For the same reason that /proc/iomem returns 0's for non-root readers
and acpi tables are root-only, make the 'resource' attribute for region
devices only readable by root. Otherwise we disclose physical address
information.
Fixes: 802f4be6fe ("libnvdimm: Add 'resource' sysfs attribute to regions")
Cc: <stable@vger.kernel.org>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Johannes Thumshirn <jthumshirn@suse.de>
Reported-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
It is useful to be able to know the position of a DIMM in an
interleave-set. Consider the case where the order of the DIMMs changes
causing a namespace to be invalidated because the interleave-set cookie no
longer matches. If the before and after state of each DIMM position is
known this state debugged by the system owner.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This state is already visible by userspace since the BLK region will not
be enabled, and it is otherwise benign as it usually indicates that the
DIMM is not configured.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The pmem driver attaches to both persistent and volatile memory ranges
advertised by the ACPI NFIT. When the region is volatile it is redundant
to spend cycles flushing caches at fsync(). Check if the hosting region
is volatile and do not set dax_write_cache() if it is.
Cc: Jan Kara <jack@suse.cz>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Allow volatile nfit ranges to participate in all the same infrastructure
provided for persistent memory regions. A resulting resulting namespace
device will still be called "pmem", but the parent region type will be
"nd_volatile". This is in preparation for disabling the dax ->flush()
operation in the pmem driver when it is hosted on a volatile range.
Cc: Jan Kara <jack@suse.cz>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The pmem driver assumes if platform firmware describes the memory
devices associated with a persistent memory range and
CONFIG_ARCH_HAS_PMEM_API=y that it has all the mechanism necessary to
flush data to a power-fail safe zone. We warn if the firmware does not
describe memory devices, but we also need to warn if the architecture
does not claim pmem support.
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Now that all callers of the pmem api have been converted to dax helpers that
call back to the pmem driver, we can remove include/linux/pmem.h and
asm/pmem.h.
Cc: <x86@kernel.org>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Toshi Kani <toshi.kani@hpe.com>
Cc: Oliver O'Halloran <oohall@gmail.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The interleave-set-cookie algorithm is extended to incorporate all the
same components that are used to generate an nvdimm unique-id. For
backwards compatibility we still maintain the old v1.1 definition.
Reported-by: Nicholas Moulin <nicholas.w.moulin@intel.com>
Reported-by: Kaushik Kanetkar <kaushik.a.kanetkar@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The pmem driver has a need to transfer data with a persistent memory
destination and be able to rely on the fact that the destination writes are not
cached. It is sufficient for the writes to be flushed to a cpu-store-buffer
(non-temporal / "movnt" in x86 terms), as we expect userspace to call fsync()
to ensure data-writes have reached a power-fail-safe zone in the platform. The
fsync() triggers a REQ_FUA or REQ_FLUSH to the pmem driver which will turn
around and fence previous writes with an "sfence".
Implement a __copy_from_user_inatomic_flushcache, memcpy_page_flushcache, and
memcpy_flushcache, that guarantee that the destination buffer is not dirty in
the cpu cache on completion. The new copy_from_iter_flushcache and sub-routines
will be used to replace the "pmem api" (include/linux/pmem.h +
arch/x86/include/asm/pmem.h). The availability of copy_from_iter_flushcache()
and memcpy_flushcache() are gated by the CONFIG_ARCH_HAS_UACCESS_FLUSHCACHE
config symbol, and fallback to copy_from_iter_nocache() and plain memcpy()
otherwise.
This is meant to satisfy the concern from Linus that if a driver wants to do
something beyond the normal nocache semantics it should be something private to
that driver [1], and Al's concern that anything uaccess related belongs with
the rest of the uaccess code [2].
The first consumer of this interface is a new 'copy_from_iter' dax operation so
that pmem can inject cache maintenance operations without imposing this
overhead on other dax-capable drivers.
[1]: https://lists.01.org/pipermail/linux-nvdimm/2017-January/008364.html
[2]: https://lists.01.org/pipermail/linux-nvdimm/2017-April/009942.html
Cc: <x86@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Toshi Kani <toshi.kani@hpe.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This is a preparation patch for handling locked nvdimm label regions, a
new concept as introduced by the latest DSM document on pmem.io [1]. A
future patch will leverage nvdimm_set_locked() at DIMM probe time to
flag regions that can not be enabled. There should be no functional
difference resulting from this change.
