This is a new slab allocator which was motivated by the complexity of the
existing code in mm/slab.c. It attempts to address a variety of concerns
with the existing implementation.
A. Management of object queues
A particular concern was the complex management of the numerous object
queues in SLAB. SLUB has no such queues. Instead we dedicate a slab for
each allocating CPU and use objects from a slab directly instead of
queueing them up.
B. Storage overhead of object queues
SLAB Object queues exist per node, per CPU. The alien cache queue even
has a queue array that contain a queue for each processor on each
node. For very large systems the number of queues and the number of
objects that may be caught in those queues grows exponentially. On our
systems with 1k nodes / processors we have several gigabytes just tied up
for storing references to objects for those queues This does not include
the objects that could be on those queues. One fears that the whole
memory of the machine could one day be consumed by those queues.
C. SLAB meta data overhead
SLAB has overhead at the beginning of each slab. This means that data
cannot be naturally aligned at the beginning of a slab block. SLUB keeps
all meta data in the corresponding page_struct. Objects can be naturally
aligned in the slab. F.e. a 128 byte object will be aligned at 128 byte
boundaries and can fit tightly into a 4k page with no bytes left over.
SLAB cannot do this.
D. SLAB has a complex cache reaper
SLUB does not need a cache reaper for UP systems. On SMP systems
the per CPU slab may be pushed back into partial list but that
operation is simple and does not require an iteration over a list
of objects. SLAB expires per CPU, shared and alien object queues
during cache reaping which may cause strange hold offs.
E. SLAB has complex NUMA policy layer support
SLUB pushes NUMA policy handling into the page allocator. This means that
allocation is coarser (SLUB does interleave on a page level) but that
situation was also present before 2.6.13. SLABs application of
policies to individual slab objects allocated in SLAB is
certainly a performance concern due to the frequent references to
memory policies which may lead a sequence of objects to come from
one node after another. SLUB will get a slab full of objects
from one node and then will switch to the next.
F. Reduction of the size of partial slab lists
SLAB has per node partial lists. This means that over time a large
number of partial slabs may accumulate on those lists. These can
only be reused if allocator occur on specific nodes. SLUB has a global
pool of partial slabs and will consume slabs from that pool to
decrease fragmentation.
G. Tunables
SLAB has sophisticated tuning abilities for each slab cache. One can
manipulate the queue sizes in detail. However, filling the queues still
requires the uses of the spin lock to check out slabs. SLUB has a global
parameter (min_slab_order) for tuning. Increasing the minimum slab
order can decrease the locking overhead. The bigger the slab order the
less motions of pages between per CPU and partial lists occur and the
better SLUB will be scaling.
G. Slab merging
We often have slab caches with similar parameters. SLUB detects those
on boot up and merges them into the corresponding general caches. This
leads to more effective memory use. About 50% of all caches can
be eliminated through slab merging. This will also decrease
slab fragmentation because partial allocated slabs can be filled
up again. Slab merging can be switched off by specifying
slub_nomerge on boot up.
Note that merging can expose heretofore unknown bugs in the kernel
because corrupted objects may now be placed differently and corrupt
differing neighboring objects. Enable sanity checks to find those.
H. Diagnostics
The current slab diagnostics are difficult to use and require a
recompilation of the kernel. SLUB contains debugging code that
is always available (but is kept out of the hot code paths).
SLUB diagnostics can be enabled via the "slab_debug" option.
Parameters can be specified to select a single or a group of
slab caches for diagnostics. This means that the system is running
with the usual performance and it is much more likely that
race conditions can be reproduced.
I. Resiliency
If basic sanity checks are on then SLUB is capable of detecting
common error conditions and recover as best as possible to allow the
system to continue.
J. Tracing
Tracing can be enabled via the slab_debug=T,<slabcache> option
during boot. SLUB will then protocol all actions on that slabcache
and dump the object contents on free.
K. On demand DMA cache creation.
Generally DMA caches are not needed. If a kmalloc is used with
__GFP_DMA then just create this single slabcache that is needed.
For systems that have no ZONE_DMA requirement the support is
completely eliminated.
L. Performance increase
Some benchmarks have shown speed improvements on kernbench in the
range of 5-10%. The locking overhead of slub is based on the
underlying base allocation size. If we can reliably allocate
larger order pages then it is possible to increase slub
performance much further. The anti-fragmentation patches may
enable further performance increases.
