Some ACPI interrupt actions may need to wait, and it's easiest to
have a thread context for this. So turn the ACPI interrupt
into a threaded interrupt.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
WARN() is not supposed to have side effects, so move the request_regions
outside.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
On Intel CPUs the processor typically uses the highest frequency
set by any logical CPU. When the system overheats
Linux first forces the frequency to the lowest available one
to lower the temperature.
However this was done only per logical CPU, which means all
logical CPUs in a package would need to go through this before
the frequency is actually lowered.
Worse this delay actually prevents real throttling, because
the real throttle code only proceeds when the lowest frequency
is already reached.
So when a throttle event happens force the lowest frequency
for all CPUs in the package where it happened. The per CPU
state is now kept per package, not per logical CPU. An alternative
would be to do it per cpufreq unit, but since we want to bring
down the temperature of the complete chip it's better
to do it for all.
In principle it may even make sense to do it for all CPUs,
but I kept it on the package for now.
With this change the frequency is actually lowered, which
in terms also allows real throttling to proceed.
I also removed an unnecessary per cpu variable initialization.
v2: Fix package mapping
Cc: <stable@vger.kernel.org>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
These are all the cases where a variable is set, but not read which are
not bugs as far as I can see, but simply leftovers.
Still needs more review.
Found by gcc 4.6's new warnings
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
These are all the cases where a variable is set, but not
read which are really bugs.
- Couple of incorrect error handling fixed.
- One incorrect use of a allocation policy
- Some other things
Still needs more review.
Found by gcc 4.6's new warnings.
[akpm@linux-foundation.org: fix build. Might have been bitrot]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
kmsg_dump takes care to sample the global variables
inside a spinlock, but then goes on to use the same
variables outside the spinlock region too.
Use the correct variable. This will make the race
window smaller.
Found by gcc 4.6's new warnings.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When the user passed in a NULL mask pass this on from the ioctl
handler.
Found by gcc 4.6's new warnings.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Fixes at least one real minor bug: the nfs4 recovery dir sysctl
would not return its status properly.
Also I finished Al's 1e41568d73 ("Take ima_path_check() in nfsd
past dentry_open() in nfsd_open()") commit, it moved the IMA
code, but left the old path initializer in there.
The rest is just dead code removed I think, although I was not
fully sure about the "is_borc" stuff. Some more review
would be still good.
Found by gcc 4.6's new warnings.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
They are only useful with CONFIG_CPUMASK_OFFSTACK
Avoids hundreds of warnings with a gcc 4.6 -Wall build.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Newer gcc has a -femit-struct-debug-baseonly option that dramatically
reduces the size of object files with debug info. What it does
is to only emit type information for structures when the structures
are defined in the same file or in a header file.
This means the type information for most headers are not included.
This is not good when the type information is actually
needed (e.g. with kgdb or systemtap)
But often kernel hackers only care about line numbers and don't
need all the type information anyways. In this case setting
the option can be a big win:
A build dir for a specific x86-64 configuration with gcc 4.5
shrunk from 2.3G to 1.2G. The compilation was also nearly a minute
faster.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
[mmarek: reformatted help text]
Signed-off-by: Michal Marek <mmarek@suse.cz>
No real bugs found, just removed some dead code.
Found by gcc 4.6's new warnings.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
oprofile used a double buffer scheme for its cpu event buffer
to avoid races on reading with the old locked ring buffer.
But that is obsolete now with the new ring buffer, so simply
use a single buffer. This greatly simplifies the code and avoids
a lot of sample drops on large runs, especially with call graph.
Based on suggestions from Steven Rostedt
For stable kernels from v2.6.32, but not earlier.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: stable <stable@kernel.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
We found a system where the MP table MPC and MPF structures overlap.
That doesn't really matter because the mptable is not used anyways with ACPI,
but it leads to a panic in the early allocator due to the overlapping
reservations in 2.6.33.
Earlier kernels handled this without problems.
Simply change these reservations to reserve_early_overlap_ok to avoid
the panic.
Reported-by: Thomas Renninger <trenn@suse.de>
Tested-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <20100329074111.GA22821@basil.fritz.box>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: <stable@kernel.org>
Convert some drivers who export a single string as class attribute
to the new class_attr_string functions. This removes redundant
code all over.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Several drivers just export a static string as class attributes.
Use the new extensible attribute support to define a simple
CLASS_ATTR_STRING() macro for this.
