___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
Looks good in general...
Add "ndctl:" to the subject.
On Wed, Aug 30, 2017 at 6:29 PM, Yasunori Goto wrote:
>
> This patch makes 2 new interfaces :
> - Call translate SPA featture of ACPI 6.2.
> - Find DIMM which SPA(System Physical Address) belongs to.
>
>
> Signed-off-by: Yasunori Goto
>
This adds a regression test for the following kernel patch:
commit 42d4a99b09cb ("ext4: fix fault handling when mounted with -o dax,ro")
The above patch fixes an issue with ext4 where executables cannot be run on
read-only filesystems mounted with the DAX option.
This issue does not appear to be
On Wed, Aug 30, 2017 at 07:51:03AM -0700, Christoph Hellwig wrote:
> > The above patch fixes an issue with ext4 where executables cannot be run on
> > read-only filesystems mounted with the DAX option.
> >
> > This issue does not appear to be present in ext2 or XFS, as they both pass
> > the test.
On Wed, Aug 30, 2017 at 06:59:30PM +0800, Eryu Guan wrote:
> On Tue, Aug 29, 2017 at 04:37:15PM -0600, Ross Zwisler wrote:
> > This adds a regression test for the following kernel patch:
> >
> > commit 42d4a99b09cb ("ext4: fix fault handling when mounted with -o dax,ro")
> >
> > The above patch f
On Wed, Aug 30, 2017 at 6:25 PM, Yasunori Goto wrote:
>
> To check what feature can be called via ND_CMD_CALL, ndctl need to read
> device/nfit/dsm_mask. This patch make an interface to check it in libndctl.c
>
>
> Signed-off-by: Yasunori Goto
>
> ---
> ndctl/lib/libndctl.c | 13 +
Make sure you also put "ndctl:" in your subject lines so that it
appears in the git commit message.
Also include Jerry on any future version of these patches.
On Wed, Aug 30, 2017 at 6:23 PM, Yasunori Goto wrote:
>
> This patch introduces libndctl-nfit.h.
>
> Since these command can be executed
The original message was received at Thu, 31 Aug 2017 10:05:52 +0800
from cixar.com [165.174.172.71]
- The following addresses had permanent fatal errors -
- Transcript of session follows -
while talking to lists.01.org.:
>>> MAIL From:kris.ko...@cixar.com
<<< 501 kris.ko...@ci
nfit_test supports translate SPA
To test ndctl list which use interface of Translate SPA,
nfit_test needs to emulates it.
This test module searches region which includes SPA and
returns 1 dimm handle which is last one.
Signed-off-by: Yasunori Goto
---
tools/testing/nvdimm/test/nfit.c | 104 +
Though nfit_test need to show what feature is supported via ND_CMD_CALL on
device/nfit/dsm_mask, currently there is no way to tell it.
This patch makes to enable it.
Signed-off-by: Yasunori Goto
---
drivers/acpi/nfit/core.c | 1 +
drivers/acpi/nfit/nfit.h | 1 +
2 files changed, 2 insertions(+)
In preparation for the error clearing rework, add sector_size in the
arena_info struct.
Signed-off-by: Vishal Verma
---
drivers/nvdimm/btt.c | 1 +
drivers/nvdimm/btt.h | 2 ++
2 files changed, 3 insertions(+)
diff --git a/drivers/nvdimm/btt.c b/drivers/nvdimm/btt.c
index 374ae62..e9dd651 10064
In btt_map_read, we read the map twice to make sure that the map entry
didn't change after we added it to the read tracking table. In
anticipation of expanding the use of the error bit, also make sure that
the error and zero flags are constant across the two map reads.
Signed-off-by: Vishal Verma
With the ACPI NFIT 'DSM' methods, acpi can be called from IO paths.
Specifically, the DSM to clear media errors is called during writes, so
that we can provide a writes-fix-errors model.
However it is easy to imagine a scenario like:
-> write through the nvdimm driver
-> acpi allocation
-
The IO context conversion for rw_bytes missed a case in the BTT write
path (btt_map_write) which should've been marked as atomic.
