This should be a plain old '&' and could easily lead to undefined
behaviour if the target of a pmd_mknotpresent invocation was the same
as the parameter.

Fixes: 9c7e535f (arm64: mm: Route pmd thp functions through pte equivalents)
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Cc: <stable@vger.kernel.org> # v3.15
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

I'm seeing this build failure for arm64:

    CC [M]  Documentation/filesystems/configfs/configfs_example_macros.o
  In file included from /usr/include/bits/sigcontext.h:27:0,
                   from /usr/include/signal.h:340,
                   from /usr/include/sys/wait.h:30,
                   from Documentation/accounting/getdelays.c:24:
  .../linux/usr/include/asm/sigcontext.h:61:2: error: unknown type name ‘u64’
    u64 esr;
    ^
  make[2]: *** [Documentation/accounting/getdelays] Error 1

This was introduced by commit 15af1942:

  arm64: Expose ESR_EL1 information to user when SIGSEGV/SIGBUS

Using __u64 instead of u64 fixes the problem.
Signed-off-by: NMark Salter <msalter@redhat.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

APM X-Gene Storm SoC supports 4 serial ports. This patch adds device nodes
for serial ports 1 to 3 (a device node for serial port 0 is already present
in the dts file).
This patch also sets the compatible property of serial nodes to "ns16550a".
Signed-off-by: NVinayak Kale <vkale@apm.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Arm64 does not define dma_get_required_mask() function.
Therefore, it should not define the ARCH_HAS_DMA_GET_REQUIRED_MASK.
This causes build errors in some device drivers (e.g. mpt2sas)
Signed-off-by: NSuravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: <stable@vger.kernel.org>

Currently core file of aarch32 process prstatus note has empty
registers set. As result aarch32 core files create by V8 kernel are
not very useful.

It happens because compat_gpr_get and compat_gpr_set functions can
copy registers values to/from either kbuf or ubuf. ELF core file
collection function fill_thread_core_info calls compat_gpr_get
with kbuf set and ubuf set to 0. But current compat_gpr_get and
compat_gpr_set function handle copy to/from only ubuf case.

Fix is to handle kbuf and ubuf as two separate cases in similar
way as other functions like user_regset_copyout, user_regset_copyin do.
Signed-off-by: NVictor Kamensky <victor.kamensky@linaro.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Cc: stable@vger.kernel.org
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Whilst native arm64 applications don't have the 16-bit UID/GID syscalls
wired up, compat tasks can still access them. The 16-bit wrappers for
these syscalls use __kernel_old_uid_t and __kernel_old_gid_t, which must
be 16-bit data types to maintain compatibility with the 16-bit UIDs used
by compat applications.

This patch defines 16-bit __kernel_old_{gid,uid}_t types for arm64
instead of using the 32-bit types provided by asm-generic.
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Our compat PTRACE_POKEUSR implementation simply passes the user data to
regset_copy_from_user after some simple range checking. Unfortunately,
the data in question has already been copied to the kernel stack by this
point, so the subsequent access_ok check fails and the ptrace request
returns -EFAULT. This causes problems tracing fork() with older versions
of strace.

This patch briefly changes the fs to KERNEL_DS, so that the access_ok
check passes even with a kernel address.
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

When the CMA buffer is allocated, it is too early to know whether
devices will require ZONE_DMA memory. This patch limits the CMA buffer
to (DMA_BIT_MASK(32) + 1) if CONFIG_ZONE_DMA is enabled.

In addition, it computes the dma_to_phys(DMA_BIT_MASK(32)) before the
increment (no current functional change).
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

The clean up of CALLER_ADDR*() functions required the archs to either
use the default __builtin_return_address(X) (where X > 0) or override
it with something the arch can use. To override it, the arch would
define ftrace_return_address(x).

The arm architecture requires this to be redefined but instead of
defining ftrace_return_address(x) it defined ftrace_return_addr(x).