[1]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example-V1.3.pdf
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Toshi noticed that the new support for a region-level badblocks missed
the case where errors are cleared due to BTT I/O.
An initial attempt to fix this ran into a "sleeping while atomic"
warning due to taking the nvdimm_bus_lock() in the BTT I/O path to
satisfy the locking requirements of __nvdimm_bus_badblocks_clear().
However, that lock is not needed since we are not acting on any data that
is subject to change under that lock. The badblocks instance has its own
internal lock to handle mutations of the error list.
So, in order to make it clear that we are just acting on region devices,
rename __nvdimm_bus_badblocks_clear() to nvdimm_clear_badblocks_regions().
Eliminate the lock and consolidate all support routines for the new
nvdimm_account_cleared_poison() in drivers/nvdimm/bus.c. Finally, to the
opportunity to cleanup to some unnecessary casts, make the calling
convention of nvdimm_clear_badblocks_regions() clearer by replacing struct
resource with the minimal struct clear_badblocks_context, and use the
DEVICE_ATTR macro.
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Vishal Verma <vishal.l.verma@intel.com>
Reported-by: Toshi Kani <toshi.kani@hpe.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The nvdimm_flush() mechanism helps to reduce the impact of an ADR
(asynchronous-dimm-refresh) failure. The ADR mechanism handles flushing
platform WPQ (write-pending-queue) buffers when power is removed. The
nvdimm_flush() mechanism performs that same function on-demand.
When a pmem namespace is associated with a block device, an
nvdimm_flush() is triggered with every block-layer REQ_FUA, or REQ_FLUSH
request. These requests are typically associated with filesystem
metadata updates. However, when a namespace is in device-dax mode,
userspace (think database metadata) needs another path to perform the
same flushing. In other words this is not required to make data
persistent, but in the case of metadata it allows for a smaller failure
domain in the unlikely event of an ADR failure.
The new 'deep_flush' attribute is visible when the individual DIMMs
backing a given interleave-set are described by platform firmware. In
ACPI terms this is "NVDIMM Region Mapping Structures" and associated
"Flush Hint Address Structures". Reads return "1" if the region supports
triggering WPQ flushes on all DIMMs. Reads return "0" the flush
operation is a platform nop, and in that case the attribute is
read-only.
Why sysfs and not an ioctl? An ioctl requires establishing a new
ioctl function number space for device-dax. Given that this would be
called on a device-dax fd an application could be forgiven for
accidentally calling this on a filesystem-dax fd. Placing this interface
in libnvdimm sysfs removes that potential for collision with a
filesystem ioctl, and it keeps ioctls out of the generic device-dax
implementation.
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In the case where a dimm does not have any associated flush hints the
ndrd->flush_wpq array may be uninitialized leading to crashes with the
following signature:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000010
IP: region_visible+0x10f/0x160 [libnvdimm]
Call Trace:
internal_create_group+0xbe/0x2f0
sysfs_create_groups+0x40/0x80
device_add+0x2d8/0x650
nd_async_device_register+0x12/0x40 [libnvdimm]
async_run_entry_fn+0x39/0x170
process_one_work+0x212/0x6c0
? process_one_work+0x197/0x6c0
worker_thread+0x4e/0x4a0
kthread+0x10c/0x140
? process_one_work+0x6c0/0x6c0
? kthread_create_on_node+0x60/0x60
ret_from_fork+0x31/0x40
Cc: <stable@vger.kernel.org>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Fixes: f284a4f237 ("libnvdimm: introduce nvdimm_flush() and nvdimm_has_flush()")
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Adding sysfs attribute in order to export the physical address of the
region. This is for supporting of user app poison clear via
ND_IOCTL_CLEAR_ERROR.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
badblocks sysfs file will be export at region level. When nvdimm event
notifier happens for NVDIMM_REVALIATE_POISON, the badblocks in the
region will be updated.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The interleave-set cookie is a sum that sanity checks the composition of
an interleave set has not changed from when the namespace was initially
created. The checksum is calculated by sorting the DIMMs by their
location in the interleave-set. The comparison for the sort must be
64-bit wide, not byte-by-byte as performed by memcmp() in the broken
case.
Fix the implementation to accept correct cookie values in addition to
the Linux "memcmp" order cookies, but only allow correct cookies to be
generated going forward. It does mean that namespaces created by
third-party-tooling, or created by newer kernels with this fix, will not
validate on older kernels. However, there are a couple mitigating
conditions:
1/ platforms with namespace-label capable NVDIMMs are not widely
available.
