Randy Dunlap [Fri, 28 Mar 2008 21:22:52 +0000 (14:22 -0700)]
PM: Remove legacy PM (fix)
Source file was removed. Need to remove docbook reference also.
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Cc: Pavel Machek <pavel@suse.cz> Cc: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Robert P. J. Day [Fri, 14 Mar 2008 02:41:52 +0000 (22:41 -0400)]
SYSFS: Explicitly include required header file slab.h.
After an experimental deletion of the unnecessary inclusion of
<linux/slab.h> from the header file <linux/percpu.h>, the following
files under fs/sysfs were exposed as needing to explicitly include
<linux/slab.h>.
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Driver core: make device_is_registered() work for class devices
device_is_registered() can use the kobject value for this, so it will
now work with devices that are associated with only a class, not a bus
and a driver.
Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Stern [Wed, 19 Mar 2008 21:39:13 +0000 (22:39 +0100)]
PM: Convert wakeup flag accessors to inline functions
This patch (as1058) improves the wakeup macros in include/linux/pm.h.
All but the trivial ones are converted to inline routines, which
requires moving them to a separate header file since they depend on
the definition of struct device.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Stern [Wed, 19 Mar 2008 21:37:42 +0000 (22:37 +0100)]
PM: Make wakeup flags available whenever CONFIG_PM is set
The various wakeup flags and their accessor macros in struct
dev_pm_info should be available whenever CONFIG_PM is enabled, not
just when CONFIG_PM_SLEEP is on. Otherwise remote wakeup won't always
be configurable for runtime power management. This patch (as1056b)
fixes the oversight.
David Brownell adds:
More accurately, fixes the "regression" ... as noted sometime
last summer, after 296699de6bdc717189a331ab6bbe90e05c94db06
introduced CONFIG_SUSPEND. But that didn't make the regression
list for that kernel, ergo the delay in fixing it.
[rjw: rebased]
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Stern [Wed, 19 Mar 2008 21:35:13 +0000 (22:35 +0100)]
PM: Fix misuse of wakeup flag accessors in serial core
This patch (as1059) fixes a mistake in the way the serial core
initializes a device's wakeup settings. It should use the accessor
routine instead of relying on a macro producing an lvalue.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Driver core: Call device_pm_add() after bus_add_device() in device_add()
Include dpm_sysfs_add() into device_pm_add(), in analogy with
device_pm_remove(), and modify device_add() to call the latter after
bus_add_device(), to avoid situations in which the PM core may
attempt to suspend a device the registration of which has not been
successful.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
PM: Handle device registrations during suspend/resume
Modify the PM core to protect its data structures, specifically the
dpm_active list, from being corrupted if a child of the currently
suspending device is registered concurrently with its ->suspend()
callback. In that case, since the new device (the child) is added
to dpm_active after its parent, the PM core will attempt to
suspend it after the parent, which is wrong.
Introduce a new member of struct dev_pm_info, called 'sleeping',
and use it to check if the parent of the device being added to
dpm_active has been suspended, in which case the device registration
fails. Also, use 'sleeping' for checking if the ordering of devices
on dpm_active is correct.
Introduce variable 'all_sleeping' that will be set to 'true' once all
devices have been suspended and make new device registrations fail
until 'all_sleeping' is reset to 'false', in order to avoid having
unsuspended devices around while the system is going into a sleep state.
Remove pm_sleep_rwsem which is not necessary any more.
Special thanks to Alan Stern for discussions and suggestions that
lead to the creation of this patch.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Kay Sievers [Sun, 9 Mar 2008 20:26:02 +0000 (21:26 +0100)]
block: send disk "change" event for rescan_partitions()
Userspace likes to get notified that the disk may have changed, when
rescan_partitions() is called after partitioning or media change. It will
make it possible to update the state of the disk with the "change" event,
before the following partition "add" events are handled.
Cc: David Zeuthen <david@fubar.dk> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ben Dooks [Tue, 4 Mar 2008 23:09:06 +0000 (15:09 -0800)]
sysdev: detect multiple driver registrations
I've just found how easy it is to accidentally register a sysdev_driver for
two different classes. When this happens, bad things happen as the
sysdev_driver structure keeps has the list entry for the driver
registration.