This will allow to remove code from drivers in followon patches.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Passing the attribute to the low level IO functions allows all kinds
of cleanups, by sharing low level IO code without requiring
an own function for every piece of data.
Also drivers can extend the attributes with own data fields
and use that in the low level function.
This makes the class attributes the same as sysdev_class attributes
and plain attributes.
This will allow further cleanups in drivers.
Full tree sweep converting all users.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This attribute is really a sysdev_class attribute, not a plain class attribute.
They are identical in layout currently, but this might not always be
the case.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Allow to create/remove arrays of sysdev attributes
Just wrappers around sysfs_create/move_files
Will be used later to clean up some drivers.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Convert the node driver to sysdev_class attribute arrays. This
greatly cleans up the code and remove a lot of code.
Saves ~150 bytes of code on x86-64.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Use sysdev_class attribute arrays in node driver
Convert the node driver to sysdev_class attribute arrays. This
greatly cleans up the code and remove a lot of code.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Add a attribute array that is automatically registered and unregistered
to struct sysdev_class. This is similar to what struct class has.
A lot of drivers add list of attributes, so it's better to do
this easily in the common sysdev layer.
This adds a new field to struct sysdev_class. I audited the
whole tree and there are no dynamically allocated sysdev classes,
so this is fully compatible.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Adding/Removing a whole array of attributes is very common. Add a standard
utility function to do this with a simple function call, instead of
requiring drivers to open code this.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Using the new attribute argument convert the cpu driver class attributes
to carry the node state. Then use a shared function to do what a lot of
individual functions did before.
This eliminates an ugly macro.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Using the new attribute argument convert the node driver class
attributes to carry the node state. Then use a shared function to do
what a lot of individual functions did before.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Passing the attribute to the low level IO functions allows all kinds
of cleanups, by sharing low level IO code without requiring
an own function for every piece of data.
Also drivers can extend the attributes with own data fields
and use that in the low level function.
Similar to sysdev_attributes and normal attributes.
This is a tree-wide sweep, converting everything in one go.
No functional changes in this patch other than passing the new
argument everywhere.
Tested on x86, the non x86 parts are uncompiled.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Better dwarf2 unwind information is a good thing, it allows better
debugging with kgdb and crash and helps systemtap.
Commit 003086497f ("Build with
-fno-dwarf2-cfi-asm") disabled some CFI information globally to work
around a module loader bug on powerpc.
But this disables the better unwind tables for all architectures, not just
powerpc. Move the workaround to powerpc and also add a suitable comment
that's it really a workaround.
This improves dwarf2 unwind tables on x86 at least.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Michal Marek <mmarek@suse.cz>
While looking for something else I noticed that the symbol
hash function used by kconfig is quite poor. It doesn't
use any of the standard hash techniques but simply
adds up the string and then uses power of two masking,
which is both known to perform poorly.
The current x86 kconfig has over 7000 symbols.
When I instrumented it showed that the minimum hash chain
length was 16 and a significant number of them was over
30.
It didn't help that the hash table size was only 256 buckets.
This patch increases the hash table size to a larger prime
and switches to a FNV32 hash. I played around with a couple of hash
functions, but that one seemed to perform best with reasonable
hash table sizes.
Increasing the hash table size even further didn't
seem like a good idea, because there are a couple of global
walks which walk the complete hash table.
I also moved the unnamed bucket to 0. It's still the longest
of all the buckets (44 entries), but hopefully it's not
often hit except for the global walk which doesn't care.
The result is a much nicer distribution:
(first column bucket length, second number of buckets with that length)
1: 3505
2: 1236
3: 294
4: 52
5: 3
47: 1 <--- this is the unnamed symbols bucket
There are still some 5+ buckets, but increasing the hash table
even more would be likely not worth it.
This also cleans up the code slightly by removing hard coded
magic numbers.
I didn't notice a big performance difference either way
on my Nehalem system, but I presume it'll help somewhat
on slower systems.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Add Xeon 7500 series support to oprofile.
Straight forward: it's the same as Core i7, so just detect
the model number. No user space changes needed.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
Right now for kfifo_*_user it's not easily possible to distingush between
a user copy failing and the FIFO not containing enough data. The problem
is that both conditions are multiplexed into the same return code.
Avoid this by moving the "copy length" into a separate output parameter
and only return 0/-EFAULT in the main return value.
I didn't fully adapt the weird "record" variants, those seem
to be unused anyways and were rather messy (should they be just removed?)
