Providing mechanism to clear poison list via the ndctl ND_CMD_CLEAR_ERROR
call. We will update the poison list and also the badblocks at region level
if the region is in dax mode or in pmem mode and not active.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/acpi/nfit/
Make sure that xlat_status is unconditionally called.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/acpi/nfit/core.c | 15 +++
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index 662036b..e
Adding sysfs attribute in order to export the physical address of the
ND region. This is for supporting of user app poison clear via
device dax.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Reviewed-by: Johannes Thumshirn <jthumsh...@suse.de>
---
drivers/nvdimm/region_d
badblocks sysfs file will be export at nd_region level. When nvdimm event
notifier happens for NVDIMM_REVALIATE_POISON, the badblocks in the nd
region will be updated.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Reviewed-by: Johannes Thumshirn <jthumsh...@suse.de>
---
drivers
badblocks.
v2:
Addressed formatting comments by Johannes.
---
Dave Jiang (5):
libnvdimm: Add mechanism to publish badblocks at nd region level
libnvdimm: Add resource sysfs attrib to nd region
acpi: cleanup acpi_nfit_ctl calling xlat_status
acpi_nfit, libnvdimm: Add support
On 03/07/2017 02:30 AM, Johannes Thumshirn wrote:
> On 03/06/2017 09:32 PM, Dave Jiang wrote:
>> Providing mechanism to clear poison list via the ndctl ND_CMD_CLEAR_ERROR
>> call. We will update the poison list and also the badblocks at region level
>> if the region is in dax
Adding unit test for device DAX 1G PUD support. This is patterned after
device-dax.c that tests the PMD support.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
test/Makefile.am | 17 +++
test/device-dax-1g.c | 336 ++
2 files c
With the addition of support to clear poison list and badblocks on a
per region level, the bus checking functions will be moved to
devdax-error test to test directly on device dax. The command issued to
active pmem are expected to fail.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
Adding helper function to acquire the physical base address of the
ND region.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
ndctl/lib/libndctl.c | 21 +
ndctl/lib/libndctl.sym |1 +
ndctl/libndctl.h.in|1 +
3 files changed, 23 insertions(+)
diff
badblocks.
---
Dave Jiang (5):
libnvdimm: Add mechanism to publish badblocks at nd region level
libnvdimm: Add resource sysfs attrib to nd region
acpi: cleanup acpi_nfit_ctl calling xlat_status
acpi_nfit, libnvdimm: Add support for clear poison list and bad blocks
Adding sysfs attribute in order to export the physical address of the
ND region. This is for supporting of user app poison clear via
device dax.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/region_devs.c | 13 +
1 file changed, 13 insertions(+)
diff
badblocks sysfs file will be export at nd_region level. When nvdimm event
notifier happens for NVDIMM_REVALIATE_POISON, the badblocks in the nd
region will be updated.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/nd.h |1 +
drivers/nvdimm/region.c
Adding dax support to the nvdimm testing in tools/testing/nvdimm. The
memory allocated by the tool is via vmalloc and non-contiguous.
Overriding pgoff_to_phys() call to support the vmalloc memory.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax-private.h
Providing mechanism to clear poison list via the ndctl ND_CMD_CLEAR_ERROR
call. We will update the poison list and also the badblocks at region level
if the region is in dax mode or in pmem mode and not active.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/acpi/nfit/
Make sure that xlat_status is unconditionally called.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/acpi/nfit/core.c | 15 +++
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index 7361d00..9
by
commit a2d581675d48 ("mm,fs,dax: change ->pmd_fault to ->huge_fault")
which missed fixing up the !CONFIG_FS_DAX_PMD case.
Reported-by: Stephen Rothwell <s...@canb.auug.org.au>
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
fs/dax.c |3 ++-
1 file changed,
On 02/16/2017 08:42 AM, Dave Hansen wrote:
> On 02/15/2017 12:31 PM, Dave Jiang wrote:
>> The fix introduced by e4decc90 to fix the UP case for 32bit x86, however
>> that broke the SMP case that was working previously. Add ifdef so the dummy
>> function only show up f
<alexander.kaps...@gmail.com>
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
arch/x86/include/asm/pgtable-3level.h |2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/x86/include/asm/pgtable-3level.h
b/arch/x86/include/asm/pgtable-3level.h
index 50d35e3..8f50fb3 100644
--
the code has a mix of partial from before and after the
patch. I'm rather confused about it
>
> url:
> https://github.com/0day-ci/linux/commits/Dave-Jiang/mm-replace-FAULT_FLAG_SIZE-with-parameter-to-huge_fault/20170204-053548
> base: git://git.cmpxchg.org/linux-mmotm.git ma
an input parameter to huge_fault that indicates
the size of the page entry. This makes the code easier to trace and
should avoid the issues we see with the fault flags where removal of the
flag was necessary in the fallback paths.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drive
On 01/30/2017 04:43 PM, Ross Zwisler wrote:
> On Thu, Jan 26, 2017 at 10:09:47AM -0700, Dave Jiang wrote:
>> In preparation for adding the ability to handle PUD pages, convert
>> ->pmd_fault to ->huge_fault. The vm_fault structure is extended to
>> include a union
On 01/26/2017 03:38 PM, Andrew Morton wrote:
> On Thu, 26 Jan 2017 10:09:53 -0700 Dave Jiang <dave.ji...@intel.com> wrote:
>
>> The current transparent hugepage code only supports PMDs. This patch
>> adds support for transparent use of PUDs with DAX. It does not include
.