Tested on:
i386 UP + SMP, x86_64 UP + SMP + NUMA emulation, IA64 NUMA + Simulator
SLUB Boot options
slub_nomerge Disable merging of slabs
slub_min_order=x Require a minimum order for slab caches. This
increases the managed chunk size and therefore
reduces meta data and locking overhead.
slub_min_objects=x Mininum objects per slab. Default is 8.
slub_max_order=x Avoid generating slabs larger than order specified.
slub_debug Enable all diagnostics for all caches
slub_debug=<options> Enable selective options for all caches
slub_debug=<o>,<cache> Enable selective options for a certain set of
caches
Available Debug options
F Double Free checking, sanity and resiliency
R Red zoning
P Object / padding poisoning
U Track last free / alloc
T Trace all allocs / frees (only use for individual slabs).
To use SLUB: Apply this patch and then select SLUB as the default slab
allocator.
[hugh@veritas.com: fix an oops-causing locking error]
[akpm@linux-foundation.org: various stupid cleanups and small fixes]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The nr_cpu_ids value is currently only calculated in smp_init. However, it
may be needed before (SLUB needs it on kmem_cache_init!) and other kernel
components may also want to allocate dynamically sized per cpu array before
smp_init. So move the determination of possible cpus into sched_init()
where we already loop over all possible cpus early in boot.
Also initialize both nr_node_ids and nr_cpu_ids with the highest value they
could take. If we have accidental users before these values are determined
then the current valud of 0 may cause too small per cpu and per node arrays
to be allocated. If it is set to the maximum possible then we only waste
some memory for early boot users.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild: (38 commits)
kconfig: fix mconf segmentation fault
kbuild: enable use of code from a different dir
kconfig: error out if recursive dependencies are found
kbuild: scripts/basic/fixdep segfault on pathological string-o-death
kconfig: correct minor typo in Kconfig warning message.
kconfig: fix path to modules.txt in Kconfig help
usr/Kconfig: fix typo
kernel-doc: alphabetically-sorted entries in index.html of 'htmldocs'
kbuild: be more explicit on missing .config file
kbuild: clarify the creation of the LOCALVERSION_AUTO string.
kbuild: propagate errors from find in scripts/gen_initramfs_list.sh
kconfig: refer to qt3 if we cannot find qt libraries
kbuild: handle compressed cpio initramfs-es
kbuild: ignore section mismatch warning for references from .paravirtprobe to .init.text
kbuild: remove stale comment in modpost.c
kbuild/mkuboot.sh: allow spaces in CROSS_COMPILE
kbuild: fix make mrproper for Documentation/DocBook/man
kbuild: remove kconfig binaries during make mrproper
kconfig/menuconfig: do not hardcode '.config'
kbuild: override build timestamp & version
...
Clarify the creation of the LOCALVERSION_AUTO string during kernel
configuration, and fix a couple typoes while we're there.
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
In init/main.c we have a reference from rest_init() to .init.text
which is intentional.
Rename the function 'init' to 'kernel_init' to make it a
kernel wide unique symbol and whitelist the reference.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Let's allow page-alignment in general for per-cpu data (wanted by Xen, and
Ingo suggested KVM as well).
Because larger alignments can use more room, we increase the max per-cpu
memory to 64k rather than 32k: it's getting a little tight.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
initramfs ended up depending on BLOCK:
INITRAMFS_SOURCE <-- BLK_DEV_INITRD <-- BLOCK
This inhibits use of customized-initramfs-over-ramfs without block layer
(ramfs would still be enabled), useful in embedded applications.
Move BLK_DEV_INITRD out of 'drivers/block/Kconfig' and into 'init/Kconfig',
make it unconditional.
Signed-off-by: Dimitri Gorokhovik <dimitri.gorokhovik@free.fr>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We frequently need the maximum number of possible processors in order to
allocate arrays for all processors. So far this was done using
highest_possible_processor_id(). However, we do need the number of
processors not the highest id. Moreover the number was so far dynamically
calculated on each invokation. The number of possible processors does not
change when the system is running. We can therefore calculate that number
once.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Frederik Deweerdt <frederik.deweerdt@gmail.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
powerpc gets:
init/main.c: In function `do_basic_setup':
init/main.c:714: warning: implicit declaration of function `init_irq_proc'
but we cannot include linux/irq.h in generic code.