Fixes: eed542d6 (ftrace: Make CALLER_ADDRx macros more generic)
Reported-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Tested-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

This patches modifies the GHASH secure hash implementation to switch to a
faster, polynomial multiplication based reduction instead of one that uses
shifts and rotates.
Signed-off-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

This fixes a bug in the GHASH algorithm resulting in the calculated hash to be
incorrect if the input is presented in chunks whose size is not a multiple of
16 bytes.
Signed-off-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
Fixes: fdd23894 ("arm64/crypto: GHASH secure hash using ARMv8 Crypto Extensions")
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

This patch adds several defconfig options required primarily by the LTP
test suite.
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

The Operating Performance Point (OPP) Layer library is a generic
library used by CPUFREQ and DEVFREQ. It can be enabled only on the
platforms that specify ARCH_HAS_OPP option.

This patch selects that option in order to allow ARM64 based platforms
to use OPP library.
Signed-off-by: NSudeep Holla <sudeep.holla@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

Signed-off-by: NSudeep Holla <sudeep.holla@arm.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

The value of ESR has been stored into x1, and should be directly pass to
do_sp_pc_abort function, "MOV x1, x25" is an extra operation and do_sp_pc_abort
will get the wrong value of ESR.
Signed-off-by: NChiaHao <andy.jhshiu@gmail.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>
Cc: <stable@vger.kernel.org>

Remove xen_enable_nmi() to fix a 64-bit guest crash when registering
the NMI callback on Xen 3.1 and earlier.

It's not needed since the NMI callback is set by a set_trap_table
hypercall (in xen_load_idt() or xen_write_idt_entry()).

It's also broken since it only set the current VCPU's callback.
Signed-off-by: NDavid Vrabel <david.vrabel@citrix.com>
Reported-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Tested-by: NVitaly Kuznetsov <vkuznets@redhat.com>

The System Type menu is getting quite long with platforms and is
inconsistent in handling of sub-arch specific options. Tidy up the menu
by making platform options a menuconfig entry containing any platform
specific config items.

[arnd: change OMAP part according to suggestion from
 Tony Lindgren <tony@atomide.com>]
Signed-off-by: NRob Herring <robh@kernel.org>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

Following 5d01b768 "mmc: simplify SDHCI Kconfig dependencies",
SDHCI drivers that use MMC_SDHCI_PLTFM no longer select it, but
instead depend on it. This means that multi_v7_defconfig no longer
selects it, and hence many SDHCI drivers are no longer enabled.
Explicitly enable MMC_SDHCI_PLTFM to solve this.

Fixes: 5d01b768 ("mmc: simplify SDHCI Kconfig dependencies")
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Tested-by: NMatt Porter <mporter@linaro.org>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

of_get_flat_dt_prop return type is now const.
Fixes the following compilation warning introduced by commit 9d0c4dfe
("of/fdt: update of_get_flat_dt_prop in prep for libfdt")

arch/arm/mach-exynos/exynos.c:259:6: warning:
assignment discards ‘const’ qualifier from pointer target type [enabled by default]
Signed-off-by: NSachin Kamat <sachin.kamat@linaro.org>
Reviewed-by: NTushar Behera <tushar.behera@linaro.org>
Cc: Rob Herring <robh@kernel.org>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

This solves a problem with building with CONFIG_SMP=n due to missing
sysram_base_addr (or sysram_ns_base_addr) variables.

The new setup method is more awkward than I'd like for it to be, but
it can't be done in init_early() since ioremap is not yet available,
but it needs to happen before SMP.
Reported-by: NRussell King <linux@arm.linux.org.uk>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NOlof Johansson <olof@lixom.net>
Reviewed-by: NTomasz Figa <t.figa@samsung.com>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

The factory bootloader on A385-DB boards expect the ECC strength to be
4 bits over 512 bytes. Hence, we need to specify this in the devicetree,
to prevent the kernel from assuming any different ECC scheme.
Signed-off-by: NEzequiel Garcia <ezequiel.garcia@free-electrons.com>
Link: https://lkml.kernel.org/r/1400941030-2123-3-git-send-email-ezequiel.garcia@free-electrons.comSigned-off-by: NJason Cooper <jason@lakedaemon.net>

