11 years agoRevert "prohibit rcutorture from being compiled into the kernel"
Ingo Molnar [Mon, 16 Jun 2008 06:40:04 +0000]
Revert "prohibit rcutorture from being compiled into the kernel"

This reverts commit 9aaffc898ff4a3df18c5fc4b9e0fa47e779ad726.

That commit was a very bad idea. RCU_TORTURE found many boot timing
bugs and other sorts of bugs in the past, so excluding it from
boot images is very silly.

The option already depends on DEBUG_KERNEL and is disabled by default.
Even when it runs, the test threads are reniced. If it annoys people
we could add a runtime sysctl.

11 years agorcu: fix nf_conntrack_helper.c build bug
Ingo Molnar [Tue, 13 May 2008 13:37:05 +0000]
rcu: fix nf_conntrack_helper.c build bug

Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agorculist.h: fix include in net/netfilter/nf_conntrack_netlink.c
Ingo Molnar [Sat, 17 May 2008 06:26:25 +0000]
rculist.h: fix include in net/netfilter/nf_conntrack_netlink.c

this file has rculist dependency but did not explicitly include it,
which broke the build.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agorcu: remove duplicated include in kernel/rcupreempt.c
Huang Weiyi [Wed, 14 May 2008 23:23:00 +0000]
rcu: remove duplicated include in kernel/rcupreempt.c

Removed duplicated include file <linux/rcupdate.h> in kernel/rcupreempt.c.

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agorcu: remove duplicated include in kernel/rcupreempt_trace.c
Huang Weiyi [Wed, 14 May 2008 23:23:00 +0000]
rcu: remove duplicated include in kernel/rcupreempt_trace.c

Removed duplicated include file <linux/rcupdate.h> in kernel/rcupreempt_trace.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agoRCU, rculist.h: fix list iterators
Paul E. McKenney [Mon, 12 May 2008 19:21:06 +0000]
RCU, rculist.h: fix list iterators

RCU list iterators: should prefetch ever be optimised out with no
side-effects, the current version will lose the barrier completely.

Pointed-out-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agorcu: fix rcu_try_flip_waitack_needed() to prevent grace-period stall
Paul E. McKenney [Mon, 12 May 2008 19:21:06 +0000]
rcu: fix rcu_try_flip_waitack_needed() to prevent grace-period stall

The comment was correct -- need to make the code match the comment.
Without this patch, if a CPU goes dynticks idle (and stays there forever)
in just the right phase of preemptible-RCU grace-period processing,
grace periods stall.  The offending sequence of events (courtesy
of Promela/spin, at least after I got the liveness criterion coded
correctly...) is as follows:

o CPU 0 is in dynticks-idle mode.  Its dynticks_progress_counter
is (say) 10.

o CPU 0 takes an interrupt, so rcu_irq_enter() increments CPU 0's
dynticks_progress_counter to 11.

o CPU 1 is doing RCU grace-period processing in rcu_try_flip_idle(),
sees rcu_pending(), so invokes dyntick_save_progress_counter(),
which in turn takes a snapshot of CPU 0's dynticks_progress_counter
into CPU 0's rcu_dyntick_snapshot -- now set to 11.  CPU 1 then
updates the RCU grace-period state to rcu_try_flip_waitack().

o CPU 0 returns from its interrupt, so rcu_irq_exit() increments
CPU 0's dynticks_progress_counter to 12.

o CPU 1 later invokes rcu_try_flip_waitack(), which notices that
CPU 0 has not yet responded, and hence in turn invokes
rcu_try_flip_waitack_needed().  This function examines the
state of CPU 0's dynticks_progress_counter and rcu_dyntick_snapshot
variables, which it copies to curr (== 12) and snap (== 11),
respectively.

Because curr!=snap, the first condition fails.

Because curr-snap is only 1 and snap is odd, the second
condition fails.

rcu_try_flip_waitack_needed() therefore incorrectly concludes
that it must wait for CPU 0 to explicitly acknowledge the
counter flip.

o CPU 0 remains forever in dynticks-idle mode, never taking
any more hardware interrupts or any NMIs, and never running
any more tasks.  (Of course, -something- will usually eventually
happen, which might be why we haven't seen this one in the
wild.  Still should be fixed!)

Therefore the grace period never ends.  Fix is to make the code match
the comment, as shown below.  With this fix, the above scenario
would be satisfied with curr being even, and allow the grace period
to proceed.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Josh Triplett <josh@kernel.org>
Cc: Dipankar Sarma <dipankar@in.ibm.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agorculist.h: use the rcu API
Franck Bui-Huu [Mon, 12 May 2008 19:21:06 +0000]
rculist.h: use the rcu API

Make almost all list mutation primitives use rcu_assign_pointer().

The main point of this being readability improvement.

Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Josh Triplett <josh@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agorcu: split list.h and move rcu-protected lists into rculist.h
Franck Bui-Huu [Mon, 12 May 2008 19:21:05 +0000]
rcu: split list.h and move rcu-protected lists into rculist.h

Move rcu-protected lists from list.h into a new header file rculist.h.

This is done because list are a very used primitive structure all over the
kernel and it's currently impossible to include other header files in this
list.h without creating some circular dependencies.

For example, list.h implements rcu-protected list and uses rcu_dereference()
without including rcupdate.h.  It actually compiles because users of
rcu_dereference() are macros.  Others RCU functions could be used too but
aren't probably because of this.

Therefore this patch creates rculist.h which includes rcupdates without to
many changes/troubles.

Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Josh Triplett <josh@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agosched: 1Q08 RCU doc update, add call_rcu_sched()
Paul E. McKenney [Mon, 12 May 2008 19:21:05 +0000]
sched: 1Q08 RCU doc update, add call_rcu_sched()

Long-delayed update to the RCU documentation, including adding the new
call_rcu_sched() and rcu_barrier_sched() APIs.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

11 years agorcu: add call_rcu_sched() and friends to rcutorture
Paul E. McKenney [Mon, 12 May 2008 19:21:05 +0000]
rcu: add call_rcu_sched() and friends to rcutorture

Add entry to rcu_torture_ops allowing the correct barrier function to
be used upon exit from rcutorture.  Also add torture options for the
new call_rcu_sched() API.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

11 years agorcu: add rcu_barrier_sched() and rcu_barrier_bh()
Paul E. McKenney [Mon, 12 May 2008 19:21:05 +0000]
rcu: add rcu_barrier_sched() and rcu_barrier_bh()

Add rcu_barrier_sched() and rcu_barrier_bh().  With these in place,
rcutorture no longer gives the occasional oops when repeatedly starting
and stopping torturing rcu_bh.  Also adds the API needed to flush out
pre-existing call_rcu_sched() callbacks.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

11 years agorcu: add memory barriers and comments to rcu_check_callbacks()
Paul E. McKenney [Mon, 12 May 2008 19:21:05 +0000]
rcu: add memory barriers and comments to rcu_check_callbacks()

Add comments to the logic that infers quiescent states when interrupting
from either user mode or the idle loop.  Also add a memory barrier: it
appears that James Huang was in fact onto something, as the scheduler
is much less synchronization happy than it once was, so we can no longer
rely on its memory barriers in all cases.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reported-by: James Huang <jamesclhuang@yahoo.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

11 years agorcu: add call_rcu_sched()
Paul E. McKenney [Mon, 12 May 2008 19:21:05 +0000]
rcu: add call_rcu_sched()

Fourth cut of patch to provide the call_rcu_sched().  This is again to
synchronize_sched() as call_rcu() is to synchronize_rcu().

Should be fine for experimental and -rt use, but not ready for inclusion.
With some luck, I will be able to tell Andrew to come out of hiding on
the next round.

Passes multi-day rcutorture sessions with concurrent CPU hotplugging.

Fixes since the first version include a bug that could result in
indefinite blocking (spotted by Gautham Shenoy), better resiliency
against CPU-hotplug operations, and other minor fixes.

Fixes since the second version include reworking grace-period detection
to avoid deadlocks that could happen when running concurrently with
CPU hotplug, adding Mathieu's fix to avoid the softlockup messages,
as well as Mathieu's fix to allow use earlier in boot.

Fixes since the third version include a wrong-CPU bug spotted by
Andrew, getting rid of the obsolete synchronize_kernel API that somehow
snuck back in, merging spin_unlock() and local_irq_restore() in a
few places, commenting the code that checks for quiescent states based
on interrupting from user-mode execution or the idle loop, removing
some inline attributes, and some code-style changes.

Known/suspected shortcomings:

o I still do not entirely trust the sleep/wakeup logic.  Next step
will be to use a private snapshot of the CPU online mask in
rcu_sched_grace_period() -- if the CPU wasn't there at the start
of the grace period, we don't need to hear from it.  And the
bit about accounting for changes in online CPUs inside of
rcu_sched_grace_period() is ugly anyway.

o It might be good for rcu_sched_grace_period() to invoke
resched_cpu() when a given CPU wasn't responding quickly,
but resched_cpu() is declared static...

This patch also fixes a long-standing bug in the earlier preemptable-RCU
implementation of synchronize_rcu() that could result in loss of
concurrent external changes to a task's CPU affinity mask.  I still cannot
remember who reported this...

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

11 years agorcupreempt: remove duplicate prototypes
Steven Rostedt [Mon, 12 May 2008 19:21:05 +0000]
rcupreempt: remove duplicate prototypes

rcu_batches_completed and rcu_patches_completed_bh are both declared
in rcuclassic.h and rcupreempt.h. This patch removes the extra
prototypes for them from rcupdate.h.

rcu_batches_completed_bh is defined as a static inline in the rcupreempt.h
header file. Trying to export this as EXPORT_SYMBOL_GPL causes linking problems
with the powerpc linker. There's no need to export a static inlined function.

Modules must be compiled with the same type of RCU implementation as the
kernel they are for.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agoLinux 2.6.26-rc3
Linus Torvalds [Sun, 18 May 2008 21:36:41 +0000]
Linux 2.6.26-rc3

11 years agoMerge branch 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6
Linus Torvalds [Sun, 18 May 2008 20:56:54 +0000]
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/max6875: Really prevent 24RF08 corruption
  i2c-amd756: Fix functionality flags
  i2c: Kill the old driver matching scheme
  i2c: Convert remaining new-style drivers to use module aliasing
  i2c: Switch pasemi to the new device/driver matching scheme
  i2c: Clean up Blackfin BF527 I2C device declarations
  i2c-nforce2: Disable the second SMBus channel on the DFI Lanparty NF4 Expert
  i2c: New co-maintainer

11 years agom68k: Add multi_defconfig
Geert Uytterhoeven [Sun, 18 May 2008 18:47:24 +0000]
m68k: Add multi_defconfig

Add multi_defconfig, to build a kernel for all supported m68k platforms,
excluding Sun 3 (Sun 3 kernels are incompatible with all other m68k platforms)

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Update defconfigs
Geert Uytterhoeven [Sun, 18 May 2008 18:47:23 +0000]
m68k: Update defconfigs

Update the m68k defconfigs

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Correctly handle multi-ISA at runtime
Geert Uytterhoeven [Sun, 18 May 2008 18:47:22 +0000]
m68k: Correctly handle multi-ISA at runtime

m68k: Correctly handle multi-ISA at runtime in multi-platform kernels

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Prefix ISA type with ISA_TYPE_
Geert Uytterhoeven [Sun, 18 May 2008 18:47:21 +0000]
m68k: Prefix ISA type with ISA_TYPE_

The *_ISA type defines are quite generic and cause namespace conflicts
(e.g. with `AMIGAHW_DECLARE(GG2_ISA)' in <asm/amigahw.h>) for some kernel
configurations. Use ISA_TYPE_* to avoid such conflicts.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: export m68k_mmutype
Geert Uytterhoeven [Sun, 18 May 2008 18:47:20 +0000]
m68k: export m68k_mmutype

UIO needs m68k_mmutype:

ERROR: "m68k_mmutype" [drivers/uio/uio.ko] undefined!

