Create struct nlm_pic_irq for interrupts handled by the PIC.
This simplifies IRQ handling for multi-SoC as well as
the single SoC cases. Also split the setup of percpu and PIC
interrupts so that we can configure the PIC interrupts for
every node.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4467Signed-off-by: John Crispin <blogic@openwrt.org>

There can be 1, 2 or 4 SoCs(nodes) in a multi-chip XLP board. Add an
option for multi-chip boards in case of XLP, and make the number of
nodes configurable.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4470Signed-off-by: John Crispin <blogic@openwrt.org>

Upto 4 Netlogic XLP SoCs can be connected over ICI links to form a
coherent multi-node system.  Each SoC has its own set of on-chip
devices including PIC.  To support this, add a per SoC stucture and
use it for the PIC and SYS block addresses instead of using global
variables.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4469Signed-off-by: John Crispin <blogic@openwrt.org>

Initial code to support more than 32 cpus. The platform CPU mask
is updated from 32-bit mask to cpumask_t. Convert places that use
cpu_/cpus_ functions to use cpumask_* functions.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4464Signed-off-by: John Crispin <blogic@openwrt.org>

Remove unused and trivial PIC accesss functions, update nlm_pic_send_ipi()
and nlm_set_irt_to_cpu() to use similar logic, and use correct type for
reg in nlm_pic_disable_irt().
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4463Signed-off-by: John Crispin <blogic@openwrt.org>

The cpuid was not passed into early_init_secondary even though the
comment indicated that it will be. Fix this.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4458Signed-off-by: John Crispin <blogic@openwrt.org>

Used the hardware thread id passed in while writing to IRT in
nlm_pic_init_irt()
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4465Signed-off-by: John Crispin <blogic@openwrt.org>

Fix Kconfig for both XLR and XLP to select ZONE_DMA32 (instead of ZONE_DMA)
in case of 64-bit compilation. This can be used for devices that can only
do DMA to 32-bit address. ZONE_DMA is not useful on XLR or XLP.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4466Signed-off-by: John Crispin <blogic@openwrt.org>

At this point early printk is available, so debugging device tree
issues is easier.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4460Signed-off-by: John Crispin <blogic@openwrt.org>

Enable Speculative Unmap Enable bit, which will enable speculative L2
cache requests for unmapped memory. This should give better performance
for kernel code/data which is in KSEG0
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4461Signed-off-by: John Crispin <blogic@openwrt.org>

This allows us to use the r2 optimized code from kernel headers
while compilation.

Disable PGD_C0_CONTEXT option for XLP, which does not work.
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4456Signed-off-by: John Crispin <blogic@openwrt.org>

Add support for XLP performance counters register in perf. Update
mips/Kconfig so that perf events can be selected for XLP.
Signed-off-by: Zi Shen Lim <zlim@netlogicmicro.com>
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4457Signed-off-by: John Crispin <blogic@openwrt.org>

Add support for XLR and XLS processors in MIPS Oprofile code. These
processors are multi-threaded and have two counters per core. Each
counter can track either all the events in the core (global mode),
or events in just one thread.

We use the counters in the global mode, and use only the first thread
in each core to handle the configuration etc.
Signed-off-by: Madhusudan Bhat <mbhat@netlogicmicro.com>
Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/4471Signed-off-by: John Crispin <blogic@openwrt.org>

Refactor nvram related functions into its own unit for easier expansion
and exposure of the values to other drivers.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Patchwork: http://patchwork.linux-mips.org/patch/4516Signed-off-by: John Crispin <blogic@openwrt.org>

Use the new reset helper where appropriate.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Patchwork: http://patchwork.linux-mips.org/patch/4453Signed-off-by: John Crispin <blogic@openwrt.org>

Add a reset helper for resetting the different cores.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Patchwork: http://patchwork.linux-mips.org/patch/4455Signed-off-by: John Crispin <blogic@openwrt.org>

The softreset register description for BCM6358 was missing, so add it.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Patchwork: http://patchwork.linux-mips.org/patch/4454Signed-off-by: John Crispin <blogic@openwrt.org>

Add a PCIe clock and use that instead of directly touching the clock
control register. While at it, fail if there is no such clock.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Acked-by: Florian Fainelli <florian@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/4452Signed-off-by: John Crispin <blogic@openwrt.org>

There are bcma based devices like the Linksys E2000 out there, which do
have one ieee80211 core, but no PCIe core and they are using no
prefixes for the sprom. In addition some values like boardtype are
stored without a prefix for the main SoC chip also when they have an
additional PCIe wifi chip with an own boardtype var on some devices.