The factory bootloader on A375-DB boards expect the ECC strength to be
4 bits over 512 bytes. Hence, we need to specify this in the devicetree,
to prevent the kernel from assuming any different ECC scheme.
Signed-off-by: NEzequiel Garcia <ezequiel.garcia@free-electrons.com>
Link: https://lkml.kernel.org/r/1400941030-2123-2-git-send-email-ezequiel.garcia@free-electrons.comSigned-off-by: NJason Cooper <jason@lakedaemon.net>

The addition of Exynos to multi-platform configs creates a mess of config
options with options appearing before the Exynos config option. This is
due to arch/arm/plat-samsung/Kconfig being included out of order with the
other Samsung platform kconfig files. Reorder the kconfig files and move
all the options into a sub-menu. Some of the options are dead, so remove
those as well.
Signed-off-by: NRob Herring <robh@kernel.org>
Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Cc: linux-samsung-soc@vger.kernel.org
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

This config exists entirely to hide the cpufreq menu from the
kernel configuration unless a platform has selected it. Nothing
is actually built if this config is 'Y' and it just leads to more
patches that add a select under a platform Kconfig so that some
other CPUfreq option can be chosen. Let's remove the option so
that we can always enable CPUfreq drivers on ARM platforms.
Acked-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

This addresses a section mismatch problem in the IM-PD1
driver in the Integrator/AP.

The IM-PD1 contains a VIC interrupt controller and therefore
the driver calls vic_init_cascaded() which is marked __init as
irqchips are simply not hot-pluggable and specifically the VIC
is assumed to initiate only on boot.

However the module driver model of the Integrator LM bus
assumes that logic tile drivers can be probed at runtime. This
is not really the case for IM-PD1: these tiles are detected at
boot and they cannot be plugged into a running system. Before
this patch it is of course possible to modprobe them later.

By first forcing the IM-PD1 to bool we make sure this driver
gets compiled into the kernel, and we know it will be probed
only at boot time when the tiles are detected, so we can tag
its probe function __init_refok as we know it won't be called
after boot now, and the section mismatch problem goes away.

As a side effect, sysfs binding from userspace becomes
impossible, so we tag the driver to suppress the bind/unbind
sysfs attributes.

Cc: Russell King <linux@arm.linux.org.uk>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

The OpenBlocks AX3-4 has a non-DT bootloader.  It also comes with 1GB of
soldered on RAM, and a DIMM slot for expansion.

Unfortunately, atags_to_fdt() doesn't work in big-endian mode, so we see
the following failure when attempting to boot a big-endian kernel:

  686 slab pages
  17 pages shared
  0 pages swap cached
  [ pid ]   uid  tgid total_vm      rss nr_ptes swapents oom_score_adj name
  Kernel panic - not syncing: Out of memory and no killable processes...

