Rewrite the vmap allocator to use rbtrees and lazy tlb flushing, and
provide a fast, scalable percpu frontend for small vmaps (requires a
slightly different API, though).
The biggest problem with vmap is actually vunmap. Presently this requires
a global kernel TLB flush, which on most architectures is a broadcast IPI
to all CPUs to flush the cache. This is all done under a global lock. As
the number of CPUs increases, so will the number of vunmaps a scaled
workload will want to perform, and so will the cost of a global TLB flush.
This gives terrible quadratic scalability characteristics.
Another problem is that the entire vmap subsystem works under a single
lock. It is a rwlock, but it is actually taken for write in all the fast
paths, and the read locking would likely never be run concurrently anyway,
so it's just pointless.
This is a rewrite of vmap subsystem to solve those problems. The existing
vmalloc API is implemented on top of the rewritten subsystem.
The TLB flushing problem is solved by using lazy TLB unmapping. vmap
addresses do not have to be flushed immediately when they are vunmapped,
because the kernel will not reuse them again (would be a use-after-free)
until they are reallocated. So the addresses aren't allocated again until
a subsequent TLB flush. A single TLB flush then can flush multiple
vunmaps from each CPU.
XEN and PAT and such do not like deferred TLB flushing because they can't
always handle multiple aliasing virtual addresses to a physical address.
They now call vm_unmap_aliases() in order to flush any deferred mappings.
That call is very expensive (well, actually not a lot more expensive than
a single vunmap under the old scheme), however it should be OK if not
called too often.
The virtual memory extent information is stored in an rbtree rather than a
linked list to improve the algorithmic scalability.
There is a per-CPU allocator for small vmaps, which amortizes or avoids
global locking.
To use the per-CPU interface, the vm_map_ram / vm_unmap_ram interfaces
must be used in place of vmap and vunmap. Vmalloc does not use these
interfaces at the moment, so it will not be quite so scalable (although it
will use lazy TLB flushing).
As a quick test of performance, I ran a test that loops in the kernel,
linearly mapping then touching then unmapping 4 pages. Different numbers
of tests were run in parallel on an 4 core, 2 socket opteron. Results are
in nanoseconds per map+touch+unmap.
threads vanilla vmap rewrite
1 14700 2900
2 33600 3000
4 49500 2800
8 70631 2900
So with a 8 cores, the rewritten version is already 25x faster.
In a slightly more realistic test (although with an older and less
scalable version of the patch), I ripped the not-very-good vunmap batching
code out of XFS, and implemented the large buffer mapping with vm_map_ram
and vm_unmap_ram... along with a couple of other tricks, I was able to
speed up a large directory workload by 20x on a 64 CPU system. I believe
vmap/vunmap is actually sped up a lot more than 20x on such a system, but
I'm running into other locks now. vmap is pretty well blown off the
profiles.
Before:
1352059 total 0.1401
798784 _write_lock 8320.6667 <- vmlist_lock
529313 default_idle 1181.5022
15242 smp_call_function 15.8771 <- vmap tlb flushing
2472 __get_vm_area_node 1.9312 <- vmap
1762 remove_vm_area 4.5885 <- vunmap
316 map_vm_area 0.2297 <- vmap
312 kfree 0.1950
300 _spin_lock 3.1250
252 sn_send_IPI_phys 0.4375 <- tlb flushing
238 vmap 0.8264 <- vmap
216 find_lock_page 0.5192
196 find_next_bit 0.3603
136 sn2_send_IPI 0.2024
130 pio_phys_write_mmr 2.0312
118 unmap_kernel_range 0.1229
After:
78406 total 0.0081
40053 default_idle 89.4040
33576 ia64_spinlock_contention 349.7500
1650 _spin_lock 17.1875
319 __reg_op 0.5538
281 _atomic_dec_and_lock 1.0977
153 mutex_unlock 1.5938
123 iget_locked 0.1671
117 xfs_dir_lookup 0.1662
117 dput 0.1406
114 xfs_iget_core 0.0268
92 xfs_da_hashname 0.1917
75 d_alloc 0.0670
68 vmap_page_range 0.0462 <- vmap
58 kmem_cache_alloc 0.0604
57 memset 0.0540
52 rb_next 0.1625
50 __copy_user 0.0208