The following patch makes a WARN_ON() if this happens, although I think
BUG_ON or returning -EAGAIN could also be valid responses to this.
Signed-off-by: Ben Dooks <ben@fluff.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Sam Ravnborg [Tue, 4 Mar 2008 23:09:05 +0000 (15:09 -0800)]
driver core: cpu: fix section mismatch in cpu.c:store_online
Fix following warning:
WARNING: vmlinux.o(.text+0x64609c): Section mismatch in reference from the function store_online() to the function .cpuinit.text:cpu_up()
store_online() is defined inside a HOTPLUG_CPU block so references are OK.
Ignore references by annotating store_online() with __ref.
Note: This is needed because cpu_up() most likely should not have been
__cpuinit but all the hotplug cpu code misuses the __cpuinit annotation.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Kay Sievers [Wed, 20 Feb 2008 23:35:54 +0000 (00:35 +0100)]
PNP: add all PNP card device id's as individual aliases
The current PNP combined card + devices module aliase can
never ever match anything, because these values are not available
all at the same time to request a module.
Instead of adding the combined alias, we add the device id's
all as individual aliases. Device id's are exported by the PNP
bus and can now properly used to request the loading of a
matching module.
With this patch it exports only the device id's, as properly
matchable aliases:
alias: pnp:dCTL0070*
alias: pnp:dCTL0045*
alias: pnp:dCTL0023*
alias: pnp:dCTL0044*
alias: pnp:dCTL0022*
alias: pnp:dCTL0042*
alias: pnp:dCTL0041*
alias: pnp:dCTL0021*
alias: pnp:dCTL0031*
Now, the exported value of the PNP bus can be used to autoload
a matching module:
$ modprobe --first-time -n -v pnp:dCTL0045
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/core/snd-rawmidi.ko
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/drivers/mpu401/snd-mpu401-uart.ko
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/core/snd-hwdep.ko
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/isa/sb/snd-sb-common.ko
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/isa/sb/snd-sb16-csp.ko
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/isa/sb/snd-sb16-dsp.ko
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/drivers/opl3/snd-opl3-lib.ko
insmod /lib/modules/2.6.24-rc6-g5b825ed2-dirty/kernel/sound/isa/sb/snd-sbawe.ko
David Rientjes [Wed, 20 Feb 2008 01:39:02 +0000 (17:39 -0800)]
sysfs: small header file cleanup for SYSFS=n
Convert sysfs_remove_bin_file() to have a return type of 'void' for
!CONFIG_SYSFS configurations. Also removes unnecessary colons from empty
void functions.
Signed-off-by: David Rientjes <rientjes@google.com> Reviewed-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When DEBUG is not defined, pr_debug and dev_dbg and some
other local debugging functions are specified as:
"inline __attribute__((format (printf, x, y)))"
This is done to validate printk arguments when not debugging.
Converting these functions to macros or statement expressions
"do { if (0) printk(fmt, ##arg); } while (0)"
or
"({ if (0) printk(fmt, ##arg); 0; })
makes at least gcc 4.2.2 produce smaller objects.
This has the additional benefit of allowing the optimizer to
avoid calling functions like print_mac that might have been
arguments to the printk.
defconfig x86 current:
$ size vmlinux
text data bss dec hex filename 4716770 474560 618496 5809826 58a6a2 vmlinux
all converted: (More patches follow)
$ size vmlinux
text data bss dec hex filename 4716642 474560 618496 5809698 58a622 vmlinux
Even kernel/sched.o, which doesn't even use these
functions, becomes smaller.
It appears that merely having an indirect include
of <linux/device.h> can cause bigger objects.
