kernel/kernel-generic.git
18 years ago[PATCH] mm: make hugepages obey cpusets.
Christoph Lameter [Sun, 8 Jan 2006 09:00:57 +0000 (01:00 -0800)]
[PATCH] mm: make hugepages obey cpusets.

See http://marc.theaimsgroup.com/?l=linux-kernel&m=113167000201265&w=2
http://marc.theaimsgroup.com/?l=linux-mm&m=113167267527312&w=2

Make hugepages obey cpusets.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: William Irwin <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] cpusets: swap migration interface
Paul Jackson [Sun, 8 Jan 2006 09:00:56 +0000 (01:00 -0800)]
[PATCH] cpusets: swap migration interface

Add a boolean "memory_migrate" to each cpuset, represented by a file
containing "0" or "1" in each directory below /dev/cpuset.

It defaults to false (file contains "0").  It can be set true by writing
"1" to the file.

If true, then anytime that a task is attached to the cpuset so marked, the
pages of that task will be moved to that cpuset, preserving, to the extent
practical, the cpuset-relative placement of the pages.

Also anytime that a cpuset so marked has its memory placement changed (by
writing to its "mems" file), the tasks in that cpuset will have their pages
moved to the cpusets new nodes, preserving, to the extent practical, the
cpuset-relative placement of the moved pages.

Signed-off-by: Paul Jackson <pj@sgi.com>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] SwapMig: Switch error handling in migrate_pages to use -Exx
Christoph Lameter [Sun, 8 Jan 2006 09:00:55 +0000 (01:00 -0800)]
[PATCH] SwapMig: Switch error handling in migrate_pages to use -Exx

Use -Exxx instead of numeric return codes and cleanup the code in
migrate_pages() using -Exx error codes.

Consolidate successful migration handling

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] SwapMig: Extend parameters for migrate_pages()
Christoph Lameter [Sun, 8 Jan 2006 09:00:55 +0000 (01:00 -0800)]
[PATCH] SwapMig: Extend parameters for migrate_pages()

Extend the parameters of migrate_pages() to allow the caller control over the
fate of successfully migrated or impossible to migrate pages.

Swap migration and direct migration will have the same interface after this
patch so that patches can be independently applied to the policy layer and the
core migration code.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] SwapMig: Drop unused pages immediately
Christoph Lameter [Sun, 8 Jan 2006 09:00:54 +0000 (01:00 -0800)]
[PATCH] SwapMig: Drop unused pages immediately

Drop unused pages immediately

If a page is encountered that is only referenced by the migration code then
there is no reason to swap or migrate the page.  Release the page by calling
move_to_lru().

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] SwapMig: add_to_swap() avoid atomic allocations
Christoph Lameter [Sun, 8 Jan 2006 09:00:53 +0000 (01:00 -0800)]
[PATCH] SwapMig: add_to_swap() avoid atomic allocations

Add gfp_mask to add_to_swap

add_to_swap does allocations with GFP_ATOMIC in order not to interfere with
swapping.  During migration we may have use add_to_swap extensively which may
lead to out of memory errors.

This patch makes add_to_swap take a parameter that specifies the gfp mask.
The page migration code can then make add_to_swap use GFP_KERNEL.

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>
18 years ago[PATCH] SwapMig: CONFIG_MIGRATION fixes
Christoph Lameter [Sun, 8 Jan 2006 09:00:52 +0000 (01:00 -0800)]
[PATCH] SwapMig: CONFIG_MIGRATION fixes

Move move_to_lru, putback_lru_pages and isolate_lru in section surrounded by
CONFIG_MIGRATION saving some codesize for single processor kernels.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Swap Migration V5: sys_migrate_pages interface
Christoph Lameter [Sun, 8 Jan 2006 09:00:51 +0000 (01:00 -0800)]
[PATCH] Swap Migration V5: sys_migrate_pages interface

sys_migrate_pages implementation using swap based page migration

This is the original API proposed by Ray Bryant in his posts during the first
half of 2005 on linux-mm@kvack.org and linux-kernel@vger.kernel.org.

The intent of sys_migrate is to migrate memory of a process.  A process may
have migrated to another node.  Memory was allocated optimally for the prior
context.  sys_migrate_pages allows to shift the memory to the new node.

sys_migrate_pages is also useful if the processes available memory nodes have
changed through cpuset operations to manually move the processes memory.  Paul
Jackson is working on an automated mechanism that will allow an automatic
migration if the cpuset of a process is changed.  However, a user may decide
to manually control the migration.

This implementation is put into the policy layer since it uses concepts and
functions that are also needed for mbind and friends.  The patch also provides
a do_migrate_pages function that may be useful for cpusets to automatically
move memory.  sys_migrate_pages does not modify policies in contrast to Ray's
implementation.

The current code here is based on the swap based page migration capability and
thus is not able to preserve the physical layout relative to it containing
nodeset (which may be a cpuset).  When direct page migration becomes available
then the implementation needs to be changed to do a isomorphic move of pages
between different nodesets.  The current implementation simply evicts all
pages in source nodeset that are not in the target nodeset.

Patch supports ia64, i386 and x86_64.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Swap Migration V5: MPOL_MF_MOVE interface
Christoph Lameter [Sun, 8 Jan 2006 09:00:50 +0000 (01:00 -0800)]
[PATCH] Swap Migration V5: MPOL_MF_MOVE interface

Add page migration support via swap to the NUMA policy layer

This patch adds page migration support to the NUMA policy layer.  An
additional flag MPOL_MF_MOVE is introduced for mbind.  If MPOL_MF_MOVE is
specified then pages that do not conform to the memory policy will be evicted
from memory.  When they get pages back in new pages will be allocated
following the numa policy.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Swap Migration V5: Add CONFIG_MIGRATION for page migration support
Christoph Lameter [Sun, 8 Jan 2006 09:00:49 +0000 (01:00 -0800)]
[PATCH] Swap Migration V5: Add CONFIG_MIGRATION for page migration support

Include page migration if the system is NUMA or having a memory model that
allows distinct areas of memory (SPARSEMEM, DISCONTIGMEM).

