Convert string compares of DT node names to use of_node_name_eq helper
instead. This removes direct access to the node name pointer.
A couple of open coded iterating thru the child node names are converted
to use for_each_child_of_node() instead.
Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
In preparation to remove the node name pointer from struct
device_node, convert printf users to use the %pOFn format specifier.
pmem.c was recently added and missed the initial conversion.
Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This comment talks about PTEs being 64-bits and PMD/PGD being 32-bits,
but that hasn't been true since 2005 when David Gibson implemented
4-level page tables in the commit titled "Four level pagetables for
ppc64".
Remove it.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
In update_lmb_associativity_index() we lookup dr_node using
of_find_node_by_path() which takes a reference for us. In the
non-error case we forget to drop the reference. Note that
find_aa_index() does modify properties of the node, but doesn't need
an extra reference held once it's returned.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
POWER9 Witherspoon machines come with 4 or 6 V100 GPUs which are not
pluggable PCIe devices but still have PCIe links which are used
for config space and MMIO. In addition to that the GPUs have 6 NVLinks
which are connected to other GPUs and the POWER9 CPU. POWER9 chips
have a special unit on a die called an NPU which is an NVLink2 host bus
adapter with p2p connections to 2 to 3 GPUs, 3 or 2 NVLinks to each.
These systems also support ATS (address translation services) which is
a part of the NVLink2 protocol. Such GPUs also share on-board RAM
(16GB or 32GB) to the system via the same NVLink2 so a CPU has
cache-coherent access to a GPU RAM.
This exports GPU RAM to the userspace as a new VFIO device region. This
preregisters the new memory as device memory as it might be used for DMA.
This inserts pfns from the fault handler as the GPU memory is not onlined
until the vendor driver is loaded and trained the NVLinks so doing this
earlier causes low level errors which we fence in the firmware so
it does not hurt the host system but still better be avoided; for the same
reason this does not map GPU RAM into the host kernel (usual thing for
emulated access otherwise).
This exports an ATSD (Address Translation Shootdown) register of NPU which
allows TLB invalidations inside GPU for an operating system. The register
conveniently occupies a single 64k page. It is also presented to
the userspace as a new VFIO device region. One NPU has 8 ATSD registers,
each of them can be used for TLB invalidation in a GPU linked to this NPU.
This allocates one ATSD register per an NVLink bridge allowing passing
up to 6 registers. Due to the host firmware bug (just recently fixed),
only 1 ATSD register per NPU was actually advertised to the host system
so this passes that alone register via the first NVLink bridge device in
the group which is still enough as QEMU collects them all back and
presents to the guest via vPHB to mimic the emulated NPU PHB on the host.
In order to provide the userspace with the information about GPU-to-NVLink
connections, this exports an additional capability called "tgt"
(which is an abbreviated host system bus address). The "tgt" property
tells the GPU its own system address and allows the guest driver to
conglomerate the routing information so each GPU knows how to get directly
to the other GPUs.
For ATS to work, the nest MMU (an NVIDIA block in a P9 CPU) needs to
know LPID (a logical partition ID or a KVM guest hardware ID in other
words) and PID (a memory context ID of a userspace process, not to be
confused with a linux pid). This assigns a GPU to LPID in the NPU and
this is why this adds a listener for KVM on an IOMMU group. A PID comes
via NVLink from a GPU and NPU uses a PID wildcard to pass it through.
This requires coherent memory and ATSD to be available on the host as
the GPU vendor only supports configurations with both features enabled
and other configurations are known not to work. Because of this and
because of the ways the features are advertised to the host system
(which is a device tree with very platform specific properties),
this requires enabled POWERNV platform.
The V100 GPUs do not advertise any of these capabilities via the config
space and there are more than just one device ID so this relies on
the platform to tell whether these GPUs have special abilities such as
NVLinks.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
VFIO regions already support region capabilities with a limited set of
fields. However the subdriver might have to report to the userspace
additional bits.
This adds an add_capability() hook to vfio_pci_regops.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
So far we only allowed mapping of MMIO BARs to the userspace. However
there are GPUs with on-board coherent RAM accessible via side
channels which we also want to map to the userspace. The first client
for this is NVIDIA V100 GPU with NVLink2 direct links to a POWER9
NPU-enabled CPU; such GPUs have 16GB RAM which is coherently mapped
to the system address space, we are going to export these as an extra
PCI region.
