David Rientjes [Mon, 28 Apr 2008 09:12:34 +0000 (02:12 -0700)]
mempolicy: small header file cleanup
Removes forward definition of vm_area_struct in linux/mempolicy.h. We already
get it from the linux/slab.h -> linux/gfp.h include.
Removes the unused mpol_set_vma_default() macro from linux/mempolicy.h.
Removes the extern definition of default_policy since it is only referenced,
as it should be, in mm/mempolicy.c.
Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This splits the implementation for the various modes out of two large
functions, mpol_new() and mpol_rebind_policy(). Eventually it may be
beneficial to add additional functions to accomodate the existing switch()
statements in mm/mempolicy.c.
[*] The ->create() function for MPOL_DEFAULT is currently NULL since no
struct mempolicy is dynamically allocated.
[Lee.Schermerhorn@hp.com: fix regression in the package mempolicy regression tests] Signed-off-by: David Rientjes <rientjes@google.com> Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Eric Whitney <eric.whitney@hp.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Rientjes [Mon, 28 Apr 2008 09:12:32 +0000 (02:12 -0700)]
mempolicy: move rebind functions
Move the mpol_rebind_{policy,task,mm}() functions after mpol_new() to avoid
having to declare function prototypes.
Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Rientjes [Mon, 28 Apr 2008 09:12:31 +0000 (02:12 -0700)]
mempolicy: update NUMA memory policy documentation
Updates Documentation/vm/numa_memory_policy.txt and
Documentation/filesystems/tmpfs.txt to describe optional mempolicy mode flags.
Cc: Christoph Lameter <clameter@sgi.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <ak@suse.de> Cc: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Paul Jackson <pj@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Rientjes [Mon, 28 Apr 2008 09:12:30 +0000 (02:12 -0700)]
mempolicy: add MPOL_F_RELATIVE_NODES flag
Adds another optional mode flag, MPOL_F_RELATIVE_NODES, that specifies
nodemasks passed via set_mempolicy() or mbind() should be considered relative
to the current task's mems_allowed.
When the mempolicy is created, the passed nodemask is folded and mapped onto
the current task's mems_allowed. For example, consider a task using
set_mempolicy() to pass MPOL_INTERLEAVE | MPOL_F_RELATIVE_NODES with a
nodemask of 1-3. If current's mems_allowed is 4-7, the effected nodemask is
5-7 (the second, third, and fourth node of mems_allowed).
If the same task is attached to a cpuset, the mempolicy nodemask is rebound
each time the mems are changed. Some possible rebinds and results are:
mems result
1-3 1-3
1-7 2-4
1,5-6 1,5-6
1,5-7 5-7
Likewise, the zonelist built for MPOL_BIND acts on the set of zones assigned
to the resultant nodemask from the relative remap.
In the MPOL_PREFERRED case, the preferred node is remapped from the currently
effected nodemask to the relative nodemask.
This mempolicy mode flag was conceived of by Paul Jackson <pj@sgi.com>.
Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Paul Jackson [Mon, 28 Apr 2008 09:12:29 +0000 (02:12 -0700)]
mempolicy: add bitmap_onto() and bitmap_fold() operations
The following adds two more bitmap operators, bitmap_onto() and bitmap_fold(),
with the usual cpumask and nodemask wrappers.
The bitmap_onto() operator computes one bitmap relative to another. If the
n-th bit in the origin mask is set, then the m-th bit of the destination mask
will be set, where m is the position of the n-th set bit in the relative mask.
The bitmap_fold() operator folds a bitmap into a second that has bit m set iff
the input bitmap has some bit n set, where m == n mod sz, for the specified sz
value.
There are two substantive changes between this patch and its
predecessor bitmap_relative:
1) Renamed bitmap_relative() to be bitmap_onto().
2) Added bitmap_fold().
The essential motivation for bitmap_onto() is to provide a mechanism for
converting a cpuset-relative CPU or Node mask to an absolute mask. Cpuset
relative masks are written as if the current task were in a cpuset whose CPUs
or Nodes were just the consecutive ones numbered 0..N-1, for some N. The
bitmap_onto() operator is provided in anticipation of adding support for the
first such cpuset relative mask, by the mbind() and set_mempolicy() system
calls, using a planned flag of MPOL_F_RELATIVE_NODES. These bitmap operators
(and their nodemask wrappers, in particular) will be used in code that
converts the user specified cpuset relative memory policy to a specific system
node numbered policy, given the current mems_allowed of the tasks cpuset.
