pandora-kernel.git
18 years agoMove code from ipw2100_wpa_enable to IPW2100_PARAM_DROP_UNENCRYPTED to
Zhu Yi [Wed, 13 Jul 2005 17:30:34 +0000 (12:30 -0500)]
Move code from ipw2100_wpa_enable to IPW2100_PARAM_DROP_UNENCRYPTED to
support wpa_supplicant with open AP. We need this to make driver_ipw
work.

driver_ext has already had the similar code with the WE-18 support
added.

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years ago[Bug 339] Fix ipw2100 iwconfig set/get txpower.
Liu Hong [Wed, 13 Jul 2005 17:29:21 +0000 (12:29 -0500)]
[Bug 339] Fix ipw2100 iwconfig set/get txpower.

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoFixed WEP on ipw2100 (priv->sec was being used instead of
<jketreno@io.(none)> [Tue, 12 Jul 2005 20:45:30 +0000 (15:45 -0500)]
Fixed WEP on ipw2100 (priv->sec was being used instead of
priv->ieee->sec)

18 years agoCatch ipw2100 up to equivelancy with v1.1.1
James Ketrenos [Thu, 25 Aug 2005 03:33:31 +0000 (22:33 -0500)]
Catch ipw2100 up to equivelancy with v1.1.1

* Added WE-18 support.  This allows the use of -Dext with wpa_supplicant
  > 0.4.x (thanks to Hong Liu)
* Fixed #339 problem with iwconfig set/get txpower (thanks to Hong Liu)
* Fixed #598 problem when with error messages when module loaded with
  'disable=1' (thanks to Hong Liu)
* Fixed #640 problem with 'iwlist retry' now showing min/max retry
* Fixed compatibility with wpa_supplicant and the new -Dipw interface
  (that included a fix for 64-bit compatibility)
* Added CFG_CRC_CHECK which allows passing through packets with bad
  CRCs while in monitor mode.

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoCatch ipw2200 up to equivelancy with v1.0.4
James Ketrenos [Thu, 25 Aug 2005 03:04:42 +0000 (22:04 -0500)]
Catch ipw2200 up to equivelancy with v1.0.4

* Fixed #627 problem with open APs not working with wpa_supplicant
* Fixed #632 problem with 'txpower auto' setting power incorrectly (thanks
  to Kai Groner)
* Fixed #634 problem with 'iwconfig eth1 frag 0' hanging the shell
* Fixed problem with adapter not fully powering off during suspend to RAM or
  when module unloaded.
* Fixed #645 problem with turning fixed rates off not taking effect until
  you reload the driver
* Fixed problem with firmware restart if wpa_supplicant was used to set a key
  that wasn't exactly 5 or 13 bytes in length.
* Fixed #623 Added iwpriv sw_reset extension to reset sw parameters
* Added managment frame export to user space with frame statistics
* Fixed #652 Modified the driver to load the EEPROM data even if RF KILL is
  active during driver load
* Global s:CX2_:IPW_:g to make code more consistent
* Fixed #572 problem with setting txpower to auto
* Fixed #656 problem with kernel oops if mode auto; modprobe -r ipw2200
* Added QoS (CONFIG_IPW_QOS) support.  This is being actively developed but
  is the first step in getting WMM support into the driver and the kernel.
* Fixed some race conditions with channel changes, association, and scan
  abort that could periodically cause a firmware restart.
* Added some extensions to export scan and network statistics to user space
  (exposed through speed_scan and net_stats sysfs entries)
* Fixed a few bugs in how monitor mode was supported (scan lists
  weren't quite right)
* Updated the firmware requirement from 2.2 to 2.3 which supports
  monitor mode.

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoCatch ipw2200 up to equivelancy with v1.0.3
James Ketrenos [Thu, 25 Aug 2005 02:56:24 +0000 (21:56 -0500)]
Catch ipw2200 up to equivelancy with v1.0.3

* Fix #616 problem with OOPS on module load (thanks to Yi Zhu)
* Fixed problem with led module parameter being described as
  'auto_create'
* Added support to merge between adhoc networks (thanks to Mohamed Abbas)
* Added semaphore lock at the driver's entry points to protect against
  re-entry (thanks to Mohamed Abbas)
* Added semaphore lock to background scheduled driver actions (thanks to
  Mohamed Abbas)
* Changed how signal quality is reported for scan output (thanks to
  Peter Jones)
* Fixed how high/low clamp values of signal quality are reported so a
  more consistent ramp is provided (thanks to Bill Moss)
* Fix #624 problem with duplicate addresses (again)  (thanks to Bernard
  Blackham)
* Fix #385 problem with fragmentation and certain sized packets (thanks
  to Mohamed Abbas)
* Modified iwconfig network name if RF kill is enabled to say 'radio off'
* Fix #382 problem with driver not responding to probe requests in Ad-Hoc
  mode (thanks to Mohamed Abbas)

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoCatch ipw2200 up to equivelancy with v1.0.2
James Ketrenos [Thu, 25 Aug 2005 02:43:11 +0000 (21:43 -0500)]
Catch ipw2200 up to equivelancy with v1.0.2

Removed unneeded parenthesis around numeric constant defines

Added support for iwspy

Put in fix for Ad-Hoc mode not passing through all packets (thanks to KKH)

Put in fix for fragmentation not working for fragment sizes between
441-464 bytes (thanks to Mohamed Abbas)

Fixed #592 problem of CONFIG_IEEE80211_WPA_MODULE not including WPA
support into the driver -- fixed as a result of no longer limiting WPAs
inclusion

Fixed #594 problem with user rates mask causing lack of association if
AP mandatory rate is masked out.  We now add back in as a supported rate
any mandatory rate.