We already support extra PCI regions and this adds support for mapping
them to the userspace.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When a page fault happens in a GPU, the GPU signals the OS and the GPU
driver calls the fault handler which populated a page table; this allows
the GPU to complete an ATS request.
On the bare metal get_user_pages() is enough as it adds a pte to
the kernel page table but under KVM the partition scope tree does not get
updated so ATS will still fail.
This reads a byte from an effective address which causes HV storage
interrupt and KVM updates the partition scope tree.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
A broken device tree might contain more than 8 values and introduce hard
to debug memory corruption bug. This adds the boundary check.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
In order to make ATS work and translate addresses for arbitrary
LPID and PID, we need to program an NPU with LPID and allow PID wildcard
matching with a specific MSR mask.
This implements a helper to assign a GPU to LPAR and program the NPU
with a wildcard for PID and a helper to do clean-up. The helper takes
MSR (only DR/HV/PR/SF bits are allowed) to program them into NPU2 for
ATS checkout requests support.
This exports pnv_npu2_unmap_lpar_dev() as following patches will use it
from the VFIO driver.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
At the moment the powernv platform registers an IOMMU group for each
PE. There is an exception though: an NVLink bridge which is attached
to the corresponding GPU's IOMMU group making it a master.
Now we have POWER9 systems with GPUs connected to each other directly
bypassing PCI. At the moment we do not control state of these links so
we have to put such interconnected GPUs to one IOMMU group which means
that the old scheme with one GPU as a master won't work - there will
be up to 3 GPUs in such group.
This introduces a npu_comp struct which represents a compound IOMMU
group made of multiple PEs - PCI PEs (for GPUs) and NPU PEs (for
NVLink bridges). This converts the existing NVLink1 code to use the
new scheme. >From now on, each PE must have a valid
iommu_table_group_ops which will either be called directly (for a
single PE group) or indirectly from a compound group handlers.
This moves IOMMU group registration for NVLink-connected GPUs to
npu-dma.c. For POWER8, this stores a new compound group pointer in the
PE (so a GPU is still a master); for POWER9 the new group pointer is
stored in an NPU (which is allocated per a PCI host controller).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[mpe: Initialise npdev to NULL in pnv_try_setup_npu_table_group()]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
At the moment NPU IOMMU is manipulated directly from the IODA2 PCI
PE code; PCI PE acts as a master to NPU PE. Soon we will have compound
IOMMU groups with several PEs from several different PHB (such as
interconnected GPUs and NPUs) so there will be no single master but
a one big IOMMU group.
This makes a first step and converts an NPU PE with a set of extern
function to a table group.
This should cause no behavioral change. Note that
pnv_npu_release_ownership() has never been implemented.
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>
Normal PCI PEs have 2 TVEs, one per a DMA window; however NPU PE has only
one which points to one of two tables of the corresponding PCI PE.
So whenever a new DMA window is programmed to PEs, the NPU PE needs to
release old table in order to use the new one.
Commit d41ce7b1bc ("powerpc/powernv/npu: Do not try invalidating 32bit
table when 64bit table is enabled") did just that but in pci-ioda.c
while it actually belongs to npu-dma.c.
This moves the single TVE handling to npu-dma.c. This does not implement
restoring though as it is highly unlikely that we can set the table to
PCI PE and cannot to NPU PE and if that fails, we could only set 32bit
table to NPU PE and this configuration is not really supported or wanted.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The iommu_table pointer stored in iommu_table_group may get stale
by accident, this adds referencing and removes a redundant comment
about this.
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>
Registering new IOMMU groups and adding devices to them are separated in
code and the latter is dug in the DMA setup code which it does not
really belong to.
This moved IOMMU groups setup to a separate helper which registers a group
and adds devices as before. This does not make a difference as IOMMU
groups are not used anyway; the only dependency here is that
iommu_add_device() requires a valid pointer to an iommu_table
(set by set_iommu_table_base()).
To keep the old behaviour, this does not add new IOMMU groups for PEs
with no DMA weight and also skips NVLink bridges which do not have
pci_controller_ops::setup_bridge (the normal way of adding PEs).
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>
The powernv platform registers IOMMU groups and adds devices to them
from the pci_controller_ops::setup_bridge() hook except one case when
virtual functions (SRIOV VFs) are added from a bus notifier.