Fix it by moving the declaration into linux/interrupt.h instead.
And make sure all code that defines init_irq_proc() is including
linux/interrupt.h.
And nuke an ifdef-in-C
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With Ingo Molnar <mingo@elte.hu>
The tick-management code is the first user of the clockevents layer. It takes
clock event devices from the clock events core and uses them to provide the
periodic tick.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With this change the sysctl inodes can be cached and nothing needs to be done
when removing a sysctl table.
For a cost of 2K code we will save about 4K of static tables (when we remove
de from ctl_table) and 70K in proc_dir_entries that we will not allocate, or
about half that on a 32bit arch.
The speed feels about the same, even though we can now cache the sysctl
dentries :(
We get the core advantage that we don't need to have a 1 to 1 mapping between
ctl table entries and proc files. Making it possible to have /proc/sys vary
depending on the namespace you are in. The currently merged namespaces don't
have an issue here but the network namespace under /proc/sys/net needs to have
different directories depending on which network adapters are visible. By
simply being a cache different directories being visible depending on who you
are is trivial to implement.
[akpm@osdl.org: fix uninitialised var]
[akpm@osdl.org: fix ARM build]
[bunk@stusta.de: make things static]
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is just a simple cleanup to keep kernel/sysctl.c from getting to crowded
with special cases, and by keeping all of the ipc logic to together it makes
the code a little more readable.
[gcoady.lk@gmail.com: build fix]
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Serge E. Hallyn <serue@us.ibm.com>
Cc: Herbert Poetzl <herbert@13thfloor.at>
Cc: Kirill Korotaev <dev@sw.ru>
Signed-off-by: Grant Coady <gcoady.lk@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
After Al Viro (finally) succeeded in removing the sched.h #include in module.h
recently, it makes sense again to remove other superfluous sched.h includes.
There are quite a lot of files which include it but don't actually need
anything defined in there. Presumably these includes were once needed for
macros that used to live in sched.h, but moved to other header files in the
course of cleaning it up.
To ease the pain, this time I did not fiddle with any header files and only
removed #includes from .c-files, which tend to cause less trouble.
Compile tested against 2.6.20-rc2 and 2.6.20-rc2-mm2 (with offsets) on alpha,
arm, i386, ia64, mips, powerpc, and x86_64 with allnoconfig, defconfig,
allmodconfig, and allyesconfig as well as a few randconfigs on x86_64 and all
configs in arch/arm/configs on arm. I also checked that no new warnings were
introduced by the patch (actually, some warnings are removed that were emitted
by unnecessarily included header files).
Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
o init() is a non __init function in .text section but it calls many
functions which are in .init.text section. Hence MODPOST generates lots
of cross reference warnings on i386 if compiled with CONFIG_RELOCATABLE=y
WARNING: vmlinux - Section mismatch: reference to .init.text:smp_prepare_cpus from .text between 'init' (at offset 0xc0101049) and 'rest_init'
WARNING: vmlinux - Section mismatch: reference to .init.text:migration_init from .text between 'init' (at offset 0xc010104e) and 'rest_init'
WARNING: vmlinux - Section mismatch: reference to .init.text:spawn_ksoftirqd from .text between 'init' (at offset 0xc0101053) and 'rest_init'
o This patch breaks down init() in two parts. One part which can go
in .init.text section and can be freed and other part which has to
be non __init(init_post()). Now init() calls init_post() and init_post()
does not call any functions present in .init sections. Hence getting
rid of warnings.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Current implementation stores a static command-line buffer allocated to
COMMAND_LINE_SIZE size. Most architectures stores two copies of this buffer,
one for future reference and one for parameter parsing.
Current kernel command-line size for most architecture is much too small for
module parameters, video settings, initramfs paramters and much more. The
problem is that setting COMMAND_LINE_SIZE to a grater value, allocates static
buffers.
In order to allow a greater command-line size, these buffers should be
dynamically allocated or marked as init disposable buffers, so unused memory
can be released.
This patch renames the static saved_command_line variable into
boot_command_line adding __initdata attribute, so that it can be disposed
after initialization. This rename is required so applications that use
saved_command_line will not be affected by this change.
It reintroduces saved_command_line as dynamically allocated buffer to match
the data in boot_command_line.
It also mark secondary command-line buffer as __initdata, and copies it to
dynamically allocated static_command_line buffer components may hold reference
to it after initialization.
