Add a read_persistent_clock function that does not just return 0.
Since timekeeping_init calls the function before time_init has been
called move reset_tod_clock to early.c to make sure that the TOD
clock is running when read_persistent_clock is invoked.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Reset the cpu timer to the maximum value and correctly initialize the
cpu accounting values in the lowcore when the cpu is started.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The steal time is calculated by subtracting the time the virtual cpu
has been running on a physical cpu from the wall clock time. To make
that work all wall time needs to be added to the steal time field first
before the virtual cpu time is subtracted.

The time between the last clock update and the load of the enabled wait
psw needs to be added to the steal_time field as well to make the sum
over all cpu accounting numbers match the wall clock.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The cpu_possible_map by default is initialized with all ones in s390.
If the kernel paramert possible_cpus=<x> is passed the cpu_possible_map
is supposed to have x bits set.
However the current code just sets the x bits without clearing the NR_CPUS
bits that were already set. So we end up with an unchanged map that has
all bits set.
To fix this just clear the map before setting any new bits.

This broke with def6cfb7
"[S390] cpumask: Use accessors code."

Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

mod_virt_timer() was used to modify/add cpu timers for cpus that were
set online. This resulted in a one-shot timer for every cpu that was
newly added or previously set offline, instead of an interval timer,
which broke the appldata vtime interval setup.

To fix this, the new mod_virt_timer_periodic() function is used, which
adds interval timers instead of one-shot timers.
Signed-off-by: NGerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

In case mod_virt_timer is used to add a non pending timer the timer
is always added as a one-shot timer. If mod_virt_timer is used for
periodic timers they may therfore be degraded to one-shot timers.

Add mod_virt_timer_periodic to the interface to allow safe re-programming
of the interval value.
Signed-off-by: NJan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Add a timer that retries the clock synchronization via the server time
protocol if there is a usable clock but the synchronization failed.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

nmi_enter/nmi_exit includes the lockdep calls and various
other calls which were missing so far.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

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

Currently the storage of the machine flags is a globally exported unsigned
long long variable. By moving the storage location into the lowcore struct we
allow assembler code to check machine_flags directly even without needing a
register. Addtionally the lowcore and therefore the machine flags too will be
in cache most of the time.
Signed-off-by: NChristian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Impact: performance regression fix for s390

The adaptive spinning mutexes will not always do what one would expect on
virtualized architectures like s390. Especially the cpu_relax() loop in
mutex_spin_on_owner might hurt if the mutex holding cpu has been scheduled
away by the hypervisor.

We would end up in a cpu_relax() loop when there is no chance that the
state of the mutex changes until the target cpu has been scheduled again by
the hypervisor.

For that reason we should change the default behaviour to no-spin on s390.

We do have an instruction which allows to yield the current cpu in favour of
a different target cpu. Also we have an instruction which allows us to figure
out if the target cpu is physically backed.

However we need to do some performance tests until we can come up with
a solution that will do the right thing on s390.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
LKML-Reference: <20090409184834.7a0df7b2@osiris.boeblingen.de.ibm.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Pass the original flags to rwlock arch-code, so that it can re-enable
interrupts if implemented for that architecture.

Initially, make __raw_read_lock_flags and __raw_write_lock_flags stubs
which just do the same thing as non-flags variants.
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NRobin Holt <holt@sgi.com>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: <linux-arch@vger.kernel.org>
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

First argument unused since 2.3.11.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This fixes a build failure with generic debug pagealloc:

  mm/debug-pagealloc.c: In function 'set_page_poison':
  mm/debug-pagealloc.c:8: error: 'struct page' has no member named 'debug_flags'
  mm/debug-pagealloc.c: In function 'clear_page_poison':
  mm/debug-pagealloc.c:13: error: 'struct page' has no member named 'debug_flags'
  mm/debug-pagealloc.c: In function 'page_poison':
  mm/debug-pagealloc.c:18: error: 'struct page' has no member named 'debug_flags'
  mm/debug-pagealloc.c: At top level:
  mm/debug-pagealloc.c:120: error: redefinition of 'kernel_map_pages'
  include/linux/mm.h:1278: error: previous definition of 'kernel_map_pages' was here
  mm/debug-pagealloc.c: In function 'kernel_map_pages':
  mm/debug-pagealloc.c:122: error: 'debug_pagealloc_enabled' undeclared (first use in this function)

by fixing

 - debug_flags should be in struct page
 - define DEBUG_PAGEALLOC config option for all architectures
Signed-off-by: NAkinobu Mita <akinobu.mita@gmail.com>
Reported-by: NAlexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

CONFIG_DEBUG_PAGEALLOC is now supported by x86, powerpc, sparc64, and
s390.  This patch implements it for the rest of the architectures by
filling the pages with poison byte patterns after free_pages() and
verifying the poison patterns before alloc_pages().

This generic one cannot detect invalid page accesses immediately but
invalid read access may cause invalid dereference by poisoned memory and
invalid write access can be detected after a long delay.
Signed-off-by: NAkinobu Mita <akinobu.mita@gmail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If a ccw device did not respond in time during internal io, we set it
into boxed state. With this patch we have the following behaviour:
 * the ccw driver will get a notification if the device was online and
   goes into the boxed state
 * if the device was disconnected and got boxed nothing special is to be
   done (it will be handled in reprobing later)
 * if the device got boxed while initial sensing it will be unregistered
Signed-off-by: NSebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Fix build breakage below which probably was introduced with
("rcu: don't include unnecessary headers, allow kmemtrace w/ tracepoints").