  CPU: 1 PID: 351 Comm: kworker/u4:0 Not tainted 3.15.0-rc8-next-20140603 #1
  [<c0215a54>] (unwind_backtrace) from [<c021160c>] (show_stack+0x10/0x14)
  [<c021160c>] (show_stack) from [<c0802500>] (dump_stack+0x78/0x94)
  [<c0802500>] (dump_stack) from [<c0800068>] (panic+0x90/0x21c)
  [<c0800068>] (panic) from [<c02b5704>] (out_of_memory+0x320/0x340)
  [<c02b5704>] (out_of_memory) from [<c02b93a0>] (__alloc_pages_nodemask+0x874/0x930)
  [<c02b93a0>] (__alloc_pages_nodemask) from [<c02d446c>] (handle_mm_fault+0x744/0x96c)
  [<c02d446c>] (handle_mm_fault) from [<c02cf250>] (__get_user_pages+0xd0/0x4c0)
  [<c02cf250>] (__get_user_pages) from [<c02f3598>] (get_arg_page+0x54/0xbc)
  [<c02f3598>] (get_arg_page) from [<c02f3878>] (copy_strings+0x278/0x29c)
  [<c02f3878>] (copy_strings) from [<c02f38bc>] (copy_strings_kernel+0x20/0x28)
  [<c02f38bc>] (copy_strings_kernel) from [<c02f4f1c>] (do_execve+0x3a8/0x4c8)
  [<c02f4f1c>] (do_execve) from [<c025ac10>] (____call_usermodehelper+0x15c/0x194)
  [<c025ac10>] (____call_usermodehelper) from [<c020e9b8>] (ret_from_fork+0x14/0x3c)
  CPU0: stopping
  CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.15.0-rc8-next-20140603 #1
  [<c0215a54>] (unwind_backtrace) from [<c021160c>] (show_stack+0x10/0x14)
  [<c021160c>] (show_stack) from [<c0802500>] (dump_stack+0x78/0x94)
  [<c0802500>] (dump_stack) from [<c021429c>] (handle_IPI+0x138/0x174)
  [<c021429c>] (handle_IPI) from [<c02087f0>] (armada_370_xp_handle_irq+0xb0/0xcc)
  [<c02087f0>] (armada_370_xp_handle_irq) from [<c0212100>] (__irq_svc+0x40/0x50)
  Exception stack(0xc0b6bf68 to 0xc0b6bfb0)
  bf60:                   e9fad598 00000000 00f509a3 00000000 c0b6a000 c0b724c4
  bf80: c0b72458 c0b6a000 00000000 00000000 c0b66da0 c0b6a000 00000000 c0b6bfb0
  bfa0: c027bb94 c027bb24 60000313 ffffffff
  [<c0212100>] (__irq_svc) from [<c027bb24>] (cpu_startup_entry+0x54/0x214)
  [<c027bb24>] (cpu_startup_entry) from [<c0ac5b30>] (start_kernel+0x318/0x37c)
  [<c0ac5b30>] (start_kernel) from [<00208078>] (0x208078)
  ---[ end Kernel panic - not syncing: Out of memory and no killable processes...

A similar failure will also occur if ARM_ATAG_DTB_COMPAT isn't selected.

Fix this by setting a sane default (1 GB) in the dts file.
Signed-off-by: NJason Cooper <jason@lakedaemon.net>
Tested-by: NKevin Hilman <khilman@linaro.org>
Cc: <stable@vger.kernel.org> #v3.13+
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

The only remaining driver using the samsung dmadev code is the broken
samsung-ac97 sound driver. However, as found by Russell's autobuilder,
the elaborate dependency chains around it cause problems with
circular dependencies.

This is an attempt to simplify those dependencies by making the
SAMSUNG_DMADEV option user-selectable. I also try to keep the
default settings for all related options unchanged, so we don't
introduce any regressions against earlier testing on linux-next.

In particular, all s3c64xx and s5p* platforms keep selecting the
pl330 and pl08x drivers they require, but the select statement
is now moved towards the main platform option, and it remains
optional by unselecting CONFIG_DMADEVICES.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Cc: Kukjin Kim <kgene.kim@samsung.com>

The dynamic relocation that the keystone platform performs
only works if we can pick the phys offset at boot time. It's
possible that there is another solution for this, but this
is the easiest workaround. Kernels with ARM_PATCH_PHYS_VIRT
are not portable across platforms, and I see no reason why
anyone would run a kernel without ARM_PATCH_PHYS_VIRT on
keystone.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NSantosh Shilimkar <santosh.shilimkar@ti.com>

Commit d941f86f ("ARM: l2c: AM43x: add L2 cache support") enabled
the L2 cache support for the am43xx SoC, but caused a build regression
when the driver for that cache controller is disabled:

arch/arm/mach-omap2/built-in.o: In function `am43xx_init_early':
:(.init.text+0xb20): undefined reference to `omap_l2_cache_init'

This did not happen for OMAP4, which has the same call, but enables
the l2x0 driver unconditionally. We could do the same thing for
am43xx, but it seems better to allow turning it off and make the
code work in either case.