Such cpuset relative mempolicies will address two deficiencies
of the existing interface between cpusets and mempolicies:
1) A task cannot at present reliably establish a cpuset
relative mempolicy because there is an essential race
condition, in that the tasks cpuset may be changed in
between the time the task can query its cpuset placement,
and the time the task can issue the applicable mbind or
set_memplicy system call.
2) A task cannot at present establish what cpuset relative
mempolicy it would like to have, if it is in a smaller
cpuset than it might have mempolicy preferences for,
because the existing interface only allows specifying
mempolicies for nodes currently allowed by the cpuset.
Cpuset relative mempolicies are useful for tasks that don't distinguish
particularly between one CPU or Node and another, but only between how many of
each are allowed, and the proper placement of threads and memory pages on the
various CPUs and Nodes available.
The motivation for the added bitmap_fold() can be seen in the following
example.
Let's say an application has specified some mempolicies that presume 16 memory
nodes, including say a mempolicy that specified MPOL_F_RELATIVE_NODES (cpuset
relative) nodes 12-15. Then lets say that application is crammed into a
cpuset that only has 8 memory nodes, 0-7. If one just uses bitmap_onto(),
this mempolicy, mapped to that cpuset, would ignore the requested relative
nodes above 7, leaving it empty of nodes. That's not good; better to fold the
higher nodes down, so that some nodes are included in the resulting mapped
mempolicy. In this case, the mempolicy nodes 12-15 are taken modulo 8 (the
weight of the mems_allowed of the confining cpuset), resulting in a mempolicy
specifying nodes 4-7.
Signed-off-by: Paul Jackson <pj@sgi.com> Signed-off-by: David Rientjes <rientjes@google.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Andi Kleen <ak@suse.de> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: <kosaki.motohiro@jp.fujitsu.com> Cc: <ray-lk@madrabbit.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
that stores the the nodemask that the user passed when he or she created the
mempolicy via set_mempolicy() or mbind(). When using MPOL_F_STATIC_NODES,
which is passed with any mempolicy mode, the user's passed nodemask
intersected with the VMA or task's allowed nodes is always used when
determining the preferred node, setting the MPOL_BIND zonelist, or creating
the interleave nodemask. This happens whenever the policy is rebound,
including when a task's cpuset assignment changes or the cpuset's mems are
changed.
This creates an interesting side-effect in that it allows the mempolicy
"intent" to lie dormant and uneffected until it has access to the node(s) that
it desires. For example, if you currently ask for an interleaved policy over
a set of nodes that you do not have access to, the mempolicy is not created
and the task continues to use the previous policy. With this change, however,
it is possible to create the same mempolicy; it is only effected when access
to nodes in the nodemask is acquired.
It is also possible to mount tmpfs with the static nodemask behavior when
specifying a node or nodemask. To do this, simply add "=static" immediately
following the mempolicy mode at mount time:
mount -o remount mpol=interleave=static:1-3
Also removes mpol_check_policy() and folds its logic into mpol_new() since it
is now obsoleted. The unused vma_mpol_equal() is also removed.
Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Rientjes [Mon, 28 Apr 2008 09:12:25 +0000 (02:12 -0700)]
mempolicy: support optional mode flags
With the evolution of mempolicies, it is necessary to support mempolicy mode
flags that specify how the policy shall behave in certain circumstances. The
most immediate need for mode flag support is to suppress remapping the
nodemask of a policy at the time of rebind.
Both the mempolicy mode and flags are passed by the user in the 'int policy'
formal of either the set_mempolicy() or mbind() syscall. A new constant,
MPOL_MODE_FLAGS, represents the union of legal optional flags that may be
passed as part of this int. Mempolicies that include illegal flags as part of
their policy are rejected as invalid.
An additional member to struct mempolicy is added to support the mode flags:
struct mempolicy {
...
unsigned short policy;
unsigned short flags;
}
The splitting of the 'int' actual passed by the user is done in
sys_set_mempolicy() and sys_mbind() for their respective syscalls. This is
done by intersecting the actual with MPOL_MODE_FLAGS, rejecting the syscall of
there are additional flags, and storing it in the new 'flags' member of struct
mempolicy. The intersection of the actual with ~MPOL_MODE_FLAGS is stored in
the 'policy' member of the struct and all current users of pol->policy remain
unchanged.
The union of the policy mode and optional mode flags is passed back to the
user in get_mempolicy().
This combination of mode and flags within the same actual does not break
userspace code that relies on get_mempolicy(&policy, ...) and either
switch (policy) {
case MPOL_BIND:
...
case MPOL_INTERLEAVE:
...
