On Wed, Mar 07, 2018 at 10:36:47AM -0800, Dan Williams wrote:
> > @@ -342,6 +372,10 @@ int util_filter_walk(struct ndctl_ctx *ctx, struct
> > util_filter_ctx *fctx,
> > param->namespace))
> > continue;
> >
> > +
If the kernel has CONFIG_NUMA unset namespaces in sysfs will lack a
numa_node attribute. In such cases ndctl will report a value of 0 for the
namespace numa_node in 'ndctl list'. Instead of reporting potentially bad
data just hide the numa_node field if it is unsupported.
Signed-off-by: Ross
Add support to 'ndctl list' so that we can filter DIMMs, regions and
namespaces based on numa node.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
Documentation/ndctl/ndctl-list.txt | 4
ndctl/list.c | 2 ++
util/filter.c
Add an API for getting the numa node of a given region and add a numa_node
field to the "ndctl list" region output.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
ndctl/lib/libndctl.c | 13 +
ndctl/lib/libndctl.sym | 1 +
ndctl/libndctl.h | 1
On Tue, Mar 06, 2018 at 08:59:11AM -0800, Dan Williams wrote:
> On Mon, Mar 5, 2018 at 7:54 PM, Ross Zwisler
> <ross.zwis...@linux.intel.com> wrote:
> > On Mon, Mar 05, 2018 at 05:09:21PM -0800, Dan Williams wrote:
> >> Dynamic debug can be instructed to add the
On Fri, Mar 02, 2018 at 01:20:49PM +0100, Johannes Thumshirn wrote:
> Dynamic debug can be instructed to add the function name to the debug
> output using the +f switch, so there is no need for the nfit module to
> do it again. If a user decides to add the +f switch for nfit's dynamic
> debug this
ynamic debug this results in double prints of the function name.
>
> Reported-by: Johannes Thumshirn <jthumsh...@suse.de>
> Reported-by: Ross Zwisler <ross.zwis...@linux.intel.com>
> Signed-off-by: Dan Williams <dan.j.willi...@intel.com>
Looks good to
ynamic debug this results in double prints of the function
> name.
>
> Reported-by: Johannes Thumshirn <jthumsh...@suse.de>
> Reported-by: Ross Zwisler <ross.zwis...@linux.intel.com>
> Signed-off-by: Dan Williams <dan.j.willi...@intel.com>
> ---
> drivers/
shirn <jthumsh...@suse.de>
I looked through this and this all looked right to me.
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
On Tue, Feb 06, 2018 at 03:19:15PM -0800, Darrick J. Wong wrote:
<>
> The last time I paid much attention to DAX was the thread "re-enable XFS
> per-inode DAX"[1] last September. Motivating me to merge anything else
> into DAX involves convincing me that we (mm, fs, dax developers) have
> some
On Tue, Feb 27, 2018 at 10:29:50AM -0700, Ross Zwisler wrote:
> In nfit_test0_setup() and nfit_test1_setup() we keep an 'offset' value
> which we use to calculate where in our 'nfit_buf' we will place our next
> structure. The handling of 'offset' and the calculation of the
On Mon, Mar 05, 2018 at 09:53:47AM -0800, Dan Williams wrote:
> On Mon, Mar 5, 2018 at 9:49 AM, Ross Zwisler
> <ross.zwis...@linux.intel.com> wrote:
> > On Fri, Mar 02, 2018 at 01:20:49PM +0100, Johannes Thumshirn wrote:
> >> Dynamic debug can be instructed to add the
On Fri, Mar 02, 2018 at 01:20:49PM +0100, Johannes Thumshirn wrote:
> Dynamic debug can be instructed to add the function name to the debug
> output using the +f switch, so there is no need for the nfit module to
> do it again. If a user decides to add the +f switch for nfit's dynamic
> debug this
On Thu, Feb 22, 2018 at 11:58:09AM -0800, Tony Luck wrote:
> EDAC driver needs to look up attributes of NVDIMMs provided in SMBIOS.
