Further, there's an additional issue that I saw in testing:
ipoib_mcast_send may get called when priv->broadcast is NULL (e.g. if
the device was downed and then upped internally because of a port
event).
If this happends and the send-only join request gets completed before
priv->broadcast is set, we get an oops.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
IPoIB: Don't start send-only joins while multicast thread is stopped
Fix the following race scenario:
- Device is up.
- Port event or set mcast list triggers ipoib_mcast_stop_thread,
this cancels the query and waits on mcast "done" completion.
- Completion is called and "done" is set.
- Meanwhile, ipoib_mcast_send arrives and starts a new query,
re-initializing "done".
Fix this by adding a "multicast started" bit and checking it before
starting a send-only join.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Ralph Campbell [Fri, 3 Feb 2006 22:28:48 +0000 (14:28 -0800)]
IB/mad: Handle DR SMPs with a LID routed part
Fix handling of directed route SMPs with a beginning or ending LID
routed part.
Signed-off-by: Ralph Campbell <ralphc@pathscale.com> Signed-off-by: Hal Rosenstock <halr@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
[PATCH] fbcon: Fix screen artifacts when moving cursor
When moving the cursor by writing to /dev/vcs*, the old cursor image is not
erased. Fix by hiding the cursor first before moving the cursor to the new
position.
Jeff Mahoney [Wed, 1 Feb 2006 11:06:51 +0000 (03:06 -0800)]
[PATCH] reiserfs: reiserfs: check for files > 2GB on 3.5.x disks
When a filesystem has been converted from 3.5.x to 3.6.x, we need an extra
check during file write to make sure we are not trying to make a 3.5.x file
> 2GB.
Signed-off-by: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Chris Mason [Wed, 1 Feb 2006 11:06:50 +0000 (03:06 -0800)]
[PATCH] reiserfs: reiserfs fix journal accounting in journal_transaction_should_end
reiserfs: journal_transaction_should_end should increase the count of
blocks allocated so the transaction subsystem can keep new writers from
creating a transaction that is too large.
Signed-off-by: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Chris Mason [Wed, 1 Feb 2006 11:06:49 +0000 (03:06 -0800)]
[PATCH] reiserfs: reiserfs hang and performance fix for data=journal mode
In data=journal mode, reiserfs writepage needs to make sure not to trigger
transactions while being run under PF_MEMALLOC. This patch makes sure to
redirty the page instead of forcing a transaction start in this case.
Also, calling filemap_fdata* in order to trigger io on the block device can
cause lock inversions on the page lock. Instead, do simple batching from
flush_commit_list.
Signed-off-by: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Chris Mason [Wed, 1 Feb 2006 11:06:48 +0000 (03:06 -0800)]
[PATCH] reiserfs: zero b_private when allocating buffer heads
The b_private field in buffer heads needs to be zero filled when the
buffers are allocated. Thanks to Nathan Scott for finding this. It was
causing problems on systems with both XFS and reiserfs.
Signed-off-by: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Chris Mason [Wed, 1 Feb 2006 11:06:47 +0000 (03:06 -0800)]
[PATCH] resierfs: fix reiserfs_invalidatepage race against data=ordered
After a transaction has closed but before it has finished commit, there is
a window where data=ordered mode requires invalidatepage to pin pages
instead of freeing them. This patch fixes a race between the
invalidatepage checks and data=ordered writeback, and it also adds a check
to the reiserfs write_ordered_buffers routines to write any anonymous
buffers that were dirtied after its first writeback loop.
That bug works like this:
proc1: transaction closes and a new one starts
proc1: write_ordered_buffers starts processing data=ordered list
proc1: buffer A is cleaned and written
proc2: buffer A is dirtied by another process
proc2: File is truncated to zero, page A goes through invalidatepage
proc2: reiserfs_invalidatepage sees dirty buffer A with reiserfs
journal head, pins it
proc1: write_ordered_buffers frees the journal head on buffer A
At this point, buffer A stays dirty forever
Signed-off-by: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use the generic_permission code with a proper wrapper and callback instead
of having a local copy.
Signed-off-by: Christoph Hellwig <hch@lst.de> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This function is completely unused since the xattr permission checking
changes. Remove it and fold __reiserfs_permission into
reiserfs_permission.
