linux-2.6
16 years agoHOWTO: Change man-page maintainer address for Japanese HOWTO
Tsugikazu Shibata [Fri, 30 Nov 2007 09:59:18 +0000 (18:59 +0900)] 
HOWTO: Change man-page maintainer address for Japanese HOWTO

Below is a patch to change email address of man-page maintainer for
Japanese HOWTO document (Documentation/ja_JP/HOWTO).
This is for sync to Documentation/HOWTO that Michael Kerrisk mentioned
to me.

From: Tsugikazu Shibata <tshibata@ab.jp.nec.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agotipar: remove obsolete module
Romain Liévin [Sat, 1 Dec 2007 07:51:58 +0000 (08:51 +0100)] 
tipar: remove obsolete module

tipar: remove obsolete module

The tipar character driver was used to implement bit-banging access
to Texas Instruments parallel link cable. A user-land method now
exists thru PPDEV & PARPORT.

Signed-off-by: Romain Liévin <roms@lpg.ticalc.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agokobject: fix the documentation of how kobject_set_name works
Greg Kroah-Hartman [Tue, 4 Dec 2007 06:45:47 +0000 (14:45 +0800)] 
kobject: fix the documentation of how kobject_set_name works

Thanks to Dave Young <hidave.darkstar@gmail.com> for pointing out that I
forgot to update the comment when I rewrote kobject_set_name.

Cc: Dave Young <hidave.darkstar@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb
Linus Torvalds [Mon, 17 Dec 2007 17:56:29 +0000 (09:56 -0800)] 
Merge git://git./linux/kernel/git/mchehab/v4l-dvb

* git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb:
  V4L/DVB (6820): s5h1409: QAM SNR related fixes
  V4L/DVB (6819): i2c: fix drivers/media/video/bt866.c
  V4L/DVB (6814): Makefile: always enter video/
  V4L/DVB (6798): saa7134: enable LNA in analog mode for Hauppauge WinTV HVR-1110
  V4L/DVB (6803): buf-core.c locking fixes

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog
Linus Torvalds [Mon, 17 Dec 2007 17:53:51 +0000 (09:53 -0800)] 
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] add Nano 7240 driver
  [WATCHDOG] ipmi: add the standard watchdog timeout ioctls
  [WATCHDOG] IT8212F watchdog driver
  [WATCHDOG] Sbus: cpwatchdog, remove SPIN_LOCK_UNLOCKED
  [WATCHDOG] bfin_wdt, remove SPIN_LOCK_UNLOCKED
  [WATCHDOG] Stop looking for device as soon as one is found
  [WATCHDOG] at32ap700x_wdt: add support for boot status and add fix for silicon errata

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sam/fix-kbuild
Linus Torvalds [Mon, 17 Dec 2007 17:53:07 +0000 (09:53 -0800)] 
Merge git://git./linux/kernel/git/sam/fix-kbuild

* git://git.kernel.org/pub/scm/linux/kernel/git/sam/fix-kbuild:
  kbuild: re-enable Makefile generation in a new O=... directory

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Mon, 17 Dec 2007 17:52:11 +0000 (09:52 -0800)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/roland/infiniband

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  IB/ehca: Fix lock flag variable location, bump version number
  IB/ehca: Serialize HCA-related hCalls if necessary
  IB/ehca: Return correct number of SGEs for SRQ

16 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Mon, 17 Dec 2007 17:49:24 +0000 (09:49 -0800)] 
Merge master.kernel.org:/home/rmk/linux-2.6-arm

* master.kernel.org:/home/rmk/linux-2.6-arm:
  [ARM] 4694/1: IXP4xx: Update clockevent support for shutdown and resume
  [ARM] 4710/1: Fix coprocessor 14 usage for debug messages via ICEDCC
  [ARM] 4690/1: PXA: fix CKEN corruption in PXA27x AC97 cold reset code
  [ARM] 4667/1: CM-X270 fixes

16 years ago[ARM] 4694/1: IXP4xx: Update clockevent support for shutdown and resume
Kevin Hilman [Tue, 11 Dec 2007 23:32:58 +0000 (00:32 +0100)] 
[ARM] 4694/1: IXP4xx: Update clockevent support for shutdown and resume

Add proper support for CLOCK_EVT_MODE_RESUME and in the process fix
CLOCK_EVT_MODE_SHUTDOWN so that only the enable bits are toggled for both.

Signed-off-by: Kevin Hilman <khilman@mvista.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
16 years ago[ARM] 4710/1: Fix coprocessor 14 usage for debug messages via ICEDCC
Uwe Kleine-König [Thu, 13 Dec 2007 08:31:34 +0000 (09:31 +0100)] 
[ARM] 4710/1: Fix coprocessor 14 usage for debug messages via ICEDCC

According to ARM7TDMI Technical Reference Manual (ARM DDI 0210C) writing
to the DCC data write register coproc dest registers are 1 and 0, not 0
and 1.

ARM920T TRM (ARM DDI 0151C) agrees on that.

Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Mon, 17 Dec 2007 16:43:49 +0000 (08:43 -0800)] 
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
  [IRDA]: irda parameters warning fixes.
  [IRDA]: stir4200 fixes.
  [IRDA]: irlmp_unregister_link() needs to free lsaps.
  [IRDA]: mcs7780 needs to free allocated rx buffer.
  [IRDA]: Race between open and disconnect in irda-usb.
  [SCTP]: Flush fragment queue when exiting partial delivery.
  [AX25]: Locking dependencies fix in ax25_disconnect().
  [IPV4]: Make tcp_input_metrics() get minimum RTO via tcp_rto_min()
  [IPV6]: Fix the return value of ipv6_getsockopt
  [BRIDGE]: Assign random address.
  [IPV4]: Updates to nfsroot documentation
  [ATM]: Fix compiler warning noise with FORE200E driver
  [NETFILTER]: bridge: fix missing link layer headers on outgoing routed packets
  [SYNCPPP]: Endianness and 64bit fixes.
  [TIPC]: Fix semaphore handling.
  [NETFILTER]: xt_hashlimit should use time_after_eq()
  [XFRM]: Display the audited SPI value in host byte order.
  [NETFILTER]: ip_tables: fix compat copy race
  [NETFILTER]: ctnetlink: set expected bit for related conntracks

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6
Linus Torvalds [Mon, 17 Dec 2007 16:43:02 +0000 (08:43 -0800)] 
Merge git://git./linux/kernel/git/davem/sparc-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  [SPARC32]: Silence sparc32 warnings on missing syscalls.
  [SPARC64]: Fix two kernel linear mapping setup bugs.

