The cpusets-formalize-intermediate-gfp_kernel-containment patch
has a deadlock problem.
This patch was part of a set of four patches to make more
extensive use of the cpuset 'mem_exclusive' attribute to
manage kernel GFP_KERNEL memory allocations and to constrain
the out-of-memory (oom) killer.
A task that is changing cpusets in particular ways on a system
when it is very short of free memory could double trip over
the global cpuset_sem semaphore (get the lock and then deadlock
trying to get it again).
The second attempt to get cpuset_sem would be in the routine
cpuset_zone_allowed(). This was discovered by code inspection.
I can not reproduce the problem except with an artifically
hacked kernel and a specialized stress test.
In real life you cannot hit this unless you are manipulating
cpusets, and are very unlikely to hit it unless you are rapidly
modifying cpusets on a memory tight system. Even then it would
be a rare occurence.
If you did hit it, the task double tripping over cpuset_sem
would deadlock in the kernel, and any other task also trying
to manipulate cpusets would deadlock there too, on cpuset_sem.
Your batch manager would be wedged solid (if it was cpuset
savvy), but classic Unix shells and utilities would work well
enough to reboot the system.
The unusual condition that led to this bug is that unlike most
semaphores, cpuset_sem _can_ be acquired while in the page
allocation code, when __alloc_pages() calls cpuset_zone_allowed.
So it easy to mistakenly perform the following sequence:
1) task makes system call to alter a cpuset
2) take cpuset_sem
3) try to allocate memory
4) memory allocator, via cpuset_zone_allowed, trys to take cpuset_sem
5) deadlock
The reason that this is not a serious bug for most users
is that almost all calls to allocate memory don't require
taking cpuset_sem. Only some code paths off the beaten
track require taking cpuset_sem -- which is good. Taking
a global semaphore on the main code path for allocating
memory would not scale well.
This patch fixes this deadlock by wrapping the up() and down()
calls on cpuset_sem in kernel/cpuset.c with code that tracks
the nesting depth of the current task on that semaphore, and
only does the real down() if the task doesn't hold the lock
already, and only does the real up() if the nesting depth
(number of unmatched downs) is exactly one.
The previous required use of refresh_mems(), anytime that
the cpuset_sem semaphore was acquired and the code executed
while holding that semaphore might try to allocate memory, is
no longer required. Two refresh_mems() calls were removed
thanks to this. This is a good change, as failing to get
all the necessary refresh_mems() calls placed was a primary
source of bugs in this cpuset code. The only remaining call
to refresh_mems() is made while doing a memory allocation,
if certain task memory placement data needs to be updated
from its cpuset, due to the cpuset having been changed behind
the tasks back.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch (written by me and also containing many suggestions of Arjan van
de Ven) does a major cleanup of the spinlock code. It does the following
things:
- consolidates and enhances the spinlock/rwlock debugging code
- simplifies the asm/spinlock.h files
- encapsulates the raw spinlock type and moves generic spinlock
features (such as ->break_lock) into the generic code.
- cleans up the spinlock code hierarchy to get rid of the spaghetti.
Most notably there's now only a single variant of the debugging code,
located in lib/spinlock_debug.c. (previously we had one SMP debugging
variant per architecture, plus a separate generic one for UP builds)
Also, i've enhanced the rwlock debugging facility, it will now track
write-owners. There is new spinlock-owner/CPU-tracking on SMP builds too.
All locks have lockup detection now, which will work for both soft and hard
spin/rwlock lockups.
The arch-level include files now only contain the minimally necessary
subset of the spinlock code - all the rest that can be generalized now
lives in the generic headers:
include/asm-i386/spinlock_types.h | 16
include/asm-x86_64/spinlock_types.h | 16
I have also split up the various spinlock variants into separate files,
making it easier to see which does what. The new layout is:
SMP | UP
----------------------------|-----------------------------------
asm/spinlock_types_smp.h | linux/spinlock_types_up.h
linux/spinlock_types.h | linux/spinlock_types.h
asm/spinlock_smp.h | linux/spinlock_up.h
linux/spinlock_api_smp.h | linux/spinlock_api_up.h
linux/spinlock.h | linux/spinlock.h
/*
* here's the role of the various spinlock/rwlock related include files:
*
* on SMP builds:
*
* asm/spinlock_types.h: contains the raw_spinlock_t/raw_rwlock_t and the
* initializers
*
* linux/spinlock_types.h:
* defines the generic type and initializers
*
* asm/spinlock.h: contains the __raw_spin_*()/etc. lowlevel
* implementations, mostly inline assembly code
*
* (also included on UP-debug builds:)
*
* linux/spinlock_api_smp.h:
* contains the prototypes for the _spin_*() APIs.