I would appreciate some double checking if I did all the conversions
correctly.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Stefani Seibold <stefani@seibold.net>
Cc: Roland Dreier <rdreier@cisco.com>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Andy Walls <awalls@radix.net>
Cc: Vikram Dhillon <dhillonv10@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When print-fatal-signals is enabled it's possible to dump any memory
reachable by the kernel to the log by simply jumping to that address from
user space.
Or crash the system if there's some hardware with read side effects.
The fatal signals handler will dump 16 bytes at the execution address,
which is fully controlled by ring 3.
In addition when something jumps to a unmapped address there will be up to
16 additional useless page faults, which might be potentially slow (and at
least is not very efficient)
Fortunately this option is off by default and only there on i386.
But fix it by checking for kernel addresses and also stopping when there's
a page fault.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
linux/kernel.h has a "clamp" macro, but r300_cmdbuf also uses a variable
with the same name. Right now it doesn't seem to include the header,
but sooner or later someone will. So better rename the variable
now.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The injector filter requires stable_page_flags() which is supplied
by procfs. So make it dependent on that.
Also add ifdefs around the filter code in memory-failure.c so that
when the filter is disabled due to missing dependencies the whole
code still builds.
Reported-by: Ingo Molnar
Signed-off-by: Andi Kleen <ak@linux.intel.com>
As predicted during code review, the sysctl(2) changes made systems with
old glibc nearly unusable. About every command gives a:
warning: process `ls' used the deprecated sysctl system call with 1.4
warning in the log.
I see this on a SUSE 10.0 system with glibc 2.3.5.
Don't warn for this common case.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Process based injection is much easier to handle for test programs,
who can first bring a page into a specific state and then test.
So add a new MADV_SOFT_OFFLINE to soft offline a page, similar
to the existing hard offline injector.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
This is a simpler, gentler variant of memory_failure() for soft page
offlining controlled from user space. It doesn't kill anything, just
tries to invalidate and if that doesn't work migrate the
page away.
This is useful for predictive failure analysis, where a page has
a high rate of corrected errors, but hasn't gone bad yet. Instead
it can be offlined early and avoided.
The offlining is controlled from sysfs, including a new generic
entry point for hard page offlining for symmetry too.
We use the page isolate facility to prevent re-allocation
race. Normally this is only used by memory hotplug. To avoid
races with memory allocation I am using lock_system_sleep().
This avoids the situation where memory hotplug is about
to isolate a page range and then hwpoison undoes that work.
This is a big hammer currently, but the simplest solution
currently.
When the page is not free or LRU we try to free pages
from slab and other caches. The slab freeing is currently
quite dumb and does not try to focus on the specific slab
cache which might own the page. This could be potentially
improved later.
Thanks to Fengguang Wu and Haicheng Li for some fixes.
[Added fix from Andrew Morton to adapt to new migrate_pages prototype]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
shake_page handles more types of page caches than
the much simpler lru_add_drain_all:
- slab (quite inefficiently for now)
- any other caches with a shrinker callback
- per cpu page allocator pages
- per CPU LRU
Use this call to try to turn pages into free or LRU pages.
Then handle the case of the page becoming free after drain everything.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
The previous version didn't take the mmap_sem before calling gup(),
which is racy.
Use get_user_pages_fast() instead which doesn't need any locks.
This is also faster of course, but then it doesn't really matter
because this is just a testing path.
Based on report from Nick Piggin.
Cc: npiggin@suse.de
Signed-off-by: Andi Kleen <ak@linux.intel.com>
The hwpoison test suite need to inject hwpoison to a collection of
selected task pages, and must not touch pages not owned by them and
thus kill important system processes such as init. (But it's OK to
mis-hwpoison free/unowned pages as well as shared clean pages.
Mis-hwpoison of shared dirty pages will kill all tasks, so the test
suite will target all or non of such tasks in the first place.)
The memory cgroup serves this purpose well. We can put the target
processes under the control of a memory cgroup, and tell the hwpoison
injection code to only kill pages associated with some active memory
cgroup.
The prerequisite for doing hwpoison stress tests with mem_cgroup is,
the mem_cgroup code tracks task pages _accurately_ (unless page is
locked). Which we believe is/should be true.
The benefits are simplification of hwpoison injector code. Also the
mem_cgroup code will automatically be tested by hwpoison test cases.
The alternative interfaces pin-pfn/unpin-pfn can also delegate the
(process and page flags) filtering functions reliably to user space.
However prototype implementation shows that this scheme adds more
complexity than we wanted.
