mpic_physmask was looping NR_CPUS times over a mask that was passed as
a u32. Since mpic is architecturaly limited to 32 physical cpus, clamp
the logical cpus to 32 when compiling (we could also clamp at runtime
to nr_cpu_ids).
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
c1854e0072 (powerpc: Set nr_cpu_ids early
and use it to free PACAs) copied the formerly static setup_nr_cpu_ids
from init/main.c but 34db18a054 (smp:
move smp setup functions to kernel/smp.c) moved it to kernel/smp.c
with a declaration in include/linux/smp.h, so we can call it instead of
replicating it.
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Now that we never set a cpu above nr_cpu_ids possible we can
limit our initial paca allocation to nr_cpu_ids. We can then
clamp the number of cpus in platforms/iseries/setup.c.
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We should not set cpus above nr_cpu_ids to possible. While we
will trigger a warning with CONFIG_CPUMASK_DEBUG, even then the mask
initializers will set the bits beyond what the iterators check and cause
nr_cpu_ids to increase.
Respecting nr_cpu_ids during setup will allow us to use it in our initial
paca allocation. It can be reduced from NR_CPUS by the existing early param
nr_cpus=, which was added in 2b633e3fac (smp:
Use nr_cpus= to set nr_cpu_ids early). We already call parse_early_parms
between finding the command line and allocating the pacas.
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
9cb82f2f46 (Make iSeries spin on
__secondary_hold_spinloop, like pSeries) added a load of current_set
but this load was repeated later and we don't even have the paca yet.
It also checked __secondary_hold_spinloop with a 32 bit compare instead
of a 64 bit compare.
b6f6b98a4e (Don't spin on sync instruction
at boot time) missed the copy of the startup code in iseries.
1426d5a3bd (Dynamically allocate pacas)
doesn't allow for pacas to be less than lppacas and recalculated the paca
location from the cpu id in r0 every time through the secondary loop.
Various revisions over time made the comments on conditional branches
confusing with respect to being a hold loop or forward progress
Mostly in-order description of the changes:
Replicate the few lines of code saved by the ugly scoped ifdef CONFIG_SMP
in the secondary loop between yielding on UP and marking time with the
hypervisor on SMP. Always compile the iseries_secondary_yield loop and
use it if the cpu id is above nr_cpu_ids. Change all forward progress
paths to be forward branches to the next numerical label. Assign a
label to all loops. Move all sync instructions from the loops to the
forward progress path. Wait to load current_set until paca is set to go.
Move the iseries_secondary_smp_loop label to cover the whole spin loop.
Add HMT_MEDIUM when we make forward progress.
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Starting with 1426d5a3bd (powerpc:
Dynamically allocate pacas) the space for pacas beyond cpu_possible
is freed, but we failed to update the loop in crash.c.
Since c1854e0072 (powerpc: Set nr_cpu_ids
early and use it to free PACAs) the number of pacas allocated is
always nr_cpu_ids.
Signed-off-by: Milton Miller <miltonm@bga.com>
Cc: <stable@kernel.org> # .34.x
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Starting with 1426d5a3bd (powerpc:
Dynamically allocate pacas) we free the memory for pacas beyond
cpu_possible, but we failed to update the loop the secondary cpus use
to find their paca. If the system has running cpu threads for which
the kernel did not allocate a paca for they will search the memory that
was freed. For instance this could happen when the device tree for
a kdump kernel was not updated after a cpu hotplug, or the kernel is
running with more cpus than the kernel was configured.
Since c1854e0072 (powerpc: Set nr_cpu_ids
early and use it to free PACAs) we set nr_cpu_ids before telling the
cpus to advance, so use that to limit the search.
We can't reference nr_cpu_ids without CONFIG_SMP because it is defined
as 1 instead of a memory location, but any extra threads should be sent
to kexec_wait in that case anyways, so make that explicit and remove
the search loop for UP.
Note to stable: The fix also requires
c1854e0072 (powerpc: Set
nr_cpu_ids early and use it to free PACAs) to function. Also
9d07bc841c (Properly handshake CPUs going
out of boot spin loop) affects the second chunk, specifically the branch
target was 3b before and is 4b after that patch, and there was a blank
line before the #ifdef CONFIG_SMP that was removed
Cc: <stable@kernel.org> # .34.x: c1854e0072 powerpc: Set nr_cpu_ids early
Cc: <stable@kernel.org> # .34.x
Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Commit 1fc711f7ff (powerpc/kexec: Fix race
in kexec shutdown) moved the write to signal the cpu had exited the kernel
from before the transition to real mode in kexec_smp_wait to kexec_wait.