>
> Provide a function that looks up an acpi_nfit_memory_map from a device
> handle (node/socket/mc/channel/dimm) and returns the SMBIOS handle.
> Also pass back the
ugh nfit_test.0 where (t->setup_hotplug == 0) this is the size of
the entire buffer we allocated, including space for the hot plug
structures, not the size that we've actually filled in.
Fix this by only trying to parse the size of the structures that we've
filled in.
Signed-off-by: Ross Zwisler <
buf' in nfit_test0_alloc(). Also add some WARN_ON()s to
nfit_test0_setup() and nfit_test1_setup() to catch future issues where the
size of the buffer doesn't match the amount of data we're writing.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
tools/testing/nvdimm/test/nfit.c | 9
each structure insertion in a
consistent way, allowing us to always calculate the position of the next
structure to be inserted by just using 'nfit_buf + offset'.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
tools/testing/nvdimm/test/nfit.c
On Fri, Feb 23, 2018 at 01:45:40PM -0800, Dan Williams wrote:
> On Fri, Feb 23, 2018 at 12:58 PM, Ross Zwisler
> <ross.zwis...@linux.intel.com> wrote:
> > On Thu, Feb 22, 2018 at 10:59:22PM -0800, Dan Williams wrote:
> >> The 'fallocate -l 196608 $image' step in
clean up and disable all the nfit_test
regions and unloading the nfit_test module, as we do in test/pmem-errors.sh
et al.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
test/firmware-update.sh | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git
.l.ve...@intel.com>
> Signed-off-by: Dan Williams <dan.j.willi...@intel.com>
Sure, seems fine.
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
On Thu, Feb 22, 2018 at 10:59:22PM -0800, Dan Williams wrote:
> The 'fallocate -l 196608 $image' step in the test fails when $image is
> on an NFS mount. Use dd instead to create a sparse file. We do not need
> to allocate anything since we are only writing zeros.
>
> Cc: Dave Jiang
.com>
> Signed-off-by: Dan Williams <dan.j.willi...@intel.com>
Yay, I was hitting this as well in my setup which uses NFS, but wasn't sure if
the flock was necessary.
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux-nvdim
On Mon, Dec 18, 2017 at 12:54:05PM -0700, Ross Zwisler wrote:
> On Fri, Dec 08, 2017 at 08:29:25AM -0800, Dan Williams wrote:
> > These tables, NFIT and HMAT, are essential for describing
> > next-generation platform memory topologies and performance
> > charac
MiB region. Vishal happened to be getting the 64 MiB region, and I
happened to get the 32 MiB.
Fix this by explicitly requesting a namespace size of 64 MiB.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
test/btt-pad-compat.sh | 2 +-
1 file changed, 1 insertion(+), 1 de
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
.gitignore | 21 +
1 file changed, 21 insertions(+)
diff --git a/.gitignore b/.gitignore
index 3c2de0e..20a04de 100644
--- a/.gitignore
+++ b/.gitignore
@@ -32,3 +32,24 @@ version.m4
cscope.files
cscop
On Tue, Feb 20, 2018 at 11:18:28AM -0700, Dave Jiang wrote:
> Adding firmware output of firmware information when ndctl list -D -F is used.
> Components displayed are current firmware version, next firmware version,
> and if a powercycle is required (firmware updated).
>
> Signed-off-by: Dave
On Tue, Feb 13, 2018 at 02:43:54PM -0700, Dave Jiang wrote:
> Adding firmware output of firmware information when ndctl list -D -F is used.
> Components displayed are current firmware version, updated firmware version,
> and if a coldboot is required (firmware updated).
>
> Signed-off-by: Dave
On Tue, Feb 13, 2018 at 01:35:19PM -0700, Dave Jiang wrote:
>
>
> On 02/13/2018 01:31 PM, Ross Zwisler wrote:
> > On Thu, Feb 08, 2018 at 05:38:19PM -0700, Dave Jiang wrote:
> >> The function return values are confusing with the way the function is
> >> named.