The Ethernet addresses are now also read out correctly without a prefix
so calling bcm47xx_fill_sprom_ethernet is not needed any more.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Patchwork: http://patchwork.linux-mips.org/patch/4364Signed-off-by: John Crispin <blogic@openwrt.org>

If there is no ieee80211 core on the devices like on the BCM4706 read
out the sprom and the other data without using a prefix.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Patchwork: http://patchwork.linux-mips.org/patch/4361Signed-off-by: John Crispin <blogic@openwrt.org>

Read out the full board data independently of the sprom version. Now we
also get the full boardflags and so on if sromrev is not set and our
code would assume a rev 1 device. When a nvram option is not set
because it is not there this is no problem.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Patchwork: http://patchwork.linux-mips.org/patch/4363Signed-off-by: John Crispin <blogic@openwrt.org>

The memory size is detected by finding a place where it repeats in
memory. Currently we are just checking when the function prom_init is
seen again, but it is better to check for a bigger part of the memory
to decrease the chance of wrong results.

This should fix a problem we saw in OpenWrt, where the detected
available memory decreed on some devices when doing a soft reboot.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Patchwork: http://patchwork.linux-mips.org/patch/4362Signed-off-by: John Crispin <blogic@openwrt.org>

Ignoring the last page when ddr size is 128M. Cached accesses to last
page is causing the processor to prefetch using address above 128M
stepping out of the ddr address space.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Patchwork: http://patchwork.linux-mips.org/patch/4365Signed-off-by: John Crispin <blogic@openwrt.org>

Commit 97ce2c88 (jump-label: initialize
jump-label subsystem much earlier) caused MIPS to break, so this was
resolved with commit 6650df3c (MIPS:
Move cache setup to setup_arch().).  Unfortunately, after this commit,
the coherency kernel parameters, cca and coherentio, are no longer
processed before their values are used.

This patch fixes this problem by marking them as early_param, which
results in them being processed before they are needed.
Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com>
Acked-by: David Daney <david.daney@cavium.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Patchwork: http://patchwork.linux-mips.org/patch/3961Signed-off-by: John Crispin <blogic@openwrt.org>

The cavium code assumed that all NOR on the boot bus was
an 8-bit NOR part and hardcoded the bankwidth. The simple
solution was to add the code that queries the configuration
register for the width of the bus that has been hardware strapped
to the Cavium. This allows both 8-bit and 16-bit parts to be
discovered during boot.
Acked-by: David Daney <david.daney@cavium.com>
Signed-off-by: Charles Hardin <ckhardin@exablox.com>
Patchwork: http://patchwork.linux-mips.org/patch/4323Signed-off-by: John Crispin <blogic@openwrt.org>

The new functions introduced in commit 02a54177 (MIPS: tlbex: Deal with
re-definition of label) should be marked __cpuinit, to eliminate a
warning that can pop up when CONFIG_EXPORT_UASM is disabled:

      LD      arch/mips/mm/built-in.o
    WARNING: arch/mips/mm/built-in.o(.text+0x2a4c): Section mismatch in reference from the function uasm_bgezl_hazard() to the function .cpuinit.text:uasm_il_bgezl()
    The function uasm_bgezl_hazard() references
    the function __cpuinit uasm_il_bgezl().
    This is often because uasm_bgezl_hazard lacks a __cpuinit
    annotation or the annotation of uasm_il_bgezl is wrong.

    WARNING: arch/mips/mm/built-in.o(.text+0x2a68): Section mismatch in reference from the function uasm_bgezl_label() to the function .cpuinit.text:uasm_build_label()
    The function uasm_bgezl_label() references
    the function __cpuinit uasm_build_label().
    This is often because uasm_bgezl_label lacks a __cpuinit
    annotation or the annotation of uasm_build_label is wrong.

(This warning might not occur if the function was inlined.)
Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Patchwork: http://patchwork.linux-mips.org/patch/4517Signed-off-by: John Crispin <blogic@openwrt.org>

Pull gfs2 fixes from Steven Whitehouse:
 "Here are a number of GFS2 bug fixes.  There are three from Andy Price
  which fix various issues spotted by automated code analysis.  There
  are two from Lukas Czerner fixing my mistaken assumptions as to how
  FITRIM should work.  Finally Ben Marzinski has fixed a bug relating to
  mmap and atime and also a bug relating to a locking issue in the
  transaction code."

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-fixes:
  GFS2: Test bufdata with buffer locked and gfs2_log_lock held
  GFS2: Don't call file_accessed() with a shared glock
  GFS2: Fix FITRIM argument handling
  GFS2: Require user to provide argument for FITRIM
  GFS2: Clean up some unused assignments
  GFS2: Fix possible null pointer deref in gfs2_rs_alloc
  GFS2: Fix an unchecked error from gfs2_rs_alloc

Pull hwmon fixes from Jean Delvare.