49 bitmap_find_free_region 0.2188 <- vmap
46 ia64_sn_udelay 0.1106
45 find_inode_fast 0.1406
42 memcmp 0.2188
42 finish_task_switch 0.1094
42 __d_lookup 0.0410
40 radix_tree_lookup_slot 0.1250
37 _spin_unlock_irqrestore 0.3854
36 xfs_bmapi 0.0050
36 kmem_cache_free 0.0256
35 xfs_vn_getattr 0.0322
34 radix_tree_lookup 0.1062
33 __link_path_walk 0.0035
31 xfs_da_do_buf 0.0091
30 _xfs_buf_find 0.0204
28 find_get_page 0.0875
27 xfs_iread 0.0241
27 __strncpy_from_user 0.2812
26 _xfs_buf_initialize 0.0406
24 _xfs_buf_lookup_pages 0.0179
24 vunmap_page_range 0.0250 <- vunmap
23 find_lock_page 0.0799
22 vm_map_ram 0.0087 <- vmap
20 kfree 0.0125
19 put_page 0.0330
18 __kmalloc 0.0176
17 xfs_da_node_lookup_int 0.0086
17 _read_lock 0.0885
17 page_waitqueue 0.0664
vmap has gone from being the top 5 on the profiles and flushing the crap
out of all TLBs, to using less than 1% of kernel time.
[akpm@linux-foundation.org: cleanups, section fix]
[akpm@linux-foundation.org: fix build on alpha]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When optimizing the kernel boot time, it's very valuable to visualize
what is going on at which time. In addition, with some of the initializing
going asynchronous soon, it's valuable to track/print which worker thread
is executing the initialization.
This patch adds a script to turn a dmesg into a SVG graph (that can be
shown with tools such as InkScape, Gimp or Firefox) and a small change
to the initcall code to print the PID of the thread calling the initcall
(so that the script can work out the parallelism).
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
.. small detail, but the silly e1000e initcall warning debugging caused
me to look at this code. Rather than gouge my eyes out with a spoon, I
just fixed it.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The kernel has this really nice facility where if you put "initcall_debug"
on the kernel commandline, it'll print which function it's going to
execute just before calling an initcall, and then after the call completes
it will
1) print if it had an error code
2) checks for a few simple bugs (like leaving irqs off)
and
3) print how long the init call took in milliseconds.
While trying to optimize the boot speed of my laptop, I have been loving
number 3 to figure out what to optimize... ... and then I wished that
the same thing was done for module loading.
This patch makes the module loader use this exact same functionality; it's
a logical extension in my view (since modules are just sort of late
binding initcalls anyway) and so far I've found it quite useful in finding
where things are too slow in my boot.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
commit fb6624ebd9 (initrd: Fix virtual/physical
mix-up in overwrite test) introduced the compiler warning below on mips,
as its virt_to_page() doesn't cast the passed address to unsigned long
internally, unlike on most other architectures:
init/main.c: In function `start_kernel':
init/main.c:633: warning: passing argument 1 of `virt_to_phys' makes pointer from integer without a cast
init/main.c:636: warning: passing argument 1 of `virt_to_phys' makes pointer from integer without a cast
For now, kill the warning by explicitly casting initrd_start to `void *', as
that's the type it should really be.
Reported-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
A previous patch added the early_initcall(), to allow a cleaner hooking of
pre-SMP initcalls. Now we remove the older interface, converting all
existing users to the new one.
[akpm@linux-foundation.org: cleanups]
[akpm@linux-foundation.org: build fix]
[kosaki.motohiro@jp.fujitsu.com: warning fix]
[kosaki.motohiro@jp.fujitsu.com: warning fix]
Signed-off-by: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Tom Zanussi <tzanussi@gmail.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Added early initcall (pre-SMP) support, using an identical interface to
that of regular initcalls. Functions called from do_pre_smp_initcalls()
could be converted to use this cleaner interface.