On Thu, Feb 08, 2018 at 05:38:19PM -0700, Dave Jiang wrote:
> The function return values are confusing with the way the function is
> named. We expect a true or false return value but it actually returns
> 0/-errno. This makes the code very confusing. Changing the return values
> to return a bool
On Tue, Feb 13, 2018 at 03:54:04PM +0100, Christoph Hellwig wrote:
> Looks good:
>
> Reviewed-by: Christoph Hellwig
>
> Can you wire up your test cases for blktests?
Is blktests really the right place for this test? This failure is highly
dependent on the configuration of the
On Mon, Feb 12, 2018 at 03:05:10PM -0800, Dan Williams wrote:
> On Mon, Feb 12, 2018 at 2:53 PM, Jeff Moyer wrote:
> > Dave Jiang writes:
> >
> >> Re-enable deep flush so that users always have a way to be sure that a
> >> write
> >> does make it all the
For ext4 we have a similar issue where writes never happen, but we don't
currently have any xfstests that use loopback and show this issue.
Fix this by restoring the WRITE flag argument to iov_iter_bvec(). This
causes the xfstests to all pass.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.
page_map
Luis de Bethencourt (1):
device-dax: Fix trailing semicolon
Ross Zwisler (2):
Merge branch 'for-4.16/dax' into libnvdimm-for-next
Merge branch 'for-4.16/nfit' into libnvdimm-for-next
Toshi Kani (1):
acpi, nfit: fix register dimm error handling
arch/arm64/mm/mmu.c
On Mon, Feb 05, 2018 at 10:44:07AM -0800, Dan Williams wrote:
> On Mon, Feb 5, 2018 at 10:20 AM, Ross Zwisler
> <ross.zwis...@linux.intel.com> wrote:
> > On Mon, Feb 05, 2018 at 02:08:52PM +, Colin King wrote:
> >> From: Colin Ian King <colin.k...@canonical.com&g
moved.
>
> Cleans up clang warning:
> drivers/nvdimm/pfn_devs.c:307:17: warning: Value stored to 'dev' during
> its initialization is never read
>
> Signed-off-by: Colin Ian King <colin.k...@canonical.com>
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
M
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 <dave.ji...@intel.com>
> R
ooks good to me. You can add:
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
> Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Looks good aside from the one nit.
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
anonical.com>
Sure, this looks good. Thanks.
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
On Fri, Dec 08, 2017 at 02:00:20PM -0700, Dave Jiang wrote:
> Providing a sysfs attribute for nd_region that shows the persistence
> capabilities for the platform.
>
> Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Sure, looks fine.
Reviewed-by: Ross Zwisler <ross.zw
On Fri, Dec 08, 2017 at 02:00:14PM -0700, Dave Jiang wrote:
> Propogate the ADR attribute flag from the NFIT platform capabilities
> sub-table to nd_region.
>
> Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Sure, looks fine.
Reviewed-by: Ross Zwisler <ross.zw
On Fri, Dec 08, 2017 at 02:00:25PM -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
> ---
>
On Fri, Dec 08, 2017 at 02:00:09PM -0700, Dave Jiang wrote:
> 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
nd_namespace_io and older v1.1 namespace labels, we will display
the default sector size of 512.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
Reviewed-by: Dave Jiang <dave.ji...@intel.com>
---
util/json.c | 29 -
1 file changed, 24 insertions(+),
nd_namespace_io and older v1.1 namespace labels, we will display
the default sector size of 512.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
util/json.c | 23 ++-
1 file changed, 18 insertions(+), 5 deletions(-)
diff --git a/util/json.c b/util/json.c
index 9
sles/ndctl.spec
util/.dirstamp
util/log.lo
util/sysfs.lo
version.m4
nothing added to commit but untracked files present (use "git add" to track)
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
.gitignore | 19 +
of
devtype nd_namespace_io.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
util/json.c | 22 +-
1 file changed, 17 insertions(+), 5 deletions(-)
diff --git a/util/json.c b/util/json.c
index 95beaa2..89306f0 100644
--- a/util/json.c
+++ b/util/json.c
@@ -
.out.bad' to see the entire
diff)
This is expected because the dm-thin target currently lacks DAX support.
