Similarly to commit 2563a70c ("powerpc/64s: Remove unnecessary relocation
branch from idle handler"), the machine check handler has a BRANCH_TO from
relocated to relocated code, which is unnecessary.

It has also caused build errors with some toolchains:

  arch/powerpc/kernel/exceptions-64s.S: Assembler messages:
  arch/powerpc/kernel/exceptions-64s.S:395: Error: operand out of range
  (0xffffffffffff8280 is not between 0x0000000000000000 and
  0x000000000000ffff)

Fixes: 1945bc45 ("powerpc/64s: Fix POWER9 machine check handler from stop state")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reported-and-tested-by : Abdul Haleem <abdhalee@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Recently in commit f6eedbba ("powerpc/mm/hash: Increase VA range to 128TB")
we increased the virtual address space for user processes to 128TB by default,
and up to 512TB if user space opts in.

This obviously required expanding the range of the Linux page tables. For Book3s
64-bit using hash and with PAGE_SIZE=64K, we increased the PGD to 2^15 entries.
This meant we could cover the full address range, while still being able to
insert a 16G hugepage at the PGD level and a 16M hugepage in the PMD.

The downside of that geometry is that it uses a lot of memory for the PGD, and
in particular makes the PGD a 4-page allocation, which means it's much more
likely to fail under memory pressure.

Instead we can make the PMD larger, so that a single PUD entry maps 16G,
allowing the 16G hugepages to sit at that level in the tree. We're then able to
split the remaining bits between the PUG and PGD. We make the PGD slightly
larger as that results in lower memory usage for typical programs.

When THP is enabled the PMD actually doubles in size, to 2^11 entries, or 2^14
bytes, which is large but still < PAGE_SIZE.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Balbir Singh <bsingharora@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>

Makefile.postlink always includes include/config/auto.conf, however
this file is not present in a clean kernel tree, causing make to fail:

  $ git clone linuxppc.git
  $ cd linuxppc.git
  $ make distclean
  arch/powerpc/Makefile.postlink:10: include/config/auto.conf: No such file or directory
  make[1]: *** No rule to make target `include/config/auto.conf'.  Stop.
  make: *** [vmlinuxclean] Error 2

Equally running 'make distclean; make distclean' will trip the error case.

Change the inclusion such that file not being found does not trigger an error.

Fixes: f188d052 ("powerpc: Use the new post-link pass to check relocations")
Reported-by: Mircea Pop <mircea.pop@nxp.com>
Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Tested-by: Justin M. Forbes <jforbes@fedoraproject.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Similar to what is done in commit b6541db1 ("powerpc/eeh: Block
PCI config access upon frozen PE"), we need block PCI config access
for BCM5719 when recovering frozen error on them. Otherwise, an
unexpected recursive fenced PHB error is observed.

   0001:06:00.0 Ethernet controller: Broadcom Corporation \
                NetXtreme BCM5718 Gigabit Ethernet PCIe (rev 10)
   0001:06:00.1 Ethernet controller: Broadcom Corporation \
                NetXtreme BCM5718 Gigabit Ethernet PCIe (rev 10)
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Power9/ISAv3 has no VRMASD field in LPCR, we shouldn't be setting reserved bits,
so don't set them on Power9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Commit 616badd2 ("powerpc/powernv: Use OPAL call for TCE kill on
NVLink2") forced all TCE kills to go via the OPAL call for
NVLink2. However the PHB3 implementation of TCE kill was still being
called directly from some functions which in some circumstances caused
a machine check.

This patch adds an equivalent IODA2 version of the function which uses
the correct invalidation method depending on PHB model and changes all
external callers to use it instead.

Fixes: 616badd2 ("powerpc/powernv: Use OPAL call for TCE kill on NVLink2")
Cc: stable@vger.kernel.org # v4.11+
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Currently the radix TLB code includes support for CPUs that do *not*
have MMU_FTR_LOCKLESS_TLBIE. On those CPUs we are required to take a
global spinlock before issuing a tlbie.

Radix can only be built for 64-bit Book3s CPUs, and of those, only
POWER4, 970, Cell and PA6T do not have MMU_FTR_LOCKLESS_TLBIE. Although
it's possible to build a kernel with Radix support that can also boot on
those CPUs, we happen to know that in reality none of those CPUs support
the Radix MMU, so the code can never actually run on those CPUs.

So remove the native_tlbie_lock in the Radix TLB code.