This is required by CPU hotplug, because early users have to register
notifiers before going SMP. One such CPU hotplug user is the relay
interface with buffer-only channels, which needs to register such a
notifier, to be usable in early code. This in turn is used by kmemtrace.
Signed-off-by: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Tom Zanussi <tzanussi@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch adds proper prototypes for pid{hash,map}_init() in
include/linux/pid_namespace.h
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'sched/for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
sched: hrtick_enabled() should use cpu_active()
sched, x86: clean up hrtick implementation
sched: fix build error, provide partition_sched_domains() unconditionally
sched: fix warning in inc_rt_tasks() to not declare variable 'rq' if it's not needed
cpu hotplug: Make cpu_active_map synchronization dependency clear
cpu hotplug, sched: Introduce cpu_active_map and redo sched domain managment (take 2)
sched: rework of "prioritize non-migratable tasks over migratable ones"
sched: reduce stack size in isolated_cpu_setup()
Revert parts of "ftrace: do not trace scheduler functions"
Fixed up conflicts in include/asm-x86/thread_info.h (due to the
TIF_SINGLESTEP unification vs TIF_HRTICK_RESCHED removal) and
kernel/sched_fair.c (due to cpu_active_map vs for_each_cpu_mask_nr()
introduction).
On recent kernels, I get the following error when using an initrd:
| initrd overwritten (0x00b78000 < 0x07668000) - disabling it.
My Amiga 4000 has 12 MiB of RAM at physical address 0x07400000 (virtual
0x00000000).
The initrd is located at the end of RAM: 0x00b78000 - 0x00c00000 (virtual).
The overwrite test compares the (virtual) initrd location to the (physical)
first available memory location, which fails.
This patch converts initrd_start to a page frame number, so it can safely be
compared with min_low_pfn.
Before the introduction of discontiguous memory support on m68k
(12d810c1b8), min_low_pfn was just left
untouched by the m68k-specific code (zero, I guess), and everything worked
fine.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is based on Linus' idea of creating cpu_active_map that prevents
scheduler load balancer from migrating tasks to the cpu that is going
down.
It allows us to simplify domain management code and avoid unecessary
domain rebuilds during cpu hotplug event handling.
Please ignore the cpusets part for now. It needs some more work in order
to avoid crazy lock nesting. Although I did simplfy and unify domain
reinitialization logic. We now simply call partition_sched_domains() in
all the cases. This means that we're using exact same code paths as in
cpusets case and hence the test below cover cpusets too.
Cpuset changes to make rebuild_sched_domains() callable from various
contexts are in the separate patch (right next after this one).
This not only boots but also easily handles
while true; do make clean; make -j 8; done
and
while true; do on-off-cpu 1; done
at the same time.
(on-off-cpu 1 simple does echo 0/1 > /sys/.../cpu1/online thing).
Suprisingly the box (dual-core Core2) is quite usable. In fact I'm typing
this on right now in gnome-terminal and things are moving just fine.
Also this is running with most of the debug features enabled (lockdep,
mutex, etc) no BUG_ONs or lockdep complaints so far.
I believe I addressed all of the Dmitry's comments for original Linus'
version. I changed both fair and rt balancer to mask out non-active cpus.
And replaced cpu_is_offline() with !cpu_active() in the main scheduler
code where it made sense (to me).
Signed-off-by: Max Krasnyanskiy <maxk@qualcomm.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Gregory Haskins <ghaskins@novell.com>
Cc: dmitry.adamushko@gmail.com
Cc: pj@sgi.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'generic-ipi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (22 commits)
generic-ipi: more merge fallout
generic-ipi: merge fix
x86, visws: use mach-default/entry_arch.h
x86, visws: fix generic-ipi build
generic-ipi: fixlet
generic-ipi: fix s390 build bug
generic-ipi: fix linux-next tree build failure
fix: "smp_call_function: get rid of the unused nonatomic/retry argument"
fix: "smp_call_function: get rid of the unused nonatomic/retry argument"
fix "smp_call_function: get rid of the unused nonatomic/retry argument"
on_each_cpu(): kill unused 'retry' parameter
smp_call_function: get rid of the unused nonatomic/retry argument
sh: convert to generic helpers for IPI function calls
parisc: convert to generic helpers for IPI function calls
mips: convert to generic helpers for IPI function calls
m32r: convert to generic helpers for IPI function calls
arm: convert to generic helpers for IPI function calls
alpha: convert to generic helpers for IPI function calls
ia64: convert to generic helpers for IPI function calls
powerpc: convert to generic helpers for IPI function calls
...
