There's no need to save the runtime map details in global variables, the
values are only required to pass to efi_runtime_map_setup().

And because 'nr_efi_runtime_map' isn't needed, get_nr_runtime_map() can
be deleted along with 'efi_data_len'.

Cc: Dave Young <dyoung@redhat.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

Currently e820_reserve_setup_data() is called before parsing early
params, it works in normal case. But for memmap=exactmap, the final
memory ranges are created after parsing memmap= cmdline params, so the
previous e820_reserve_setup_data() has no effect. For example,
setup_data ranges will still be marked as normal system ram, thus when
later sysfs driver ioremap them kernel will warn about mapping normal
ram.

This patch fix it by moving the e820_reserve_setup_data() callback after
parsing early params so they can be set as reserved ranges and later
ioremap will be fine with it.
Signed-off-by: NDave Young <dyoung@redhat.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

kexec-tools use boot_params for getting the 1st kernel hardware_subarch,
the kexec kernel EFI runtime support also needs to read the old efi_info
from boot_params. Currently it exists in debugfs which is not a good
place for such infomation. Per HPA, we should avoid "sploit debugfs".

In this patch /sys/kernel/boot_params are exported, also the setup_data is
exported as a subdirectory. kexec-tools is using debugfs for hardware_subarch
for a long time now so we're not removing it yet.

Structure is like below:

/sys/kernel/boot_params
|__ data                /* boot_params in binary*/
|__ setup_data
|   |__ 0               /* the first setup_data node */
|   |   |__ data        /* setup_data node 0 in binary*/
|   |   |__ type        /* setup_data type of setup_data node 0, hex string */
[snip]
|__ version             /* boot protocal version (in hex, "0x" prefixed)*/
Signed-off-by: NDave Young <dyoung@redhat.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

Old kexec-tools can not load new kernels. The reason is kexec-tools does
not fill efi_info in x86 setup header previously, thus EFI failed to
initialize.  In new kexec-tools it will by default to fill efi_info and
pass other EFI required infomation to 2nd kernel so kexec kernel EFI
initialization can succeed finally.

To prevent from breaking userspace, add a new xloadflags bit so
kexec-tools can check the flag and switch to old logic.
Signed-off-by: NDave Young <dyoung@redhat.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

Add a new setup_data type SETUP_EFI for kexec use.  Passing the saved
fw_vendor, runtime, config tables and EFI runtime mappings.

When entering virtual mode, directly mapping the EFI runtime regions
which we passed in previously. And skip the step to call
SetVirtualAddressMap().

Specially for HP z420 workstation we need save the smbios physical
address.  The kernel boot sequence proceeds in the following order.
Step 2 requires efi.smbios to be the physical address.  However, I found
that on HP z420 EFI system table has a virtual address of SMBIOS in step
1.  Hence, we need set it back to the physical address with the smbios
in efi_setup_data.  (When it is still the physical address, it simply
sets the same value.)

1. efi_init() - Set efi.smbios from EFI system table
2. dmi_scan_machine() - Temporary map efi.smbios to access SMBIOS table
3. efi_enter_virtual_mode() - Map EFI ranges

Tested on ovmf+qemu, lenovo thinkpad, a dell laptop and an
HP z420 workstation.
Signed-off-by: NDave Young <dyoung@redhat.com>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

kexec kernel will need exactly same mapping for EFI runtime memory
ranges. Thus here export the runtime ranges mapping to sysfs,
kexec-tools will assemble them and pass to 2nd kernel via setup_data.

Introducing a new directory /sys/firmware/efi/runtime-map just like
/sys/firmware/memmap. Containing below attribute in each file of that
directory:

attribute  num_pages  phys_addr  type  virt_addr
Signed-off-by: NDave Young <dyoung@redhat.com>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

Export fw_vendor, runtime and config table physical addresses to
/sys/firmware/efi/{fw_vendor,runtime,config_table} because kexec kernels
need them.

