drivers/block/floppy.c: remove most uses of CALL and ECALL macros

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove [U]CLEARF, [U]SETF, and [U]TESTF macros

Use clear_bit, set_bit, and test_bit functions directly

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: add debug_dcl(...) macro

Converted #ifdef DCL_DEBUG if (test) DPRINTK(...); #endif
to debug_dcl(test, ...);

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove macro LOCK_FDC

Macros with hidden returns aren't nice.

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove a few spaces from function casts

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove IN/OUT macros, indent switch/case

Remove ugly IN/OUT macros, use direct case and code
Add missing semicolon after ECALL

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: indent a comment

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove CLEARSTRUCT macro, use memset

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: comment neatening and remove naked ;

Spacing, column alignment and a for loop with
a naked semicolon converted to an assign and while

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove LAST_OUT macro

Macros with hidden returns are not nice.
Convert the 2 uses to use direct code.

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: hoist assigns from if()s, neatening

Move assigns above if()s
Remove unnecessary parentheses from returns
Use a temporary for a duplicated test

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove used once CHECK_READY macro

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: remove unnecessary braces

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: use pr_<level>

Convert bare printk to pr_info and pr_cont
Convert printk(KERN_ERR to pr_err

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: #define space and column neatening

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/block/floppy.c: convert some #include <asm/ to #include <linux/

Signed-off-by: Joe Perches <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Marcin Slusarz <[email protected]>
Cc: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

mm: introduce dump_page() and print symbolic flag names

- introduce dump_page() to print the page info for debugging some error
  condition.

- convert three mm users: bad_page(), print_bad_pte() and memory offline
  failure.

- print an extra field: the symbolic names of page->flags

Example dump_page() output:

[  157.521694] page:ffffea0000a7cba8 count:2 mapcount:1 mapping:ffff88001c901791 index:0x147
[  157.525570] page flags: 0x100000000100068(uptodate|lru|active|swapbacked)

Signed-off-by: Wu Fengguang <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Alex Chiang <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: KOSAKI Motohiro <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

drivers/scsi/ses.c: eliminate double free

The few lines below the kfree of hdr_buf may go to the label err_free
which will also free hdr_buf.  The most straightforward solution seems to
be to just move the kfree of hdr_buf after these gotos.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@r@
identifier E;
expression E1;
iterator I;
statement S;
@@

*kfree(E);
... when != E = E1
    when != I(E,...) S
    when != &E
*kfree(E);
// </smpl>

Signed-off-by: Julia Lawall <[email protected]>
Cc: James Bottomley <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

mm: do not iterate over NR_CPUS in __zone_pcp_update()

__zone_pcp_update() iterates over NR_CPUS instead of limiting the access
to the possible cpus.  This might result in access to uninitialized areas
as the per cpu allocator only populates the per cpu memory for possible
cpus.

This problem was created as a result of the dynamic allocation of pagesets
from percpu memory that went in during the merge window - commit
99dcc3e5a94ed491fbef402831d8c0bbb267f995 ("this_cpu: Page allocator
conversion").

Signed-off-by: Thomas Gleixner <[email protected]>
Acked-by: Pekka Enberg <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Acked-by: Christoph Lameter <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Reviewed-by: KOSAKI Motohiro <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

nommu: fix build breakage

Commit 34e55232e59f7b19050267a05ff1226e5cd122a5 ("mm: avoid false sharing
of mm_counter") added sync_mm_rss() for syncing loosely accounted rss
counters.  It's for CONFIG_MMU but sync_mm_rss is called even in NOMMU
enviroment (kerne/exit.c, fs/exec.c).  Above commit doesn't handle it
well.

This patch changes
  SPLIT_RSS_COUNTING depends on SPLIT_PTLOCKS && CONFIG_MMU

And for avoid unnecessary function calls, sync_mm_rss changed to be inlined
noop function in header file.

Reported-by: David Howells <[email protected]>
Signed-off-by: KAMEZAWA Hiroyuki <[email protected]>
Signed-off-by: Mike Frysinger <[email protected]>
Signed-off-by: Michal Simek <[email protected]>
Signed-off-by: David Howells <[email protected]>
Cc: Greg Ungerer <[email protected]>
Cc: Geert Uytterhoeven <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

MAINTAINERS: add bfin_sdh driver

Signed-off-by: Mike Frysinger <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

sdio: recognize io card without powercycle

SDIO Simplified Specification V2.00 states that it is strongly recommended
that the host executes either a power reset or issues a CMD52 (I/O Reset)
to re-initialize an I/O only card or the I/O portion of a combo card.
Additionally, the CMD52 must be issued first because it cannot be issued
after a CMD0.

With this patch the Nintendo Wii SDIO-based WLAN card is detected after a
system reset, without requiring a complete system powercycle.

Signed-off-by: Albert Herranz <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

rtc/hctosys: only claim the RTC provided the system time if it did

Without this patch /sys/class/rtc/$CONFIG_RTC_HCTOSYS_DEVICE/hctosys
contains a 1 (meaning "This rtc was used to initialize the system clock")
even if reading the time at bootup failed.