Fix trivial conflicts due to rcu updates in kernel/rcupdate.c manually
This adds kernel/smp.c which contains helpers for IPI function calls. In
addition to supporting the existing smp_call_function() in a more efficient
manner, it also adds a more scalable variant called smp_call_function_single()
for calling a given function on a single CPU only.
The core of this is based on the x86-64 patch from Nick Piggin, lots of
changes since then. "Alan D. Brunelle" <Alan.Brunelle@hp.com> has
contributed lots of fixes and suggestions as well. Also thanks to
Paul E. McKenney <paulmck@linux.vnet.ibm.com> for reviewing RCU usage
and getting rid of the data allocation fallback deadlock.
Acked-by: Ingo Molnar <mingo@elte.hu>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Fourth cut of patch to provide the call_rcu_sched(). This is again to
synchronize_sched() as call_rcu() is to synchronize_rcu().
Should be fine for experimental and -rt use, but not ready for inclusion.
With some luck, I will be able to tell Andrew to come out of hiding on
the next round.
Passes multi-day rcutorture sessions with concurrent CPU hotplugging.
Fixes since the first version include a bug that could result in
indefinite blocking (spotted by Gautham Shenoy), better resiliency
against CPU-hotplug operations, and other minor fixes.
Fixes since the second version include reworking grace-period detection
to avoid deadlocks that could happen when running concurrently with
CPU hotplug, adding Mathieu's fix to avoid the softlockup messages,
as well as Mathieu's fix to allow use earlier in boot.
Fixes since the third version include a wrong-CPU bug spotted by
Andrew, getting rid of the obsolete synchronize_kernel API that somehow
snuck back in, merging spin_unlock() and local_irq_restore() in a
few places, commenting the code that checks for quiescent states based
on interrupting from user-mode execution or the idle loop, removing
some inline attributes, and some code-style changes.
Known/suspected shortcomings:
o I still do not entirely trust the sleep/wakeup logic. Next step
will be to use a private snapshot of the CPU online mask in
rcu_sched_grace_period() -- if the CPU wasn't there at the start
of the grace period, we don't need to hear from it. And the
bit about accounting for changes in online CPUs inside of
rcu_sched_grace_period() is ugly anyway.
o It might be good for rcu_sched_grace_period() to invoke
resched_cpu() when a given CPU wasn't responding quickly,
but resched_cpu() is declared static...
This patch also fixes a long-standing bug in the earlier preemptable-RCU
implementation of synchronize_rcu() that could result in loss of
concurrent external changes to a task's CPU affinity mask. I still cannot
remember who reported this...
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This patch fixes a build bug on m68k - gcc decides to emit a call to the
strlen library function, which we don't implement.
More importantly - my previous patch "init: don't lose initcall return
values" (commit e662e1cfd4) had introduced
potential buffer overflow by wrong calculation of string accumulator
size.
Use strlcat() instead, fixing both bugs.
Many thanks Andreas Schwab and Geert Uytterhoeven for helping
to catch and fix the bug.
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
One function to just loop over the entries, one function to actually do
the call and the associated debugging code.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Everybody wants to pass it a function pointer, and in fact, that is what
you _must_ pass it for it to make sense (since it knows that ia64 and
ppc64 use descriptors for function pointers and fetches the actual
address from there).
So don't make the argument be a 'unsigned long' and force everybody to
add a cast.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There is an ability to lose an initcall return value if it happened with irq
disabled or imbalanced preemption (and if we debug initcall).
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
this replaces the rq->clock stuff (and possibly cpu_clock()).