The pseries platform registers IOMMU groups from
the pci_controller_ops::dma_bus_setup() hook and adds devices from
the pci_controller_ops::dma_dev_setup() hook. The very same bus notifier
used for powernv does not add devices for pseries though as
__of_scan_bus() adds devices first, then it does the bus/dev DMA setup.
Both platforms use iommu_add_device() which takes a device and expects
it to have a valid IOMMU table struct with an iommu_table_group pointer
which in turn points the iommu_group struct (which represents
an IOMMU group). Although the helper seems easy to use, it relies on
some pre-existing device configuration and associated data structures
which it does not really need.
This simplifies iommu_add_device() to take the table_group pointer
directly. Pseries already has a table_group pointer handy and the bus
notified is not used anyway. For powernv, this copies the existing bus
notifier, makes it work for powernv only which means an easy way of
getting to the table_group pointer. This was tested on VFs but should
also support physical PCI hotplug.
Since iommu_add_device() receives the table_group pointer directly,
pseries does not do TCE cache invalidation (the hypervisor does) nor
allow multiple groups per a VFIO container (in other words sharing
an IOMMU table between partitionable endpoints), this removes
iommu_table_group_link from pseries.
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>
The pci_dma_bus_setup_pSeries and pci_dma_dev_setup_pSeries hooks are
registered for the pseries platform which does not have FW_FEATURE_LPAR;
these would be pre-powernv platforms which we never supported PCI pass
through for anyway so remove it.
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>
We already changed NPU API for GPUs to not to call OPAL and the remaining
bit is initializing NPU structures.
This searches for POWER9 NVLinks attached to any device on a PHB and
initializes an NPU structure if any found.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
We might have memory@ nodes with "linux,usable-memory" set to zero
(for example, to replicate powernv's behaviour for GPU coherent memory)
which means that the memory needs an extra initialization but since
it can be used afterwards, the pseries platform will try mapping it
for DMA so the DMA window needs to cover those memory regions too;
if the window cannot cover new memory regions, the memory onlining fails.
This walks through the memory nodes to find the highest RAM address to
let a huge DMA window cover that too in case this memory gets onlined
later.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When introduced, the NPU context init/destroy helpers called OPAL which
enabled/disabled PID (a userspace memory context ID) filtering in an NPU
per a GPU; this was a requirement for P9 DD1.0. However newer chip
revision added a PID wildcard support so there is no more need to
call OPAL every time a new context is initialized. Also, since the PID
wildcard support was added, skiboot does not clear wildcard entries
in the NPU so these remain in the hardware till the system reboot.
This moves LPID and wildcard programming to the PE setup code which
executes once during the booting process so NPU2 context init/destroy
won't need to do additional configuration.
This replaces the check for FW_FEATURE_OPAL with a check for npu!=NULL as
this is the way to tell if the NPU support is present and configured.
This moves pnv_npu2_init() declaration as pseries should be able to use it.
This keeps pnv_npu2_map_lpar() in powernv as pseries is not allowed to
call that. This exports pnv_npu2_map_lpar_dev() as following patches
will use it from the VFIO driver.
While at it, replace redundant list_for_each_entry_safe() with
a simpler list_for_each_entry().
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The powernv PCI code stores NPU data in the pnv_phb struct. The latter
is referenced by pci_controller::private_data. We are going to have NPU2
support in the pseries platform as well but it does not store any
private_data in in the pci_controller struct; and even if it did,
it would be a different data structure.
This makes npu a pointer and stores it one level higher in
the pci_controller struct.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This new memory does not have page structs as it is not plugged to
the host so gup() will fail anyway.
This adds 2 helpers:
- mm_iommu_newdev() to preregister the "memory device" memory so
the rest of API can still be used;
- mm_iommu_is_devmem() to know if the physical address is one of thise
new regions which we must avoid unpinning of.
This adds @mm to tce_page_is_contained() and iommu_tce_xchg() to test
if the memory is device memory to avoid pfn_to_page().
This adds a check for device memory in mm_iommu_ua_mark_dirty_rm() which
does delayed pages dirtying.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Normally mm_iommu_get() should add a reference and mm_iommu_put() should
remove it. However historically mm_iommu_find() does the referencing and
mm_iommu_get() is doing allocation and referencing.
We are going to add another helper to preregister device memory so
instead of having mm_iommu_new() (which pre-registers the normal memory
and references the region), we need separate helpers for pre-registering
and referencing.
