sfrench/cifs-2.6.git
16 years agoDocumentation: prune redundant SubmitChecklist items
J. Bruce Fields [Wed, 13 Feb 2008 23:03:20 +0000 (15:03 -0800)]
Documentation: prune redundant SubmitChecklist items

Kernel style is mentioned twice, and the git apply trick is a bit redundant
given the checkpatch.pl recommendation (which also checks for bad
whitespace).

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agohugetlb: fix overcommit locking
Nishanth Aravamudan [Wed, 13 Feb 2008 23:03:19 +0000 (15:03 -0800)]
hugetlb: fix overcommit locking

proc_doulongvec_minmax() calls copy_to_user()/copy_from_user(), so we can't
hold hugetlb_lock over the call.  Use a dummy variable to store the sysctl
result, like in hugetlb_sysctl_handler(), then grab the lock to update
nr_overcommit_huge_pages.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Reported-by: Miles Lane <miles.lane@gmail.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoSC26XX: missing PORT define in serial_core.h
Thomas Bogendoerfer [Wed, 13 Feb 2008 23:03:17 +0000 (15:03 -0800)]
SC26XX: missing PORT define in serial_core.h

When submitting the driver for inclusion to 2.6.25 I've missed the change to
serial_core.h. This patch fixes this.

Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoFinal removal of FASTCALL()/fastcall
Harvey Harrison [Wed, 13 Feb 2008 23:03:17 +0000 (15:03 -0800)]
Final removal of FASTCALL()/fastcall

All users are gone, remove definitions and comments referring
to them.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoremove final fastcall users
Harvey Harrison [Wed, 13 Feb 2008 23:03:16 +0000 (15:03 -0800)]
remove final fastcall users

fastcall always expands to empty, remove it.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agorcupdate: fix comment
Paul E. McKenney [Wed, 13 Feb 2008 23:03:15 +0000 (15:03 -0800)]
rcupdate: fix comment

This comment caused some consternation during fastcall removal.  Make it
truthful.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoinclude/linux: Remove all users of FASTCALL() macro
Harvey Harrison [Wed, 13 Feb 2008 23:03:15 +0000 (15:03 -0800)]
include/linux: Remove all users of FASTCALL() macro

FASTCALL() is always expanded to empty, remove it.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoxtime_lock vs update_process_times
Peter Zijlstra [Wed, 13 Feb 2008 20:33:16 +0000 (21:33 +0100)]
xtime_lock vs update_process_times

Commit d3d74453c34f8fd87674a8cf5b8a327c68f22e99 ("hrtimer: fixup the
HRTIMER_CB_IRQSAFE_NO_SOFTIRQ fallback") broke several archs, and since
only Russell bothered to merge the fix, and Greg to ACK his arch, I'm
sending this for merger.

I have confirmation that the Alpha bit results in a booting kernel.
That leaves: blackfin, frv, sh and sparc untested.

The deadlock in question was found by Russell:

  IRQ handle
    -> timer_tick() - xtime seqlock held for write
      -> update_process_times()
        -> run_local_timers()
          -> hrtimer_run_queues()
            -> hrtimer_get_softirq_time() - tries to get a read lock

Now, Thomas assures me the fix is trivial, only do_timer() needs to be
done under the xtime_lock, and update_process_times() can savely be
removed from under it.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Greg Ungerer <gerg@uclinux.org>
CC: Richard Henderson <rth@twiddle.net>
CC: Bryan Wu <bryan.wu@analog.com>
CC: David Howells <dhowells@redhat.com>
CC: Paul Mundt <lethal@linux-sh.org>
CC: William Irwin <wli@holomorphy.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoacpi: fix acpi_os_read_pci_configuration() misuse of raw_pci_read()
Linus Torvalds [Wed, 13 Feb 2008 17:56:14 +0000 (09:56 -0800)]
acpi: fix acpi_os_read_pci_configuration() misuse of raw_pci_read()

The raw_pci_read() interface (as the raw_pci_ops->read() before it)
unconditionally fills in a 32-bit integer return value regardless of the
size of the operation requested.

So claiming to take a "void *" is wrong, as is passing in a pointer to
just a byte variable.

Noticed by pageexec when enabling -fstack-protector (which needs other
patches too to actually work, but that's a separate issue).

Acked-by: Len Brown <len.brown@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoFRV: Fix up parse error in linker script
David Howells [Wed, 13 Feb 2008 16:10:08 +0000 (16:10 +0000)]
FRV: Fix up parse error in linker script

Fix up parse error in FRV linker script, presumably introduced through changes
to the INIT_TEXT and EXIT_TEXT macros.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched
Linus Torvalds [Wed, 13 Feb 2008 16:22:41 +0000 (08:22 -0800)]
Merge git://git./linux/kernel/git/mingo/linux-2.6-sched

* git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched:
  sched: rt-group: refure unrunnable tasks
  sched: rt-group: clean up the ifdeffery
  sched: rt-group: make rt groups scheduling configurable
  sched: rt-group: interface
  sched: rt-group: deal with PI
  sched: fix incorrect irq lock usage in normalize_rt_tasks()
  sched: fair-group: separate tg->shares from task_group_lock
  hrtimer: more hrtimer_init_sleeper() fallout.

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86
Linus Torvalds [Wed, 13 Feb 2008 16:20:31 +0000 (08:20 -0800)]
Merge git://git./linux/kernel/git/x86/linux-2.6-x86

* git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86:
  x86: pit_clockevent can be static
  x86: EFI runtime code mapping enhancement
  x86: EFI: fix use of unitialized variable and the cache logic
  x86: CPA: fix gbpages support in try_preserve_large_page
  xen: unpin initial Xen pagetable once we're finished with it
  x86/early_ioremap: don't assume we're using swapper_pg_dir
  x86: fixup machine_ops reboot_{32|64}.c unification fallout
  x86: fix sigcontext.h user export

16 years agodrivers/memstick/host/tifm_ms.c breakage
Al Viro [Wed, 13 Feb 2008 03:56:59 +0000 (03:56 +0000)]
drivers/memstick/host/tifm_ms.c breakage

writel(sock + ...) that should've been writel(sock->addr + ...)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agodm-raid1 breakage on 64bit
Al Viro [Wed, 13 Feb 2008 03:53:00 +0000 (03:53 +0000)]
dm-raid1 breakage on 64bit

test_and_set_bit() on address of uint32_t is a Bad Idea(tm)...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agowmi: (!x & y) strikes again
Al Viro [Wed, 13 Feb 2008 04:03:25 +0000 (04:03 +0000)]
wmi: (!x & y) strikes again

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Carlos Corbacho <carlos@strangeworlds.co.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agox86: pit_clockevent can be static
Harvey Harrison [Tue, 12 Feb 2008 20:10:27 +0000 (12:10 -0800)]
x86: pit_clockevent can be static

arch/x86/kernel/i8253.c:98:27: warning: symbol 'pit_clockevent' was not declared. Should it be static?

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86: EFI runtime code mapping enhancement
Huang, Ying [Wed, 13 Feb 2008 09:22:41 +0000 (17:22 +0800)]
x86: EFI runtime code mapping enhancement

This patch enhances EFI runtime code memory mapping as following:

- Move __supported_pte_mask & _PAGE_NX checking before invoking
  runtime_code_page_mkexec(). This makes it possible for compiler to
  eliminate runtime_code_page_mkexec() on machine without NX support.

- Use set_memory_x/nx in early_mapping_set_exec(). This eliminates the
  duplicated implementation.