(noticed by Christian T. Steigies)

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Q40/Q60 floppy support is broken
Geert Uytterhoeven [Sun, 18 May 2008 18:47:19 +0000]
m68k: Q40/Q60 floppy support is broken

Mark Q40/Q60 floppy support broken:

    arch/m68k/q40/q40ints.c: In function 'q40_irq_handler':
    arch/m68k/q40/q40ints.c:214: error: implicit declaration of function 'floppy_hardint'

Including <asm/floppy.h> doesn't help, as it causes a lot of additional error
messages (cfr. Sun 3x).

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Return -ENODEV if no device is found
Geert Uytterhoeven [Sun, 18 May 2008 18:47:18 +0000]
m68k: Return -ENODEV if no device is found

According to the tests in do_initcalls(), the proper error code in case no
device is found is -ENODEV, not -ENXIO or -EIO.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Some input drivers do not check the platform
Geert Uytterhoeven [Sun, 18 May 2008 18:47:17 +0000]
m68k: Some input drivers do not check the platform

Some input drivers do not check whether they're actually running on the
correct platform, causing multi-platform kernels to crash if they are not.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Some network drivers do not check the platform
Geert Uytterhoeven [Sun, 18 May 2008 18:47:16 +0000]
m68k: Some network drivers do not check the platform

Some network drivers do not check whether they're actually running on the
correct platform, causing multi-platform kernels to crash if they are not.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: dnfb doesnt check for Apollo
Geert Uytterhoeven [Sun, 18 May 2008 18:47:15 +0000]
m68k: dnfb doesnt check for Apollo

The Apollo frame buffer device driver (dnfb) doesn't check whether it's
actually running on Apollo hardware, causing a crash if it isn't.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: macide doesnt check for Mac
Geert Uytterhoeven [Sun, 18 May 2008 18:47:14 +0000]
m68k: macide doesnt check for Mac

The Macintosh IDE driver (macide) doesn't check whether it's actually running
on Mac hardware, causing a crash if it isn't.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Correct jump if not running on HP300
Geert Uytterhoeven [Sun, 18 May 2008 18:47:13 +0000]
m68k: Correct jump if not running on HP300

When running a HP300-enabled kernel on non-HP300 hardware, a test in the early
startup code jumps to the wrong label, causing a double bus fault.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Make gcc aware that BUG() does not return
Geert Uytterhoeven [Sun, 18 May 2008 18:47:12 +0000]
m68k: Make gcc aware that BUG() does not return

Use `__builtin_trap()' instead of `asm volatile("illegal")' in the m68k BUG()
macros (as suggested by Andrew Pinski), to kill warnings in code that assumes
BUG() does not return.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k vme_scc: avoid global namespace pollution
Geert Uytterhoeven [Sun, 18 May 2008 18:47:11 +0000]
m68k vme_scc: avoid global namespace pollution

m68k vme_scc:
  - make scc_ports[] static
  - kill unused global scc_initialized

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Kill CONFIG_WHIPPET_SERIAL
Geert Uytterhoeven [Sun, 18 May 2008 18:47:10 +0000]
m68k: Kill CONFIG_WHIPPET_SERIAL

The Hisoft Whippet PCMCIA serial driver has been removed a long time ago, but
it's Kconfig symbol still existed.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: FB_HP300 depends on DIO and doesnt need FB_CFB_FILLRECT
Geert Uytterhoeven [Sun, 18 May 2008 18:47:09 +0000]
m68k: FB_HP300 depends on DIO and doesnt need FB_CFB_FILLRECT

Correct FB_HP300 dependencies:
  - FB_HP300 doesn't depend only on HP300, but also on DIO (which depends on
    HP300)
  - FB_HP300 does not need FB_CFB_FILLRECT

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Kill CONFIG_FB_DAFB
Geert Uytterhoeven [Sun, 18 May 2008 18:47:08 +0000]
m68k: Kill CONFIG_FB_DAFB

CONFIG_FB_DAFB is a leftover from pre-Kconfig

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agom68k: Convert access_ok() to an inline function
Geert Uytterhoeven [Sun, 18 May 2008 18:47:07 +0000]
m68k: Convert access_ok() to an inline function

Convert access_ok() from a macro to an inline function, so the compiler no
longer complains about unused variables:

    fs/read_write.c: In function 'rw_copy_check_uvector':
    fs/read_write.c:556: warning: unused variable 'buf'

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agoi2c/max6875: Really prevent 24RF08 corruption
Jean Delvare [Sun, 18 May 2008 18:49:41 +0000]
i2c/max6875: Really prevent 24RF08 corruption

i2c-core takes care of the possible corruption of 24RF08 chips for
quite some times, so device devices no longer need to do it. And they
really should not, as applying the prevention twice voids it.

I thought that I had fixed all drivers long ago but apparently I had
missed that one.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Ben Gardner <bgardner@wabtec.com>

11 years agoi2c-amd756: Fix functionality flags
Jean Delvare [Sun, 18 May 2008 18:49:41 +0000]
i2c-amd756: Fix functionality flags

The i2c-amd756 driver pretends to support SMBus process call
transactions but actually does not. Fix it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>

11 years agoi2c: Kill the old driver matching scheme
Jean Delvare [Sun, 18 May 2008 18:49:41 +0000]
i2c: Kill the old driver matching scheme

Remove the old driver_name/type scheme for i2c driver matching. Only the
standard aliasing model will be used from now on.

Signed-off-by: Jean Delvare <khali@linux-fr.org>

11 years agoi2c: Convert remaining new-style drivers to use module aliasing
Jean Delvare [Sun, 18 May 2008 18:49:40 +0000]
i2c: Convert remaining new-style drivers to use module aliasing

Update all the remaining new-style i2c drivers to use standard module
aliasing instead of the old driver_name/type driver matching scheme.