Fixed #597 kernel oops due to calling dev_kfree_skb on an skb multiple times.

Added code to control LEDs that can be controlled through the wireless
NIC (vs. non-wireless HW interfaces) -- this is currently disabled by
default due to reports by some users of it hanging their laptop.

Added some more debug messages around fragmentation logic

Added locking around STATUS_HCMD_ACTIVE to prevent re-entry race
conditions

Moved ipw_adapter_restart to only execute on the priv->workqueue to
keep keyboard errors from occuring during adapter restart

Added CFG_BACKGROUND_SCAN to easily allow people to play with
background scanning implementations

Modified WPA logic to send WPA IE if one is set (vs. being based on
wpa_enabled)

Modified scan result logic to report WPA and RSN IEs if set (vs. being
based on wpa_enabled)

Fixed issues with endianess compatability between the host and
wireless adapter (thanks to York Liu and Yi Zhu)

Fixed problem with Ad-Hoc network creation causing a firmware error if
a scan was actively running (thanks to Mohamed Abbas)

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoCatch ipw2200 up to equivelancy with v1.0.1
James Ketrenos [Thu, 25 Aug 2005 02:25:16 +0000 (21:25 -0500)]
Catch ipw2200 up to equivelancy with v1.0.1

This commit contains the following fixes:

Fixed #559: iwconfig rate support (thanks to Florian Hackenberger)

Improved link signal quality calculation (thanks to Bill Moss)

Fixed a problem with sensitivity threshold during association

Added iwpriv for turning forcing long preamble support:
  % iwpriv eth1 set_preamble 1|0

Fixed #542 and #377 support for short preamble

Fixed locked BSSID reporting channel number (thanks to Pedro
Ramalhais)

Fixed type-o with scan watchdog timeout message (thanks to Pedro
Ramalhais)

Changed logic for displaying get_mode output so the code is easier to
follow (thanks to Pedro Ramalhais)

Added initial support for WPA (thanks to Yi Zhu) -- tested with
wpa_supplicant (either tip w/ ipw driver, or with -Dipw2100) with
both CCMP and TKIP

Fixed problem with CCMP not working due to uninitialized 802.11
header fields (thanks to Pedro Ramalhais)

Bug references are to defects stored on http://bughost.org

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoIPW_DEBUG has already included DRV_NAME, remove double prefix print.
Zhu Yi [Mon, 4 Jul 2005 06:06:00 +0000 (14:06 +0800)]
IPW_DEBUG has already included DRV_NAME, remove double prefix print.

18 years agoRan scripts/Lindent on drivers/net/wireless/ipw2{1,2}00.{c,h}
James Ketrenos [Wed, 14 Sep 2005 14:47:29 +0000 (09:47 -0500)]
Ran scripts/Lindent on drivers/net/wireless/ipw2{1,2}00.{c,h}

No other changes.

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoUpdate version ieee80211 stamp to 1.1.7
James Ketrenos [Tue, 25 Oct 2005 01:44:06 +0000 (20:44 -0500)]
Update version ieee80211 stamp to 1.1.7

18 years agoFix problem with WEP unicast key > index 0
Volker Braun [Mon, 24 Oct 2005 15:15:36 +0000 (10:15 -0500)]
Fix problem with WEP unicast key > index 0

The functions ieee80211_wx_{get,set}_encodeext fail if one tries to set
unicast (IW_ENCODE_EXT_GROUP_KEY not set) keys at key indices>0. But at
least some Cisco APs dish out dynamic WEP unicast keys at index !=0.

Signed-off-by: Volker Braun <volker.braun@physik.hu-berlin.de>
Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years agoscripts/Lindent on ieee80211 subsystem.
James Ketrenos [Mon, 24 Oct 2005 15:20:53 +0000 (10:20 -0500)]
scripts/Lindent on ieee80211 subsystem.

Signed-off-by: James Ketrenos <jketreno@linux.intel.com>
18 years ago[PATCH] b44: late request_irq in b44_open
Francois Romieu [Mon, 7 Nov 2005 00:52:57 +0000 (01:52 +0100)]
[PATCH] b44: late request_irq in b44_open

Don't request_irq before the registers are reset/init.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] b44: s/spin_lock_irqsave/spin_lock/ in b44_interrupt
Francois Romieu [Mon, 7 Nov 2005 00:52:06 +0000 (01:52 +0100)]
[PATCH] b44: s/spin_lock_irqsave/spin_lock/ in b44_interrupt

There is no need to save/restore the irq state as the irq are always
locally disabled when b44_interrupt is issued.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] b44: expose counters through ethtool
Francois Romieu [Mon, 7 Nov 2005 00:51:34 +0000 (01:51 +0100)]
[PATCH] b44: expose counters through ethtool

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] b44: miscellaneous cleanup
Francois Romieu [Mon, 7 Nov 2005 00:50:46 +0000 (01:50 +0100)]
[PATCH] b44: miscellaneous cleanup

- remove unneeded forward declarations
- s/kmalloc + memset/kzalloc/
- whitespace readjustement can't hurt
- wrong comment: b44_init_rings _is_ called with a spinlock held in
  b44_{open/set_ringparam/set_pauseparam/etc}.
  Actually, it does not need to be able to sleep
- b44_remove_one() can not be issued with a NULL device in its
  private member: remove the test.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] b44: b44_start_xmit returns with a lock held when it fails allocating
Francois Romieu [Mon, 7 Nov 2005 00:50:03 +0000 (01:50 +0100)]
[PATCH] b44: b44_start_xmit returns with a lock held when it fails allocating