This patch has been tested on Intel x86_64 platform with EFI64/32
firmware.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86: EFI: fix use of unitialized variable and the cache logic
Thomas Gleixner [Tue, 12 Feb 2008 18:46:48 +0000 (19:46 +0100)]
x86: EFI: fix use of unitialized variable and the cache logic

Andi Kleen pointed out that the cache attribute logic is reverse in
efi_enter_virtual_mode(). This problem alone is harmless as we do not
(yet) do cache attribute conflict resolution. (This bug was not present
in the original EFI submission - I introduced it while fixing up rejects.)

While reviewing this code I noticed a second, worse problem: the use of
uninitialized md->virt_addr.

Fix both problems.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86: CPA: fix gbpages support in try_preserve_large_page
Andi Kleen [Wed, 13 Feb 2008 15:20:35 +0000 (16:20 +0100)]
x86: CPA: fix gbpages support in try_preserve_large_page

[ mingo@elte.hu: while gbpages cannot be enabled on mainline currently,
  keep the code uptodate and this fix is easy enough. ]

Use correct page sizes and masks for GB pages in try_preserve_large_page()

This prevents a boot hang on a GB capable system with CONFIG_DIRECT_GBPAGES
enabled.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
16 years agoxen: unpin initial Xen pagetable once we're finished with it
Jeremy Fitzhardinge [Wed, 13 Feb 2008 15:20:35 +0000 (16:20 +0100)]
xen: unpin initial Xen pagetable once we're finished with it

Unpin the Xen-provided pagetable once we've finished with it, so it
doesn't cause stray references which cause later swapper_pg_dir
pagetable updates to fail.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Tested-by: Jody Belka <knew-linux@pimb.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86/early_ioremap: don't assume we're using swapper_pg_dir
Jeremy Fitzhardinge [Wed, 13 Feb 2008 15:20:35 +0000 (16:20 +0100)]
x86/early_ioremap: don't assume we're using swapper_pg_dir

At the early stages of boot, before the kernel pagetable has been
fully initialized, a Xen kernel will still be running off the
Xen-provided pagetables rather than swapper_pg_dir[].  Therefore,
readback cr3 to determine the base of the pagetable rather than
assuming swapper_pg_dir[].

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Tested-by: Jody Belka <knew-linux@pimb.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86: fixup machine_ops reboot_{32|64}.c unification fallout
Jody Belka [Tue, 12 Feb 2008 23:37:48 +0000 (23:37 +0000)]
x86: fixup machine_ops reboot_{32|64}.c unification fallout

When reboot_32.c and reboot_64.c were unified (commit 4d022e35fd...),
the machine_ops code was broken, leading to xen pvops kernels failing
to properly halt/poweroff/reboot etc. This fixes that up.

Signed-off-by: Jody Belka <knew-linux@pimb.org>
Cc: Miguel Boton <mboton@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86: fix sigcontext.h user export
Ingo Molnar [Wed, 13 Feb 2008 15:20:35 +0000 (16:20 +0100)]
x86: fix sigcontext.h user export

Jakub Jelinek reported that some user-space code that relies on
kernel headers has built dependency on the sigcontext->eip/rip
register names - which have been unified in commit:

  commit 742fa54a62be6a263df14a553bf832724471dfbe
  Author: H. Peter Anvin <hpa@zytor.com>
  Date:   Wed Jan 30 13:30:56 2008 +0100

      x86: use generic register names in struct sigcontext

so give the old layout to user-space. This is not particularly
pretty, but it's an ABI so there's no danger of the two definitions
getting out of sync.

Reported-by: Jakub Jelinek <jakub@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosched: rt-group: refure unrunnable tasks
Peter Zijlstra [Wed, 13 Feb 2008 14:45:40 +0000 (15:45 +0100)]
sched: rt-group: refure unrunnable tasks

Refuse to accept or create RT tasks in groups that can't run them.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosched: rt-group: clean up the ifdeffery
Peter Zijlstra [Wed, 13 Feb 2008 14:45:40 +0000 (15:45 +0100)]
sched: rt-group: clean up the ifdeffery

Clean up some of the excessive ifdeffery introduces in the last patch.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosched: rt-group: make rt groups scheduling configurable
Peter Zijlstra [Wed, 13 Feb 2008 14:45:40 +0000 (15:45 +0100)]
sched: rt-group: make rt groups scheduling configurable

Make the rt group scheduler compile time configurable.
Keep it experimental for now.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosched: rt-group: interface
Peter Zijlstra [Wed, 13 Feb 2008 14:45:39 +0000 (15:45 +0100)]
sched: rt-group: interface

Change the rt_ratio interface to rt_runtime_us, to match rt_period_us.
This avoids picking a granularity for the ratio.

Extend the /sys/kernel/uids/<uid>/ interface to allow setting
the group's rt_runtime.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosched: rt-group: deal with PI
Peter Zijlstra [Wed, 13 Feb 2008 14:45:39 +0000 (15:45 +0100)]
sched: rt-group: deal with PI

Steven mentioned the fun case where a lock holding task will be throttled.

Simple fix: allow groups that have boosted tasks to run anyway.

If a runnable task in a throttled group gets boosted the dequeue/enqueue
done by rt_mutex_setprio() is enough to unthrottle the group.

This is ofcourse not quite correct. Two possible ways forward are:
  - second prio array for boosted tasks
  - boost to a prio ceiling (this would also work for deadline scheduling)

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosched: fix incorrect irq lock usage in normalize_rt_tasks()
Peter Zijlstra [Wed, 13 Feb 2008 14:45:39 +0000 (15:45 +0100)]
sched: fix incorrect irq lock usage in normalize_rt_tasks()

lockdep spotted this bogus irq locking. normalize_rt_tasks() can be called
from hardirq context through sysrq-n

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agosched: fair-group: separate tg->shares from task_group_lock
Peter Zijlstra [Wed, 13 Feb 2008 14:45:39 +0000 (15:45 +0100)]
sched: fair-group: separate tg->shares from task_group_lock

On Mon, 2008-02-11 at 15:09 +0300, Denis V. Lunev wrote:
> BUG: sleeping function called from invalid context
> at /home/den/src/linux-netns26/kernel/mutex.c:209
> in_atomic():1, irqs_disabled():0
> no locks held by swapper/0.
> Pid: 0, comm: swapper Not tainted 2.6.24 #304
>
> Call Trace:
>  <IRQ>  [<ffffffff80252d1e>] ? __debug_show_held_locks+0x15/0x27
>  [<ffffffff8022c2a8>] __might_sleep+0xc0/0xdf
>  [<ffffffff8049f1df>] mutex_lock_nested+0x28/0x2a9
>  [<ffffffff80231294>] sched_destroy_group+0x18/0xea
>  [<ffffffff8023e835>] sched_destroy_user+0xd/0xf
>  [<ffffffff8023e8c1>] free_uid+0x8a/0xab
>  [<ffffffff80233e24>] __put_task_struct+0x3f/0xd3
>  [<ffffffff80236708>] delayed_put_task_struct+0x23/0x25
>  [<ffffffff8026fda7>] __rcu_process_callbacks+0x8d/0x215
>  [<ffffffff8026ff52>] rcu_process_callbacks+0x23/0x44
>  [<ffffffff8023a2ae>] __do_softirq+0x79/0xf8
>  [<ffffffff8020f8c3>] ? profile_pc+0x2a/0x67
>  [<ffffffff8020d38c>] call_softirq+0x1c/0x30
>  [<ffffffff8020f689>] do_softirq+0x61/0x9c
>  [<ffffffff8023a233>] irq_exit+0x51/0x53
>  [<ffffffff8021bd1a>] smp_apic_timer_interrupt+0x77/0xad
>  [<ffffffff8020ce3b>] apic_timer_interrupt+0x6b/0x70
>  <EOI>  [<ffffffff8020b0dd>] ? default_idle+0x43/0x76
>  [<ffffffff8020b0db>] ? default_idle+0x41/0x76
>  [<ffffffff8020b09a>] ? default_idle+0x0/0x76
>  [<ffffffff8020b186>] ? cpu_idle+0x76/0x98

separate the tg->shares protection from the task_group lock.