Note that the tuner driver is a bit quirky at the moment, as it
overwrites i2c_client.name with arbitrary strings. We write "tuner"
back on remove, to make sure that driver cycling will work properly,
but there may still be troublesome corner cases.

Signed-off-by: Jean Delvare <khali@linux-fr.org>

11 years agoi2c: Switch pasemi to the new device/driver matching scheme
Jean Delvare [Sun, 18 May 2008 18:49:40 +0000]
i2c: Switch pasemi to the new device/driver matching scheme

The old device/driver matching scheme is going away so stop using it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Olof Johansson <olof@lixom.net>

11 years agoi2c: Clean up Blackfin BF527 I2C device declarations
Jean Delvare [Sun, 18 May 2008 18:49:40 +0000]
i2c: Clean up Blackfin BF527 I2C device declarations

I2C_BOARD_INFO() now sets the type field so no need to set it
separatetly.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Bryan Wu <cooloney@kernel.org>

11 years agoi2c-nforce2: Disable the second SMBus channel on the DFI Lanparty NF4 Expert
Jean Delvare [Sun, 18 May 2008 18:49:40 +0000]
i2c-nforce2: Disable the second SMBus channel on the DFI Lanparty NF4 Expert

There is a strange chip at 0x2e on the second SMBus channel of the
DFI Lanparty NF4 Expert motherboard. Accessing the chip reboots the
system. As there's nothing interesting on this SMBus channel, the
easiest and safest thing to do is to disable it on that board.

This is a better fix to bug #5889 than the it87 driver update that was
done originally:
http://bugzilla.kernel.org/show_bug.cgi?id=5889

Signed-off-by: Jean Delvare <khali@linux-fr.org>

11 years agoi2c: New co-maintainer
Jean Delvare [Sun, 18 May 2008 18:49:40 +0000]
i2c: New co-maintainer

Ben Dooks agreed to become my co-maintainer for the i2c subsystem. In
particular, Ben will help with drivers for embedded systems, of which
my experience is inexistent. Thanks Ben and welcome on board!

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Ben Dooks <ben-linux@fluff.org>

11 years agoMerge branch 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Sat, 17 May 2008 22:17:10 +0000]
Merge branch 'for-linus' of /home/rmk/linux-2.6-arm

* 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm:
  [ARM] pxa: spitz wants PXA27x UDC definitions
  [ARM] pxa: fix pxafb build when cpufreq is enabled
  [ARM] fix parenthesis in include/asm-arm/arch-omap/control.h
  [ARM] colibri: fix support for DM9000 ethernet device
  [ARM] arm/kernel/arthur.c: add MODULE_LICENSE
  [ARM] 5037/1: Orion: fix DNS323/Kurobox Pro PCI initialisation
  [ARM] 5034/1: fix arm{925,926,940,946} dma_flush_range() in WT mode
  [ARM] export copy_page
  [ARM] 5026/1: locomo: add .settype for gpio and several small fixes
  ARM: OMAP: Fixed comments on global PRM register usage
  ARM: OMAP: Add PARENT_CONTROLS_CLOCK flag to dpll5_m2_ck
  ARM: OMAP: PRCM fixes to ssi clock handling
  ARM: OMAP: Add fuctional clock enabler for iva2
  ARM: OMAP: Fix 34xx to use correct shift values for gpio2-6 fclks
  ARM: OMAP: Keymap fix for palmte and palmz71
  ARM: OMAP: Fix Unbalanced enable for IRQ in omap mailbox
  ARM: OMAP: DMA: Fix incorrect channel linking
  ARM: OMAP: Warn on disabling clocks with no users
  ARM: OMAP: Add calls to omap2_set_globals_*()
  ARM: OMAP: Update MMC header to fix compile

11 years agoMerge branch 'omap-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind...
Russell King [Sat, 17 May 2008 21:56:08 +0000]
Merge branch 'omap-fixes' of git://git./linux/kernel/git/tmlind/linux-omap-2.6

11 years agoMerge branch 'sa1100'
Russell King [Sat, 17 May 2008 21:55:51 +0000]
Merge branch 'sa1100'

11 years ago[ARM] pxa: spitz wants PXA27x UDC definitions
Russell King [Sat, 17 May 2008 21:51:35 +0000]
[ARM] pxa: spitz wants PXA27x UDC definitions

... so include the header file.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] pxa: fix pxafb build when cpufreq is enabled
Russell King [Sat, 17 May 2008 21:51:14 +0000]
[ARM] pxa: fix pxafb build when cpufreq is enabled

If cpufreq is enabled, pxafb wants to call the removed
get_clk_frequency_khz() function for a debug printk.  Remove
this reference.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] fix parenthesis in include/asm-arm/arch-omap/control.h
Mariusz Kozlowski [Thu, 15 May 2008 18:52:24 +0000]
[ARM] fix parenthesis in include/asm-arm/arch-omap/control.h

Parenthesis fix in include/asm-arm/arch-omap/control.h

Signed-off-by: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Tony Lindgren <tony@atomide.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] colibri: fix support for DM9000 ethernet device
Michael Abbott [Wed, 14 May 2008 23:29:24 +0000]
[ARM] colibri: fix support for DM9000 ethernet device

Two changes are necessary to enable proper operation of the DM9000 device with
the Colibri PXA 270 board: firstly, the IRQ type needs to be configured for
rising edge interrupts, and secondly this configuration needs to be
communicated through to the DM9000.

[akpm@linux-foundation.org: remove set_irq_type() call as per ben-linux request]
Signed-off-by: Michael Abbott <michael.abbott@diamond.ac.uk>
Cc: Daniel Mack <daniel@caiaq.org>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] arm/kernel/arthur.c: add MODULE_LICENSE
Adrian Bunk [Mon, 31 Mar 2008 09:08:16 +0000]
[ARM] arm/kernel/arthur.c: add MODULE_LICENSE

This patch adds the missing MODULE_LICENSE("GPL").