Unfornately it missed that kexec_wait is used both by cpus leaving the
kernel and by secondary slave cpus that were not allocated a paca for
what ever reason -- they could be beyond nr_cpus or not described in
the current device tree for whatever reason (for example, kexec-load
was not refreshed after a cpu hotplug operation). Cpus coming through
that path they will write to paca[NR_CPUS] which is beyond the space
allocated for the paca data and overwrite memory not allocated to pacas
but very likely still real mode accessable).
Move the write back to kexec_smp_wait, which is used only by cpus that
found their paca, but after the transition to real mode.
Signed-off-by: Milton Miller <miltonm@bga.com>
Cc: <stable@kernel.org> # (1fc711f was backported to 2.6.32)
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
I have a report of an FWNMI with an r3 value that we think is
corrupt, but since we don't print r3 we have no idea what was
wrong with it.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When we swtich to direct dma ops, we set the dma data union to have the
dma offset. When we switch back to iommu table ops because of a later
dma_set_mask, we need to restore the iommu table pointer. Without this
change, crashes have been observed on kexec where (for reasons still
being investigated) we fall back to a 32-bit dma mask on a particular
device and then panic because the table pointer is not valid.
The easiset way to find this value is to call
pci_dma_dev_setup_pSeriesLP which will search up the pci tree until it
finds the node with the table.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Milton Miller <miltonm@bga.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Anton Blanchard <anton@samba.org>
Cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We have a confusing number of ioremap functions. Make things just a
bit simpler by merging ioremap_flags and ioremap_prot.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Add ioremap_wc so drivers can request write combining on kernel
mappings.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
After looking at our system call path, Mary Brown suggested that we
should put all mfspr SRR* instructions before any mtspr SRR*.
To test this I used a very simple null syscall (actually getppid)
testcase at http://ozlabs.org/~anton/junkcode/null_syscall.c
I tested with the following changes against the pseries_defconfig:
CONFIG_VIRT_CPU_ACCOUNTING=n
CONFIG_AUDIT=n
to remove the overhead of virtual CPU accounting and syscall
auditing.
POWER6:
baseline: mean = 757.2 cycles sd = 2.108
modified: mean = 759.1 cycles sd = 2.020
POWER7:
baseline: mean = 411.4 cycles sd = 0.138
modified: mean = 404.1 cycles sd = 0.109
So we have 1.77% improvement on POWER7 which looks significant. The
POWER6 suggest a 0.25% slowdown, but the results are within 1
standard deviation and may be in the noise.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
A static branch hint will override dynamic branch prediction on
recent POWER CPUs. Since we are about to use more altivec in the
kernel remove the static hint in giveup_altivec that assumes
a userspace task is using altivec.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To make it easier to add optimised versions of copy_page, remove
the 4kB loop for 64kB pages and just do all the work in copy_page.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Enable iSCSI support for a number of cards. We had the base
networking devices enabled but forgot to enable iSCSI.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Enable the Qlogic and Emulex 10Gbit adapters.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The variable 'old' is set but not used in the wrprotect functions in
arch/powerpc/include/asm/pgtable-ppc64.h, which can trigger a compiler warning.
Remove the variable, since it's not used anyway.
Signed-off-by: Stratos Psomadakis <psomas@ece.ntua.gr>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Future releases of fimrware will enforce a requirement that DTL buffers
do not cross a 4k boundary. Commit
127493d5dc satisfies this requirement for
CONFIG_VIRT_CPU_ACCOUNTING=y kernels, but if !CONFIG_VIRT_CPU_ACCOUNTING
&& CONFIG_DTL=y, the current code will fail at dtl registration time.
Fix this by making the kmem cache from
127493d5dc visible outside of setup.c and
using the same cache in both dtl.c and setup.c. This requires a bit of
reorganization to ensure ordering of the kmem cache and buffer
allocations.
Note: Since firmware now limits the size of the buffer, I made
dtl_buf_entries read-only in debugfs.
Tested with upcoming firmware with the 4 combinations of
CONFIG_VIRT_CPU_ACCOUNTING and CONFIG_DTL.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Anton Blanchard <anton@samba.org>
Cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When we kexec we look for a particular property added by the first
kernel, "linux,direct64-ddr-window-info", per-device where we already
have set up dynamic dma windows. The current code, though, wasn't
initializing the size of this property and thus when we kexec'd, we
would find the property but read uninitialized memory resulting in
garbage ddw values for the kexec'd kernel and panics. Fix this by
setting the size at enable_ddw() time and ensuring that the size of the
found property is valid at dupe_ddw_if_kexec() time.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch below removes an unused config variable found by using a kernel
cleanup script.