Reported-by: Denis V. Lunev <den@openvz.org>
Tested-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agohrtimer: more hrtimer_init_sleeper() fallout.
Peter Zijlstra [Wed, 13 Feb 2008 14:45:36 +0000 (15:45 +0100)]
hrtimer: more hrtimer_init_sleeper() fallout.

Missed an instance...

  futex_lock_pi()
    hrtimer_init_sleeper()
    rt_mutex_timed_lock()
      rt_mutex_timed_fastlock()
        rt_mutex_slowlock()
          hrtimer_start()

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agotimeconst.pl: correct reversal of USEC_TO_HZ and HZ_TO_USEC
H. Peter Anvin [Tue, 12 Feb 2008 21:52:37 +0000 (13:52 -0800)]
timeconst.pl: correct reversal of USEC_TO_HZ and HZ_TO_USEC

The USEC_TO_HZ and HZ_TO_USEC constant sets were mislabelled, with
seriously incorrect results.  This among other things manifested
itself as cpufreq not working when a tickless kernel was configured.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Tested-by: Carlos R. Mafra <crmafra@ift.unesp.br>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoWMI: initialize wmi_blocks.list even if ACPI is disabled
Linus Torvalds [Tue, 12 Feb 2008 04:52:01 +0000 (20:52 -0800)]
WMI: initialize wmi_blocks.list even if ACPI is disabled

Even if we don't want to register the WMI driver, we should initialize
the wmi_blocks list to be empty, since we don't want the wmi helper
functions to oops just because that basic list has not even been set up.

With this, "find_guid()" will happily return "not found" rather than
oopsing all over the place, and the callers will then just automatically
return false or AE_NOT_FOUND as appropriate.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agox86: vdso_install fix
Roland McGrath [Mon, 11 Feb 2008 22:38:51 +0000 (14:38 -0800)]
x86: vdso_install fix

The makefile magic for installing the 32-bit vdso images on disk had a
little error.  A single-line change would fix that bug, but this does a
little more to reduce the error-prone duplication of this bit of
makefile variable magic.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agomempolicy: silently restrict nodemask to allowed nodes
KOSAKI Motohiro [Tue, 12 Feb 2008 04:30:22 +0000 (13:30 +0900)]
mempolicy: silently restrict nodemask to allowed nodes

Kosaki Motohito noted that "numactl --interleave=all ..." failed in the
presence of memoryless nodes.  This patch attempts to fix that problem.

Some background:

numactl --interleave=all calls set_mempolicy(2) with a fully populated
[out to MAXNUMNODES] nodemask.  set_mempolicy() [in do_set_mempolicy()]
calls contextualize_policy() which requires that the nodemask be a
subset of the current task's mems_allowed; else EINVAL will be returned.

A task's mems_allowed will always be a subset of node_states[N_HIGH_MEMORY]
i.e., nodes with memory.  So, a fully populated nodemask will be
declared invalid if it includes memoryless nodes.

  NOTE:  the same thing will occur when running in a cpuset
         with restricted mem_allowed--for the same reason:
         node mask contains dis-allowed nodes.

mbind(2), on the other hand, just masks off any nodes in the nodemask
that are not included in the caller's mems_allowed.

In each case [mbind() and set_mempolicy()], mpol_check_policy() will
complain [again, resulting in EINVAL] if the nodemask contains any
memoryless nodes.  This is somewhat redundant as mpol_new() will remove
memoryless nodes for interleave policy, as will bind_zonelist()--called
by mpol_new() for BIND policy.

Proposed fix:

1) modify contextualize_policy logic to:
   a) remember whether the incoming node mask is empty.
   b) if not, restrict the nodemask to allowed nodes, as is
      currently done in-line for mbind().  This guarantees
      that the resulting mask includes only nodes with memory.

      NOTE:  this is a [benign, IMO] change in behavior for
             set_mempolicy().  Dis-allowed nodes will be
             silently ignored, rather than returning an error.

   c) fold this code into mpol_check_policy(), replace 2 calls to
      contextualize_policy() to call mpol_check_policy() directly
      and remove contextualize_policy().

2) In existing mpol_check_policy() logic, after "contextualization":
   a) MPOL_DEFAULT:  require that in coming mask "was_empty"
   b) MPOL_{BIND|INTERLEAVE}:  require that contextualized nodemask
      contains at least one node.
   c) add a case for MPOL_PREFERRED:  if in coming was not empty
      and resulting mask IS empty, user specified invalid nodes.
      Return EINVAL.
   c) remove the now redundant check for memoryless nodes

3) remove the now redundant masking of policy nodes for interleave
   policy from mpol_new().

4) Now that mpol_check_policy() contextualizes the nodemask, remove
   the in-line nodes_and() from sys_mbind().  I believe that this
   restores mbind() to the behavior before the memoryless-nodes
   patch series.  E.g., we'll no longer treat an invalid nodemask
   with MPOL_PREFERRED as local allocation.

[ Patch history:

  v1 -> v2:
   - Communicate whether or not incoming node mask was empty to
     mpol_check_policy() for better error checking.
   - As suggested by David Rientjes, remove the now unused
     cpuset_nodes_subset_current_mems_allowed() from cpuset.h

  v2 -> v3:
   - As suggested by Kosaki Motohito, fold the "contextualization"
     of policy nodemask into mpol_check_policy().  Looks a little
     cleaner. ]

Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Tested-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6
Linus Torvalds [Tue, 12 Feb 2008 04:44:58 +0000 (20:44 -0800)]
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 build for sim_defconfig

16 years agoBe more robust about bad arguments in get_user_pages()
Jonathan Corbet [Mon, 11 Feb 2008 23:17:33 +0000 (16:17 -0700)]
Be more robust about bad arguments in get_user_pages()

So I spent a while pounding my head against my monitor trying to figure
out the vmsplice() vulnerability - how could a failure to check for
*read* access turn into a root exploit? It turns out that it's a buffer
overflow problem which is made easy by the way get_user_pages() is
coded.

In particular, "len" is a signed int, and it is only checked at the
*end* of a do {} while() loop.  So, if it is passed in as zero, the loop
will execute once and decrement len to -1.  At that point, the loop will
proceed until the next invalid address is found; in the process, it will
likely overflow the pages array passed in to get_user_pages().

I think that, if get_user_pages() has been asked to grab zero pages,
that's what it should do.  Thus this patch; it is, among other things,
enough to block the (already fixed) root exploit and any others which
might be lurking in similar code.  I also think that the number of pages
should be unsigned, but changing the prototype of this function probably
requires some more careful review.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Tue, 12 Feb 2008 04:43:14 +0000 (20:43 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/roland/infiniband

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  mlx4_core: Fix build break (missing include)

16 years agoAdd Matt to MAINTAINERS as a SLAB allocator maintainer
Pekka Enberg [Mon, 11 Feb 2008 21:52:47 +0000 (23:52 +0200)]
Add Matt to MAINTAINERS as a SLAB allocator maintainer

Matt is already the maintainer of SLOB which is one of the "SLAB" allocators in
the kernel so add him to MAINTAINERS.

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzi...
Linus Torvalds [Tue, 12 Feb 2008 04:42:11 +0000 (20:42 -0800)]
Merge branch 'upstream-fixes' of git://git./linux/kernel/git/jgarzik/libata-dev

* 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  sata_mv: platform driver allocs dma without create
  pata_ninja32: setup changes
  pata_legacy: typo fix
  pata_amd: Note in the module description it handles Nvidia
  sata_mv: fix loop with last port
  libata: ignore deverr on SETXFER if mode is configured
  pata_via: fix SATA cable detection on cx700

16 years agoMake topology fallback macros reference their arguments.
Andi Kleen [Mon, 11 Feb 2008 19:03:17 +0000 (20:03 +0100)]
Make topology fallback macros reference their arguments.