The patch simply factors out the release of the lock.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years agoMerge branch 'master'
Jeff Garzik [Mon, 7 Nov 2005 04:01:34 +0000 (23:01 -0500)]
Merge branch 'master'

18 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/sam/kbuild
Linus Torvalds [Mon, 7 Nov 2005 00:59:14 +0000 (16:59 -0800)]
Merge /pub/scm/linux/kernel/git/sam/kbuild

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-drvmodel
Linus Torvalds [Mon, 7 Nov 2005 00:58:38 +0000 (16:58 -0800)]
Merge kernel.org:/home/rmk/linux-2.6-drvmodel

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Mon, 7 Nov 2005 00:58:11 +0000 (16:58 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-arm

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-serial
Linus Torvalds [Mon, 7 Nov 2005 00:57:34 +0000 (16:57 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-serial

18 years ago[PATCH] ppc64: support 64k pages
Benjamin Herrenschmidt [Mon, 7 Nov 2005 00:06:55 +0000 (11:06 +1100)]
[PATCH] ppc64: support 64k pages

Adds a new CONFIG_PPC_64K_PAGES which, when enabled, changes the kernel
base page size to 64K.  The resulting kernel still boots on any
hardware.  On current machines with 4K pages support only, the kernel
will maintain 16 "subpages" for each 64K page transparently.

Note that while real 64K capable HW has been tested, the current patch
will not enable it yet as such hardware is not released yet, and I'm
still verifying with the firmware architects the proper to get the
information from the newer hypervisors.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[ARM] Fix /proc/cpuinfo format for ARM SMP
Russell King [Sun, 6 Nov 2005 21:41:08 +0000 (21:41 +0000)]
[ARM] Fix /proc/cpuinfo format for ARM SMP

glibc expects to count lines beginning with "processor" to determine
the number of processors, not lines beginning with "Processor".  So,
give glibc the format it expects.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] Don't call dump_cpu_info unless we're booting
Russell King [Sun, 6 Nov 2005 19:49:21 +0000 (19:49 +0000)]
[ARM] Don't call dump_cpu_info unless we're booting

We don't want to call dump_cpu_info() from cpu_init() after boot since
it produces a lot of unnecessary noise - since cpu_init() gets called
on resume and hotplug cpu insertion events.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] Fix second missing declaration of cache_is_vivt()
Russell King [Sun, 6 Nov 2005 19:47:04 +0000 (19:47 +0000)]
[ARM] Fix second missing declaration of cache_is_vivt()

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] Fix missing declaration of cache_is_vivt()
Russell King [Sun, 6 Nov 2005 15:46:57 +0000 (15:46 +0000)]
[ARM] Fix missing declaration of cache_is_vivt()

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] Fix another use of // as a comment
Russell King [Sun, 6 Nov 2005 15:45:00 +0000 (15:45 +0000)]
[ARM] Fix another use of // as a comment

// disagrees with ld's script parsing ability.  Don't use it.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3113/1: PXA: Allow machines to override (and also reuse) pxa pm functions
Richard Purdie [Sun, 6 Nov 2005 15:03:23 +0000 (15:03 +0000)]
[ARM] 3113/1: PXA: Allow machines to override (and also reuse) pxa pm functions

Patch from Richard Purdie

Update the PXA pm.c file to allow machines (such as the Sharp
Zaurus) to override the standard pm functions but reuse/wrap them
where needed.

The init call is made slightly earlier to give machine code an init
level to override them in removing any race.

Signed-off-by: Richard Purdie
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3115/1: small optimizations to exception vector entry code
Nicolas Pitre [Sun, 6 Nov 2005 14:42:37 +0000 (14:42 +0000)]
[ARM] 3115/1: small optimizations to exception vector entry code

Patch from Nicolas Pitre

Since we know the value of cpsr on entry, we can replace the bic+orr with
a single eor.  Also remove a possible result delay (at least on XScale).

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3114/1: use ixp2000_reg_wrb in ixp2000 uengine loader
Lennert Buytenhek [Sun, 6 Nov 2005 14:34:13 +0000 (14:34 +0000)]
[ARM] 3114/1: use ixp2000_reg_wrb in ixp2000 uengine loader

Patch from Lennert Buytenhek

Make the uengine loader use ixp2000_reg_wrb in the right places.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3089/1: ixp4xx AHB/PCI endianness fix
Alessandro Zummo [Sun, 6 Nov 2005 14:34:12 +0000 (14:34 +0000)]
[ARM] 3089/1: ixp4xx AHB/PCI endianness fix

Patch from Alessandro Zummo

  This patch fixes AHB/PCI endianness problems when the
 processor is in little-endian mode.

 The patch configures the CSR register closely following the directives
 in [1], paragraph 4.1, page 19.

 According to the considerations in [1], page 11, while the AHB bus
 supports both endian modes, on the IXP4XX it always uses big-endian.

 The PCI bus is connected to the South AHB. A wrong setting in the CSR
 register will thus cause a malfunctional PCI bus.

 A schematic diagram of the bus interconnections on the IXP4XX
 can be found in [1], page 18.

 The patch has been verified to work on the NSLU2 in
 both LE and BE modes.

 The author is Peter Korsgaard.