16 years agoMerge branch 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6
Linus Torvalds [Mon, 17 Dec 2007 16:42:23 +0000 (08:42 -0800)] 
Merge branch 'for-linus' of git://git390.osdl.marist.edu/linux-2.6

* 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6:
  [S390] pud_present/pmd_present bug.

16 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc
Linus Torvalds [Mon, 17 Dec 2007 16:41:56 +0000 (08:41 -0800)] 
Merge branch 'merge' of git://git./linux/kernel/git/paulus/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc:
  [POWERPC] Make PS3_SYS_MANAGER default y, not m
  [POWERPC] Fix rounding bug in emulation for double float operating
  [POWERPC] iSeries: don't printk with HV spinlock held
  [POWERPC] 82xx: mpc8272ads, pq2fads: Update defconfig with CONFIG_FS_ENET_MDIO_FCC
  [POWRPC] CPM2: Eliminate section mismatch warning in cpm2_reset().
  [POWERPC] Kill non-existent symbols from ksyms and commproc.h
  [POWERPC] Fix typo #ifdef -> #ifndef

16 years agoMerge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
Linus Torvalds [Mon, 17 Dec 2007 16:41:25 +0000 (08:41 -0800)] 
Merge branch 'upstream' of git://ftp.linux-mips.org/upstream-linus

* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
  [MIPS] Ensure that ST0_FR is never set on a 32 bit kernel
  [MIPS] time: Delete weak definition of plat_time_init() due to gcc bug.
  [MIPS] PCI: Make pcibios_fixup_device_resources ignore legacy resources.
  [MIPS] Atlas, Malta: Don't free firmware memory on free_initmem.
  [MIPS] Alchemy: fix off by two error in __fixup_bigphys_addr()
  [MIPS] Alchemy: fix PCI resource conflict
  [MIPS] time: Set up Cobalt's mips_hpt_frequency

16 years ago[S390] pud_present/pmd_present bug.
Martin Schwidefsky [Mon, 17 Dec 2007 15:25:48 +0000 (16:25 +0100)] 
[S390] pud_present/pmd_present bug.

Git commit 3610cce87af0693603db171d5b6f6735f5e3dc5b (yeah my own :-/)
introduced a bug in regard to pud/pmd table entries.
If the address of the page table refered to by a pud/pmd value happens
to have zeroes in the lower 32 bits, pud_present and pmd_present return
false. The obvious effect is that this triggers the BUG_ON in exit_mmap
because some ptes will not get released on process end.  Worse is that
the next fault for memory covered by that pud/pmd will allocate another
pmd/pte table and populate the pud/pmd entry. The old page table
entries hanging below this entry are lost!

The fix is simple, properly check against 0. The check is added for
pud_none/pmd_none as well even if these two functions work because
the invalid bit is in the lower 32 bits.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
16 years ago[IRDA]: irda parameters warning fixes.
Richard Knutsson [Sun, 16 Dec 2007 22:10:33 +0000 (14:10 -0800)] 
[IRDA]: irda parameters warning fixes.

This patch fixes:
  CHECK   /home/kernel/src/net/irda/parameters.c
/home/kernel/src/net/irda/parameters.c:466:2: warning: Using plain integer as NULL pointer
/home/kernel/src/net/irda/parameters.c:520:2: warning: Using plain integer as NULL pointer
/home/kernel/src/net/irda/parameters.c:573:2: warning: Using plain integer as NULL pointer

Signed-off-by: Richard Knutsson <ricknu-0@student.ltu.se>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IRDA]: stir4200 fixes.
Olaf Hartmann [Sun, 16 Dec 2007 22:09:44 +0000 (14:09 -0800)] 
[IRDA]: stir4200 fixes.

From: Olaf Hartmann <olaf.hartmann@s1998.tu-chemnitz.de>

The attached patch observes the stir4200 fifo size and will clear the
fifo, if the size is increasing, while it should be transmitting bytes

Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IRDA]: irlmp_unregister_link() needs to free lsaps.
Hinko Kocevar [Sun, 16 Dec 2007 22:08:58 +0000 (14:08 -0800)] 
[IRDA]: irlmp_unregister_link() needs to free lsaps.

While testing the mcs7780 based IrDA USB dongle I've stumbled upon
memory leak in irlmp_unregister_link(). Hashbin for lsaps is created in
irlmp_register_link and should probably be freed in irlmp_unregister_link().

Signed-off-by: Hinko Kocevar <hinko.kocevar@cetrtapot.si>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IRDA]: mcs7780 needs to free allocated rx buffer.
Hinko Kocevar [Sun, 16 Dec 2007 22:08:22 +0000 (14:08 -0800)] 
[IRDA]: mcs7780 needs to free allocated rx buffer.

While testing the mcs7780 based IrDA USB dongle I've stumbled upon
memory leak in mcs_net_close(). Patch below fixes it.

Signed-off-by: Hinko Kocevar <hinko.kocevar@cetrtapot.si>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IRDA]: Race between open and disconnect in irda-usb.
Oliver Neukum [Sun, 16 Dec 2007 22:07:36 +0000 (14:07 -0800)] 
[IRDA]: Race between open and disconnect in irda-usb.

It seems to me that irda_usb_net_open() must set self->netopen
under spinlock or disconnect() may fail to kill all URBs, if it is called
while an interface is opened.

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[SCTP]: Flush fragment queue when exiting partial delivery.
Vlad Yasevich [Sun, 16 Dec 2007 22:05:45 +0000 (14:05 -0800)] 
[SCTP]: Flush fragment queue when exiting partial delivery.

At the end of partial delivery, we may have complete messages
sitting on the fragment queue.  These messages are stuck there
until a new fragment arrives.  This can comletely stall a
given association.  When clearing partial delivery state, flush
any complete messages from the fragment queue and send them on
their way up.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[AX25]: Locking dependencies fix in ax25_disconnect().
Jarek Poplawski [Sun, 16 Dec 2007 22:02:07 +0000 (14:02 -0800)] 
[AX25]: Locking dependencies fix in ax25_disconnect().

Bernard Pidoux reported these lockdep warnings:

[ INFO: possible irq lock inversion dependency detected ]
2.6.23.1 #1
---------------------------------------------------------
fpac/4933 just changed the state of lock:
 (slock-AF_AX25){--..}, at: [<d8be3312>] ax25_disconnect+0x46/0xaf
 [ax25]
 but this lock was taken by another, soft-irq-safe lock in the past:
  (ax25_list_lock){-+..}

  and interrupts could create inverse lock ordering between them.
  [...]

