2015-06-10 02:13:37 +08:00
|
|
|
menuconfig LIBNVDIMM
|
2015-05-20 10:54:31 +08:00
|
|
|
tristate "NVDIMM (Non-Volatile Memory Device) Support"
|
|
|
|
|
depends on PHYS_ADDR_T_64BIT
|
|
|
|
|
depends on BLK_DEV
|
|
|
|
|
help
|
|
|
|
|
Generic support for non-volatile memory devices including
|
|
|
|
|
ACPI-6-NFIT defined resources. On platforms that define an
|
|
|
|
|
NFIT, or otherwise can discover NVDIMM resources, a libnvdimm
|
|
|
|
|
bus is registered to advertise PMEM (persistent memory)
|
|
|
|
|
namespaces (/dev/pmemX) and BLK (sliding mmio window(s))
|
nd_btt: atomic sector updates
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>
2015-06-25 16:20:32 +08:00
|
|
|
namespaces (/dev/ndblkX.Y). A PMEM namespace refers to a
|
|
|
|
|
memory resource that may span multiple DIMMs and support DAX
|
|
|
|
|
(see CONFIG_DAX). A BLK namespace refers to an NVDIMM control
|
|
|
|
|
region which exposes an mmio register set for windowed access
|
|
|
|
|
mode to non-volatile memory.
|
2015-06-10 02:13:37 +08:00
|
|
|
|
|
|
|
|
if LIBNVDIMM
|
|
|
|
|
|
|
|
|
|
config BLK_DEV_PMEM
|
|
|
|
|
tristate "PMEM: Persistent memory block device support"
|
|
|
|
|
default LIBNVDIMM
|
|
|
|
|
depends on HAS_IOMEM
|
nd_btt: atomic sector updates
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>
2015-06-25 16:20:32 +08:00
|
|
|
select ND_BTT if BTT
|
2015-06-10 02:13:37 +08:00
|
|
|
help
|
|
|
|
|
Memory ranges for PMEM are described by either an NFIT
|
|
|
|
|
(NVDIMM Firmware Interface Table, see CONFIG_NFIT_ACPI), a
|
|
|
|
|
non-standard OEM-specific E820 memory type (type-12, see
|
|
|
|
|
CONFIG_X86_PMEM_LEGACY), or it is manually specified by the
|
|
|
|
|
'memmap=nn[KMG]!ss[KMG]' kernel command line (see
|
|
|
|
|
Documentation/kernel-parameters.txt). This driver converts
|
|
|
|
|
these persistent memory ranges into block devices that are
|
|
|
|
|
capable of DAX (direct-access) file system mappings. See
|
|
|
|
|
Documentation/nvdimm/nvdimm.txt for more details.
|
|
|
|
|
|
|
|
|
|
Say Y if you want to use an NVDIMM
|
|
|
|
|
|
2015-06-25 16:21:02 +08:00
|
|
|
config ND_BLK
|
|
|
|
|
tristate "BLK: Block data window (aperture) device support"
|
|
|
|
|
default LIBNVDIMM
|
|
|
|
|
select ND_BTT if BTT
|
|
|
|
|
help
|
|
|
|
|
Support NVDIMMs, or other devices, that implement a BLK-mode
|
|
|
|
|
access capability. BLK-mode access uses memory-mapped-i/o
|
|
|
|
|
apertures to access persistent media.
|
|
|
|
|
|
|
|
|
|
Say Y if your platform firmware emits an ACPI.NFIT table
|
|
|
|
|
(CONFIG_ACPI_NFIT), or otherwise exposes BLK-mode
|
|
|
|
|
capabilities.
|
|
|
|
|
|
nd_btt: atomic sector updates
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>
2015-06-25 16:20:32 +08:00
|
|
|
config ND_BTT
|
|
|
|
|
tristate
|
|
|
|
|
|
2015-06-25 16:20:04 +08:00
|
|
|
config BTT
|
nd_btt: atomic sector updates
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>
2015-06-25 16:20:32 +08:00
|
|
|
bool "BTT: Block Translation Table (atomic sector updates)"
|
|
|
|
|
default y if LIBNVDIMM
|
|
|
|
|
help
|
|
|
|
|
The Block Translation Table (BTT) provides atomic sector
|
|
|
|
|
update semantics for persistent memory devices, so that
|
|
|
|
|
applications that rely on sector writes not being torn (a
|
|
|
|
|
guarantee that typical disks provide) can continue to do so.
|
|
|
|
|
The BTT manifests itself as an alternate personality for an
|
|
|
|
|
NVDIMM namespace, i.e. a namespace can be in raw mode (pmemX,
|
|
|
|
|
ndblkX.Y, etc...), or 'sectored' mode, (pmemXs, ndblkX.Ys,
|
|
|
|
|
etc...).
|
|
|
|
|
|
|
|
|
|
Select Y if unsure
|
2015-06-25 16:20:04 +08:00
|
|
|
|
2015-06-10 02:13:37 +08:00
|
|
|
endif
|