|
2013-12-18 18:08:54 -07:00
|
|
|
/* overcommit hugetlb page will be freed to buddy */
|
|
|
|
if (PageHuge(page)) {
|
|
|
|
set_page_hwpoison_huge_page(hpage);
|
|
|
|
dequeue_hwpoisoned_huge_page(hpage);
|
|
|
|
atomic_long_add(1 << compound_order(hpage),
|
|
|
|
&num_poisoned_pages);
|
|
|
|
} else {
|
|
|
|
SetPageHWPoison(page);
|
|
|
|
atomic_long_inc(&num_poisoned_pages);
|
|
|
|
}
|
2010-09-07 19:19:39 -06:00
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2013-02-22 17:34:03 -07:00
|
|
|
static int __soft_offline_page(struct page *page, int flags)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
unsigned long pfn = page_to_pfn(page);
|
2009-12-16 04:20:00 -07:00
|
|
|
|
|
|
|
/*
|
2013-02-22 17:34:03 -07:00
|
|
|
* Check PageHWPoison again inside page lock because PageHWPoison
|
|
|
|
* is set by memory_failure() outside page lock. Note that
|
|
|
|
* memory_failure() also double-checks PageHWPoison inside page lock,
|
|
|
|
* so there's no race between soft_offline_page() and memory_failure().
|
2009-12-16 04:20:00 -07:00
|
|
|
*/
|
2013-02-22 17:33:59 -07:00
|
|
|
lock_page(page);
|
|
|
|
wait_on_page_writeback(page);
|
2013-02-22 17:34:03 -07:00
|
|
|
if (PageHWPoison(page)) {
|
|
|
|
unlock_page(page);
|
|
|
|
put_page(page);
|
|
|
|
pr_info("soft offline: %#lx page already poisoned\n", pfn);
|
|
|
|
return -EBUSY;
|
|
|
|
}
|
2009-12-16 04:20:00 -07:00
|
|
|
/*
|
|
|
|
* Try to invalidate first. This should work for
|
|
|
|
* non dirty unmapped page cache pages.
|
|
|
|
*/
|
|
|
|
ret = invalidate_inode_page(page);
|
|
|
|
unlock_page(page);
|
|
|
|
/*
|
|
|
|
* RED-PEN would be better to keep it isolated here, but we
|
|
|
|
* would need to fix isolation locking first.
|
|
|
|
*/
|
|
|
|
if (ret == 1) {
|
2011-05-24 18:12:20 -06:00
|
|
|
put_page(page);
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("soft_offline: %#lx: invalidated\n", pfn);
|
2013-02-22 17:34:03 -07:00
|
|
|
SetPageHWPoison(page);
|
|
|
|
atomic_long_inc(&num_poisoned_pages);
|
|
|
|
return 0;
|
2009-12-16 04:20:00 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Simple invalidation didn't work.
|
|
|
|
* Try to migrate to a new page instead. migrate.c
|
|
|
|
* handles a large number of cases for us.
|
|
|
|
*/
|
|
|
|
ret = isolate_lru_page(page);
|
2011-05-24 18:12:20 -06:00
|
|
|
/*
|
|
|
|
* Drop page reference which is came from get_any_page()
|
|
|
|
* successful isolate_lru_page() already took another one.
|
|
|
|
*/
|
|
|
|
put_page(page);
|
2009-12-16 04:20:00 -07:00
|
|
|
if (!ret) {
|
|
|
|
LIST_HEAD(pagelist);
|
2011-06-15 16:08:48 -06:00
|
|
|
inc_zone_page_state(page, NR_ISOLATED_ANON +
|
2013-02-22 17:35:14 -07:00
|
|
|
page_is_file_cache(page));
|
2009-12-16 04:20:00 -07:00
|
|
|
list_add(&page->lru, &pagelist);
|
2014-06-04 17:08:25 -06:00
|
|
|
ret = migrate_pages(&pagelist, new_page, NULL, MPOL_MF_MOVE_ALL,
|
2013-02-22 17:35:14 -07:00
|
|
|
MIGRATE_SYNC, MR_MEMORY_FAILURE);
|
2009-12-16 04:20:00 -07:00
|
|
|
if (ret) {
|
2014-01-21 16:51:17 -07:00
|
|
|
if (!list_empty(&pagelist)) {
|
|
|
|
list_del(&page->lru);
|
|
|
|
dec_zone_page_state(page, NR_ISOLATED_ANON +
|
|
|
|
page_is_file_cache(page));
|
|
|
|
putback_lru_page(page);
|
|
|
|
}
|
|
|
|
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("soft offline: %#lx: migration failed %d, type %lx\n",
|
2009-12-16 04:20:00 -07:00
|
|
|
pfn, ret, page->flags);
|
|
|
|
if (ret > 0)
|
|
|
|
ret = -EIO;
|
2013-02-22 17:34:03 -07:00
|
|
|
} else {
|
2013-07-03 16:02:37 -06:00
|
|
|
/*
|
|
|
|
* After page migration succeeds, the source page can
|
|
|
|
* be trapped in pagevec and actual freeing is delayed.