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] 5037/1: Orion: fix DNS323/Kurobox Pro PCI initialisation
Lennert Buytenhek [Thu, 15 May 2008 09:31:14 +0000]
[ARM] 5037/1: Orion: fix DNS323/Kurobox Pro PCI initialisation

Whereas most Orion 5x machine support code would initialise the PCI
subsystem with nr_controllers in their struct hw_pci set to 2, the
DNS323 and Kurobox Pro machine support code had nr_controllers set
to 1.

This was presumably done because on those two machines, the PCI(-X)
controller (nr == 1) isn't used, requiring initialisation of only
the PCIe controller (nr == 0.)  However, not initialising the PCI(-X)
controller on boards that don't use it leads to a situation where
both the PCIe and the PCI(-X) controller think that their root bus is
zero, and it messes up IRQ assignment.

This patch changes the DNS323 and Kurobox Pro support code to always
use nr_controllers == 2.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] 5034/1: fix arm{925,926,940,946} dma_flush_range() in WT mode
Lennert Buytenhek [Sat, 10 May 2008 20:05:31 +0000]
[ARM] 5034/1: fix arm{925,926,940,946} dma_flush_range() in WT mode

The CPU's dma_flush_range() operation needs to clean+invalidate the
given memory area if the cache is in writeback mode, or do just the
invalidate part if the cache is in writethrough mode, but the current
proc-arm{925,926,940,946} (incorrectly) do a cache clean in the
latter case.  This patch fixes that.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] export copy_page
Russell King [Sun, 11 May 2008 21:36:03 +0000]
[ARM] export copy_page

Martin Michlmayr reported that fuse complains:
  ERROR: "copy_page" [fs/fuse/fuse.ko] undefined!

so export the needed function.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years ago[ARM] 5026/1: locomo: add .settype for gpio and several small fixes
Thomas Kunze [Tue, 29 Apr 2008 16:44:54 +0000]
[ARM] 5026/1: locomo: add .settype for gpio and several small fixes

irqs.h:
    * rename IRQ_LOCOMO_SPI_OVRN to IRQ_LOCOMO_SPI_REND
locomo.h:
    * add some definition for locomo spi controller
    * correct some errors
locomo.c:
    * correct some errors
    * add set_type for locomo gpio irq chip

Signed-off-by: Thomas Kunze <thommycheck@gmx.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

11 years agoMerge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 17 May 2008 21:21:43 +0000]
Merge branch 'x86-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: disable mwait for AMD family 10H/11H CPUs
  x86: fix crash on cpu hotplug on pat-incapable machines
  x86: remove mwait capability C-state check

11 years agox86: disable mwait for AMD family 10H/11H CPUs
Thomas Gleixner [Fri, 16 May 2008 20:55:26 +0000]
x86: disable mwait for AMD family 10H/11H CPUs

The previous revert of 0c07ee38c9d4eb081758f5ad14bbffa7197e1aec left
out the mwait disable condition for AMD family 10H/11H CPUs.

Andreas Herrman said:

It depends on the CPU. For AMD CPUs that support MWAIT this is wrong.
Family 0x10 and 0x11 CPUs will enter C1 on HLT. Powersavings then
depend on a clock divisor and current Pstate of the core.

If all cores of a processor are in halt state (C1) the processor can
enter the C1E (C1 enhanced) state. If mwait is used this will never
happen.

Thus HLT saves more power than MWAIT here.

It might be best to switch off the mwait flag for these AMD CPU
families like it was introduced with commit
f039b754714a422959027cb18bb33760eb8153f0 (x86: Don't use MWAIT on AMD
Family 10)

Re-add the AMD families 10H/11H check and disable the mwait usage for
those.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

11 years agox86: fix crash on cpu hotplug on pat-incapable machines
Avi Kivity [Wed, 14 May 2008 09:20:32 +0000]
x86: fix crash on cpu hotplug on pat-incapable machines

pat_disable() is __init, which means it goes away after booting is complete.
Unfortunately it is used by the hotplug code if the machine is not
pat-capable, causing a crash.

Fix by marking pat_disable() as __cpuinit.

Signed-off-by: Avi Kivity <avi@qumranet.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agox86: remove mwait capability C-state check
Ingo Molnar [Wed, 14 May 2008 06:47:40 +0000]
x86: remove mwait capability C-state check

Vegard Nossum reports:

| powertop shows between 200-400 wakeups/second with the description
| "<kernel IPI>: Rescheduling interrupts" when all processors have load (e.g.
| I need to run two busy-loops on my 2-CPU system for this to show up).
|
| The bisect resulted in this commit:
|
| commit 0c07ee38c9d4eb081758f5ad14bbffa7197e1aec
| Date:   Wed Jan 30 13:33:16 2008 +0100
|
|     x86: use the correct cpuid method to detect MWAIT support for C states

remove the functional effects of this patch and make mwait unconditional.

A future patch will turn off mwait on specific CPUs where that causes
power to be wasted.

Bisected-by: Vegard Nossum <vegard.nossum@gmail.com>
Tested-by: Vegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

11 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6
Linus Torvalds [Fri, 16 May 2008 22:39:28 +0000]
Merge git://git./linux/kernel/git/davem/sparc-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  sparc64: Update defconfig.
  sparc64: Fix lmb_reserve() args in find_ramdisk().

11 years agosparc64: Update defconfig.
David S. Miller [Fri, 16 May 2008 20:36:27 +0000]
sparc64: Update defconfig.

Signed-off-by: David S. Miller <davem@davemloft.net>

11 years agosparc64: Fix lmb_reserve() args in find_ramdisk().
David S. Miller [Thu, 15 May 2008 06:10:33 +0000]
sparc64: Fix lmb_reserve() args in find_ramdisk().

This fixes the missing ram regression reported by
Mikael Pettersson <mikpe@it.uu.se>, much thanks for
all of this help in diagnosing this.