Moreover change error handling in rtc_hctosys() to use goto and so reduce
the indention level.

Signed-off-by: Uwe Kleine-König <[email protected]>
Cc: Paul Gortmaker <[email protected]>
Acked-by: Alessandro Zummo <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

fs: buffer_head: remove kmem_cache constructor to reduce memory usage under slub

When using slub, having a kmem_cache constructor forces slub to add a free
pointer to the size of the cached object, which can have a significant
impact to the number of small objects that can fit into a slab.

As buffer_head is relatively small and we can have large numbers of them,
removing the constructor is a definite win.

On x86_64 removing the constructor gives me 39 objects/slab, 3 more than
without the patch.  And on x86_32 73 objects/slab, which is 9 more.

As alloc_buffer_head() already initializes each new object there is very
little difference in actual code run.

Signed-off-by: Richard Kennedy <[email protected]>
Cc: Alexander Viro <[email protected]>
Cc: Jens Axboe <[email protected]>
Acked-by: Nick Piggin <[email protected]>
Cc: "Theodore Ts'o" <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

fs/ocfs2/cluster/tcp.c: remove use of NIPQUAD, use %pI4

Signed-off-by: Joe Perches <[email protected]>
Cc: Mark Fasheh <[email protected]>
Acked-by: Joel Becker <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

Merge branch 'for-linus' of git://git./linux/kernel/git/jbarnes/pci-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
  PCI/PM Runtime: Make runtime PM of PCI devices inactive by default

Merge git://git./linux/kernel/git/mason/btrfs-unstable

* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
  Btrfs: kfree correct pointer during mount option parsing
  Btrfs: use RB_ROOT to intialize rb_trees instead of setting rb_node to NULL

Btrfs: kfree correct pointer during mount option parsing

We kstrdup the options string, but then strsep screws with the pointer,
so when we kfree() it, we're not giving it the right pointer.

Tested-by: Andy Lutomirski <[email protected]>
Signed-off-by: Chris Mason <[email protected]>

Btrfs: use RB_ROOT to intialize rb_trees instead of setting rb_node to NULL

btrfs inialize rb trees in quite a number of places by settin rb_node =
NULL;  The problem with this is that 17d9ddc72fb8bba0d4f678 in the
linux-next tree adds a new field to that struct which needs to be NULL for
the new rbtree library code to work properly.  This patch uses RB_ROOT as
the intializer so all of the relevant fields will be NULL'd.  Without the
patch I get a panic.

Signed-off-by: Eric Paris <[email protected]>
Acked-by: Venkatesh Pallipadi <[email protected]>
Signed-off-by: Chris Mason <[email protected]>

Linux 2.6.34-rc1

Merge git://git./linux/kernel/git/gregkh/driver-core-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core-2.6: (62 commits)
  msi-laptop: depends on RFKILL
  msi-laptop: Detect 3G device exists by standard ec command
  msi-laptop: Add resume method for set the SCM load again
  msi-laptop: Support some MSI 3G netbook that is need load SCM
  msi-laptop: Add threeg sysfs file for support query 3G state by standard 66/62 ec command
  msi-laptop: Support standard ec 66/62 command on MSI notebook and nebook
  Driver core: create lock/unlock functions for struct device
  sysfs: fix for thinko with sysfs_bin_attr_init()
  sysfs: Kill unused sysfs_sb variable.
  sysfs: Pass super_block to sysfs_get_inode
  driver core: Use sysfs_rename_link in device_rename
  sysfs: Implement sysfs_rename_link
  sysfs: Pack sysfs_dirent more tightly.
  sysfs: Serialize updates to the vfs inode
  sysfs: windfarm: init sysfs attributes
  sysfs: Use sysfs_attr_init and sysfs_bin_attr_init on module dynamic attributes
  sysfs: Document sysfs_attr_init and sysfs_bin_attr_init
  sysfs: Use sysfs_attr_init and sysfs_bin_attr_init on dynamic attributes
  sysfs: Use one lockdep class per sysfs attribute.
  sysfs: Only take active references on attributes.
  ...

Merge git://git./linux/kernel/git/wim/linux-2.6-watchdog

* git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog:
  [WATCHDOG] i6300esb.c: change platform_driver to pci_driver
  [WATCHDOG] i6300esb: fix unlock register with
  [WATCHDOG] drivers/watchdog/wdt.c:wdt_ioctl(): make `ident' non-static
  [WATCHDOG] change reboot_notifier to platform-shutdown method.
  [WATCHDOG] watchdog_info constify
  [WATCHDOG] gef_wdt: Author corrections following split of GE Fanuc joint venture
  [WATCHDOG] iTCO_wdt: clean up probe(), modify err msg
  [WATCHDOG] ep93xx: watchdog timer driver for TS-72xx SBCs cleanup
  [WATCHDOG] support for max63xx watchdog timer chips
  [WATCHDOG] ep93xx: added platform side support for TS-72xx WDT driver
  [WATCHDOG] ep93xx: implemented watchdog timer driver for TS-72xx SBCs