Note: I did try to cross compile these but hit erros while doing so..
(gcc is not setup to cross compile) and am unsure if anymore needs to be done.
Please have a look if/when anybody has free time.
Signed-off-by: Justin P. Mattock <justinmattock@gmail.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The timestamps recorded in the .gz files add no value.
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Jack Miller <jack@codezen.org>
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Add a platform for the Wire Speed Processor, based on the PPC A2.
This includes code for the ICS & OPB interrupt controllers, as well
as a SCOM backend, and SCOM based cpu bringup.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Jack Miller <jack@codezen.org>
Signed-off-by: Ian Munsie <imunsie@au1.ibm.com>
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
For adapters which have devices under a PCIe switch/bridge it is informative
to display information for both the PCIe switch/bridge and the device on
which the bus error was detected.
rebased to powerpc-next
Signed-off-by: Richard A Lary <rlary@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
slb0_limit() wasn't a very descriptive name. This changes it along with
a comment explaining what it's used for, and provides a 64-bit BookE
implementation.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This patch adds support for handling IO Event interrupts which come
through at the /event-sources/ibm,io-events device tree node.
The interrupts come through ibm,io-events device tree node are generated
by the firmware to report IO events. The firmware uses the same interrupt
to report multiple types of events for multiple devices. Each device may
have its own event handler. This patch implements a plateform interrupt
handler that is triggered by the IO event interrupts come through
ibm,io-events device tree node, pull in the IO events from RTAS and call
device event handlers registered in the notifier list.
Device event handlers are expected to use atomic_notifier_chain_register()
and atomic_notifier_chain_unregister() to register/unregister their
event handler in pseries_ioei_notifier_list list with IO event interrupt.
Device event handlers are responsible to identify if the event belongs
to the device event handler. The device event handle should return NOTIFY_OK
after the event is handled if the event belongs to the device event handler,
or NOTIFY_DONE otherwise.
Signed-off-by: Tseng-Hui (Frank) Lin <thlin@us.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This patch adds definitions of non-IBM specific v6 extended log
definitions to rtas.h.
Signed-off-by: Tseng-Hui (Frank) Lin <tsenglin@us.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Due to a collision between NO_CONTEXT->MMU_NO_CONTEXT change and
Anton's patch.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Fundamental reset is an optional reset type supported only by PCIe adapters.
Handle the unexpected case where a non-PCIe device has requested a
fundamental reset. Try hot-reset as a fallback to handle this case.
Signed-off-by: Richard A Lary <rlary@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
For multifunction adapters with a PCI bridge or switch as the device
at the Partitionable Endpoint(PE), if one or more devices below PE
sets dev->needs_freset, that value will be set for the PE device.
In other words, if any device below PE requires a fundamental reset
the PE will request a fundamental reset.
Signed-off-by: Richard A Lary <rlary@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Adds support for page coalescing, which is a feature on IBM Power servers
which allows for coalescing identical pages between logical partitions.
Hint text pages as coalesce candidates, since they are the most likely
pages to be able to be coalesced between partitions. This patch also
exports some page coalescing statistics available from firmware via
lparcfg.
[BenH: Moved a couple of things around to fix compile problems]
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Commit b987812b3f left
crash_kexec_wait_realmode() undefined for UP.
Commit 7c7a81b53e defined it for UP but
left it undefined for 32-bit SMP.
Seems like people are getting confused by nested #ifdef's, so move the
definitions of crash_kexec_wait_realmode() after the #ifdef CONFIG_SMP
section.
Compile-tested with 32-bit UP, 32-bit SMP and 64-bit SMP configurations.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Adapt new API.
Almost change is trivial. Most important change is the below line
because we plan to change task->cpus_allowed implementation.
- ctx->cpus_allowed = current->cpus_allowed;
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Recent 64-bit server processors (POWER6 and POWER7) have a "Come-From
Address Register" (CFAR), that records the address of the most recent
branch or rfid (return from interrupt) instruction for debugging purposes.
This saves the value of the CFAR in the exception entry code and stores
it in the exception frame. We also make xmon print the CFAR value in
its register dump code.