*
* linux/spinlock.h: builds the final spin_*() APIs.
*
* on UP builds:
*
* linux/spinlock_type_up.h:
* contains the generic, simplified UP spinlock type.
* (which is an empty structure on non-debug builds)
*
* linux/spinlock_types.h:
* defines the generic type and initializers
*
* linux/spinlock_up.h:
* contains the __raw_spin_*()/etc. version of UP
* builds. (which are NOPs on non-debug, non-preempt
* builds)
*
* (included on UP-non-debug builds:)
*
* linux/spinlock_api_up.h:
* builds the _spin_*() APIs.
*
* linux/spinlock.h: builds the final spin_*() APIs.
*/
All SMP and UP architectures are converted by this patch.
arm, i386, ia64, ppc, ppc64, s390/s390x, x64 was build-tested via
crosscompilers. m32r, mips, sh, sparc, have not been tested yet, but should
be mostly fine.
From: Grant Grundler <grundler@parisc-linux.org>
Booted and lightly tested on a500-44 (64-bit, SMP kernel, dual CPU).
Builds 32-bit SMP kernel (not booted or tested). I did not try to build
non-SMP kernels. That should be trivial to fix up later if necessary.
I converted bit ops atomic_hash lock to raw_spinlock_t. Doing so avoids
some ugly nesting of linux/*.h and asm/*.h files. Those particular locks
are well tested and contained entirely inside arch specific code. I do NOT
expect any new issues to arise with them.
If someone does ever need to use debug/metrics with them, then they will
need to unravel this hairball between spinlocks, atomic ops, and bit ops
that exist only because parisc has exactly one atomic instruction: LDCW
(load and clear word).
From: "Luck, Tony" <tony.luck@intel.com>
ia64 fix
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjanv@infradead.org>
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Cc: Matthew Wilcox <willy@debian.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Mikael Pettersson <mikpe@csd.uu.se>
Signed-off-by: Benoit Boissinot <benoit.boissinot@ens-lyon.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
pcibus_to_cpumask expands into more than just an initialiser so gcc
moans about code before variable declarations.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
GCC 4 emits more DWARF debugging information than before and there is now a
.debug_loc section as well. This causes "make buildcheck" to fail. Rather
than just add that one to the special case list, I used a regexp to ignore
any .debug_ANYTHING sections in case more show up in the future.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Splitting of autoconf.h requires that split-include was built before,
and
needs to be-re-done when split-include changes. This dependency was
previously missing. Additionally, since autoconf.h is (suppoosed to
be)
generated as a side effect of executing config targets, include/linux
should be created prior to running the respective sub-make.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
In order to maintain a more correct build number, updates to the
version
number should only be commited after a successful link of vmlinux, not
before (so that errors in the link process don't lead to pointless
increments).
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Allows to add to sparse arguments without mutilating makefiles - just
pass CF=<arguments> and they will be added to CHECKFLAGS.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Dunno if there was a conscious decision to leave it out, but if you're
happy with adding some help text for it here's a patch against 2.6.13-mm1..
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Convert wbsd to use the new delay functionality in mmc_detect_change()
rather than implementing its own timer.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Brian Haley <brian.haley@hp.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the CM REQ handling function gets to error2, then it frees
cm_id_priv->timewait_info. But the next line goes through
ib_destroy_cm_id() -> ib_send_cm_rej() -> cm_reset_to_idle(),
which ends up calling cm_cleanup_timewait(), which dereferences the
pointer we just freed. Make sure we clear cm_id_priv->timewait_info
after freeing it, so that doesn't happen.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Changes to CM to support CM and port redirection (REJ reason 24).
Signed-off-by: John Kingman <kingman <at> storagegear.com>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This shouldn't be a BUG. We should cope.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If you try to assemble an array with too many missing devices, raid10 will now
reject the attempt, instead of allowing it.
Also check when hot-adding a drive and refuse the hot-add if the array is
beyond hope.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There was another case where sb_size wasn't being set, so instead do the
sensible thing and set if when filling in the content of a superblock. That
ensures that whenever we write a superblock, the sb_size MUST be set.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There are two ways to add devices to an md/raid array.
It can have superblock written to it, and then given to the md driver,
which will read the superblock (the new way)
or
md can be told (through SET_ARRAY_INFO) the shape of the array, and
the told about individual drives, and md will create the required
superblock (the old way).
The newly introduced sb_size was only set for drives being added the
new way, not the old ways. Oops :-(
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Just like failed drives have (F), so spare drives now have (S).
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Leave it unchanged if the original (0.90) is used, incase it might be a
compatability problem.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Doh. I want the physical hard-sector-size, not the current block size...