Signed-off-by: Christoph Hellwig <hch@lst.de> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diego Calleja [Wed, 1 Feb 2006 11:06:44 +0000 (03:06 -0800)]
[PATCH] reiserfs: missing kmalloc failure check
According to http://bugzilla.kernel.org/show_bug.cgi?id=5778
fs/reiserfs/file.c is missing this check.
Signed-off-by: Diego Calleja <diegocg@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <mason@suse.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Horst Hummel [Wed, 1 Feb 2006 11:06:37 +0000 (03:06 -0800)]
[PATCH] s390: dasd wait for clear i/o interrupt
The sleep_on function clears a running cqr without waiting for the related
interrupt. This can lead to a panic at the time the interrupt is processed
because the related memory might already be freed. Wait for clear-interrupt
and de-queue cqr prior to return.
Signed-off-by: Horst Hummel <horst.hummel@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Horst Hummel [Wed, 1 Feb 2006 11:06:36 +0000 (03:06 -0800)]
[PATCH] s390: dasd open counter
The open_count is increased for every opener, that includes the blkdev_get in
dasd_scan_partitions. This tampers the open_count in BIODASDINFO. Hide the
internal open from user-space.
Signed-off-by: Horst Hummel <horst.hummel@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] s390: ccw_device_probe_console return value
The return code of ccw_device_probe_console() is not properly handled. It
should only return a valid ccw device pointer or a error value converted by
ERR_PTR. Fix the console driver code to check with IS_ERR instead against
NULL.
Signed-off-by: Peter Oberparleiter <peter.oberparleiter@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jan Glauber [Wed, 1 Feb 2006 11:06:34 +0000 (03:06 -0800)]
[PATCH] s390: hangcheck timer support
Remove useless s390 define from hangcheck-timer, remove wrong definition of a
TOD second and other s390 ifdefs. Use monotonic_clock instead.
Add hangcheck-timer option, copied from drivers/char/Kconfig. This is ugly
but unless we have a big Kconfig cleanup we cannot include
drivers/char/Kconfig...
Signed-off-by: Jan Glauber <jan.glauber@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jan Glauber [Wed, 1 Feb 2006 11:06:32 +0000 (03:06 -0800)]
[PATCH] s390: overflow in sched_clock
The least significant bit of the TOD clock value returned by get_clock
is the 4096th part of a microsecond. To get to nanoseconds the value
needs to be divided by 4096 and multiplied with 1000.
The current method multiplies first and then shifts the value to make the
result as precise as possible. The disadvantage is that the multiplication
with 1000 will overflow shortly after 52 days. sched_clock is used by the
scheduler for time stamp deltas, if an overflow occurs between two time stamps
the scheduler will get confused.
With the patch the problem occurs only after approx. one year, so the chance
to run into this overflow is extremly low.
Signed-off-by: Jan Glauber <jan.glauber@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Olaf reported UML doesn't build for him with a clear analisys of what happened
- we're using NR_CPUS in files linked against glibc headers. Seems like it
defines CONFIG_SMP but not CONFIG_NR_CPUS, so we get CONFIG_NR_CPUS
undeclared.
The fix is to move the declaration away from that header file and move it in
asm-um headers, and to add that header where needed.
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>
Fix some simple sparse warnings - a lot more staticness and a misplaced
__user.
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>
In a recent fixup i386 code was copied raw to x86_64 subarch to make it
compile again.
Here there are some little fixups and resyncs needed for it (mainly for
cleanliness sake) - I did an audit and found the rest of the code to be safe.
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>
I'm bothering with this mainly because it gives a correct warning when the
config option is enabled, even if the code is for a almost unused debugging
option.
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>
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>
The C99 initialization, with GCC's bad handling, for 6K wide structs (which
_aren't_ on the stack), is causing GCC to use 12K for these silly procs with 3
vars. Workaround this.
Note that .name = { '\0' } translates to memset(->name, 0, '->name' size) - I verified
this with GCC's docs and a testprogram.
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>
While fixing myself the mktime conflict (which someone already merged), I also
improved a few comments. Merge them up.