Example test case:
mkdir /cgroup/hwpoison
usemem -m 100 -s 1000 &
echo `jobs -p` > /cgroup/hwpoison/tasks
memcg_ino=$(ls -id /cgroup/hwpoison | cut -f1 -d' ')
echo $memcg_ino > /debug/hwpoison/corrupt-filter-memcg
page-types -p `pidof init` --hwpoison # shall do nothing
page-types -p `pidof usemem` --hwpoison # poison its pages
[AK: Fix documentation]
[Add fix for problem noticed by Li Zefan <lizf@cn.fujitsu.com>;
dentry in the css could be NULL]
CC: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
CC: Hugh Dickins <hugh.dickins@tiscali.co.uk>
CC: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
CC: Balbir Singh <balbir@linux.vnet.ibm.com>
CC: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
CC: Li Zefan <lizf@cn.fujitsu.com>
CC: Paul Menage <menage@google.com>
CC: Nick Piggin <npiggin@suse.de>
CC: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Now that "ref" is just a boolean turn it into
a flags argument. First step is only a single flag
that makes the code's intention more clear, but more
may follow.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
shake_page handles more types of page caches than lru_drain_all()
- per cpu page allocator pages
- per CPU LRU
Stops early when the page became free.
Used in followon patches.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Allow memory hotplug and hibernation in the same kernel
Memory hotplug and hibernation were exclusive in Kconfig. This is
obviously a problem for distribution kernels who want to support both in
the same image.
After some discussions with Rafael and others the only problem is with
parallel memory hotadd or removal while a hibernation operation is in
process. It was also working for s390 before.
This patch removes the Kconfig level exclusion, and simply makes the
memory add / remove functions grab the pm_mutex to exclude against
hibernation.
Fixes a regression - old kernels didn't exclude memory hotadd and
hibernation.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Do a similar optimization as earlier for touch_atime. Getting the lock in
mnt_get_write is relatively costly, so try all avenues to avoid it first.
This patch is careful to still only update inode fields inside the lock
region.
This didn't show up in benchmarks, but it's easy enough to do.
[akpm@linux-foundation.org: fix typo in comment]
[hugh.dickins@tiscali.co.uk: fix inverted test of mnt_want_write_file()]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Valerie Aurora <vaurora@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Some benchmark testing shows touch_atime to be high up in profile logs for
IO intensive workloads. Most likely that's due to the lock in
mnt_want_write(). Unfortunately touch_atime first takes the lock, and
then does all the other tests that could avoid atime updates (like noatime
or relatime).
Do it the other way round -- first try to avoid the update and only then
if that didn't succeed take the lock. That works because none of the
atime avoidance tests rely on locking.
This also eliminates a goto.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Reviewed-by: Valerie Aurora <vaurora@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
I tend to use a 'D' debugging macro a lot during debugging. When I define
it before includes I often get conflicts with kmap_types.h's use of 'D'
too. It's not very nice when a global include pollutes the name space
like this.
Rename the kmap_types.h D to KMAP_D. It is only used temporarily in the
header so has no effect on anything else.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: WANG Cong <xiyou.wangcong@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When this script fails the build should fail too. Otherwise there
are mysterious build failures later.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
The upcomming gcc 4.5 has a new -fconserve-stack option that tells the
inliner to take stack frame size in account. Set it if the compiler
supports it.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
I had some problems with record_mcount in the Makefile and it was hard
to track down. Echo it by default to make it easier to diagnose.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
The new alternative `gold' linker in recent binutils doesn't support
the -X option. This breaks allyesconfig builds that have
CONFIG_STRIP_ASM_SYMS enabled. Check if the linker really supports
the option using ld-option.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
When booting with pci=nomsi aer causes lost interrupts and
lockdep inversions.
So check if MSIs are not disabled before initializing the aer
driver.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Useful for some testing scenarios, although specific testing is often
done better through MADV_POISON
This can be done with the x86 level MCE injector too, but this interface
allows it to do independently from low level x86 changes.
v2: Add module license (Haicheng Li)
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Impact: optional, useful for debugging
Add a new madvice sub command to inject poison for some
pages in a process' address space. This is useful for
testing the poison page handling.
This patch can allow root to tie up large amounts of memory.
I got feedback from container developers and they didn't see any
problem.
v2: Use write flag for get_user_pages to make sure to always get
a fresh page
v3: Don't request write mapping (Fengguang Wu)
v4: Move MADV_* number to avoid conflict with KSM (Hugh Dickins)
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Enable hardware memory error handling for NFS
Truncation of data pages at runtime should be safe in NFS,
even when it doesn't support migration so far.