Merge branch 'for-linus' of git://git./linux/kernel/git/tiwai/sound-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6: (26 commits)
  ALSA: hdmi - show debug message on changing audio infoframe
  ALSA: hdmi - merge common code for intelhdmi and nvhdmi
  ALSA: hda - Add ASRock mobo to MSI blacklist
  ALSA: hda: uninitialized variable fix
  ALSA: hda: Use LPIB for a Biostar Microtech board
  ALSA: usb/audio.h: Fix field order
  ALSA: fix jazz16 compile (udelay)
  ALSA: hda: Use LPIB for Dell Latitude 131L
  ALSA: hda - Build hda_eld into snd-hda-codec module
  ALSA: hda - Support NVIDIA MCP89 and GT21x hdmi audio
  ALSA: hda - Support max codecs to 8 for nvidia hda controller
  ALSA: riptide: clean up while loop
  ALSA: usbaudio - remove debug "SAMPLE BYTES" printk line
  ALSA: timer - pass real event in snd_timer_notify1() to instance callback
  ALSA: oxygen: change || to &&
  ALSA: opti92x: use PnP data to select Master Control port
  ASoC: fix ak4104 register array access
  ASoC: soc_pcm_open: Add missing bailout tag
  ALSA: usbaudio: Fix wrong bitrate for Creative Creative VF0470 Live Cam
  ALSA: ua101: removing debugging code
  ...

Merge branch 'for-linus' of git://git390.marist.edu/linux-2.6

* 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6:
  [S390] smsgiucv_app: deliver z/VM CP special messages (SMSG) as uevents
  [S390] smsgiucv: declare char pointers as "const"
  [S390] dasd: automatic recognition of read-only devices
  [S390] remove unused qdio flags in zfcp and qeth
  [S390] Cleanup xtime usage
  [S390] qdio: add missing bracket
  [S390] cio: fix init_count in case of recognition after steal lock
  [S390] dasd: security and PSF update patch for EMC CKD ioctl
  [S390] hvc_iucv: allocate memory buffers for IUCV in zone DMA
  [S390] uaccess: make sure copy_from_user_overflow is builtin

FS-Cache: Remove the EXPERIMENTAL flag

Remove the EXPERIMENTAL flag from FS-Cache so that Ubuntu can make use of the
facility.

Signed-off-by: Christian Kujau <[email protected]>
Signed-off-by: David Howells <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

[WATCHDOG] i6300esb.c: change platform_driver to pci_driver

Since the intel 6300esb platform uses a dedicated PCI ID for
it's watchdog timer (00:1d.4), it's better and easier to have
this as a pci_driver.

Signed-off-by: Wim Van Sebroeck <[email protected]>

[S390] smsgiucv_app: deliver z/VM CP special messages (SMSG) as uevents

The smsgiucv_app driver registers a callback with the smsgiucv driver
to receive z/VM CP special messages (SMSG) starting with "APP".

When the callback is called for special messages, the driver creates
an uevent for the received message.  The uevent consists of additional
environment data containing the message prefix ("APP"), message sender,
and message content.

udev rules can be used to trigger application specific actions through
matching the content or sender of the special message.

Signed-off-by: Hendrik Brueckner <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] smsgiucv: declare char pointers as "const"

Declare the smsgiucv prefix char pointer as "const" and use
use const char pointers in callback functions.

Signed-off-by: Hendrik Brueckner <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] dasd: automatic recognition of read-only devices

In z/VM it is possible to attach a device as read-only. To prevent
unintentional write requests and subsequent I/O errors, we can detect
this configuration using the z/VM DIAG 210 interface and set the
respective linux block device to read-only as well.

Signed-off-by: Stefan Weinhuber <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] remove unused qdio flags in zfcp and qeth

zfcp and qeth are setting flags for the qdio-layer, but these flags
are not used in qdio. Patch removes the flag definitions from qdio
and their settings in zfcp and qeth.

Cc: Jan Glauber <[email protected]>
Cc: Christof Schmitt <[email protected]>
Signed-off-by: Ursula Braun <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] Cleanup xtime usage

This replaces direct xtime usage in the s390 arch with timekeeping accessors,
so we can further clean up the timekeeping core.

Cc: Thomas Gleixner <[email protected]>
Signed-off-by: John Stultz <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] qdio: add missing bracket

Add a missing bracket to only log the outbound handler event in the
appropriate case.

Signed-off-by: Ursula Braun <[email protected]>
Signed-off-by: Jan Glauber <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] cio: fix init_count in case of recognition after steal lock

After we try to steal a lock on a ccw device in boxed state,
we have to restart device recognition and potentially reprobing.

In this case ccw_device_init_count was erroneously decreased
twice. This patch fixes the issue.

Signed-off-by: Sebastian Ott <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] dasd: security and PSF update patch for EMC CKD ioctl

Remove the PSF order/suborder check from the Symmetrix CKD dasd ioctl.
In exchange restrict the ioctl to CAP_SYS_ADMIN and CAP_SYS_RAWIO.