This avoids warnings with unreferenced variables in the !NUMA case.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agomlx4_core: Fix build break (missing include)
Olof Johansson [Mon, 11 Feb 2008 02:22:57 +0000 (20:22 -0600)]
mlx4_core: Fix build break (missing include)

Commit 313abe55 ("mlx4_core: For 64-bit systems, vmap() kernel queue
buffers") caused this to pop up on powerpc allyesconfig, looks like a
missing include file:

    drivers/net/mlx4/alloc.c: In function 'mlx4_buf_alloc':
    drivers/net/mlx4/alloc.c:162: error: implicit declaration of function 'vmap'
    drivers/net/mlx4/alloc.c:162: error: 'VM_MAP' undeclared (first use in this function)
    drivers/net/mlx4/alloc.c:162: error: (Each undeclared identifier is reported only once
    drivers/net/mlx4/alloc.c:162: error: for each function it appears in.)
    drivers/net/mlx4/alloc.c:162: warning: assignment makes pointer from integer without a cast
    drivers/net/mlx4/alloc.c: In function 'mlx4_buf_free':
    drivers/net/mlx4/alloc.c:187: error: implicit declaration of function 'vunmap'

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
16 years ago[IA64] Fix build for sim_defconfig
Tony Luck [Mon, 11 Feb 2008 21:23:46 +0000 (13:23 -0800)]
[IA64] Fix build for sim_defconfig

Commit bdc807871d58285737d50dc6163d0feb72cb0dc2 broke the build
for this config because the sim_defconfig selects CONFIG_HZ=250
but include/asm-ia64/param.h has an ifdef for the simulator to
force HZ to 32.  So we ended up with a kernel/timeconst.h set
for HZ=250 ... which then failed the check for the right HZ
value and died with:

Drop the #ifdef magic from param.h and make force CONFIG_HZ=32
directly for the simulator.

Signed-off-by: Tony Luck <tony.luck@intel.com>
16 years agosata_mv: platform driver allocs dma without create
Byron Bradley [Sun, 10 Feb 2008 21:17:30 +0000 (21:17 +0000)]
sata_mv: platform driver allocs dma without create

When the sata_mv driver is used as a platform driver,
mv_create_dma_pools() is never called so it fails when trying
to alloc in mv_pool_start().

Signed-off-by: Byron Bradley <byron.bbradley@gmail.com>
Acked-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agopata_ninja32: setup changes
Alan Cox [Fri, 8 Feb 2008 15:25:10 +0000 (15:25 +0000)]
pata_ninja32: setup changes

Forcibly set more of the configuration at init time. This seems to fix at
least one problem reported. We don't know what most of these bits do, but
we do know what windows stuffs there.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agopata_legacy: typo fix
Alan Cox [Fri, 8 Feb 2008 15:23:38 +0000 (15:23 +0000)]
pata_legacy: typo fix

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agopata_amd: Note in the module description it handles Nvidia
Alan Cox [Fri, 8 Feb 2008 15:22:39 +0000 (15:22 +0000)]
pata_amd: Note in the module description it handles Nvidia

This has confused a few people so fix it

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agosata_mv: fix loop with last port
Yinghai Lu [Thu, 7 Feb 2008 23:06:17 +0000 (15:06 -0800)]
sata_mv: fix loop with last port

commit f351b2d638c3cb0b95adde3549b7bfaf3f991dfa
        sata_mv: Support SoC controllers

cause panic:

scsi 4:0:0:0: Direct-Access     ATA      HITACHI HDS7225S V44O PQ: 0 ANSI: 5
sd 4:0:0:0: [sde] 488390625 512-byte hardware sectors (250056 MB)
sd 4:0:0:0: [sde] Write Protect is off
sd 4:0:0:0: [sde] Mode Sense: 00 3a 00 00
sd 4:0:0:0: [sde] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
sd 4:0:0:0: [sde] 488390625 512-byte hardware sectors (250056 MB)
sd 4:0:0:0: [sde] Write Protect is off
sd 4:0:0:0: [sde] Mode Sense: 00 3a 00 00
sd 4:0:0:0: [sde] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
 sde:<1>BUG: unable to handle kernel NULL pointer dereference at 000000000000001a
IP: [<ffffffff806262c7>] mv_interrupt+0x21c/0x4cc
PGD 0
Oops: 0000 [1] SMP
CPU 3
Modules linked in:
Pid: 0, comm: swapper Not tainted 2.6.24-smp-08636-g0afc2ed-dirty #26
RIP: 0010:[<ffffffff806262c7>]  [<ffffffff806262c7>] mv_interrupt+0x21c/0x4cc
RSP: 0000:ffff8102050bbec8  EFLAGS: 00010297
RAX: 0000000000000008 RBX: 0000000000000000 RCX: 0000000000000003
RDX: 0000000000008000 RSI: 0000000000000286 RDI: ffff8102035180e0
RBP: 0000000000000001 R08: 0000000000000003 R09: ffff8102036613e0
R10: 0000000000000002 R11: ffffffff8061474c R12: ffff8102035bf828
R13: 0000000000000008 R14: ffff81020348ece8 R15: ffffc20002cb2000
FS:  0000000000000000(0000) GS:ffff810405025700(0000) knlGS:0000000000000000
CS:  0010 DS: 0018 ES: 0018 CR0: 000000008005003b
CR2: 000000000000001a CR3: 0000000000201000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process swapper (pid: 0, threadinfo ffff810405094000, task ffff8102050b28c0)
Stack:  000000010000000c 0002040000220400 0000001100000002 ffff81020348eda8
 0000000000000001 ffff8102035f2cc0 0000000000000000 0000000000000000
 0000000000000018 0000000000000000 0000000000000000 ffffffff80269ee8
Call Trace:
 <IRQ>  [<ffffffff80269ee8>] ? handle_IRQ_event+0x25/0x53
 [<ffffffff8026b393>] ? handle_fasteoi_irq+0x90/0xc8
 [<ffffffff802218e2>] ? do_IRQ+0xf1/0x15f
 [<ffffffff8021df24>] ? default_idle+0x0/0x55
 [<ffffffff8021f361>] ? ret_from_intr+0x0/0xa
 <EOI>  [<ffffffff8023010c>] ? lapic_next_event+0x0/0xa
 [<ffffffff8021df55>] ? default_idle+0x31/0x55
 [<ffffffff8021df50>] ? default_idle+0x2c/0x55
 [<ffffffff8021df24>] ? default_idle+0x0/0x55
 [<ffffffff8021e00b>] ? cpu_idle+0x92/0xb8

Code: 41 14 85 c0 89 44 24 14 0f 84 9d 02 00 00 f7 d0 01 d6 41 89 d5 89 41 14 8b 41 14 89 34 24 e9 7e 02 00 00 49 63 c5 49 8b 5c c6 48 <f6> 43 1a 80 4c 8b a3 20 37 00 00 0f 85 62 02 00 00 31 c9 41 83
RIP  [<ffffffff806262c7>] mv_interrupt+0x21c/0x4cc
 RSP <ffff8102050bbec8>
CR2: 000000000000001a
---[ end trace 2583b5f7a5350584 ]---
Kernel panic - not syncing: Aiee, killing interrupt handler!

last_port already include port0 base.
this patch change use last_port directly, and move pp assignment later.

Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agolibata: ignore deverr on SETXFER if mode is configured
Tejun Heo [Thu, 7 Feb 2008 01:34:08 +0000 (10:34 +0900)]
libata: ignore deverr on SETXFER if mode is configured

Some controllers (VIA CX700) raise device error on SETXFER even after
mode configuration succeeded.  Update ata_dev_set_mode() such that
device error is ignored if transfer mode is configured correctly.  To
implement this, device is revalidated even after device error on
SETXFER.

This fixes kernel bugzilla bug 8563.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agopata_via: fix SATA cable detection on cx700
Tejun Heo [Thu, 7 Feb 2008 01:18:53 +0000 (10:18 +0900)]
pata_via: fix SATA cable detection on cx700

The first port of cx700 is SATA.  Fix cable detection.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
16 years agox86: remove over noisy debug printk
Thomas Gleixner [Sun, 10 Feb 2008 22:57:36 +0000 (23:57 +0100)]
x86: remove over noisy debug printk

pageattr-test.c contains a noisy debug printk that people reported.
The condition under which it prints (randomly tapping into a mem_map[]
hole and not being able to c_p_a() there) is valid behavior and not
interesting to report.

Remove it.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild
Linus Torvalds [Mon, 11 Feb 2008 17:22:23 +0000 (09:22 -0800)]
Merge git://git./linux/kernel/git/sam/kbuild

* git://git.kernel.org/pub/scm/linux/kernel/git/sam/kbuild:
  kbuild: fix make V=1

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris...
Linus Torvalds [Mon, 11 Feb 2008 17:21:26 +0000 (09:21 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jmorris/selinux-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/selinux-2.6:
  selinux: support 64-bit capabilities

16 years agoPrevent IDE boot ops on NUMA system
Andi Kleen [Mon, 11 Feb 2008 00:35:20 +0000 (01:35 +0100)]
Prevent IDE boot ops on NUMA system

Without this patch a Opteron test system here oopses at boot with
current git.

Calling to_pci_dev() on a NULL pointer gives a negative value so the
following NULL pointer check never triggers and then an illegal address
is referenced.  Check the unadjusted original device pointer for NULL
instead.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'for-linus' of git://linux-nfs.org/~bfields/linux
Linus Torvalds [Mon, 11 Feb 2008 17:19:47 +0000 (09:19 -0800)]
Merge branch 'for-linus' of git://linux-nfs.org/~bfields/linux

* 'for-linus' of git://linux-nfs.org/~bfields/linux:
  SUNPRC: Fix printk format warning
  nfsd: clean up svc_reserve_auth()
  NLM: don't requeue block if it was invalidated while GRANT_MSG was in flight
  NLM: don't reattempt GRANT_MSG when there is already an RPC in flight
  NLM: have server-side RPC clients default to soft RPC tasks
  NLM: set RPC_CLNT_CREATE_NOPING for NLM RPC clients

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6
Linus Torvalds [Mon, 11 Feb 2008 17:19:22 +0000 (09:19 -0800)]
Merge git://git./linux/kernel/git/bart/ide-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6:
  ide: remove stale comment from ide-lib.c
  ide: fix comment in init_irq()
  ide: ide_init_port() bugfix
  ide-disk: fix flush requests (take 2)
  ide: introduce CONFIG_BLK_DEV_IDEDMA_SFF option
  bast-ide: build fix
  ide-tape: remove never executed code
  ide: fix ide/legacy/gayle.c compilation
  ide-cd: replace ntohs with generic byteorder macro be16_to_cpu
  ide: remove stale version number
  pdc202xx_old: always enable burst mode
  palm_bk3710: use struct ide_port_info
  palm_bk3710: port initialization/probing bugfix
  palm_bk3710: fix ide_unregister() usage
  palm_bk3710: ide_register_hw() -> ide_device_add()
  ide: insert BUG_ON() into __ide_set_handler() (take 2)
  cs5520: remove stale comment
  ide: another possible ide panic fix for blk-end-request

16 years agokbuild: fix make V=1
Sam Ravnborg [Mon, 11 Feb 2008 13:26:26 +0000 (14:26 +0100)]
kbuild: fix make V=1

When make -s support were added to filechk to
combination created with make V=1 were not
covered.
Fix it by explicitly cover this case too.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Mike Frysinger <vapier@gentoo.org>
16 years agoUse proper abstractions in quirk_intel_irqbalance
Matthew Wilcox [Mon, 11 Feb 2008 04:18:15 +0000 (23:18 -0500)]
Use proper abstractions in quirk_intel_irqbalance

Since we may not have a pci_dev for the device we need to access, we can't
use pci_read_config_word.  But raw_pci_read is an internal implementation
detail; it's better to use the architected pci_bus_read_config_word
interface.  Using PCI_DEVFN instead of a mysterious constant helps
reassure everyone that we really do intend to access device 8.

[ Thanks to Grant Grundler for pointing out to me that this is exactly
  what the write immediately above this is doing -- enabling device 8 to
  respond to config space cycles.
- Matthew

  Grant also says:

"Can you also add a comment which points at the Intel
 documentation?

 The 'Intel E7320 Memory Controller Hub (MCH) Datasheet' at

  http://download.intel.com/design/chipsets/datashts/30300702.pdf

 Page 69 documents register F4h (DEVPRES1).

 And I just doubled checked that the 0xf4 register value is
 restored later in the quirk (obvious when you look at the code
 but not from the patch"

  so here it is.
 - Linus ]

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Acked-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoselinux: support 64-bit capabilities
Stephen Smalley [Thu, 7 Feb 2008 16:21:04 +0000 (11:21 -0500)]
selinux: support 64-bit capabilities

Fix SELinux to handle 64-bit capabilities correctly, and to catch
future extensions of capabilities beyond 64 bits to ensure that SELinux
is properly updated.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
16 years agoide: remove stale comment from ide-lib.c
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:15 +0000 (00:32 +0100)]
ide: remove stale comment from ide-lib.c

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: fix comment in init_irq()
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:15 +0000 (00:32 +0100)]
ide: fix comment in init_irq()

APUS support is gone...

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: ide_init_port() bugfix
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:15 +0000 (00:32 +0100)]
ide: ide_init_port() bugfix

On Sunday 10 February 2008, Atsushi Nemoto wrote:
> On Sun, 06 Jan 2008 18:03:10 +0100, Bartlomiej Zolnierkiewicz <bzolnier@gmail.com> wrote:
> > + /* reset DMA masks only for SFF-style DMA controllers */
> > + if ((d->host_flags && IDE_HFLAG_NO_DMA) == 0 && hwif->dma_base == 0)
> > + hwif->swdma_mask = hwif->mwdma_mask = hwif->ultra_mask = 0;
>
> It might be too late, but "host_flags && IDE_HFLAGS_NO_DMA" seems
> wrong for me.

Fix regression caused by commmit c413b9b94d9a8e7548cc4b2e04b7df0439ce76fd
("ide: add struct ide_port_info instances to legacy host drivers").

Reported-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-disk: fix flush requests (take 2)
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:14 +0000 (00:32 +0100)]
ide-disk: fix flush requests (take 2)

commit 813a0eb233ee67d7166241a8b389b6a76f2247f9
Author: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date:   Fri Jan 25 22:17:10 2008 +0100

    ide: switch idedisk_prepare_flush() to use REQ_TYPE_ATA_TASKFILE requests

...

broke flush requests.

Allocating IDE command structure on the stack for flush requests is not
a very brilliant idea:

- idedisk_prepare_flush() only prepares the request and it doesn't wait
  for it to be completed

- there are can be multiple flush requests queued in the queue

Fix the problem (per hints from James Bottomley) by:
- dynamically allocating ide_task_t instance using kmalloc(..., GFP_ATOMIC)
- adding new taskfile flag (IDE_TFLAG_DYN)
- calling kfree() in ide_end_drive_command() if IDE_TFLAG_DYN is set
  (while at it rename 'args' to 'task' and fix whitespace damage)

[ This will be fixed properly before 2.6.25 but this bug is rather
  critical and the proper solution requires some more work + testing. ]

Thanks to Sebastian Siewior and Christoph Hellwig for reporting the
problem and testing patches (extra thanks to Sebastian for bisecting
it to the guilty commmit).