  CC      arch/s390/hypfs/hypfs_diag.o
arch/s390/hypfs/hypfs_diag.c: In function 'diag204_free_buffer':
arch/s390/hypfs/hypfs_diag.c:364: error: implicit declaration of function 'free_pages'
arch/s390/hypfs/hypfs_diag.c: In function 'diag204_alloc_rbuf':
arch/s390/hypfs/hypfs_diag.c:384: error: implicit declaration of function '__get_free_pages'
arch/s390/hypfs/hypfs_diag.c:384: error: 'GFP_KERNEL' undeclared (first use in this function)
arch/s390/hypfs/hypfs_diag.c:384: error: (Each undeclared identifier is reported only once
arch/s390/hypfs/hypfs_diag.c:384: error: for each function it appears in.)
Reported-by: NSachin Sant <sachinp@in.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Due to a different size of ino_t ustat needs a compat handler, but
currently only x86 and mips provide one.  Add a generic compat_sys_ustat
and switch all architectures over to it.  Instead of doing various
user copy hacks compat_sys_ustat just reimplements sys_ustat as
it's trivial.  This was suggested by Arnd Bergmann.

Found by Eric Sandeen when running xfstests/017 on ppc64, which causes
stack smashing warnings on RHEL/Fedora due to the too large amount of
data writen by the syscall.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Remove duplicated #include's in arch/s390/kernel/ipl.c.
Signed-off-by: NHuang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Makes code futureproof against the impending change to mm->cpu_vm_mask.

It's also a chance to use the new cpumask_ ops which take a pointer
(the older ones are deprecated, but there's no hurry for arch code).
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Impact: use new API

Use the accessors rather than frobbing bits directly.  Most of this is
in arch code I haven't even compiled, but is straightforward.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NMike Travis <travis@sgi.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Impact: cleanup, futureproof

In fact, all cpumask ops will only be valid (in general) for bit
numbers < nr_cpu_ids.  So use that instead of NR_CPUS in various
places (I also updated the immediate sites to use the new cpumask_
operators).

This is always safe: no cpu number can be >= nr_cpu_ids, and
nr_cpu_ids is initialized to NR_CPUS at boot.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NMike Travis <travis@sgi.com>
Acked-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Impact: cleanup

cpu_coregroup_mask is the New Hotness.

As S/390 uses theirs internally, so we just make it static.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The lowcore clock comparator save area on 64 bit machines is defined to
contain only the seven most significant bits of the register.
That's also why it starts at an uneven address (0x1331).
The current code however writes eight bytes to the address and
therefore overwrites the first byte of the access register save area.
Fix this and write only seven bytes to the save area.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The Extended Translation Facility 3 (ETF3) added instructions which
allow conversions between different unicode character maps (UTF-8,
UTF-32 ...).  These instructions got enhanced with a later version of
the ETF3 allowing malformed multibyte chars to be recognized and
reported correctly.  The attached patch reserves bit 8 in the elf
hwcaps vector for the enhanced version of ETF3. The bit corresponds to
the stfle bits 22 and 30 and will only be set if both of the stfle
bits are set.
Signed-off-by: NAndreas Krebbel <krebbel@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The s390 ipl panic notifier will stop the system or trigger a system dump.
This should be done as final action on the panic path. All other panic
notifiers should be executed before. Currently we use priority 0 for the ipl
notifier. In order to be called late, this patch changes the priority to
INT_MIN which is the lowest possible priority.
Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The old dfp detection wanted to check bit 43 (dfp high performance), but due
to a wrong calculation always used to check bit 42. Additionally the
"userspace expectation" is, that the dfp capability bit is set is if facility
bit 42 (decimal floating point facility available) and bit 44 (perform floating
point operation facility avail).
The patch fixes the bit calculation and extends the check to work like:
  elf hw cap dfp bit = facility bits 42 (dfp) & 44 (pfpo) available
Signed-off-by: NChristian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Performing an initial cpu reset makes sure all registers and tlbs of
the targeted cpu are initialized and flushed.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

If sigp_set_prefix fails on __cpu_up we leak the lowcore structures
and async+panic stacks for the failed cpu.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

A code analysis tool reported two warnings:
"The expression `ipl_info.type == IPL_TYPE_FCP' is true whenever evaluated."
and "Default is not possible". This patch improves the corresponding if
statement logic and removes the unnecessary switch defaults.
Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The check for NULL pointer has already be done before. Therefore we can remove
the second check.
Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Use kzfree() instead of memset() + kfree().
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NPekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Signed-off-by: NStoyan Gaydarov <stoyboyker@gmail.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The zcore code switches to real addressing mode when creating a kernel dump.
This is not possible, if it is built as a kernel module. With this patch
zcore (zfcpdump) can't be built as a kernel module any more.
Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The cksm function in system.h is duplicate to csum_partial in checksum.h.
Remove cksm and use csum_partial instead.
Signed-off-by: NFrank Munzert <munzert@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The lowcore.h header has quite a lot of whitespace damage and a rather
wild collection of entries. Remove all that whitespace and tidy up the
order of the lowcore fields.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

This patch fixes two addresses in the comments for the
lowcore structure. Looks like an copy-paste bug.
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>