 [1] IntelĀ® IXP4XX Product Line of Network Processors and IXC1100
 Control Plane Processor:
 Understanding Big Endian and Little Endian Modes

 http://www.intel.com/design/network/applnots/25423701.pdf

Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] 3088/1: PXA: Add machine support for the Sharp SL-6000x series of PDAs
Dirk Opfer [Sun, 6 Nov 2005 14:27:52 +0000 (14:27 +0000)]
[ARM] 3088/1: PXA: Add machine support for the Sharp SL-6000x series of PDAs

Patch from Dirk Opfer

This patch adds basic machine support for the Sharp SL-6000x (Tosa) PDAs.

Signed-off-by: Dirk Opfer
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years agokconfig: fix xconfig on fedora 2 & 3 (x86_64)
Sam Ravnborg [Sun, 6 Nov 2005 10:05:21 +0000 (11:05 +0100)]
kconfig: fix xconfig on fedora 2 & 3 (x86_64)

From: Than Ngo <than@redhat.com>
qt as installed on fedora core (2 and 3) does not work with vanilla
kernel. The linker fails to locate the qt lib:

Actual Results:  # make xconfig
  HOSTLD  scripts/kconfig/qconf
  /usr/bin/ld: cannot find -lqt
  collect2: ld returned 1 exit status

Than Ngo has provided following fix for the bug.

Cc: Than Ngo <than@redhat.com>
Acked-by: Dave Jones <davej@redhat.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
18 years ago[PATCH] kbuild: permanently fix kernel configuration include mess
Russell King [Sun, 30 Oct 2005 21:42:11 +0000 (22:42 +0100)]
[PATCH] kbuild: permanently fix kernel configuration include mess

Include autoconf.h into every kernel compilation via the gcc command line
using -imacros.  This ensures that we have the kernel configuration
included from the start, rather than relying on each file having #include
<linux/config.h> as appropriate.  History has shown that this is something
which is difficult to get right.

Since we now include the kernel configuration automatically, make
configcheck becomes meaningless, so remove it.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
18 years ago[SERIAL] Support Au1x00 8250 UARTs using the generic 8250 driver.
Pantelis Antoniou [Sun, 6 Nov 2005 09:07:03 +0000 (09:07 +0000)]
[SERIAL] Support Au1x00 8250 UARTs using the generic 8250 driver.

The offsets of the registers are in a different place, and
some parts cannot handle a full set of modem control signals.

Signed-off-by: Pantelis Antoniou <pantelis@embeddedalley.ocm>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[netdrvr s2io] warning fixes
Jeff Garzik [Sun, 6 Nov 2005 04:40:46 +0000 (23:40 -0500)]
[netdrvr s2io] warning fixes

From Andrew Morton.

18 years ago[netdrvr] fac_8xx build fix
Jeff Garzik [Sun, 6 Nov 2005 04:40:16 +0000 (23:40 -0500)]
[netdrvr] fac_8xx build fix

18 years agoRemove linux/version.h include from drivers/net/phy/* and net/ieee80211/*.
Jeff Garzik [Sun, 6 Nov 2005 04:39:54 +0000 (23:39 -0500)]
Remove linux/version.h include from drivers/net/phy/* and net/ieee80211/*.

Unused, and causes the files to be needlessly rebuilt in some cases.

18 years agoMerge git://git.tuxdriver.com/git/netdev-jwl
Jeff Garzik [Sun, 6 Nov 2005 03:20:11 +0000 (22:20 -0500)]
Merge git://git.tuxdriver.com/git/netdev-jwl

18 years ago[PATCH] airo.c/airo_cs.c: correct prototypes
Adrian Bunk [Sat, 5 Nov 2005 16:42:27 +0000 (17:42 +0100)]
[PATCH] airo.c/airo_cs.c: correct prototypes

This patch creates a file airo.h containing prototypes of the global
functions in airo.c used by airo_cs.c .

If you got strange problems with either airo_cs devices or in any other
completely unrelated part of the kernel shortly or long after a airo_cs
device was detected by the kernel, this might have been caused by the
fact that caller and callee disagreed regarding the size of the first
argument to init_airo_card()...

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] bnx2: update version and minor fixes
Michael Chan [Fri, 4 Nov 2005 16:53:48 +0000 (08:53 -0800)]
[PATCH] bnx2: update version and minor fixes

Some book keeping and a style fix.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] bnx2: refine bnx2_poll
Michael Chan [Fri, 4 Nov 2005 16:53:48 +0000 (08:53 -0800)]
[PATCH] bnx2: refine bnx2_poll

Refine bnx2_poll() logic to write back the most up-to-date status tag
when all work has been processed. This eliminates some occasional
extra interrupts when a older status tag is written even though all
work has been processed.

The idea is to read the status tag just before exiting bnx2_poll() and
then check again for any new work. If no new work is pending, the
status tag written back will not generate any extra interrupt. This
logic is similar to the changes David Miller did to tg3_poll().

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] bnx2: update firmware handshake for 5708
Michael Chan [Fri, 4 Nov 2005 16:51:21 +0000 (08:51 -0800)]
[PATCH] bnx2: update firmware handshake for 5708

Dynamically determine the shared memory location where eeprom
parameters are stored instead of using a fixed location.