2/ interleave-sets with a single-dimm are by definition not affected
(nothing to sort). This covers the QEMU-KVM NVDIMM emulation case.
The cookie stored in the namespace label will be fixed by any write the
namespace label, the most straightforward way to achieve this is to
write to the "alt_name" attribute of a namespace in sysfs.
Cc: <stable@vger.kernel.org>
Fixes: eaf961536e ("libnvdimm, nfit: add interleave-set state-tracking infrastructure")
Reported-by: Nicholas Moulin <nicholas.w.moulin@linux.intel.com>
Tested-by: Nicholas Moulin <nicholas.w.moulin@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
For warnings that should only ever trigger during development and
testing replace WARN statements with lockdep_assert_held. The lockdep
pattern is prevalent, and these paths are are well covered by libnvdimm
unit tests.
Reported-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Similar to BLK regions, publish new seed namespace devices to allow
unused PMEM region capacity to be consumed by additional namespaces.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The free dpa (dimm-physical-address) space calculation reports how much
free space is available with consideration for aliased BLK + PMEM
regions. Recall that BLK capacity is allocated from high addresses and
PMEM is allocated from low addresses in their respective regions.
nd_region_available_dpa() accounts for the fact that the largest
encroachment (lowest starting address) into PMEM capacity by a BLK
allocation limits the available capacity to that point, regardless if
there is BLK allocation hole at a higher address. Similarly, for the
multi-pmem case we need to track the largest encroachment (highest
ending address) of a PMEM allocation in BLK capacity regardless of
whether there is an allocation hole that a BLK allocation could fill at
a lower address.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In preparation for enabling multiple namespaces per pmem region, convert
the label tracking to use a linked list. In particular this will allow
select_pmem_id() to move labels from the unvalidated state to the
validated state. Currently we only track one validated set per-region.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Before we add more libnvdimm-private fields to nd_mapping make it clear
which parameters are input vs libnvdimm internals. Use struct
nd_mapping_desc instead of struct nd_mapping in nd_region_desc and make
struct nd_mapping private to libnvdimm.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Existing implemenetation writes to all the flush hint addresses for a
given ND region. This is not necessary as the flushes are per imc and
not per DIMM. Search the mappings and clear out the duplicates at init
to avoid multiple flush to the same imc.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The definition of the flush hint table as:
void __iomem *flush_wpq[0][0];
...passed the unit test, but is broken as flush_wpq[0][1] and
flush_wpq[1][0] refer to the same entry. Fix this to use a helper that
calculates a slot in the table based on the geometry of flush hints in
the region. This is important to get right since virtualization
solutions use this mechanism to trigger hypervisor flushes to platform
persistence.
Reported-by: Dave Jiang <dave.jiang@intel.com>
Tested-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
nd_activate_region() iomaps any hint addresses required when activating
a region. To prevent duplicate mappings it checks the PFN of the hint to
be mapped against the PFNs of the already mapped hints. Unfortunately it
doesn't convert the PFN back into a physical address before passing it
to devm_nvdimm_ioremap(). Instead it applies PHYS_PFN a second time
which ends about as well as you would imagine.
Signed-off-by: Oliver O'Halloran <oohall@gmail.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
When the NFIT provides multiple flush hint addresses per-dimm it is
expressing that the platform is capable of processing multiple flush
requests in parallel. There is some fixed cost per flush request, let
the cost be shared in parallel on multiple cpus.
Since there may not be enough flush hint addresses for each cpu to have
one, keep a per-cpu index of the last used hint, hash it with current
pid, and assume that access pattern and scheduler randomness will keep
the flush-hint usage somewhat staggered across cpus.
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
nvdimm_flush() is a replacement for the x86 'pcommit' instruction. It is
an optional write flushing mechanism that an nvdimm bus can provide for
the pmem driver to consume. In the case of the NFIT nvdimm-bus-provider
nvdimm_flush() is implemented as a series of flush-hint-address [1]
writes to each dimm in the interleave set (region) that backs the
namespace.