Just skip this test if we are using DAX.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
common/dmthin | 5 +
1 file changed, 5 insertions(+)
diff --git a/
On Thu, Jan 18, 2018 at 12:23:43PM +0800, Eryu Guan wrote:
> On Wed, Jan 17, 2018 at 04:23:51PM -0700, Ross Zwisler wrote:
> > generic/347 currently fails when run in cojunction with the DAX mount
> > option:
> >
> > generic/347 72s ... - output mismatch (see
> &g
ip shared/272 if the DAX mount option is in use.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
tests/shared/272 | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/shared/272 b/tests/shared/272
index 7023b657..0c9763df 100755
--- a/tests/shared/272
+++ b/tests/sha
.out.bad' to see the entire
diff)
This is expected because the dm-thin target currently lacks DAX support.
Just skip this test if we are using DAX.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
tests/generic/347 | 1 +
1 file changed, 1 insertion(+)
diff --git a
On Thu, Jan 11, 2018 at 06:49:17PM +, Cheng-mean Liu (SOCCER) wrote:
> In the case of emulated NVDIMM devices in the VM environment, there
> are scenarios that NVDIMM device with much smaller sizes are desired, for
> example, we might
> use a single enumerated NVDIMM DAX device for
On Wed, Aug 23, 2017 at 10:21 PM, Wanpeng Li wrote:
> From: Wanpeng Li
>
> vmx_complete_interrupts() assumes that the exception is always injected,
> so it would be dropped by kvm_clear_exception_queue(). This patch separates
> exception.pending from
-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
Changes since v1:
- Changed ordering of .gitignore entries. (Eryu)
- Added '_require_scratch_ext4_feature "inline_data"' test. (Eryu)
---
.gitignore | 1 +
src/Makefile
on or off journaling. The latter is how we
prevent this issue in the kernel.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
Changes since v1:
- Reordered .gitignore entry. (Eryu)
- Added comments about how "chattr +j" turns off DAX and about why we need
the 'n
On Sat, Dec 23, 2017 at 04:56:43PM -0800, Dan Williams wrote:
> In support of testing truncate colliding with dma add a mechanism that
> delays the completion of block I/O requests by a programmable number of
> seconds. This allows a truncate operation to be issued while page
> references are held
Christoph Hellwig <h...@lst.de>
> Cc: Matthew Wilcox <mawil...@microsoft.com>
> Cc: Ross Zwisler <ross.zwis...@linux.intel.com>
> Signed-off-by: Dan Williams <dan.j.willi...@intel.com>
> ---
> drivers/dax/super.c | 15 --
> fs/dax.c| 75
On Fri, Dec 22, 2017 at 02:53:42PM -0800, Dan Williams wrote:
> On Thu, Dec 21, 2017 at 12:31 PM, Brice Goglin <brice.gog...@gmail.com> wrote:
> > Le 20/12/2017 à 23:41, Ross Zwisler a écrit :
> [..]
> > Hello
> >
> > I can confirm that HPC runtimes are going to
On Fri, Dec 22, 2017 at 08:39:41AM +0530, Anshuman Khandual wrote:
> On 12/14/2017 07:40 AM, Ross Zwisler wrote:
<>
> > We solve this issue by providing userspace with performance information on
> > individual memory ranges. This performance information is exposed via
> &g
On Fri, Dec 22, 2017 at 08:39:41AM +0530, Anshuman Khandual wrote:
> On 12/14/2017 07:40 AM, Ross Zwisler wrote:
> > Quick Summary
> >
> > Platforms exist today which have multiple types of memory attached to a
> > single CPU. These disparate memory range
xfs_io operation such as mwrite. This isolates the log replay from other
operations that happen as part of xfs_io exiting (file handles being
closed, mmaps being torn down, etc.). This also allows users to insert
multiple marks between different xfs_io commands.