Signed-off-by: Nigel Hislop <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] hvc_iucv: allocate memory buffers for IUCV in zone DMA

The device driver must allocate memory for IUCV buffers with GFP_DMA,
because IUCV cannot address memory above 2GB (31bit addresses only).

Because the IUCV ignores the higher bits of the address, sending and
receiving IUCV data with this driver might cause memory corruptions.

Signed-off-by: Hendrik Brueckner <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[S390] uaccess: make sure copy_from_user_overflow is builtin

If there is no in kernel image caller modules will suffer:

ERROR: "copy_from_user_overflow" [net/core/pktgen.ko] undefined!
ERROR: "copy_from_user_overflow" [net/can/can-raw.ko] undefined!
ERROR: "copy_from_user_overflow" [fs/cifs/cifs.ko] undefined!

Signed-off-by: Heiko Carstens <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

[WATCHDOG] i6300esb: fix unlock register with

Before you can read or write from the i6300esb memeory mapped registers,
you need to unlock these. This is done by writing the magic numbers 0x80
and 0x86 to the reload register. The size of the reload register is 32bit
though.

Also binary 11 is hex 0x03 and not 0x11.

Signed-off-by: Wim Van Sebroeck <[email protected]>

Merge branch 'topic/misc' into for-linus

Merge branch 'topic/asoc' into for-linus

Merge branch 'topic/hda' into for-linus

ALSA: hdmi - show debug message on changing audio infoframe

Also change printk level for the two others.

Signed-off-by: Wu Fengguang <[email protected]>
Signed-off-by: Takashi Iwai <[email protected]>

ALSA: hdmi - merge common code for intelhdmi and nvhdmi

Create patch_hdmi.c to hold common code from intelhdmi and nvhdmi.

For now the patch_hdmi.c file is simply included by patch_intelhdmi.c
and patch_nvhdmi.c, and does not represent a real codec.

There are no behavior changes to intelhdmi. However nvhdmi made several
changes when copying code out of intelhdmi, which are all reverted in
this patch. Wei Ni confirmed that the reverted code actually works fine.

Tested-by: Wei Ni <[email protected]>
Signed-off-by: Wu Fengguang <[email protected]>
Signed-off-by: Takashi Iwai <[email protected]>

msi-laptop: depends on RFKILL

msi-laptop uses rfkill*() interfaces so it should depend on RFKILL.

msi-laptop.c:(.text+0x1fcd1b): undefined reference to `rfkill_alloc'
msi-laptop.c:(.text+0x1fcd76): undefined reference to `rfkill_register'
msi-laptop.c:(.text+0x1fcdc8): undefined reference to `rfkill_destroy'
msi-laptop.c:(.text+0x1fcdd9): undefined reference to `rfkill_unregister'

This repairs "msi-laptop: Detect 3G device exists by standard ec command",
which is in some gregkh tree.

Signed-off-by: Randy Dunlap <[email protected]>
Cc: Lennart Poettering <[email protected]>
Cc: Lee, Chun-Yi <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

msi-laptop: Detect 3G device exists by standard ec command

Detect 3G device exists by standard ec command. Driver will not create the threeg sysfs
file and threeg rfkill interface if there have no internal 3G device in MSI notebook/netbook.

Signed-off-by: Lee, Chun-Yi <[email protected]>
Cc: Lennart Poettering <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

msi-laptop: Add resume method for set the SCM load again

Implement the resume method for set the load SCM flag after system reusme.
Without this patch, the wifi function key on SCM model will back to BIOS
control mode then confuse with the userland software control.
e.g. MSI N034

Signed-off-by: Lee, Chun-Yi <[email protected]>
Cc: Lennart Poettering <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

msi-laptop: Support some MSI 3G netbook that is need load SCM

Some MSI 3G netbook only have one fn key to control Wlan/Bluetooth/3G,
those netbook will load the SCM (windows app) to disable the original
Wlan/Bluetooth control by BIOS when user press fn key, then control
Wlan/Bluetooth/3G by SCM (software control by OS). Without SCM, user
cann't on/off 3G module on those 3G netbook.
On Linux, msi-laptop driver will do the same thing to disable the
original BIOS control, then might need use HAL or other userland
application to do the software control that simulate with SCM.
e.g. MSI N034 netbook

Signed-off-by: Lee, Chun-Yi <[email protected]>
Cc: Lennart Poettering <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

msi-laptop: Add threeg sysfs file for support query 3G state by standard 66/62 ec command

Add threeg sysfs file for support query 3G state by standard 66/62 ec
command, the MSI standard ec interface supported this feature.

Signed-off-by: Lee, Chun-Yi <[email protected]>
Cc: Lennart Poettering <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

msi-laptop: Support standard ec 66/62 command on MSI notebook and nebook

Suppport standard ec 66/62 command on MSI notebook and nebook. MSI
netbook and notebook already support 66/62 command, so, add new
get_state function, and put the old model to non-standard model, but
driver still support those old model.