Add speed reporting to management firmware. This allows management
firmware to know the current speed without contending for MII
registers.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] bnx2: update nvram code for 5708
Michael Chan [Fri, 4 Nov 2005 16:49:17 +0000 (08:49 -0800)]
[PATCH] bnx2: update nvram code for 5708

Update bnx2 nvram code with support for 5708.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] bnx2: update firmware for 5708
Michael Chan [Fri, 4 Nov 2005 16:48:02 +0000 (08:48 -0800)]
[PATCH] bnx2: update firmware for 5708

Update bnx2 firmware with support for 5708.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] bnx2: add 5708 support
Michael Chan [Fri, 4 Nov 2005 16:45:49 +0000 (08:45 -0800)]
[PATCH] bnx2: add 5708 support

Add 5708 copper and serdes basic support, including 2.5 Gbps support
on 5708 serdes. SPEED_2500 is also added to ethtool.h

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] prism54: Remove redundant assignment
Daniel Drake [Thu, 3 Nov 2005 22:45:02 +0000 (22:45 +0000)]
[PATCH] prism54: Remove redundant assignment

The last patch I sent in ("prism54: Free skb after disabling
interrupts") included a redundant NULL assignment. Thanks to Herbert
Xu for pointing it out.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] drivers/net/hamradio/dmascc.c: remove dmascc_setup()
Adrian Bunk [Mon, 31 Oct 2005 00:33:45 +0000 (01:33 +0100)]
[PATCH] drivers/net/hamradio/dmascc.c: remove dmascc_setup()

It seems dmascc_setup() is a leftover time before dmascc_init() was
there.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] drivers/net/e1000/: possible cleanups
Adrian Bunk [Sun, 30 Oct 2005 15:53:34 +0000 (16:53 +0100)]
[PATCH] drivers/net/e1000/: possible cleanups

This patch contains the following possible cleanups:
- make needlessly global code static
- #if 0 the following unused global functions:
  - e1000_hw.c: e1000_mc_addr_list_update
  - e1000_hw.c: e1000_read_reg_io
  - e1000_hw.c: e1000_enable_pciex_master

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] drivers/net/ixgb/: make some code static
Adrian Bunk [Sun, 30 Oct 2005 15:53:30 +0000 (16:53 +0100)]
[PATCH] drivers/net/ixgb/: make some code static

This patch makes some needlessly global code static.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] reset tss->io_bitmap_owner in sys_ioperm()
Bart Oldeman [Sat, 5 Nov 2005 23:54:07 +0000 (12:54 +1300)]
[PATCH] reset tss->io_bitmap_owner in sys_ioperm()

my patch "x86: initialise tss->io_bitmap_owner to something" (commit ID
d5cd4aadd3d220afac8e3e6d922e333592551f7d) introduced a problem with a
program (DOSEMU) that called ioperm after already doing some port i/o.

The problem is that a process switch return causes tss->io_bitmap_base
to be set to IO_BITMAP_OFFSET so that the fault (that *really* sets the
io bitmap) never triggers.

This fixes that regression.

Signed-off-by: Bart Oldeman <bartoldeman@users.sourceforge.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Set the vga cursor even when hidden
Samuel Thibault [Sat, 5 Nov 2005 21:19:50 +0000 (22:19 +0100)]
[PATCH] Set the vga cursor even when hidden

Some visually impaired people use hardware devices which directly read
the vga screen. When newt for instance asks to hide the cursor for
better visual aspect, the kernel puts the vga cursor out of the screen,
so that the cursor position can't be read by the hardware device. This
is a great loss for such people.

Here is a patch which uses the same technique as CUR_NONE for hiding the
cursor while still moving it.

Mario, you should apply it to the speakup kernel for access floppies
asap. I'll submit a 2.4 patch too.

Signed-off-by: samuel.thibault@ens-lyon.org
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[DRIVER MODEL] Fix sgivwfb
Russell King [Sat, 5 Nov 2005 21:22:39 +0000 (21:22 +0000)]
[DRIVER MODEL] Fix sgivwfb

Statically allocated devices in module data is a potential cause
of oopsen.  The device may be in use by a userspace process, which
will keep a reference to the device.  If the module is unloaded,
the module data will be freed.  Subsequent use of the platform
device will cause a kernel oops.

Use generic platform device allocation/release code in modules.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[DRIVER MODEL] Fix gbefb
Russell King [Sat, 5 Nov 2005 21:22:13 +0000 (21:22 +0000)]
[DRIVER MODEL] Fix gbefb

Statically allocated devices in module data is a potential cause
of oopsen.  The device may be in use by a userspace process, which
will keep a reference to the device.  If the module is unloaded,
the module data will be freed.  Subsequent use of the platform
device will cause a kernel oops.

Use generic platform device allocation/release code in modules.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[DRIVER MODEL] Fix arcfb
Russell King [Sat, 5 Nov 2005 21:21:38 +0000 (21:21 +0000)]
[DRIVER MODEL] Fix arcfb

Release code in driver modules is a potential cause of oopsen.
The device may be in use by a userspace process, which will keep
a reference to the device.  If the module is unloaded, the module
text will be freed.  Subsequently, when the last reference is
dropped, the release code will be called, which no longer exists.

Use generic platform device allocation/release code in modules.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[DRIVER MODEL] Fix macsonic
Russell King [Sat, 5 Nov 2005 21:21:10 +0000 (21:21 +0000)]
[DRIVER MODEL] Fix macsonic

Release code in driver modules is a potential cause of oopsen.
The device may be in use by a userspace process, which will keep
a reference to the device.  If the module is unloaded, the module
text will be freed.  Subsequently, when the last reference is
dropped, the release code will be called, which no longer exists.

Use generic platform device allocation/release code in modules.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[DRIVER MODEL] Fix jazzsonic
Russell King [Sat, 5 Nov 2005 21:20:47 +0000 (21:20 +0000)]
[DRIVER MODEL] Fix jazzsonic

Release code in driver modules is a potential cause of oopsen.
The device may be in use by a userspace process, which will keep
a reference to the device.  If the module is unloaded, the module
text will be freed.  Subsequently, when the last reference is
dropped, the release code will be called, which no longer exists.