And:
- Only include lru_add_drain_per_cpu if building for an SMP system.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Swap Migration V5: migrate_pages() function
Christoph Lameter [Sun, 8 Jan 2006 09:00:48 +0000 (01:00 -0800)]
[PATCH] Swap Migration V5: migrate_pages() function

This adds the basic page migration function with a minimal implementation that
only allows the eviction of pages to swap space.

Page eviction and migration may be useful to migrate pages, to suspend
programs or for remapping single pages (useful for faulty pages or pages with
soft ECC failures)

The process is as follows:

The function wanting to migrate pages must first build a list of pages to be
migrated or evicted and take them off the lru lists via isolate_lru_page().
isolate_lru_page determines that a page is freeable based on the LRU bit set.

Then the actual migration or swapout can happen by calling migrate_pages().

migrate_pages does its best to migrate or swapout the pages and does multiple
passes over the list.  Some pages may only be swappable if they are not dirty.
 migrate_pages may start writing out dirty pages in the initial passes over
the pages.  However, migrate_pages may not be able to migrate or evict all
pages for a variety of reasons.

The remaining pages may be returned to the LRU lists using putback_lru_pages().

Changelog V4->V5:
- Use the lru caches to return pages to the LRU

Changelog V3->V4:
- Restructure code so that applying patches to support full migration does
  require minimal changes. Rename swapout_pages() to migrate_pages().

Changelog V2->V3:
- Extract common code from shrink_list() and swapout_pages()

Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: "Michael Kerrisk" <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Swap Migration V5: PF_SWAPWRITE to allow writing to swap
Christoph Lameter [Sun, 8 Jan 2006 09:00:47 +0000 (01:00 -0800)]
[PATCH] Swap Migration V5: PF_SWAPWRITE to allow writing to swap

Add PF_SWAPWRITE to control a processes permission to write to swap.

- Use PF_SWAPWRITE in may_write_to_queue() instead of checking for kswapd
  and pdflush

- Set PF_SWAPWRITE flag for kswapd and pdflush

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Swap Migration V5: LRU operations
Christoph Lameter [Sun, 8 Jan 2006 09:00:45 +0000 (01:00 -0800)]
[PATCH] Swap Migration V5: LRU operations

This is the start of the `swap migration' patch series.

Swap migration allows the moving of the physical location of pages between
nodes in a numa system while the process is running.  This means that the
virtual addresses that the process sees do not change.  However, the system
rearranges the physical location of those pages.

The main intent of page migration patches here is to reduce the latency of
memory access by moving pages near to the processor where the process
accessing that memory is running.

The patchset allows a process to manually relocate the node on which its
pages are located through the MF_MOVE and MF_MOVE_ALL options while
setting a new memory policy.

The pages of process can also be relocated from another process using the
sys_migrate_pages() function call.  Requires CAP_SYS_ADMIN.  The migrate_pages
function call takes two sets of nodes and moves pages of a process that are
located on the from nodes to the destination nodes.

Manual migration is very useful if for example the scheduler has relocated a
process to a processor on a distant node.  A batch scheduler or an
administrator can detect the situation and move the pages of the process
nearer to the new processor.

sys_migrate_pages() could be used on non-numa machines as well, to force all
of a particualr process's pages out to swap, if someone thinks that's useful.

Larger installations usually partition the system using cpusets into sections
of nodes.  Paul has equipped cpusets with the ability to move pages when a
task is moved to another cpuset.  This allows automatic control over locality
of a process.  If a task is moved to a new cpuset then also all its pages are
moved with it so that the performance of the process does not sink
dramatically (as is the case today).

Swap migration works by simply evicting the page.  The pages must be faulted
back in.  The pages are then typically reallocated by the system near the node
where the process is executing.

For swap migration the destination of the move is controlled by the allocation
policy.  Cpusets set the allocation policy before calling sys_migrate_pages()
in order to move the pages as intended.

No allocation policy changes are performed for sys_migrate_pages().  This
means that the pages may not faulted in to the specified nodes if no
allocation policy was set by other means.  The pages will just end up near the
node where the fault occurred.

There's another patch series in the pipeline which implements "direct
migration".

The direct migration patchset extends the migration functionality to avoid
going through swap.  The destination node of the relation is controllable
during the actual moving of pages.  The crutch of using the allocation policy
to relocate is not necessary and the pages are moved directly to the target.
Its also faster since swap is not used.

And sys_migrate_pages() can then move pages directly to the specified node.
Implement functions to isolate pages from the LRU and put them back later.

This patch:

An earlier implementation was provided by Hirokazu Takahashi
<taka@valinux.co.jp> and IWAMOTO Toshihiro <iwamoto@valinux.co.jp> for the
memory hotplug project.

From: Magnus

This breaks out isolate_lru_page() and putpack_lru_page().  Needed for swap
migration.

Signed-off-by: Magnus Damm <magnus.damm@gmail.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>
18 years ago[PATCH] add schedule_on_each_cpu()
Christoph Lameter [Sun, 8 Jan 2006 09:00:43 +0000 (01:00 -0800)]
[PATCH] add schedule_on_each_cpu()

swap migration's isolate_lru_page() currently uses an IPI to notify other
processors that the lru caches need to be drained if the page cannot be
found on the LRU.  The IPI interrupt may interrupt a processor that is just
processing lru requests and cause a race condition.

This patch introduces a new function run_on_each_cpu() that uses the
keventd() to run the LRU draining on each processor.  Processors disable
preemption when dealing the LRU caches (these are per processor) and thus
executing LRU draining from another process is safe.

Thanks to Lee Schermerhorn <lee.schermerhorn@hp.com> for finding this race
condition.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] mm: free_pages opt
Nick Piggin [Sun, 8 Jan 2006 09:00:42 +0000 (01:00 -0800)]
[PATCH] mm: free_pages opt

Try to streamline free_pages_bulk by ensuring callers don't pass in a
'count' that exceeds the list size.

Some cleanups:
Rename __free_pages_bulk to __free_one_page.
Put the page list manipulation from __free_pages_ok into free_one_page.
Make __free_pages_ok static.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] mm: cleanup zone_pcp
Nick Piggin [Sun, 8 Jan 2006 09:00:41 +0000 (01:00 -0800)]
[PATCH] mm: cleanup zone_pcp

Use zone_pcp everywhere even though NUMA code "knows" the internal details
of the zone.  Stop other people trying to copy, and it looks nicer.