This patch is for linux-2.6.20-rc4-mm1 and is divided to target each
architecture. I could not check this in any architecture so please forgive me
if I got it wrong.
The per-architecture modification is very simple, use boot_command_line in
place of saved_command_line. The common code is the change into dynamic
command-line.
This patch:
1. Rename saved_command_line into boot_command_line, mark as init
disposable.
2. Add dynamic allocated saved_command_line.
3. Add dynamic allocated static_command_line.
4. During startup copy: boot_command_line into saved_command_line. arch
command_line into static_command_line.
5. Parse static_command_line and not arch command_line, so arch
command_line may be freed.
Signed-off-by: Alon Bar-Lev <alon.barlev@gmail.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ian Molton <spyro@f2s.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Cc: Richard Curnow <rc@rc0.org.uk>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp>
Cc: Chris Zankel <chris@zankel.net>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Since they depends on TASKSTATS, it would be nice to move them closer to
another options depending on TASKSTATS.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Remove the last (and commented out) invocation of the obsolete
smp_commence() call.
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Acked-by: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The file init/initramfs.c is always compiled and linked in the kernel
vmlinux even when BLK_DEV_RAM and BLK_DEV_INITRD are disabled and the
system isn't using any form of an initramfs or initrd. In this situation
the code is only used to unpack a (static) default initial rootfilesystem.
The current init/initramfs.c code. usr/initramfs_data.o compiles to a size
of ~15 kbytes. Disabling BLK_DEV_RAM and BLK_DEV_INTRD shrinks the kernel
code size with ~60 Kbytes.
This patch avoids compiling in the code and data for initramfs support if
CONFIG_BLK_DEV_INITRD is not defined. Instead of the initramfs code and
data it uses a small routine in init/noinitramfs.c to setup an initial
static default environment for mounting a rootfilesystem later on in the
kernel initialisation process. The new code is: 164 bytes of size.
The patch is separated in two parts:
1) doesn't compile initramfs code when CONFIG_BLK_DEV_INITRD is not set
2) changing all plaforms vmlinux.lds.S files to not reserve an area of
PAGE_SIZE when CONFIG_BLK_DEV_INITRD is not set.
[deweerdt@free.fr: warning fix]
Signed-off-by: Jean-Paul Saman <jean-paul.saman@nxp.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add retain_initrd option to control freeing of initrd memory after
extraction. By default, free memory as previously.
The first boot will need to hold a copy of the in memory fs for the second
boot. This image can be large (much larger than the kernel), hence we can
save time when the memory loader is slow. Also, it reduces the memory
footprint while extracting the first boot since you don't need another copy
of the fs.
Signed-off-by: Michael Neuling <mikey@neuling.org>
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It might save a few bytes after bootup, but it causes the string to be
linked in at the end of the final vmlinux image, which defeats the whole
point of doing all this, namely allowing some broken user-space binaries
to search for the kernel version string in the kernel binary.
So just remove the __init specifier.
Cc: Olaf Hering <olaf@aepfle.de>
Cc: Jean Delvare <khali@linux-fr.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Andrey Borzenkov <arvidjaar@mail.ru>
Cc: Andrew Morton <akpm@osdl.org>
Acked-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
o Some functions which should have been in init sections as they are called
only once. Put them in init sections. Otherwise MODPOST generates warning
as these functions are placed in .text and they end up accessing something
in init sections.
WARNING: vmlinux - Section mismatch: reference to .init.text:migration_init
from .text between 'do_pre_smp_initcalls' (at offset 0xc01000d1) and
'run_init_process'
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Revert previous attempts at messing with the linux banner string and
simply use a separate format string for proc.
Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Acked-by: Olaf Hering <olaf@aepfle.de>
Acked-by: Jean Delvare <khali@linux-fr.org>
Cc: Andrey Borzenkov <arvidjaar@mail.ru>
Cc: Andrew Morton <akpm@osdl.org>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Herbert Poetzl <herbert@13thfloor.at>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The calls made by parse_parms to other initialization code might enable
interrupts again way too early.
Having interrupts on this early can make systems PANIC when they initialize
the IRQ controllers (which happens later in the code). This patch detects
that irq's are enabled again, barfs about it and disables them again as a
safety net.