Signed-off-by: Lee, Chun-Yi <[email protected]>
Cc: Lennart Poettering <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver core: create lock/unlock functions for struct device

In the future, we are going to be changing the lock type for struct
device (once we get the lockdep infrastructure properly worked out)  To
make that changeover easier, and to possibly burry the lock in a
different part of struct device, let's create some functions to lock and
unlock a device so that no out-of-core code needs to be changed in the
future.

This patch creates the device_lock/unlock/trylock() functions, and
converts all in-tree users to them.

Cc: Thomas Gleixner <[email protected]>
Cc: Jean Delvare <[email protected]>
Cc: Dave Young <[email protected]>
Cc: Ming Lei <[email protected]>
Cc: Jiri Kosina <[email protected]>
Cc: Phil Carmody <[email protected]>
Cc: Arjan van de Ven <[email protected]>
Cc: Cornelia Huck <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>
Cc: Pavel Machek <[email protected]>
Cc: Len Brown <[email protected]>
Cc: Magnus Damm <[email protected]>
Cc: Alan Stern <[email protected]>
Cc: Randy Dunlap <[email protected]>
Cc: Stefan Richter <[email protected]>
Cc: David Brownell <[email protected]>
Cc: Vegard Nossum <[email protected]>
Cc: Jesse Barnes <[email protected]>
Cc: Alex Chiang <[email protected]>
Cc: Kenji Kaneshige <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Andrew Patterson <[email protected]>
Cc: Yu Zhao <[email protected]>
Cc: Dominik Brodowski <[email protected]>
Cc: Samuel Ortiz <[email protected]>
Cc: Wolfram Sang <[email protected]>
Cc: CHENG Renquan <[email protected]>
Cc: Oliver Neukum <[email protected]>
Cc: Frans Pop <[email protected]>
Cc: David Vrabel <[email protected]>
Cc: Kay Sievers <[email protected]>
Cc: Sarah Sharp <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: fix for thinko with sysfs_bin_attr_init()

After merging the final tree, today's linux-next build (powerpc
allyesconfig) failed like this:

drivers/pci/pci-sysfs.c: In function 'pci_create_legacy_files':
drivers/pci/pci-sysfs.c:645: error: lvalue required as unary '&' operand
drivers/pci/pci-sysfs.c:658: error: lvalue required as unary '&' operand

Caused by commit "sysfs: Use sysfs_attr_init and sysfs_bin_attr_init on
dynamic attributes" interacting with commit "sysfs: Use one lockdep
class per sysfs attribute") both from the driver-core tree.

Signed-off-by: Stephen Rothwell <[email protected]>
Cc: "Eric W. Biederman" <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Kill unused sysfs_sb variable.

Now that there are no more users we can remove
the sysfs_sb variable.

Acked-by: Tejun Heo <[email protected]>
Acked-by: Serge Hallyn <[email protected]>
Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Pass super_block to sysfs_get_inode

Currently sysfs_get_inode magically returns an inode on
sysfs_sb.  Make the super_block parameter explicit and
the code becomes clearer.

Acked-by: Tejun Heo <[email protected]>
Acked-by: Serge Hallyn <[email protected]>
Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

driver core: Use sysfs_rename_link in device_rename

Don't open code the renaming of symlinks in sysfs
instead use the new helper function sysfs_rename_link

Acked-by: Tejun Heo <[email protected]>
Acked-by: Serge Hallyn <[email protected]>
Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Implement sysfs_rename_link

Because of rename ordering problems we occassionally give false
warnings about invalid sysfs operations.  So using sysfs_rename
create a sysfs_rename_link function that doesn't need strange
workarounds.

Cc: Benjamin Thery <[email protected]>
Cc: Daniel Lezcano <[email protected]>
Acked-by: Serge Hallyn <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Pack sysfs_dirent more tightly.

Placing the 16bit s_mode between a pointer and a long doesn't pack well
especailly on 64bit where we wast 48 bits.  So move s_mode and
declare it as a unsigned short.  This is the sysfs backing store
after all we don't need fields extra large just in case someday
we want userspace to be able to use a larger value.

Acked-by: Tejun Heo <[email protected]>
Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Serialize updates to the vfs inode

The vfs depends upon filesystem methods to update the
vfs inode.   Sysfs adds to the normal number of places
where the vfs inode is updated by also updatng the
vfs inode in sysfs_refresh_inode.

Typically the inode mutex is used to serialize updates
to the vfs inode, but grabbing the inode mutex in
sysfs_permission and sysfs_getattr causes deadlocks,
because sometimes the vfs calls those operations with
the inode mutex held.  Therefore sysfs  can not use the
inode mutex to serial updates to the vfs inode.

The sysfs_mutex is acquired in all of the routines
where sysfs updates the vfs inode, and with a small
change we can consistently protext sysfs vfs inode
updates with the sysfs_mutex. To protect the sysfs
vfs inode updates with the sysfs_mutex simply requires
extending the scope of sysfs_mutex in sysfs_setattr
over inode_setattr, and over inode_change_ok (so we
have an unchanging inode when we perform the check).

Acked-by: Serge Hallyn <[email protected]>
Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: windfarm: init sysfs attributes

This is required for lockdep.

Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Use sysfs_attr_init and sysfs_bin_attr_init on module dynamic attributes

A little more whack-a-mole annotating the dynamic sysfs attributes.  I
had everything built into my earlier test kernel, and so I missed
these.

Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Document sysfs_attr_init and sysfs_bin_attr_init

I have added a new requirement to the external sysfs interface
that dynamically allocated sysfs attributes must call sysfs_attr_init
if lockdep is enabled.  For the time being callying sysfs_attr_init
is only mandatory if lockdep is enabled, so we can live with a few
unconverted instances until we find them all.  As this is part of
the public interface of sysfs it is a good idea to document these
pseudo functions so someone inspeciting the code can find out
what has happened.

Signed-off-by: Eric W. Biederman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Use sysfs_attr_init and sysfs_bin_attr_init on dynamic attributes

These are the non-static sysfs attributes that exist on
my test machine.  Fix them to use sysfs_attr_init or
sysfs_bin_attr_init as appropriate.   It simply requires
making a sysfs attribute present to see this.  So this
is a little bit tedious but otherwise not too bad.

Signed-off-by: Eric W. Biederman <[email protected]>
Acked-by: WANG Cong <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Use one lockdep class per sysfs attribute.

Acknowledge that the logical sysfs rwsem has one instance per
sysfs attribute with different locking depencencies for different
attributes.

There is a sysfs idiom where writing to one sysfs file causes the
addition or removal of other sysfs files.   Lumping all of the
sysfs attributes together in one lock class causes lockdep to
generate lots of false positives.

This introduces the requirement that non-static sysfs attributes
need to be initialized with sysfs_attr_init or sysfs_bin_attr_init.
Strictly speaking this requirement only exists when lockdep is
enabled, and when lockdep is enabled we get a bit fat warning
if this requirement is not met.

Signed-off-by: Eric W. Biederman <[email protected]>
Acked-by: WANG Cong <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Only take active references on attributes.

If we exclude directories and symlinks from the set of sysfs
dirents where we need active references we are left with
sysfs attributes (binary or not).

- Tweak sysfs_deactivate to only do something on attributes
- Move lockdep initialization into sysfs_file_add_mode to
  limit it to just attributes.

Signed-off-by: Eric W. Biederman <[email protected]>
Acked-by: WANG Cong <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Remove sysfs_get/put_active_two

It turns out that holding an active reference on a directory is
pointless.  The purpose of the active references are to allows us to
block when removing sysfs entries that have custom methods so we don't
remove modules while running modular code and to keep those custom
methods from accessing data structures after the files have been
removed.  Further sysfs_remove_dir remove all elements in the
directory before removing the directory itself, so there is no chance
we will remove a directory with active children.

Signed-off-by: Eric W. Biederman <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver core: Fix first line of kernel-doc for a few functions

The function name must be followed by a space, hypen, space, and a
short description.

Signed-off-by: Ben Hutchings <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

UIO: Remove SMX Cryptengine driver

Ben Nizette, the author of this driver, told me in a private mail that this
project has been cancelled. He suggested to remove the driver for now, and
will come back with a new version should the hardware really exist.
This patch completely removes the driver.

Signed-off-by: Hans J. Koch <[email protected]>
Acked-by: Ben Nizette <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

UIO: minor Kconfig fixes

Two trivial fixes for the Userspace IO Kconfig file:
1) uio_sercos3 is a PCI driver, so let it depend on PCI.
2) "default n" under UIO_PCI_GENERIC is luxury since it is already the default.

Acked-by: John Ogness <[email protected]>
Signed-off-by: Hans J. Koch <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

UIO: Add a driver for Hilscher netX-based fieldbus cards

This patch adds a Userspace IO driver for netX-based fieldbus cards by
Hilscher (see http://www.hilscher.com). ATM, cifX and comX cards are
supported. The userspace part for this driver is provided by Hilscher
and should come with the card.
The driver is in use for several months now and has been tested by
people at Hilscher and Linutronix.

Signed-off-by: Hans J. Koch <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

auxdisplay: move cfag12864bfb's probe function to .devinit.text

A pointer to cfag12864bfb_probe is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Acked-by: Miguel Ojeda <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Avuton Olrich <[email protected]>
Cc: Antonino Daplas <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

media: move omap24xxcam's probe function to .devinit.text

A pointer to omap24xxcam_probe is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Acked-by: Sakari Ailus <[email protected]>
Acked-by: Trilok Soni <[email protected]>
Cc: Hans Verkuil <[email protected]>
Cc: Mauro Carvalho Chehab <[email protected]>
Cc: Andrew Morton <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

w1: move omap_hdq's probe function to .devinit.text

A pointer to omap_hdq_probe is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Cc: Stanley.Miao <[email protected]>
Acked-by: Evgeniy Polyakov <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Madhusudhan Chikkature <[email protected]>
Cc: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

