When reading from the LPC, the OPAL FW calls return the value via pointer
to a uint32_t which is always returned big endian. Our internal inb/outb
implementation byteswaps that fine but our debugfs code is still broken.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
CC: <stable@vger.kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Add printk levels to powernv platform code, and convert to
pr_err() etc while here.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Move MSI checks from arch_msi_check_device() to arch_setup_msi_irqs().
This makes the code more compact and allows removing
arch_msi_check_device() from generic MSI code.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
The dma_get_required_mask() function is used by some drivers to
query the platform about what DMA mask is needed to cover all of
memory. This is a bit of a strange semantic when we have to choose
between IOMMU translation or bypass, but essentially what it means
is "what DMA mask will give best performances".
Currently, our IOMMU backend always returns a 32-bit mask here, we
don't do anything special to it when we have bypass available. This
causes some drivers to choose a 32-bit mask, thus losing the ability
to use the bypass window, thinking this is more efficient. The problem
was reported from the driver of following device:
0004:03:00.0 0107: 1000:0087 (rev 05)
0004:03:00.0 Serial Attached SCSI controller: LSI Logic / Symbios \
Logic SAS2308 PCI-Express Fusion-MPT SAS-2 (rev 05)
This patch adds an override of that function in order to, instead,
return a 64-bit mask whenever a bypass window is available in order
for drivers to prefer this configuration.
Reported-by: Murali N. Iyer <mniyer@us.ibm.com>
Suggested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
It should have been part of commit 1ad7a72c5 ("powerpc/eeh: Report
frozen parent PE prior to child PE"). There are 2 ways to report
EEH errors: proactively polling because of 0xFF's returned from
PCI config or IO read, or interrupt driven event. We missed to
report and handle parent frozen PE prior to child frozen PE for
the later case on PowerNV platform.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The names of PCI reset scopes aren't sychronized with firmware.
The patch fixes it.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The frozen state on one specific PE is probably caused by error
injection, which is done with help of PAPR error injection registers.
According to the hardware spec, those registers should be cleared
automatically after one-shot frozen PE. However, that's not always
true, at least on P7IOC of Firebird-L. So we have to clear them
before doing PE reset to avoid recursive EEH errors at recovery
stage.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The patch adds debugfs file (/sys/kernel/debug/powerpc/PCIxxxx/
err_injct), which accepts following formated string, to support
error injection. It will be used to support userland utility
"errinjct" in future.
"pe_no:0:function:address:mask" - 32-bits PCI errors
"pe_no:1:function:address:mask" - 64-bits PCI errors
Signed-off-by: Mike Qiu <qiudayu@linux.vnet.ibm.com>
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The patch introduces eeh_ops::err_inject(), which allows to inject
specified errors to indicated PE for testing purpose. The functionality
isn't support on pSeries platform. On PowerNV, the functionality
relies on OPAL API opal_pci_err_inject().
Signed-off-by: Mike Qiu <qiudayu@linux.vnet.ibm.com>
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The patch adds one more option (EEH_OPT_FREEZE_PE) to set_option()
method to proactively freeze PE, which will be issued before resetting
pass-throughed PE to drop MMIO access during reset because it's
always contributing to recursive EEH error.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
No need for 3 functions when a single one will do.
Modify the function declaring macros to call the single function.
Reduces object code size a little:
$ size arch/powerpc/platforms/powernv/pci-ioda.o*
text data bss dec hex filename
22303 1073 6680 30056 7568 arch/powerpc/platforms/powernv/pci-ioda.o.new
22840 1121 6776 30737 7811 arch/powerpc/platforms/powernv/pci-ioda.o.old
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The return value is unnecessary and unused, so make the functions
void instead of int.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
On PowerNV platforms, when a CPU is offline, we put it into nap mode.
It's possible that the CPU wakes up from nap mode while it is still
offline due to a stray IPI. A misdirected device interrupt could also
potentially cause it to wake up. In that circumstance, we need to clear
the interrupt so that the CPU can go back to nap mode.
In the past the clearing of the interrupt was accomplished by briefly
enabling interrupts and allowing the normal interrupt handling code
(do_IRQ() etc.) to handle the interrupt. This has the problem that
this code calls irq_enter() and irq_exit(), which call functions such
as account_system_vtime() which use RCU internally. Use of RCU is not
permitted on offline CPUs and will trigger errors if RCU checking is
enabled.