[ INFO: inconsistent lock state ]
2.6.23.1 #1
---------------------------------
inconsistent {in-softirq-W} -> {softirq-on-W} usage.
ax25_call/4005 [HC0[0]:SC0[0]:HE1:SE1] takes:
 (slock-AF_AX25){-+..}, at: [<d8b79312>] ax25_disconnect+0x46/0xaf [ax25]
 [...]

This means slock-AF_AX25 could be taken both from softirq and process
context with softirqs enabled, so it's endangered itself, but also makes
ax25_list_lock vulnerable. It was not 100% verified if the real lockup
can happen, but this fix isn't very costly and looks safe anyway.
(It was tested by Bernard with 2.6.23.9 and 2.6.24-rc5 kernels.)

Reported_by: Bernard Pidoux <pidoux@ccr.jussieu.fr>
Tested_by: Bernard Pidoux <pidoux@ccr.jussieu.fr>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IPV4]: Make tcp_input_metrics() get minimum RTO via tcp_rto_min()
Satoru SATOH [Sun, 16 Dec 2007 22:00:19 +0000 (14:00 -0800)] 
[IPV4]: Make tcp_input_metrics() get minimum RTO via tcp_rto_min()

tcp_input_metrics() refers to the built-time constant TCP_RTO_MIN
regardless of configured minimum RTO with iproute2.

Signed-off-by: Satoru SATOH <satoru.satoh@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IPV6]: Fix the return value of ipv6_getsockopt
Wei Yongjun [Sun, 16 Dec 2007 21:39:57 +0000 (13:39 -0800)] 
[IPV6]: Fix the return value of ipv6_getsockopt

If CONFIG_NETFILTER if not selected when compile the kernel source code,
ipv6_getsockopt will returen an EINVAL error if optname is not supported by
the kernel. But if CONFIG_NETFILTER is selected, ENOPROTOOPT error will
be return.

This patch fix to always return ENOPROTOOPT error if optname argument of
ipv6_getsockopt is not supported by the kernel.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[BRIDGE]: Assign random address.
Stephen Hemminger [Sun, 16 Dec 2007 21:35:51 +0000 (13:35 -0800)] 
[BRIDGE]: Assign random address.

Assigning a valid random address to bridge device solves problems
when bridge device is brought up before adding real device to bridge.
When the first real device is added to the bridge, it's address
will overide the bridges random address.

Note: any device added to a bridge must already have a valid
ethernet address.
 br_add_if -> br_fdb_insert -> fdb_insert -> is_valid_ether_addr

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IPV4]: Updates to nfsroot documentation
Amos Waterland [Fri, 14 Dec 2007 19:30:22 +0000 (11:30 -0800)] 
[IPV4]: Updates to nfsroot documentation

The difference between ip=off and ip=::::::off has been a cause of much
confusion.  Document how each behaves, and do not contradict ourselves by
saying that "off" is the default when in fact "any" is the default and is
descibed as being so lower in the file.

Signed-off-by: Amos Waterland <apw@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[ATM]: Fix compiler warning noise with FORE200E driver
Tom "spot" Callaway [Thu, 13 Dec 2007 17:33:50 +0000 (09:33 -0800)] 
[ATM]: Fix compiler warning noise with FORE200E driver

gcc throws these warnings with:

CONFIG_ATM_FORE200E=m
# CONFIG_ATM_FORE200E_PCA is not set

drivers/atm/fore200e.c:2695: warning: 'fore200e_pca_detect' defined but
not used
drivers/atm/fore200e.c:2748: warning: 'fore200e_pca_remove_one' defined
but not used

By moving the #ifdef CONFIG_ATM_FORE200E_PCA around those two functions,
the compiler warnings are silenced.

Signed-off-by: Tom "spot" Callaway <tcallawa@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[NETFILTER]: bridge: fix missing link layer headers on outgoing routed packets
Patrick McHardy [Thu, 13 Dec 2007 17:32:04 +0000 (09:32 -0800)] 
[NETFILTER]: bridge: fix missing link layer headers on outgoing routed packets

As reported by Damien Thebault, the double POSTROUTING hook invocation
fix caused outgoing packets routed between two bridges to appear without
a link-layer header. The reason for this is that we're skipping the
br_nf_post_routing hook for routed packets now and don't save the
original link layer header, but nevertheless tries to restore it on
output, causing corruption.

The root cause for this is that skb->nf_bridge has no clearly defined
lifetime and is used to indicate all kind of things, but that is
quite complicated to fix. For now simply don't touch these packets
and handle them like packets from any other device.

Tested-by: Damien Thebault <damien.thebault@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[SYNCPPP]: Endianness and 64bit fixes.
Al Viro [Thu, 13 Dec 2007 02:50:18 +0000 (18:50 -0800)] 
[SYNCPPP]: Endianness and 64bit fixes.

* trivial annotations
* long != 32bit, use __be32
* wrong endianness in sending CISCO_ADDR_REPLY

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[TIPC]: Fix semaphore handling.
Andrew Morton [Wed, 12 Dec 2007 19:24:19 +0000 (11:24 -0800)] 
[TIPC]: Fix semaphore handling.

As noted by Kevin, tipc's release() does down_interruptible() and
ignores the return value.  So if signal_pending() we'll end up doing
up() on a non-downed semaphore.  Fix.

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[NETFILTER]: xt_hashlimit should use time_after_eq()
Eric Dumazet [Wed, 12 Dec 2007 19:11:28 +0000 (11:11 -0800)] 
[NETFILTER]: xt_hashlimit should use time_after_eq()

In order to avoid jiffies wraparound and its effect, special care must
be taken
when doing comparisons ...

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[XFRM]: Display the audited SPI value in host byte order.
Paul Moore [Wed, 12 Dec 2007 19:10:16 +0000 (11:10 -0800)] 
[XFRM]: Display the audited SPI value in host byte order.

Currently the IPsec protocol SPI values are written to the audit log in
network byte order which is different from almost all other values which
are recorded in host byte order.  This patch corrects this inconsistency
by writing the SPI values to the audit record in host byte order.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[NETFILTER]: ip_tables: fix compat copy race
Patrick McHardy [Wed, 12 Dec 2007 18:35:16 +0000 (10:35 -0800)] 
[NETFILTER]: ip_tables: fix compat copy race

When copying entries to user, the kernel makes two passes through the
data, first copying all the entries, then fixing up names and counters.
On the second pass it copies the kernel and match data from userspace
to the kernel again to find the corresponding structures, expecting
that kernel pointers contained in the data are still valid.