};
statements or
if (policy == MPOL_INTERLEAVE) {
...
}
statements. Such applications would need to use optional mode flags when
calling set_mempolicy() or mbind() for these previously implemented statements
to stop working. If an application does start using optional mode flags, it
will need to mask the optional flags off the policy in switch and conditional
statements that only test mode.
An additional member is also added to struct shmem_sb_info to store the
optional mode flags.
[hugh@veritas.com: shmem mpol: fix build warning] Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: David Rientjes <rientjes@google.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>
David Rientjes [Mon, 28 Apr 2008 09:12:23 +0000 (02:12 -0700)]
mempolicy: convert MPOL constants to enum
The mempolicy mode constants, MPOL_DEFAULT, MPOL_PREFERRED, MPOL_BIND, and
MPOL_INTERLEAVE, are better declared as part of an enum since they are
sequentially numbered and cannot be combined.
The policy member of struct mempolicy is also converted from type short to
type unsigned short. A negative policy does not have any legitimate meaning,
so it is possible to change its type in preparation for adding optional mode
flags later.
The equivalent member of struct shmem_sb_info is also changed from int to
unsigned short.
For compatibility, the policy formal to get_mempolicy() remains as a pointer
to an int:
int get_mempolicy(int *policy, unsigned long *nmask,
unsigned long maxnode, unsigned long addr,
unsigned long flags);
although the only possible values is the range of type unsigned short.
Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@sgi.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pekka Enberg [Mon, 28 Apr 2008 09:12:22 +0000 (02:12 -0700)]
mm: move cache_line_size() to <linux/cache.h>
Not all architectures define cache_line_size() so as suggested by Andrew move
the private implementations in mm/slab.c and mm/slob.c to <linux/cache.h>.
Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Reviewed-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Adam Litke [Mon, 28 Apr 2008 09:12:20 +0000 (02:12 -0700)]
hugetlb: decrease hugetlb_lock cycling in gather_surplus_huge_pages
To reduce hugetlb_lock acquisitions and releases when freeing excess surplus
pages, scan the page list in two parts. First, transfer the needed pages to
the hugetlb pool. Then drop the lock and free the remaining pages back to the
buddy allocator.
In the common case there are zero excess pages and no lock operations are
required.
Thanks Mel Gorman for this improvement.
Signed-off-by: Adam Litke <agl@us.ibm.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Dave Hansen <haveblue@us.ibm.com> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: Andy Whitcroft <apw@shadowen.org> Cc: Mel Gorman <mel@csn.ul.ie> Cc: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Chris Dearman [Mon, 28 Apr 2008 09:12:19 +0000 (02:12 -0700)]
mm: try both endianess when checking for endianess
When checking for the swap header try byteswapping the endianess dependent
fields to allow the swap partition to be shared between big & little endian
systems.
Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Acked-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: filter based on a nodemask as well as a gfp_mask
The MPOL_BIND policy creates a zonelist that is used for allocations
controlled by that mempolicy. As the per-node zonelist is already being
filtered based on a zone id, this patch adds a version of __alloc_pages() that
takes a nodemask for further filtering. This eliminates the need for
MPOL_BIND to create a custom zonelist.
A positive benefit of this is that allocations using MPOL_BIND now use the
local node's distance-ordered zonelist instead of a custom node-id-ordered
zonelist. I.e., pages will be allocated from the closest allowed node with
available memory.
[Lee.Schermerhorn@hp.com: Mempolicy: update stale documentation and comments]
[Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask]
[Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask rework] Signed-off-by: Mel Gorman <mel@csn.ul.ie> Acked-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Hugh Dickins <hugh@veritas.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: have zonelist contains structs with both a zone pointer and zone_idx
Filtering zonelists requires very frequent use of zone_idx(). This is costly
as it involves a lookup of another structure and a substraction operation. As
the zone_idx is often required, it should be quickly accessible. The node idx
could also be stored here if it was found that accessing zone->node is
significant which may be the case on workloads where nodemasks are heavily
used.
This patch introduces a struct zoneref to store a zone pointer and a zone
index. The zonelist then consists of an array of these struct zonerefs which
are looked up as necessary. Helpers are given for accessing the zone index as
well as the node index.
