Avoid merging a VMA with another VMA which is cloned from the parent process.
The cloned VMA shares the anon_vma lock with the parent process's VMA. If
we do the merge, more vmas (even the new range is only for current
process) use the perent process's anon_vma lock. This introduces
scalability issues. find_mergeable_anon_vma() already considers this
case.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If we only change vma->vm_end, we can avoid taking anon_vma lock even if
'insert' isn't NULL, which is the case of split_vma.
As I understand it, we need the lock before because rmap must get the
'insert' VMA when we adjust old VMA's vm_end (the 'insert' VMA is linked
to anon_vma list in __insert_vm_struct before).
But now this isn't true any more. The 'insert' VMA is already linked to
anon_vma list in __split_vma(with anon_vma_clone()) instead of
__insert_vm_struct. There is no race rmap can't get required VMAs. So
the anon_vma lock is unnecessary, and this can reduce one locking in brk
case and improve scalability.
Signed-off-by: Shaohua Li<shaohua.li@intel.com>
Cc: Rik van Riel <riel@redhat.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make some variables have correct alignment/section to avoid cache issue.
In a workload which heavily does mmap/munmap, the variables will be used
frequently.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Architectures that implement their own show_mem() function did not pass
the filter argument to show_free_areas() to appropriately avoid emitting
the state of nodes that are disallowed in the current context. This patch
now passes the filter argument to show_free_areas() so those nodes are now
avoided.
This patch also removes the show_free_areas() wrapper around
__show_free_areas() and converts existing callers to pass an empty filter.
ia64 emits additional information for each node, so skip_free_areas_zone()
must be made global to filter disallowed nodes and it is converted to use
a nid argument rather than a zone for this use case.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Helge Deller <deller@gmx.de>
Cc: James Bottomley <jejb@parisc-linux.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Guan Xuetao <gxt@mprc.pku.edu.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is not useful: it provides page->virtual and is used with highmem.
xtensa has no support for highmem and those HIGHMEM bits which are found
by grep are partly implemented. The interesting functions like kmap() are
missing. If someone actually implements the complete HIGHMEM support he
could use HASHED_PAGE_VIRTUAL like most others do.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Cc: Chris Zankel <chris@zankel.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It is not referenced by any Makefile. pte_alloc_one_kernel() and
pte_alloc_one() is implemented in arch/xtensa/include/asm/pgalloc.h.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Cc: Chris Zankel <chris@zankel.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It should check if strict_strtoul() succeeds.
[akpm@linux-foundation.org: don't override strict_strtoul() return value]
Signed-off-by: Liu Yuan <tailai.ly@taobao.com>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It has been reported on some laptops that kswapd is consuming large
amounts of CPU and not being scheduled when SLUB is enabled during large
amounts of file copying. It is expected that this is due to kswapd
missing every cond_resched() point because;
shrink_page_list() calls cond_resched() if inactive pages were isolated
which in turn may not happen if all_unreclaimable is set in
shrink_zones(). If for whatver reason, all_unreclaimable is
set on all zones, we can miss calling cond_resched().
balance_pgdat() only calls cond_resched if the zones are not
balanced. For a high-order allocation that is balanced, it
checks order-0 again. During that window, order-0 might have
become unbalanced so it loops again for order-0 and returns
that it was reclaiming for order-0 to kswapd(). It can then
find that a caller has rewoken kswapd for a high-order and
re-enters balance_pgdat() without ever calling cond_resched().
shrink_slab only calls cond_resched() if we are reclaiming slab
pages. If there are a large number of direct reclaimers, the
shrinker_rwsem can be contended and prevent kswapd calling
cond_resched().
This patch modifies the shrink_slab() case. If the semaphore is
contended, the caller will still check cond_resched(). After each
successful call into a shrinker, the check for cond_resched() remains in
case one shrinker is particularly slow.