To avoid calling into any generic code which might use RCU, we adopt
a different method of clearing interrupts on offline CPUs. Since we
are on the PowerNV platform, we know that the system interrupt
controller is a XICS being driven directly (i.e. not via hcalls) by
the kernel. Hence this adds a new icp_native_flush_interrupt()
function to the native-mode XICS driver and arranges to call that
when an offline CPU is woken from nap. This new function reads the
interrupt from the XICS. If it is an IPI, it clears the IPI; if it
is a device interrupt, it prints a warning and disables the source.
Then it does the end-of-interrupt processing for the interrupt.
The other thing that briefly enabling interrupts did was to check and
clear the irq_happened flag in this CPU's PACA. Therefore, after
flushing the interrupt from the XICS, we also clear all bits except
the PACA_IRQ_HARD_DIS (interrupts are hard disabled) bit from the
irq_happened flag. The PACA_IRQ_HARD_DIS flag is set by power7_nap()
and is left set to indicate that interrupts are hard disabled. This
means we then have to ignore that flag in power7_nap(), which is
reasonable since it doesn't indicate that any interrupt event needs
servicing.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
CONFIG_JUMP_LABEL doesn't ensure HAVE_JUMP_LABEL, if it
is not the case use maintainers's own mutex to guard
the modification of global values.
Signed-off-by: Zhouyi Zhou <yizhouzhou@ict.ac.cn>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Fix a number of places where global functions were not including
their prototype. This ensures the prototype and the function match.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Check that the OPAL_DUMP_READ token exists before initalising the elog
infrastructure.
This avoids littering the OPAL console with:
"OPAL: Called with bad token 91"
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Check that the OPAL_ELOG_READ token exists before initalising the elog
infrastructure.
This avoids littering the OPAL console with:
"OPAL: Called with bad token 74"
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Check that the OPAL_RTC_READ token exists before we use the OPAL RTC.
Refactors the code a little to merge error paths.
This avoids littering the OPAL console with:
"OPAL: Called with bad token 3".
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Currently there is no way to generically check if an OPAL call exists or not
from the host kernel.
This adds an OPAL call opal_check_token() which tells you if the given token is
present in OPAL or not.
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Presently we only support initiating Service Processor dump from host.
Hence update sysfs message. Also update couple of other error/info
messages.
Signed-off-by: Vasant Hegde <hegdevasant@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
As opal_message_init() uses machine_early_initcall(powernv, ), and
opal_hmi_handler_init() depends on that early initcall, so it also needs
use machine_* to check the machine_id.
Signed-off-by: Li Zhong <zhong@linux.vnet.ibm.com>
Pull more powerpc updates from Ben Herrenschmidt:
"Here are some more powerpc bits for 3.17, essentially fixes.
The biggest series, also aimed at -stable, is from Aneesh and is the
result of weeks and weeks of debugging to find out why the heck or THP
implementation was occasionally triggering multi-hit errors in our
level 1 TLB. It ended up being a combination of issues including
subtleties as to how we should invalidate those special 'MPSS' pages
we use to allow the use of 16M pages inside 4K/64K "base page size"
segments (you really have to love our MMU !)
Another interesting one in the "OMG" category is the series from
Michael adding memory barriers to spin_is_locked(). That's also the
result of many days of debugging to figure out why the semaphore code
would occasionally crash in ways that made no sense. It ended up
being some creative lock stacking that was defeated by the fact that
our locks allow a load inside the locked section to be re-ordered with
the load of the lock value itself (I'm still of two mind about whether
to kill that once and for all by putting a heavier barrier back into
our lock implementation...). The fixes come with a long explanation
in the cset comments, feel free to read it if you feel like having a
headache today"
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (25 commits)
powerpc/thp: Add tracepoints to track hugepage invalidate
powerpc/mm: Use read barrier when creating real_pte
powerpc/thp: Use ACCESS_ONCE when loading pmdp
powerpc/thp: Invalidate with vpn in loop
powerpc/thp: Handle combo pages in invalidate
powerpc/thp: Invalidate old 64K based hash page mapping before insert of 4k pte
powerpc/thp: Don't recompute vsid and ssize in loop on invalidate
powerpc/thp: Add write barrier after updating the valid bit
powerpc: reorder per-cpu NUMA information's initialization
powerpc/perf/hv-24x7: Use kmem_cache_free
powerpc/pseries/hvcserver: Fix endian issue in hvcs_get_partner_info
powerpc: Hard disable interrupts in xmon
powerpc: remove duplicate definition of TEXASR_FS
powerpc/pseries: Avoid deadlock on removing ddw
powerpc/pseries: Failure on removing device node
powerpc/boot: Use correct zlib types for comparison
powerpc/powernv: Interface to register/unregister opal dump region
printk: Add function to return log buffer address and size
powerpc: Add POWER8 features to CPU_FTRS_POSSIBLE/ALWAYS
powerpc/ppc476: Disable BTAC
...