Use generic platform device allocation/release code in modules.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[DRIVER MODEL] Fix depca
Russell King [Sat, 5 Nov 2005 21:20:21 +0000 (21:20 +0000)]
[DRIVER MODEL] Fix depca

Release code in driver modules is a potential cause of oopsen.
The device may be in use by a userspace process, which will keep
a reference to the device.  If the module is unloaded, the module
text will be freed.  Subsequently, when the last reference is
dropped, the release code will be called, which no longer exists.

Use generic platform device allocation/release code in modules.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[DRIVER MODEL] Improved dynamically allocated platform_device interface
Russell King [Sat, 5 Nov 2005 21:19:33 +0000 (21:19 +0000)]
[DRIVER MODEL] Improved dynamically allocated platform_device interface

Re-jig the simple platform device support to allow private data
to be attached to a platform device, as well as allowing the
parent device to be set.

Example usage:

pdev = platform_device_alloc("mydev", id);
if (pdev) {
err = platform_device_add_resources(pdev, &resources,
    ARRAY_SIZE(resources));
if (err == 0)
err = platform_device_add_data(pdev, &platform_data,
       sizeof(platform_data));
if (err == 0)
err = platform_device_add(pdev);
} else {
err = -ENOMEM;
}
if (err)
platform_device_put(pdev);

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years agoMerge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik...
Linus Torvalds [Sat, 5 Nov 2005 20:04:02 +0000 (12:04 -0800)]
Merge branch 'upstream-linus' of /linux/kernel/git/jgarzik/netdev-2.6

18 years ago[PATCH] phy address mask support for generic phy layer
Matt Porter [Wed, 2 Nov 2005 23:13:06 +0000 (16:13 -0700)]
[PATCH] phy address mask support for generic phy layer

Adds a phy_mask field to struct mii_bus and uses it.  This field
indicates each phy address to be ignored when probing the mdio bus.

This support is needed for the fs_enet and ibm_emac drivers to be
converted to the generic phy layer among other drivers. Many systems
lock up on probing certain phy addresses or probing doesn't return
0xffff when nothing is found at the address. A new driver I'm
working on also makes use of this mask.

Signed-off-by: Matt Porter <mporter@kernel.crashing.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] pcnet32: Prevent hang with 79c976
Don Fry [Tue, 1 Nov 2005 21:13:35 +0000 (13:13 -0800)]
[PATCH] pcnet32: Prevent hang with 79c976

Some boards using the 79c976 pcnet32 chip will hang the system if the
ethtool --register-dump is performed with the device operational.  The
request to read bcr30 is retried by the PCI device infinitely without
returning data, hanging the system.

Tested ia32 and ppc64.

Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] pcnet32: AT2700/2701 and Bugzilla 2699 & 4551
Don Fry [Tue, 1 Nov 2005 20:50:57 +0000 (12:50 -0800)]
[PATCH] pcnet32: AT2700/2701 and Bugzilla 2699 & 4551

This patch is a better fix for Allied Telesyn 2700/2701 FX boards than
the change made in early January this year.  It allows the user to
select the speed/duplex via module_param, but if no selection is made,
forces the speed to 100 FD.  It fixes both Bugzilla bugs 2669 and 4551.
Tested ia32 and ppc64 by myself, and by the originator of bug 2669.

Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] pcnet32: show name of failing device
Don Fry [Tue, 1 Nov 2005 20:04:33 +0000 (12:04 -0800)]
[PATCH] pcnet32: show name of failing device

Display the name eth%d or pci_name() of device which fails to allocate
memory.  When changing ring size via ethtool, it also releases the
lock before returning on error.  Added comment that the caller of
pcnet32_alloc_ring must call pcnet32_free_ring on error, to avoid leak.
Tested ia32 by forcing allocation errors.

Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] S2io: Updated documentation
Ravinandan Arakali [Tue, 1 Nov 2005 20:24:25 +0000 (15:24 -0500)]
[PATCH] S2io: Updated documentation

Hi,
This patch provides updated documentation on the Neterion(S2io) driver.

Please review the patch.

Signed-off-by: Ravinandan Arakali <ravinandan.arakali@neterion.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] S2io: Multi buffer mode support
Ananda Raju [Mon, 31 Oct 2005 21:55:31 +0000 (16:55 -0500)]
[PATCH] S2io: Multi buffer mode support

Hi,
This patch  provides dynamic two buffer-mode and 3 buffer-mode options.
Previously 2 buffer-mode was compilation option. Now with this patch applied
one can load driver in 2 buffer-mode with module-load parameter

ie.
#insmod s2io.ko rx_ring_mode=2

This patch also provides 3 buffer-mode which provides header separation
functionality. In 3 buffer-mode skb->data will have L2/L3/L4 headers and
"skb_shinfo(skb)->frag_list->data" will have have L4 payload.
one can load driver in 3 buffer-mode with same above module-load parameter

ie.
#insmod s2io.ko rx_ring_mode=3

Please review the patch.

Signed-off-by: Ananda Raju <ananda.raju@neterion.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years ago[PATCH] drivers/net/wireless/airo.c unsigned comparason
Gabriel A. Devenyi [Fri, 4 Nov 2005 00:30:47 +0000 (19:30 -0500)]
[PATCH] drivers/net/wireless/airo.c unsigned comparason

fid is declared as a u32 (unsigned int), and then a few lines later, it is checked for a value < 0, which is clearly useless.
In the two locations this function is used, in one it is *explicitly* given a negative number, which would be ignored with the
current definition.