[akpm@osdl.org: cleanups]
Signed-off-by: Ard van Breemen <ard@telegraafnet.nl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
compile.h is created super-late in the build. But proc_misc.c want to include
it, and it's generally not sane to have a header file in include/linux be
created at the end of the build: it's either not present or, worse, wrong for
most of the build.
So the patch arranges for compile.h to be built at the start of the build
process. It also consolidates the compile.h rules with those for version.h
and utsname.h, so they all get built together.
I hope. My chances of having got this right are about 2%.
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is to disallow to make SLOB with SMP or SPARSEMEM. This avoids latent
troubles of SLOB with SLAB_DESTROY_BY_RCU. And fix compile error.
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The VM event counters, enabled by CONFIG_VM_EVENT_COUNTERS, which provides
VM event counters in /proc/vmstat, has become more essential to
non-EMBEDDED kernel configurations than they were in the past. Comments in
the code and the Kconfig configuration explanation were stale, downplaying
their role excessively.
Refresh those comments to correctly reflect the current role of VM event
counters.
Signed-off-by: Paul Jackson <pj@sgi.com>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a prototype for driver_init() in include/linux/device.h.
Also remove a static function of the same name in drivers/acpi/ibm_acpi.c to
ibm_acpi_driver_init() to fix the namespace collision.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
We should not initialize rootfs before all the core initializers have
run. So do it as a separate stage just before starting the regular
driver initializers.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As reported by Andy Whitcroft, at least the SLES9 initrd build process
depends on getting the kernel version from the kernel binary. It does
that by simply trawling the binary and looking for the signature of the
"linux_banner" string (the string "Linux version " to be exact. Which
is really broken in itself, but whatever..)
That got broken when the string was changed to allow /proc/version to
change the UTS release information dynamically, and "get_kernel_version"
thus returned "%s" (see commit a2ee8649ba6d71416712e798276bf7c40b64e6e5:
"[PATCH] Fix linux banner utsname information").
This just restores "linux_banner" as a static string, which should fix
the version finding. And /proc/version simply uses a different string.
To avoid wasting even that miniscule amount of memory, the early boot
string should really be marked __initdata, but that just causes the same
bug in SLES9 to re-appear, since it will then find other occurrences of
"Linux version " first.
Cc: Andy Whitcroft <apw@shadowen.org>
Acked-by: Herbert Poetzl <herbert@13thfloor.at>
Cc: Andi Kleen <ak@suse.de>
Cc: Andrew Morton <akpm@osdl.org>
Cc: Steve Fox <drfickle@us.ibm.com>
Acked-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The present per-task IO accounting isn't very useful. It simply counts the
number of bytes passed into read() and write(). So if a process reads 1MB
from an already-cached file, it is accused of having performed 1MB of I/O,
which is wrong.
(David Wright had some comments on the applicability of the present logical IO accounting:
For billing purposes it is useless but for workload analysis it is very
useful
read_bytes/read_calls average read request size
write_bytes/write_calls average write request size
read_bytes/read_blocks ie logical/physical can indicate hit rate or thrashing
write_bytes/write_blocks ie logical/physical guess since pdflush writes can
be missed
I often look for logical larger than physical to see filesystem cache
problems. And the bytes/cpusec can help find applications that are
dominating the cache and causing slow interactive response from page cache
contention.
I want to find the IO intensive applications and make sure they are doing
efficient IO. Thus the acctcms(sysV) or csacms command would give the high
IO commands).
This patchset adds new accounting which tries to be more accurate. We account
for three things:
reads:
attempt to count the number of bytes which this process really did cause
to be fetched from the storage layer. Done at the submit_bio() level, so it
is accurate for block-backed filesystems. I also attempt to wire up NFS and
CIFS.
writes:
attempt to count the number of bytes which this process caused to be sent
to the storage layer. This is done at page-dirtying time.
The big inaccuracy here is truncate. If a process writes 1MB to a file
and then deletes the file, it will in fact perform no writeout. But it will
have been accounted as having caused 1MB of write.
So...
cancelled_writes:
account the number of bytes which this process caused to not happen, by
truncating pagecache.
We _could_ just subtract this from the process's `write' accounting. But
that means that some processes would be reported to have done negative
amounts of write IO, which is silly.
So we just report the raw number and punt this decision up to userspace.
Now, we _could_ account for writes at the physical I/O level. But
- This would require that we track memory-dirtying tasks at the per-page
level (would require a new pointer in struct page).