[kamezawa.hiroyu@jp.fujitsu.com: Suggested struct zoneref instead of embedding information in pointers]
[hugh@veritas.com: mm-have-zonelist: fix memcg ooms]
[hugh@veritas.com: just return do_try_to_free_pages]
[hugh@veritas.com: do_try_to_free_pages gfp_mask redundant] Signed-off-by: Mel Gorman <mel@csn.ul.ie> Acked-by: Christoph Lameter <clameter@sgi.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Christoph Lameter <clameter@sgi.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> 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>
mm: use two zonelist that are filtered by GFP mask
Currently a node has two sets of zonelists, one for each zone type in the
system and a second set for GFP_THISNODE allocations. Based on the zones
allowed by a gfp mask, one of these zonelists is selected. All of these
zonelists consume memory and occupy cache lines.
This patch replaces the multiple zonelists per-node with two zonelists. The
first contains all populated zones in the system, ordered by distance, for
fallback allocations when the target/preferred node has no free pages. The
second contains all populated zones in the node suitable for GFP_THISNODE
allocations.
An iterator macro is introduced called for_each_zone_zonelist() that interates
through each zone allowed by the GFP flags in the selected zonelist.
Signed-off-by: Mel Gorman <mel@csn.ul.ie> Acked-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Christoph Lameter <clameter@sgi.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: remember what the preferred zone is for zone_statistics
On NUMA, zone_statistics() is used to record events like numa hit, miss and
foreign. It assumes that the first zone in a zonelist is the preferred zone.
When multiple zonelists are replaced by one that is filtered, this is no
longer the case.
This patch records what the preferred zone is rather than assuming the first
zone in the zonelist is it. This simplifies the reading of later patches in
this set.
Signed-off-by: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Christoph Lameter <clameter@sgi.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: introduce node_zonelist() for accessing the zonelist for a GFP mask
Introduce a node_zonelist() helper function. It is used to lookup the
appropriate zonelist given a node and a GFP mask. The patch on its own is a
cleanup but it helps clarify parts of the two-zonelist-per-node patchset. If
necessary, it can be merged with the next patch in this set without problems.
Reviewed-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Christoph Lameter <clameter@sgi.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: use zonelists instead of zones when direct reclaiming pages
The following patches replace multiple zonelists per node with two zonelists
that are filtered based on the GFP flags. The patches as a set fix a bug with
regard to the use of MPOL_BIND and ZONE_MOVABLE. With this patchset, the
MPOL_BIND will apply to the two highest zones when the highest zone is
ZONE_MOVABLE. This should be considered as an alternative fix for the
MPOL_BIND+ZONE_MOVABLE in 2.6.23 to the previously discussed hack that filters
only custom zonelists.
The first patch cleans up an inconsistency where direct reclaim uses
zonelist->zones where other places use zonelist.
The second patch introduces a helper function node_zonelist() for looking up
the appropriate zonelist for a GFP mask which simplifies patches later in the
set.
The third patch defines/remembers the "preferred zone" for numa statistics, as
it is no longer always the first zone in a zonelist.
The forth patch replaces multiple zonelists with two zonelists that are
filtered. The two zonelists are due to the fact that the memoryless patchset
introduces a second set of zonelists for __GFP_THISNODE.
The fifth patch introduces helper macros for retrieving the zone and node
indices of entries in a zonelist.
The final patch introduces filtering of the zonelists based on a nodemask.
Two zonelists exist per node, one for normal allocations and one for
__GFP_THISNODE.
Performance results varied depending on the machine configuration. In real
workloads the gain/loss will depend on how much the userspace portion of the
benchmark benefits from having more cache available due to reduced referencing
of zonelists.
These are the range of performance losses/gains when running against
2.6.24-rc4-mm1. The set and these machines are a mix of i386, x86_64 and
ppc64 both NUMA and non-NUMA.
loss to gain
Total CPU time on Kernbench: -0.86% to 1.13%
Elapsed time on Kernbench: -0.79% to 0.76%
page_test from aim9: -4.37% to 0.79%
brk_test from aim9: -0.71% to 4.07%
fork_test from aim9: -1.84% to 4.60%
exec_test from aim9: -0.71% to 1.08%
This patch:
The allocator deals with zonelists which indicate the order in which zones
should be targeted for an allocation. Similarly, direct reclaim of pages
iterates over an array of zones. For consistency, this patch converts direct
reclaim to use a zonelist. No functionality is changed by this patch. This
simplifies zonelist iterators in the next patch.
Signed-off-by: Mel Gorman <mel@csn.ul.ie> Acked-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Christoph Lameter <clameter@sgi.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mmap_region: cleanup the final vma_merge() related code
It is not easy to actually understand the "if (!file || !vma_merge())"
code, turn it into "if (file && vma_merge())". This makes immediately
obvious that the subsequent "if (file)" is superfluous.