- architectures that have an 'imperfect' hardware clock can set
CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
- the 'jiffie' window might be superfulous when we update tick_gtod
before the __update_sched_clock() call in sched_clock_tick()
- cpu_clock() might be implemented as:
sched_clock_cpu(smp_processor_id())
if the accuracy proves good enough - how far can TSC drift in a
single jiffie when considering the filtering and idle hooks?
[ mingo@elte.hu: various fixes and cleanups ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We can see an ever repeating problem pattern with objects of any kind in the
kernel:
1) freeing of active objects
2) reinitialization of active objects
Both problems can be hard to debug because the crash happens at a point where
we have no chance to decode the root cause anymore. One problem spot are
kernel timers, where the detection of the problem often happens in interrupt
context and usually causes the machine to panic.
While working on a timer related bug report I had to hack specialized code
into the timer subsystem to get a reasonable hint for the root cause. This
debug hack was fine for temporary use, but far from a mergeable solution due
to the intrusiveness into the timer code.
The code further lacked the ability to detect and report the root cause
instantly and keep the system operational.
Keeping the system operational is important to get hold of the debug
information without special debugging aids like serial consoles and special
knowledge of the bug reporter.
The problems described above are not restricted to timers, but timers tend to
expose it usually in a full system crash. Other objects are less explosive,
but the symptoms caused by such mistakes can be even harder to debug.
Instead of creating specialized debugging code for the timer subsystem a
generic infrastructure is created which allows developers to verify their code
and provides an easy to enable debug facility for users in case of trouble.
The debugobjects core code keeps track of operations on static and dynamic
objects by inserting them into a hashed list and sanity checking them on
object operations and provides additional checks whenever kernel memory is
freed.
The tracked object operations are:
- initializing an object
- adding an object to a subsystem list
- deleting an object from a subsystem list
Each operation is sanity checked before the operation is executed and the
subsystem specific code can provide a fixup function which allows to prevent
the damage of the operation. When the sanity check triggers a warning message
and a stack trace is printed.
The list of operations can be extended if the need arises. For now it's
limited to the requirements of the first user (timers).
The core code enqueues the objects into hash buckets. The hash index is
generated from the address of the object to simplify the lookup for the check
on kfree/vfree. Each bucket has it's own spinlock to avoid contention on a
global lock.
The debug code can be compiled in without being active. The runtime overhead
is minimal and could be optimized by asm alternatives. A kernel command line
option enables the debugging code.
Thanks to Ingo Molnar for review, suggestions and cleanup patches.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Greg KH <greg@kroah.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are some places that are known to operate on tasks'
global pids only:
* the rest_init() call (called on boot)
* the kgdb's getthread
* the create_kthread() (since the kthread is run in init ns)
So use the find_task_by_pid_ns(..., &init_pid_ns) there
and schedule the find_task_by_pid for removal.
[sukadev@us.ibm.com: Fix warning in kernel/pid.c]
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The global init has a lot of long standing problems with the unhandled fatal
signals.
- The "is_global_init(current)" check in get_signal_to_deliver()
protects only the main thread. Sub-thread can dequee the fatal
signal and shutdown the whole thread group except the main thread.
If it dequeues SIGSTOP /sbin/init will be stopped, this is not
right too. Note that we can't use is_global_init(->group_leader),
this breaks exec and this can't solve other problems we have.
- Even if afterwards ignored, the fatal signals sets SIGNAL_GROUP_EXIT
on delivery. This breaks exec, has other bad implications, and this
is just wrong.
Introduce the new SIGNAL_UNKILLABLE flag to fix these problems. It also helps
to solve some other problems addressed by the subsequent patches.
Currently we use this flag for the global init only, but it could also be used
by kthreads and (perhaps) by the sub-namespace inits.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Avoid a possible kmem_cache_create() failure by creating idr_layer_cache
unconditionary at boot time rather than creating it on-demand when idr_init()
is called the first time.
This change also enables us to eliminate the check every time idr_init() is
called.
[akpm@linux-foundation.org: rename init_id_cache() to idr_init_cache()]
[akpm@linux-foundation.org: fix alpha build]
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Remove the mem_cgroup member from mm_struct and instead adds an owner.