Tested-by: Sebastian Siewior <ide-bug@ml.breakpoint.cc>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Tejun Heo <htejun@gmail.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: introduce CONFIG_BLK_DEV_IDEDMA_SFF option
Sergei Shtylyov [Sun, 10 Feb 2008 23:32:14 +0000 (00:32 +0100)]
ide: introduce CONFIG_BLK_DEV_IDEDMA_SFF option

Introduce new option CONFIG_BLK_DEV_IDEDMA_SFF for non-PCI SFF-8038i compatible
bus mastering IDE controllers (which there are a few known), thus fixing a hack
made for Palmchip BK3710 controller...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Anton Salnikov <asalnikov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agobast-ide: build fix
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:14 +0000 (00:32 +0100)]
bast-ide: build fix

On Saturday 09 February 2008, Adrian Bunk wrote:
> Commit 9e016a719209d95338e314b46c3012cc7feaaeec causes the following
> compile error:
>
> <--  snip  -->
>
> ...
>   CC      drivers/ide/arm/bast-ide.o
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/arm/bast-ide.c: In function 'bastide_register':
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/arm/bast-ide.c:31: error: 'hwif' redeclared as different kind of symbol
> /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/arm/bast-ide.c:29: error: previous definition of 'hwif' was here
> make[4]: *** [drivers/ide/arm/bast-ide.o] Error 1
>
> <--  snip  -->

Remove 'ide_hwif_t **hwif' argument from bastide_register()
(together with write-only ifs[]).

Cc: Adrian Bunk <bunk@kernel.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: remove never executed code
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:14 +0000 (00:32 +0100)]
ide-tape: remove never executed code

rq->cmd[0] is never set to REQ_IDETAPE_READ_BUFFER so remove
REQ_IDETAPE_READ_BUFFER handling from idetape_create_write_cmd()
and the define itself.

Then remove no longer used idetape_create_read_buffer_cmd()
and IDETAPE_RETRIEVE_FAULTY_BLOCK define.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: fix ide/legacy/gayle.c compilation
Adrian Bunk [Sun, 10 Feb 2008 23:32:14 +0000 (00:32 +0100)]
ide: fix ide/legacy/gayle.c compilation

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-cd: replace ntohs with generic byteorder macro be16_to_cpu
Borislav Petkov [Sun, 10 Feb 2008 23:32:13 +0000 (00:32 +0100)]
ide-cd: replace ntohs with generic byteorder macro be16_to_cpu

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove stale version number
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:13 +0000 (00:32 +0100)]
ide: remove stale version number

On Thursday 03 January 2008, Robert Hancock wrote:

[...]

> How about getting rid of this stupid thing in drivers/ide/ide.c:
>
> #define       REVISION        "Revision: 7.00alpha2"
>
> which is used in:
>
> printk(KERN_INFO "Uniform Multi-Platform E-IDE driver " REVISION "\n");
>
> It's been 7.00alpha2 for god knows how long, so clearly this version
> number is not useful..

Cc: Robert Hancock <hancockr@shaw.ca>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agopdc202xx_old: always enable burst mode
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:13 +0000 (00:32 +0100)]
pdc202xx_old: always enable burst mode

Alan has noticed that distros always enabled burst mode
(+ datasheet confirms that it is the right thing to do).

Thus fix pdc202xx_old host driver to do it unconditionally
and remove no longer needed CONFIG_PDC202XX_BURST option.

Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agopalm_bk3710: use struct ide_port_info
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:13 +0000 (00:32 +0100)]
palm_bk3710: use struct ide_port_info

* Factor out cable detection to palm_bk3710_cable_detect().

* Add palm_bk3710_init_hwif() (->init_hwif method implementation).

* Remove needless ->quirkproc initialization.

* Add missing ->pio_mask initialization.

* Use ATA_* defines for setting ->{ultra,mwdma}_mask.

* Add 'struct ide_port_info palm_bk3710_port_info' and pass it to
  ide_device_add().  Then remove open-coded 'hwif' initialization.

Cc: Anton Salnikov <asalnikov@ru.mvista.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agopalm_bk3710: port initialization/probing bugfix
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:13 +0000 (00:32 +0100)]
palm_bk3710: port initialization/probing bugfix

Probe port _after_ it is fully initialized.

Cc: Anton Salnikov <asalnikov@ru.mvista.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agopalm_bk3710: fix ide_unregister() usage
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:12 +0000 (00:32 +0100)]
palm_bk3710: fix ide_unregister() usage

Don't set 'restore' flag for ide_unregister() when initializing new
interface.

[ identical change as done to bast-ide/ide-cs/delkin_cb host drivers
  by commit 909f4369bca30f9a186316a3bf2b4a9c1e702a25 ]

Cc: Anton Salnikov <asalnikov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agopalm_bk3710: ide_register_hw() -> ide_device_add()
Bartlomiej Zolnierkiewicz [Sun, 10 Feb 2008 23:32:12 +0000 (00:32 +0100)]
palm_bk3710: ide_register_hw() -> ide_device_add()

* Convert palm_bk3710 host driver to use ide_device_add() instead of
  ide_register_hw() (while at it drop doing "ide_unregister()" loop which
  tries to unregister _all_ IDE interfaces if useable ide_hwifs[] slot
  cannot be find).

  [ identical change as done to bast-ide/ide-cs/delkin_cb host drivers
    by commit 9e016a719209d95338e314b46c3012cc7feaaeec ]

* Rename 'ide_ctlr_info' to 'hw' and 'index' to 'i' while at it.

Cc: Anton Salnikov <asalnikov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: insert BUG_ON() into __ide_set_handler() (take 2)
Sergei Shtylyov [Sun, 10 Feb 2008 23:32:12 +0000 (00:32 +0100)]
ide: insert BUG_ON() into __ide_set_handler() (take 2)

Replace the check for hwgroup->handler and printk(KERN_CRIT, ...) at the start
of __ide_set_handler() with mere BUG_ON() while removing such from the caller,
ide_execute_command(). Fix up the code formatting, while at it...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agocs5520: remove stale comment
Benjamin Herrenschmidt [Sun, 10 Feb 2008 23:32:12 +0000 (00:32 +0100)]
cs5520: remove stale comment

Remove stale comment from the cs5520 IDE driver.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: another possible ide panic fix for blk-end-request
Kiyoshi Ueda [Sun, 10 Feb 2008 23:32:11 +0000 (00:32 +0100)]
ide: another possible ide panic fix for blk-end-request

I have reviewed all blk-end-request patches again to confirm whether
there are any similar problems with the last week's ide-cd panic:
    http://lkml.org/lkml/2008/1/29/140

And I found a possible similar bug in ide-io change:
ide_end_drive_cmd() could be called for blk_pc_request() which could
have bios.  To complete such requests correctly, we need to pass
the actual size of the request.
Otherwise, __blk_end_request() returns 1 because the request still has
bios, and the system will BUG() unnecessarily.