From EFI spec these 3 variables will be updated to virtual address after
entering virtual mode. But kernel startup code will need the physical
address.
Signed-off-by: NDave Young <dyoung@redhat.com>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

Add two small functions:
efi_merge_regions() and efi_map_regions(), efi_enter_virtual_mode()
calls them instead of embedding two long for loop.
Signed-off-by: NDave Young <dyoung@redhat.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

Current code check boot service region with kernel text region by:
start+size >= __pa_symbol(_text)
The end of the above region should be start + size - 1 instead.

I see this problem in ovmf + Fedora 19 grub boot:
text start: 1000000 md start: 800000 md size: 800000
Signed-off-by: NDave Young <dyoung@redhat.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Acked-by: NToshi Kani <toshi.kani@hp.com>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Cc: stable@vger.kernel.org
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

Kexec kernel will use saved runtime virtual mapping, so add a new
function efi_map_region_fixed() for directly mapping a md to md->virt.

The md is passed in from 1st kernel, the virtual addr is saved in
md->virt_addr.
Signed-off-by: NDave Young <dyoung@redhat.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

variables size and end is useless in this function, thus remove them.
Reported-by: NToshi Kani <toshi.kani@hp.com>
Tested-by: NToshi Kani <toshi.kani@hp.com>
Signed-off-by: NDave Young <dyoung@redhat.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Signed-off-by: NMatt Fleming <matt.fleming@intel.com>

There are two code paths how page with pmd page table can be freed:
pmd_free() and pmd_free_tlb().

I've missed the second one and didn't add page table destructor call
there.  It leads to leak of page->ptl for pmd page tables, if
dynamically allocated page->ptl is in use.

The patch adds the missed destructor and modifies documentation
accordingly.
Signed-off-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reported-by: NAndrey Vagin <avagin@openvz.org>
Tested-by: NAndrey Vagin <avagin@openvz.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The VSX MSR bit in the user context indicates if the context contains VSX
state.  Currently we set this when the process has touched VSX at any stage.

Unfortunately, if the user has not provided enough space to save the VSX state,
we can't save it but we currently still set the MSR VSX bit.

This patch changes this to clear the MSR VSX bit when the user doesn't provide
enough space.  This indicates that there is no valid VSX state in the user
context.

This is needed to support get/set/make/swapcontext for applications that use
VSX but only provide a small context.  For example, getcontext in glibc
provides a smaller context since the VSX registers don't need to be saved over
the glibc function call.  But since the program calling getcontext may have
used VSX, the kernel currently says the VSX state is valid when it's not.  If
the returned context is then used in setcontext (ie. a small context without
VSX but with MSR VSX set), the kernel will refuse the context.  This situation
has been reported by the glibc community.

Based on patch from Carlos O'Donell.
Tested-by: NHaren Myneni <haren@linux.vnet.ibm.com>
Signed-off-by: NMichael Neuling <mikey@neuling.org>
Cc: stable@vger.kernel.org
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

In commit a489043f "Implement arch_get_random_long() based on H_RANDOM" I
broke the SMP=n build. We were getting plpar_wrappers.h via spinlock.h
which breaks when SMP=n.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Up until now we have only used cpu_to_chip_id() in the topology code,
which is only used on SMP builds. However my recent commit a4da0d50
"Implement arch_get_random_long/int() for powernv" added a usage when
SMP=n, breaking the build.

Move cpu_to_chip_id() into prom.c so it is available for SMP=n builds.

We would move the extern to prom.h, but that breaks the include in
topology.h. Instead we leave it in smp.h, but move it out of the
CONFIG_SMP #ifdef. We also need to include asm/smp.h in rng.c, because
the linux version skips asm/smp.h on UP. What a mess.
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

I encountered following issue:
[    0.283035] ibmvscsi 30000015: couldn't initialize event pool
[    5.688822] ibmvscsi: probe of 30000015 failed with error -1

which prevents the storage from being recognized, and the machine from
booting.

After some digging, it seems that it is caused by commit 4886c399

as dma_mask pointer in viodev->dev is not set, so in
dma_set_mask_and_coherent(), dma_set_coherent_mask() is not called
because dma_set_mask(), which is dma_set_mask_pSeriesLP() returned EIO.
While before the commit, dma_set_coherent_mask() is always called.