[mgorman@suse.de: preserve call to cond_resched after each call into shrinker]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Tested-by: Colin King <colin.king@canonical.com>
Cc: Raghavendra D Prabhu <raghu.prabhu13@gmail.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: <stable@kernel.org> [2.6.38+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are a few reports of people experiencing hangs when copying large
amounts of data with kswapd using a large amount of CPU which appear to be
due to recent reclaim changes. SLUB using high orders is the trigger but
not the root cause as SLUB has been using high orders for a while. The
root cause was bugs introduced into reclaim which are addressed by the
following two patches.
Patch 1 corrects logic introduced by commit 1741c877 ("mm: kswapd:
keep kswapd awake for high-order allocations until a percentage of
the node is balanced") to allow kswapd to go to sleep when
balanced for high orders.
Patch 2 notes that it is possible for kswapd to miss every
cond_resched() and updates shrink_slab() so it'll at least reach
that scheduling point.
Chris Wood reports that these two patches in isolation are sufficient to
prevent the system hanging. AFAIK, they should also resolve similar hangs
experienced by James Bottomley.
This patch:
Johannes Weiner poined out that the logic in commit 1741c877 ("mm: kswapd:
keep kswapd awake for high-order allocations until a percentage of the
node is balanced") is backwards. Instead of allowing kswapd to go to
sleep when balancing for high order allocations, it keeps it kswapd
running uselessly.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Wu Fengguang <fengguang.wu@intel.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Tested-by: Colin King <colin.king@canonical.com>
Cc: Raghavendra D Prabhu <raghu.prabhu13@gmail.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Rik van Riel <riel@redhat.com>
Reviewed-by: Minchan Kim <minchan.kim@gmail.com>
Reviewed-by: Wu Fengguang <fengguang.wu@intel.com>
Cc: <stable@kernel.org> [2.6.38+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 442b06bcea ("slub: Remove node check in slab_free") added a
call to deactivate_slab() in the debug case in __slab_alloc(), which
unlocks the current slab used for allocation. Going to the label
'unlock_out' then does it again.
Also, in the debug case we do not need all the other processing that the
'unlock_out' path does. We always fall back to the slow path in the
debug case. So the tid update is useless.
Similarly, ALLOC_SLOWPATH would just be incremented for all allocations.
Also a pretty useless thing.
So simply restore irq flags and return the object.
Signed-off-by: Christoph Lameter <cl@linux.com>
Reported-and-bisected-by: James Morris <jmorris@namei.org>
Reported-by: Ingo Molnar <mingo@elte.hu>
Reported-by: Jens Axboe <jaxboe@fusionio.com>
Cc: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Update documentation pointers to include virtfs publication, 9p RFC
as well as updated list of servers and alternative clients.
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Teach 9p filesystem to work in container with non-default network namespace.
(Note: I also patched the unix domain socket code but don't have a test case
for that. It's the same fix, I just don't have a server for it...)
To test, run diod server (http://code.google.com/p/diod):
diod -n -f -L stderr -l 172.23.255.1:9999 -c /dev/null -e /root
and then mount like so:
mount -t 9p -o port=9999,aname=/root,version=9p2000.L 172.23.255.1 /mnt
A container test environment is described at http://landley.net/lxc
Signed-off-by: Rob Landley <rob@landley.net>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
We need to return -1 on error. Also handle error properly
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
The dentry fid get clunked via the dput.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
There are two small cleanups in this patch:
- p9_errstr2errno was declared twice - remove one declaration.
- A uint8_t type was mixed in, change it to u8 to match
with the rest of the type names and remove dependency.
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Cc: Ron Minnich <rminnich@sandia.gov>
Cc: Latchesar Ionkov <lucho@ionkov.net>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
The 9p client is currently undergoing regular regresssion and
stress testing as a by-product of the virtfs work. I think its
finally time to take off the experimental tags from the well-tested
code paths.
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Typo fixes and minor cleanups for v9fs
Signed-off-by: Rob Landley <rob@landley.net>
Reviewed-by: Venkateswararao Jujjuri (JV) <jvrao@linux.vnet.ibm.com>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
As on Jeopardy, my question is in the form of a patch: Does this have
some special meaning, or is it an accident? (I looked at other
filesystems but they didn't bother having doc entries for their
init/exit function that I could find.)