This renames:
- mm_iommu_get to mm_iommu_new;
- mm_iommu_find to mm_iommu_get.
This changes mm_iommu_get() to reference the region so the name now
reflects what it does.
This removes the check for exact match from mm_iommu_new() as we want it
to fail on existing regions; mm_iommu_get() should be used instead.
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>
The skiboot firmware has a hot reset handler which fences the NVIDIA V100
GPU RAM on Witherspoons and makes accesses no-op instead of throwing HMIs:
https://github.com/open-power/skiboot/commit/fca2b2b839a67
Now we are going to pass V100 via VFIO which most certainly involves
KVM guests which are often terminated without getting a chance to offline
GPU RAM so we end up with a running machine with misconfigured memory.
Accessing this memory produces hardware management interrupts (HMI)
which bring the host down.
To suppress HMIs, this wires up this hot reset hook to vfio_pci_disable()
via pci_disable_device() which switches NPU2 to a safe mode and prevents
HMIs.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Alistair Popple <alistair@popple.id.au>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
On the 8xx, no-execute is set via PPP bits in the PTE. Therefore
a no-exec fault generates DSISR_PROTFAULT error bits,
not DSISR_NOEXEC_OR_G.
This patch adds DSISR_PROTFAULT in the test mask.
Fixes: d3ca587404 ("powerpc/mm: Fix reporting of kernel execute faults")
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
System call table generation script must be run to gener-
ate unistd_32/64.h and syscall_table_32/64/c32/spu.h files.
This patch will have changes which will invokes the script.
This patch will generate unistd_32/64.h and syscall_table-
_32/64/c32/spu.h files by the syscall table generation
script invoked by parisc/Makefile and the generated files
against the removed files must be identical.
The generated uapi header file will be included in uapi/-
asm/unistd.h and generated system call table header file
will be included by kernel/systbl.S file.
Signed-off-by: Firoz Khan <firoz.khan@linaro.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The system call tables are in different format in all
architecture and it will be difficult to manually add or
modify the system calls in the respective files. To make
it easy by keeping a script and which will generate the
uapi header and syscall table file. This change will also
help to unify the implementation across all architectures.
The system call table generation script is added in
syscalls directory which contain the script to generate
both uapi header file and system call table files.
The syscall.tbl file will be the input for the scripts.
syscall.tbl contains the list of available system calls
along with system call number and corresponding entry point.
Add a new system call in this architecture will be possible
by adding new entry in the syscall.tbl file.
Adding a new table entry consisting of:
- System call number.
- ABI.
- System call name.
- Entry point name.
- Compat entry name, if required.
syscallhdr.sh and syscalltbl.sh will generate uapi header-
unistd_32/64.h and syscall_table_32/64/c32/spu.h files
respectively. File syscall_table_32/64/c32/spu.h is incl-
uded by syscall.S - the real system call table. Both *.sh
files will parse the content syscall.tbl to generate the
header and table files.
ARM, s390 and x86 architecuture does have similar support.
I leverage their implementation to come up with a generic
solution.
Signed-off-by: Firoz Khan <firoz.khan@linaro.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
PowerPC uses a syscall table with native and compat calls
interleaved, which is a slightly simpler way to define two
matching tables.
As we move to having the tables generated, that advantage
is no longer important, but the interleaved table gets in
the way of using the same scripts as on the other archit-
ectures.
Split out a new compat_sys_call_table symbol that contains
all the compat calls, and leave the main table for the nat-
ive calls, to more closely match the method we use every-
where else.
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Firoz Khan <firoz.khan@linaro.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Move the macro definition for compat_sys_sigsuspend from
asm/systbl.h to the file which it is getting included.
One of the patch in this patch series is generating uapi
header and syscall table files. In order to come up with
a common implimentation across all architecture, we need
to do this change.
This change will simplify the implementation of system
call table generation script and help to come up a common
implementation across all architecture.
Signed-off-by: Firoz Khan <firoz.khan@linaro.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
NR_syscalls macro holds the number of system call exist
in powerpc architecture. We have to change the value of
NR_syscalls, if we add or delete a system call.
One of the patch in this patch series has a script which
will generate a uapi header based on syscall.tbl file.
The syscall.tbl file contains the number of system call
information. So we have two option to update NR_syscalls
value.
1. Update NR_syscalls in asm/unistd.h manually by count-
ing the no.of system calls. No need to update NR_sys-
calls until we either add a new system call or delete
existing system call.