Also, only print the pagesets of online cpus in zoneinfo.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Make high and batch sizes of per_cpu_pagelists configurable
Rohit Seth [Sun, 8 Jan 2006 09:00:40 +0000 (01:00 -0800)]
[PATCH] Make high and batch sizes of per_cpu_pagelists configurable

As recently there has been lot of traffic on the right values for batch and
high water marks for per_cpu_pagelists.  This patch makes these two
variables configurable through /proc interface.

A new tunable /proc/sys/vm/percpu_pagelist_fraction is added.  This entry
controls the fraction of pages at most in each zone that are allocated for
each per cpu page list.  The min value for this is 8.  It means that we
don't allow more than 1/8th of pages in each zone to be allocated in any
single per_cpu_pagelist.

The batch value of each per cpu pagelist is also updated as a result.  It
is set to pcp->high/4.  The upper limit of batch is (PAGE_SHIFT * 8)

Signed-off-by: Rohit Seth <rohit.seth@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] drop-pagecache
Andrew Morton [Sun, 8 Jan 2006 09:00:39 +0000 (01:00 -0800)]
[PATCH] drop-pagecache

Add /proc/sys/vm/drop_caches.  When written to, this will cause the kernel to
discard as much pagecache and/or reclaimable slab objects as it can.  THis
operation requires root permissions.

It won't drop dirty data, so the user should run `sync' first.

Caveats:

a) Holds inode_lock for exorbitant amounts of time.

b) Needs to be taught about NUMA nodes: propagate these all the way through
   so the discarding can be controlled on a per-node basis.

This is a debugging feature: useful for getting consistent results between
filesystem benchmarks.  We could possibly put it under a config option, but
it's less than 300 bytes.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] slab: remove nested #ifdef CONFIG_NUMA
Christoph Lameter [Sun, 8 Jan 2006 09:00:38 +0000 (01:00 -0800)]
[PATCH] slab: remove nested #ifdef CONFIG_NUMA

For some reason there is an #ifdef CONFIG_NUMA within another #ifdef
CONFIG_NUMA in the page allocator.  Remove innermost #ifdef CONFIG_NUMA

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] slab: fix code formatting
Pekka Enberg [Sun, 8 Jan 2006 09:00:37 +0000 (01:00 -0800)]
[PATCH] slab: fix code formatting

The slab allocator code is inconsistent in coding style and messy.  For this
patch, I ran Lindent for mm/slab.c and fixed up goofs by hand.

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>
18 years ago[PATCH] slab: extract slab order calculation to separate function
Pekka Enberg [Sun, 8 Jan 2006 09:00:36 +0000 (01:00 -0800)]
[PATCH] slab: extract slab order calculation to separate function

This patch moves the ugly loop that determines the 'optimal' size (page order)
of cache slabs from kmem_cache_create() to a separate function and cleans it
up a bit.

Thanks to Matthew Wilcox for the help with this patch.

Signed-off-by: Matthew Dobson <colpatch@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] slab: extract slabinfo header printing to separate function
Pekka Enberg [Sun, 8 Jan 2006 09:00:36 +0000 (01:00 -0800)]
[PATCH] slab: extract slabinfo header printing to separate function

This patch extracts slabinfo header printing to a separate function
print_slabinfo_header() to make s_start() more readable.

Signed-off-by: Matthew Dobson <colpatch@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] slab: remove unused align parameter from alloc_percpu
Pekka Enberg [Sun, 8 Jan 2006 09:00:33 +0000 (01:00 -0800)]
[PATCH] slab: remove unused align parameter from alloc_percpu

__alloc_percpu and alloc_percpu both take an 'align' argument which is
completely ignored.  snmp6_mib_init() in net/ipv6/af_inet6.c attempts to use
it, but it will be ignored.  Therefore, remove the 'align' argument and fixup
the lone caller.

Signed-off-by: Matthew Dobson <colpatch@us.ibm.com>
Acked-by: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Fix compilation with CONFIG_MEMORY_HOTPLUG=y and gcc41.
Olaf Hering [Sun, 8 Jan 2006 09:00:32 +0000 (01:00 -0800)]
[PATCH] Fix compilation with CONFIG_MEMORY_HOTPLUG=y and gcc41.

Fix compilation with CONFIG_MEMORY_HOTPLUG=y and gcc41.
Also remove unneeded declations, add a public function.

drivers/base/memory.c:53: error: static declaration of 'register_memory_notifier' follows non-static declaration
include/linux/memory.h:85: error: previous declaration of 'register_memory_notifier' was here
drivers/base/memory.c:58: error: static declaration of 'unregister_memory_notifier' follows non-static declaration
include/linux/memory.h:86: error: previous declaration of 'unregister_memory_notifier' was here
drivers/base/memory.c:68: error: static declaration of 'register_memory' follows non-static declaration
include/linux/memory.h:73: error: previous declaration of 'register_memory' was here

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] ARM Netwinder watchdog wdt977 update
Woody Suwalski [Sun, 8 Jan 2006 09:00:31 +0000 (01:00 -0800)]
[PATCH] ARM Netwinder watchdog wdt977 update

Cleanup for the ARM-only watchdog driver wdt977.

This is probably the last update, since we want to merge with w83977f_wdt.
Jose Goncalves has ported this driver to i386, so probably we can iron out
configuration differences.

Signed-off-by: Woody Suwalski <woodys@xandros.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] small hp_sdc_rtc cleanup: use no_llseek
Marcelo Tosatti [Sun, 8 Jan 2006 09:00:29 +0000 (01:00 -0800)]
[PATCH] small hp_sdc_rtc cleanup: use no_llseek

Use no_llseek function.

Signed-off-by: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
Cc: "Brian S. Julin" <bri@calyx.com>
Acked-by: Vojtech Pavlik <vojtech@suse.cz>
Cc: Dmitry Torokhov <dtor_core@ameritech.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] asm-generic/atomic.h needs types.h
Andrew Morton [Sun, 8 Jan 2006 09:00:29 +0000 (01:00 -0800)]
[PATCH] asm-generic/atomic.h needs types.h

For BITS_PER_LONG

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] revert "mm: page_state fixes"
Andrew Morton [Sun, 8 Jan 2006 09:00:28 +0000 (01:00 -0800)]
[PATCH] revert "mm: page_state fixes"

Hugh says:

page_alloc_cpu_notify() specifically contains code to

  /* Add dead cpu's page_states to our own. */

which handles this more efficiently.

Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[IPV6]: small cleanups
Adrian Bunk [Sat, 7 Jan 2006 21:24:25 +0000 (13:24 -0800)]
[IPV6]: small cleanups

This patch contains the following cleanups:
- addrconf.c: make addrconf_dad_stop() static
- inet6_connection_sock.c should #include <net/inet6_connection_sock.h>
  for getting the prototypes of it's global functions

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[IPV4]: make ip_fragment() static
Adrian Bunk [Sat, 7 Jan 2006 21:23:39 +0000 (13:23 -0800)]
[IPV4]: make ip_fragment() static

Since there's no longer any external user of ip_fragment() we can make
it static.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: Add dummy nf_hook{_thresh}() when NETFILTER is disabled.
David S. Miller [Sat, 7 Jan 2006 20:50:27 +0000 (12:50 -0800)]
[NETFILTER]: Add dummy nf_hook{_thresh}() when NETFILTER is disabled.

Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: ip_conntrack_proto_sctp.c needs linux/interrupt.h
Joe Kappus [Sat, 7 Jan 2006 07:15:04 +0000 (23:15 -0800)]
[NETFILTER]: ip_conntrack_proto_sctp.c needs linux/interrupt.h

Signed-off-by: Joe Kappus <joecool1029@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[AX25/MKISS]: unbalanced spinlock_bh in ax_encaps()
Francois Romieu [Sat, 7 Jan 2006 07:08:42 +0000 (23:08 -0800)]
[AX25/MKISS]: unbalanced spinlock_bh in ax_encaps()

The unlocking disappeared during commit
5793f4be23f0171b4999ca68a39a9157b44139f3.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: Add ipt_policy/ip6t_policy matches
Patrick McHardy [Sat, 7 Jan 2006 07:06:48 +0000 (23:06 -0800)]
[NETFILTER]: Add ipt_policy/ip6t_policy matches

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: Handle NAT in IPsec policy checks
Patrick McHardy [Sat, 7 Jan 2006 07:06:30 +0000 (23:06 -0800)]
[NETFILTER]: Handle NAT in IPsec policy checks

Handle NAT of decapsulated IPsec packets by reconstructing the struct flowi
of the original packet from the conntrack information for IPsec policy
checks.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: Keep conntrack reference until IPsec policy checks are done
Patrick McHardy [Sat, 7 Jan 2006 07:06:10 +0000 (23:06 -0800)]
[NETFILTER]: Keep conntrack reference until IPsec policy checks are done

Keep the conntrack reference until policy checks have been performed for
IPsec NAT support. The reference needs to be dropped before a packet is
queued to avoid having the conntrack module unloadable.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: Redo policy lookups after NAT when neccessary
Patrick McHardy [Sat, 7 Jan 2006 07:05:36 +0000 (23:05 -0800)]
[NETFILTER]: Redo policy lookups after NAT when neccessary

When NAT changes the key used for the xfrm lookup it needs to be done
again. If a new policy is returned in POST_ROUTING the packet needs
to be passed to xfrm4_output_one manually after all hooks were called
because POST_ROUTING is called with fixed okfn (ip_finish_output).

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: Use conntrack information to determine if packet was NATed
Patrick McHardy [Sat, 7 Jan 2006 07:05:17 +0000 (23:05 -0800)]
[NETFILTER]: Use conntrack information to determine if packet was NATed

Preparation for IPsec support for NAT:
Use conntrack information instead of saving the saving and comparing the
addresses to determine if a packet was NATed and needs to be rerouted to
make it easier to extend the key.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NETFILTER]: Fix xfrm lookup in ip_route_me_harder/ip6_route_me_harder
Patrick McHardy [Sat, 7 Jan 2006 07:04:54 +0000 (23:04 -0800)]
[NETFILTER]: Fix xfrm lookup in ip_route_me_harder/ip6_route_me_harder

ip_route_me_harder doesn't use the port numbers of the xfrm lookup and
uses ip_route_input for non-local addresses which doesn't do a xfrm
lookup, ip6_route_me_harder doesn't do a xfrm lookup at all.

Use xfrm_decode_session and do the lookup manually, make sure both
only do the lookup if the packet hasn't been transformed already.

Makeing sure the lookup only happens once needs a new field in the
IP6CB, which exceeds the size of skb->cb. The size of skb->cb is
increased to 48b. Apparently the IPv6 mobile extensions need some
more room anyway.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[IPV4]: reset IPCB flags when neccessary
Patrick McHardy [Sat, 7 Jan 2006 07:04:01 +0000 (23:04 -0800)]
[IPV4]: reset IPCB flags when neccessary

Reset IPSKB_XFRM_TUNNEL_SIZE flags in ipip and ip_gre hard_start_xmit
function before the packet reenters IP. This is neccessary so the
encapsulated packets are checked not to be oversized in xfrm4_output.c
again. Reset all flags in sit when a packet changes its address family.

Also remove some obsolete IPSKB flags.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[IPV4/6]: Netfilter IPsec input hooks
Patrick McHardy [Sat, 7 Jan 2006 07:03:34 +0000 (23:03 -0800)]
[IPV4/6]: Netfilter IPsec input hooks

When the innermost transform uses transport mode the decapsulated packet
is not visible to netfilter. Pass the packet through the PRE_ROUTING and
LOCAL_IN hooks again before handing it to upper layer protocols to make
netfilter-visibility symetrical to the output path.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[IPV6]: Move nextheader offset to the IP6CB
Patrick McHardy [Sat, 7 Jan 2006 07:02:34 +0000 (23:02 -0800)]
[IPV6]: Move nextheader offset to the IP6CB

Move nextheader offset to the IP6CB to make it possible to pass a
packet to ip6_input_finish multiple times and have it skip already
parsed headers. As a nice side effect this gets rid of the manual
hopopts skipping in ip6_input_finish.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[XFRM]: Netfilter IPsec output hooks
Patrick McHardy [Sat, 7 Jan 2006 07:01:48 +0000 (23:01 -0800)]
[XFRM]: Netfilter IPsec output hooks

Call netfilter hooks before IPsec transforms. Packets visit the
FORWARD/LOCAL_OUT and POST_ROUTING hook before the first encapsulation
and the LOCAL_OUT and POST_ROUTING hook before each following tunnel mode
transform.

