Signed-off-by: NDominik Dingel <dingel@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Signed-off-by: NDominik Dingel <dingel@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Signed-off-by: NDominik Dingel <dingel@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Limit the number of bits to the maximum number of cpus a machine
can have.
possible_cpu_mask typically will have more bits set than a machine
may physically have. This results in wasted memory during per-cpu
memory allocations, if the possible mask contains more cpus than
physically possible for a given configuration.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The "weight_min" attribute got the wrong name. The value represents
the number of non-stopped (operating) CPUS. Therefore add a note and
rename the struct member to "ocpus".
Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.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>

The PTRACE_SINGLEBLOCK option is used to get control whenever
the inferior has executed a successful branch. The PER option to
implement block stepping is successful-branching event, bit 32
in the PER-event mask.
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>

The call to update_cpu_masks() from within topology_init()
is completely redundant. This patch removes it.
Signed-off-by: NPhilipp Hachtmann <phacht@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The whole compat exec domain code doesn't make any difference.

From the registered s390_exec_domain:

- exec domain name is only displayed in /proc/execdomains
- handler is unused
- pers_low and pers_high are only used internally to find this specific
  exec domain otherwise the default exec domain will be used
- all other fields match the default exec domain

So let's get rid of this.
Reported-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The unconditionally built keyboard driver, drivers/s390/char/keyboard.c,
requires CONFIG_TTY, so select it from CONFIG_S390 to prevent a build
error.
Signed-off-by: NJosh Triplett <josh@joshtriplett.org>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The cpu array size calculation uses the NR_CPUS config option, which
was recently increased from 64 to 256. With a value of 256, the cpu
array will no longer fit into one APPLDATA record and loading the
appldata_os module fails with the following error:
could not insert 'appldata_os': Cannot allocate memory

Use num_possible_cpus() instead of NR_CPUS. For z/VM, this will still
result in a value of 64. This is not true for LPAR, but the appldata
feature is not available for LPAR.
Signed-off-by: NGerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The memset() within csum_partial_copy_from_user() is rather pointless since
copy_from_user() already cleared the rest of the destination buffer if an
exception happened.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

There is no user left, so remove it.
It was also potentially broken, since the function didn't clear destination
memory if copy_from_user() failed. Which would allow for information leaks.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Currently we return a negative block count if SCLP returns HSA block
count zero. The reason is that we subtract one for the header page.

So fix this and correctly return zero block count if SCLP returns zero.
Signed-off-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Add entries for s390/cio, s390/dasd and s390/pci.
Signed-off-by: NSebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The maximum usable buffer size of the s390 debug feature (when using
the sprintf_view) is 11 * sizeof(long) (1 pointer for the format
string + 10 arguments). When a larger buffer size is specified the
additional memory is unused and wasted per debug entry. So reducing
the buffer size to its maximum (or to the actual buffer size used)
will make more precious debug feature space usable.

For pci_msg, chsc_msg, and cio_crw we use the additional usable dbf
space to reduce the number of allocated pages.
Signed-off-by: NSebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Add airq_iv_alloc and airq_iv_free to allocate and free consecutive
ranges of irqs from the interrupt vector.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Add the pgtable_pmd_page_ctor/pgtable_pmd_page_dtor calls to the pmd
allocation and free functions and enable ARCH_ENABLE_SPLIT_PMD_PTLOCK
for 64 bit.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The pmd pointer passed to pmd_lockptr/pmd_huge_pte can point to any
entry in a pmd table. With USE_SPLIT_PMD_PTLOCKS==1 the code uses
virt_to_page to get a struct page for the pmd table. The virt_to_page
function automatically masks the lower PAGE_SHIFT bits from the
address. But if the size of a pmd table is larger than PAGE_SIZE the
additional bits are not removed from the pmd address and the wrong
page struct is used.