This approach was suggested by Paul Menage. The advantage of this approach
is that, once the mm->owner is known, using the subsystem id, the cgroup
can be determined. It also allows several control groups that are
virtually grouped by mm_struct, to exist independent of the memory
controller i.e., without adding mem_cgroup's for each controller, to
mm_struct.
A new config option CONFIG_MM_OWNER is added and the memory resource
controller selects this config option.
This patch also adds cgroup callbacks to notify subsystems when mm->owner
changes. The mm_cgroup_changed callback is called with the task_lock() of
the new task held and is called just prior to changing the mm->owner.
I am indebted to Paul Menage for the several reviews of this patchset and
helping me make it lighter and simpler.
This patch was tested on a powerpc box, it was compiled with both the
MM_OWNER config turned on and off.
After the thread group leader exits, it's moved to init_css_state by
cgroup_exit(), thus all future charges from runnings threads would be
redirected to the init_css_set's subsystem.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Pavel Emelianov <xemul@openvz.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Sudhir Kumar <skumar@linux.vnet.ibm.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Hirokazu Takahashi <taka@valinux.co.jp>
Cc: David Rientjes <rientjes@google.com>,
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Reviewed-by: Paul Menage <menage@google.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
print_fn_descriptor_symbol() prints the address if we don't have a symbol, so
no need to print both.
Also, combine printing return value with elapsed time. Changes this:
Calling initcall 0xc05b7a70: pci_mmcfg_late_insert_resources+0x0/0x50()
initcall 0xc05b7a70: pci_mmcfg_late_insert_resources+0x0/0x50() returned 1.
initcall 0xc05b7a70 ran for 0 msecs: pci_mmcfg_late_insert_resources+0x0/0x50()
initcall at 0xc05b7a70: pci_mmcfg_late_insert_resources+0x0/0x50(): returned with error code 1
to this:
calling pci_mmcfg_late_insert_resources+0x0/0x50()
initcall pci_mmcfg_late_insert_resources+0x0/0x50() returned 1 after 0 msecs
initcall pci_mmcfg_late_insert_resources+0x0/0x50() returned with error code 1
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use the __weak macro instead of the longer __attribute__ ((weak)) form
in one place in init/main.c.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
--
init/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Signed-off-by: Paul Mackerras <paulus@samba.org>
Some architectures need to maintain a kmem cache for thread info
structures. The next commit adds that to powerpc to fix an alignment
problem.
There is no good arch callback to use to initialize that cache
that I can find, so this adds a new one in the form of a weak
function whose default is empty.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Move the setting of nr_cpu_ids from sched_init() to start_kernel()
so that it's available as early as possible.
Note that an arch has the option of setting it even earlier if need be,
but it should not result in a different value than the setup_nr_cpu_ids()
function.
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* Add a static cpumask_t variable "CPU_MASK_ALL_PTR" to use as
a pointer reference to CPU_MASK_ALL. This reduces where possible
the instances where CPU_MASK_ALL allocates and fills a large
array on the stack. Used only if NR_CPUS > BITS_PER_LONG.
* Change init/main.c to use new set_cpus_allowed_ptr().
Depends on:
[sched-devel]: sched: add new set_cpus_allowed_ptr function
Cc: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This essentially reverts commit 71fc47a9ad
("ACPI: basic initramfs DSDT override support"), because the code simply
isn't ready.
It did ugly things to the init sequence to populate the rootfs image
early, but that just ended up showing other problems with the whole
approach. The fact is, the VFS layer simply isn't initialized this
early, and the relevant ACPI code should either run much later, or this
shouldn't be done at all.
For 2.6.25, we'll just pick the latter option. We can revisit this
concept later if necessary.
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Tilman Schmidt <tilman@imap.cc>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Renninger <trenn@suse.de>
Cc: Eric Piel <eric.piel@tremplin-utc.net>
Cc: Len Brown <len.brown@intel.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Markus Gaugusch <dsdt@gaugusch.at>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Keith Mannthey said:
The parameter hotadd_percent is setup right but there is a "Malformed
early option 'numa'" message.
Rusty Russell said:
This happens when the function registered with early_param() returns
non-zero. __setup() functions return 1 if OK, module_param() and
early_param() return 0 or a -ve error code.