I tried to replace dma_set_mask_and_coherent() with
dma_coerce_mask_and_coherent(), and the machine could boot again.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

arch/powerpc/platforms/wsp/wsp.c: In function ‘wsp_probe_devices’:
arch/powerpc/platforms/wsp/wsp.c:76:3: error: implicit declaration of function ‘of_address_to_resource’ [-Werror=implicit-function-declaration]
Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Commit fba2369e (mm: use vm_unmapped_area() on powerpc architecture)
has a bug in slice_scan_available() where we compare an unsigned long
(high_slices) against a shifted int. As a result, comparisons against
the top 32 bits of high_slices (representing the top 32TB) always
returns 0 and the top of our mmap region is clamped at 32TB

This also breaks mmap randomisation since the randomised address is
always up near the top of the address space and it gets clamped down
to 32TB.

Cc: stable@vger.kernel.org # v3.10+
Signed-off-by: NAnton Blanchard <anton@samba.org>
Acked-by: NMichel Lespinasse <walken@google.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

If TM is not active there is no need to print PACATMSCRATCH
so we can save ourselves a line.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Acked-by: NMichael Neuling <mikey@neuling.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

When reading from the dispatch trace log (dtl) userspace interface, I
sometimes see duplicate entries. One example:

# hexdump -C dtl.out

00000000  07 04 00 0c 00 00 48 44  00 00 00 00 00 00 00 00
00000010  00 0c a0 b4 16 83 6d 68  00 00 00 00 00 00 00 00
00000020  00 00 00 00 10 00 13 50  80 00 00 00 00 00 d0 32

00000030  07 04 00 0c 00 00 48 44  00 00 00 00 00 00 00 00
00000040  00 0c a0 b4 16 83 6d 68  00 00 00 00 00 00 00 00
00000050  00 00 00 00 10 00 13 50  80 00 00 00 00 00 d0 32

The problem is in scan_dispatch_log() where we call dtl_consumer()
but bail out before incrementing the index.

To fix this I moved dtl_consumer() after the timebase comparison.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Cc: stable@vger.kernel.org
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

We waste quite a few lines in our oops output:

...
MSR: 8000000000009032 <SF,EE,ME,IR,DR,RI>  CR: 28044024  XER: 00000000
SOFTE: 0
CFAR: 0000000000009088
DAR: 000000000000001c, DSISR: 40000000

GPR00: c0000000000c74f0 c00000037cc1b010 c000000000d2bb30 0000000000000000
...

We can do a better job here and remove 3 lines:

MSR: 8000000000009032 <SF,EE,ME,IR,DR,RI>  CR: 28044024  XER: 00000000
CFAR: 0000000000009088 DAR: 0000000000000010, DSISR: 40000000 SOFTE: 1
GPR00: c0000000000e3d10 c00000037cc2fda0 c000000000d2c3a8 0000000000000001

Also move PACATMSCRATCH up, it doesn't really belong in the stack
trace section.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Machine check exceptions set DAR and DSISR, so print them in our
oops output.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

__get_user_pages_fast() may be called with interrupts disabled (see e.g.
get_futex_key() in kernel/futex.c) and therefore should use local_irq_save()
and local_irq_restore() instead of local_irq_disable()/enable().
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
CC: <stable@vger.kernel.org> [v3.12]
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

This clarifies in the log whether the error is a global PHB error
or an individual PE being frozen.
Signed-off-by: NGavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

On PHB3, we will fail to fetch IODA tables without PCI_COMMAND_MASTER
on PCI bridges. According to one experiment I had, the MSIx interrupts
didn't raise from the adapter without the bit applied to all upstream
PCI bridges including root port of the adapter. The patch forces to
have that bit enabled accordingly.
Signed-off-by: NGavin Shan <shangw@linux.vnet.ibm.com>
CC: <stable@vger.kernel.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

The ARM architected timer driver doesn't compile without
GENERIC_CLOCKEVENTS selected, so ensure that we select it when building
for a platform that has the timer.