Rather than extend the pt_regs struct at this time, we steal the orig_gpr3
field, which is only used for system calls, and use it for the CFAR value
for all exceptions/interrupts other than system calls. This means we
don't save the CFAR on system calls, which is not a great problem since
system calls tend not to happen unexpectedly, and also avoids adding the
overhead of reading the CFAR to the system call entry path.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When we take an interrupt or exception from kernel mode and the stack
pointer is obviously not a kernel address (i.e. the top bit is 0), we
switch to an emergency stack, save register values and panic. However,
on 64-bit server machines, we don't actually save the values of r9 - r13
at the time of the interrupt, but rather values corrupted by the
exception entry code for r12-r13, and nothing at all for r9-r11.
This fixes it by passing a pointer to the register save area in the paca
through to the bad_stack code in r3. The register values are saved in
one of the paca register save areas (depending on which exception this
is). Using the pointer in r3, the bad_stack code now retrieves the
saved values of r9 - r13 and stores them in the exception frame on the
emergency stack. This also stores the normal exception frame marker
("regshere") in the exception frame.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Icswx is a PowerPC instruction to send data to a co-processor. On Book-S
processors the LPAR_ID and process ID (PID) of the owning process are
registered in the window context of the co-processor at initialization
time. When the icswx instruction is executed the L2 generates a cop-reg
transaction on PowerBus. The transaction has no address and the
processor does not perform an MMU access to authenticate the transaction.
The co-processor compares the LPAR_ID and the PID included in the
transaction and the LPAR_ID and PID held in the window context to
determine if the process is authorized to generate the transaction.
The OS needs to assign a 16-bit PID for the process. This cop-PID needs
to be updated during context switch. The cop-PID needs to be destroyed
when the context is destroyed.
Signed-off-by: Sonny Rao <sonnyrao@linux.vnet.ibm.com>
Signed-off-by: Tseng-Hui (Frank) Lin <thlin@linux.vnet.ibm.com>
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This removes MMU_FTR_TLBIE_206 as we can now use CPU_FTR_HVMODE_206. It
also changes the logic to select which tlbie to use to be based on this
new CPU feature bit.
This also duplicates the ASM_FTR_IF/SET/CLR defines for CPU features
(copied from MMU features).
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Added support for ibm,configure-pe RTAS call introduced with
PAPR 2.2.
Signed-off-by: Richard A. Lary <rlary@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Some of the 64bit PPC CPU features are MMU-related, so this patch moves
them to MMU_FTR_ bits. All cpu_has_feature()-style tests are moved to
mmu_has_feature(), and seven feature bits are freed as a result.
Signed-off-by: Matt Evans <matt@ozlabs.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
RTAS returns extended error codes as a hint of how long the
OS might want to wait before retrying a call. If we have nothing
else useful to do we may as well call back straight away.
This was found when testing the new dynamic dma window feature.
Firmware split the zeroing of the TCE table into 32k chunks but
returned 9901 (which is a suggested wait of 10ms). All up this took
about 10 minutes to complete since msleep is jiffies based and will
round 10ms up to 20ms.
With the patch below we take 3 seconds to complete the same test.
The hint firmware is returning in the RTAS call should definitely
be decreased, but even if we slept 1ms each iteration this would
take 32s.
Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The calculation of the size for the exception save area of the TLB
miss handler is wrong, luckily it's too big not too small.
Rework it to make it a bit clearer, and also correct. We want 3 save
areas, each EX_TLB_SIZE _bytes_.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We check MSR_SF a lot in sstep.c, to decide if we need to emulate the
truncation of values when running in 32-bit mode. Factor out that code
into a helper, and convert it and the other uses to use MSR_64BIT.
This fixes a bug on BOOK3E where kprobes would end up returning to a
32-bit address, because regs->nip was truncated, because (msr & MSR_SF)
was false.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Use the new MSR_64BIT in a few places. Some of these are already ifdef'ed
for BOOKE vs BOOKS, but it's still clearer, MSR_SF does not immediately
parse as "MSR bit for 64bit".
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The MSR bit which indicates 64-bit-ness is different between server and
booke, so add a #define which gives you the right mask regardless.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
BT_L2CAP and BT_SCO have changed to bool .
Value 'm' has invalid .
Signed-off-by: Wanlong Gao <wanlong.gao@gmail.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
commit ec775d0e70 ("powerpc: Convert to new irq_*
function names") changed a call from set_irq_chip_data() to
irq_set_chip_data(), but forgot to update the corresponding debug message
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Geoff Levand <geoff@infradead.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This can be useful for differentiating interrupts on the same host
but with different chip data.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