Note that there is another lock of the same name in the hash code, which
is unaffected by this patch.
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

machine_check_early() gets called in real mode. The very first time when
add_taint() is called, it prints a warning which ends up calling opal
call (that uses OPAL_CALL wrapper) for writing it to console. If we get a
very first machine check while we are in opal we are doomed. OPAL_CALL
overwrites the PACASAVEDMSR in r13 and in this case when we are done with
MCE handling the original opal call will use this new MSR on it's way
back to opal_return. This usually leads to unexpected behaviour or the
kernel to panic. Instead move the add_taint() call later in the virtual
mode where it is safe to call.

This is broken with current FW level. We got lucky so far for not getting
very first MCE hit while in OPAL. But easily reproducible on Mambo.

Fixes: 27ea2c42 ("powerpc: Set the correct kernel taint on machine check errors.")
Cc: stable@vger.kernel.org # v4.2+
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

The entire body of unregister_cpu_online() is inside an #ifdef
CONFIG_HOTPLUG_CPU block. This is ugly and means we create an empty function
when hotplug is disabled for no reason.

Instead move the #ifdef out of the function body and define the function to be
NULL in the else case. This means we'll pass NULL to cpuhp_setup_state(), but
that's fine because it accepts NULL to mean there is no teardown callback, which
is exactly what we want.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

This code was until recently completely undocumented and even now the comment is
not very verbose.

We've already had one patch sent to remove the IRQ enable/disable because it's
"paradoxical and unnecessary". So document it thoroughly to save anyone else
from puzzling over it.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Otherwise we might select it when its dependenices aren't enabled,
leading to a build break.

It's default y anyway, so will be on unless someone disables it
manually.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

pnv_eeh_reset() has special handling for PEs whose primary bus is the
root bus or the bus immediately underneath the root port.

The cxl bi-modal card support added in b0b5e591 ("cxl: Add
cxl_check_and_switch_mode() API to switch bi-modal cards") relies on
this behaviour when hot-resetting the CAPI adapter following a mode
switch.  Document this in pnv_eeh_reset() so we don't accidentally break
it.
Suggested-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Remove unnecessary tags in eeh_handle_normal_event(), and add function
comments for eeh_handle_normal_event() and eeh_handle_special_event().

The only functional difference is that in the case of a PE reaching the
maximum number of failures, rather than one message telling you of this
and suggesting you reseat the device, there are two separate messages.
Suggested-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Russell Currey <ruscur@russell.cc>
Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

eeh_handle_special_event() is called when an EEH event is detected but
can't be narrowed down to a specific PE.  This function looks through
every PE to find one in an erroneous state, then calls the regular event
handler eeh_handle_normal_event() once it knows which PE has an error.

However, if eeh_handle_normal_event() found that the PE cannot possibly
be recovered, it will free it, rendering the passed PE stale.
This leads to a use after free in eeh_handle_special_event() as it attempts to
clear the "recovering" state on the PE after eeh_handle_normal_event() returns.

Thus, make sure the PE is valid when attempting to clear state in
eeh_handle_special_event().

Fixes: 8a6b1bc7 ("powerpc/eeh: EEH core to handle special event")
Cc: stable@vger.kernel.org # v3.11+
Reported-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Russell Currey <ruscur@russell.cc>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

In some situations, a faulty AFU slice may create an interrupt storm of
slice errors, rendering the machine unusable. Since these interrupts are
informational only, present the interrupt once, then mask it off to
prevent it from being retriggered until the AFU is reset.
Signed-off-by: Alastair D'Silva <alastair@d-silva.org>
Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Reviewed-by: Vaibhav Jain <vaibhav@linux.vnet.ibm.com>
Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Fix a boundary condition where in some cases an eeh event that results
in card reset isn't passed on to a driver attached to the virtual PCI
device associated with a slice. This will happen in case when a slice
attached device driver returns a value other than
PCI_ERS_RESULT_NEED_RESET from the eeh error_detected() callback. This
would result in an early return from cxl_pci_error_detected() and
other drivers attached to other AFUs on the card wont be notified.

The patch fixes this by making sure that all slice attached
device-drivers are notified and the return values from
error_detected() callback are aggregated in a scheme where request for
'disconnect' trumps all and 'none' trumps 'need_reset'.

Fixes: 9e8df8a2 ("cxl: EEH support")
Cc: stable@vger.kernel.org # v4.3+
Signed-off-by: Vaibhav Jain <vaibhav@linux.vnet.ibm.com>
Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

During an eeh event when the cxl card is fenced and card sysfs attr
perst_reloads_same_image is set following warning message is seen in the
kernel logs:

  Adapter context unlocked with 0 active contexts
  ------------[ cut here ]------------
  WARNING: CPU: 12 PID: 627 at
  ../drivers/misc/cxl/main.c:325 cxl_adapter_context_unlock+0x60/0x80 [cxl]

Even though this warning is harmless, it clutters the kernel log
during an eeh event. This warning is triggered as the EEH callback
cxl_pci_error_detected doesn't obtain a context-lock before forcibly
detaching all active context and when context-lock is released during
call to cxl_configure_adapter from cxl_pci_slot_reset, a warning in
cxl_adapter_context_unlock is triggered.