This is obviously broken, fix by avoiding the second pass completely
and fixing names and counters while dumping the ruleset, using the
kernel-internal data structures.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[NETFILTER]: ctnetlink: set expected bit for related conntracks
Pablo Neira Ayuso [Wed, 12 Dec 2007 18:34:29 +0000 (10:34 -0800)] 
[NETFILTER]: ctnetlink: set expected bit for related conntracks

This patch is a fix. It sets IPS_EXPECTED for related conntracks.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoixgb: make sure jumbos stay enabled after reset
Matheos Worku [Fri, 14 Dec 2007 19:48:29 +0000 (11:48 -0800)] 
ixgb: make sure jumbos stay enabled after reset

Currently a device reset (ethtool -r ethX) would cause the
adapter to fall back to regular MTU sizes.

Signed-off-by: Matheos Worku <matheos.worku@sun.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agoNet: ibm_newemac, remove SPIN_LOCK_UNLOCKED
Jiri Slaby [Fri, 14 Dec 2007 00:02:57 +0000 (16:02 -0800)] 
Net: ibm_newemac, remove SPIN_LOCK_UNLOCKED

SPIN_LOCK_UNLOCKED is deprecated, use DEFINE_SPINLOCK instead

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agonet: smc911x: shut up compiler warnings
Paul Mundt [Fri, 14 Dec 2007 00:02:59 +0000 (16:02 -0800)] 
net: smc911x: shut up compiler warnings

Trivial fix to shut up gcc.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agoucc_geth: minor whitespace fix
Andrew Morton [Fri, 14 Dec 2007 00:02:55 +0000 (16:02 -0800)] 
ucc_geth: minor whitespace fix

The zombie whitespace from outer space that will not die!

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Emil Medve <Emilian.Medve@Freescale.com>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Kumar Gala <galak@gate.crashing.org>
Cc: Li Yang <leoli@freescale.com>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agodrivers/net/s2io.c section fixes
Adrian Bunk [Tue, 11 Dec 2007 22:23:06 +0000 (23:23 +0100)] 
drivers/net/s2io.c section fixes

Code used by the non-__devinit s2io_open() mustn't be __devinit.

This patch fixes the following section mismatch with CONFIG_HOTPLUG=n:

<--  snip  -->

...
WARNING: vmlinux.o(.text+0x6f6e3e): Section mismatch: reference to .init.text.20:s2io_test_intr (between 's2io_open' and 's2io_ethtool_sset')
...

<--  snip  -->

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agodrivers/net/sis190.c section fix
Adrian Bunk [Tue, 11 Dec 2007 22:23:56 +0000 (23:23 +0100)] 
drivers/net/sis190.c section fix

This patch fixes the following section mismatch with CONFIG_HOTPLUG=n:

<--  snip  -->

...
WARNING: vmlinux.o(.init.text.20+0x4cb25): Section mismatch: reference to .exit.text:sis190_mii_remove (between 'sis190_init_one' and 'read_eeprom')
...

<--  snip  -->

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agohamachi endianness fixes
Al Viro [Mon, 10 Dec 2007 18:39:29 +0000 (18:39 +0000)] 
hamachi endianness fixes

badly broken on big-endian

* passing little-endian to pci_unmap_single() et.al.
* cpu_to_le32() before passing value to writel()
* worse, cpu_to_le64() and shifting/masking result before the same
*  hmp->tx_ring[i].status_n_length = cpu_to_le32(
DescEndRing |
(hmp->tx_ring[i].status_n_length & 0x0000FFFF));
  is obviously bogus on big-endian.  Not hard to untangle, fortunately...
* poisoning addresses in rx_ring is better done after we'd done
pci_unmap_single() on them, not before that.  [this one affects little-endian
as well, obviously, provided that pci_unmap_single() is not a no-op on target
in question]

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agoe100: free IRQ to remove warningwhenrebooting
Auke Kok [Thu, 13 Dec 2007 00:30:42 +0000 (16:30 -0800)] 
e100: free IRQ to remove warningwhenrebooting

Adapted from Ian Wienand <ianw@gelato.unsw.edu.au>

Explicitly free the IRQ before removing the device to remove a
warning "Destroying IRQ without calling free_irq"

Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Cc: Ian Wienand <ianw@gelato.unsw.edu.au>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agostarfire VLAN fix
Al Viro [Sun, 9 Dec 2007 17:00:54 +0000 (17:00 +0000)] 
starfire VLAN fix

Recognized VLAN ids are set via writew(), should go in host-endian.
That's a long-standing bug, BTW - see http://lkml.org/lkml/2004/2/27/180
for example.  What happens is that card gets VLAN id table populated by
byteswapped values on little-endian boxen (so 257 works as expected, 256
and 258 do not, etc.).  Bug is easily reproduced, patch fixes it.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agosundance fixes
Al Viro [Sun, 9 Dec 2007 16:50:47 +0000 (16:50 +0000)] 
sundance fixes

* all places where we assign ->addr get cpu_to_le32(pci_map_single(....)), so
we ought to convert back to host-endian before doing pci_unmap_single() et.al.
* poisoning addresses in netdev_close() should be done _after_ unmapping them,
not before it...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agosky2: RX lockup fix
Stephen Hemminger [Fri, 7 Dec 2007 23:22:15 +0000 (15:22 -0800)] 
sky2: RX lockup fix

I'm using a Marvell 88E8062 on a custom PPC64 blade and ran into RX
lockups while validating the sky2 driver.  The receive MAC FIFO would
become stuck during testing with high traffic.  One port of the 88E8062
would lockup, while the other port remained functional.  Re-inserting
the sky2 module would not fix the problem - only a power cycle would.

I looked over Marvell's most recent sk98lin driver and it looks like
they had a "workaround" for the Yukon XL that the sky2 doesn't have yet.
The sk98lin driver disables the RX MAC FIFO flush feature for all
revisions of the Yukon XL.

According to skgeinit.c of the sk98lin driver, "Flushing must be enabled
(needed for ASF see dev. #4.29), but the flushing mask should be
disabled (see dev. #4.115)".  Nice. I implemented this same change in
the sky2 driver and verified that the RX lockup I was seeing was
resolved.

Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agoPull bugzilla-9362 into release branch
Len Brown [Fri, 14 Dec 2007 20:14:52 +0000 (15:14 -0500)] 
Pull bugzilla-9362 into release branch

16 years agoACPI: SBS: Return rate in mW if capacity in mWh
Alexey Starikovskiy [Sat, 8 Dec 2007 10:02:46 +0000 (13:02 +0300)] 
ACPI: SBS: Return rate in mW if capacity in mWh

klaptopd assumes rate to be in same units as capacity.

http://bugzilla.kernel.org/show_bug.cgi?id=9362

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
16 years agoACPI: SBS: Ignore alarms coming from unknown devices
Alexey Starikovskiy [Sat, 8 Dec 2007 10:02:40 +0000 (13:02 +0300)] 
ACPI: SBS: Ignore alarms coming from unknown devices

http://bugzilla.kernel.org/show_bug.cgi?id=9362

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
16 years agoACPI: SBS: Reset alarm bit
Alexey Starikovskiy [Sat, 8 Dec 2007 10:02:33 +0000 (13:02 +0300)] 
ACPI: SBS: Reset alarm bit

Alarm bit should be cleared in order for other alarms to be sent.
http://bugzilla.kernel.org/show_bug.cgi?id=9362

Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
16 years agoPull hotplug into release branch
Len Brown [Fri, 14 Dec 2007 19:54:09 +0000 (14:54 -0500)] 
Pull hotplug into release branch

16 years ago[SPARC32]: Silence sparc32 warnings on missing syscalls.
Martin Habets [Fri, 14 Dec 2007 18:59:50 +0000 (10:59 -0800)] 
[SPARC32]: Silence sparc32 warnings on missing syscalls.

Silence sparc32 warnings on missing syscalls, these won't be added.
This patch is based on this mail:
http://www.mail-archive.com/linux-arch@vger.kernel.org/msg02571.html

Signed-off-by: Martin Habets <errandir_news@mph.eclipse.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[MIPS] Ensure that ST0_FR is never set on a 32 bit kernel
Chris Dearman [Thu, 13 Dec 2007 22:42:19 +0000 (22:42 +0000)] 
[MIPS] Ensure that ST0_FR is never set on a 32 bit kernel

Signed-off-by: Chris Dearman <chris@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] time: Delete weak definition of plat_time_init() due to gcc bug.
Ralf Baechle [Fri, 14 Dec 2007 00:05:34 +0000 (00:05 +0000)] 
[MIPS] time: Delete weak definition of plat_time_init() due to gcc bug.

Frank Rowand <frank.rowand@am.sony.com> reports:

> In linux-2.6.24-rc4 the Toshiba RBTX4927 hangs on boot.
>
> The cause is that plat_time_init() from arch/mips/tx4927/common/
> tx4927_setup.c does not override the __weak plat_time_init() from
> arch/mips/kernel/time.c.  This is due to a compiler bug in gcc 4.1.1.  The
> bug is reported to not exist in earlier versions of gcc, and to be fixed in
> 4.1.2.  The problem is that the __weak plat_time_init() is empty and thus
> gets optimized out of existence (thus the linker is never given the option
> to replace the __weak function).

[ He meant the call to plat_time_init() from time_init() gets optimized away ]

> For more info on the gcc bug see
>
>    http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27781
>
> The attached patch is one workaround.  Another possible workaround

[ His patch adds -fno-unit-at-a-time for time.c ]

> would be to change the __weak plat_time_init() to be a non-empty
> function.

The __weak definition of plat_time_init was only ever meant to be a
migration helper to keep platforms that don't have a plat_time_init
compiling.  A few greps says that all platforms now supply their own
plat_time_init() so the weak definition is no longer needed.  So I
instead delete it.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] PCI: Make pcibios_fixup_device_resources ignore legacy resources.
Ralf Baechle [Tue, 11 Dec 2007 19:49:24 +0000 (19:49 +0000)] 
[MIPS] PCI: Make pcibios_fixup_device_resources ignore legacy resources.

There might be other reasons why a resource might be marked as fixed
such as a PCI UART holding the system console but until we use
IORESOURCE_PCI_FIXED that way also this will work.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] Atlas, Malta: Don't free firmware memory on free_initmem.
Ralf Baechle [Tue, 11 Dec 2007 19:38:12 +0000 (19:38 +0000)] 
[MIPS] Atlas, Malta: Don't free firmware memory on free_initmem.

A proper fix for this needs to turn a few MIPS-generic bits which I
don't want at this stage.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] Alchemy: fix off by two error in __fixup_bigphys_addr()
Sergei Shtylyov [Mon, 10 Dec 2007 17:36:50 +0000 (20:36 +0300)] 
[MIPS] Alchemy: fix off by two error in __fixup_bigphys_addr()

the PCI specific code in this function doesn't check for the address range
being under the upper bound of the PCI memory window correctly -- fix this,
somewhat beautifying the code around the check, while at it...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] Alchemy: fix PCI resource conflict
Sergei Shtylyov [Mon, 10 Dec 2007 17:28:51 +0000 (20:28 +0300)] 
[MIPS] Alchemy: fix PCI resource conflict

... by getting the PCI resources back into the 32-bit range -- there's no
need therefore for CONFIG_RESOURCES_64BIT either. This makes Alchemy PCI
work again while currently the kernel skips the bus scan.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years ago[MIPS] time: Set up Cobalt's mips_hpt_frequency
Yoichi Yuasa [Sun, 9 Dec 2007 12:22:04 +0000 (21:22 +0900)] 
[MIPS] time: Set up Cobalt's mips_hpt_frequency

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
16 years agoMerge branch 'for-2.6.24' of master.kernel.org:/pub/scm/linux/kernel/git/galak/powerp...
Paul Mackerras [Fri, 14 Dec 2007 05:46:50 +0000 (16:46 +1100)] 
Merge branch 'for-2.6.24' of /linux/kernel/git/galak/powerpc into merge

16 years ago[POWERPC] Make PS3_SYS_MANAGER default y, not m
Michael Ellerman [Mon, 10 Dec 2007 04:17:43 +0000 (15:17 +1100)] 
[POWERPC] Make PS3_SYS_MANAGER default y, not m

Currently it's impossible to build a ps3_defconfig which will reboot
without modules installed. This makes it all too easy to find yourself
with a PS3 that won't reboot.

This is because the system manager driver, which provides the reboot
mechanism, is only selectable if PS3_ADVANCED is set, else it defaults
to m. In ps3_defconfig PS3_ADVANCED is not set, therefore the system
manager is built as a module.