$ size sched.inline.o sched.if0.o
text data bss dec hex filename
31385 2854 328 34567 8707 sched.inline.o
31366 2854 328 34548 86f4 sched.if0.o
The current preprocessed only kernel/sched.i file contains:
Badari Pulavarty [Mon, 11 Feb 2008 17:23:18 +0000 (09:23 -0800)]
driver core: register_memory/unregister_memory clean ups and bugfix
register_memory()/unregister_memory() never gets called with
"root". unregister_memory() is accessing kobject_name of
the object just freed up. Since no one uses the code,
lets take the code out. And also, make register_memory() static.
Another bug fix - before calling unregister_memory()
remove_memory_block() gets a ref on kobject. unregister_memory()
need to drop that ref before calling sysdev_unregister().
Ben Nizette [Thu, 13 Mar 2008 11:27:30 +0000 (22:27 +1100)]
UIO: Implement a UIO interface for the SMX Cryptengine
This patch implements a UIO interface for the SMX Cryptengine.
The Cryptengine found on the Nias Digital SMX board is best suited
for a UIO interface. It is not wired in to the cryptographic API
as the engine handles it's own keys, algorithms, everything. All
that we know about is that if there's room in the buffer, you can
write data to it and when there's data ready, you read it out again.
There isn't necessarily even any direct correlation between data
going in and data coming out again, the engine may consume or
generate data all on its own.
This driver is for proprietary hardware but we're always told to
submit the drivers anyway; here you are. :-)
This is version 4 of this patch and addresses all issues raised by
Hans-Jürgen Koch and Paul Mundt in their reviews. Slightly altered
is Paul's suggestion to use DRV_NAME and DRV_VERSION as the UIO
version and name. While at the moment they are the same, there
is no reason for them to stay that way. Nevertheless we now at
least provide a MODULE_VERSION macro to keep modinfo happy.
Signed-off-by: Ben Nizette <bn@niasdigital.com> Acked-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Hans J Koch <hjk@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Bernard Pidoux [Sun, 20 Apr 2008 01:41:51 +0000 (18:41 -0700)]
rose: Socket lock was not released before returning to user space
================================================
[ BUG: lock held when returning to user space! ]
------------------------------------------------
xfbbd/3683 is leaving the kernel with locks still held!
1 lock held by xfbbd/3683:
#0: (sk_lock-AF_ROSE){--..}, at: [<c8cd1eb3>] rose_connect+0x73/0x420 [rose]
Pavel Machek [Sun, 20 Apr 2008 01:17:26 +0000 (18:17 -0700)]
hci_usb: remove code obfuscation
_urb_free is an alias for kfree... making code longer & harder to
read. Remove it.
Signed-off-by: Pavel Machek <pavel@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Pavel Machek [Sun, 20 Apr 2008 01:13:40 +0000 (18:13 -0700)]
hci_usb: do not initialize static variables to 0
hci_usb: do not initialize static variables to 0.
Signed-off-by: Pavel Machek <pavel@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Matt Carlson [Sun, 20 Apr 2008 01:12:33 +0000 (18:12 -0700)]
tg3: 5701 DMA corruption fix
Herbert Xu's commit fb93134dfc2a6e6fbedc7c270a31da03fce88db9, entitled
"[TCP]: Fix size calculation in sk_stream_alloc_pskb", has triggered a
bug in the 5701 where the 5701 DMA engine will corrupt outgoing
packets. This problem only happens when the starting address of the
packet matches a certain range of offsets and only when the 5701 is
placed downstream of a particular Intel bridge.
This patch detects the problematic bridge and if present, readjusts the
starting address of the packet data to a dword aligned boundary.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mark Asselstine [Sun, 20 Apr 2008 01:10:46 +0000 (18:10 -0700)]
atm nicstar: Removal of debug code containing deprecated calls to cli()/sti()
Code within NS_DEBUG_SPINLOCKS contained deprecated cli()/sti()
function calls. NS_DEBUG_SPINLOCKS and the associated code seems to
be of little use these days so the strategy of removing this code
rather then updating it to use spinlocks has been taken.
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com> Reviewed-by: Matthew Wilcox <willy@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sun, 20 Apr 2008 01:09:39 +0000 (18:09 -0700)]
iwlwifi: Fix unconditional access to station->tidp[].agg.