Fix this by explicitely masking the offset in the pmd table from
the pmd pointer.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Fix some numbers in the comments describing the layout of the bit maps.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The guest page state needs to be reset to stable for all pages
on initial program load via diagnose 0x308.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

This patch enables Collaborative Memory Management (CMM) for kvm
on s390. CMM allows the guest to inform the host about page usage
(see arch/s390/mm/cmm.c). The host uses this information to avoid
swapping in unused pages in the page fault handler. Further, a CPU
provided list of unused invalid pages is processed to reclaim swap
space of not yet accessed unused pages.

[ Martin Schwidefsky: patch reordering and cleanup ]
Signed-off-by: NKonstantin Weitz <konstantin.weitz@gmail.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

In a virtualized environment and given an appropriate interface the guest
can mark pages as unused while they are free (for the s390 implementation
see git commit 45e576b1 "guest page hinting light"). For the host
the unused state is a property of the pte.

This patch adds the primitive 'pte_unused' and code to the host swap out
handler so that pages marked as unused by all mappers are not swapped out
but discarded instead, thus saving one IO for swap out and potentially
another one for swap in.

[ Martin Schwidefsky: patch reordering and simplification ]
Signed-off-by: NKonstantin Weitz <konstantin.weitz@gmail.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Git commit 050eef36 "[S390] fix tlb flushing vs. concurrent
/proc accesses" introduced the attach counter to avoid using the
mm_users value to decide between IPTE for every PTE and lazy TLB
flushing with IDTE. That fixed the problem with mm_users but it
introduced another subtle race, fortunately one that is very hard
to hit.
The background is the requirement of the architecture that a valid
PTE may not be changed while it can be used concurrently by another
cpu. The decision between IPTE and lazy TLB flushing needs to be
done while the PTE is still valid. Now if the virtual cpu is
temporarily stopped after the decision to use lazy TLB flushing but
before the invalid bit of the PTE has been set, another cpu can attach
the mm, find that flush_mm is set, do the IDTE, return to userspace,
and recreate a TLB that uses the PTE in question. When the first,
stopped cpu continues it will change the PTE while it is attached on
another cpu. The first cpu will do another IDTE shortly after the
modification of the PTE which makes the race window quite short.

To fix this race the CPU that wants to attach the address space of a
user space thread needs to wait for the end of the PTE modification.
The number of concurrent TLB flushers for an mm is tracked in the
upper 16 bits of the attach_count and finish_arch_post_lock_switch
is used to wait for the end of the flush operation if required.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The finish_arch_post_lock_switch is called at the end of the task
switch after all locks have been released. In concept it is paired
with the switch_mm function, but the current code only does the
call in finish_task_switch. Add the call to idle_task_exit and
use_mm. One use case for the additional calls is s390 which will
use finish_arch_post_lock_switch to wait for the completion of
TLB flush operations.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The uaccesspt kernel parameter allows to enforce using the uaccess page
table walk variant. This is mainly for debugging purposes, so this mode
can also be enabled on machines which support the mvcos instruction.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Remove another leftover from the time when we supported running
user space in either home or primary address space.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

MACHINE_HAS_MVCOS is used exactly once when the machine is brought up.
There is no need to cache the flag in the machine_flags.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The types 'size_t' and 'unsigned long' have been used randomly for the
uaccess functions. This looks rather confusing.
So let's change all functions to use unsigned long instead and get rid
of size_t in order to have a consistent interface.

The only exception is strncpy_from_user() which uses 'long' since it
may return a signed value (-EFAULT).
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

There are only two uaccess variants on s390 left: the version that is used
if the mvcos instruction is available, and the page table walk variant.
So there is no need for expensive indirect function calls.

By default the mvcos variant will be called. If the mvcos instruction is not
available it will call the page table walk variant.