Thanks to LinuxICC (http://linuxicc.sf.net).

Signed-off-by: Gabriel A. Devenyi <ace@staticwave.ca>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
18 years agoMerge branch 'master'
Jeff Garzik [Sat, 5 Nov 2005 19:38:55 +0000 (14:38 -0500)]
Merge branch 'master'

18 years ago[libata] restore sg on DMA mapping failure
Tejun Heo [Sat, 5 Nov 2005 19:29:01 +0000 (14:29 -0500)]
[libata] restore sg on DMA mapping failure

18 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-for-linus-2.6
Linus Torvalds [Sat, 5 Nov 2005 18:31:35 +0000 (10:31 -0800)]
Merge /linux/kernel/git/jejb/scsi-for-linus-2.6

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-mmc
Linus Torvalds [Sat, 5 Nov 2005 17:47:46 +0000 (09:47 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-mmc

18 years ago[SCSI] sd: Fix refcounting
Alan Stern [Fri, 4 Nov 2005 19:44:41 +0000 (14:44 -0500)]
[SCSI] sd: Fix refcounting

Currently the driver takes a reference only for requests coming by way
of the gendisk, not for requests coming by way of the struct device or
struct scsi_device.  Such requests can arrive in the rescan, flush,
and shutdown pathways.

The patch also makes the scsi_disk keep a reference to the underlying
scsi_device, and it erases the scsi_device's pointer to the scsi_disk
when the scsi_device is removed (since the pointer should no longer be
used).

This resolves Bugzilla entry #5237.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
18 years ago[MMC] Use controller id instead of driver name for printks
Pierre Ossman [Sat, 5 Nov 2005 10:36:35 +0000 (10:36 +0000)]
[MMC] Use controller id instead of driver name for printks

The printks that aren't for debugging should use the name of the controller,
not the driver name. Multiple MMC controllers aren't that common today, but
this is the right way to do things.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[ARM] Documentation/arm/README: small update
Adrian Bunk [Sat, 5 Nov 2005 10:20:56 +0000 (10:20 +0000)]
[ARM] Documentation/arm/README: small update

- egcs is not supported by kernel 2.6
- gcc 3.3 seems to be a good choice on ARM

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[MMC] Fix chip config in wbsd
Pierre Ossman [Sat, 5 Nov 2005 10:16:50 +0000 (10:16 +0000)]
[MMC] Fix chip config in wbsd

There is a broken if clause in the wbsd driver that can cause the
driver to try and configure the chip even though none is found. This
results in i/o on invalid ports.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years agoFix ips.c compile
James Bottomley [Sat, 5 Nov 2005 05:22:55 +0000 (23:22 -0600)]
Fix ips.c compile

It looks like one of the ips patches was missing a closing brace in a
function

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
18 years agoMerge by Hand
James Bottomley [Sat, 5 Nov 2005 04:29:52 +0000 (22:29 -0600)]
Merge by Hand

Conflicts in dec_esp.c (Thanks Bacchus), scsi_transport_iscsi.c and
scsi_transport_fc.h

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
18 years ago[libata] ATAPI pad allocation fixes/cleanup
Jeff Garzik [Sat, 5 Nov 2005 03:08:00 +0000 (22:08 -0500)]
[libata] ATAPI pad allocation fixes/cleanup

Use ata_pad_{alloc,free} in two drivers, to factor out common code.

Add ata_pad_{alloc,free} to two other drivers, which needed the padding
but had not been updated.

18 years agoMerge branch 'master'
Jeff Garzik [Sat, 5 Nov 2005 02:39:31 +0000 (21:39 -0500)]
Merge branch 'master'

18 years ago[PATCH] nvidiafb: Geforce 7800 series support added
Calin A. Culianu [Sat, 5 Nov 2005 01:38:04 +0000 (20:38 -0500)]
[PATCH] nvidiafb: Geforce 7800 series support added

This adds support for the Nvidia Geforce 7800 series of cards to the
nvidiafb framebuffer driver.  All it does is add the PCI device id for
the 7800, 7800 GTX, 7800 GO, and 7800 GTX GO cards to the module device
table for the nvidiafb.ko driver, so that nvidiafb.ko will actually work
on these cards.

I also added the relevant PCI device ids to linux/pci_ids.h

I tested it on my 7800 GTX here and it works like a charm.  I now can
get framebuffer support on this card! Woo hoo!! Nothing like 200x75 text
mode to make your eyes BLEED.  ;)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years agoMerge branch 'srp' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband
Linus Torvalds [Sat, 5 Nov 2005 00:32:36 +0000 (16:32 -0800)]
Merge branch 'srp' of /linux/kernel/git/roland/infiniband

18 years agoMerge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband
Linus Torvalds [Sat, 5 Nov 2005 00:31:54 +0000 (16:31 -0800)]
Merge branch 'for-linus' of /linux/kernel/git/roland/infiniband

18 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/paulus/powerpc-merge
Linus Torvalds [Sat, 5 Nov 2005 00:27:50 +0000 (16:27 -0800)]
Merge /linux/kernel/git/paulus/powerpc-merge

18 years agopowerpc: Fix vmlinux.lds.S for 32-bit
Paul Mackerras [Fri, 4 Nov 2005 23:36:59 +0000 (10:36 +1100)]
powerpc: Fix vmlinux.lds.S for 32-bit

We can't currently use asm-ppc/page.h in vmlinux.lds.S, so until
we have a merged page.h, define PAGE_SIZE and KERNELBASE locally.
Also gets rid of some dynamic executable cruft that we had for
32-bit.  With -Ttext=$(KERNELBASE) this didn't cause any problem,
but when we changed to putting . = KERNELBASE in the vmlinux.lds.S
this cruft caused the text to get linked at 0xa0 instead of
0xc0000000.  Oops.