PowerNV platform is capable of capturing host memory region when system
crashes (because of host/firmware). We have new OPAL API to register/
unregister memory region to be captured when system crashes.
This patch adds support for new API. Also during boot time we register
kernel log buffer and unregister before doing kexec.
Signed-off-by: Vasant Hegde <hegdevasant@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When we take full hotplug to recover from EEH errors, PCI buses
could be involved. For the case, the child devices of involved
PCI buses can't be attached to IOMMU group properly, which is
caused by commit 3f28c5a ("powerpc/powernv: Reduce multi-hit of
iommu_add_device()").
When adding the PCI devices of the newly created PCI buses to
the system, the IOMMU group is expected to be added in (C).
(A) fails to bind the IOMMU group because bus->is_added is
false. (B) fails because the device doesn't have binding IOMMU
table yet. bus->is_added is set to true at end of (C) and
pdev->is_added is set to true at (D).
pcibios_add_pci_devices()
pci_scan_bridge()
pci_scan_child_bus()
pci_scan_slot()
pci_scan_single_device()
pci_scan_device()
pci_device_add()
pcibios_add_device() A: Ignore
device_add() B: Ignore
pcibios_fixup_bus()
pcibios_setup_bus_devices()
pcibios_setup_device() C: Hit
pcibios_finish_adding_to_bus()
pci_bus_add_devices()
pci_bus_add_device() D: Add device
If the parent PCI bus isn't involved in hotplug, the IOMMU
group is expected to be bound in (B). (A) should fail as the
sysfs entries aren't populated.
The patch fixes the issue by reverting commit 3f28c5a and remove
WARN_ON() in iommu_add_device() to allow calling the function
even the specified device already has associated IOMMU group.
Cc: <stable@vger.kernel.org> # 3.16+
Reported-by: Thadeu Lima de Souza Cascardo <cascardo@linux.vnet.ibm.com>
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Acked-by: Wei Yang <weiyang@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When we hit the HMI in Linux, invoke opal call to handle/recover from HMI
errors in real mode and then in virtual mode during check_irq_replay()
invoke opal_poll_events()/opal_do_notifier() to retrieve HMI event from
OPAL and act accordingly.
Now that we are ready to handle HMI interrupt directly in linux, remove
the HMI interrupt registration with firmware.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Handle Hypervisor Maintenance Interrupt (HMI) in Linux. This patch implements
basic infrastructure to handle HMI in Linux host. The design is to invoke
opal handle hmi in real mode for recovery and set irq_pending when we hit HMI.
During check_irq_replay pull opal hmi event and print hmi info on console.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The PCI config accessors check for PE frozen state and clear it if
EEH isn't functional. The patch handles compound PE in config accessors
if PHB supports it. For consistency, all PEs will be put into frozen
state if any one in compound group gets frozen by hardware.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch handles compound PE for EEH backend. If one specific
PE in compound group has been frozen, we enforces to freeze
all PEs in the group. If we're enable DMA or MMIO for one PE
in compound group, DMA or MMIO of all PEs in the group will be
enabled.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch introduces 3 PHB callbacks: compound PE state retrieval,
force freezing and unfreezing compound PE. The PCI config accessors
and PowerNV EEH backend can use them in subsequent patches.
We don't export the capability of compound PE to EEH core, which
helps avoiding more complexity to EEH core.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Function ioda_eeh_get_state() is used to fetch EEH state for PHB
or PE. We're going to support compound PE and the function becomes
more complicated with that. The patch splits the function into two
functions for PHB and PE cases separately to improve readability.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch synchronizes header file with firmware to have new OPAL
API opal_pci_eeh_freeze_set(), which is used to freeze the specified
PE in order to support "compound" PE.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This patch enables M64 aperatus for PHB3.
We already had platform hook (ppc_md.pcibios_window_alignment) to affect
the PCI resource assignment done in PCI core so that each PE's M32 resource
was built on basis of M32 segment size. Similarly, we're using that for
M64 assignment on basis of M64 segment size.