The nvdimm_has_flush() routine relies on platform firmware to describe
the flushing capabilities of a platform. It uses the heuristic of
whether an nvdimm bus provider provides flush address data to return a
ternary result:
1: flush addresses defined
0: dimm topology described without flush addresses (assume ADR)
-errno: no topology information, unable to determine flush mechanism
The pmem driver is expected to take the following actions on this ternary
result:
1: nvdimm_flush() in response to REQ_FUA / REQ_FLUSH and shutdown
0: do not set, WC or FUA on the queue, take no further action
-errno: warn and then operate as if nvdimm_has_flush() returned '0'
The caveat of this heuristic is that it can not distinguish the "dimm
does not have flush address" case from the "platform firmware is broken
and failed to describe a flush address". Given we are already
explicitly trusting the NFIT there's not much more we can do beyond
blacklisting broken firmwares if they are ever encountered.
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In preparation for triggering flushes of a DIMM's writes-posted-queue
(WPQ) via the pmem driver move mapping of flush hint addresses to the
region driver. Since this uses devm_nvdimm_memremap() the flush
addresses will remain mapped while any region to which the dimm belongs
is active.
We need to communicate more information to the nvdimm core to facilitate
this mapping, namely each dimm object now carries an array of flush hint
address resources.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Now that all shared mappings are handled by devm_nvdimm_memremap() we no
longer need nfit_spa_map() nor do we need to trigger a callback to the
bus provider at region disable time.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
ida instances allocate some internal memory for ->free_bitmap in
addition to the base 'struct ida'. Use ida_destroy() to release that
memory at module_exit().
Reported-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Device DAX is the device-centric analogue of Filesystem DAX
(CONFIG_FS_DAX). It allows persistent memory ranges to be allocated and
mapped without need of an intervening file system. This initial
infrastructure arranges for a libnvdimm pfn-device to be represented as
a different device-type so that it can be attached to a driver other
than the pmem driver.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
When btt devices were re-worked to be child devices of regions this
routine was overlooked. It mistakenly attempts to_nd_namespace_pmem()
or to_nd_namespace_blk() conversions on btt and pfn devices. By luck to
date we have happened to be hitting valid memory leading to a uuid
miscompare, but a recent change to struct nd_namespace_common causes:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000001
IP: [<ffffffff814610dc>] memcmp+0xc/0x40
[..]
Call Trace:
[<ffffffffa0028631>] is_uuid_busy+0xc1/0x2a0 [libnvdimm]
[<ffffffffa0028570>] ? to_nd_blk_region+0x50/0x50 [libnvdimm]
[<ffffffff8158c9c0>] device_for_each_child+0x50/0x90
Cc: <stable@vger.kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This simple change hides pfn_seed attribute for non pmem
regions because they don't support pfn anyway.
Signed-off-by: Dmitry V. Krivenok <krivenok.dmitry@gmail.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The expectation is that the legacy / non-standard pmem discovery method
(e820 type-12) will only ever be used to describe small quantities of
persistent memory. Larger capacities will be described via the ACPI
NFIT. When "allocate struct page from pmem" support is added this default
policy can be overridden by assigning a legacy pmem namespace to a pfn
device, however this would be only be necessary if a platform used the
legacy mechanism to define a very large range.
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Implement the base infrastructure for libnvdimm PFN devices. Similar to
BTT devices they take a namespace as a backing device and layer
functionality on top. In this case the functionality is reserving space
for an array of 'struct page' entries to be handed out through
pfn_to_page(). For now this is just the basic libnvdimm-device-model for
configuring the base PFN device.
As the namespace claiming mechanism for PFN devices is mostly identical
to BTT devices drivers/nvdimm/claim.c is created to house the common
bits.
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
A new BLK namespace "seed" device is created whenever the current seed
is successfully probed. However, if that namespace is assigned to a BTT
it may never directly experience a successful probe as it is a
subordinate device to a BTT configuration.
The effect of the current code is that no new namespaces can be
instantiated, after the seed namespace, to consume available BLK DPA
capacity. Fix this by treating a successful BTT probe event as a
successful probe event for the backing namespace.
Reported-by: Nicholas Moulin <nicholas.w.moulin@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
ACPI NFIT table has System Physical Address Range Structure entries that
describe a proximity ID of each range when ACPI_NFIT_PROXIMITY_VALID is
set in the flags.
Change acpi_nfit_register_region() to map a proximity ID to its node ID,
and set it to a new numa_node field of nd_region_desc, which is then
conveyed to the nd_region device.
The device core arranges for btt and namespace devices to inherit their
node from their parent region.
Signed-off-by: Toshi Kani <toshi.kani@hp.com>
[djbw: move set_dev_node() from region.c to bus.c]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Upon detection of an unarmed dimm in a region, arrange for descendant
BTT, PMEM, or BLK instances to be read-only. A dimm is primarily marked
"unarmed" via flags passed by platform firmware (NFIT).