In reality this should not cause a problem, because map writes are to
small for nsio_rw_bytes to attempt error clearing, but it should be
fixed for posterity.
Add a mi
Clearing errors or badblocks during a BTT write requires sending an ACPI
DSM, which means potentially sleeping. Since a BTT IO happens in atomic
context (preemption disabled, spinlocks may be held), we cannot perform
error clearing in the course of an IO. Due to this error clearing for
BTT IOs has
changes in v7:
- There were cases where any problem while attempting to clear errors
would result in the BTT thread doing IO to be locked in an infinite
retry loop. Fix those conditions by detecting any error in clearing
and bailing on that write. (Toshi, Dan).
changes in v6:
- Remove t
Add helpers for converting a raw map entry to just the block number, or
either of the 'e' or 'z' flags in preparation for actually using the
error flag to mark blocks with media errors.
Signed-off-by: Vishal Verma
---
drivers/nvdimm/btt.c | 8
drivers/nvdimm/btt.h | 4
2 files chan
Make private definitions to command emulation.
These definitions were originally defined at include/uapi/linux/ndctl.h,
but they are only for nfit_test emulation now.
Signed-off-by: Yasunori Goto
---
tools/testing/nvdimm/test/nfit_test.h | 48 +++
1 file changed,
This patch set is to emulate translate SPA feature by nfit_test.ko.
Certainly the nfit acpi driver supports translate SPA interface
via ND_CMD_CALL. But nfit_test does not support it yet.
To test translate SPA with nfit_test, this patch set is needed.
---
Change log since v2 [1]:
- Make priva
Show dimm's name which has badblock by ndctl list command.
This patch uses translate SPA interface to get bad dimm info.
Signed-off-by: Yasunori Goto
---
util/json.c | 39 +++
1 file changed, 39 insertions(+)
diff --git a/util/json.c b/util/json.c
index 981
This patch makes 2 new interfaces :
- Call translate SPA featture of ACPI 6.2.
- Find DIMM which SPA(System Physical Address) belongs to.
Signed-off-by: Yasunori Goto
---
ndctl/Makefile.am | 1 +
ndctl/lib/Makefile.am | 4 +-
ndctl/lib/libndctl-nfit.c | 147
Currently ndctl supports ND_CMD_CALL only for DIMM,
but Translate SPA is the feature of bus.
So ND_CMD_CALL must be allowed bus's ioctl().
Signed-off-by: Yasunori Goto
---
ndctl/lib/libndctl.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/ndctl/lib/libndctl.c b/ndctl/lib/libndctl.c
index
To check what feature can be called via ND_CMD_CALL, ndctl need to read
device/nfit/dsm_mask. This patch make an interface to check it in libndctl.c
Signed-off-by: Yasunori Goto
---
ndctl/lib/libndctl.c | 13 +
ndctl/lib/libndctl.sym | 1 +
ndctl/libndctl.h.in| 1 +
3 file
This patch introduces libndctl-nfit.h.
Since these command can be executed via ND_CMD_CALL,
libndctl.h which is shared between ndctl command and kernel does not
have to include these defintions.
So, libndctl-nfit.h, which is defined for only ndctl, is created instead,
and move definitions from n
Hi,
I wrote v3 patch set to show broken NVDIMM for ndctl list command.
Please check it.
---
Change log since v2 [1]:
- Make libndctl-nfit.h and libndctl-nfit.c and define new interfaces
which use translate spa on them.
- Add sanity checks for new interfaces.