Patch from Herbert Xu <herbert@gondor.apana.org.au>:

Move the loop from dst_output into xfrm4_output/xfrm6_output since they're
the only ones who need to it. xfrm{4,6}_output_one() processes the first SA
all subsequent transport mode SAs and is called in a loop that calls the
netfilter hooks between each two calls.

In order to avoid the tail call issue, I've added the inline function
nf_hook which is nf_hook_slow plus the empty list check.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[XFRM]: Fix sparse warning.
Luiz Capitulino [Sat, 7 Jan 2006 06:59:43 +0000 (22:59 -0800)]
[XFRM]: Fix sparse warning.

security/selinux/xfrm.c:155:10: warning: Using plain integer as NULL pointer

Signed-off-by: Luiz Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[DCCP]: ipv6.c needs net/ip6_checksum.c
David S. Miller [Sat, 7 Jan 2006 06:55:39 +0000 (22:55 -0800)]
[DCCP]: ipv6.c needs net/ip6_checksum.c

Reported by Dave Jones.

Signed-off-by: David S. Miller <davem@davemloft.net>
18 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Sat, 7 Jan 2006 19:29:51 +0000 (11:29 -0800)]
Merge git://git./linux/kernel/git/dtor/input

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Sat, 7 Jan 2006 18:45:22 +0000 (10:45 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-arm

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-serial
Linus Torvalds [Sat, 7 Jan 2006 18:44:22 +0000 (10:44 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-serial

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-mmc
Linus Torvalds [Sat, 7 Jan 2006 18:43:40 +0000 (10:43 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-mmc

18 years ago[PATCH] fbcon: don´t call set_par() in fbcon_init() if vc_mode == KD_GRAPHICS
Knut Petersen [Sat, 7 Jan 2006 09:22:04 +0000 (10:22 +0100)]
[PATCH] fbcon: don´t call set_par() in fbcon_init() if vc_mode == KD_GRAPHICS

Nothing prevents a user to modprobe a framebuffer driver from e.g.  the
xterm prompt.  As a result, the set_par() function of the driver will be
called from fbcon_init().

This is fatal as a lot of X / framebuffer combinations are unable to
recover from set_par() reprogramming the graphics controller in
KD_GRAPHICS mode.

It is also unnecessary as the set_par() function will be called during a
switch to KD_TEXT anyway.  Because of this no side effects are possible.

Signed-off-by: Knut Petersen <Knut_Petersen@t-online.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[ARM] byteorder.h needs linux/compiler.h
Russell King [Sat, 7 Jan 2006 16:55:50 +0000 (16:55 +0000)]
[ARM] byteorder.h needs linux/compiler.h

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years agoInput: ibmasm - convert to dynamic input_dev allocation
Vernon Mauery [Sat, 7 Jan 2006 16:35:05 +0000 (11:35 -0500)]
Input: ibmasm - convert to dynamic input_dev allocation

Update the ibmasm driver to use the dynamic allocation of input_dev
structs to work with the sysfs subsystem.

Vojtech: Fixed some problems/bugs in the patch.
Dmitry: Fixed some more.

Signed-off-by: Vernon Mauery <vernux@us.ibm.com>
Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
18 years ago[ARM] Move asm/hardware/clock.h to linux/clk.h
Russell King [Sat, 7 Jan 2006 16:15:52 +0000 (16:15 +0000)]
[ARM] Move asm/hardware/clock.h to linux/clk.h

This is needs to be visible to other architectures using the AMBA
bus and peripherals.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] Move AMBA bus code to drivers/amba/
Russell King [Sat, 7 Jan 2006 14:54:15 +0000 (14:54 +0000)]
[ARM] Move AMBA bus code to drivers/amba/

Make the AMBA bus code visible to other architectures.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years agoMerge with Linus' kernel.
Russell King [Sat, 7 Jan 2006 14:40:05 +0000 (14:40 +0000)]
Merge with Linus' kernel.

18 years ago[ARM] Move AMBA include files to include/linux/amba/
Russell King [Sat, 7 Jan 2006 13:52:45 +0000 (13:52 +0000)]
[ARM] Move AMBA include files to include/linux/amba/

Since the ARM AMBA bus is used on MIPS as well as ARM, we need
to make the bus available for other architectures to use.  Move
the AMBA include files from include/asm-arm/hardware/ to
include/linux/amba/

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3239/1: Add ARM optimised swab32
Andre McCurdy [Sat, 7 Jan 2006 11:39:20 +0000 (11:39 +0000)]
[ARM] 3239/1: Add ARM optimised swab32

Patch from Andre McCurdy

Replaces generic swab32 routine with a more ARM friendly version.
Reduces kernel text size by approx 1200 bytes when compiled with
3.4.4 and approx 2400 bytes with 4.0.2

Probably some performance benefit as well.

Signed-off-by: Andre McCurdy <armccurdy@yahoo.co.uk>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3235/1: SharpSL PM: Fix a gcc4 build error
Richard Purdie [Sat, 7 Jan 2006 11:12:26 +0000 (11:12 +0000)]
[ARM] 3235/1: SharpSL PM: Fix a gcc4 build error

Patch from Richard Purdie

Fix a gcc4 build error (incomplete element type) in the pxa SharpSL
PM code.

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3232/1: i.MX Frame Buffer undeclared "dev" variable fix
Pavel Pisa [Sat, 7 Jan 2006 10:44:32 +0000 (10:44 +0000)]
[ARM] 3232/1: i.MX Frame Buffer undeclared "dev" variable fix

Patch from Pavel Pisa

Correction of the code broken by update
whole-tree platform devices update.

Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ACPI] remove Kconfig "default y" for laptop drivers
Borislav Petkov [Fri, 6 Jan 2006 06:31:00 +0000 (01:31 -0500)]
[ACPI] remove Kconfig "default y" for laptop drivers

Signed-off-by: Borislav Petkov <petkov@uni-muenster.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Len Brown <len.brown@intel.com>
18 years ago[ACPI] fix kernel-doc warnings in acpi/scan.c
Randy Dunlap [Fri, 6 Jan 2006 06:31:00 +0000 (01:31 -0500)]
[ACPI] fix kernel-doc warnings in acpi/scan.c

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Len Brown <len.brown@intel.com>
18 years agoAuto-update from upstream
Len Brown [Sat, 7 Jan 2006 07:05:40 +0000 (02:05 -0500)]
Auto-update from upstream

18 years ago[PATCH] Fix posix-cpu-timers sched_time accumulation
David S. Miller [Fri, 6 Jan 2006 23:36:48 +0000 (15:36 -0800)]
[PATCH] Fix posix-cpu-timers sched_time accumulation

I've spent the past 3 days digging into a glibc testsuite failure in
current CVS, specifically libc/rt/tst-cputimer1.c The thr1 and thr2
timers fire too early in the second pass of this test.  The second
pass is noteworthy because it makes use of intervals, whereas the
first pass does not.

All throughout the posix-cpu-timers.c code, the calculation of the
process sched_time sum is implemented roughly as:

unsigned long long sum;

sum = tsk->signal->sched_time;
t = tsk;
do {
sum += t->sched_time;
t = next_thread(t);
} while (t != tsk);

In fact this is the exact scheme used by check_process_timers().

In the case of check_process_timers(), current->sched_time has just
been updated (via scheduler_tick(), which is invoked by
update_process_times(), which subsequently invokes
run_posix_cpu_timers()) So there is no special processing necessary
wrt. that.

In other contexts, we have to allot for the fact that tsk->sched_time
might be a bit out of date if we are current.  And the
posix-cpu-timers.c code uses current_sched_time() to deal with that.

Unfortunately it does so in an erroneous and inconsistent manner in
one spot which is what results in the early timer firing.

In cpu_clock_sample_group_locked(), it does this:

cpu->sched = p->signal->sched_time;
/* Add in each other live thread.  */
while ((t = next_thread(t)) != p) {
cpu->sched += t->sched_time;
}
if (p->tgid == current->tgid) {
/*
 * We're sampling ourselves, so include the
 * cycles not yet banked.  We still omit
 * other threads running on other CPUs,
 * so the total can always be behind as
 * much as max(nthreads-1,ncpus) * (NSEC_PER_SEC/HZ).
 */
cpu->sched += current_sched_time(current);
} else {
cpu->sched += p->sched_time;
}

The problem is the "p->tgid == current->tgid" test.  If "p" is
not current, and the tgids are the same, we will add the process
t->sched_time twice into cpu->sched and omit "p"'s sched_time
which is very very very wrong.

posix-cpu-timers.c has a helper function, sched_ns(p) which takes care
of this, so my fix is to use that here instead of this special tgid
test.

The fact that current can be one of the sub-threads of "p" points out
that we could make things a little bit more accurate, perhaps by using
sched_ns() on every thread we process in these loops.  It also points
out that we don't use the most accurate value for threads in the group
actively running other cpus (and this is mentioned in the comment).

But that is a future enhancement, and this fix here definitely makes
sense.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/gregkh/i2c-2.6
Linus Torvalds [Fri, 6 Jan 2006 23:25:08 +0000 (15:25 -0800)]
Merge /pub/scm/linux/kernel/git/gregkh/i2c-2.6

18 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Fri, 6 Jan 2006 23:24:28 +0000 (15:24 -0800)]
Merge /pub/scm/linux/kernel/git/davem/net-2.6

18 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild
Linus Torvalds [Fri, 6 Jan 2006 23:23:56 +0000 (15:23 -0800)]
Merge git://git./linux/kernel/git/sam/kbuild

18 years agoMerge git://git.linux-nfs.org/pub/linux/nfs-2.6
Linus Torvalds [Fri, 6 Jan 2006 23:22:57 +0000 (15:22 -0800)]
Merge git://git.linux-nfs.org/pub/linux/nfs-2.6

18 years ago[PATCH] fs/ufs: debug mode compilation failure
Evgeniy [Fri, 6 Jan 2006 18:18:01 +0000 (21:18 +0300)]
[PATCH] fs/ufs: debug mode compilation failure

This patch should fix compilation failure of fs/ufs/dir.c with defined UFS_DIR_DEBUG

Signed-off-by: Evgeniy Dushistov <dushistov@mail.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] shrink mmtimer memory size
Dimitri Sivanich [Fri, 6 Jan 2006 17:33:41 +0000 (11:33 -0600)]
[PATCH] shrink mmtimer memory size

This greatly reduces the amount of memory used by mmtimer on smaller
machines with large values of MAX_COMPACT_NODES.

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] MAINTAINERS: remove the outdated DAC960 entry
Adrian Bunk [Fri, 6 Jan 2006 17:26:29 +0000 (18:26 +0100)]
[PATCH] MAINTAINERS: remove the outdated DAC960 entry

Randy Dunlap: "Dave is no longer at OSDL and is no longer maintaining
that driver."

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years agoAuto-update from upstream
Len Brown [Fri, 6 Jan 2006 21:34:21 +0000 (16:34 -0500)]
Auto-update from upstream

18 years ago[NET]: Endian-annotate in_aton()
Alexey Dobriyan [Fri, 6 Jan 2006 21:24:54 +0000 (13:24 -0800)]
[NET]: Endian-annotate in_aton()

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NET]: Endian-annotate struct iphdr
Alexey Dobriyan [Fri, 6 Jan 2006 21:24:29 +0000 (13:24 -0800)]
[NET]: Endian-annotate struct iphdr

And fix trivial warnings that emerged.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[LSM-IPSec]: Corrections to LSM-IPSec Nethooks
Trent Jaeger [Fri, 6 Jan 2006 21:22:39 +0000 (13:22 -0800)]
[LSM-IPSec]: Corrections to LSM-IPSec Nethooks

This patch contains two corrections to the LSM-IPsec Nethooks patches
previously applied.

(1) free a security context on a failed insert via xfrm_user
interface in xfrm_add_policy.  Memory leak.