The second argument to lmb_reserve() is a size,
not an end address bounds.

Tested-by: Mikael Pettersson <mikpe@it.uu.se>

Signed-off-by: David S. Miller <davem@davemloft.net>

11 years agobyteorder: don't directly include linux/byteorder/generic.h
Harvey Harrison [Fri, 16 May 2008 18:20:25 +0000]
byteorder: don't directly include linux/byteorder/generic.h

Use asm/byteorder.h instead.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc
Linus Torvalds [Fri, 16 May 2008 01:28:46 +0000]
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] macintosh: Replace deprecated __initcall with device_initcall
  [POWERPC] cell: Fix section mismatches in io-workarounds code
  [POWERPC] spufs: Fix compile error
  [POWERPC] Fix uninitialized variable bug in copy_{to|from}_user
  [POWERPC] Add null pointer check to of_find_property
  [POWERPC] vmemmap fixes to use smaller pages
  [POWERPC] spufs: Fix pointer reference in find_victim
  [POWERPC] 85xx: SBC8548 - Add flash support and HW Rev reporting
  [POWERPC] 85xx: Fix some sparse warnings for 85xx MDS
  [POWERPC] 83xx: Enable DMA engine on the MPC8377 MDS board.
  [POWERPC] 86xx: mpc8610_hpcd: fix second serial port
  [POWERPC] 86xx: mpc8610_hpcd: add support for NOR and NAND flashes
  [POWERPC] 85xx: Add 8568 PHY workarounds to board code
  [POWERPC] 86xx: mpc8610_hpcd: use ULI526X driver for on-board ethernet

11 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
Linus Torvalds [Fri, 16 May 2008 01:28:28 +0000]
Merge branch 'for_linus' of git://git./linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  jbd2: update transaction t_state to T_COMMIT fix
  ext4: Retry block allocation if new blocks are allocated from system zone.
  ext4: mballoc fix mb_normalize_request algorithm for 1KB block size filesystems
  ext4: fix typos in messages and comments (journalled -> journaled)
  ext4: fix synchronization of quota files in journal=data mode
  ext4: Fix mount messages when quota disabled
  ext4: correct mount option parsing to detect when quota options can be changed

11 years agoinitcalls: Fix m68k build and possible buffer overflow
Cyrill Gorcunov [Thu, 15 May 2008 20:52:41 +0000]
initcalls: Fix m68k build and possible buffer overflow

This patch fixes a build bug on m68k - gcc decides to emit a call to the
strlen library function, which we don't implement.

More importantly - my previous patch "init: don't lose initcall return
values" (commit e662e1cfd434aa234b72fbc781f1d70211cb785b) had introduced
potential buffer overflow by wrong calculation of string accumulator
size.

Use strlcat() instead, fixing both bugs.

Many thanks Andreas Schwab and Geert Uytterhoeven for helping
to catch and fix the bug.

Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agoSplit up 'do_initcalls()' into two simpler functions
Linus Torvalds [Fri, 16 May 2008 01:14:01 +0000]
Split up 'do_initcalls()' into two simpler functions

One function to just loop over the entries, one function to actually do
the call and the associated debugging code.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agoClean up 'print_fn_descriptor_symbol()' types
Linus Torvalds [Fri, 16 May 2008 00:50:37 +0000]
Clean up 'print_fn_descriptor_symbol()' types

Everybody wants to pass it a function pointer, and in fact, that is what
you _must_ pass it for it to make sense (since it knows that ia64 and
ppc64 use descriptors for function pointers and fetches the actual
address from there).

So don't make the argument be a 'unsigned long' and force everybody to
add a cast.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years agoMerge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6
Linus Torvalds [Thu, 15 May 2008 20:56:42 +0000]
Merge branch 'release' of git://git./linux/kernel/git/aegl/linux-2.6

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6:
  [IA64] fix personality(PER_LINUX32) performance issue
  [IA64] Properly unregister legacy interrupts
  [IA64] Remove NULL pointer check for argument never passed as NULL.
  [IA64] trivial cleanup for perfmon.c
  [IA64] trivial cleanup for entry.S
  [IA64] fix interrupt masking for pending works on kernel leave
  [IA64] allow user to force_pal_cache_flush
  [IA64] Don't reserve crashkernel memory > 4 GB
  [IA64] machvec support for SGI UV platform
  [IA64] Add header files for SGI UV platform

11 years agojbd2: update transaction t_state to T_COMMIT fix
Mingming Cao [Thu, 15 May 2008 18:46:17 +0000]
jbd2: update transaction t_state to T_COMMIT fix

Updating the current transaction's t_state is protected by j_state_lock.  We
need to do the same when updating the t_state to T_COMMIT.

Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

11 years agoext4: Retry block allocation if new blocks are allocated from system zone.
Aneesh Kumar K.V [Thu, 15 May 2008 18:43:20 +0000]
ext4: Retry block allocation if new blocks are allocated from system zone.

If the block allocator gets blocks out of system zone ext4 calls
ext4_error. But if the file system is mounted with errors=continue
retry block allocation. We need to mark the system zone blocks as
in use to make sure retry don't pick them again

System zone is the block range mapping block bitmap, inode bitmap and inode
table.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>

11 years agotty: fix BKL related leak and crash
Ingo Molnar [Wed, 14 May 2008 15:11:46 +0000]
tty: fix BKL related leak and crash

Enabling the BKL to be lockdep tracked uncovered the following
upstream kernel bug in the tty code, which caused a BKL
reference leak:

  ================================================
  [ BUG: lock held when returning to user space! ]
  ------------------------------------------------
  dmesg/3121 is leaving the kernel with locks still held!
  1 lock held by dmesg/3121:
   #0:  (kernel_mutex){--..}, at: [<c02f34d9>] opost+0x24/0x194

this might explain some of the atomicity warnings and crashes
that -tip tree testing has been experiencing since the BKL
was converted back to a spinlock.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

11 years ago[IA64] fix personality(PER_LINUX32) performance issue
Huang, Xiaolan [Thu, 15 May 2008 02:18:41 +0000]
[IA64] fix personality(PER_LINUX32) performance issue

The patch aims to fix a performance issue for the syscall
personality(PER_LINUX32).