Trond tells me migration is also queued up for 2.6.32.
Acked-by: Trond.Myklebust@netapp.com
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Enable removing of corrupted pages through truncation
for a bunch of file systems: ext*, xfs, gfs2, ocfs2, ntfs
These should cover most server needs.
I chose the set of migration aware file systems for this
for now, assuming they have been especially audited.
But in general it should be safe for all file systems
on the data area that support read/write and truncate.
Caveat: the hardware error handler does not take i_mutex
for now before calling the truncate function. Is that ok?
Cc: tytso@mit.edu
Cc: hch@infradead.org
Cc: mfasheh@suse.com
Cc: aia21@cantab.net
Cc: hugh.dickins@tiscali.co.uk
Cc: swhiteho@redhat.com
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Add the high level memory handler that poisons pages
that got corrupted by hardware (typically by a two bit flip in a DIMM
or a cache) on the Linux level. The goal is to prevent everyone
from accessing these pages in the future.
This done at the VM level by marking a page hwpoisoned
and doing the appropriate action based on the type of page
it is.
The code that does this is portable and lives in mm/memory-failure.c
To quote the overview comment:
High level machine check handler. Handles pages reported by the
hardware as being corrupted usually due to a 2bit ECC memory or cache
failure.
This focuses on pages detected as corrupted in the background.
When the current CPU tries to consume corruption the currently
running process can just be killed directly instead. This implies
that if the error cannot be handled for some reason it's safe to
just ignore it because no corruption has been consumed yet. Instead
when that happens another machine check will happen.
Handles page cache pages in various states. The tricky part
here is that we can access any page asynchronous to other VM
users, because memory failures could happen anytime and anywhere,
possibly violating 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.
Some of the operations here are somewhat inefficient and have non
linear algorithmic complexity, because the data structures have not
been optimized for this case. This is in particular the case
for the mapping from a vma to a process. Since this case is expected
to be rare we hope we can get away with this.
There are in principle two strategies to kill processes on poison:
- just unmap the data and wait for an actual reference before
killing
- kill as soon as corruption is detected.
Both have advantages and disadvantages and should be used
in different situations. Right now both are implemented and can
be switched with a new sysctl vm.memory_failure_early_kill
The default is early kill.
The patch does some rmap data structure walking on its own to collect
processes to kill. This is unusual because normally all rmap data structure
knowledge is in rmap.c only. I put it here for now to keep
everything together and rmap knowledge has been seeping out anyways
Includes contributions from Johannes Weiner, Chris Mason, Fengguang Wu,
Nick Piggin (who did a lot of great work) and others.
Cc: npiggin@suse.de
Cc: riel@redhat.com
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
This allows processes to override their early/late kill
behaviour on hardware memory errors.
Typically applications which are memory error aware is
better of with early kill (see the error as soon
as possible), all others with late kill (only
see the error when the error is really impacting execution)
There's a global sysctl, but this way an application
can set its specific policy.
We're using two bits, one to signify that the process
stated its intention and that
I also made the prctl future proof by enforcing
the unused arguments are 0.
The state is inherited to children.
Note this makes us officially run out of process flags
on 32bit, but the next patch can easily add another field.
Manpage patch will be supplied separately.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Truncating metadata pages is not safe right now before
we haven't audited all file systems.
To enable truncation only for data address space define
a new address_space callback error_remove_page.
This is used for memory_failure.c memory error handling.
This can be then set to truncate_inode_page()
This patch just defines the new operation and adds documentation.
Callers and users come in followon patches.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
try_to_unmap currently has multiple modi (migration, munlock, normal unmap)
which are selected by magic flag variables. The logic is not very straight
forward, because each of these flag change multiple behaviours (e.g.
migration turns off aging, not only sets up migration ptes etc.)
Also the different flags interact in magic ways.
A later patch in this series adds another mode to try_to_unmap, so
this becomes quickly unmanageable.
Replace the different flags with a action code (migration, munlock, munmap)
and some additional flags as modifiers (ignore mlock, ignore aging).
This makes the logic more straight forward and allows easier extension
to new behaviours. Change all the caller to declare what they want to
do.
This patch is supposed to be a nop in behaviour. If anyone can prove
it is not that would be a bug.