(2) change the authorization of the allocation of a security context
in a xfrm_policy or xfrm_state from both relabelfrom and relabelto
to setcontext.

Signed-off-by: Trent Jaeger <tjaeger@cse.psu.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[PKTGEN]: Adds missing __init.
Luiz Capitulino [Fri, 6 Jan 2006 21:19:31 +0000 (13:19 -0800)]
[PKTGEN]: Adds missing __init.

pktgen_find_thread() and pktgen_create_thread() are only called at
initialization time.

Signed-off-by: Luiz Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[ACPI] linux-acpi@vger.kernel.org replaces acpi-devel@lists.sourceforge.net
Len Brown [Fri, 6 Jan 2006 21:19:26 +0000 (16:19 -0500)]
[ACPI] linux-acpi@vger. replaces acpi-devel@lists.sourceforge.net

Signed-off-by: Len Brown <len.brown@intel.com>
18 years ago[NETFILTER]: ipt_helper.c needs linux/interrupt.h
Joe [Fri, 6 Jan 2006 21:15:11 +0000 (13:15 -0800)]
[NETFILTER]: ipt_helper.c needs linux/interrupt.h

From: Joe <joecool1029@gmail.com>

Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[BRIDGE] netfilter: vlan + hw checksum = bug?
Stephen Hemminger [Fri, 6 Jan 2006 21:13:29 +0000 (13:13 -0800)]
[BRIDGE] netfilter: vlan + hw checksum = bug?

It looks like the bridge netfilter code does not correctly update
the hardware checksum after popping off the VLAN header.

This is by inspection, I have *not* tested this.
To test you would need to set up a filtering bridge with vlans
and a device the does hardware receive checksum (skge, or sungem)

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[X25]: Fix for broken x25 module.
Shaun Pereira [Fri, 6 Jan 2006 21:11:35 +0000 (13:11 -0800)]
[X25]: Fix for broken x25 module.

When a user-space server application calls bind on a socket, then in kernel
space this bound socket is considered 'x25-linked' and the SOCK_ZAPPED flag
is unset.(As in x25_bind()/af_x25.c).

Now when a user-space client application attempts to connect to the server
on the listening socket, if the kernel accepts this in-coming call, then it
returns a new socket to userland and attempts to reply to the caller.

The reply/x25_sendmsg() will fail, because the new socket created on
call-accept has its SOCK_ZAPPED flag set by x25_make_new().
(sock_init_data() called by x25_alloc_socket() called by x25_make_new()
sets the flag to SOCK_ZAPPED)).

Fix: Using the sock_copy_flag() routine available in sock.h fixes this.

Tested on 32 and 64 bit kernels with x25 over tcp.

Signed-off-by: Shaun Pereira <pereira.shaun@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NET]: Change sk_run_filter()'s return type in net/core/filter.c
Kris Katterjohn [Fri, 6 Jan 2006 21:08:20 +0000 (13:08 -0800)]
[NET]: Change sk_run_filter()'s return type in net/core/filter.c

It should return an unsigned value, and fix sk_filter() as well.

Signed-off-by: Kris Katterjohn <kjak@ispwest.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[NET]: Use newer is_multicast_ether_addr() in some files
Kris Katterjohn [Fri, 6 Jan 2006 21:05:58 +0000 (13:05 -0800)]
[NET]: Use newer is_multicast_ether_addr() in some files

This uses is_multicast_ether_addr() because it has recently been
changed to do the same thing these seperate tests are doing.

Signed-off-by: Kris Katterjohn <kjak@users.sourceforge.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years agoMerge ../torvalds-2.6/
Greg Kroah-Hartman [Fri, 6 Jan 2006 20:59:59 +0000 (12:59 -0800)]
Merge ../torvalds-2.6/

18 years agokconfig: fix gconfig with POSIXLY_CORRECT=1
Adrian Bunk [Mon, 2 Jan 2006 10:25:30 +0000 (11:25 +0100)]
kconfig: fix gconfig with POSIXLY_CORRECT=1

This patch fixed "make gconfig" with POSIXLY_CORRECT=1 set.

This issue was reported by Jens Elkner <elkner@linofee.org> in kernel
Bugzilla #2919.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
18 years agokbuild: un-stringnify KBUILD_MODNAME
Sam Ravnborg [Fri, 6 Jan 2006 20:17:50 +0000 (21:17 +0100)]
kbuild: un-stringnify KBUILD_MODNAME

Now when kbuild passes KBUILD_MODNAME with "" do not __stringify it when
used. Remove __stringnify for all users.
This also fixes the output of:

$ ls -l /sys/module/
drwxr-xr-x 4 root root 0 2006-01-05 14:24 pcmcia
drwxr-xr-x 4 root root 0 2006-01-05 14:24 pcmcia_core
drwxr-xr-x 3 root root 0 2006-01-05 14:24 "processor"
drwxr-xr-x 3 root root 0 2006-01-05 14:24 "psmouse"

The quoting of the module names will be gone again.
Thanks to GregKH + Kay Sievers for reproting this.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
18 years agoSUNRPC: Make krb5 report unsupported encryption types
J. Bruce Fields [Tue, 3 Jan 2006 08:56:01 +0000 (09:56 +0100)]
SUNRPC: Make krb5 report unsupported encryption types

 Print messages when an unsupported encrytion algorthm is requested or
 there is an error locating a supported algorthm.

Signed-off-by: Kevin Coffman <kwc@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: Make spkm3 report unsupported encryption types
J. Bruce Fields [Tue, 3 Jan 2006 08:56:01 +0000 (09:56 +0100)]
SUNRPC: Make spkm3 report unsupported encryption types

 Print messages when an unsupported encrytion algorthm is requested or
 there is an error locating a supported algorthm.

Signed-off-by: Kevin Coffman <kwc@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: Update the spkm3 code to use the make_checksum interface
J. Bruce Fields [Tue, 3 Jan 2006 08:56:00 +0000 (09:56 +0100)]
SUNRPC: Update the spkm3 code to use the make_checksum interface

 Also update the tokenlen calculations to accomodate g_token_size().