Reportred by Ingo Molnar:
drivers/net/wireless/iwlwifi/iwl-debugfs.c: In function 'iwl_dbgfs_stations_read':
drivers/net/wireless/iwlwifi/iwl-debugfs.c:256: error: 'struct iwl4965_tid_data' has no member named 'agg'
Needs CONFIG_IWL4965_HT protection.
Signed-off-by: David S. Miller <davem@davemloft.net>
Adrian Bunk [Mon, 14 Apr 2008 18:42:20 +0000 (21:42 +0300)]
avr32: add include/asm-avr32/serial.h
On Mon, Apr 14, 2008 at 03:36:24PM +0100, Alan Cox wrote:
> On Mon, 14 Apr 2008 17:17:21 +0300
> Adrian Bunk <bunk@kernel.org> wrote:
>
> > This patch fixes the following build error:
> >
> > <-- snip -->
> >
> > ...
> > CC [M] drivers/serial/8250.o
> > /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/serial/8250.c:95:24: error: asm/serial.h: No such file or directory
> > make[3]: *** [drivers/serial/8250.o] Error 1
> >
> > <-- snip -->
> >
> > Signed-off-by: Adrian Bunk <bunk@kernel.org>
> >
> > ---
> > 3cb4ef80d75e118ccfd44f7006aea3db54afb31c diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
> > index b1bbaa0..b0e216d 100644
> > --- a/drivers/serial/Kconfig
> > +++ b/drivers/serial/Kconfig
> > @@ -11,7 +11,7 @@ menu "Serial drivers"
> > # The new 8250/16550 serial drivers
> > config SERIAL_8250
> > tristate "8250/16550 and compatible serial support"
> > - depends on (BROKEN || !SPARC)
> > + depends on (BROKEN || !SPARC) && !AVR32
> > select SERIAL_CORE
> > ---help---
>
> NAK.
>
> Add an asm/serial.h to the platform as it has PCI so will have 8250 PCI
> devices available to it. A copy of the MIPS one should be right.
Patch below.
> Alan
cu
Adrian
<-- snip -->
This patch fixes the following build error with CONFIG_SERIAL_8250:
<-- snip -->
...
CC [M] drivers/serial/8250.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/serial/8250.c:95:24: error: asm/serial.h: No such file or directory
make[3]: *** [drivers/serial/8250.o] Error 1
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Adrian Bunk [Mon, 14 Apr 2008 14:16:53 +0000 (17:16 +0300)]
avr32: don't offer PARPORT_PC
This patch fixes the following compile error:
<-- snip -->
...
CC [M] drivers/parport/parport_pc.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/parport/parport_pc.c:67:25: error: asm/parport.h: No such file or directory
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/parport/parport_pc.c: In function 'parport_pc_find_ports':
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/parport/parport_pc.c:3215: error: implicit declaration of function 'parport_pc_find_nonpci_ports'
make[3]: *** [drivers/parport/parport_pc.o] Error 1
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Adrian Bunk [Mon, 14 Apr 2008 14:16:44 +0000 (17:16 +0300)]
avr32: don't offer CONFIG_GEN_RTC
This patch fixes the following compile error:
<-- snip -->
...
CC [M] drivers/char/genrtc.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/char/genrtc.c:58:21: error: asm/rtc.h: No such file or directory
...
make[3]: *** [drivers/char/genrtc.o] Error 1
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Adrian Bunk [Mon, 14 Apr 2008 14:16:40 +0000 (17:16 +0300)]
avr32: don't offer CONFIG_RTC
This patch fixes the following compile error:
<-- snip -->
...
CC [M] drivers/char/rtc.o
In file included from /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/char/rtc.c:70:
/home/bunk/linux/kernel-2.6/git/linux-2.6/include/linux/mc146818rtc.h:16:59: error: asm/mc146818rtc.h: No such file or directory
...
make[3]: *** [drivers/char/rtc.o] Error 1
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Adrian Bunk [Mon, 14 Apr 2008 14:16:32 +0000 (17:16 +0300)]
add include/asm-avr32/xor.h
This patch fixes the following compile error with CONFIG_MD_RAID456
on avr32:
<-- snip -->
...