Signed-off-by: Paul Mackerras <paulus@samba.org>
18 years agopowerpc: Merge smp.c and smp.h
Paul Mackerras [Fri, 4 Nov 2005 23:33:55 +0000 (10:33 +1100)]
powerpc: Merge smp.c and smp.h

This also moves setup_cpu_maps to setup-common.c (calling it
smp_setup_cpu_maps) and uses it on both 32-bit and 64-bit.

Signed-off-by: Paul Mackerras <paulus@samba.org>
18 years agoNFS,SUNRPC,NLM: fix unused variable warnings when CONFIG_SYSCTL is disabled
Chuck Lever [Tue, 1 Nov 2005 21:53:32 +0000 (16:53 -0500)]
NFS,SUNRPC,NLM: fix unused variable warnings when CONFIG_SYSCTL is disabled

 Fix some dprintk's so that NLM, NFS client, and RPC client compile
 cleanly if CONFIG_SYSCTL is disabled.

 Test plan:
 Compile kernel with CONFIG_NFS enabled and CONFIG_SYSCTL disabled.

Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoSUNRPC: allow sunrpc.o to link when CONFIG_SYSCTL is disabled
Chuck Lever [Tue, 1 Nov 2005 17:24:48 +0000 (12:24 -0500)]
SUNRPC: allow sunrpc.o to link when CONFIG_SYSCTL is disabled

 The sunrpc module should build properly even when CONFIG_SYSCTL is
 disabled.

 Reported by Jan-Benedict Glaw.

 Test plan:
 Compile kernel with CONFIG_NFS as a module and built-in, and CONFIG_SYSCTL
 enabled and disabled.

Signed-off-by: Chuck Lever <cel@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Teach NFSv4 to cache locks when we hold a delegation
Trond Myklebust [Fri, 4 Nov 2005 20:39:36 +0000 (15:39 -0500)]
NFSv4: Teach NFSv4 to cache locks when we hold a delegation

 Now that we have a method of dealing with delegation recalls, actually
 enable the caching of posix and BSD locks.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Recover locks too when returning a delegation
Trond Myklebust [Fri, 4 Nov 2005 20:38:11 +0000 (15:38 -0500)]
NFSv4: Recover locks too when returning a delegation

 Delegations allow us to cache posix and BSD locks, however when the
 delegation is recalled, we need to "flush the cache" and send
 the cached LOCK requests to the server.

 This patch sets up the mechanism for doing so.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Fix recovery of flock() locks.
Trond Myklebust [Fri, 4 Nov 2005 20:35:30 +0000 (15:35 -0500)]
NFSv4: Fix recovery of flock() locks.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Return any delegations before sillyrenaming the file
Trond Myklebust [Fri, 4 Nov 2005 20:35:02 +0000 (15:35 -0500)]
NFSv4: Return any delegations before sillyrenaming the file

 I missed this one... Any form of rename will result in a delegation
 recall, so it is more efficient to return the one we hold before
 trying the rename.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Fix the handling of the error NFS4ERR_OLD_STATEID
Trond Myklebust [Fri, 4 Nov 2005 20:33:50 +0000 (15:33 -0500)]
NFSv4: Fix the handling of the error NFS4ERR_OLD_STATEID

 Ensure that we retry the failed operation...

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Fix problem with OPEN_DOWNGRADE
Trond Myklebust [Fri, 4 Nov 2005 20:33:38 +0000 (15:33 -0500)]
NFSv4: Fix problem with OPEN_DOWNGRADE

 RFC 3530 states that for OPEN_DOWNGRADE "The share_access and share_deny
 bits specified must be exactly equal to the union of the share_access and
 share_deny bits specified for some subset of the OPENs in effect for
 current openowner on the current file.

 Setattr is currently violating the NFSv4 rules for OPEN_DOWNGRADE in that
 it may cause a downgrade from OPEN4_SHARE_ACCESS_BOTH to
 OPEN4_SHARE_ACCESS_WRITE despite the fact that there exists no open file
 with O_WRONLY access mode.

 Fix the problem by replacing nfs4_find_state() with a modified version of
 nfs_find_open_context().

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years agoNFSv4: Fix a race between open() and close()
Trond Myklebust [Fri, 4 Nov 2005 20:32:58 +0000 (15:32 -0500)]
NFSv4: Fix a race between open() and close()

 We must not remove the nfs4_state structure from the inode open lists
 before we are in sequence lock.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
18 years ago[USB]: Make early handoff a final fixup instead of a header one.
David S. Miller [Fri, 4 Nov 2005 19:17:24 +0000 (11:17 -0800)]
[USB]: Make early handoff a final fixup instead of a header one.

At header fixup time, it is not yet legal to ioremap() PCI
device registers, yet that is what this quirk code needs to
do.

Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[PATCH] improve scheduler fairness a bit
Oleg Nesterov [Fri, 4 Nov 2005 15:54:30 +0000 (18:54 +0300)]
[PATCH] improve scheduler fairness a bit

Do not transfer remaining time slice to another cpu on process exit.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[MMC] Response to write commands is R1 nor R1b
Russell King [Fri, 4 Nov 2005 18:43:39 +0000 (18:43 +0000)]
[MMC] Response to write commands is R1 nor R1b

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-serial
Linus Torvalds [Fri, 4 Nov 2005 18:42:53 +0000 (10:42 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-serial