Signed-off-by: Andy Adamson <andros@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Fix an Oops in nfs_do_expire_all_delegations
Trond Myklebust [Tue, 3 Jan 2006 08:55:58 +0000 (09:55 +0100)]
NFSv4: Fix an Oops in nfs_do_expire_all_delegations

 If the loop errors, we need to exit.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Allow entries in the idmap cache to expire
Trond Myklebust [Tue, 3 Jan 2006 08:55:57 +0000 (09:55 +0100)]
NFSv4: Allow entries in the idmap cache to expire

 If someone changes the uid/gid mapping in userland, then we do eventually
 want those changes to be propagated to the kernel. Currently the kernel
 assumes that it may cache entries forever.

 Add an expiration time + garbage collector for idmap entries.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: Clean up xprt_destroy()
Trond Myklebust [Tue, 3 Jan 2006 08:55:56 +0000 (09:55 +0100)]
SUNRPC: Clean up xprt_destroy()

 We ought never to be calling xprt_destroy() if there are still active
 rpc_tasks. Optimise away the broken code that attempts to "fix" that case.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: Ensure client closes the socket when server initiates a close
Trond Myklebust [Tue, 3 Jan 2006 08:55:55 +0000 (09:55 +0100)]
SUNRPC: Ensure client closes the socket when server initiates a close

 If the server decides to close the RPC socket, we currently don't actually
 respond until either another RPC call is scheduled, or until xprt_autoclose()
 gets called by the socket expiry timer (which may be up to 5 minutes
 later).

 This patch ensures that xprt_autoclose() is called much sooner if the
 server closes the socket.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFS: get rid of some needless code obfuscation in xdr_encode_sattr().
Trond Myklebust [Tue, 3 Jan 2006 08:55:54 +0000 (09:55 +0100)]
NFS: get rid of some needless code obfuscation in xdr_encode_sattr().

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFS: Send valid mode bits to the server
Trond Myklebust [Tue, 3 Jan 2006 08:55:53 +0000 (09:55 +0100)]
NFS: Send valid mode bits to the server

 inode->i_mode contains a lot more than just the mode bits. Make sure that
 we mask away this extra stuff in SETATTR calls to the server.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: get rid of cl_chatty
Chuck Lever [Tue, 3 Jan 2006 08:55:52 +0000 (09:55 +0100)]
SUNRPC: get rid of cl_chatty

 Clean up: Every ULP that uses the in-kernel RPC client, except the NLM
 client, sets cl_chatty.  There's no reason why NLM shouldn't set it, so
 just get rid of cl_chatty and always be verbose.

 Test-plan:
 Compile with CONFIG_NFS enabled.

Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: transport switch API for setting port number
Chuck Lever [Tue, 3 Jan 2006 08:55:51 +0000 (09:55 +0100)]
SUNRPC: transport switch API for setting port number

 At some point, transport endpoint addresses will no longer be IPv4.  To hide
 the structure of the rpc_xprt's address field from ULPs and port mappers,
 add an API for setting the port number during an RPC bind operation.

 Test-plan:
 Destructive testing (unplugging the network temporarily).  Connectathon
 with UDP and TCP.  NFSv2/3 and NFSv4 mounting should be carefully checked.
 Probably need to rig a server where certain services aren't running, or
 that returns an error for some typical operation.

Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: new interface to force an RPC rebind
Chuck Lever [Tue, 3 Jan 2006 08:55:50 +0000 (09:55 +0100)]
SUNRPC: new interface to force an RPC rebind

 We'd like to hide fields in rpc_xprt and rpc_clnt from upper layer protocols.
 Start by creating an API to force RPC rebind, replacing logic that simply
 sets cl_port to zero.

 Test-plan:
 Destructive testing (unplugging the network temporarily).  Connectathon
 with UDP and TCP.  NFSv2/3 and NFSv4 mounting should be carefully checked.
 Probably need to rig a server where certain services aren't running, or
 that returns an error for some typical operation.

Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: switchable buffer allocation
Chuck Lever [Tue, 3 Jan 2006 08:55:49 +0000 (09:55 +0100)]
SUNRPC: switchable buffer allocation

 Add RPC client transport switch support for replacing buffer management
 on a per-transport basis.

 In the current IPv4 socket transport implementation, RPC buffers are
 allocated as needed for each RPC message that is sent.  Some transport
 implementations may choose to use pre-allocated buffers for encoding,
 sending, receiving, and unmarshalling RPC messages, however.  For
 transports capable of direct data placement, the buffers can be carved
 out of a pre-registered area of memory rather than from a slab cache.

 Test-plan:
 Millions of fsx operations.  Performance characterization with "sio" and
 "iozone".  Use oprofile and other tools to look for significant regression
 in CPU utilization.

Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv3: try get_root user-supplied security_flavor
J. Bruce Fields [Tue, 3 Jan 2006 08:55:48 +0000 (09:55 +0100)]
NFSv3: try get_root user-supplied security_flavor

 Thanks to Ed Keizer for bug and root cause.  He says: "... we could only mount
 the top-level Solaris share. We could not mount deeper into the tree.
 Investigation showed that Solaris allows UNIX authenticated FSINFO only on the
 top level of the share. This is a problem because we share/export our home
 directories one level higher than we mount them. I.e. we share the partition
 and not the individual home directories. This prevented access to home
 directories."

 We still may need to try auth_sys for the case where the client doesn't have
 appropriate credentials.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNLM: fix parsing of sm notify procedure
J. Bruce Fields [Tue, 3 Jan 2006 08:55:46 +0000 (09:55 +0100)]
NLM: fix parsing of sm notify procedure

 The procedure that decodes statd sm_notify call seems to be skipping a
 few arguments.  How did this ever work?

 >From folks at Polyserve.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNLM: Further cancel fixes
J. Bruce Fields [Tue, 3 Jan 2006 08:55:46 +0000 (09:55 +0100)]
NLM: Further cancel fixes

 If the server receives an NLM cancel call and finds no waiting lock to
 cancel, then chances are the lock has already been applied, and the client
 just hadn't yet processed the NLM granted callback before it sent the
 cancel.

 The Open Group text, for example, perimts a server to return either success
 (LCK_GRANTED) or failure (LCK_DENIED) in this case.  But returning an error
 seems more helpful; the client may be able to use it to recognize that a
 race has occurred and to recover from the race.

 So, modify the relevant functions to return an error in this case.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>