CC [M] crypto/xor.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/crypto/xor.c:23:21: error: asm/xor.h: No such file or directory
/home/bunk/linux/kernel-2.6/git/linux-2.6/crypto/xor.c: In function 'calibrate_xor_blocks':
/home/bunk/linux/kernel-2.6/git/linux-2.6/crypto/xor.c:131: error: 'XOR_TRY_TEMPLATES' undeclared (first use in this function)
/home/bunk/linux/kernel-2.6/git/linux-2.6/crypto/xor.c:131: error: (Each undeclared identifier is reported only once
/home/bunk/linux/kernel-2.6/git/linux-2.6/crypto/xor.c:131: error: for each function it appears in.)
make[2]: *** [crypto/xor.o] Error 1
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
avr32: Implement set_rate(), set_parent() and mode() for pll1
This patch is a take two of adding full functionality to PLL1 on
AT32AP7000. This allows board-specific code and drivers to configure
and enable PLL1. This is useful when precise control over the
frequency of e.g. a genclock is needed and requested by users for the
ABDAC device.
The patch is based upon previous patches from both Haavard Skinnemoen
and David Brownell.
David Brownell [Thu, 14 Feb 2008 19:24:02 +0000 (11:24 -0800)]
avr32: Generic clockevents support
This combines three patches from David Brownell:
* avr32: tclib support
* avr32: simplify clocksources
* avr32: Turn count/compare into a oneshot clockevent device
Register both TC blocks (instead of just the first one) so that
the AT32/AT91 tclib code will pick them up (instead of just the
avr32-only PIT-style clocksource).
Rename the first one and its resources appropriately.
More cleanups to the cycle counter clocksource code
- Disable all the weak symbol magic; remove the AVR32-only TCB-based
clocksource code (source and header).
- Mark the __init code properly.
- Don't forget to report IRQF_TIMER.
- Make the system work properly with this clocksource, by preventing
use of the CPU "idle" sleep state in the idle loop when it's used.
Package the avr32 count/compare timekeeping support as a oneshot
clockevent device, so it supports NO_HZ and high res timers.
This means it also supports plugging in other clockevent devices
and clocksources.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
David Brownell [Tue, 12 Feb 2008 22:45:49 +0000 (14:45 -0800)]
avr32: start clocksource cleanup
Start cleaning up the AVR32 clocksource mess, starting with the cycle
counter clocksource: remove unneeded pseudo-RTC (just inline that
call to mktime) and associated build warning, and unused sysdev.
Add comment about the problem using the cycle counter register,
and adjust the clocksource rating accordingly. Later patches can
make this usable again (by disabling use of the idle state and
providing a proper clocksource without the weak binding hacks)
and move towards TCB-based clockevent support (including high
resolution timers) that's shared between AT91 and AVR32.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Ben Nizette [Thu, 7 Feb 2008 04:28:57 +0000 (15:28 +1100)]
avr32: pass i2c board info through at32_add_device_twi
New-style I2C drivers require that motherboard-mounted I2C devices are
registered with the I2C core, typically at arch_initcall time. This
can be done nice and neat by passing the struct i2c_board_info[]
through at32_add_device_twi just like we do for the SPI board info.
While we've got the hood up, remove a duplicate declaration of
at32_add_device_twi() in board.h.
Cyrill Gorcunov [Sun, 3 Feb 2008 11:22:18 +0000 (14:22 +0300)]
avr32: cleanup - use _AC macro to define PAGE_SIZE
PAGE_SIZE is used both from assembly and C code. We want to have type
specifiers when using it from C, but this will make the assembler
confused, so we need to make it conditional.
This is exactly what the _AC macro is for, so using it allows us to
get rid of a few lines of cpp noise.
SUNRPC: Don't change the RPCSEC_GSS context on a credential that is in use
When a server rejects our credential with an AUTH_REJECTEDCRED or similar,
we need to refresh the credential and then retry the request.