Cc: Lee.Schermerhorn@hp.com
Cc: npiggin@suse.de
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Add VM_FAULT_HWPOISON handling to the x86 page fault handler. This is
very similar to VM_FAULT_OOM, the only difference is that a different
si_code is passed to user space and the new addr_lsb field is initialized.
v2: Make the printk more verbose/unique
Cc: x86@kernel.org
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Bail out early when hardware poisoned pages are found in page fault handling.
Since they are poisoned they should not be mapped freshly into processes,
because that would cause another (potentially deadly) machine check
This is generally handled in the same way as OOM, just a different
error code is returned to the architecture code.
v2: Do a page unlock if needed (Fengguang Wu)
Signed-off-by: Andi Kleen <ak@linux.intel.com>
- Add a new VM_FAULT_HWPOISON error code to handle_mm_fault. Right now
architectures have to explicitely enable poison page support, so
this is forward compatible to all architectures. They only need
to add it when they enable poison page support.
- Add poison page handling in swap in fault code
v2: Add missing delayacct_clear_flag (Hidehiro Kawai)
v3: Really use delayacct_clear_flag (Hidehiro Kawai)
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Add new SIGBUS codes for reporting machine checks as signals. When
the hardware detects an uncorrected ECC error it can trigger these
signals.
This is needed for telling KVM's qemu about machine checks that happen to
guests, so that it can inject them, but might be also useful for other programs.
I find it useful in my test programs.
This patch merely defines the new types.
- Define two new si_codes for SIGBUS. BUS_MCEERR_AO and BUS_MCEERR_AR
* BUS_MCEERR_AO is for "Action Optional" machine checks, which means that some
corruption has been detected in the background, but nothing has been consumed
so far. The program can ignore those if it wants (but most programs would
already get killed)
* BUS_MCEERR_AR is for "Action Required" machine checks. This happens
when corrupted data is consumed or the application ran into an area
which has been known to be corrupted earlier. These require immediate
action and cannot just returned to. Most programs would kill themselves.
- They report the address of the corruption in the user address space
in si_addr.
- Define a new si_addr_lsb field that reports the extent of the corruption
to user space. That's currently always a (small) page. The user application
cannot tell where in this page the corruption happened.
AK: I plan to write a man page update before anyone asks.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Memory migration uses special swap entry types to trigger special actions on
page faults. Extend this mechanism to also support poisoned swap entries, to
trigger poison handling on page faults. This allows follow-on patches to
prevent processes from faulting in poisoned pages again.
v2: Fix overflow in MAX_SWAPFILES (Fengguang Wu)
v3: Better overflow fix (Hidehiro Kawai)
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Needed for later patch that walks rmap entries on its own.
This used to be very frowned upon, but memory-failure.c does
some rather specialized rmap walking and rmap has been stable
for quite some time, so I think it's ok now to export it.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Hardware poisoned pages need special handling in the VM and shouldn't be
touched again. This requires a new page flag. Define it here.
The page flags wars seem to be over, so it shouldn't be a problem
to get a new one.
v2: Add TestSetHWPoison (suggested by Johannes Weiner)
Acked-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Shut off the long standing
linux/drivers/pci/search.c:144: warning: 'pci_find_device' is deprecated (declared at linux/drivers/pci/search.c:136)
linux/drivers/pci/search.c:144: warning: 'pci_find_device' is deprecated (declared at linux/drivers/pci/search.c:136)
warnings that appear on every build when CONFIG_PCI_LEGACY is enabled.
gcc warns for the use in EXPORT_SYMBOL
I moved these to a separate file and disabled the warning in the Makefile for that file.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
While looking for something else I spent some time adding
one liner comments to the tcp_output.c functions that
didn't have any. That makes the comments more consistent.
I hope I documented everything right.
No code changes.
v2: Incorporated feedback from Ilpo.
v3: Change style of one liner comments, add a few more comments.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
The Intel x86 architecture right now only supports 32 machine check
banks, more would bump into other MSRs.
So lower the max define to 32.
This only affects a few bitmaps, most data structures are dynamically
sized anyways.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Instead of open coded calculations for bank MSRs hide the indexing of higher
banks MCE register MSRs in new macros.
No semantic changes.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
This addresses one of the leftover review comments.
Move the per bank data into a single structure. This avoids
several separate variables and also separate allocation of sysfs objects.
I didn't move the CMCI ownership information so far because
that would have needed some non trivial changes in the algorithms.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Now that the X86_OLD_MCE ifdefs are gone move some code that
used to be outside the big ifdef to a more natural place
near its user.
No code change.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Andi Kleen