- It would mean that IO statistics for a process are usually only available
long after that process has exitted. Which means that we probably cannot
communicate this info via taskstats.
This patch:
Wire up the kernel-private data structures and the accessor functions to
manipulate them.
Cc: Jay Lan <jlan@sgi.com>
Cc: Shailabh Nagar <nagar@watson.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Chris Sturtivant <csturtiv@sgi.com>
Cc: Tony Ernst <tee@sgi.com>
Cc: Guillaume Thouvenin <guillaume.thouvenin@bull.net>
Cc: David Wright <daw@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a per pid_namespace child-reaper. This is needed so processes are reaped
within the same pid space and do not spill over to the parent pid space. Its
also needed so containers preserve existing semantic that pid == 1 would reap
orphaned children.
This is based on Eric Biederman's patch: http://lkml.org/lkml/2006/2/6/285
Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Signed-off-by: Cedric Le Goater <clg@fr.ibm.com>
Cc: Kirill Korotaev <dev@openvz.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Herbert Poetzl <herbert@13thfloor.at>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
utsname information is shown in the linux banner, which also is used for
/proc/version (which can have different utsname values inside a uts
namespaces). this patch makes the varying data arguments and changes the
string to a format string, using those arguments.
Signed-off-by: Herbert Poetzl <herbert@13thfloor.at>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Keith says
Compiling 2.6.19-rc6 with gcc version 4.1.0 (SUSE Linux), wait_hpet_tick is
optimized away to a never ending loop and the kernel hangs on boot in timer
setup.
0000001a <wait_hpet_tick>:
1a: 55 push %ebp
1b: 89 e5 mov %esp,%ebp
1d: eb fe jmp 1d <wait_hpet_tick+0x3>
This is not a problem with gcc 3.3.5. Adding barrier() calls to
wait_hpet_tick does not help, making the variables volatile does.
And the consensus is that gcc-4.1.0 is busted. Suse went and shipped
gcc-4.1.0 so we cannot ban it. Add a warning.
Cc: Keith Owens <kaos@ocs.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It turns out that the "-c" option of cpio is highly unportable even between
distros let alone unix variants, and may actually make the wrong type of
cpio archive. I just wasted quite some time on this, and the kernel can
detect this and warn about it (it's __init memory so it gets thrown away
and thus there is no runtime overhead)
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Move process freezing functions from include/linux/sched.h to freezer.h, so
that modifications to the freezer or the kernel configuration don't require
recompiling just about everything.
[akpm@osdl.org: fix ueagle driver]
Signed-off-by: Nigel Cunningham <nigel@suspend2.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Both lhype and Xen want to call the core of the x86 cpu detect code before
calling start_kernel.
(extracted from larger patch)
AK: folded in start_kernel header patch
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Provide a way to support older versions of udev that are shipped in
older distros. If this option is disabled, it will also turn off the
compatible symlinks in sysfs that older programs might rely on.
When in doubt, or if running a distro older than 2006, say Yes here.
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The basic issue is that despite have been deprecated and warned about as a
very bad thing in the man pages since its inception there are a few real
users of sys_sysctl. It was my assumption that because sysctl had been
deprecated for all of 2.6 there would be no user space users by this point,
so I initially gave sys_sysctl a very short deprecation period.
Now that I know there are a few real users the only sane way to proceed
with deprecation is to push the time limit out to a year or two work and
work with distributions that have big testing pools like fedora core to
find these last remaining users.
Which means that the sys_sysctl interface needs to be maintained in the
meantime.
Since I have provided a technical measure that allows us to add new sysctl
entries without reserving more binary numbers I believe that is enough to
fix the sys_sysctl binary interface maintenance problems, because there is
no longer a need to change the binary interface at all.
Since the sys_sysctl implementation needs to stay around for a while and
the worst of the maintenance issues that caused us to occasionally break
the ABI have been addressed I don't see any advantage in continuing with
the removal of sys_sysctl.
So instead of merely increasing the deprecation period this patch removes
the deprecation of sys_sysctl and modifies the kernel to compile the code
in by default.
With committing to maintain sys_sysctl we get all of the advantages of a
fast interface for anything that needs it. Currently sys_sysctl is about
5x faster than /proc/sys, for the same string data.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This changes the dwarf2 unwinder to do a binary search for CIEs
instead of a linear work. The linker is unfortunately not
able to build a proper lookup table at link time, instead it creates
one at runtime as soon as the bootmem allocator is usable (so you'll continue
using the linear lookup for the first [hopefully] few calls).