For minimal performance impact the "if (mvcos_is_available)" is implemented
with a jump label, which will be a six byte nop on machines with mvcos.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

For some unknown reason the indirect uaccess functions on s390 implement a
different parameter order than what is usual.

e.g.:

unsigned long copy_to_user(void *to, const void *from, unsigned long n);
vs.
size_t (*copy_to_user)(size_t n, void __user * to, const void *from);

Let's get rid of this confusing parameter reordering.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Interrupts which happen on ccw consoles prior to their registration
with the driver core are not accounted to the respective device
driver. Fix this by setting the proper interrupt class during
initialization of ccw consoles.
Reviewed-by: NPeter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: NSebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Drivers for ccw consoles use ccw_device_probe_console to receive
an initialized ccw device which is already enabled for interrupts.
After that the device driver does the initialization of its private
data. This can race with unsolicited interrupts which can happen
once the device is enabled for interrupts.

Split ccw_device_probe_console into ccw_device_create_console and
ccw_device_enable_console and reorder the initialization of the ccw
console drivers.

While at it mark these functions as __init.
Reviewed-by: NPeter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: NSebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

ccw consoles are in use before they can be properly registered with
the driver core. For devices which are in use by a device driver we
rely on the ccw_device's pointer to the driver callbacks to be valid.
For ccw consoles this pointer is NULL until they are registered later
during boot and we dereferenced this pointer. This worked by
chance on 64 bit builds (cdev->drv was NULL but the optional callback
cdev->drv->path_event was also NULL by coincidence) and was unnoticed
until we received reports about boot failures on 31 bit systems.
Fix it by initializing the driver pointer for ccw consoles.

Cc: <stable@vger.kernel.org> # 3.10+
Reported-by: NMike Frysinger <vapier@gentoo.org>
Reported-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Reviewed-by: NPeter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: NSebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Pull drm fixes from Dave Airlie:
 "Lots of little small things, nothing too major: nouveau regression
  fixes, vmware fixes for the new hw support, memory leaks in error path
  fixes"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (31 commits)
  drm/radeon/ni: fix typo in dpm sq ramping setup
  drm/radeon/si: fix typo in dpm sq ramping setup
  drm/radeon: fix CP semaphores on CIK
  drm/radeon: delete a stray tab
  drm/radeon: fix display tiling setup on SI
  drm/radeon/dpm: reduce r7xx vblank mclk threshold to 200
  drm/radeon: fill in DRM_CAPs for cursor size
  drm: add DRM_CAPs for cursor size
  drm/radeon: unify bpc handling
  drm/ttm: Fix memory leak in ttm_agp_backend.c
  drm/ttm: declare 'struct device' in ttm_page_alloc.h
  drm/nouveau: fix TTM_PL_TT memtype on pre-nv50
  drm/nv50/disp: use correct register to determine DP display bpp
  drm/nouveau/fb: use correct ram oclass for nv1a hardware
  drm/nv50/gr: add missing nv_error parameter priv
  drm/nouveau: fix ENG_RUNLIST register address
  drm/nv4c/bios: disallow retrieving from prom on nv4x igp's
  drm/nv4c/vga: decode register is in a different place on nv4x igp's
  drm/nv4c/mc: nv4x igp's have a different msi rearm register
  drm/nouveau: set irq_enabled manually
  ...

Pull HID update from Jiri Kosina:

 - fixes for several bugs in incorrect allocations of buffers by David
   Herrmann and Benjamin Tissoires.

 - support for a few new device IDs by Archana Patni, Benjamin
   Tissoires, Huei-Horng Yo, Reyad Attiyat and Yufeng Shen

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:
  HID: hyperv: make sure input buffer is big enough
  HID: Bluetooth: hidp: make sure input buffers are big enough
  HID: hid-sensor-hub: quirk for STM Sensor hub
  HID: apple: add Apple wireless keyboard 2011 JIS model support
  HID: fix buffer allocations
  HID: multitouch: add FocalTech FTxxxx support
  HID: microsoft: Add ID's for Surface Type/Touch Cover 2
  HID: usbhid: quirk for CY-TM75 75 inch Touch Overlay

Pull networking fixes from David Miller:

 1) kvaser CAN driver has fixed limits of some of it's table, validate
    that we won't exceed those limits at probe time.  Fix from Olivier
    Sobrie.