Signed-off-by: Ross Zwisler
Add support for a new -S flag to xfs_io's mmap command. This opens the
mapping with the (MAP_SYNC | MAP_SHARED_VALIDATE) flags instead of the
standard MAP_SHARED flag.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
Suggested-by: Dave Chinner <da...@fromorbit.com>
---
On Thu, Dec 21, 2017 at 09:09:08AM -0800, Darrick J. Wong wrote:
> On Tue, Dec 05, 2017 at 04:56:50PM -0700, Ross Zwisler wrote:
> > @@ -195,6 +200,13 @@ mmap_f(
> > case 'x':
> > prot |= PROT_EXEC;
> > bre
te the error from ext4_iomap_begin() and implement do standard
> allocation retry loop in ext4_dax_huge_fault().
>
> Signed-off-by: Jan Kara <j...@suse.cz>
Looks good.
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux
_iomap_fault() for passing such error.
>
> Signed-off-by: Jan Kara <j...@suse.cz>
I like how much simpler this version is. Looks good.
Reviewed-by: Ross Zwisler <ross.zwis...@linux.intel.com>
___
Linux-nvdimm mailing list
Linux-nvdimm@lis
On Wed, Dec 13, 2017 at 09:45:52AM -0700, Ross Zwisler wrote:
> On Tue, Dec 05, 2017 at 04:56:48PM -0700, Ross Zwisler wrote:
> > This is the second revision of my MAP_SYNC + dm-log-writes support for
> > xfsprogs.
>
> Friendly ping on this x
On Wed, Dec 20, 2017 at 10:19:37AM -0800, Matthew Wilcox wrote:
> On Mon, Dec 18, 2017 at 01:35:47PM -0700, Ross Zwisler wrote:
> > What I'm hoping to do with this series is to just provide a sysfs
> > representation of the HMAT so that applications can know which NUMA node
On Mon, Dec 18, 2017 at 01:35:47PM -0700, Ross Zwisler wrote:
> On Thu, Dec 14, 2017 at 02:00:32PM +0100, Michal Hocko wrote:
<>
> > What is the testing procedure? How can I setup qemu to simlate such HW?
>
> Well, the QEMU table simulation is gross, so I'd rather not g
On Tue, Dec 19, 2017 at 03:30:51PM +0100, Jan Kara wrote:
> On Fri 15-12-17 12:36:25, Ross Zwisler wrote:
> > On Wed, Dec 13, 2017 at 10:13:50AM +0100, Jan Kara wrote:
> > > Hello,
> > >
> > > these two patches fix handling of ENOSPC during DAX faults. The pro
On Thu, Dec 14, 2017 at 02:00:32PM +0100, Michal Hocko wrote:
> [CC linix-api]
Oh, thanks. I'll add them to my CC list for sysfs related changes in the
future.
> On Wed 13-12-17 19:10:16, Ross Zwisler wrote:
> > This is the third revision of my patches adding a sysfs re
On Fri, Dec 08, 2017 at 08:29:25AM -0800, Dan Williams wrote:
> These tables, NFIT and HMAT, are essential for describing
> next-generation platform memory topologies and performance
> characteristics. Allow them to be overridden for debug and test and
> purposes.
>
> Cc: Ross
On Fri, Dec 15, 2017 at 01:52:03AM +0100, Rafael J. Wysocki wrote:
<>
> > diff --git a/drivers/acpi/hmat/core.c b/drivers/acpi/hmat/core.c
> > new file mode 100644
> > index ..61b90dadf84b
> > --- /dev/null
> > +++ b/drivers/acpi/hmat/core.c
> > @@ -0,0 +1,536 @@
> > +/*
> > + *
On Wed, Dec 13, 2017 at 10:13:50AM +0100, Jan Kara wrote:
> Hello,
>
> these two patches fix handling of ENOSPC during DAX faults. The problem is
> that currently running transaction may be holding lots of already freed
> blocks which can be reallocated only once the transaction commits. Standard
they have subtable
headers of type struct acpi_hmat_structure which has a 2 byte type and a 4
byte length.