2. We can keep this feature in above mentioned script,
that will count the number of syscalls and keep it in
a generated file. In this case we don't need to expli-
citly update NR_syscalls in asm/unistd.h file.
The 2nd option will be the recommended one. For that, I
added the __NR_syscalls macro in uapi/asm/unistd.h along
with NR_syscalls asm/unistd.h. The macro __NR_syscalls
also added for making the name convention same across all
architecture. While __NR_syscalls isn't strictly part of
the uapi, having it as part of the generated header to
simplifies the implementation. We also need to enclose
this macro with #ifdef __KERNEL__ to avoid side effects.
Signed-off-by: Firoz Khan <firoz.khan@linaro.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Protection key tracking information is not copied over to the
mm_struct of the child during fork(). This can cause the child to
erroneously allocate keys that were already allocated. Any allocated
execute-only key is lost aswell.
Add code; called by dup_mmap(), to copy the pkey state from parent to
child explicitly.
This problem was originally found by Dave Hansen on x86, which turns
out to be a problem on powerpc aswell.
Fixes: cf43d3b264 ("powerpc: Enable pkey subsystem")
Cc: stable@vger.kernel.org # v4.16+
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The AFU Descriptor Template in the PCI config space has a Name Space
field which is a 24 Byte ASCII character string of descriptive name
space for the AFU. The OCXL driver read the string four characters at
a time with pci_read_config_dword().
This optimization is valid on a little-endian system since this is PCI,
but a big-endian system ends up with each subset of four characters in
reverse order.
This could be fixed by switching to read characters one by one. Another
option is to swap the bytes if we're big-endian.
Go for the latter with le32_to_cpu().
Cc: stable@vger.kernel.org # v4.16
Signed-off-by: Greg Kurz <groug@kaod.org>
Acked-by: Frederic Barrat <fbarrat@linux.ibm.com>
Acked-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This is a new test case that creates a signal and starts a suspended
transaction inside the signal handler.
It returns from the signal handler with the CPU at suspended state, but
without setting user context MSR Transaction State (TS) field.
The kernel signal handler code should be able to handle this discrepancy
instead of crashing.
This code could be compiled and used to test 32 and 64-bits signal
handlers.
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Gustavo Romero <gromero@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
There is a TM Bad Thing bug that can be caused when you return from a
signal context in a suspended transaction but with ucontext MSR[TS] unset.
This forces regs->msr[TS] to be set at syscall entrance (since the CPU
state is transactional). It also calls treclaim() to flush the transaction
state, which is done based on the live (mfmsr) MSR state.
Since user context MSR[TS] is not set, then restore_tm_sigcontexts() is not
called, thus, not executing recheckpoint, keeping the CPU state as not
transactional. When calling rfid, SRR1 will have MSR[TS] set, but the CPU
state is non transactional, causing the TM Bad Thing with the following
stack:
[ 33.862316] Bad kernel stack pointer 3fffd9dce3e0 at c00000000000c47c
cpu 0x8: Vector: 700 (Program Check) at [c00000003ff7fd40]
pc: c00000000000c47c: fast_exception_return+0xac/0xb4
lr: 00003fff865f442c
sp: 3fffd9dce3e0
msr: 8000000102a03031
current = 0xc00000041f68b700
paca = 0xc00000000fb84800 softe: 0 irq_happened: 0x01
pid = 1721, comm = tm-signal-sigre
Linux version 4.9.0-3-powerpc64le (debian-kernel@lists.debian.org) (gcc version 6.3.0 20170516 (Debian 6.3.0-18) ) #1 SMP Debian 4.9.30-2+deb9u2 (2017-06-26)
WARNING: exception is not recoverable, can't continue
The same problem happens on 32-bits signal handler, and the fix is very
similar, if tm_recheckpoint() is not executed, then regs->msr[TS] should be
zeroed.
This patch also fixes a sparse warning related to lack of indentation when
CONFIG_PPC_TRANSACTIONAL_MEM is set.
Fixes: 2b0a576d15 ("powerpc: Add new transactional memory state to the signal context")
CC: Stable <stable@vger.kernel.org> # 3.10+
Signed-off-by: Breno Leitao <leitao@debian.org>
Tested-by: Michal Suchánek <msuchanek@suse.de>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Usually a TM Bad Thing exception is raised due to three different problems.
a) touching SPRs in an active transaction; b) using TM instruction with the
facility disabled and c) setting a wrong MSR/SRR1 at RFID.