As Hugh Dickins pointed out, we can also factor out the ->i_writecount
corrections, and add a small comment about that.
fix invalidate_inode_pages2_range() to not clear ret
DIO invalidates page cache through invalidate_inode_pages2_range().
invalidate_inode_pages2_range() sets ret=-EIO when
invalidate_complete_page2() fails, but this ret is cleared if
do_launder_page() succeed on a page of next index.
In this case, dio is carried out even if invalidate_complete_page2() fails
on some pages.
This can cause inconsistency between memory and blocks on HDD because the
page cache still exists.
[akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Badari Pulavarty <pbadari@us.ibm.com> Cc: Ken Chen <kenchen@google.com> Cc: Zach Brown <zach.brown@oracle.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Chuck Lever <cel@citi.umich.edu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Migrate flags must be set on slab creation as agreed upon when the antifrag
logic was reviewed. Otherwise some slabs of a slabcache will end up in the
unmovable and others in the reclaimable section depending on which flag was
active when a new slab page was allocated.
This likely slid in somehow when antifrag was merged. Remove it.
The buffer_heads are always allocated with __GFP_RECLAIMABLE because the
SLAB_RECLAIM_ACCOUNT option is set. The set_migrateflags() never had any
effect there.
Radix tree allocations are not directly reclaimable but they are allocated
with __GFP_RECLAIMABLE set on each allocation. We now set
SLAB_RECLAIM_ACCOUNT on radix tree slab creation making sure that radix
tree slabs are consistently placed in the reclaimable section. Radix tree
slabs will also be accounted as such.
There is then no user left of set_migratepages. So remove it.
Signed-off-by: Christoph Lameter <clameter@sgi.com> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jeff Moyer [Mon, 28 Apr 2008 09:12:04 +0000 (02:12 -0700)]
aio: io_getevents() should return if io_destroy() is invoked
This patch wakes up a thread waiting in io_getevents if another thread
destroys the context. This was tested using a small program that spawns a
thread to wait in io_getevents while the parent thread destroys the io context
and then waits for the getevents thread to exit. Without this patch, the
program hangs indefinitely. With the patch, the program exits as expected.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com> Cc: Zach Brown <zach.brown@oracle.com> Cc: Christopher Smith <x@xman.org> Cc: Benjamin LaHaise <bcrl@kvack.org> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
All architectures use an effectively identical definition of online_page(), so
just make it common code. x86-64, ia64, powerpc and sh are actually
identical; x86-32 is slightly different.
x86-32's differences arise because it puts its hotplug pages in the highmem
zone. We can handle this in the generic code by inspecting the page to see if
its in highmem, and update the totalhigh_pages count appropriately. This
leaves init_32.c:free_new_highpage with a single caller, so I folded it into
add_one_highpage_init.
I also removed an incorrect comment referring to the NUMA case; any NUMA
details have already been dealt with by the time online_page() is called.
hotplug memory remove: generic __remove_pages() support
Generic helper function to remove section mappings and sysfs entries for the
section of the memory we are removing. offline_pages() correctly adjusted
zone and marked the pages reserved.
TODO: Yasunori Goto is working on patches to free up allocations from bootmem.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
rtc: add the support for alarm time relative to current time in sysfs
In current kernel if we want to set the alarm time, the absolute time the
seconds relative to 1970-01-01 00:00:00) should be written into
/sys/class/rtc/rtc0/wakealarm. It is not convenient.
It is more reasonable to add the support for the alarm time relative to
current RTC time.(the unit is second)
For example:
If the RTC is required to generate alarm after 2 minutes, the following
will be OK.
echo +120 > /sys/class/rtc/rtc0/wakealarm
or echo +0x78 > /sys/class/rtc/rtc0/wakealarm
Paul Mundt [Mon, 28 Apr 2008 09:11:57 +0000 (02:11 -0700)]
rtc: rtc-rs5c372: fix up NULL name in transfer error path
rs5c_get_regs() currently uses rs5c->rtc->name for its debug printk when
i2c_transfer() fails, though it is used several times before the rtc dev
has been registered. The earliest we can get at the symbolic name is via
the i2c client's struct device, which can be handled by moving the first
rs5c_get_regs() until after the client pointer is assigned.