For instance:
Linux version 2.6.25-rc3-t (raa@steel) (gcc version 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)) #22 SMP PREEMPT Tue Feb 26
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 00000000000a0000 (usable)
BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 000000003fff0000 (usable)
BIOS-e820: 000000003fff0000 - 000000003fff3000 (ACPI NVS)
BIOS-e820: 000000003fff3000 - 0000000040000000 (ACPI data)
BIOS-e820: 00000000fec00000 - 0000000100000000 (reserved)
Malformed early option 'loglevel'
127MB HIGHMEM available.
896MB LOWMEM available.
Command line:
BOOT_IMAGE=2.6.25-t ro root=809 ro console=ttyS0,57600n8 console=tty0 loglevel=5
Acked-by: Yinghai Lu <yhlu.kernel@gmai.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Keith Mannthey <kmannth@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
DEBUG_PAGEALLOC must not be enabled before mem_init(). Before this
point there is nothing to allocate.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
So we can use them for the early console like console=uart8250 or
earlycon=uart8250 or early_printk
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As Eric pointed out, there is no problem with init starting with sid == pgid
== 0, and this was historical linux behavior changed in 2.6.18.
Remove kernel_init()->__set_special_pids(), this is unneeded and complicates
the rules for sys_setsid().
This change and the previous change in daemonize() mean that /sbin/init does
not need the special "session != 1" hack in sys_setsid() any longer. We can't
remove this check yet, we should cleanup copy_process(CLONE_NEWPID) first, so
update the comment only.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Change set_special_pids() to work with struct pid, not pid_t from global name
space. This again speedups and imho cleanups the code, also a preparation for
the next patch.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The basics of DSDT from initramfs. In case this option is selected,
populate_rootfs() is called a bit earlier to have the initramfs content
available during ACPI initialization.
This is a very similar path to the one available at
http://gaugusch.at/kernel.shtml but with some update in the
documentation, default set to No and the change of populate_rootfs() the
"Jeff Mahony way" (which avoids reading the initramfs twice).
Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Eric Piel <eric.piel@tremplin-utc.net>
Signed-off-by: Len Brown <len.brown@intel.com>
Add a proper prototype for signals_init() in include/linux/signal.h
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
get more testing of the c_p_a() code done by not turning off
PSE on DEBUG_PAGEALLOC.
this simplifies the early pagetable setup code, and tests
the largepage-splitup code quite heavily.
In the end, all the largepages will be split up pretty quickly,
so there's no difference to how DEBUG_PAGEALLOC worked before.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
On VMs implemented using JITs that cache translated code changing the lock
prefixes is a quite costly operation that forces the JIT to throw away and
retranslate a lot of code.
Previously a SMP kernel would rewrite the locks once for each CPU which
is quite unnecessary. This patch changes the code to never switch at boot in
the normal case (SMP kernel booting with >1 CPU) or only once for SMP kernel
on UP.
This makes a significant difference in boot up performance on AMD SimNow!
Also I expect it to be a little faster on native systems too because a smp
switch does a lot of text_poke()s which each synchronize the pipeline.
v1->v2: Rename max_cpus
v1->v2: Fix off by one in UP check (Thomas Gleixner)
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The use of the __GENERIC_PERCPU is a bit problematic since arches
may want to run their own percpu setup while using the generic
percpu definitions. Replace it through a kconfig variable.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch implements a Refcount + Waitqueue based model for
cpu-hotplug.
Now, a thread which wants to prevent cpu-hotplug, will bump up a global
refcount and the thread which wants to perform a cpu-hotplug operation
will block till the global refcount goes to zero.
The readers, if any, during an ongoing cpu-hotplug operation are blocked
until the cpu-hotplug operation is over.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Paul Jackson <pj@sgi.com> [For !CONFIG_HOTPLUG_CPU ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch adds a proper prototype for migration_init() in
include/linux/sched.h
Since there's no point in always returning 0 to a caller that doesn't check
the return value it also changes the function to return void.
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The header file <linux/compiler.h> already enforces a suitably recent
version of gcc, so there's no point checking for that again.