|
|
|
|
* Freeing code works differently based on PG_hwpoison,
|
|
|
|
* so there's a race. We need to make sure that the
|
|
|
|
* source page should be freed back to buddy before
|
|
|
|
* setting PG_hwpoison.
|
|
|
|
*/
|
|
|
|
if (!is_free_buddy_page(page))
|
|
|
|
lru_add_drain_all();
|
|
|
|
if (!is_free_buddy_page(page))
|
|
|
|
drain_all_pages();
|
2013-02-22 17:34:03 -07:00
|
|
|
SetPageHWPoison(page);
|
2013-07-03 16:02:37 -06:00
|
|
|
if (!is_free_buddy_page(page))
|
|
|
|
pr_info("soft offline: %#lx: page leaked\n",
|
|
|
|
pfn);
|
2013-02-22 17:34:03 -07:00
|
|
|
atomic_long_inc(&num_poisoned_pages);
|
2009-12-16 04:20:00 -07:00
|
|
|
}
|
|
|
|
} else {
|
2010-09-27 15:31:30 -06:00
|
|
|
pr_info("soft offline: %#lx: isolation failed: %d, page count %d, type %lx\n",
|
2011-10-31 18:09:04 -06:00
|
|
|
pfn, ret, page_count(page), page->flags);
|
2009-12-16 04:20:00 -07:00
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
2013-09-11 15:22:56 -06:00
|
|
|
|
|
|
|
/**
|
|
|
|
* soft_offline_page - Soft offline a page.
|
|
|
|
* @page: page to offline
|
|
|
|
* @flags: flags. Same as memory_failure().
|
|
|
|
*
|
|
|
|
* Returns 0 on success, otherwise negated errno.
|
|
|
|
*
|
|
|
|
* Soft offline a page, by migration or invalidation,
|
|
|
|
* without killing anything. This is for the case when
|
|
|
|
* a page is not corrupted yet (so it's still valid to access),
|
|
|
|
* but has had a number of corrected errors and is better taken
|
|
|
|
* out.
|
|
|
|
*
|
|
|
|
* The actual policy on when to do that is maintained by
|
|
|
|
* user space.
|
|
|
|
*
|
|
|
|
* This should never impact any application or cause data loss,
|
|
|
|
* however it might take some time.
|
|
|
|
*
|
|
|
|
* This is not a 100% solution for all memory, but tries to be
|
|
|
|
* ``good enough'' for the majority of memory.
|
|
|
|
*/
|
|
|
|
int soft_offline_page(struct page *page, int flags)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
unsigned long pfn = page_to_pfn(page);
|
2014-03-03 16:38:18 -07:00
|
|
|
struct page *hpage = compound_head(page);
|
2013-09-11 15:22:56 -06:00
|
|
|
|
|
|
|
if (PageHWPoison(page)) {
|
|
|
|
pr_info("soft offline: %#lx page already poisoned\n", pfn);
|
|
|
|
return -EBUSY;
|
|
|
|
}
|
|
|
|
if (!PageHuge(page) && PageTransHuge(hpage)) {
|
|
|
|
if (PageAnon(hpage) && unlikely(split_huge_page(hpage))) {
|
|
|
|
pr_info("soft offline: %#lx: failed to split THP\n",
|
|
|
|
pfn);
|
|
|
|
return -EBUSY;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
mem-hotplug: implement get/put_online_mems
kmem_cache_{create,destroy,shrink} need to get a stable value of
cpu/node online mask, because they init/destroy/access per-cpu/node
kmem_cache parts, which can be allocated or destroyed on cpu/mem
hotplug. To protect against cpu hotplug, these functions use
{get,put}_online_cpus. However, they do nothing to synchronize with
memory hotplug - taking the slab_mutex does not eliminate the
possibility of race as described in patch 2.
What we need there is something like get_online_cpus, but for memory.
We already have lock_memory_hotplug, which serves for the purpose, but
it's a bit of a hammer right now, because it's backed by a mutex. As a
result, it imposes some limitations to locking order, which are not
desirable, and can't be used just like get_online_cpus. That's why in
patch 1 I substitute it with get/put_online_mems, which work exactly
like get/put_online_cpus except they block not cpu, but memory hotplug.