Signed-off-by: Paul Mundt <lethal@linux-sh.org> Cc: David Brownell <david-b@pacbell.net> Cc: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sam Ravnborg [Mon, 28 Apr 2008 09:11:55 +0000 (02:11 -0700)]
rtc: silence section mismatch warning in rtc-test
Fix following warning:
WARNING: vmlinux.o(.data+0x253e28): Section mismatch in reference from the variable test_drv to the function .devexit.text:test_remove()
Fix by renaming the platfrom_driver variable from *_drv to *_driver
so modpost ignore the reference to an __devexit section.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org> Cc: Alessandro Zummo <a.zummo@towertech.it> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Mon, 28 Apr 2008 09:11:52 +0000 (02:11 -0700)]
rtc: avoid legacy drivers with generic framework
Kconfig tweaks to help reduce RTC configuration bugs, by avoiding
legacy RTC drivers when the generic RTC framework is enabled:
- If rtc-cmos is selected, disable the legacy rtc driver;
- When using generic RTC on x86, enable rtc-cmos by default;
- In the old "chardev RTC" section of Kconfig, add a comment
warning people off these (seven) legacy RTC drivers when
the generic framework is in use.
People can still use the legacy drivers if they want (or need) to.
This doesn't fix the broken dependencies for the legacy "CMOS" RTC driver.
Ideally it would be a full list of platforms where it works, not a partial
list of ones where it won't. Or better yet, it would depend on a
"HAVE_CMOS_RTC" flag defined by various platforms ... surely there's a
Kconfig style guideline lurking there.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Acked-by: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ingo already fixed one of these at my request (in "x86 PAT: tone down
debugging messages", commit 1ebcc654f010d4a63f3ebf8ddd2cab5a709b1824),
but there was another one he missed.
the platform modalias is prefixed with "platform:". Add MODULE_ALIAS()
to the hotpluggable SCSI platform drivers, to re-enable auto loading.
[dbrownell@users.sourceforge.net: more drivers, registration fixes]
[akpm@linux-foundation.org: fix sgiwd93.c] Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Hannes Reinecke [Fri, 25 Apr 2008 13:01:41 +0000 (15:01 +0200)]
[SCSI] aic7xxx: teach aicasm to not emit unused debug code/data
Add a 'count' variable to each symbol which gets increased every time
the symbol is referenced. And then modify the register definition to
include counts for symbols which are referenced from the source code
only and not from the sequencer code.
This will give us an automatic usage count for the symbols with only
minimal hand-crafting.
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Andrew Vasquez [Thu, 24 Apr 2008 22:21:30 +0000 (15:21 -0700)]
[SCSI] qla2xxx: Correct regression in relogin code.
Commit 63a8651f2548c6bb5132c0b4e7dad4f57a9274db ([SCSI] qla2xxx:
Correct infinite-login-retry issue.) introduced a small
regression where a successful relogin would result in an fcport's
loop_id to be incorrectly reset to FC_NO_LOOP_ID. Only clear-out
loopid, if retries have been 'truly' exhausted.
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Cc: Stable Tree <stable@kernel.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Adrian Bunk [Thu, 24 Apr 2008 22:21:28 +0000 (15:21 -0700)]
[SCSI] qla2xxx: make qla2x00_issue_iocb_timeout() static
This patch makes the needlessly global qla2x00_issue_iocb_timeout()
static.
Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Adrian Bunk [Thu, 24 Apr 2008 22:21:27 +0000 (15:21 -0700)]
[SCSI] qla2xxx: qla_os.c, make 2 functions static
This patch makes the following needlessly global functions static:
- qla2x00_alloc_work()
- qla2x00_post_work()
Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
The driver is incorrectly assuming that the 'sp' reference held
in qla2[x00|4xx]_abort_command() is valid after the mailbox
command is issued to abort the exchange. It is *not*, as the
command may be completed during interrupt context before control
is returned to the mailbox caller.
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Andrew Vasquez [Thu, 24 Apr 2008 22:21:22 +0000 (15:21 -0700)]
[SCSI] qla2xxx: Collapse RISC-RAM retrieval code during a firmware-dump.
Use the more efficient read-DMA'ble-buffer mailbox commands
rather than reading a single word/dword at a time. We also
remove a bulk of the duplicate mailbox command-handling codes in
favor of more generic read-memory() routines (qla2xxx_dump_ram()
and qla24xx_dump_ram()).
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Finn Thain [Fri, 25 Apr 2008 15:06:05 +0000 (10:06 -0500)]
[SCSI] m68k: new mac_esp scsi driver
Replace the mac_esp driver with a new one based on the esp_scsi core.