The two initial cases are easy to identify by looking at the instructions.
The latter case is harder, because the MSR is masked after RFID, so, it is
very useful to look at the previous MSR (SRR1) before RFID as also the
current and masked MSR.
Since MSR is saved at paca just before RFID, this patch prints it if a TM
Bad thing happen, helping to understand what is the invalid TM transition
that is causing the exception.
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
As other exit points, move SRR1 (MSR) into paca->tm_scratch, so, if
there is a TM Bad Thing in RFID, it is easy to understand what was the
SRR1 value being used.
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
On a signal handler return, the user could set a context with MSR[TS] bits
set, and these bits would be copied to task regs->msr.
At restore_tm_sigcontexts(), after current task regs->msr[TS] bits are set,
several __get_user() are called and then a recheckpoint is executed.
This is a problem since a page fault (in kernel space) could happen when
calling __get_user(). If it happens, the process MSR[TS] bits were
already set, but recheckpoint was not executed, and SPRs are still invalid.
The page fault can cause the current process to be de-scheduled, with
MSR[TS] active and without tm_recheckpoint() being called. More
importantly, without TEXASR[FS] bit set also.
Since TEXASR might not have the FS bit set, and when the process is
scheduled back, it will try to reclaim, which will be aborted because of
the CPU is not in the suspended state, and, then, recheckpoint. This
recheckpoint will restore thread->texasr into TEXASR SPR, which might be
zero, hitting a BUG_ON().
kernel BUG at /build/linux-sf3Co9/linux-4.9.30/arch/powerpc/kernel/tm.S:434!
cpu 0xb: Vector: 700 (Program Check) at [c00000041f1576d0]
pc: c000000000054550: restore_gprs+0xb0/0x180
lr: 0000000000000000
sp: c00000041f157950
msr: 8000000100021033
current = 0xc00000041f143000
paca = 0xc00000000fb86300 softe: 0 irq_happened: 0x01
pid = 1021, comm = kworker/11:1
kernel BUG at /build/linux-sf3Co9/linux-4.9.30/arch/powerpc/kernel/tm.S:434!
Linux version 4.9.0-3-powerpc64le (debian-kernel@lists.debian.org) (gcc version 6.3.0 20170516 (Debian 6.3.0-18) ) #1 SMP Debian 4.9.30-2+deb9u2 (2017-06-26)
enter ? for help
[c00000041f157b30] c00000000001bc3c tm_recheckpoint.part.11+0x6c/0xa0
[c00000041f157b70] c00000000001d184 __switch_to+0x1e4/0x4c0
[c00000041f157bd0] c00000000082eeb8 __schedule+0x2f8/0x990
[c00000041f157cb0] c00000000082f598 schedule+0x48/0xc0
[c00000041f157ce0] c0000000000f0d28 worker_thread+0x148/0x610
[c00000041f157d80] c0000000000f96b0 kthread+0x120/0x140
[c00000041f157e30] c00000000000c0e0 ret_from_kernel_thread+0x5c/0x7c
This patch simply delays the MSR[TS] set, so, if there is any page fault in
the __get_user() section, it does not have regs->msr[TS] set, since the TM
structures are still invalid, thus avoiding doing TM operations for
in-kernel exceptions and possible process reschedule.
With this patch, the MSR[TS] will only be set just before recheckpointing
and setting TEXASR[FS] = 1, thus avoiding an interrupt with TM registers in
invalid state.
Other than that, if CONFIG_PREEMPT is set, there might be a preemption just
after setting MSR[TS] and before tm_recheckpoint(), thus, this block must
be atomic from a preemption perspective, thus, calling
preempt_disable/enable() on this code.
It is not possible to move tm_recheckpoint to happen earlier, because it is
required to get the checkpointed registers from userspace, with
__get_user(), thus, the only way to avoid this undesired behavior is
delaying the MSR[TS] set.
The 32-bits signal handler seems to be safe this current issue, but, it
might be exposed to the preemption issue, thus, disabling preemption in
this chunk of code.
Changes from v2:
* Run the critical section with preempt_disable.