i2c: move i2c_omap's probe function to .devinit.text

A pointer to omap_i2c_probe is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Cc: Kalle Jokiniemi <[email protected]>
Cc: Tony Lindgren <[email protected]>
Cc: Paul Walmsley <[email protected]>
Cc: Richard Woodruff <[email protected]>
Cc: chandra shekhar <[email protected]>
Cc: Jason P Marini <[email protected]>
Cc: Syed Mohammed Khasim <[email protected]>
Cc: Jarkko Nikula <[email protected]>
Cc: Juha Yrjola <[email protected]>
Cc: Andrew Morton <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

x86: move hp-wmi's probe function to .devinit.text

A pointer to hp_wmi_bios_setup is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Cc: Frans Pop <[email protected]>
Cc: Larry Finger <[email protected]>
Cc: Len Brown <[email protected]>
Cc: Helge Deller <[email protected]>
Cc: Andrew Morton <[email protected]>
Acked-by: Matthew Garrett <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

platform-drivers: move probe to .devinit.text in drivers/video

A pointer to a probe callback is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Cc: Adrian Bunk <[email protected]>
Cc: Alberto Mardegan <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Andriy Skulysh <[email protected]>
Cc: Antonino Daplas <[email protected]>
Cc: Anton Vorontsov <[email protected]>
Cc: Ben Dooks <[email protected]>
Cc: Chandramouli Narayanan <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Frans Pop <[email protected]>
Cc: Geert Uytterhoeven <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Helge Deller <[email protected]>
Cc: Huang Ying <[email protected]>
Cc: Ian Molton <[email protected]>
Cc: Joshua Kinard <[email protected]>
Cc: Kaj-Michael Lang <[email protected]>
Cc: Krzysztof Helt <[email protected]>
Cc: [email protected]
Cc: Maciej W. Rozycki <[email protected]>
Cc: Magnus Damm <[email protected]>
Cc: Martin Michlmayr <[email protected]>
Cc: Matthias Kaehlcke <[email protected]>
Cc: Paul Mundt <[email protected]>
Cc: Pavel Machek <[email protected]>
Cc: Philipp Zabel <[email protected]>
Cc: Richard Purdie <[email protected]>
Cc: Roel Kluin <[email protected]>
Cc: Roland Stigge <[email protected]>
Cc: Russell King <[email protected]>
Cc: Thomas Bogendoerfer <[email protected]>
Cc: Vincent Sanders <[email protected]>
Cc: Yoichi Yuasa <[email protected]>
Acked-by: Ralf Baechle <[email protected]>
Acked-by: Arnaud Patard <[email protected]>
Acked-by: James Simmons <[email protected]>
Acked-by: Peter Jones <[email protected]>
Acked-by: Jaya Kumar <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

platform-drivers: move probe to .devinit.text in drivers/scsi

A pointer to a probe callback is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: David Brownell <[email protected]>
Cc: Dmitri Vorobiev <[email protected]>
Cc: Henrik Kretzschmar <[email protected]>
Cc: James Bottomley <[email protected]>
Cc: Kay Sievers <[email protected]>
Cc: peter fuerst <[email protected]>
Cc: Thomas Bogendoerfer <[email protected]>
Acked-by: Ralf Baechle <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

platform-drivers: move probe to .devinit.text in arch/arm

A pointer to a probe callback is passed to the core via
platform_driver_register and so the function must not disappear when the
.init sections are discarded.  Otherwise (if also having HOTPLUG=y)
unbinding and binding a device to the driver via sysfs will result in an
oops as does a device being registered late.

An alternative to this patch is using platform_driver_probe instead of
platform_driver_register plus removing the pointer to the probe function
from the struct platform_driver.

Signed-off-by: Uwe Kleine-König <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Ben Dooks <[email protected]>
Cc: Dmitry Baryshkov <[email protected]>
Cc: Eric Miao <[email protected]>
Cc: Liam Girdwood <[email protected]>
Cc: Paul Sokolovsky <[email protected]>
Cc: Richard Purdie <[email protected]>
Cc: Russell King <[email protected]>
Acked-by: Arnaud Patard <[email protected]>
Acked-by: Dmitry Eremin-Solenikov <[email protected]>
Acked-by: Kristoffer Ericson <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver core: make struct platform_driver.id_table const

This fixes a warning on several pxa based machines:

arch/arm/mach-pxa/ssp.c:475: warning: initialization discards qualifiers from pointer target type

Signed-off-by: Uwe Kleine-König <[email protected]>
Acked-by: Vikram Dhillon <[email protected]>
Acked-by: Eric Miao <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver core: Constify struct sysfs_ops in struct kobj_type

Constify struct sysfs_ops.

This is part of the ops structure constification
effort started by Arjan van de Ven et al.