To fix this warning, we acquire the adapter context-lock via
cxl_adapter_context_lock() in the eeh callback
cxl_pci_error_detected() once all the virtual AFU PHBs are notified
and their contexts detached. The context-lock is released in
cxl_pci_slot_reset() after the adapter is successfully reconfigured
and before the we call the slot_reset callback on slice attached
device-drivers.

Fixes: 70b565bb ("cxl: Prevent adapter reset if an active context exists")
Cc: stable@vger.kernel.org # v4.9+
Reported-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Signed-off-by: Vaibhav Jain <vaibhav@linux.vnet.ibm.com>
Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com>
Reviewed-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Tested-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

This was a hack we added to work around the allmodconfig build breaking, see
commit fb43e847 ("powerpc: Disable RELOCATABLE for COMPILE_TEST with
PPC64").

Since we merged the thin archives support in commit 43c9127d ("powerpc: Add
option to use thin archives") this hasn't been necessary, so remove it.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Currently if we take an oops caused by an 0x380 or 0x480 exception, we get a
print which assumes SLB problems. With radix, these vectors have different
meanings.

This patch updates the oops message to reflect these different meanings.
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Michal Suchánek noticed a comment in book3s/64/mmu-hash.h about the context ids
we use for the kernel was inconsistent with the code and other comments in the
same file.

It should read 1-4 not 1-5.

While we're touching it, update "address" to "addresses" which makes more sense
as it's referring to more than one address below.
Reported-by: Michal Suchánek <msuchanek@suse.de>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

This enables VFIO on pseries host in order to allow VFIO in nested guest under
PR KVM or DPDK in a HV guest. This adds support of the VFIO_SPAPR_TCE_IOMMU
type.

This adds exchange() callback to allow TCE updates by the SPAPR TCE IOMMU
driver in VFIO.

This initializes DMA32 window parameters in iommu_table_group as as this does
not implement VFIO_SPAPR_TCE_v2_IOMMU and VFIO_SPAPR_TCE_IOMMU just reuses the
existing DMA32 window.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

When the userspace requests a small TCE table (which takes less than
the system page size) and more than 1 TCE level, the existing code
returns a single page size which is a bug as each additional TCE level
requires at least one page and this is what
pnv_pci_ioda2_table_alloc_pages() does. And we end up seeing
WARN_ON(!ret && ((*ptbl)->it_allocated_size != table_size))
in drivers/vfio/vfio_iommu_spapr_tce.c.

This replaces incorrect _ALIGN_UP() (which aligns zero up to zero) with
max_t() to fix the bug.

Besides removing WARN_ON(), there should be no other changes in
behaviour.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

pnv_pci_table_alloc() ignores possible failure from kzalloc_node(),
this adds a check. There are 2 callers of pnv_pci_table_alloc(),
one already checks for tbl!=NULL, this adds WARN_ON() to the other path
which only happens during boot time in IODA1 and not expected to fail.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Move a couple of existing scripts under there. Remove scripts directory:
a script is a tool, a tool is not a script.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Currently powerpc has to introduce a dependency on its default build
target zImage in order to run a relocation check pass over the linked
vmlinux. This is deficient because the check is not run if the plain
vmlinux target is built, or if one of the other boot targets is built.

Switch to using the kbuild post-link pass, added in commit fbe6e37d
("kbuild: add arch specific post-link Makefile") in order to run this
check. In future powerpc will use this to do more complicated operations,
but initially using it for something simple is a good first step.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

An externally triggered system reset (e.g., via QEMU nmi command, or pseries
reset button) can cause system reset interrupts on all CPUs. In case this causes
xmon to be entered, it is undesirable for the primary (first) CPU into xmon to
trigger an NMI IPI to others, because this may cause a nested system reset
interrupt.

So spin for a time waiting for secondaries to join xmon before performing the
NMI IPI, similarly to what the crash dump code does.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Only do it when we come in from system reset, not via sysrq etc.]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Have the NMI IPI code use this op when the platform defines it.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Add a simple NMI IPI system that handles concurrency and reentrancy.