- Fix some names
o trans_
The mmap(2) syscall suffers from the ABI anti-pattern of not validating
unknown flags. However, proposals like MAP_SYNC and MAP_DIRECT need a
mechanism to define new behavior that is known to fail on older kernels
without the support. Define a new MAP_VALIDATE flag pattern that is
guaranteed to fai
As noted in patch2:
The mmap(2) syscall suffers from the ABI anti-pattern of not validating
unknown flags. However, proposals like MAP_SYNC and MAP_DIRECT need a
mechanism to define new behavior that is known to fail on older kernels
without the support. Define a new MAP_VALIDATE f
We are running running short of vma->vm_flags. We can avoid needing a
new VM_* flag in some cases if the original @flags submitted to mmap(2)
is made available to the ->mmap() 'struct file_operations'
implementation. For example, the proposed addition of MAP_DIRECT can be
implemented without taking
On Wed, Aug 30, 2017 at 12:43:58PM -0700, Dan Williams wrote:
> The ->iomap_begin() operation is a hot path, so cache the
> fs_dax_get_by_host() result at mount time to avoid the incurring the
> hash lookup overhead on a per-i/o basis.
>
> Cc: "Darrick J. Wong"
> Reported-by: Christoph Hellwig
>
On Wed, Aug 30, 2017 at 12:43:52PM -0700, Dan Williams wrote:
> Add a helper that can replace the following common pattern:
>
> if (blk_queue_dax(bdev->bd_queue))
> fs_dax_get_by_host(bdev->bd_disk->disk_name);
>
> This will be used to move dax_device lookup from iomap-operati
Hi Linus, please pull from:
git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm libnvdimm-fixes
...to receive a single patch removing some structure definitions from a
uapi header file. These payloads are never processed directly by the
kernel they are simply passed through an ioctl as
Adding a function that will export the transfer capability per descriptor
for a DMA device for the dmaengine subsystem.
Signed-off-by: Dave Jiang
---
drivers/dma/ioat/init.c |1 +
include/linux/dmaengine.h | 10 ++
2 files changed, 11 insertions(+)
diff --git a/drivers/dma/ioat/
Adding a dmaengine transaction operation that allows copy to/from a
scatterlist and a flat buffer.
Signed-off-by: Dave Jiang
---
Documentation/dmaengine/provider.txt |3 +++
include/linux/dmaengine.h| 19 +++
2 files changed, 22 insertions(+)
diff --git a/Docum
Commit 7618d0359c16 ("dmaengine: ioatdma: Set non RAID channels to be
private capable") makes all non-RAID ioatdma channels as private to be
requestable by dma_request_channel(). With PQ CAP support going away for
ioatdma, this would make all channels private. To support the usage of
ioatdma for bl
This provides support to unmap scatterlist with the
dmaengine_unmap_data. We will support only 1 scatterlist per
direction.
Signed-off-by: Dave Jiang
---
drivers/dma/dmaengine.c | 27 +++
include/linux/dmaengine.h | 13 -
2 files changed, 39 insertions(+
Add a DMA supported blk-mq driver for pmem. This provides significant
CPU utilization reduction at the cost of some increased latency and
bandwidth reduction in some cases. By default the current cpu-copy based
pmem driver will load, but this driver can be manually selected with a
modprobe configu
Adding a dmaengine support function to provide the number of available
channels that can be shared with support of a filter function.
Signed-off-by: Dave Jiang
---
drivers/dma/dmaengine.c | 45 +
include/linux/dmaengine.h |7 +++
2 files ch
In preparation for adding another pmem driver, moving all common functions
in pmem that will be used by new driver to a pmem_core module to share.
Signed-off-by: Dave Jiang
---
drivers/nvdimm/Kconfig | 10 +
drivers/nvdimm/Makefile|3
drivers/nvdimm/pmem.c | 401 +---
According to driver comment REQ_FLUSH is replaced with REQ_PREFLUSH after
v4.8-rc1. Removing definition and putting in new definition.
Signed-off-by: Dave Jiang
---
drivers/nvdimm/pmem.c |7 +--
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/drivers/nvdimm/pmem.c b/drivers/
Adding ioatdma support to copy from a physically contiguous buffer to a
provided scatterlist and vice versa. This is used to support
reading/writing persistent memory in the pmem driver.
Signed-off-by: Dave Jiang
---
drivers/dma/ioat/dma.h |4 +++
drivers/dma/ioat/init.c |2 ++
drivers/
- Per Dan's suggestions
- Moved all common code from attach_disk to pmem_core as helper functions.