The following patch fixes the bug and should be applied on top of
Linus' git.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Cc: Borislav Petkov <petkovbb@googlemail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoSUNPRC: Fix printk format warning
Roland Dreier [Sat, 9 Feb 2008 00:02:04 +0000 (16:02 -0800)]
SUNPRC: Fix printk format warning

net/sunrpc/xprtrdma/svc_rdma_sendto.c:160: warning: format '%llx'
expects type 'long long unsigned int', but argument 3 has type 'u64'

Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
16 years agonfsd: clean up svc_reserve_auth()
J. Bruce Fields [Fri, 8 Feb 2008 04:10:21 +0000 (23:10 -0500)]
nfsd: clean up svc_reserve_auth()

This is a void function attempting to return the return value from
another void function, which seems harmless but extremely weird, and
apparently makes some compilers complain.

While we're there, clean up a little (e.g. the switch statement had a
minor style problem and seemed overkill as long as there's only one
case).

Thanks to Trond for noticing this.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
16 years agoNLM: don't requeue block if it was invalidated while GRANT_MSG was in flight
Jeff Layton [Wed, 6 Feb 2008 16:34:13 +0000 (11:34 -0500)]
NLM: don't requeue block if it was invalidated while GRANT_MSG was in flight

It's possible for lockd to catch a SIGKILL while a GRANT_MSG callback
is in flight. If this happens we don't want lockd to insert the block
back into the nlm_blocked list.

This helps that situation, but there's still a possible race. Fixing
that will mean adding real locking for nlm_blocked.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
16 years agoNLM: don't reattempt GRANT_MSG when there is already an RPC in flight
Jeff Layton [Wed, 6 Feb 2008 16:34:12 +0000 (11:34 -0500)]
NLM: don't reattempt GRANT_MSG when there is already an RPC in flight

With the current scheme in nlmsvc_grant_blocked, we can end up with more
than one GRANT_MSG callback for a block in flight. Right now, we requeue
the block unconditionally so that a GRANT_MSG callback is done again in
30s. If the client is unresponsive, it can take more than 30s for the
call already in flight to time out.

There's no benefit to having more than one GRANT_MSG RPC queued up at a
time, so put it on the list with a timeout of NLM_NEVER before doing the
RPC call. If the RPC call submission fails, we requeue it with a short
timeout. If it works, then nlmsvc_grant_callback will end up requeueing
it with a shorter timeout after it completes.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
16 years agoNLM: have server-side RPC clients default to soft RPC tasks
Jeff Layton [Wed, 6 Feb 2008 16:34:11 +0000 (11:34 -0500)]
NLM: have server-side RPC clients default to soft RPC tasks

Now that it no longer does an RPC ping, lockd always ends up queueing
an RPC task for the GRANT_MSG callback. But, it also requeues the block
for later attempts. Since these are hard RPC tasks, if the client we're
calling back goes unresponsive the GRANT_MSG callbacks can stack up in
the RPC queue.

Fix this by making server-side RPC clients default to soft RPC tasks.
lockd requeues the block anyway, so this should be OK.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
16 years agoNLM: set RPC_CLNT_CREATE_NOPING for NLM RPC clients
Jeff Layton [Wed, 6 Feb 2008 16:34:10 +0000 (11:34 -0500)]
NLM: set RPC_CLNT_CREATE_NOPING for NLM RPC clients

It's currently possible for an unresponsive NLM client to completely
lock up a server's lockd. The scenario is something like this:

1) client1 (or a process on the server) takes a lock on a file
2) client2 tries to take a blocking lock on the same file and
   awaits the callback
3) client2 goes unresponsive (plug pulled, network partition, etc)
4) client1 releases the lock

...at that point the server's lockd will try to queue up a GRANT_MSG
callback for client2, but first it requeues the block with a timeout of
30s. nlm_async_call will attempt to bind the RPC client to client2 and
will call rpc_ping. rpc_ping entails a sync RPC call and if client2 is
unresponsive it will take around 60s for that to time out. Once it times
out, it's already time to retry the block and the whole process repeats.

Once in this situation, nlmsvc_retry_blocked will never return until
the host starts responding again. lockd won't service new calls.

Fix this by skipping the RPC ping on NLM RPC clients. This makes
nlm_async_call return quickly when called.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
16 years agoLinux 2.6.25-rc1 v2.6.25-rc1
Linus Torvalds [Sun, 10 Feb 2008 22:18:14 +0000 (14:18 -0800)]
Linux 2.6.25-rc1

.. and I really need to call it something else.  Maybe it is time to
bring back the weasel series, since weasels always make me feel good
about a kernel.

16 years agoMerge branch 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Sun, 10 Feb 2008 22:09:44 +0000 (14:09 -0800)]
Merge branch 'for-linus' of /home/rmk/linux-2.6-arm

* 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm: (30 commits)
  [ARM] constify function pointer tables
  [ARM] 4823/1: AT91 section fix
  [ARM] 4824/1: pxa: clear RDH bit after any reset
  [ARM] pxa: remove debugging PM: printk
  ARM: OMAP1: Misc clean-up
  ARM: OMAP1: Update defconfigs for omap1
  ARM: OMAP1: Palm Tungsten E board clean-up
  ARM: OMAP1: Use I2C bus registration helper for omap1
  ARM: OMAP1: Remove omap_sram_idle()
  ARM: OMAP1: PM fixes for OMAP1
  ARM: OMAP1: Use MMC multislot structures for Siemens SX1 board
  ARM: OMAP1: Make omap1 use MMC multislot structures
  ARM: OMAP1: Change the comments to C style
  ARM: OMAP1: Make omap1 boards to use omap_nand_platform_data
  ARM: OMAP: Add helper module for board specific I2C bus registration
  ARM: OMAP: Add dmtimer support for OMAP3
  ARM: OMAP: Pre-3430 clean-up for dmtimer.c
  ARM: OMAP: Add DMA support for chaining and 3430
  ARM: OMAP: Add 24xx GPIO debounce support
  ARM: OMAP: Get rid of unnecessary ifdefs in GPIO code
  ...

16 years agoChange pci_raw_ops to pci_raw_read/write
Matthew Wilcox [Sun, 10 Feb 2008 14:45:28 +0000 (09:45 -0500)]
Change pci_raw_ops to pci_raw_read/write

We want to allow different implementations of pci_raw_ops for standard
and extended config space on x86.  Rather than clutter generic code with
knowledge of this, we make pci_raw_ops private to x86 and use it to
implement the new raw interface -- raw_pci_read() and raw_pci_write().

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoPCI x86: always use conf1 to access config space below 256 bytes
Ivan Kokshaysky [Mon, 14 Jan 2008 22:31:09 +0000 (17:31 -0500)]
PCI x86: always use conf1 to access config space below 256 bytes

Thanks to Loic Prylli <loic@myri.com>, who originally proposed
this idea.

Always using legacy configuration mechanism for the legacy config space
and extended mechanism (mmconf) for the extended config space is
a simple and very logical approach. It's supposed to resolve all
known mmconf problems. It still allows per-device quirks (tweaking
dev->cfg_size). It also allows to get rid of mmconf fallback code.

Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Sun, 10 Feb 2008 20:03:57 +0000 (12:03 -0800)]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
  [PKT_SCHED] ematch: Fix build warning.

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/tglx/linux-2.6-hrt
Linus Torvalds [Sun, 10 Feb 2008 20:02:45 +0000 (12:02 -0800)]
Merge git://git./linux/kernel/git/tglx/linux-2.6-hrt

* git://git.kernel.org/pub/scm/linux/kernel/git/tglx/linux-2.6-hrt:
  hrtimer: don't modify restart_block->fn in restart functions
  hrtimer: fix *rmtp/restarts handling in compat_sys_nanosleep()
  hrtimer: fix *rmtp handling in hrtimer_nanosleep()
  ntp: correct inconsistent interval/tick_length usage

16 years agosplice: fix user pointer access in get_iovec_page_array()
Bastian Blank [Sun, 10 Feb 2008 14:47:57 +0000 (16:47 +0200)]
splice: fix user pointer access in get_iovec_page_array()

Commit 8811930dc74a503415b35c4a79d14fb0b408a361 ("splice: missing user
pointer access verification") added the proper access_ok() calls to
copy_from_user_mmap_sem() which ensures we can copy the struct iovecs
from userspace to the kernel.