It would be desirable IMHO for the defconfig to produce a kernel that
boots and reboots, without needing modules to be installed.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
16 years ago[POWERPC] Fix rounding bug in emulation for double float operating
Liu Yu [Mon, 10 Dec 2007 05:00:52 +0000 (13:00 +0800)] 
[POWERPC] Fix rounding bug in emulation for double float operating

This patch fixes rounding bug in emulation for double float operating on PowerPC platform.

When pack double float operand, it need to truncate the tail due to the limited precision.
If the truncated part is not zero, the last bit of work bit (totally 3 bits) need to '|' 1.

This patch is completed in _FP_FRAC_SRS_2(X,N,sz) (arch/powerpc/math-emu/op-2.h).
Originally the code leftwards rotates the operand to just keep the truncated part,
then check whether it is zero. However, the number it rotates is not correct when
N is not smaller than _FP_W_TYPE_SIZE, and it will cause the work bit '|' 1 in the improper case.

This patch fixes this issue.

Signed-off-by: Liu Yu <b13201@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years ago[POWERPC] iSeries: don't printk with HV spinlock held
Stephen Rothwell [Wed, 12 Dec 2007 03:58:12 +0000 (14:58 +1100)] 
[POWERPC] iSeries: don't printk with HV spinlock held

Printk was observed to hang during module unload due to a limited
window of characters that may be sent to the hypervisor.  The window
only reexpands when we receive an ack from the HV and the spinlock here
prevents us from ever processing that ack.  This fixes it by dropping
the lock before doing the printk, then looping back to the top to
reacquire the lock.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
16 years ago[POWERPC] 82xx: mpc8272ads, pq2fads: Update defconfig with CONFIG_FS_ENET_MDIO_FCC
Scott Wood [Mon, 10 Dec 2007 17:10:45 +0000 (11:10 -0600)] 
[POWERPC] 82xx: mpc8272ads, pq2fads: Update defconfig with CONFIG_FS_ENET_MDIO_FCC

This was recently made configurable, and needs to be set for these boards.

Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years ago[POWRPC] CPM2: Eliminate section mismatch warning in cpm2_reset().
Scott Wood [Wed, 12 Dec 2007 22:54:32 +0000 (16:54 -0600)] 
[POWRPC] CPM2: Eliminate section mismatch warning in cpm2_reset().

Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years ago[POWERPC] Kill non-existent symbols from ksyms and commproc.h
Jochen Friedrich [Thu, 22 Nov 2007 16:53:54 +0000 (17:53 +0100)] 
[POWERPC] Kill non-existent symbols from ksyms and commproc.h

Remove exports of __res and cpm_install_handler/cpm_free_handler.  Remove
cpm_install_handler/cpm_free_handler from the commproc.h as well.  Both
were used for ARCH=ppc and aren't defined for ARCH=powerpc.

CC      arch/powerpc/kernel/ppc_ksyms.o
arch/powerpc/kernel/ppc_ksyms.c:180: error: '__res' undeclared here (not in a function)
arch/powerpc/kernel/ppc_ksyms.c:180: warning: type defaults to 'int' in declaration of '__res'
make[1]: *** [arch/powerpc/kernel/ppc_ksyms.o] Error 1
make: *** [arch/powerpc/kernel] Error 2

LD      .tmp_vmlinux1
arch/powerpc/kernel/built-in.o:(__ksymtab+0x198): undefined reference to `cpm_free_handler'
arch/powerpc/kernel/built-in.o:(__ksymtab+0x1a0): undefined reference to `cpm_install_handler'
make: *** [.tmp_vmlinux1] Error 1

Signed-off-by: Jochen Friedrich <jochen@scram.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Vitaly Bordug <vitb@kernel.crashing.org>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years ago[POWERPC] Fix typo #ifdef -> #ifndef
Jochen Friedrich [Thu, 22 Nov 2007 16:53:47 +0000 (17:53 +0100)] 
[POWERPC] Fix typo #ifdef -> #ifndef

fpi->cp_command should be overwritten only if CONFIG_PPC_CPM_NEW_BINDING
is NOT set. Otherwise it is already set from the device tree.

Signed-off-by: Jochen Friedrich <jochen@scram.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Vitaly Bordug <vitb@kernel.crashing.org>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
16 years agoPull battery-2.6.24 into release branch
Len Brown [Fri, 14 Dec 2007 03:16:14 +0000 (22:16 -0500)] 
Pull battery-2.6.24 into release branch

16 years agoPull thinkpad-2.6.24 into release branch
Len Brown [Fri, 14 Dec 2007 03:12:38 +0000 (22:12 -0500)] 
Pull thinkpad-2.6.24 into release branch

16 years agoACPI: thinkpad-acpi: fix lenovo keymap for brightness
Henrique de Moraes Holschuh [Thu, 13 Dec 2007 14:14:09 +0000 (12:14 -0200)] 
ACPI: thinkpad-acpi: fix lenovo keymap for brightness

Starting in 2.6.23...

Several reports from X60 users complained that the default Lenovo keymap
issuing EV_KEY KEY_BRIGHTNESS_UP/DOWN input events caused major issues when
the proper brightness support through ACPI video.c was loaded.

Therefore, remove the generation of these events by default, which is the
right thing for T60, X60, R60, T61, X61 and R61 with their latest BIOSes.

Distros that want to misuse these events into OSD reporting (which requires
an ugly hack from hell in HAL) are welcome to set up the key map they need
through HAL.  That way, we don't break everyone else's systems.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Len Brown <len.brown@intel.com>
16 years agoACPI: fix modpost warnings
Jan Beulich [Thu, 13 Dec 2007 08:32:26 +0000 (08:32 +0000)] 
ACPI: fix modpost warnings

for sn2_defconfig:

WARNING: vmlinux.o(.text+0x4b8601): Section mismatch: reference to .init.data:node_to_pxm_map (between '__acpi_map_pxm_to_node' and 'acpi_get_pxm')
WARNING: vmlinux.o(.text+0x4b8741): Section mismatch: reference to .init.data:pxm_to_node_map (between 'acpi_map_pxm_to_node' and 'acpi_get_node')

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Len Brown <len.brown@intel.com>
16 years agoACPI: video_device_list corruption
William Lee Irwin III [Wed, 12 Dec 2007 11:56:55 +0000 (03:56 -0800)] 
ACPI: video_device_list corruption

The ->cap fields of struct acpi_video_device and struct acpi_video_bus
are 1B each, not 4B. The oversized memset()'s corrupted the subsequent
list_head fields. This resulted in silent corruption without
CONFIG_DEBUG_LIST and BUG's with it. This patch uses sizeof() to pass
the proper bounds to the memset() calls and thereby correct the bugs.