Without this patch, mach-virt fails to build without something like
mach-vexpress also selected.
Acked-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Uwe reported a build failure when targetting a NOMMU platform with my
recent prefetch changes:

  arch/arm/lib/changebit.S: Assembler messages:
  arch/arm/lib/changebit.S:15: Error: architectural extension `mp' is
			not allowed for the current base architecture

This is due to use of the .arch_extension mp directive immediately prior
to an ALT_SMP(...) instruction. Whilst the ALT_SMP macro will expand to
nothing if !CONFIG_SMP, gas will still choke on the directive.

This patch fixes the issue by only emitting the sequence (including the
directive) if CONFIG_SMP=y.
Tested-by: NUwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Note that pmds[i] is simply uninitialized at that point...

Granted, it's very hard to hit (you need split page locks *and*
kmalloc(sizeof(spinlock_t), GFP_KERNEL) failing), but the code is
obviously bogus.

Introduced by commit 09ef4939 ("x86: add missed
pgtable_pmd_page_ctor/dtor calls for preallocated pmds")
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

I've finally tracked down why my CR signal-unwind test case still
fails on little-endian.  The problem turned to be that the kernel
installs a signal trampoline in the vDSO, and provides a DWARF CFI
record for that trampoline.  This CFI describes the save location
for CR:

  rsave (70, 38*RSIZE + (RSIZE - CRSIZE))

which is correct for big-endian, but points to the wrong word on
little-endian.   This is wrong no matter which ABI.

In addition, for the ELFv2 ABI, we should not only provide a CFI
record for register 70 (cr2), but for all CR fields separately.
Strictly speaking, I guess this would mean providing two separate
vDSO images, one for ELFv1 processes and one for ELFv2 processes (or
maybe playing some tricks with conditional DWARF expressions).
However, having CFI records for the other CR fields in ELFv1 is not
actually wrong, they just will be ignored.   So it seems the simplest
fix would be just to always provide CFI for all the fields.
Signed-off-by: NUlrich Weigand <Ulrich.Weigand@de.ibm.com>
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

With the little endian support merged, we can add the
CONFIG_CPU_LITTLE_ENDIAN kernel config option.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

The kernel doesn't build correctly using the ELFv2 ABI.  This patch
ensures that the ELFv1 ABI is used when building a kernel with an
ELFv2 enabled compiler.
Signed-off-by: NAlistair Popple <alistair@popple.id.au>
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

For the ELFv2 ABI, the hander is the entry point, not a function descriptor.
We also need to set up r12, and fortunately the fast_exception_return
exit path restores r12 for us so nothing else is required.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

No function descriptor, but we set r12 up and set TIF_RESTOREALL as it
normally isn't restored on return from syscall.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

We leave it at zero (though it could be 1) for old tasks.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Little endian ppc64 is getting an exciting new ABI.  This is reflected
by the bottom two bits of e_flags in the ELF header:

	0 == legacy binaries (v1 ABI)
	1 == binaries using the old ABI (compiled with a new toolchain)
	2 == binaries using the new ABI.

We store this in a thread flag, because we need to set it in core
dumps and for signal delivery.  Our chief concern is that it doesn't
use function descriptors.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

On little endian builds call H_SET_MODE so exceptions have the
correct endianness. We need to reset the endian during kexec
so do that in the MMU hashtable clear callback.
Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

We should not be using jump labels before they were initialized. Push back
the callback to until after jump label initialization.
Signed-off-by: NSasha Levin <sasha.levin@oracle.com>
Signed-off-by: NGleb Natapov <gleb@redhat.com>

For all uapi headers, need use "_UAPI" prefix for its guard macro
(which will be stripped by "scripts/headers_installer.sh").

Also remove redundant files (bitsperlong.h, errno.h, fcntl.h, ioctl.h,
ioctls.h, ipcbuf.h, kvm_para.h, mman.h, poll.h, resource.h, siginfo.h,
statfs.h, and unistd.h) which are already in Kbuild.

Also be sure that all "#endif" only have one empty line above, and each
file has guard macro.
Signed-off-by: NChen Gang <gang.chen@asianux.com>
Signed-off-by: NHans-Christian Egtvedt <hegtvedt@cisco.com>