[DJ: Forward ported to 4.10-rc]
Signed-off-by: Matthew Wilcox <mawil...@microsoft.com>
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax.c | 34 +-
fs/dax.c| 43 ++-
fs/ext2/
Adding transparent huge PUD pages support for device DAX by adding a
pud_fault handler.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax.c | 48
1 file changed, 48 insertions(+)
diff --git a/drivers/dax/dax.c b/drive
will keep
increasing.
An argument can be made to convert the applications from process
model to thread model, but in the real world that may not be
always practical.
Hopefully this helps explain the use case where this is valuable.
v2: Fixup build issues from 0-day build.
---
Dave Jiang (1
__GFP_IO got accidentally dropped in the dax fault handler during the
vmf parameter conversion. Putting it back.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
fs/dax.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/fs/dax.c b/fs/dax.c
index 2e90f7a..20e9db8 100644
--- a/fs
With only vmf being passed in, the pgoff in vmf that is passed in is for
the PTE and not the PMD. Fix up so the pgoff is corrected.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
include/trace/events/fs_dax.h |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/i
Adding transparent huge PUD pages support for device DAX by adding a
pud_fault handler.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax.c | 48
1 file changed, 48 insertions(+)
diff --git a/drivers/dax/dax.c b/drive
.
[DJ: Forward ported to 4.10-rc]
Signed-off-by: Matthew Wilcox <mawil...@microsoft.com>
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax.c | 34 +-
fs/dax.c| 43 ++-
fs/ext2/
f locking the PUD before calling ->pud_walk,
so handlers do not need to worry whether the PUD is stable.
[DJ: Forward ported to 4.10-rc]
Signed-off-by: Matthew Wilcox <mawil...@microsoft.com>
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
arch/Kconfig
code to support device DAX.
---
Dave Jiang (1):
dax: Support for transparent PUD pages for device DAX
Matthew Wilcox (2):
mm,fs,dax: Change ->pmd_fault to ->huge_fault
mm,x86: Add support for PUD-sized transparent hugepages
arch/Kconfig |3
->fault(), ->page_mkwrite(), and ->pfn_mkwrite() calls do not need to take
a vma and vmf parameter when the vma already resides in vmf. Remove the vma
parameter to simplify things.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Reviewed-by: Ross Zwisler <ross.zwis.
pmd_fault() and related functions really only need the vmf parameter since
the additional parameters are all included in the vmf struct. Removing
additional parameter and simplify pmd_fault() and friends.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Reviewed-by: Ross Zwisler <
Instead of passing in multiple parameters in the pmd_fault() handler,
a vmf can be passed in just like a fault() handler. This will simplify
code and remove the need for the actual pmd fault handlers to allocate a
vmf. Related functions are also modified to do the same.
Signed-off-by: Dave Jiang
On 01/17/2017 11:13 AM, Dave Jiang wrote:
>
>
> On 01/17/2017 09:05 AM, Dave Jiang wrote:
>> On 01/16/2017 03:45 AM, Stephen Rothwell wrote:
>>> Hi Xiong,
>>>
>>> Just cc'ing Andrew.
>>>
>>> On Mon, 16 Jan 2017 17:29:16 +0800 Xiong Z
On 01/17/2017 09:05 AM, Dave Jiang wrote:
> On 01/16/2017 03:45 AM, Stephen Rothwell wrote:
>> Hi Xiong,
>>
>> Just cc'ing Andrew.
>>
>> On Mon, 16 Jan 2017 17:29:16 +0800 Xiong Zhou <xz...@redhat.com> wrote:
>>>
>>> On Fri, Ja
On 01/16/2017 03:45 AM, Stephen Rothwell wrote:
> Hi Xiong,
>
> Just cc'ing Andrew.