Signed-off-by: William Irwin <wli@holomorphy.com>
Acked-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Len Brown <len.brown@intel.com>
16 years agoV4L/DVB (6820): s5h1409: QAM SNR related fixes
Steven Toth [Thu, 13 Dec 2007 01:14:00 +0000 (22:14 -0300)] 
V4L/DVB (6820): s5h1409: QAM SNR related fixes

QAM SNR values were incorrect when the cable was disconnected. This
patch extends the lookup tables to ensure correct values are being
returned.

Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
16 years agoV4L/DVB (6819): i2c: fix drivers/media/video/bt866.c
Andrew Morton [Thu, 13 Dec 2007 05:30:26 +0000 (02:30 -0300)] 
V4L/DVB (6819): i2c: fix drivers/media/video/bt866.c

usage_count got removed.

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
16 years agoV4L/DVB (6814): Makefile: always enter video/
Adrian Bunk [Wed, 12 Dec 2007 18:40:32 +0000 (15:40 -0300)] 
V4L/DVB (6814): Makefile: always enter video/

Since not all code under drivers/media/video/ depends on
CONFIG_VIDEO_DEV we cannot only enter it depending
on CONFIG_VIDEO_DEV.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
16 years agoV4L/DVB (6798): saa7134: enable LNA in analog mode for Hauppauge WinTV HVR-1110
Michael Krufky [Mon, 10 Dec 2007 00:13:38 +0000 (21:13 -0300)] 
V4L/DVB (6798): saa7134: enable LNA in analog mode for Hauppauge WinTV HVR-1110

Thanks to Hermann Pitton for noticing that this was missing.

Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Cc: Hermann Pitton <hermann-pitton@arcor.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
16 years agokbuild: re-enable Makefile generation in a new O=... directory
Guillaume Chazarain [Tue, 11 Dec 2007 12:22:08 +0000 (13:22 +0100)] 
kbuild: re-enable Makefile generation in a new O=... directory

The commit:
18c32dac75b187d1a4e858f3cfdf03e844129f5e "kbuild: fix
building with O=.. options"
disabled the creation of a Makefile in a new O=... directory. Restore it.

Signed-off-by: Guillaume Chazarain <guichaz@yahoo.fr>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
16 years agoIB/ehca: Fix lock flag variable location, bump version number
Joachim Fenkes [Thu, 13 Dec 2007 12:35:57 +0000 (13:35 +0100)] 
IB/ehca: Fix lock flag variable location, bump version number

Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
16 years ago[SPARC64]: Fix two kernel linear mapping setup bugs.
David S. Miller [Thu, 13 Dec 2007 14:13:38 +0000 (06:13 -0800)] 
[SPARC64]: Fix two kernel linear mapping setup bugs.

This was caught and identified by Greg Onufer.

Since we setup the 256M/4M bitmap table after taking over the trap
table, it's possible for some 4M mapping to get loaded in the TLB
beforhand which later will be 256M mappings.

This can cause illegal TLB multiple-match conditions.  Fix this by
setting up the bitmap before we take over the trap table.

Next, __flush_tlb_all() was not doing anything on hypervisor
platforms.  Fix by adding sun4v_mmu_demap_all() and calling it.

Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agoV4L/DVB (6803): buf-core.c locking fixes
Adrian Bunk [Wed, 12 Dec 2007 19:44:54 +0000 (16:44 -0300)] 
V4L/DVB (6803): buf-core.c locking fixes

After commit 19fb1457990b6b7e15586ec7331541a184233acc the callers in
videobuf-core.c that already hold the lock must call
__videobuf_read_start() instead of videobuf_read_start().

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
16 years agoMAINTAINERS: update the NFS CLIENT entry
Trond Myklebust [Thu, 13 Dec 2007 01:16:06 +0000 (20:16 -0500)] 
MAINTAINERS: update the NFS CLIENT entry

Add in the new NFS mailing list on vger, website, and git tree info, and
update my email address to reflect the fact that I've been working for
netapp for the past 2 years.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
16 years agoide: fix ->io_32bit race in set_io_32bit()
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:32:00 +0000 (23:32 +0100)] 
ide: fix ->io_32bit race in set_io_32bit()

set_io_32bit() (ide_procset_t function) can race against running
PIO transfers.  Fix it by using ide_spin_wait_hwgroup().

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove stale changelog from ide-probe.c
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:32:00 +0000 (23:32 +0100)] 
ide: remove stale changelog from ide-probe.c

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove stale changelog from ide-disk.c
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:32:00 +0000 (23:32 +0100)] 
ide: remove stale changelog from ide-disk.c

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove dead code from __ide_dma_test_irq()
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:32:00 +0000 (23:32 +0100)] 
ide: remove dead code from __ide_dma_test_irq()

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agohpt366: fix HPT37x PIO mode timings (take 2)
Sergei Shtylyov [Wed, 12 Dec 2007 22:31:58 +0000 (23:31 +0100)] 
hpt366: fix HPT37x PIO mode timings (take 2)

After looking into the HPT370 manual (now that I have it) and re-checking all
the timing tables, here's what I have discovered:

- at 33 MHz clock, PIO mode 0 timings turned to be overclocked, and all other
  PIO modes underclocked;

- at 50 MHz clock, PIO modes 0 to 2 turned to be overclocked;

- at 66 MHz clock, PIO mode 0 was overclocked too.

Finally, the taskfile timing (matching PIO mode 0) turned to be overclocked at
all clock frequencies (and in all manuals)...

The new timings have been tested on HPT370 chip (at 33 MHz PCI clock) and on
HPT371N chip (at both 50 and 66 MHz DPLL clock).

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agopdc202xx_new: fix Promise TX4 support
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:58 +0000 (23:31 +0100)] 
pdc202xx_new: fix Promise TX4 support

In case of Promise TX4 the first PCI device is located at slot 1
and the second one is at slot 2 so the offset used by pci_get_slot()
should be "+1" and not "+2".

Thanks goes out to Markus Dietz for bugreport and testing this patch.