The flags in the NFIT memory device sub-structure indicate the state of
the data on the nvdimm relative to its energy source or last "flush to
persistence". For the most part there is nothing the driver can do but
advertise the state of these flags in sysfs and emit a message if
firmware indicates that the contents of the device may be corrupted.
However, for the case of ACPI_NFIT_MEM_ARMED, the driver can arrange for
the block devices incorporating that nvdimm to be marked read-only.
This is a safe default as the data is still available and new writes are
held off until the administrator either forces read-write mode, or the
energy source becomes armed.
A 'read_only' attribute is added to REGION devices to allow for
overriding the default read-only policy of all descendant block devices.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The libnvdimm implementation handles allocating dimm address space (DPA)
between PMEM and BLK mode interfaces. After DPA has been allocated from
a BLK-region to a BLK-namespace the nd_blk driver attaches to handle I/O
as a struct bio based block device. Unlike PMEM, BLK is required to
handle platform specific details like mmio register formats and memory
controller interleave. For this reason the libnvdimm generic nd_blk
driver calls back into the bus provider to carry out the I/O.
This initial implementation handles the BLK interface defined by the
ACPI 6 NFIT [1] and the NVDIMM DSM Interface Example [2] composed from
DCR (dimm control region), BDW (block data window), IDT (interleave
descriptor) NFIT structures and the hardware register format.
[1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf
[2]: http://pmem.io/documents/NVDIMM_DSM_Interface_Example.pdf
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Boaz Harrosh <boaz@plexistor.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jens Axboe <axboe@fb.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
BTT stands for Block Translation Table, and is a way to provide power
fail sector atomicity semantics for block devices that have the ability
to perform byte granularity IO. It relies on the capability of libnvdimm
namespace devices to do byte aligned IO.
The BTT works as a stacked blocked device, and reserves a chunk of space
from the backing device for its accounting metadata. It is a bio-based
driver because all IO is done synchronously, and there is no queuing or
asynchronous completions at either the device or the driver level.
The BTT uses 'lanes' to index into various 'on-disk' data structures,
and lanes also act as a synchronization mechanism in case there are more
CPUs than available lanes. We did a comparison between two lane lock
strategies - first where we kept an atomic counter around that tracked
which was the last lane that was used, and 'our' lane was determined by
atomically incrementing that. That way, for the nr_cpus > nr_lanes case,
theoretically, no CPU would be blocked waiting for a lane. The other
strategy was to use the cpu number we're scheduled on to and hash it to
a lane number. Theoretically, this could block an IO that could've
otherwise run using a different, free lane. But some fio workloads
showed that the direct cpu -> lane hash performed faster than tracking
'last lane' - my reasoning is the cache thrash caused by moving the
atomic variable made that approach slower than simply waiting out the
in-progress IO. This supports the conclusion that the driver can be a
very simple bio-based one that does synchronous IOs instead of queuing.
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Boaz Harrosh <boaz@plexistor.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jens Axboe <axboe@fb.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Neil Brown <neilb@suse.de>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Greg KH <gregkh@linuxfoundation.org>
[jmoyer: fix nmi watchdog timeout in btt_map_init]
[jmoyer: move btt initialization to module load path]
[jmoyer: fix memory leak in the btt initialization path]
[jmoyer: Don't overwrite corrupted arenas]
Signed-off-by: Vishal Verma <vishal.l.verma@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
NVDIMM namespaces, in addition to accepting "struct bio" based requests,
also have the capability to perform byte-aligned accesses. By default
only the bio/block interface is used. However, if another driver can
make effective use of the byte-aligned capability it can claim namespace
interface and use the byte-aligned ->rw_bytes() interface.
The BTT driver is the initial first consumer of this mechanism to allow
adding atomic sector update semantics to a pmem or blk namespace. This
patch is the sysfs infrastructure to allow configuring a BTT instance
for a namespace. Enabling that BTT and performing i/o is in a
subsequent patch.
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
A blk label set describes a namespace comprised of one or more
discontiguous dpa ranges on a single dimm. They may alias with one or
more pmem interleave sets that include the given dimm.
This is the runtime/volatile configuration infrastructure for sysfs
manipulation of 'alt_name', 'uuid', 'size', and 'sector_size'. A later
patch will make these settings persistent by writing back the label(s).