- Fixed up Kconfig dependencies
- Cleaned up header file inclusions
- Removed module parameters
- Split pmem_core refactor into own patch
- Removed REQ_FLUSH define
v6:
- Put all common c
The ->iomap_begin() operation is a hot path, so cache the
fs_dax_get_by_host() result at mount time to avoid the incurring the
hash lookup overhead on a per-i/o basis.
Cc: "Theodore Ts'o"
Cc: Andreas Dilger
Cc: Jan Kara
Reported-by: Christoph Hellwig
Signed-off-by: Dan Williams
---
fs/ext4/e
The ->iomap_begin() operation is a hot path, so cache the
fs_dax_get_by_host() result at mount time to avoid the incurring the
hash lookup overhead on a per-i/o basis.
Cc: "Theodore Ts'o"
Cc: Andreas Dilger
Cc: Jan Kara
Reported-by: Christoph Hellwig
Signed-off-by: Dan Williams
---
fs/ext2/e
The ->iomap_begin() operation is a hot path, so cache the
fs_dax_get_by_host() result at mount time to avoid the incurring the
hash lookup overhead on a per-i/o basis.
Cc: "Darrick J. Wong"
Reported-by: Christoph Hellwig
Signed-off-by: Dan Williams
---
fs/xfs/xfs_aops.c | 13 +
Changes since v2 [1]:
* Split fs_dax_get_by_bdev() to its own patch (Christoph)
* Push dax_device reference management into _{fill,put}_super()
rather than the generic vfs. (Christoph)
[1]: https://lists.01.org/pipermail/linux-nvdimm/2017-August/012133.html
---
Christoph notes:
I just no
Add a helper that can replace the following common pattern:
if (blk_queue_dax(bdev->bd_queue))
fs_dax_get_by_host(bdev->bd_disk->disk_name);
This will be used to move dax_device lookup from iomap-operation time to
fs-mount time.
Cc: Jan Kara
Cc: Christoph Hellwig
Signed
On Wed, Aug 30, 2017 at 11:19 AM, Robin Murphy wrote:
> mmio_flush_range() suffers from a lack of clearly-defined semantics,
> and is somewhat ambiguous to port to other architectures where the
> scope of the writeback implied by "flush" and ordering might matter,
> but MMIO would tend to imply no
On 08/30/2017 11:18 AM, Robin Murphy wrote:
> On 25/08/17 21:59, Dave Jiang wrote:
>> Adding a dmaengine transaction operation that allows copy to/from a
>> scatterlist and a flat buffer.
>
> Apologies if I'm late to the party, but doesn't DMA_SG already cover
> this use-case? As far as I can see,
mmio_flush_range() suffers from a lack of clearly-defined semantics,
and is somewhat ambiguous to port to other architectures where the
scope of the writeback implied by "flush" and ordering might matter,
but MMIO would tend to imply non-cacheable anyway. Per the rationale
in 67a3e8fe9015 ("nd_blk:
On 25/08/17 21:59, Dave Jiang wrote:
> Adding a dmaengine transaction operation that allows copy to/from a
> scatterlist and a flat buffer.
Apologies if I'm late to the party, but doesn't DMA_SG already cover
this use-case? As far as I can see, all this does is save the caller
from setting up a si
> The above patch fixes an issue with ext4 where executables cannot be run on
> read-only filesystems mounted with the DAX option.
>
> This issue does not appear to be present in ext2 or XFS, as they both pass
> the test. I've also confirmed outside of the test that they are both
> indeed able to
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
The old calculation assumed that the label space was 128k and the label
size is 128. With v1.2 labels where the label size is 256 this
calculation will return zero. We are saved by the fact that the
nsindex_size is always pre-initialized from a previous 128 byte
assumption and we are lucky that the
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
On Tue, Aug 29, 2017 at 04:37:15PM -0600, Ross Zwisler wrote:
> This adds a regression test for the following kernel patch:
>
> commit 42d4a99b09cb ("ext4: fix fault handling when mounted with -o dax,ro")
>
> The above patch fixes an issue with ext4 where executables cannot be run on
> read-only
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
61 matches
Mail list logo