Enhance the subtable parsing in acpi_parse_entries_array() so that it can
handle these new HMAT subtables.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
drivers/acpi/ta
── node2 -> ../../node/node2
├── power
│ ├── async
│ ...
├── subsystem -> ../../../../bus/hmat
└── uevent
Users can discover information about the memory owned by this memory target
by following the node2 symlink and looking at meminfo, vmstat and at the
memory* memory section symlin
d be present in very
large systems, so in this series we only surface performance information
for local (initiator,target) pairings. The changelog for patch 5
discusses this in detail.
Ross Zwisler (3):
acpi: HMAT support in acpi_parse_entries_array()
hmat: add heterogeneous memory sysfs sup
regardless of how
we nest them in a directory hierarchy.
By only representing performance information for local (initiator,target)
pairings, we reduce the number of sysfs entries to O(num_targets).
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
dri
On Tue, Dec 05, 2017 at 04:56:48PM -0700, Ross Zwisler wrote:
> This is the second revision of my MAP_SYNC + dm-log-writes support for
> xfsprogs.
Friendly ping on this xfsprogs series.
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
On Fri, Dec 08, 2017 at 02:36:10PM +0800, Eryu Guan wrote:
> (Test was re-numbered as generic/470, BTW.)
Thanks! For future reference, does the pattern of us submitting tests with
high numbers (generic/999) to avoid merge conflicts and asking you to renumber
them when you merge work for you?
On Thu, Dec 07, 2017 at 06:36:57PM +0800, Eryu Guan wrote:
> Now we have two _require rules to test log_writes dm target:
> _require_log_writes # _notrun explicitly when MOUNT_OPTIONS contains dax
> _require_log_writes_dax # _notrun if log-writes target doesn't support
> dax
>
> I think
On Wed, Dec 06, 2017 at 12:47:49PM -0800, Darrick J. Wong wrote:
> On Wed, Dec 06, 2017 at 01:10:14PM -0700, Ross Zwisler wrote:
> > On Wed, Dec 06, 2017 at 11:27:43AM +1100, Dave Chinner wrote:
> > > On Tue, Dec 05, 2017 at 04:56:49PM -0700, Ross Zwisler wrote:
> > &g
On Wed, Dec 06, 2017 at 11:27:43AM +1100, Dave Chinner wrote:
> On Tue, Dec 05, 2017 at 04:56:49PM -0700, Ross Zwisler wrote:
> > I recently upgraded my compiler from
> > gcc (GCC) 6.4.1 20170727 (Red Hat 6.4.1-1)
> > to
> > gcc (GCC) 7.2.1 20170915 (Red H
xfs_io operation such as mwrite. This isolates the log replay from other
operations that happen as part of xfs_io exiting (file handles being
closed, mmaps being torn down, etc.). This also allows users to insert
multiple marks between different xfs_io commands.
Signed-off-by: Ross Zwisler
integrity checking. We can only verify
that the metadata writes for the page faults happened.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
common/dmlogwrites| 20 +
tests/generic/999 | 82 +++
tests/gene
54775807 is the max value). The format we're using for bbuf is
"[%lld..%lld]:" which has 2 signed long longs plus 6 other characters
"[..]:\0", which means it's possible we'll print up to 44 characters,
overflowing our 32 char buffer.
Fix this by bumping all the buffer sizes i
xfs_io operation such as mwrite. This isolates the log replay from other
operations that happen as part of xfs_io exiting (file handles being
closed, mmaps being torn down, etc.). This also allows users to insert
multiple marks between different xfs_io commands.
Signed-off-by: Ross Zwisler
s enhances xfs_io by adding support for the MAP_SYNC mmap() flag
and for dm-log-writes marks. This allows the resulting xfstest for
MAP_SYNC to be much simpler and have no custom C programs.