Unlike pmem namespaces, multiple blk namespaces can be created per
region. Once a blk namespace has been created a new seed device
(unconfigured child of a parent blk region) is instantiated. As long as
a region has 'available_size' != 0 new child namespaces may be created.
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Neil Brown <neilb@suse.de>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
A complete label set is a PMEM-label per-dimm per-interleave-set where
all the UUIDs match and the interleave set cookie matches the hosting
interleave set.
Present sysfs attributes for manipulation of a PMEM-namespace's
'alt_name', 'uuid', and 'size' attributes. A later patch will make
these settings persistent by writing back the label.
Note that PMEM allocations grow forwards from the start of an interleave
set (lowest dimm-physical-address (DPA)). BLK-namespaces that alias
with a PMEM interleave set will grow allocations backward from the
highest DPA.
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Neil Brown <neilb@suse.de>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
On platforms that have firmware support for reading/writing per-dimm
label space, a portion of the dimm may be accessible via an interleave
set PMEM mapping in addition to the dimm's BLK (block-data-window
aperture(s)) interface. A label, stored in a "configuration data
region" on the dimm, disambiguates which dimm addresses are accessed
through which exclusive interface.
Add infrastructure that allows the kernel to block modifications to a
label in the set while any member dimm is active. Note that this is
meant only for enforcing "no modifications of active labels" via the
coarse ioctl command. Adding/deleting namespaces from an active
interleave set is always possible via sysfs.
Another aspect of tracking interleave sets is tracking their integrity
when DIMMs in a set are physically re-ordered. For this purpose we
generate an "interleave-set cookie" that can be recorded in a label and
validated against the current configuration. It is the bus provider
implementation's responsibility to calculate the interleave set cookie
and attach it to a given region.
Cc: Neil Brown <neilb@suse.de>
Cc: <linux-acpi@vger.kernel.org>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Robert Moore <robert.moore@intel.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The libnvdimm region driver is an intermediary driver that translates
non-volatile "region"s into "namespace" sub-devices that are surfaced by
persistent memory block-device drivers (PMEM and BLK).
ACPI 6 introduces the concept that a given nvdimm may simultaneously
offer multiple access modes to its media through direct PMEM load/store
access, or windowed BLK mode. Existing nvdimms mostly implement a PMEM
interface, some offer a BLK-like mode, but never both as ACPI 6 defines.
If an nvdimm is single interfaced, then there is no need for dimm
metadata labels. For these devices we can take the region boundaries
directly to create a child namespace device (nd_namespace_io).
Acked-by: Christoph Hellwig <hch@lst.de>
Tested-by: Toshi Kani <toshi.kani@hp.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
A "region" device represents the maximum capacity of a BLK range (mmio
block-data-window(s)), or a PMEM range (DAX-capable persistent memory or
volatile memory), without regard for aliasing. Aliasing, in the
dimm-local address space (DPA), is resolved by metadata on a dimm to
designate which exclusive interface will access the aliased DPA ranges.
Support for the per-dimm metadata/label arrvies is in a subsequent
patch.
The name format of "region" devices is "regionN" where, like dimms, N is
a global ida index assigned at discovery time. This id is not reliable
across reboots nor in the presence of hotplug. Look to attributes of
the region or static id-data of the sub-namespace to generate a
persistent name. However, if the platform configuration does not change
it is reasonable to expect the same region id to be assigned at the next
boot.
"region"s have 2 generic attributes "size", and "mapping"s where:
- size: the BLK accessible capacity or the span of the
system physical address range in the case of PMEM.
- mappingN: a tuple describing a dimm's contribution to the region's
capacity in the format (<nmemX>,<dpa>,<size>). For a PMEM-region
there will be at least one mapping per dimm in the interleave set. For
a BLK-region there is only "mapping0" listing the starting DPA of the
BLK-region and the available DPA capacity of that space (matches "size"
above).
The max number of mappings per "region" is hard coded per the
constraints of sysfs attribute groups. That said the number of mappings
per region should never exceed the maximum number of possible dimms in
the system. If the current number turns out to not be enough then the
"mappings" attribute clarifies how many there are supposed to be. "32
should be enough for anybody...".
Cc: Neil Brown <neilb@suse.de>
Cc: <linux-acpi@vger.kernel.org>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Robert Moore <robert.moore@intel.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: Toshi Kani <toshi.kani@hp.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Dan Williams