The platform does not have to implement a true non-maskable interrupt,
the default is to simply use the debugger break IPI message. This has
now been co-opted for a general IPI message, and users (debugger and
crash) have been reimplemented on top of the NMI system.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Incorporate incremental fixes from Nick]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

System reset is a non-maskable interrupt from Linux's point of view
(occurs under local_irq_disable()), so it should use nmi_enter/exit.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

The system reset interrupt is used for crash/debug situations, so it is
desirable to have as little impact on the normal state of the system as
possible.

Currently it uses the current kernel stack to process the exception.
This stores into the stack which may be involved with the crash. The
stack pointer may be corrupted, or it may have overflowed.

Avoid or minimise these problems by creating a dedicated NMI stack for
the system reset interrupt to use.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

In preparation for using a dedicated stack for system reset interrupts,
prevent a nested system reset from recovering, in order to simplify
code that is called in crash/debug path. This allows a system reset
interrupt to just use the base stack pointer.

Keep an in_nmi nesting counter similarly to the in_mce counter. Consider
the interrrupt non-recoverable if it is taken inside another system
reset.

Interrupt nesting could be allowed similarly to MCE, but system reset
is a special case that's not for normal operation, so simplicity wins
until there is requirement for nested system reset interrupts.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

The system reset interrupt can occur when MSR_EE=0, and it currently
uses the PACA_EXGEN save area.

Some PACA_EXGEN interrupts have a window where MSR_RI=1 and MSR_EE=0
when the save area is still in use. A system reset interrupt in this
window can lead to undetected corruption when the save area gets
overwritten.

This patch introduces PACA_EXNMI save area for system reset exceptions,
which closes this corruption window. It's also helpful to retain the
EXGEN state for debugging situations, even if not considering the
recoverability aspect.

This patch also moves the PACA_EXMC area down to a less frequently used
part of the paca with the new save area.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

This code is common to a few exceptions, and another user will be added.
This causes a trivial change to generated code:

-     604: std     r9,416(r1)
-     608: mfspr   r11,314
-     60c: std     r11,368(r1)
-     610: mfspr   r12,315
+     604: mfspr   r11,314
+     608: mfspr   r12,315
+     60c: std     r9,416(r1)
+     610: std     r11,368(r1)

machine_check_powernv_early could also use this, but that requires non
trivial changes to generated code, so that's for another patch.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Subsequent patches will add more non-RI variant exceptions, so
create a macro for it rather than open-code it.

This does not change generated instructions.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Cell will wake from low power state at the system reset interrupt,
with the event encoded in SRR1, rather than waking at the interrupt
vector that corresponds to that event.

The system reset handler for this platform decodes SRR1 event reason
and calls the interrupt handler to process it directly from the system
reset handlre.

A subsequent change will treat the system reset interrupt as a Linux NMI
with its own per-CPU stack, and this will no longer work. Remove the
external and decrementer handlers from the system reset handler.

- The external exception remains raised and will fire again at the
  EE interrupt vector when system reset returns.

- The decrementer is set to 1 so it will be raised again and fire when
  the system reset returns.

It is possible to branch to an idle handler from the system reset
interrupt (like POWER does), then restore a normal stack and restore
this optimisation. But simplicity wins for now.
Tested-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

PA Semi will wake from low power state at the system reset interrupt,
with the event encoded in SRR1, rather than waking at the interrupt
vector that corresponds to that event.

The system reset handler for this platform decodes SRR1 event reason
and calls the interrupt handler to process it directly from the system
reset handlre.

A subsequent change will treat the system reset interrupt as a Linux NMI
with its own per-CPU stack, and this will no longer work. Remove the
external and decrementer handlers from the system reset handler.

- The external exception remains raised and will fire again at the
  EE interrupt vector when system reset returns.

- The decrementer is set to 1 so it will be raised again and fire when
  the system reset returns.

It is possible to branch to an idle handler from the system reset
interrupt (like POWER does), then restore a normal stack and restore
this optimisation. But simplicity wins for now.
Tested-by: Christian Zigotzky <chzigotzky@xenosoft.de>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

Merge the topic branch we were sharing with kvm-ppc, Paul has also
merged it.

Split ftrace_64.S further retaining the core ftrace 64-bit aspects
in ftrace_64.S and moving ftrace_caller() and ftrace_graph_caller() into
separate files based on -mprofile-kernel. The livepatch routines are all
now contained within the mprofile file.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

entry_*.S now includes a lot more than just kernel entry/exit code. As a
first step at cleaning this up, let's split out the ftrace bits into
separate files. Also move all related tracing code into a new trace/
subdirectory.

No functional changes.
Suggested-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>