Fixes: 87b4e5393a ("powerpc/tm: Fix return of active 64bit signals")
Cc: stable@vger.kernel.org (v3.9+)
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
For fadump to work successfully there should not be any holes in reserved
memory ranges where kernel has asked firmware to move the content of old
kernel memory in event of crash. Now that fadump uses CMA for reserved
area, this memory area is now not protected from hot-remove operations
unless it is cma allocated. Hence, fadump service can fail to re-register
after the hot-remove operation, if hot-removed memory belongs to fadump
reserved region. To avoid this make sure that memory from fadump reserved
area is not hot-removable if fadump is registered.
However, if user still wants to remove that memory, he can do so by
manually stopping fadump service before hot-remove operation.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
fadump fails to register when there are holes in reserved memory area.
This can happen if user has hot-removed a memory that falls in the
fadump reserved memory area. Throw a meaningful error message to the
user in such case.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
[mpe: is_reserved_memory_area_contiguous() returns bool, unsplit string]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
One of the primary issues with Firmware Assisted Dump (fadump) on Power
is that it needs a large amount of memory to be reserved. On large
systems with TeraBytes of memory, this reservation can be quite
significant.
In some cases, fadump fails if the memory reserved is insufficient, or
if the reserved memory was DLPAR hot-removed.
In the normal case, post reboot, the preserved memory is filtered to
extract only relevant areas of interest using the makedumpfile tool.
While the tool provides flexibility to determine what needs to be part
of the dump and what memory to filter out, all supported distributions
default this to "Capture only kernel data and nothing else".
We take advantage of this default and the Linux kernel's Contiguous
Memory Allocator (CMA) to fundamentally change the memory reservation
model for fadump.
Instead of setting aside a significant chunk of memory nobody can use,
this patch uses CMA instead, to reserve a significant chunk of memory
that the kernel is prevented from using (due to MIGRATE_CMA), but
applications are free to use it. With this fadump will still be able
to capture all of the kernel memory and most of the user space memory
except the user pages that were present in CMA region.
Essentially, on a P9 LPAR with 2 cores, 8GB RAM and current upstream:
[root@zzxx-yy10 ~]# free -m
total used free shared buff/cache available
Mem: 7557 193 6822 12 541 6725
Swap: 4095 0 4095
With this patch:
[root@zzxx-yy10 ~]# free -m
total used free shared buff/cache available
Mem: 8133 194 7464 12 475 7338
Swap: 4095 0 4095
Changes made here are completely transparent to how fadump has
traditionally worked.
Thanks to Aneesh Kumar and Anshuman Khandual for helping us understand
CMA and its usage.
TODO:
- Handle case where CMA reservation spans nodes.
Signed-off-by: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
opal_power_control_init() depends on opal message notifier to be
initialized, which is done in opal_init()->opal_message_init(). But both
these initialization are called through machine initcalls and it all
depends on in which order they being called. So far these are called in
correct order (may be we got lucky) and never saw any issue. But it is
clearer to control initialization order explicitly by moving
opal_power_control_init() into opal_init().
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The script "checkpatch.pl" pointed information out like the following.
WARNING: void function return statements are not generally useful
Thus remove such a statement in the affected functions.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Omit an extra message for a memory allocation failure in these
functions.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
A single character (line break) should be put into a sequence.
Thus use the corresponding function "seq_putc".
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Some data were printed into a sequence by four separate function calls.
Print the same data by two single function calls instead.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
CONFIG_EARLY_DEBUG_CPM requires IMMR area TLB to be pinned
otherwise it doesn't survive MMU_init, and the boot fails.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
CONFIG_PCI_MSI was made mandatory by commit a311e738b6
("powerpc/powernv: Make PCI non-optional") so the #ifdef
checks around CONFIG_PCI_MSI here can be removed entirely.
Signed-off-by: Oliver O'Halloran <oohall@gmail.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Fix a spelling mistake in a register description.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Commit 14c63f17b1 ("perf: Drop sample rate when sampling is too
slow") introduced a way to throttle PMU interrupts if we're spending
too much time just processing those. Wire up powerpc PMI handler to
use this infrastructure.
We have throttling of the *rate* of interrupts, but this adds
throttling based on the *time taken* to process the interrupts.
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The current implementation of the OPAL_PCI_EEH_FREEZE_STATUS call in
skiboot's NPU driver does not touch the pci_error_type parameter so
it might have garbage but the powernv code analyzes it nevertheless.
This initializes pcierr and fstate to zero in all call sites.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Sam Bobroff <sbobroff@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