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>
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>
Alexey Dobriyan [Wed, 1 Feb 2006 11:06:14 +0000 (03:06 -0800)]
[PATCH] arm26: fix find_first_zero_bit related warnings
include/linux/nodemask.h: In function `__first_unset_node':
include/linux/nodemask.h:254: warning: passing arg 1 of `_find_first_zero_bit_le' discards qualifiers from pointer target type
fs/minix/bitmap.c: In function `minix_new_block':
fs/minix/bitmap.c:89: warning: passing arg 1 of `_find_first_zero_bit_le' from incompatible pointer type
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Acked-by: Ian Molton <spyro@f2s.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Eric Sesterhenn [Wed, 1 Feb 2006 11:06:13 +0000 (03:06 -0800)]
[PATCH] alpha show_interrups() trashes argument
This is a bug found by cpminer. The show_interrupts function reuses i as a
for loop counter, and therefore trashes its contents, which are needed
later.
(akpm: rename local `i' to `irq', use for_each_inline_cpu())
Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] alpha: Fix getxpid on alpha so it works for threads
While looking in the code I discovered that alpha has fallen behind because
it doesn't use sys_getppid. The problem is that it doesn't follow the task
struct to the task_group_leader.
Acked-by: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mark Lord [Wed, 1 Feb 2006 11:06:11 +0000 (03:06 -0800)]
[PATCH] VMSPLIT config options
Enable selection of different user/kernel VM splits for i386, including an
optimized mode for 1GB physical RAM, which gives the kernel a direct (non
HIGHMEM) mapping to the entire 1GB rather than just the first 896MB.
There is a similarly a similarly optimized mode for machines with exactly 2GB
of physical RAM.
This can speed up the kernel by avoiding having to create/destroy temporary
HIGHMEM mappings, and by not having to include HIGHMEM support at all on such
machines. The flip side is that there's less virtual addressing left for
userspace in these alternatives, and some binary-only kernel modules may
misbehave unless rebuilt with the same VMSPLIT option as the main kernel
image.
Original idea/patch from Jens Axboe, modified based on suggestions from Linus
et al.
Signed-off-by: Mark Lord <mlord@pobox.com> Signed-off-by: Jens Axboe <axboe@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Alexey Dobriyan [Wed, 1 Feb 2006 11:06:09 +0000 (03:06 -0800)]
[PATCH] arch/sh64/kernel/time.c: add module.h
It uses EXPORT_SYMBOL.
arch/sh64/kernel/time.c:254: warning: type defaults to `int' in declaration of `EXPORT_SYMBOL'
arch/sh64/kernel/time.c:254: warning: parameter names (without types) in function declaration
arch/sh64/kernel/time.c:254: warning: data definition has no type or storage class
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Cc: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As reported by Russell King, sh and sh64 currently have bogus definitions for
TIOCGICOUNT, particularly referencing a kernel only structure. Switch to
using a sensible ioctl value.
Signed-off-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Paul Mundt [Wed, 1 Feb 2006 11:06:05 +0000 (03:06 -0800)]
[PATCH] sh: convert voyagergx to platform device, drop sh-bus
Trivial patch updating the voyagergx cchip code to reference a platform device
instead, now that the dma mask is taken care of. Given this, there's no
longer any reason to drag around the SH-bus code, so kill that off entirely.
Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net> Signed-off-by: Paul Mundt <lethal@linux-sh.org> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Paul Mundt [Wed, 1 Feb 2006 11:06:04 +0000 (03:06 -0800)]
[PATCH] sh: drop maskpos from make_ipr_irq(), remove duplicate irq definitions
Clean up some of the subtype IRQ definitions for IPR IRQ, and consolidate the
make_ipr_irq() definitions by dropping maskpos. SH-4A was the only thing
interested in the maskpos, and this should be handled through INTC2 rather
than IPR.
Signed-off-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Paul Mundt [Wed, 1 Feb 2006 11:06:01 +0000 (03:06 -0800)]
[PATCH] sh: Make peripheral clock frequency setting mandatory
Pretty much every subtype does this now anyways, and as we depend on it in a
few places being set to something sensible quite early on, it's better for a
new subtype to simply set a sensible default.
Signed-off-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Paul Mundt [Wed, 1 Feb 2006 11:05:59 +0000 (03:05 -0800)]
[PATCH] sh: SH4-202 microdev updates
A few trivial updates for the microdev board support code:
- Update for __IO_PREFIX changes.
- Consolidate headers into a single microdev.h.
- Update the microdev_defconfig.
- Add init values for the S1D13806 used by s1d13xxxfb.
Signed-off-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Olaf Hering [Wed, 1 Feb 2006 11:05:58 +0000 (03:05 -0800)]
[PATCH] MODALIAS= for macio
Prodive a MODALIAS= enviroment variable for devices on the mac-io bus.