On IA-64 box, the syscall personality (PER_LINUX32) has poor performance
because it failed to find the Linux/x86 execution domain. Then it tried
to load the kernel module however it failed always and it used the default
execution domain PER_LINUX instead. Requesting kernel modules is very
expensive. It caused the performance issue. (see the function
lookup_exec_domain in kernel/exec_domain.c).

To resolve the issue, execution domain Linux/x86 is always registered in
initialization time for IA-64 architecture.

Signed-off-by: Xiaolan Huang <xiaolan.huang@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

11 years agoMerge branch 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6
Linus Torvalds [Thu, 15 May 2008 16:10:13 +0000]
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] show_interrupts: prevent cpu hotplug when walking cpu_online_map.
  [S390] smp: __smp_call_function_map vs cpu_online_map fix.
  [S390] tape: Use ccw_dev_id to build cdev_id.
  [S390] dasd: fix timeout handling in interrupt handler
  [S390] s390dbf: Use const char * for dbf name.
  [S390] dasd: Use const in busid functions.
  [S390] blacklist.c: removed duplicated include
  [S390] vmlogrdr: module initialization function should return negative errors
  [S390] sparsemem vmemmap: initialize memmap.
  [S390] Remove last traces of cio_msg=.
  [S390] cio: Remove CCW_CMD_SUSPEND_RECONN in front of CCW_CMD_SET_PGID.

11 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/kyle/parisc-2.6
Linus Torvalds [Thu, 15 May 2008 16:09:43 +0000]
Merge git://git./linux/kernel/git/kyle/parisc-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/kyle/parisc-2.6:
  arch/parisc/kernel/perf_asm.S: build fix
  parisc: remove -traditional from assembler flags
  parisc: use conditional macro for 64-bit wide ops
  parisc: Remove ioctl.h content picked up from <asm-generic/ioctl.h>.
  arch/parisc/kernel/unaligned.c: use time_* macros
  parisc: remove redundant display of free swap space in show_mem()
  drivers/parisc: replace remaining __FUNCTION__ occurrences
  parisc: replace remaining __FUNCTION__ occurrences
  parisc: new termios definitions
  parisc: fix trivial section name warnings

11 years agoarch/parisc/kernel/perf_asm.S: build fix
Kyle McMartin [Thu, 15 May 2008 14:57:33 +0000]
arch/parisc/kernel/perf_asm.S: build fix

Missing <linux/init.h> header for __HEAD macro.

Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agoparisc: remove -traditional from assembler flags
Kyle McMartin [Thu, 15 May 2008 14:56:36 +0000]
parisc: remove -traditional from assembler flags

Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agoparisc: use conditional macro for 64-bit wide ops
Kyle McMartin [Thu, 15 May 2008 14:53:57 +0000]
parisc: use conditional macro for 64-bit wide ops

This work enables us to remove -traditional from $AFLAGS on
parisc.

Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agoparisc: Remove ioctl.h content picked up from <asm-generic/ioctl.h>.
Robert P. J. Day [Wed, 14 May 2008 19:14:52 +0000]
parisc: Remove ioctl.h content picked up from <asm-generic/ioctl.h>.

Now that <asm-generic/ioctl.h> allows overriding of the most commonly
changed macro values, take advantage of that.

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years ago[S390] show_interrupts: prevent cpu hotplug when walking cpu_online_map.
Heiko Carstens [Thu, 15 May 2008 14:52:39 +0000]
[S390] show_interrupts: prevent cpu hotplug when walking cpu_online_map.

Surround all the code withing show_interrupts() with
get/put_online_cpus() to prevent strange results wrt cpu hotplug.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] smp: __smp_call_function_map vs cpu_online_map fix.
Heiko Carstens [Thu, 15 May 2008 14:52:38 +0000]
[S390] smp: __smp_call_function_map vs cpu_online_map fix.

Both smp_call_function() and __smp_call_function_map() access
cpu_online_map. Both functions run with preemption disabled which
protects for cpus going offline. However new cpus can be added and
therefore the cpu_online_map can change unexpectedly.
So use the call_lock to protect against changes to the cpu_online_map
in start_secondary() and all smp_call_* functions.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] tape: Use ccw_dev_id to build cdev_id.
Cornelia Huck [Thu, 15 May 2008 14:52:37 +0000]
[S390] tape: Use ccw_dev_id to build cdev_id.

To construct the integer containing the information from the bus_id,
it is easier to use the data from ccw_dev_id than to parse the bus_id.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] dasd: fix timeout handling in interrupt handler
Stefan Weinhuber [Thu, 15 May 2008 14:52:36 +0000]
[S390] dasd: fix timeout handling in interrupt handler

When the dasd_int_handler is called with an error code instead of
an irb, the associated request should be restarted. This handling
was missing from the -ETIMEDOUT case. In fact it should be done in
any case.

Signed-off-by: Stefan Weinhuber <wein@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] s390dbf: Use const char * for dbf name.
Cornelia Huck [Thu, 15 May 2008 14:52:35 +0000]
[S390] s390dbf: Use const char * for dbf name.

We should use const char * for passing the name of the debug feature
around since it will not be changed.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] dasd: Use const in busid functions.
Cornelia Huck [Thu, 15 May 2008 14:52:34 +0000]
[S390] dasd: Use const in busid functions.

We should use 'const char *' in the busid functions since the
strings are not modified anyway.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] blacklist.c: removed duplicated include
Huang Weiyi [Thu, 15 May 2008 14:52:33 +0000]
[S390] blacklist.c: removed duplicated include

Removed duplicated include <asm/cio.h>
in drivers/s390/cio/blacklist.c.