Signed-off-by: Rob Landley <rob@landley.net>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Avoid to have multiple path saving the config. This fixes an error check
miss when the window is being closed and the user requested the config
to be written.
Reported-by: Hiromu Yakura <hiromu1996@gmail.com>
Pointed-out-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Michal Marek <mmarek@suse.cz>
The Blackfin percpu input sections are outdated, so rather than update
them, drop them completely and use the new common define.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
There is a small typo in the Analog Devices driver mailinglist address.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Now that the Blackfin machine drivers have been updated to the
multicomponent support, update the resources to match. The pin
settings are now a board issue and removed from the driver.
Signed-off-by: Scott Jiang <scott.jiang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Anomaly 05000480 on BF537 rev 0.0, 0.1, 0.2:
Multiple Simultaneous Urgent DMA Requests May Cause DMA System Instability
Suggested Workaround:
Program the DMA Traffic Control Period to a non-zero value. This forces
the DMA block to group accesses together rather than allow arbitration for
each piece of data placed on the internal DMA bus.
Signed-off-by: Steven Miao <realmz6@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Since the bfin_write() func needs proper type information in order to
expand into the right bfin_writeX() variant, preserve the addr's type
when setting up the local __addr. Otherwise the helpers will detect
the variant based upon sizeof(void) which is almost never right.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
old cpu_xxx() APIs is planned to removed later. then, converted.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Every callter (except kthread_bind) should use proper
set_cpus_allowed_ptr() APIs.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
We plan to remove cpu_possible_map and cpu_present_map later and we
have proper init_cpu_possible() and init_cpu_present() APIs.
Therefore this patch rewrites platform_init_cpus and platform_prepare_cpus
by their APIs.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Both the BF548-EZKIT and the BF561-EZKIT use top boot flashes, so now
that Das U-Boot uses the last small sector for its environment, update
their nor layout in the kernel accordingly.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Turns out the documentation is wrong and doing "RAISE 1" does not result
in a software reset, only a core reset. So when the on-chip rom has a
functioning reset helper, use it.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
These boards don't have AD1980 modules, and that device is deprecated,
so don't bother building it up by default anymore.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Since the value of these MMRs aren't changing, store the value in a local
variable and work off of that. This avoids multiple MMR reads which are
implicitly forced by the volatile markings.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The BF537 SIC combines the gpio port H mask A interrupts with the
emac rx interrupt, so we need to demux this in software.
It also combines the gpio port H mask B and the emac tx interrupts,
and the watchdog and port F mask B interrupts, but since we don't
support mask B yet, just add the defines for now.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The search logic in the gpio demux walks all possible gpio blocks starting
at the specified pin. The trouble on bf537 parts when we demux the port
F and port G mask A interrupts is that we also demux port H mask A ints.
Most of the time this isn't an issue as people don't usually use port H,
but might as well avoid it when possible.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
This functionality was merged into the common bfin_pm_standby_ctrl func
some time ago, so punt these now unused funcs and data, and localize the
wake funcs that aren't needed externally anymore.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The SIC interrupt line muxing that the bf537 does is specific to this
CPU (thankfully), so rip it out of the common code and move it to a
bf537-specific file. This tidies up the common code significantly.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
These are only used in a few internal Blackfin places, so move the irq
prototypes out of the global header and into the internal irq one. No
functional changes other than shuffling locales.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Prefer MMR named checks over part-specific lists, condense duplicated
code across different #ifdef branches, simplify CONFIG_PM ifdefs, and
drop unused kgdb header.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Make sure we mark cache flushing as unsafe to kgdb in SMP mode so that
kgdb doesn't flush things incorrectly on us.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
After some cache setup reordering changesets, the blackfin_cpudata init
was left behind. While cpu0's data was correct, cpu1's data was not.
Not that big of a deal as these are only used in the cpuinfo output, but
should still be fixed. So move the setup of these fields to the common
cache setup function to avoid this happening again in the future.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The smp flush lines are too long and have too many newlines, so scale
them back to match the other lines.
The %p modifier shows "(null)" for 0, so use %08x instead.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Build the sound pieces that the board actually has into the kernel, and
punt older devices that we discourage.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Shaohua Li