>
> On Mon, 16 Jan 2017 17:29:16 +0800 Xiong Zhou wrote:
>>
>> On Fri, Jan 13, 2017 at 06:16:41PM +0800, Xiong Zhou wrote:
>>> Hi,
>>>
>>> These cases "hang" when testing with -o dax mount
. The nn@ss ranges will be skipped by the
mem_avoid set since it indicates memory useable.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Acked-by: Kees Cook <keesc...@chromium.org>
Acked-by: Baoquan He <b...@redhat.com>
---
arch/x86/boot/boot.h |1
arch/x
On 01/11/2017 05:00 AM, Thomas Gleixner wrote:
> On Tue, 10 Jan 2017, Dave Jiang wrote:
>> +unsigned long simple_strtoul(const char *cp, char **endp, unsigned int
>> base);
>> +long simple_strtol(const char *cp, char **endp, unsigned int base);
>
&g
Clearing out the poison in the metadata block of the namespace before
we use it. Range from start + 8k to pfn_sb->dataoff.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/pfn_devs.c | 25 +
1 file changed, 25 insertions(+)
diff --git
On 12/19/2016 11:27 AM, Dave Jiang wrote:
> Instead of passing in multiple parameters in the pmd_fault() handler,
> a vmf can be passed in just like a fault() handler. This will simplify
> code and remove the need for the actual pmd fault handlers to allocate a
> vmf. Related functi
. The nn@ss ranges will be skipped by the
mem_avoid set since it indicates memory useable.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
v2:
Addressing comments from Ingo.
- Handle entire list of memmaps
v3:
Fix 32bit build issue
v4:
Addressing comments from Baoquan
- Not exclude nn@ss
On 01/03/2017 07:37 PM, Baoquan He wrote:
> Hi Dave,
>
> I have several concerns, please see the inline comments.
>
> On 01/03/17 at 01:48pm, Dave Jiang wrote:
>> CONFIG_RANDOMIZE_BASE relocates the kernel to a random base address.
>> However it does not take into ac
Adding fallocate support to device-dax. This implements FALLOC_FL_PUNCH_HOLE
in order to allow clearing of badblocks/poison list.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax-private.h |1 +
drivers/dax/dax.c | 24 +++-
drivers/dax
pmd_fault() and related functions really only need the vmf parameter since
the additional parameters are all included in the vmf struct. Removing
additional parameter and simplify pmd_fault() and friends.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Reviewed-by: Ross Zwisler <
Instead of passing in multiple parameters in the pmd_fault() handler,
a vmf can be passed in just like a fault() handler. This will simplify
code and remove the need for the actual pmd fault handlers to allocate a
vmf. Related functions are also modified to do the same.
Signed-off-by: Dave Jiang
On 11/29/2016 12:26 PM, Dave Jiang wrote:
> CONFIG_RANDOMIZE_BASE relocates the kernel to a random base address.
> However it does not take into account the memmap= parameter passed in from
> the kernel cmdline. This results in the kernel sometimes being put in
> the middle of the
pmd_fault() and related functions really only need the vmf parameter since
the additional parameters are all included in the vmf struct. Removing
additional parameter and simplify pmd_fault() and friends.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax.c
On 12/14/2016 02:12 PM, Jeff Moyer wrote:
> Dave Jiang <dave.ji...@intel.com> writes:
>
>> Adding support to show badblocks in the pmem region that's provided
>> by the poison_list. This should show up in
>> /sys/class/dax/daxN.N/badblocks as read only. Current
On 12/13/2016 05:15 AM, Jan Kara wrote:
> On Thu 08-12-16 14:34:21, Dave Jiang wrote:
>> Instead of passing in multiple parameters in the pmd_fault() handler,
>> a vmf can be passed in just like a fault() handler. This will simplify
>> code and remove the need for the actu
Adding dax support to the nvdimm testing in tools/testing/nvdimm. The
memory allocated by the tool is via vmalloc and non-contiguous.
Overriding pgoff_to_phys() call to support the vmalloc memory.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax-private.h
Adding fallocate support to device-dax. This implements FALLOC_FL_PUNCH_HOLE
in order to allow clearing of badblocks/poison list.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax-private.h |1 +
drivers/dax/dax.c | 28 +--
drivers/dax
.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax.c | 27 ++-
drivers/dax/dax.h |8 +++-
drivers/dax/pmem.c | 37 -
3 files changed, 65 insertions(+), 7 deletions(-)
diff --git a/drivers/dax/d
support to clear the errors.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
test/Makefile.am |8 ++-
test/device-dax-errors.c | 130 +
test/device-dax-errors.sh | 52 ++
3 files changed, 188 insertions(+), 2 del
-by: Dave Jiang <dave.ji...@intel.com>
---
ndctl/builtin-xaction-namespace.c | 34 +++---
util/size.h |1 +
2 files changed, 32 insertions(+), 3 deletions(-)
diff --git a/ndctl/builtin-xaction-namespace.c
b/ndctl/builtin-xaction-names
Instead of passing in multiple parameters in the pmd_fault() handler,
a vmf can be passed in just like a fault() handler. This will simplify
code and remove the need for the actual pmd fault handlers to allocate a
vmf. Related functions are also modified to do the same.