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
On reflection, a better default location for hot-adding bitmaps with version-1
superblocks is immediately after the superblock. There might not be much room
there, but there is usually atleast 3k, and that is a good start.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The bitmap code used to have two daemons, so there is some 'common' start/stop
code. But now there is only one, so the common code is just noise.
This patch tidies this up somewhat.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
mddev->bitmap gets clearred before the writeback daemon is stopped. So the
write_back daemon needs to be careful not to dereference the 'bitmap' if it is
NULL.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Switch MD to use the kthread infrastructure, to simplify the code and get rid
of tasklist_lock abuse in md_unregister_thread.
Also don't flush signals in md_thread, as the called thread will always do
that.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is a direct port of the raid5 patch.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Most awkward part of this is delaying write requests until bitmap updates have
been flushed.
To achieve this, we have a sequence number (seq_flush) which is incremented
each time the raid5 is unplugged.
If the raid thread notices that this has changed, it flushes bitmap changes,
and assigned the value of seq_flush to seq_write.
When a write request arrives, it is given the number from seq_write, and that
write request may not complete until seq_flush is larger than the saved seq
number.
We have a new queue for storing stripes which are waiting for a bitmap flush
and an extra flag for stripes to record if the write was 'degraded' and so
should not clear the a bit in the bitmap.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
version-1 superblocks are not (normally) 4K long, and can be of variable size.
Writing the full 4K can cause corruption (but only in non-default
configurations).
With this patch the super-block-flavour can choose a size to read, and set a
size to write based on what it finds.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These inlines haven't been used for ages, they should go.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
read_sb_page() assumed that if sync_page_io fails, the device would be marked
faultly. However it isn't. So in the face of error, read_sb_page would loop
forever.
Redo the logic so that this cannot happen.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As this is used to flag an internal bitmap.
Also, introduce symbolic names for feature bits.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It is possibly (and occasionally useful) to have a raid1 without persistent
superblocks. The code in add_new_disk for adding a device to such an array
always tries to read a superblock.
This will obviously fail.
So do the appropriate test and call md_import_device with
appropriate args.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When hot-adding a bitmap, bitmap_daemon_work could get called while the bitmap
is being created, so don't set mddev->bitmap until the bitmap is ready.
This requires freeing the bitmap inside bitmap_create if creation failed
part-way through.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The 'lastrun' time wasn't being initialised, so it could be half a
jiffie-cycle before it seemed to be time to do work again.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A state of 0 mean 'not quiesced'
A state of 1 means 'is quiesced'
The original code got this wrong.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
linear currently uses division by the size of the smallest componenet device
to find which device a request goes to. If that smallest device is larger
than 2 terabytes, then the division will not work on some systems.
So we introduce a pre-shift, and take care not to make the hash table too
large, much like the code in raid0.
Also get rid of conf->nr_zones, which is not needed.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If a device is flagged 'WriteMostly' and the array has a bitmap, and the
bitmap superblock indicates that write_behind is allowed, then write_behind is
enabled for WriteMostly devices.
Write requests will be acknowledges as complete to the caller (via b_end_io)
when all non-WriteMostly devices have completed the write, but will not be
cleared from the bitmap until all devices complete.
This requires memory allocation to make a local copy of the data being
written. If there is insufficient memory, then we fall-back on normal write
semantics.
Signed-Off-By: Paul Clements <paul.clements@steeleye.com>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This allows a device in a raid1 to be marked as "write mostly". Read requests
will only be sent if there is no other option.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Both file-bitmaps and superblock bitmaps are supported.
If you add a bitmap file on the array device, you lose.
This introduces a 'default_bitmap_offset' field in mddev, as the ioctl used
for adding a superblock bitmap doesn't have room for giving an offset. Later,
this value will be setable via sysfs.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When we find a 'stale' bitmap, possibly because it is new, we should just
assume every bit needs to be set, but rather base the setting of bits on the
current state of the array (degraded and recovery_cp).
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
... otherwise we loose a reference and can never free the file.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix another bug in dm-raid1.c that the dirty region may stay in or be moved
to clean list and freed while in use.
It happens as follows:
CPU0 CPU1
------------------------------------------------------------------------------
rh_dec()
if (atomic_dec_and_test(pending))
<the region is still marked dirty>
rh_inc()
if the region is clean
mark the region dirty
and remove from clean list
mark the region clean
and move to clean list
atomic_inc(pending)
At this stage, the region is in clean list and will be mistakenly reclaimed
by rh_update_states() later.
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
md does not yet support BIO_RW_BARRIER, so be honest about it and fail
(-EOPNOTSUPP) any such requests.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