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Generic Process Control Groups
--------------------------
There have recently been various proposals floating around for
resource management/accounting and other task grouping subsystems in
the kernel, including ResGroups, User BeanCounters, NSProxy
cgroups, and others. These all need the basic abstraction of being
able to group together multiple processes in an aggregate, in order to
track/limit the resources permitted to those processes, or control
other behaviour of the processes, and all implement this grouping in
different ways.
This patchset provides a framework for tracking and grouping processes
into arbitrary "cgroups" and assigning arbitrary state to those
groupings, in order to control the behaviour of the cgroup as an
aggregate.
The intention is that the various resource management and
virtualization/cgroup efforts can also become task cgroup
clients, with the result that:
- the userspace APIs are (somewhat) normalised
- it's easier to test e.g. the ResGroups CPU controller in
conjunction with the BeanCounters memory controller, or use either of
them as the resource-control portion of a virtual server system.
- the additional kernel footprint of any of the competing resource
management systems is substantially reduced, since it doesn't need
to provide process grouping/containment, hence improving their
chances of getting into the kernel
This patch:
Add the main task cgroups framework - the cgroup filesystem, and the
basic structures for tracking membership and associating subsystem state
objects to tasks.
Signed-off-by: Paul Menage <menage@google.com>
Cc: Serge E. Hallyn <serue@us.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Kirill Korotaev <dev@openvz.org>
Cc: Herbert Poetzl <herbert@13thfloor.at>
Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alexey Dobriyan reports that maxcpus=1 is still broken in 2.6.23-rc4:
if CONFIG_HOTPLUG_CPU is not set, x86_64 bootup oopses in show_stat() -
for_each_possible_cpu accesses a per-cpu area which was never set up.
Alexey identified commit 61ec7567db
(ACPI: boot correctly with "nosmp" or "maxcpus=0") as the origin;
but it's not really to blame, just exposes a bug in 2.6.23-rc1's commit
8b3b295502 (Especially when !CONFIG_HOTPLUG_CPU,
avoid needlessy allocating resources for CPUs that can never become available).
rc1's test for max_cpus < 2 in start_kernel() wasn't working because
max_cpus was still NR_CPUS at that point: until rc4 moved the maxcpus
parsing earlier. Now it sets cpu_possible_map to 1 before allocating
all possible per-cpu areas; then smp_init() expands cpu_possible_map
to cpu_present_map (0xf in my case) later on.
rc1's commit has good intentions, but expects cpu_present_map to be
limited by maxcpus, which is only the case on i386. cpus_and(possible,
possible,present) might be good, but needs an audit of cpu_present_map
uses - there may well be assumptions that any cpu present is possible.
So stay safe for now and just revert those #ifndef CONFIG_HOTPLUG_CPU
optimizations in rc1's commit.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Alexey Dobriyan <adobriyan@sw.ru>
Cc: Len Brown <lenb@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 61ec7567db ('ACPI: boot correctly
with "nosmp" or "maxcpus=0"') broke 'maxcpus=' handling on x86[-64].
maxcpus=N is now having no effect on x86_64, and freezing bootup on i386
(because of inconsistency with the separate maxcpus parsing down in
arch/i386, I guess). That's because early_param parsing is a little
different from __setup parsing, and needs the "=" omitted: then it seems
to work as the original commit intended (no mention of IO-APIC in
/proc/interrupts when maxcpus=0).
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Len Brown <len.brown@intel.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In MPS mode, "nosmp" and "maxcpus=0" boot a UP kernel with IOAPIC disabled.
However, in ACPI mode, these parameters didn't completely disable
the IO APIC initialization code and boot failed.
init/main.c:
Disable the IO_APIC if "nosmp" or "maxcpus=0"
undefine disable_ioapic_setup() when it doesn't apply.
i386:
delete ioapic_setup(), it was a duplicate of parse_noapic()
delete undefinition of disable_ioapic_setup()
x86_64:
rename disable_ioapic_setup() to parse_noapic() to match i386
define disable_ioapic_setup() in header to match i386
http://bugzilla.kernel.org/show_bug.cgi?id=1641
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
Nick Piggin