Signed-off-by: Dave Jiang
pmd_fault() and relate functions really only need the vmf parameter since
the additional parameters are all included in the vmf struct. Removing
additional parameter and simplify pmd_fault() and friends.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
Reviewed-by: Ross Zwisler <
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
On 11/18/2016 10:07 AM, Dan Williams wrote:
> On Fri, Nov 18, 2016 at 8:47 AM, Dave Jiang <dave.ji...@intel.com>
> wrote:
>> -BEGIN PGP SIGNED MESSAGE- Hash: SHA256
>>
>>
>>
>> On 11/18/2016 0
Clearing out the poison in the metadata block of the namespace before
we use it. Range from start + 8k to pfn_sb->dataoff.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/pfn_devs.c | 30 ++
1 file changed, 30 insertions(+)
diff --git
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
On 11/17/2016 10:25 AM, Dave Jiang wrote:
> On 11/17/2016 09:35 AM, Kani, Toshimitsu wrote:
>> On Mon, 2016-11-14 at 17:21 -0700, Dave Jiang wrote:
>>> Clearing out the poison in the metadata block of the namespace
>>> b
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
On 11/18/2016 04:33 AM, Thomas Gleixner wrote:
> On Thu, 17 Nov 2016, Dave Jiang wrote:
>> CONFIG_RANDOMIZE_BASE can place the kernel anywhere. This causes
>> a problem for when memmap=nn!ss is used. This information is not
>&
is not
reliable in the presence of CONFIG_RANDOMIZE_BASE.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
Documentation/kernel-parameters.txt |5 -
arch/x86/kernel/e820.c |2 ++
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/Documentation/kernel-parameters.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
On 11/17/2016 09:35 AM, Kani, Toshimitsu wrote:
> On Mon, 2016-11-14 at 17:21 -0700, Dave Jiang wrote:
>> Clearing out the poison in the metadata block of the namespace
>> before we use it. Range from start + 8k to pfn_sb->dataoff.
We need to clear any poison when we are writing to pmem. The granularity
will be sector size. If it's less then we can't do anything about it
barring corruption.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/claim.c | 30 ++
1 file chang
We need to clear any poison when we are writing to pmem. The granularity
will be sector size. If it's less then we can't do anything about it
barring corruption.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/claim.c | 24 +---
1 file chang
We need to clear any poison when we are writing to pmem. The granularity
will be sector size. If it's less then we can't do anything about it
barring corruption.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/claim.c | 20 ++--
1 file changed, 18 inse
-by: Dave Jiang <dave.ji...@intel.com>
---
ndctl/builtin-xaction-namespace.c | 22 --
util/size.h |1 +
2 files changed, 21 insertions(+), 2 deletions(-)
diff --git a/ndctl/builtin-xaction-namespace.c
b/ndctl/builtin-xaction-namespace.c
.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/dax/dax.c | 25 +
drivers/dax/dax.h |8 +++-
drivers/dax/pmem.c | 37 -
3 files changed, 64 insertions(+), 6 deletions(-)
diff --git a/drivers/dax/d
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
-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/md/dm-flakey.c | 21 -
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c
index 6a2e8dd..cf182a7 100644
--- a/drivers/md/dm-flakey.c
+++ b/drivers/md/dm-fl
On 09/18/2016 09:42 AM, Elliott, Robert (Persistent Memory) wrote:
>
>
>> -Original Message-----
>> From: Dave Jiang [mailto:dave.ji...@intel.com]
>> Sent: Friday, September 16, 2016 12:24 PM
> ...
>> Subject: Re: [PATCH] ndctl: move test/dax-errors
On 09/15/2016 06:18 PM, Elliott, Robert (Persistent Memory) wrote:
>
>
>> -Original Message-
>> From: Linux-nvdimm [mailto:linux-nvdimm-boun...@lists.01.org] On
>> Behalf Of Dave Jiang
>> Sent: Thursday, September 15, 2016 5:28 PM
>> To: vishal.l.ve...
On 09/15/2016 06:18 PM, Elliott, Robert (Persistent Memory) wrote:
>
>
>> -Original Message-
>> From: Linux-nvdimm [mailto:linux-nvdimm-boun...@lists.01.org] On
>> Behalf Of Dave Jiang
>> Sent: Thursday, September 15, 2016 5:28 PM
>> To: vishal.l
need to just return as handled and allow the bad blocks to
be cleared rather than fail.
Signed-off-by: Dave Jiang <dave.ji...@intel.com>
---
drivers/nvdimm/bus.c |6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/nvdimm/bus.c b/drivers/nvdimm/bus.c
index 4
801 - 874 of 874 matches
Mail list logo