This adds an inline wrapper for omap_l2_cache_init for the disabled
case, and removes the 'select' from OMAP4 so it becomes a user
visible option.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NTony Lindgren <tony@atomide.com>
Cc: linux-omap@vger.kernel.org

"make vdso_install" installs unstripped versions of the vdso objects
for the benefit of the debugger.  This was broken by checkin:

6f121e54 x86, vdso: Reimplement vdso.so preparation in build-time C

The filenames are different now, so update the Makefile to cope.

This still installs the 64-bit vdso as vdso64.so.  We believe this
will be okay, as the only known user is a patched gdb which is known
to use build-ids, but if it turns out to be a problem we may have to
add a link.

Inspired by a patch from Sam Ravnborg.
Acked-by: NSam Ravnborg <sam@ravnborg.org>
Reported-by: NJosh Boyer <jwboyer@fedoraproject.org>
Tested-by: NJosh Boyer <jwboyer@fedoraproject.org>
Signed-off-by: NAndy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/b10299edd8ba98d17e07dafcd895b8ecf4d99eff.1402586707.git.luto@amacapital.netSigned-off-by: NH. Peter Anvin <hpa@zytor.com>

The Go runtime has a buggy vDSO parser that currently segfaults.
This writes an empty SHT_DYNSYM entry that causes Go's runtime to
malfunction by thinking that the vDSO is empty rather than
malfunctioning by running off the end and segfaulting.

This affects x86-64 only as far as we know, so we do not need this for
the i386 and x32 vdsos.
Signed-off-by: NAndy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/d10618176c4bd39b457a5e85c497295c90cab1bc.1402620737.git.luto@amacapital.netSigned-off-by: NH. Peter Anvin <hpa@zytor.com>

Add PUT_LE() by analogy with GET_LE() to write littleendian values in
addition to reading them.
Signed-off-by: NAndy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/3d9b27e92745b27b6fda1b9a98f70dc9c1246c7a.1402620737.git.luto@amacapital.netSigned-off-by: NH. Peter Anvin <hpa@zytor.com>

These defconfigs contain the CONFIG_M25P80 symbol, which is now
dependent on the MTD_SPI_NOR symbol. Add CONFIG_MTD_SPI_NOR to the
relevant defconfigs.

At the same time, drop the now-nonexistent CONFIG_MTD_CHAR symbol.
Signed-off-by: NBrian Norris <computersforpeace@gmail.com>
Cc: Steven Miao <realmz6@gmail.com>
Cc: adi-buildroot-devel@lists.sourceforge.net
Cc: linux-kernel@vger.kernel.org

The rule to create the final images uses a zImage.% pattern.
Unfortunately, this also matches the names of the zImage.*.lds linker
scripts, which appear as a dependency of the final images. This somehow
worked when $(srctree) used to be an absolute path, but now the pattern
matches too much. List only the images from $(image-y) as the target of
the rule, to avoid the circular dependency.
Reported-and-tested-by: NMike Qiu <qiudayu@linux.vnet.ibm.com>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Commit 2749a2f2 (powerpc/book3s: Fix machine check handling for
unhandled errors) introduced a few ABIv2 issues.

We can maintain ABIv1 and ABIv2 compatibility by branching to the
function rather than the dot symbol.

Fixes: 2749a2f2 ("powerpc/book3s: Fix machine check handling for unhandled errors")
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Currently we forward MCEs to guest which have been recovered by guest.
And for unhandled errors we do not deliver the MCE to guest. It looks like
with no support of FWNMI in qemu, guest just panics whenever we deliver the
recovered MCEs to guest. Also, the existig code used to return to host for
unhandled errors which was casuing guest to hang with soft lockups inside
guest and makes it difficult to recover guest instance.