But we also must check whether we can access the actual memory region
pointed to by the struct iovec to fix the access checks properly.

Signed-off-by: Bastian Blank <waldi@debian.org>
Acked-by: Oliver Pinter <oliver.pntr@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years ago[PKT_SCHED] ematch: Fix build warning.
David S. Miller [Sun, 10 Feb 2008 11:48:15 +0000 (03:48 -0800)]
[PKT_SCHED] ematch: Fix build warning.

Commit 954415e33ed6cfa932c13e8c2460bd05e50723b5 ("[PKT_SCHED] ematch:
tcf_em_destroy robustness") removed a cast on em->data when
passing it to kfree(), but em->data is an integer type that can
hold pointers as well as other values so the cast is necessary.

Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agohrtimer: don't modify restart_block->fn in restart functions
Oleg Nesterov [Fri, 1 Feb 2008 17:41:30 +0000 (20:41 +0300)]
hrtimer: don't modify restart_block->fn in restart functions

hrtimer_nanosleep_restart() clears/restores restart_block->fn. This is
pointless and complicates its usage. Note that if sys_restart_syscall()
doesn't actually happen, we have a bogus "pending" restart->fn anyway,
this is harmless.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Alexey Dobriyan <adobriyan@sw.ru>
Cc: Pavel Emelyanov <xemul@sw.ru>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Toyo Abe <toyoa@mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
16 years agohrtimer: fix *rmtp/restarts handling in compat_sys_nanosleep()
Oleg Nesterov [Fri, 1 Feb 2008 17:35:31 +0000 (20:35 +0300)]
hrtimer: fix *rmtp/restarts handling in compat_sys_nanosleep()

Spotted by Pavel Emelyanov and Alexey Dobriyan.

compat_sys_nanosleep() implicitly uses hrtimer_nanosleep_restart(), this can't
work. Make a suitable compat_nanosleep_restart() helper.

Introduced by commit c70878b4e0b6cf8d2f1e46319e48e821ef4a8aba
hrtimer: hook compat_sys_nanosleep up to high res timer code

Also, set ->addr_limit = KERNEL_DS before doing hrtimer_nanosleep(), this func
was changed by the previous patch and now takes the "__user *" parameter.

Thanks to Ingo Molnar for fixing the bug in this patch.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Alexey Dobriyan <adobriyan@sw.ru>
Cc: Pavel Emelyanov <xemul@sw.ru>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Toyo Abe <toyoa@mvista.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
16 years agohrtimer: fix *rmtp handling in hrtimer_nanosleep()
Oleg Nesterov [Fri, 1 Feb 2008 14:29:05 +0000 (17:29 +0300)]
hrtimer: fix *rmtp handling in hrtimer_nanosleep()

Spotted by Pavel Emelyanov and Alexey Dobriyan.

hrtimer_nanosleep() sets restart_block->arg1 = rmtp, but this rmtp points to
the local variable which lives in the caller's stack frame. This means that
if sys_restart_syscall() actually happens and it is interrupted as well, we
don't update the user-space variable, but write into the already dead stack
frame.

Introduced by commit 04c227140fed77587432667a574b14736a06dd7f
hrtimer: Rework hrtimer_nanosleep to make sys_compat_nanosleep easier

Change the callers to pass "__user *rmtp" to hrtimer_nanosleep(), and change
hrtimer_nanosleep() to use copy_to_user() to actually update *rmtp.

Small problem remains. man 2 nanosleep states that *rtmp should be written if
nanosleep() was interrupted (it says nothing whether it is OK to update *rmtp
if nanosleep returns 0), but (with or without this patch) we can dirty *rem
even if nanosleep() returns 0.

NOTE: this patch doesn't change compat_sys_nanosleep(), because it has other
bugs. Fixed by the next patch.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Alexey Dobriyan <adobriyan@sw.ru>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: Pavel Emelyanov <xemul@sw.ru>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Toyo Abe <toyoa@mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
 include/linux/hrtimer.h |    2 -
 kernel/hrtimer.c        |   51 +++++++++++++++++++++++++-----------------------
 kernel/posix-timers.c   |   14 +------------
 3 files changed, 30 insertions(+), 37 deletions(-)

16 years agontp: correct inconsistent interval/tick_length usage
john stultz [Sun, 10 Feb 2008 09:48:03 +0000 (10:48 +0100)]
ntp: correct inconsistent interval/tick_length usage

clocksource initialization and error accumulation.  This corrects a 280ppm
drift seen on some systems using acpi_pm, and affects other clocksources as
well (likely to a lesser degree).

Signed-off-by: John Stultz <johnstul@us.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Sun, 10 Feb 2008 08:04:35 +0000 (00:04 -0800)]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (28 commits)
  [NET_SCHED] sch_htb: htb_requeue fix
  [IPV6]: Replace using the magic constant "1024" with IP6_RT_PRIO_USER for fc_metric.
  starfire: secton fix
  via-velocity: section fix
  natsemi: section fix
  typhoon: section fix
  isdn: fix section mismatch warning for ISACVer
  isdn: fix section mismatch warnings from hisax_cs_setup_card
  isdn: fix section mismatch warnings in isac.c and isar.c
  isdn: fix section mismatch warning in hfc_sx.c
  [PKT_SCHED] ematch: tcf_em_destroy robustness
  [PKT_SCHED]: deinline functions in meta match
  [SCTP]: Convert sctp_dbg_objcnt to seq files.
  [SCTP]: Use snmp_fold_field instead of a homebrew analogue.
  [IGMP]: Optimize kfree_skb in igmp_rcv.
  [KEY]: Convert net/pfkey to use seq files.
  [KEY]: Clean up proc files creation a bit.
  pppol2tp: fix printk warnings
  bnx2: section fix
  bnx2x: section fix
  ...

16 years ago[NET_SCHED] sch_htb: htb_requeue fix
Jarek Poplawski [Sun, 10 Feb 2008 07:44:00 +0000 (23:44 -0800)]
[NET_SCHED] sch_htb: htb_requeue fix

htb_requeue() enqueues skbs for which htb_classify() returns NULL.
This is wrong because such skbs could be handled by NET_CLS_ACT code,
and the decision could be different than earlier in htb_enqueue().
So htb_requeue() is changed to work and look more like htb_enqueue().

Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years ago[IPV6]: Replace using the magic constant "1024" with IP6_RT_PRIO_USER for fc_metric.
Rami Rosen [Sun, 10 Feb 2008 07:43:11 +0000 (23:43 -0800)]
[IPV6]: Replace using the magic constant "1024" with IP6_RT_PRIO_USER for fc_metric.

This patch replaces the explicit usage of the magic constant "1024"
with IP6_RT_PRIO_USER in the IPV6 tree.

Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agostarfire: secton fix
Andrew Morton [Sun, 10 Feb 2008 07:42:17 +0000 (23:42 -0800)]
starfire: secton fix

gcc-3.4.4 on powerpc:

drivers/net/starfire.c:219: error: version causes a section type conflict

Cc: Jeff Garzik <jeff@garzik.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
16 years agovia-velocity: section fix
Andrew Morton [Sun, 10 Feb 2008 07:41:40 +0000 (23:41 -0800)]
via-velocity: section fix

From: Andrew Morton <akpm@linux-foundation.org>

gcc-3.4.4 on powerpc:

drivers/net/via-velocity.c:443: error: chip_info_table causes a section type conflict

on this one I had to remove the __devinitdata too.  Don't know why.

Cc: Jeff Garzik <jeff@garzik.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>