However, we do want to allow any requests that are in flight to finish
executing, so that we can at least attempt to process the replies that
depend on this instance of the credential.
The solution is to ensure that gss_refresh() looks up an entirely new
RPCSEC_GSS credential instead of attempting to create a context for the
existing invalid credential.
SUNRPC: Don't disconnect more than once if retransmitting NFSv4 requests
NFSv4 requires us to ensure that we break the TCP connection before we're
allowed to retransmit a request. However in the case where we're
retransmitting several requests that have been sent on the same
connection, we need to ensure that we don't interfere with the attempt to
reconnect and/or break the connection again once it has been established.
We therefore introduce a 'connection' cookie that is bumped every time a
connection is broken. This allows requests to track if they need to force a
disconnection.
We need to try to ensure that we always use the same credentials whenever
we re-establish the clientid on the server. If not, the server won't
recognise that we're the same client, and so may not allow us to recover
state.
NFSv4: Don't use cred->cr_ops->cr_name in nfs4_proc_setclientid()
With the recent change to generic creds, we can no longer use
cred->cr_ops->cr_name to distinguish between RPCSEC_GSS principals and
AUTH_SYS/AUTH_NULL identities. Replace it with the rpc_authops->au_name
instead...
Fred Isaman [Wed, 2 Apr 2008 12:21:15 +0000 (15:21 +0300)]
nfs: fix printout of multiword bitfields
Benny points out that zero-padding of multiword bitfields is necessary,
and that delimiting each word is nice to avoid endianess confusion.
bhalevy: without zero padding output can be ambiguous. Also,
since the printed array of two 32-bit unsigned integers is not a
64-bit number, delimiting the output with a semicolon makes more sense.
Benny Halevy [Mon, 31 Mar 2008 14:39:06 +0000 (17:39 +0300)]
nfs: return negative error value from nfs{,4}_stat_to_errno
All use sites for nfs{,4}_stat_to_errno negate their return value.
It's more efficient to return a negative error from the stat_to_errno convertors
rather than negating its return value everywhere. This also produces slightly
smaller code.
NLM/lockd: Ensure client locking calls use correct credentials
Now that we've added the 'generic' credentials (that are independent of the
rpc_client) to the nfs_open_context, we can use those in the NLM client to
ensure that the lock/unlock requests are authenticated to whoever
originally opened the file.
Trond Myklebust [Fri, 28 Mar 2008 20:04:36 +0000 (16:04 -0400)]
NLM/lockd: convert __nlm_async_call to use rpc_run_task()
Peter Staubach comments:
> In the course of investigating testing failures in the locking phase of
> the Connectathon testsuite, I discovered a couple of things. One was
> that one of the tests in the locking tests was racy when it didn't seem
> to need to be and two, that the NFS client asynchronously releases locks
> when a process is exiting.
...
> The Single UNIX Specification Version 3 specifies that: "All locks
> associated with a file for a given process shall be removed when a file
> descriptor for that file is closed by that process or the process holding
> that file descriptor terminates.".
>
> This does not specify whether those locks must be released prior to the
> completion of the exit processing for the process or not. However,
> general assumptions seem to be that those locks will be released. This
> leads to more deterministic behavior under normal circumstances.
The following patch converts the NFSv2/v3 locking code to use the same
mechanism as NFSv4 for sending asynchronous RPC calls and then waiting for
them to complete. This ensures that the UNLOCK and CANCEL RPC calls will
complete even if the user interrupts the call, yet satisfies the
above request for synchronous behaviour on process exit.
NLM/lockd: Add a reference counter to struct nlm_rqst
When we replace the existing synchronous RPC calls with asynchronous calls,
the reference count will be needed in order to allow us to examine the
result of the RPC call.
Trond Myklebust [Fri, 21 Mar 2008 20:19:41 +0000 (16:19 -0400)]
SUNRPC: Fix read ordering problems with req->rq_private_buf.len
We want to ensure that req->rq_private_buf.len is updated before
req->rq_received, so that call_decode() doesn't use an old value for
req->rq_rcv_buf.len.