For esp_scsi: add support for sync transfers for the PIO mode, add a new
esp_driver_ops method to get the maximum dma transfer size (like the old
NCR53C9x driver), and some cleanups.
Signed-off-by: Finn Thain <fthain@telegraphics.com.au> Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
[SCSI] zfcp: Add some statistics provided by the FCP adapter to the sysfs
The new FCP adapter statistics provide a variety of information about
the virtual adapter (subchannel). In order to collect this information
the zfcp driver is extended to query this information.
The information provided by the new FCP adapter statistics can be
fetched by reading from the following files in the sysfs filesystem
These are the statistics on a virtual adapter (subchannel) level.
The information provided is raw and not modified or interpreted by any
means. No interpretation or modification of the values is done by the
zfcp driver.
When statistics are polled from sysfs, the statistics use the same
commands as the adapter initialization. Change the messages printed
here, so they are only printed during initialization and not for each
poll of adapter data.
[SCSI] zfcp: Wait for free SBAL during exchange config
When sending a exchange config data command, wait for a free SBAL.
This does not matter during adapter initialization, but this is
required for pulling adapter statistics during high I/O load.
James Smart [Thu, 24 Apr 2008 16:12:46 +0000 (12:12 -0400)]
[SCSI] scsi_transport_fc: fc_user_scan correction
Way back when, when the fc_user_scan routine was created, it kept some
of its original logic that walked the rport list and kicked off a scan.
Unfortunately, it didn't keep any of the locking around the rport list,
nor did it consider the synchronous nature of the scan invoked. The result,
there are some scan requests where the rport list changes, thus a subsequent
scan is called on a bogus rport structure and the system NMI's.
Signed-off-by: James Smart <james.smart@emulex.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Jeff Garzik [Thu, 24 Apr 2008 23:45:32 +0000 (19:45 -0400)]
[SCSI] aha1542: minor irq handler cleanups
- where the 'irq' function argument is known never to be used, rename
it to 'dummy' to make this more obvious
- replace per-irq lookup functions and tables with a direct reference
to data object obtained via 'dev_id' function argument, passed from
request_irq()
Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Adrian Bunk [Wed, 23 Apr 2008 09:51:10 +0000 (12:51 +0300)]
[SCSI] FlashPoint: fix off-by-one errors
This patch fixes off-by-one errors in error checks (the variables are
used as array indexes for arrays with MAX_SCSI_TAR resp. MAX_LUN
elements) spotted by the Coverity checker.
Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Hannes Reinecke [Wed, 23 Apr 2008 09:39:49 +0000 (11:39 +0200)]
[SCSI] aic7xxx: Update type check in aicasm grammar
The function type_check() in aicasm grammar code was
never used properly due to a bug.
This patch fixes it up and ensures it's only called if appropriate.
In addition the unused 16bit instruction are disabled, but left in
the code for reference.
Signed-off-by: Hannes Reinecke <hare@suse.de> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
James Bottomley [Mon, 21 Apr 2008 15:57:20 +0000 (10:57 -0500)]
[SCSI] fix SLUB WARN_ON
We're getting a WARN_ON from SLUB indicating that we're trying to free
caches with in-use objects. The root cause is a new dependency in the
command/sense free on unchecked_isa_dma. The WARN_ON is caused by
drivers which change this in their setup after the command/sense cache
is allocated.
The fix is to move the allocation of this cache into scsi_add_host()
so things like gdth have an opportunity to modify it between alloc and
add (but *not* after).
The true fix would be to move unchecked_isa_dma into the template and
out of the host, so it because a truly read only variable.
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Merge branch 'kvm-updates-2.6.26' of git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm
* 'kvm-updates-2.6.26' of git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm: (147 commits)
KVM: kill file->f_count abuse in kvm
KVM: MMU: kvm_pv_mmu_op should not take mmap_sem
KVM: SVM: remove selective CR0 comment
KVM: SVM: remove now obsolete FIXME comment
KVM: SVM: disable CR8 intercept when tpr is not masking interrupts
KVM: SVM: sync V_TPR with LAPIC.TPR if CR8 write intercept is disabled
KVM: export kvm_lapic_set_tpr() to modules
KVM: SVM: sync TPR value to V_TPR field in the VMCB
KVM: ppc: PowerPC 440 KVM implementation
KVM: Add MAINTAINERS entry for PowerPC KVM
KVM: ppc: Add DCR access information to struct kvm_run
ppc: Export tlb_44x_hwater for KVM
KVM: Rename debugfs_dir to kvm_debugfs_dir
KVM: x86 emulator: fix lea to really get the effective address
KVM: x86 emulator: fix smsw and lmsw with a memory operand
KVM: x86 emulator: initialize src.val and dst.val for register operands
KVM: SVM: force a new asid when initializing the vmcb
KVM: fix kvm_vcpu_kick vs __vcpu_run race
KVM: add ioctls to save/store mpstate
KVM: Rename VCPU_MP_STATE_* to KVM_MP_STATE_*
...
