Re: [LSF/MM TOPIC] Native NVMM file systems

On Wed, Jan 31, 2018 at 04:45:36PM -0800, Andiry Xu wrote: > PMEM/DAX should allow for significant improvements in file system > performance and enable new programming models that allow direct, > efficient access to PMEM from userspace. Achieving these gains in > existing file systems built for

Re: [LSF/MM TOPIC] Native NVMM file systems

[ adding Jan and Ross ] On Wed, Jan 31, 2018 at 5:03 PM, Dan Williams wrote: > On Wed, Jan 31, 2018 at 4:45 PM, Andiry Xu wrote: >> PMEM/DAX should allow for significant improvements in file system >> performance and enable new programming models that

Re: [LSF/MM TOPIC] Native NVMM file systems

On Wed, Jan 31, 2018 at 4:45 PM, Andiry Xu wrote: > PMEM/DAX should allow for significant improvements in file system > performance and enable new programming models that allow direct, > efficient access to PMEM from userspace. Achieving these gains in > existing file systems

Re: [ndctl PATCH] ndctl: remove a double put in injection status json

On Wed, Jan 31, 2018 at 4:49 PM, Vishal Verma wrote: > newer json-c catches a double put of the json object in printing the > injection status. We only need to put the containing jobj. > > Cc: Dan Williams > Signed-off-by: Vishal Verma

[ndctl PATCH] ndctl: remove a double put in injection status json

newer json-c catches a double put of the json object in printing the injection status. We only need to put the containing jobj. Cc: Dan Williams Signed-off-by: Vishal Verma --- ndctl/inject-error.c | 2 -- 1 file changed, 2 deletions(-) diff

[LSF/MM TOPIC] Native NVMM file systems

PMEM/DAX should allow for significant improvements in file system performance and enable new programming models that allow direct, efficient access to PMEM from userspace. Achieving these gains in existing file systems built for block devices (e.g., XFS and EXT4…) presents a range of challenges

Re: [PATCH] xfs: fix rt_dev usage for DAX

On 01/31/2018 05:29 PM, Darrick J. Wong wrote: On Wed, Jan 31, 2018 at 05:13:20PM -0700, Dave Jiang wrote: When using realtime device (rtdev) with xfs where the data device is not DAX capable, two issues arise. One is when data device is not DAX but the realtime device is DAX capable, we

Re: [PATCH] xfs: fix rt_dev usage for DAX

On Wed, Jan 31, 2018 at 05:13:20PM -0700, Dave Jiang wrote: > When using realtime device (rtdev) with xfs where the data device is not > DAX capable, two issues arise. One is when data device is not DAX but the > realtime device is DAX capable, we currently disable DAX. > After passing this check,

[PATCH] xfs: fix rt_dev usage for DAX

When using realtime device (rtdev) with xfs where the data device is not DAX capable, two issues arise. One is when data device is not DAX but the realtime device is DAX capable, we currently disable DAX. After passing this check, we are also not marking the inode as DAX capable. This change will

[PATCH v5 4/4] ndctl, test: firmware update unit test

Adding a unit test that will use nfit_test kernel module to test the firmware update sequence. Signed-off-by: Dave Jiang --- test/Makefile.am|3 +- test/firmware-update.sh | 65 +++ 2 files changed, 67

[PATCH v5 2/4] ndctl: add firmware download support functions in libndctl

Adding the DSM commands from Intel DSM v1.6 to support firmware update sequence in the libndctl library as additional dimm_ops. Signed-off-by: Dave Jiang --- ndctl/lib/Makefile.am |1 ndctl/lib/firmware.c | 109 ndctl/lib/intel.c | 262

[PATCH v5 3/4] ndctl: add firmware update command option for ndctl

Adding option "update-firmware" to ndctl for update firmware support from Intel DSM v1.6. ndctl update-firmware takes an option of -f for a firmware binary and a -d for the DIMM name: ndctl update-firmware -d nmem0 -f new_firmware.bin Signed-off-by: Dave Jiang ---

[PATCH v5 1/4] ndctl: add support to alloc_intel_cmd for variable payload

Certain payloads have variable size. The existing alloc_intel_cmd() does not take into account of that. Adding additional size for allocation. We do waste a little bit of the space because of the smart payload. Signed-off-by: Dave Jiang --- ndctl/lib/intel.c |2 +- 1

[PATCH v5 0/4] ndctl: add DIMM firmware update support

The following series implements support for DIMM firmware update in ndctl. v5: Update all to SPDX license headers v4: Addressed Vishal's comments - replaced copyright header - changed gettimeofday to clock_gettime - changed return error for verify_fw_size() v3: Addressed Dan's comments -

Re: [PATCH 1/2] libnvdimm/nfit_test: add firmware download emulation

On Fri, 2018-01-26 at 10:55 -0700, Dave Jiang wrote: > Adding support in nfit_test for DSM v1.6 firmware update sequence. > The test > will simulate the flashing of firmware to the DIMM. A bogus version > string > will be returned as the test has no idea how to parse the firmware > binary. > Any

[PATCH v5 4/4] nfit-test: Add platform cap support from ACPI 6.2a to test

Adding NFIT platform capabilities sub table in nfit_test simulated ACPI NFIT table. Only the first NFIT table is added with the capability sub-table. Signed-off-by: Dave Jiang Reviewed-by: Ross Zwisler --- tools/testing/nvdimm/test/nfit.c |

[PATCH v5 3/4] libnvdimm: expose platform persistence attribute for nd_region

Providing a sysfs attribute for nd_region that shows the persistence capabilities for the platform. Signed-off-by: Dave Jiang Reviewed-by: Ross Zwisler --- drivers/nvdimm/region_devs.c | 13 + 1 file changed, 13 insertions(+)

[PATCH v5 0/4] add support for platform persistence capabilities

ACPI 6.2a provides an NFIT sub-table that informs if the platform has auto CPU flush and memory flush on unexpected power loss events. This series propogates those attributes to nd_region and add sysfs attribute to show those capabilities. v5: Fix value of mask in nfit_test. v4: Fixed typos and

[PATCH v5 2/4] acpi: nfit: add persistent memory control flag for nd_region

Propagate the ADR attribute flag from the NFIT platform capabilities sub-table to nd_region. Signed-off-by: Dave Jiang Reviewed-by: Ross Zwisler --- drivers/acpi/nfit/core.c |3 +++ include/linux/libnvdimm.h |6 ++ 2 files

[PATCH v5 1/4] acpi: nfit: Add support for detect platform CPU cache flush on power loss

In ACPI 6.2a the platform capability structure has been added to the NFIT tables. That provides software the ability to determine whether a system supports the auto flushing of CPU caches on power loss. If the capability is supported, we do not need to do dax_flush(). Plumbing the path to set the

Re: [PATCH v4 4/4] nfit-test: Add platform cap support from ACPI 6.2a to test

On Tue, Jan 30, 2018 at 05:21:39PM -0700, Dave Jiang wrote: > Adding NFIT platform capabilities sub table in nfit_test simulated ACPI > NFIT table. Only the first NFIT table is added with the capability > sub-table. > > Signed-off-by: Dave Jiang > Reviewed-by: Ross Zwisler