In 'call_decode()' itself, instead of using task->tk_status (which is set
using req->rq_received) must use the actual value of
req->rq_private_buf.len when deciding whether or not the received RPC reply
is too short.
Finally ensure that we set req->rq_rcv_buf.len to zero when retrying a
request. A typo meant that we were resetting req->rq_private_buf.len in
call_decode(), and then clobbering that value with the old rq_rcv_buf.len
again in xprt_transmit().
NFSv4: Only increment the sequence id if the server saw it
It is quite possible that the OPEN, CLOSE, LOCK, LOCKU,... compounds fail
before the actual stateful operation has been executed (for instance in the
PUTFH call). There is no way to tell from the overall status result which
operations were executed from the COMPOUND.
The fix is to move incrementing of the sequence id into the XDR layer,
so that we do it as we process the results from the stateful operation.
NFSv4: Remove bogus call to nfs4_drop_state_owner() in _nfs4_open_expired()
There should be no need to invalidate a perfectly good state owner just
because of a stale filehandle. Doing so can cause the state recovery code
to break, since nfs4_get_renew_cred() and nfs4_get_setclientid_cred() rely
on finding active state owners.
NFS: Fix nfs_wb_page() to always exit with an error or a clean page
It is possible for nfs_wb_page() to sometimes exit with 0 return value, yet
the page is left in a dirty state.
For instance in the case where the server rebooted, and the COMMIT request
failed, then all the previously "clean" pages which were cached by the
server, but were not guaranteed to have been writted out to disk,
have to be redirtied and resent to the server.
The fix is to have nfs_wb_page_priority() check that the page is clean
before it exits...
This fixes a condition that triggers the BUG_ON(PagePrivate(page)) in
nfs_create_request() when we're in the nfs_readpage() path.
Also eliminate a redundant BUG_ON(!PageLocked(page)) while we're at it. It
turns out that clear_page_dirty_for_io() has the exact same test.
SUNRPC: Don't attempt to destroy expired RPCSEC_GSS credentials..
..and always destroy using a 'soft' RPC call. Destroying GSS credentials
isn't mandatory; the server can always cope with a few credentials not
getting destroyed in a timely fashion.
This actually fixes a hang situation. Basically, some servers will decide
that the client is crazy if it tries to destroy an RPC context for which
they have sent an RPCSEC_GSS_CREDPROBLEM, and so will refuse to talk to it
for a while.
The regression therefor probably was introduced by commit 0df7fb74fbb709591301871a38aac7735a1d6583.
The rest of the networking layer uses SOCK_ASYNC_NOSPACE to signal whether
or not we have someone waiting for buffer memory. Convert the SUNRPC layer
to use the same idiom.
Remove the unlikely()s in xs_udp_write_space and xs_tcp_write_space. In
fact, the most common case will be that there is nobody waiting for buffer
space.
SOCK_NOSPACE is there to tell the TCP layer whether or not the cwnd was
limited by the application window. Ensure that we follow the same idiom as
the rest of the networking layer here too.
Finally, ensure that we clear SOCK_ASYNC_NOSPACE once we wake up, so that
write_space() doesn't keep waking things up on xprt->pending.
call_verify() can, under certain circumstances, free the RPC slot. In that
case, our cached pointer 'req = task->tk_rqstp' is invalid. Bug was
introduced in commit 220bcc2afd7011b3e0569fc178331fa983c92c1b (SUNRPC:
Don't call xprt_release in call refresh).
(a + b) / (c + d) != a / c + b / d. The old code errornously
assumed this incorrect formuld. Instead, just sum all erase
counters in a 64-bit variable and divide to the number of EBs
at the end.
Peter Zijlstra [Sat, 19 Apr 2008 17:45:00 +0000 (19:45 +0200)]
sched: rt-group: optimize dequeue_rt_stack
Now that the group hierarchy can have an arbitrary depth the O(n^2) nature
of RT task dequeues will really hurt. Optimize this by providing space to
store the tree path, so we can walk it the other way.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
projects / linux-2.6-omap-h63xx.git / log