This patch now forwards all fatal MCEs to guest causing guest to crash/panic.
And, for recovered errors we just go back to normal functioning of guest
instead of returning to host. This fixes soft lockup issues in guest.
This patch also fixes an issue where guest MCE events were not logged to
host console.
Signed-off-by: NMahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

We don't see MCE counter getting increased in /proc/interrupts which gives
false impression of no MCE occurred even when there were MCE events.
The machine check early handling was added for PowerKVM and we missed to
increment the MCE count in the early handler.

We also increment mce counters in the machine_check_exception call, but
in most cases where we handle the error hypervisor never reaches there
unless its fatal and we want to crash. Only during fatal situation we may
see double increment of mce count. We need to fix that. But for
now it always good to have some count increased instead of zero.
Signed-off-by: NMahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Currently machine check handler does not check for stack overflow for
nested machine check. If we hit another MCE while inside the machine check
handler repeatedly from same address then we get into risk of stack
overflow which can cause huge memory corruption. This patch limits the
nested MCE level to 4 and panic when we cross level 4.
Signed-off-by: NMahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Current code does not check for unhandled/unrecovered errors and return from
interrupt if it is recoverable exception which in-turn triggers same machine
check exception in a loop causing hypervisor to be unresponsive.

This patch fixes this situation and forces hypervisor to panic for
unhandled/unrecovered errors.

This patch also fixes another issue where unrecoverable_exception routine
was called in real mode in case of unrecoverable exception (MSR_RI = 0).
This causes another exception vector 0x300 (data access) during system crash
leading to confusion while debugging cause of the system crash.

Also turn ME bit off while going down, so that when another MCE is hit during
panic path, system will checkstop and hypervisor will get restarted cleanly
by SP.

With the above fixes we now throw correct console messages (see below) while
crashing the system in case of unhandled/unrecoverable machine checks.

--------------
Severe Machine check interrupt [[Not recovered]
  Initiator: CPU
  Error type: UE [Instruction fetch]
    Effective address: 0000000030002864
Oops: Machine check, sig: 7 [#1]
SMP NR_CPUS=2048 NUMA PowerNV
Modules linked in: bork(O) bridge stp llc kvm [last unloaded: bork]
CPU: 36 PID: 55162 Comm: bash Tainted: G           O 3.14.0mce #1
task: c000002d72d022d0 ti: c000000007ec0000 task.ti: c000002d72de4000
NIP: 0000000030002864 LR: 00000000300151a4 CTR: 000000003001518c
REGS: c000000007ec3d80 TRAP: 0200   Tainted: G           O  (3.14.0mce)
MSR: 9000000000041002 <SF,HV,ME,RI>  CR: 28222848  XER: 20000000
CFAR: 0000000030002838 DAR: d0000000004d0000 DSISR: 00000000 SOFTE: 1
GPR00: 000000003001512c 0000000031f92cb0 0000000030078af0 0000000030002864
GPR04: d0000000004d0000 0000000000000000 0000000030002864 ffffffffffffffc9
GPR08: 0000000000000024 0000000030008af0 000000000000002c c00000000150e728
GPR12: 9000000000041002 0000000031f90000 0000000010142550 0000000040000000
GPR16: 0000000010143cdc 0000000000000000 00000000101306fc 00000000101424dc
GPR20: 00000000101424e0 000000001013c6f0 0000000000000000 0000000000000000
GPR24: 0000000010143ce0 00000000100f6440 c000002d72de7e00 c000002d72860250
GPR28: c000002d72860240 c000002d72ac0038 0000000000000008 0000000000040000
NIP [0000000030002864] 0x30002864
LR [00000000300151a4] 0x300151a4
Call Trace:
Instruction dump:
XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX
XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX
---[ end trace 7285f0beac1e29d3 ]---

Sending IPI to other CPUs
IPI complete
OPAL V3 detected !
--------------
Signed-off-by: NMahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>