Ross Zwisler (3):
xfs_io: fix compiler warnings in getfsmap code
xfs_io: add MAP_SYNC support to mmap(
On Sat, Nov 18, 2017 at 02:17:03PM +0800, Eryu Guan wrote:
> On Fri, Nov 17, 2017 at 01:28:27PM -0700, Ross Zwisler wrote:
> > Enhance _require_dm_target so that a user can request a minimum version of a
> > given dm target.
> >
> > DAX support was added to v1.1.0
On Fri, Nov 17, 2017 at 12:40:21PM -0800, Darrick J. Wong wrote:
> On Fri, Nov 17, 2017 at 01:25:23PM -0700, Ross Zwisler wrote:
> > Add support for a new -S flag to xfs_io's mmap command. This opens the
> > mapping with the (MAP_SYNC | MAP_SHARED_VALIDATE) flags instead of t
On Fri, Nov 17, 2017 at 12:35:43PM -0800, Dan Williams wrote:
> On Fri, Nov 17, 2017 at 12:25 PM, Ross Zwisler
> <ross.zwis...@linux.intel.com> wrote:
> > Add support for a new -S flag to xfs_io's mmap command. This opens the
> > mapping with the (MAP_SYNC | MAP_SHARED
On Fri, Nov 17, 2017 at 03:03:39PM -0600, Eric Sandeen wrote:
> On 11/17/17 2:48 PM, Ross Zwisler wrote:
> > On Fri, Nov 17, 2017 at 02:39:07PM -0600, Eric Sandeen wrote:
> >> On 11/17/17 2:25 PM, Ross Zwisler wrote:
> >>> Add a new 'log_writes' command to xfs_io so
On Fri, Nov 17, 2017 at 02:39:07PM -0600, Eric Sandeen wrote:
> On 11/17/17 2:25 PM, Ross Zwisler wrote:
> > Add a new 'log_writes' command to xfs_io so that we can add dm-log-writes
> > log marks via the external 'dmsetup' executable. It's helpful to allow
> > users of xfs_i
Enhance _require_dm_target so that a user can request a minimum version of a
given dm target.
DAX support was added to v1.1.0 of the dm-log-writes kernel module, so
allow the DAX mount option starting with that version.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
Sug
Allow tests to inquire about whether a mount option is set, rather than
just disallowing it via _exclude_scratch_mount_option().
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
Suggested-by: Amir Goldstein <amir7...@gmail.com>
---
common/rc | 10 ++
1 file
. This can cause unexpected test results.
Fix this by making sure that the mark we're given actually exists in the
log before we allow the replay.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
common/dmlogwrites | 4
1 file changed, 4 insertions(+)
diff --git a/
integrity checking. We can only verify
that the metadata writes for the page faults happened.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
---
tests/generic/468 | 83 +++
tests/generic/468.out | 3 ++
tests/generic/group | 1 +
/scm/linux/kernel/git/zwisler/xfstests-dev.git/log/?h=map_sync_test_v4
Ross Zwisler (4):
common/rc: add _scratch_has_mount_option()
dm-log-writes: only replay log to marks that exist
dm-log-writes: allow DAX to be used when possible
generic: add test for DAX MAP_SYNC support
common
will be found in
the upcoming v4.15-rc1. For ease of testing I've posted a kernel that is
v4.14 plus just those two patch series here:
https://git.kernel.org/pub/scm/linux/kernel/git/zwisler/linux.git/log/?h=map_sync_dm_log_writes
Ross Zwisler (2):
xfs_io: add MAP_SYNC support to mmap()
xfs_io: add
commands.
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
Suggested-by: Dave Chinner <da...@fromorbit.com>
---
io/Makefile | 5 ++--
io/init.c | 1 +
io/io.h | 1 +
io/log_writes.c | 78 +++
man/man8/x
On Thu, Oct 26, 2017 at 07:59:39AM +0300, Amir Goldstein wrote:
> On Wed, Oct 25, 2017 at 11:47 PM, Ross Zwisler
> <ross.zwis...@linux.intel.com> wrote:
> > Add a test that exercises DAX's new MAP_SYNC flag.
> >
> > This test creates a file and writes to it via an m
201 - 300 of 754 matches
Mail list logo