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
mlx4_core: Add helper to move QP to ready-to-send
mlx4_core: Add HW queues allocation helpers
RDMA/nes: Remove volatile qualifier from struct nes_hw_cq.cq_vbase
mlx4_core: CQ resizing should pass a 0 opcode modifier to MODIFY_CQ
mlx4_core: Move kernel doorbell management into core
IB/ehca: Bump version number to 0026
IB/ehca: Make some module parameters bool, update descriptions
IB/ehca: Remove mr_largepage parameter
IB/ehca: Move high-volume debug output to higher debug levels
IB/ehca: Prevent posting of SQ WQEs if QP not in RTS
IPoIB: Handle 4K IB MTU for UD (datagram) mode
RDMA/nes: Fix adapter reset after PXE boot
RDMA/nes: Print IPv4 addresses in a readable format
RDMA/nes: Use print_mac() to format ethernet addresses for printing
Al Viro [Sun, 27 Apr 2008 05:15:42 +0000 (06:15 +0100)]
e1000e triggers sparc32 gcc bug
... and isn't possible on sparc32 boxen anyway, unless somebody
had done JavaStation with PCIE lately.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Al Viro [Sat, 19 Apr 2008 19:33:56 +0000 (20:33 +0100)]
KVM: kill file->f_count abuse in kvm
Use kvm own refcounting instead of playing with ->filp->f_count.
That will allow to get rid of a lot of crap in anon_inode_getfd() and
kill a race in kvm_dev_ioctl_create_vm() (file might have been closed
immediately by another thread, so ->filp might point to already freed
struct file when we get around to setting it).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Avi Kivity <avi@qumranet.com>
There is not selective cr0 intercept bug. The code in the comment sets the
CR0.PG bit. But KVM sets the CR4.PG bit for SVM always to implement the paged
real mode. So the 'mov %eax,%cr0' instruction does not change the CR0.PG bit.
Selective CR0 intercepts only occur when a bit is actually changed. So its the
right behavior that there is no intercept on this instruction.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
KVM: SVM: disable CR8 intercept when tpr is not masking interrupts
This patch disables the intercept of CR8 writes if the TPR is not masking
interrupts. This reduces the total number CR8 intercepts to below 1 percent of
what we have without this patch using Windows 64 bit guests.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
This functionality is definitely experimental, but is capable of running
unmodified PowerPC 440 Linux kernels as guests on a PowerPC 440 host. (Only
tested with 440EP "Bamboo" guests so far, but with appropriate userspace
support other SoC/board combinations should work.)
See Documentation/powerpc/kvm_440.txt for technical details.
[stephen: build fix]
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Acked-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Avi Kivity <avi@qumranet.com>
KVM: ppc: Add DCR access information to struct kvm_run
Device Control Registers are essentially another address space found on PowerPC
4xx processors, analogous to PIO on x86. DCRs are always 32 bits, and can be
identified by a 32-bit number. We forward most DCR accesses to userspace for
emulation (with the exception of CPR0 registers, which can be read directly
for simplicity in timebase frequency determination).
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
PowerPC 440 KVM needs to know how many TLB entries are used for the host kernel
linear mapping (it does not modify these mappings when switching between guest
and host execution).
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Acked-by: Josh Boyer <jwboyer@linux.vnet.ibm.com> Acked-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Avi Kivity <avi@qumranet.com>
Avi Kivity [Mon, 14 Apr 2008 11:40:50 +0000 (14:40 +0300)]
KVM: x86 emulator: fix smsw and lmsw with a memory operand
lmsw and smsw were implemented only with a register operand. Extend them
to support a memory operand as well. Fixes Windows running some display
compatibility test on AMD hosts.
Get rid of the pipeline merge stage but retain the chrdev req caching
functionality by using a merge buffer tape->merge_bh which is flushed in chunks
of several blocks at a time. Also, remove last references to pipelining, e.g.
typedef idetape_stage_s.
projects / linux-2.6 / log