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-cd: remove dead post_transform_command()
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:58 +0000 (23:31 +0100)] 
ide-cd: remove dead post_transform_command()

post_transform_command() call in cdrom_newpc_intr() has no effect because
it is done after the request has already been fully completed (rq->bio and
rq->data are always NULL).  It was verified to be true regardless whether
INQUIRY command is using DMA or PIO to transfer data (by using modified
Tejun Heo's test-shortsg.c utility and adding a few printk()-s to ide-cd).

This was uncovered thanks to the "blk_end_request: full I/O completion
handler (take 3)" patch series from Kiyoshi Ueda.

Cc: jens.axboe@oracle.com
Cc: bharrosh@panasas.com
Cc: Kiyoshi Ueda <k-ueda@ct.jp.nec.com
Cc: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Cc: Tejun Heo <htejun@gmail.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: DMA reporting and validity checking fixes (take 3)
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:58 +0000 (23:31 +0100)] 
ide: DMA reporting and validity checking fixes (take 3)

* ide_xfer_verbose() fixups:
  - beautify returned mode names
  - fix PIO5 reporting
  - make it return 'const char *'

* Change printk() level from KERN_DEBUG to KERN_INFO in ide_find_dma_mode().

* Add ide_id_dma_bug() helper based on ide_dma_verbose() to check for invalid
  DMA info in identify block.

* Use ide_id_dma_bug() in ide_tune_dma() and ide_driveid_update().

  As a result DMA won't be tuned or will be disabled after tuning if device
  reports inconsistent info about enabled DMA mode (ide_dma_verbose() does the
  same checks while the IDE device is probed by ide-{cd,disk} device driver).

* Remove no longer needed ide_dma_verbose().

This patch should fix the following problem with out-of-sync IDE messages
reported by Nick Warne:

       hdd: ATAPI 48X DVD-ROM DVD-R-RAM CD-R/RW drive, 2048kB Cache<7>hdd:
       skipping word 93 validity check
        , UDMA(66)

and later debugged by Mark Lord to be caused by:

        ide_dma_verbose()
                printk( ... "2048kB Cache");
        eighty_ninty_three()
                printk(KERN_DEBUG "%s: skipping word 93 validity check\n");
        ide_dma_verbose()
                printk(", UDMA(66)"

Please note that as a result ide-{cd,disk} device drivers won't report the
DMA speed used but this is intended since now DMA mode being used is always
reported by IDE core code.

v2:
* fixes suggested by Randy:
  - use KERN_CONT for printk()-s in ide-{cd,disk}.c
  - don't remove argument name from ide_xfer_verbose() declaration

v3:
* Remove incorrect check for (id->field_valid & 1) from ide_id_dma_bug()
  (spotted by Sergei).

* "XFER SLOW" -> "PIO SLOW" in ide_xfer_verbose() (suggested by Sergei).

* Fix ide_find_dma_mode() to report the correct mode ('mode' after being
  limited by 'req_mode').

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Nick Warne <nick@ukfsn.org>
Cc: Mark Lord <lkml@rtr.ca>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: add /sys/bus/ide/devices/*/{model,firmware,serial} sysfs entries
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:58 +0000 (23:31 +0100)] 
ide: add /sys/bus/ide/devices/*/{model,firmware,serial} sysfs entries

Cc: Dan Kegel <dank@kegel.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Xavier Bestel <xavier.bestel@free.fr>
Cc: Andrey Borzenkov <arvidjaar@mail.ru>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: coding style fixes for drivers/ide/setup-pci.c
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:57 +0000 (23:31 +0100)] 
ide: coding style fixes for drivers/ide/setup-pci.c

* remove trailing whitespaces
* 'if()' -> 'if ()'
* remove extra new-line before EXPORT_SYMBOL_GPL()
* add extra new-line after 'id' definition
* respect 80-columns limit

There should be no functionality changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: fix ide_scan_pcibus() error message
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:57 +0000 (23:31 +0100)] 
ide: fix ide_scan_pcibus() error message

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: deprecate CONFIG_BLK_DEV_OFFBOARD
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:57 +0000 (23:31 +0100)] 
ide: deprecate CONFIG_BLK_DEV_OFFBOARD

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: add missing checks for control register existence
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:57 +0000 (23:31 +0100)] 
ide: add missing checks for control register existence

Add missing checks for control register existence (some legacy m68k specific
IDE controllers don't have it).  Also use drive->ctl while at it.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: add ide_scsi_hex_dump() helper
Bartlomiej Zolnierkiewicz [Wed, 12 Dec 2007 22:31:57 +0000 (23:31 +0100)] 
ide-scsi: add ide_scsi_hex_dump() helper

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Denis Cheng <crquan@gmail.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoIB/ehca: Serialize HCA-related hCalls if necessary
Joachim Fenkes [Mon, 10 Dec 2007 17:59:10 +0000 (18:59 +0100)] 
IB/ehca: Serialize HCA-related hCalls if necessary

Several pSeries firmware versions share a rare locking issue in the
HCA-related hCalls. Check for a feature flag that indicates the issue
being fixed and serialize all HCA hCalls if not.

Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
16 years agoIB/ehca: Return correct number of SGEs for SRQ
Joachim Fenkes [Mon, 10 Dec 2007 11:20:57 +0000 (12:20 +0100)] 
IB/ehca: Return correct number of SGEs for SRQ

Firmware would round up the number of SGEs to four, because the WQE
structure holds four SGEs. For SRQ, only three are supported, so return
a fixed value instead.

Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
16 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6
Linus Torvalds [Wed, 12 Dec 2007 17:52:12 +0000 (09:52 -0800)] 
Merge /pub/scm/linux/kernel/git/davem/sparc-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6:
  [SPARC64]: Fix endless loop in cheetah_xcall_deliver().
  [SERIAL] sparc: Infrastructure to fix section mismatch bugs.

16 years agoMerge branch 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6
Linus Torvalds [Wed, 12 Dec 2007 16:43:51 +0000 (08:43 -0800)] 
Merge branch 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6

* 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6:
  i2c/isp1301_omap: Build fix
  i2c: Add missing spaces in split log messages
  i2c-gpio: Initialize adapter class
  i2c: Delete an outdated piece of documentation

16 years agoNFS: Fix an Oops in NFS unmount
Trond Myklebust [Wed, 12 Dec 2007 16:12:15 +0000 (11:12 -0500)] 
NFS: Fix an Oops in NFS unmount

Ensure that the dummy 'root dentry' is invisible to d_find_alias(). If not,
then it may be spliced into the tree if a parent directory from the same
filesystem gets mounted at a later time.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>