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] vmlogrdr: module initialization function should return negative errors
Marcin Slusarz [Thu, 15 May 2008 14:52:32 +0000]
[S390] vmlogrdr: module initialization function should return negative errors

Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] sparsemem vmemmap: initialize memmap.
Heiko Carstens [Thu, 15 May 2008 14:52:31 +0000]
[S390] sparsemem vmemmap: initialize memmap.

Let's just use the generic vmmemmap_alloc_block() function which
always returns initialized memory.

Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] Remove last traces of cio_msg=.
Cornelia Huck [Thu, 15 May 2008 14:52:30 +0000]
[S390] Remove last traces of cio_msg=.

cio_msg= is gone, also remove it from kernel-parameters.txt.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years ago[S390] cio: Remove CCW_CMD_SUSPEND_RECONN in front of CCW_CMD_SET_PGID.
Michael Ernst [Thu, 15 May 2008 14:52:29 +0000]
[S390] cio: Remove CCW_CMD_SUSPEND_RECONN in front of CCW_CMD_SET_PGID.

CCW_CMD_SUSPEND_RECONN causes a system hang if the cable of a
reserved DASD is disconnected and connected again.

Signed-off-by: Michael Ernst <mernst@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

11 years agoarch/parisc/kernel/unaligned.c: use time_* macros
S.Caglar Onur [Wed, 14 May 2008 23:21:57 +0000]
arch/parisc/kernel/unaligned.c: use time_* macros

The functions time_before, time_before_eq, time_after, and time_after_eq are
more robust for comparing jiffies against other values.

So use the time_after() macro, defined in linux/jiffies.h, which deals with
wrapping correctl

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: S.Caglar Onur <caglar@pardus.org.tr>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agoparisc: remove redundant display of free swap space in show_mem()
Johannes Weiner [Wed, 14 May 2008 23:21:56 +0000]
parisc: remove redundant display of free swap space in show_mem()

show_mem() has no need to print the amount of free swap space manually because
show_free_areas() does this already and is called by the former.

The two outputs only differ in text formatting:

  printk("Free swap  = %lukB\n", ...);
  printk("Free swap:       %6ldkB\n", ...);

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agodrivers/parisc: replace remaining __FUNCTION__ occurrences
Harvey Harrison [Wed, 14 May 2008 23:21:56 +0000]
drivers/parisc: replace remaining __FUNCTION__ occurrences

__FUNCTION__ is gcc-specific, use __func__

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Matthew Wilcox <willy@debian.org>
Cc: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agoparisc: replace remaining __FUNCTION__ occurrences
Harvey Harrison [Wed, 14 May 2008 23:21:55 +0000]
parisc: replace remaining __FUNCTION__ occurrences

__FUNCTION__ is gcc-specific, use __func__

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Matthew Wilcox <willy@debian.org>
Cc: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agoparisc: new termios definitions
Alan Cox [Wed, 14 May 2008 23:21:54 +0000]
parisc: new termios definitions

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Grant Grundler <grundler@parisc-linux.org>
Cc: Matthew Wilcox <willy@debian.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years agoparisc: fix trivial section name warnings
Helge Deller [Fri, 2 May 2008 20:02:48 +0000]
parisc: fix trivial section name warnings

This trivial patch fixes the following section warnings on PARISC:
> WARNING: vmlinux.o (.text.1): unexpected section name.
>The (.[number]+) following section name are ld generated and not expected.
> Did you forget to use "ax"/"aw" in a .S file?
> Note that for example <linux/init.h> contains
> section definitions for use in .S files.

Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>

11 years ago[POWERPC] macintosh: Replace deprecated __initcall with device_initcall
Robert P. J. Day [Wed, 14 May 2008 23:12:53 +0000]
[POWERPC] macintosh: Replace deprecated __initcall with device_initcall

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>

11 years ago[POWERPC] cell: Fix section mismatches in io-workarounds code
Ishizaki Kou [Wed, 14 May 2008 09:05:19 +0000]
[POWERPC] cell: Fix section mismatches in io-workarounds code

Fix following warnings:
WARNING: arch/powerpc/platforms/cell/built-in.o(.devinit.text+0x9c): Section mismatch in reference from the function .cell_setup_phb() to the function .init.text:.iowa_register_bus()
WARNING: arch/powerpc/platforms/cell/built-in.o(.devinit.text+0xa4): Section mismatch in reference from the function .cell_setup_phb() to the function .init.text:.io_workaround_init()

Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Signed-off-by: Paul Mackerras <paulus@samba.org>

11 years ago[POWERPC] spufs: Fix compile error
FUJITA Tomonori [Tue, 13 May 2008 09:07:42 +0000]
[POWERPC] spufs: Fix compile error

With CONFIG_VIRT_CPU_ACCOUNTING disabled, I got the following error:

linux-2.6/arch/powerpc/platforms/cell/spufs/file.c: In function 'spu_switch_log_notify':
linux-2.6/arch/powerpc/platforms/cell/spufs/file.c:2542: error: implicit declaration of function 'get_tb'
make[4]: *** [arch/powerpc/platforms/cell/spufs/file.o] Error 1

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Paul Mackerras <paulus@samba.org>

11 years ago[POWERPC] Fix uninitialized variable bug in copy_{to|from}_user
Nate Case [Mon, 12 May 2008 20:14:14 +0000]
[POWERPC] Fix uninitialized variable bug in copy_{to|from}_user

Calls to copy_to_user() or copy_from_user() can fail when copying N
bytes, where N is a constant less than 8, but not 1, 2, 4, or 8,
because 'ret' is not initialized and is only set if the size is 1,
2, 4 or 8, but is tested after the switch statement for any constant
size <= 8.  This fixes it by initializing 'ret' to 1, causing the
code to fall through to the __copy_tofrom_user call for sizes other
than 1, 2, 4 or 8.

Signed-off-by: Dave Scidmore <dscidmore@xes-inc.com>
Signed-off-by: Nate Case <ncase@xes-inc.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>