Benefits of this constification:

 * prevents modification of data that is shared
   (referenced) by many other structure instances
   at runtime

 * detects/prevents accidental (but not intentional)
   modification attempts on archs that enforce
   read-only kernel data at runtime

 * potentially better optimized code as the compiler
   can assume that the const data cannot be changed

 * the compiler/linker move const data into .rodata
   and therefore exclude them from false sharing

Signed-off-by: Emese Revfy <[email protected]>
Acked-by: David Teigland <[email protected]>
Acked-by: Matt Domsch <[email protected]>
Acked-by: Maciej Sosnowski <[email protected]>
Acked-by: Hans J. Koch <[email protected]>
Acked-by: Pekka Enberg <[email protected]>
Acked-by: Jens Axboe <[email protected]>
Acked-by: Stephen Hemminger <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysdev: fix up the probe/release attributes

These should be sysdev attributes, not class attributes.  This patch
should resolve the problem.

Thanks to Stephen Rothwell for pointing out the problem.

Reported-by: Stephen Rothwell <[email protected]>
Cc: Andi Kleen <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

HOWTO: Updates on subsystem trees, patchwork, -next (vs. -mm)

One of the roles which -mm fulfilled some time ago (to offer an
integration testing ground) has been taken over by -next.  This is still
news to Documentation/HOWTO, so mention it there.

Also add a word on how patchwork is used to track patches as they make
their way into subsystem trees.  Remove some arbitrary links to
subsystem repositories; they can all be found in the MAINTAINERS
database.

Acked-by: Randy Dunlap <[email protected]>
Signed-off-by: Stefan Richter <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

kobject: Constify struct kset_uevent_ops

Constify struct kset_uevent_ops.

This is part of the ops structure constification
effort started by Arjan van de Ven et al.

Benefits of this constification:

 * prevents modification of data that is shared
   (referenced) by many other structure instances
   at runtime

 * detects/prevents accidental (but not intentional)
   modification attempts on archs that enforce
   read-only kernel data at runtime

 * potentially better optimized code as the compiler
   can assume that the const data cannot be changed

 * the compiler/linker move const data into .rodata
   and therefore exclude them from false sharing

Signed-off-by: Emese Revfy <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

driver-core: firmware_class: remove base.h header inclusion

base.h is used by base drivers for sharing internal structures.
Turns out firmware_class does not depend on it at all so remove it.

Cc: Johannes Berg <[email protected]>
Signed-off-by: Luis R. Rodriguez <[email protected]>
Acked-by: Marcel Holtmann <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysfs: Cache the last sysfs_dirent to improve readdir scalability v2

When sysfs_readdir stops short we now cache the next
sysfs_dirent to return to user space in filp->private_data.
There is no impact on the rest of sysfs by doing this and
in the common case it allows us to pick up exactly where
we left off with no seeking.

Additionally I drop and regrab the sysfs_mutex around
filldir to avoid a page fault abritrarily increasing the
hold time on the sysfs_mutex.

v2: Returned to using INT_MAX as the EOF condition.
    seekdir is ambiguous unless all directory entries have
    a unique f_pos value.

Fixes http://bugzilla.kernel.org/show_bug.cgi?id=14949

Signed-off-by: Eric W. Biederman <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: stable <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver-Core: require valid action string in uevent trigger

No longer fall back to "add" and warn, but always require a valid
action-string written to the "uevent" file.

Signed-off-by: Kay Sievers <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver-Core: disable /sbin/hotplug by default

No recent mainstream system uses the /sbin/hotplug fork-bomb any more.
Disable it by default to reflect how it is used these days.

Signed-off-by: Kay Sievers <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver-Core: devtmpfs - remove EXPERIMENTAL and flush out the description

All major distros enable devtmpfs on recent systems, so remove
the EXPERIMENTAL flag, and make the description a bit more instructive.

Signed-off-by: Kay Sievers <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Driver-Core: devtmpfs - reset inode permissions before unlinking

Before unlinking the inode, reset the current permissions of possible
references like hardlinks, so granted permissions can not be retained
across the device lifetime by creating hardlinks, in the unusual case
that there is a user-writable directory on the same filesystem.

Signed-off-by: Kay Sievers <[email protected]>
Cc: stable <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

driver core: Convert some drivers to CLASS_ATTR_STRING

Convert some drivers who export a single string as class attribute
to the new class_attr_string functions. This removes redundant
code all over.

Signed-off-by: Andi Kleen <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

driver core: Add class_attr_string for simple read-only string

Several drivers just export a static string as class attributes.

Use the new extensible attribute support to define a simple
CLASS_ATTR_STRING() macro for this.

This will allow to remove code from drivers in followon patches.

Signed-off-by: Andi Kleen <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

driver-core: Add attribute argument to class_attribute show/store

Passing the attribute to the low level IO functions allows all kinds
of cleanups, by sharing low level IO code without requiring
an own function for every piece of data.

Also drivers can extend the attributes with own data fields
and use that in the low level function.

This makes the class attributes the same as sysdev_class attributes
and plain attributes.

This will allow further cleanups in drivers.

Full tree sweep converting all users.

Signed-off-by: Andi Kleen <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

sysdev: fix missing s390 conversion

In linux-next "sysdev: Pass attribute in sysdev_class attributes show/store"
forgot to convert one place in s390 code. Here is the missing part.

Cc: Andi Kleen <[email protected]>
Cc: Martin Schwidefsky <[email protected]>
Signed-off-by: Heiko Carstens <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>