* We're using last M64 BAR to cover M64 aperatus, and it's shared by all
256 PEs.
* We don't support P7IOC yet. However, some function callbacks are added
to (struct pnv_phb) so that we can reuse them on P7IOC in future.
* PE, corresponding to PCI bus with large M64 BAR device attached, might
span multiple M64 segments. We introduce "compound" PE to cover the case.
The compound PE is a list of PEs and the master PE is used as before.
The slave PEs are just for MMIO isolation.
Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch allows PE (struct eeh_pe) instance to have auxillary data,
whose size is configurable on basis of platform. For PowerNV, the
auxillary data will be used to cache PHB diag-data for that PE
(frozen PE or fenced PHB). In turn, we can retrieve the diag-data
at any later points.
It's useful for the case of VFIO PCI devices where the error log
should be cached, and then be retrieved by the guest at later point.
Also, it can avoid PHB diag-data overwritting if another frozen PE
reported and the previous diag-data isn't fetched by guest.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
pr_warn() is equal to pr_warning(), but the former is a bit more
formal according to commit fc62f2f ("kernel.h: add pr_warn for
symmetry to dev_warn, netdev_warn").
The patch replaces pr_warning() with pr_warn().
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
According to the experiment I did, PCI config access is blocked
on P7IOC frozen PE by hardware, but PHB3 doesn't do that. That
means we always get 0xFF's while dumping PCI config space of the
frozen PE on P7IOC. We don't have the problem on PHB3. So we have
to enable I/O prioir to collecting error log. Otherwise, meaningless
0xFF's are always returned.
The patch fixes it by EEH flag (EEH_ENABLE_IO_FOR_LOG), which is
selectively set to indicate the case for: P7IOC on PowerNV platform,
pSeries platform.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
There are multiple global EEH flags. Almost each flag has its own
accessor, which doesn't make sense. The patch refactors EEH flag
accessors so that they look unified:
eeh_add_flag(): Add EEH flag
eeh_clear_flag(): Clear EEH flag
eeh_has_flag(): Check if one specific flag has been set
eeh_enabled(): Check if EEH functionality has been enabled
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
On PHB3, PCI devices can bypass IOMMU for DMA access. If we pass
through one PCI device, whose hose driver ever enable the bypass
mode, pdev->dev.archdata.dma_data.iommu_table_base isn't IOMMU
table. However, EEH needs access the IOMMU table when the device
is owned by guest.
The patch fixes pdev->dev.archdata.dma_data.iommu_table when
passing through the device to guest in pnv_pci_ioda2_set_bypass().
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
powerpc defines various machine-specific routines for handling
pci_set_dma_mask(). The routines for machine "PowerNV" may neglect
to set dev->dma_mask. This could confuse anyone (e.g. drivers) that
consult dev->dma_mask to find the current mask. Set the dma_mask in
the PowerNV leaf routine.
Signed-off-by: Brian W. Hart <hartb@linux.vnet.ibm.com>
CC: <stable@vger.kernel.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The sysfs entries are lost because of commit 2213fb1 ("powerpc/eeh:
Skip eeh sysfs when eeh is disabled"). That commit added condition
to create sysfs entries with EEH_ENABLED, which isn't populated
when trying to create sysfs entries on PowerNV platform during system
boot time. The patch fixes the issue by:
* Reoder EEH initialization functions so that they're same on
PowerNV/pSeries.
* Cache PE's primary bus by PowerNV platform instead of EEH core
to avoid kernel crash caused by the function reorder. Another
benefit with this is to avoid one eeh_probe_mode_dev() in EEH
core.
Signed-off-by: Mike Qiu <qiudayu@linux.vnet.ibm.com>
Acked-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We must not handle EEH error on devices which are passed to somebody
else. Instead, we expect that the frozen device owner detects an EEH
error and recovers from it.
This avoids EEH error handling on passed through devices so the device
owner gets a chance to handle them.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
A lot of the code in platforms/powernv is using non-machine initcalls.
That means if a kernel built with powernv support runs on another
platform, for example pseries, the initcalls will still run.
That is usually OK, because the initcalls will check for something in
the device tree or elsewhere before doing anything, so on other
platforms they will usually just return.
But it's fishy for powernv code to be running on other platforms, so
switch them all to be machine initcalls. If we want any of them to run
on other platforms in future they should move to sysdev.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