* 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvare/staging:
  hwmon: Fix chip feature table headers
  hwmon: (w83627ehf) Force initial bank selection

In gfs2_trans_add_bh(), gfs2 was testing if a there was a bd attached to the
buffer without having the gfs2_log_lock held. It was then assuming it would
stay attached for the rest of the function. However, without either the log
lock being held of the buffer locked, __gfs2_ail_flush() could detach bd at any
time.  This patch moves the locking before the test.  If there isn't a bd
already attached, gfs2 can safely allocate one and attach it before locking.
There is no way that the newly allocated bd could be on the ail list,
and thus no way for __gfs2_ail_flush() to detach it.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

file_accessed() was being called by gfs2_mmap() with a shared glock. If it
needed to update the atime, it was crashing because it dirtied the inode in
gfs2_dirty_inode() without holding an exclusive lock. gfs2_dirty_inode()
checked if the caller was already holding a glock, but it didn't make sure that
the glock was in the exclusive state. Now, instead of calling file_accessed()
while holding the shared lock in gfs2_mmap(), file_accessed() is called after
grabbing and releasing the glock to update the inode.  If file_accessed() needs
to update the atime, it will grab an exclusive lock in gfs2_dirty_inode().

gfs2_dirty_inode() now also checks to make sure that if the calling process has
already locked the glock, it has an exclusive lock.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Currently implementation in gfs2 uses FITRIM arguments as it were in
file system blocks units which is wrong. The FITRIM arguments
(fstrim_range.start, fstrim_range.len and fstrim_range.minlen) are
actually in bytes.

Moreover, check for start argument beyond the end of file system, len
argument being smaller than file system block and minlen argument being
bigger than biggest resource group were missing.

This commit converts the code to convert FITRIM argument to file system
blocks and also adds appropriate checks mentioned above.

All the problems were recognised by xfstests 251 and 260.
Signed-off-by: Lukas Czerner <lczerner@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

When the fstrim_range argument is not provided by user in FITRIM ioctl
we should just return EFAULT and not promoting bad behaviour by filling
the structure in kernel. Let the user deal with it.
Signed-off-by: Lukas Czerner <lczerner@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Cleans up two cases where variables were assigned values but then never
used again.
Signed-off-by: Andrew Price <anprice@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Despite the return value from kmem_cache_zalloc() being checked, the
error wasn't being returned until after a possible null pointer
dereference. This patch returns the error immediately, allowing the
removal of the error variable.
Signed-off-by: Andrew Price <anprice@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Check the return value of gfs2_rs_alloc(ip) and avoid a possible null
pointer dereference.
Signed-off-by: Andrew Price <anprice@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Pull drm fixes from Dave Airlie:
 "A single radeon typo fix for a regressions and two fixes for a
  regression in the open helper address space stuff."

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
  drm/radeon: fix typo in evergreen_mc_resume()
  drm: set dev_mapping before calling drm_open_helper
  drm: restore open_count if drm_setup fails

Pull arm fixes from Russell King:
 "Not much here again.

  The two most notable things here are the sched_clock() fix, which was
  causing problems with the scheduling of threaded IRQs after a suspend
  event, and the vfp fix, which afaik has only been seen on some older
  OMAP boards.  Nevertheless, both are fairly important fixes."

* 'fixes' of git://git.linaro.org/people/rmk/linux-arm:
  ARM: 7569/1: mm: uninitialized warning corrections
  ARM: 7567/1: io: avoid GCC's offsettable addressing modes for halfword accesses
  ARM: 7566/1: vfp: fix save and restore when running on pre-VFPv3 and CONFIG_VFPv3 set
  ARM: 7565/1: sched: stop sched_clock() during suspend

Add missing index that may have led us to enabling
more crtcs than necessary.

May also fix:
https://bugs.freedesktop.org/show_bug.cgi?id=56139Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

Some drivers (specifically vmwgfx) look at dev_mapping
in their open hook, so we have to set dev->dev_mapping
earlier in the process.

Reference:
http://lists.freedesktop.org/archives/dri-devel/2012-October/029420.htmlSigned-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com>
Reported-by: Thomas Hellstrom <thellstrom@vmware.com>
Cc: stable@vger.kernel.org
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

If drm_setup (called at first open) fails, the whole
open call has failed, so we should not keep the
open_count incremented.
Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com>
Cc: stable@vger.kernel.org
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>