The code should be ready to utilize a build-time created table once
a fixed linker becomes available.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
This should make sure that, for UML, host's configuration files are not
considered, which avoids various pains to the user. Our dependency are such
that the obtained Kconfig will be valid and will lead to successful
compilation - however they cannot prevent an user from disabling any boot
device, and if an option is not set in the read .config (say
/boot/config-XXX), with make menuconfig ARCH=um, it is not set. This always
disables UBD and all console I/O channels, which leads to non-working UML
kernels, so this bothers users - especially now, since it will happen on
almost every machine (/boot/config-`uname -r` exists almost on every machine).
It can be workarounded with make defconfig ARCH=um, but it is non-obvious and
can be avoided, so please _do_ merge this patch.
Given the existence of options, it could be interesting to implement
(additionally) "option required" - with it, Kconfig will refuse reading a
.config file (from wherever it comes) if the given option is not set. With
this, one could mark with it the option characteristic of the given
architecture (it was an old proposal of Roman Zippel, when I pointed out our
problem):
config UML
option required
default y
However this should be further discussed:
*) for x86, it must support constructs like:
==arch/i386/Kconfig==
config 64BIT
option required
default n
where Kconfig must require that CONFIG_64BIT is disabled or not present in the
read .config.
*) do we want to do such checks only for the starting defconfig or also for
.config? Which leads to:
*) I may want to port a x86_64 .config to x86 and viceversa, or even among more
different archs. Should that be allowed, and in which measure (the user may
force skipping the check for a .config or it is only given a warning by
default)?
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: <kbuild-devel@lists.sourceforge.net>
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Once upon a time we needed to fixed limit to the number of md devices,
probably because we preallocated some array. This need no longer exists, but
we still have an arbitrary limit.
So remove MAX_MD_DEVS and allow as many devices as we can fit into the 'minor'
part of a device number.
Also remove some useless noise at init time (which reports MAX_MD_DEVS) and
remove MD_THREAD_NAME_MAX which hasn't been used for a while.
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 a few places in the kernel where the init task is signaled. The
ctrl+alt+del sequence is one them. It kills a task, usually init, using a
cached pid (cad_pid).
This patch replaces the pid_t by a struct pid to avoid pid wrap around
problem. The struct pid is initialized at boot time in init() and can be
modified through systctl with
/proc/sys/kernel/cad_pid
[ I haven't found any distro using it ? ]
It also introduces a small helper routine kill_cad_pid() which is used
where it seemed ok to use cad_pid instead of pid 1.
[akpm@osdl.org: cleanups, build fix]
Signed-off-by: Cedric Le Goater <clg@fr.ibm.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The use of execve() in the kernel is dubious, since it relies on the
__KERNEL_SYSCALLS__ mechanism that stores the result in a global errno
variable. As a first step of getting rid of this, change all users to a
global kernel_execve function that returns a proper error code.
This function is a terrible hack, and a later patch removes it again after the
kernel syscalls are gone.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Andi Kleen <ak@muc.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ian Molton <spyro@f2s.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Hirokazu Takata <takata.hirokazu@renesas.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Cc: Richard Curnow <rc@rc0.org.uk>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp>
Cc: Chris Zankel <chris@zankel.net>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch set allows to unshare IPCs and have a private set of IPC objects
(sem, shm, msg) inside namespace. Basically, it is another building block of
containers functionality.
This patch implements core IPC namespace changes:
- ipc_namespace structure
- new config option CONFIG_IPC_NS
- adds CLONE_NEWIPC flag
- unshare support
[clg@fr.ibm.com: small fix for unshare of ipc namespace]
[akpm@osdl.org: build fix]
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
Signed-off-by: Cedric Le Goater <clg@fr.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch defines the uts namespace and some manipulators.
Adds the uts namespace to task_struct, and initializes a
system-wide init namespace.
It leaves a #define for system_utsname so sysctl will compile.
This define will be removed in a separate patch.
[akpm@osdl.org: build fix, cleanup]
Signed-off-by: Serge Hallyn <serue@us.ibm.com>
Cc: Kirill Korotaev <dev@openvz.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Herbert Poetzl <herbert@13thfloor.at>
Cc: Andrey Savochkin <saw@sw.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Christoph Lameter