3 Upcoming Intel CPUs have support for recovering from some memory errors
4 (``MCA recovery''). This requires the OS to declare a page "poisoned",
5 kill the processes associated with it and avoid using it in the future.
7 This patchkit implements the necessary infrastructure in the VM.
9 To quote the overview comment:
11 * High level machine check handler. Handles pages reported by the
12 * hardware as being corrupted usually due to a 2bit ECC memory or cache
15 * This focusses on pages detected as corrupted in the background.
16 * When the current CPU tries to consume corruption the currently
17 * running process can just be killed directly instead. This implies
18 * that if the error cannot be handled for some reason it's safe to
19 * just ignore it because no corruption has been consumed yet. Instead
20 * when that happens another machine check will happen.
22 * Handles page cache pages in various states. The tricky part
23 * here is that we can access any page asynchronous to other VM
24 * users, because memory failures could happen anytime and anywhere,
25 * possibly violating some of their assumptions. This is why this code
26 * has to be extremely careful. Generally it tries to use normal locking
27 * rules, as in get the standard locks, even if that means the
28 * error handling takes potentially a long time.
30 * Some of the operations here are somewhat inefficient and have non
31 * linear algorithmic complexity, because the data structures have not
32 * been optimized for this case. This is in particular the case
33 * for the mapping from a vma to a process. Since this case is expected
34 * to be rare we hope we can get away with this.
36 The code consists of a the high level handler in mm/memory-failure.c,
37 a new page poison bit and various checks in the VM to handle poisoned
40 The main target right now is KVM guests, but it works for all kinds
41 of applications. KVM support requires a recent qemu-kvm release.
43 For the KVM use there was need for a new signal type so that
44 KVM can inject the machine check into the guest with the proper
45 address. This in theory allows other applications to handle
46 memory failures too. The expection is that near all applications
47 won't do that, but some very specialized ones might.
51 There are two (actually three) modi memory failure recovery can be in:
53 vm.memory_failure_recovery sysctl set to zero:
54 All memory failures cause a panic. Do not attempt recovery.
55 (on x86 this can be also affected by the tolerant level of the
59 (can be controlled globally and per process)
60 Send SIGBUS to the application as soon as the error is detected
61 This allows applications who can process memory errors in a gentle
62 way (e.g. drop affected object)
63 This is the mode used by KVM qemu.
66 Send SIGBUS when the application runs into the corrupted page.
67 This is best for memory error unaware applications and default
68 Note some pages are always handled as late kill.
74 vm.memory_failure_recovery
77 vm.memory_failure_early_kill
78 Enable early kill mode globally
81 Set early/late kill mode/revert to system default
82 arg1: PR_MCE_KILL_CLEAR: Revert to system default
83 arg1: PR_MCE_KILL_SET: arg2 defines thread specific mode
84 PR_MCE_KILL_EARLY: Early kill
85 PR_MCE_KILL_LATE: Late kill
86 PR_MCE_KILL_DEFAULT: Use system global default
95 madvise(MADV_HWPOISON, ....)
97 Poison a page in the process for testing
100 hwpoison-inject module through debugfs
106 Inject hwpoison fault at PFN echoed into this file. This does
107 some early filtering to avoid corrupted unintended pages in test suites.
111 Software-unpoison page at PFN echoed into this file. This
112 way a page can be reused again.
113 This only works for Linux injected failures, not for real
116 Note these injection interfaces are not stable and might change between
119 corrupt-filter-dev-major
120 corrupt-filter-dev-minor
122 Only handle memory failures to pages associated with the file system defined
123 by block device major/minor. -1U is the wildcard value.
124 This should be only used for testing with artificial injection.
128 Limit injection to pages owned by memgroup. Specified by inode number
132 mkdir /sys/fs/cgroup/mem/hwpoison
134 usemem -m 100 -s 1000 &
135 echo `jobs -p` > /sys/fs/cgroup/mem/hwpoison/tasks
137 memcg_ino=$(ls -id /sys/fs/cgroup/mem/hwpoison | cut -f1 -d' ')
138 echo $memcg_ino > /debug/hwpoison/corrupt-filter-memcg
140 page-types -p `pidof init` --hwpoison # shall do nothing
141 page-types -p `pidof usemem` --hwpoison # poison its pages
143 corrupt-filter-flags-mask
144 corrupt-filter-flags-value
146 When specified, only poison pages if ((page_flags & mask) == value).
147 This allows stress testing of many kinds of pages. The page_flags
148 are the same as in /proc/kpageflags. The flag bits are defined in
149 include/linux/kernel-page-flags.h and documented in
150 Documentation/vm/pagemap.txt
152 Architecture specific MCE injector
154 x86 has mce-inject, mce-test
156 Some portable hwpoison test programs in mce-test, see blow.
162 http://halobates.de/mce-lc09-2.pdf
163 Overview presentation from LinuxCon 09
165 git://git.kernel.org/pub/scm/utils/cpu/mce/mce-test.git
166 Test suite (hwpoison specific portable tests in tsrc)
168 git://git.kernel.org/pub/scm/utils/cpu/mce/mce-inject.git
169 x86 specific injector
176 - Not all page types are supported and never will. Most kernel internal
177 objects cannot be recovered, only LRU pages for now.
178 - Right now hugepage support is missing.