[ v1 can be found at https://lkml.org/lkml/2014/4/6/68. I NAK'ed it by
myself, because it used an rw semaphore for get/put_online_mems,
making them dead lock prune. ]
This patch (of 2):
{un}lock_memory_hotplug, which is used to synchronize against memory
hotplug, is currently backed by a mutex, which makes it a bit of a
hammer - threads that only want to get a stable value of online nodes
mask won't be able to proceed concurrently. Also, it imposes some
strong locking ordering rules on it, which narrows down the set of its
usage scenarios.
This patch introduces get/put_online_mems, which are the same as
get/put_online_cpus, but for memory hotplug, i.e. executing a code
inside a get/put_online_mems section will guarantee a stable value of
online nodes, present pages, etc.
lock_memory_hotplug()/unlock_memory_hotplug() are removed altogether.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Toshi Kani <toshi.kani@hp.com>
Cc: Xishi Qiu <qiuxishi@huawei.com>
Cc: Jiang Liu <liuj97@gmail.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Wen Congyang <wency@cn.fujitsu.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-06-04 17:07:18 -06:00
|
|
|
get_online_mems();
|
2013-11-12 16:07:26 -07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Isolate the page, so that it doesn't get reallocated if it
|
|
|
|
* was free. This flag should be kept set until the source page
|
|
|
|
* is freed and PG_hwpoison on it is set.
|
|
|
|
*/
|
|
|
|
if (get_pageblock_migratetype(page) != MIGRATE_ISOLATE)
|
|
|
|
set_migratetype_isolate(page, true);
|
|
|
|
|
2013-09-11 15:22:56 -06:00
|
|
|
ret = get_any_page(page, pfn, flags);
|
mem-hotplug: implement get/put_online_mems
kmem_cache_{create,destroy,shrink} need to get a stable value of
cpu/node online mask, because they init/destroy/access per-cpu/node
kmem_cache parts, which can be allocated or destroyed on cpu/mem
hotplug. To protect against cpu hotplug, these functions use
{get,put}_online_cpus. However, they do nothing to synchronize with
memory hotplug - taking the slab_mutex does not eliminate the
possibility of race as described in patch 2.
What we need there is something like get_online_cpus, but for memory.
We already have lock_memory_hotplug, which serves for the purpose, but
it's a bit of a hammer right now, because it's backed by a mutex. As a
result, it imposes some limitations to locking order, which are not
desirable, and can't be used just like get_online_cpus. That's why in
patch 1 I substitute it with get/put_online_mems, which work exactly
like get/put_online_cpus except they block not cpu, but memory hotplug.
[ v1 can be found at https://lkml.org/lkml/2014/4/6/68. I NAK'ed it by
myself, because it used an rw semaphore for get/put_online_mems,
making them dead lock prune. ]
This patch (of 2):
{un}lock_memory_hotplug, which is used to synchronize against memory
hotplug, is currently backed by a mutex, which makes it a bit of a
hammer - threads that only want to get a stable value of online nodes
mask won't be able to proceed concurrently. Also, it imposes some
strong locking ordering rules on it, which narrows down the set of its
usage scenarios.
This patch introduces get/put_online_mems, which are the same as
get/put_online_cpus, but for memory hotplug, i.e. executing a code
inside a get/put_online_mems section will guarantee a stable value of
online nodes, present pages, etc.
lock_memory_hotplug()/unlock_memory_hotplug() are removed altogether.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Toshi Kani <toshi.kani@hp.com>
Cc: Xishi Qiu <qiuxishi@huawei.com>
Cc: Jiang Liu <liuj97@gmail.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Wen Congyang <wency@cn.fujitsu.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-06-04 17:07:18 -06:00
|
|
|
put_online_mems();
|
2013-11-12 16:07:26 -07:00
|
|
|
if (ret > 0) { /* for in-use pages */
|
2013-09-11 15:22:56 -06:00
|
|
|
if (PageHuge(page))
|
|
|
|
ret = soft_offline_huge_page(page, flags);
|
|
|
|
else
|
|
|
|
ret = __soft_offline_page(page, flags);
|
2013-11-12 16:07:26 -07:00
|
|
|
} else if (ret == 0) { /* for free pages */
|
2013-09-11 15:22:56 -06:00
|
|
|
if (PageHuge(page)) {
|
|
|
|
set_page_hwpoison_huge_page(hpage);
|
|
|
|
dequeue_hwpoisoned_huge_page(hpage);
|
|
|
|
atomic_long_add(1 << compound_order(hpage),
|
|
|
|
&num_poisoned_pages);
|
|
|
|
} else {
|
|
|
|
SetPageHWPoison(page);
|
|
|
|
atomic_long_inc(&num_poisoned_pages);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
unset_migratetype_isolate(page, MIGRATE_MOVABLE);
|
|
|
|
return ret;
|
|
|
|
}
|