Change the buffer length counter to not waste memory by advancing the
pointer for the next string too far. Tested on an ibook1 with modular
pmac_zilog.
Signed-off-by: Olaf Hering <olh@suse.de> 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@osdl.org>
Stephen Smalley [Wed, 1 Feb 2006 11:05:56 +0000 (03:05 -0800)]
[PATCH] selinux: remove security struct magic number fields and tests
Remove the SELinux security structure magic number fields and tests, along
with some unnecessary tests for NULL security pointers. These fields and
tests are leftovers from the early attempts to support SELinux as a
loadable module during LSM development.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Acked-by: James Morris <jmorris@namei.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Stephen Smalley [Wed, 1 Feb 2006 11:05:55 +0000 (03:05 -0800)]
[PATCH] selinux: change file_alloc_security to use GFP_KERNEL
This patch changes the SELinux file_alloc_security function to use
GFP_KERNEL rather than GFP_ATOMIC; the use of GFP_ATOMIC appears to be a
remnant of when this function was being called with the files_lock spinlock
held, and is no longer necessary. Please apply.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Acked-by: James Morris <jmorris@namei.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Stephen Smalley [Wed, 1 Feb 2006 11:05:54 +0000 (03:05 -0800)]
[PATCH] selinux: fix and cleanup mprotect checks
Fix the SELinux mprotect checks on executable mappings so that they are not
re-applied when the mapping is already executable as well as cleaning up
the code. This avoids a situation where e.g. an application is prevented
from removing PROT_WRITE on an already executable mapping previously
authorized via execmem permission due to an execmod denial.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Acked-by: James Morris <jmorris@namei.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Pekka Enberg [Wed, 1 Feb 2006 11:05:52 +0000 (03:05 -0800)]
[PATCH] slab: fix kzalloc and kstrdup caller report for CONFIG_DEBUG_SLAB
Fix kzalloc() and kstrdup() caller report for CONFIG_DEBUG_SLAB. We must
pass the caller to __cache_alloc() instead of directly doing
__builtin_return_address(0) there; otherwise kzalloc() and kstrdup() are
reported as the allocation site instead of the real one.
Thanks to Valdis Kletnieks for reporting the problem and Steven Rostedt for
the original idea.
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Pekka Enberg [Wed, 1 Feb 2006 11:05:49 +0000 (03:05 -0800)]
[PATCH] slab: extract virt_to_{cache|slab}
Introduce virt_to_cache() and virt_to_slab() functions to reduce duplicate
code and introduce a proper abstraction should we want to support other kind
of mapping for address to slab and cache (eg. for vmalloc() or I/O memory).
Acked-by: Manfred Spraul <manfred@colorfullife.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Matthew Dobson [Wed, 1 Feb 2006 11:05:46 +0000 (03:05 -0800)]
[PATCH] slab: extract slab_destroy_objs()
Create a helper function, slab_destroy_objs() which called from
slab_destroy(). This makes slab_destroy() smaller and more readable, and
moves ifdefs outside the function body.
Signed-off-by: Matthew Dobson <colpatch@us.ibm.com> Acked-by: Manfred Spraul <manfred@colorfullife.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Steven Rostedt [Wed, 1 Feb 2006 11:05:45 +0000 (03:05 -0800)]
[PATCH] slab: cache_estimate cleanup
Clean up cache_estimate() in mm/slab.c and improves the algorithm from O(n) to
O(1). We first calculate the maximum number of objects a slab can hold after
struct slab and kmem_bufctl_t for each object has been given enough space.
After that, to respect alignment rules, we decrease the number of objects if
necessary. As required padding is at most align-1 and memory of obj_size is
at least align, it is always enough to decrease number of objects by one.
The optimization was originally made by Balbir Singh with more improvements
from Steven Rostedt. Manfred Spraul provider further modifications: no loop
at all for the off-slab case and added comments to explain the background.
Acked-by: Balbir Singh <bsingharora@gmail.com> Signed-off-by: Manfred Spraul <manfred@colorfullife.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Steven Rostedt [Wed, 1 Feb 2006 11:05:44 +0000 (03:05 -0800)]
[PATCH] slab: have index_of bug at compile time
I noticed the code for index_of is a creative way of finding the cache
index using the compiler to optimize to a single hard coded number. But
I couldn't help noticing that it uses two methods to let you know that
someone used it wrong. One is at compile time (the correct way), and
the other is at run time (not good).
Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Acked-by: Manfred Spraul <manfred@colorfullife.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] Direct Migration V9: Avoid writeback / page_migrate() method
Migrate a page with buffers without requiring writeback
This introduces a new address space operation migratepage() that may be used
by a filesystem to implement its own version of page migration.
A version is provided that migrates buffers attached to pages. Some
filesystems (ext2, ext3, xfs) are modified to utilize this feature.
The swapper address space operation are modified so that a regular
migrate_page() will occur for anonymous pages without writeback (migrate_pages
forces every anonymous page to have a swap entry).
Signed-off-by: Mike Kravetz <kravetz@us.ibm.com> Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] Direct Migration V9: upgrade MPOL_MF_MOVE and sys_migrate_pages()
Modify policy layer to support direct page migration
- Add migrate_pages_to() allowing the migration of a list of pages to a a
specified node or to vma with a specific allocation policy in sets of
MIGRATE_CHUNK_SIZE pages
- Modify do_migrate_pages() to do a staged move of pages from the source
nodes to the target nodes.
Signed-off-by: Paul Jackson <pj@sgi.com> Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] Direct Migration V9: remove_from_swap() to remove swap ptes
Add remove_from_swap
remove_from_swap() allows the restoration of the pte entries that existed
before page migration occurred for anonymous pages by walking the reverse
maps. This reduces swap use and establishes regular pte's without the need
for page faults.
Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] Direct Migration V9: migrate_pages() extension
Add direct migration support with fall back to swap.
Direct migration support on top of the swap based page migration facility.
This allows the direct migration of anonymous pages and the migration of file
backed pages by dropping the associated buffers (requires writeout).
Fall back to swap out if necessary.
The patch is based on lots of patches from the hotplug project but the code
was restructured, documented and simplified as much as possible.
Note that an additional patch that defines the migrate_page() method for
filesystems is necessary in order to avoid writeback for anonymous and file
backed pages.
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Mike Kravetz <kravetz@us.ibm.com> Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Check for PageSwapCache after looking up and locking a swap page.
The page migration code may change a swap pte to point to a different page
under lock_page().
If that happens then the vm must retry the lookup operation in the swap space
to find the correct page number. There are a couple of locations in the VM
where a lock_page() is done on a swap page. In these locations we need to
check afterwards if the page was migrated. If the page was migrated then the
old page that was looked up before was freed and no longer has the
PageSwapCache bit set.
Signed-off-by: Hirokazu Takahashi <taka@valinux.co.jp> Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Christoph Lameter <clameter@@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If large amounts of zone memory are used by empty slabs then zone_reclaim
becomes uneffective. This patch shakes the slab a bit.
The problem with this patch is that the slab reclaim is not containable to a
zone. Thus slab reclaim may affect the whole system and be extremely slow.
This also means that we cannot determine how many pages were freed in this
zone. Thus we need to go off node for at least one allocation.
The functionality is disabled by default.
We could modify the shrinkers to take a zone parameter but that would be quite
invasive. Better ideas are welcome.
Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] Zone reclaim: Allow modification of zone reclaim behavior
In some situations one may want zone_reclaim to behave differently. For
example a process writing large amounts of memory will spew unto other nodes
to cache the writes if many pages in a zone become dirty. This may impact the
performance of processes running on other nodes.
Allowing writes during reclaim puts a stop to that behavior and throttles the
process by restricting the pages to the local zone.
Similarly one may want to contain processes to local memory by enabling
regular swap behavior during zone_reclaim. Off node memory allocation can
then be controlled through memory policies and cpusets.
Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] zone_reclaim: configurable off node allocation period.
Currently the zone_reclaim code has a fixed window of 30 seconds of off node
allocations should a local zone have no unused pagecache pages left. Reclaim
will be attempted again after this timeout period to avoid repeated useless
scans for memory. This is also useful to established sufficiently large off
node allocation chunks to relieve the local node.
It may be beneficial to adjust that time period for some special situations.
For example if memory use was exceeding node capacity one may want to give up
for longer periods of time. If memory spikes intermittendly then one may want
to shorten the time period to reduce the number of off node allocations.
This patch allows just that....
Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
projects / linux-2.6-omap-h63xx.git / log