 2) Fix rtl8192ce disabling interrupts for too long, from Olivier
    Langlois.

 3) Fix botched shift in ath5k driver, from Dan Carpenter.

 4) Fix corruption of deferred packets in TIPC, from Erik Hugne.

 5) Fix newlink error path in macvlan driver, from Cong Wang.

 6) Fix netpoll deadlock in bonding, from Ding Tianhong.

 7) Handle GSO packets properly in forwarding path when fragmentation is
    necessary on egress, from Florian Westphal.

 8) Fix axienet build errors, from Michal Simek.

 9) Fix refcounting of ubufs on tx in vhost net driver, from Michael S
    Tsirkin.

10) Carrier status isn't set properly in hyperv driver, from Haiyang
    Zhang.

11) Missing pci_disable_device() in tulip_remove_one), from Ingo Molnar.

12) AF_PACKET qdisc bypass mode doesn't adhere to driver provided TX
    queue selection method.  Add a fallback method mechanism to fix this
    bug, from Daniel Borkmann.

13) Fix regression in link local route handling on GRE tunnels, from
    Nicolas Dichtel.

14) Bonding can assign dup aggregator IDs in some sequences of
    configuration, fix by making the allocation counter per-bond instead
    of global.  From Jiri Bohac.

15) sctp_connectx() needs compat translations, from Daniel Borkmann.

16) Fix of_mdio PHY interrupt parsing, from Ben Dooks

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (62 commits)
  MAINTAINERS: add entry for the PHY library
  of_mdio: fix phy interrupt passing
  net: ethernet: update dependency and help text of mvneta
  NET: fec: only enable napi if we are successful
  af_packet: remove a stray tab in packet_set_ring()
  net: sctp: fix sctp_connectx abi for ia32 emulation/compat mode
  ipv4: fix counter in_slow_tot
  irtty-sir.c: Do not set_termios() on irtty_close()
  bonding: 802.3ad: make aggregator_identifier bond-private
  usbnet: remove generic hard_header_len check
  gre: add link local route when local addr is any
  batman-adv: fix potential kernel paging error for unicast transmissions
  batman-adv: avoid double free when orig_node initialization fails
  batman-adv: free skb on TVLV parsing success
  batman-adv: fix TT CRC computation by ensuring byte order
  batman-adv: fix potential orig_node reference leak
  batman-adv: avoid potential race condition when adding a new neighbour
  batman-adv: properly check pskb_may_pull return value
  batman-adv: release vlan object after checking the CRC
  batman-adv: fix TT-TVLV parsing on OGM reception
  ...

Pull ARM fixes from Russell King:
 "A range of ARM fixes.  Biggest change is the stage-2 attributes used
  for for hyp mode which were wrong.  I've killed some bits in a couple
  of DT files which turned out not to be required, and a few other
  fixes.

  One fix touches code outside of arch/arm, which is related to sorting
  out the DMA masks correctly.  There is a long standing issue with the
  conversion from PFNs to addresses where people assume that shifting an
  unsigned long left by PAGE_SHIFT results in a correct address.  This
  is not the case with C: the integer promotion happens at assignment
  after evaluation.  This fixes the recently introduced dma_max_pfn()
  function, but there's a number of other places where we try this
  directly on an unsigned long in the mm code"

* 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm:
  ARM: 7957/1: add DSB after icache flush in __flush_icache_all()
  Fix uses of dma_max_pfn() when converting to a limiting address
  ARM: 7955/1: spinlock: ensure we have a compiler barrier before sev
  ARM: 7953/1: mm: ensure TLB invalidation is complete before enabling MMU
  ARM: 7952/1: mm: Fix the memblock allocation for LPAE machines
  ARM: 7950/1: mm: Fix stage-2 device memory attributes
  ARM: dts: fix spdif pinmux configuration

Pull jfs fix from David Kleikamp:
 "Another ACL regression. This one more subtle"

* tag 'jfs-3.14-rc4' of git://github.com/kleikamp/linux-shaggy:
  jfs: set i_ctime when setting ACL