linux-toradex.git/drivers/nvdimm, branch v6.6-rc3

Merge tag 'libnvdimm-for-6.6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

2023-08-31T03:52:08+00:00

Pull nvdimm updates from Dave Jiang:
 "This is mostly small cleanups, fixes, and with a change to prevent
  zero-sized namespace exposed to user for nvdimm.

  Summary:

   - kstrtobool() conversion for nvdimm

   - Add REQ_OP_WRITE for virtio_pmem

   - Header files update for of_pmem

   - Restrict zero-sized namespace from being exposed to user

   - Avoid unnecessary endian conversion

   - Fix mem leak in nvdimm pmu

   - Fix dereference after free in nvdimm pmu"

* tag 'libnvdimm-for-6.6' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  nvdimm: Fix dereference after free in register_nvdimm_pmu()
  nvdimm: Fix memleak of pmu attr_groups in unregister_nvdimm_pmu()
  nvdimm/pfn_dev: Avoid unnecessary endian conversion
  nvdimm/pfn_dev: Prevent the creation of zero-sized namespaces
  nvdimm: Explicitly include correct DT includes
  virtio_pmem: add the missing REQ_OP_WRITE for flush bio
  nvdimm: Use kstrtobool() instead of strtobool()

mm/hugepage pud: allow arch-specific helper function to check huge page pud support

2023-08-18T17:12:53+00:00

Patch series "Add support for DAX vmemmap optimization for ppc64", v6.

This patch series implements changes required to support DAX vmemmap
optimization for ppc64.  The vmemmap optimization is only enabled with
radix MMU translation and 1GB PUD mapping with 64K page size.

The patch series also splits the hugetlb vmemmap optimization as a
separate Kconfig variable so that architectures can enable DAX vmemmap
optimization without enabling hugetlb vmemmap optimization.  This should
enable architectures like arm64 to enable DAX vmemmap optimization while
they can't enable hugetlb vmemmap optimization.  More details of the same
are in patch "mm/vmemmap optimization: Split hugetlb and devdax vmemmap
optimization".

With 64K page size for 16384 pages added (1G) we save 14 pages
With 4K page size for 262144 pages added (1G) we save 4094 pages
With 4K page size for 512 pages added (2M) we save 6 pages


This patch (of 13):

Architectures like powerpc would like to enable transparent huge page pud
support only with radix translation.  To support that add
has_transparent_pud_hugepage() helper that architectures can override.

[aneesh.kumar@linux.ibm.com: use the new has_transparent_pud_hugepage()]
  Link: https://lkml.kernel.org/r/87tttrvtaj.fsf@linux.ibm.com
Link: https://lkml.kernel.org/r/20230724190759.483013-1-aneesh.kumar@linux.ibm.com
Link: https://lkml.kernel.org/r/20230724190759.483013-2-aneesh.kumar@linux.ibm.com
Signed-off-by: Aneesh Kumar K.V 
Reviewed-by: Christophe Leroy 
Cc: Catalin Marinas 
Cc: Dan Williams 
Cc: Joao Martins 
Cc: Michael Ellerman 
Cc: Mike Kravetz 
Cc: Muchun Song 
Cc: Nicholas Piggin 
Cc: Oscar Salvador 
Cc: Will Deacon 
Signed-off-by: Andrew Morton

nvdimm: Fix dereference after free in register_nvdimm_pmu()

2023-08-17T16:34:03+00:00

'nd_pmu->pmu.attr_groups' is dereferenced in function
'nvdimm_pmu_free_hotplug_memory' call after it has been freed. Because in
function 'nvdimm_pmu_free_hotplug_memory' memory pointed by the fields of
'nd_pmu->pmu.attr_groups' is deallocated it is necessary to call 'kfree'
after 'nvdimm_pmu_free_hotplug_memory'.

Fixes: 0fab1ba6ad6b ("drivers/nvdimm: Add perf interface to expose nvdimm performance stats")
Co-developed-by: Ivanov Mikhail 
Signed-off-by: Konstantin Meskhidze 
Reviewed-by: Jeff Moyer 
Link: https://lore.kernel.org/r/20230817114103.754977-1-konstantin.meskhidze@huawei.com
Signed-off-by: Dave Jiang

nvdimm: Fix memleak of pmu attr_groups in unregister_nvdimm_pmu()

2023-08-17T16:34:02+00:00

Memory pointed by 'nd_pmu->pmu.attr_groups' is allocated in function
'register_nvdimm_pmu' and is lost after 'kfree(nd_pmu)' call in function
'unregister_nvdimm_pmu'.

Fixes: 0fab1ba6ad6b ("drivers/nvdimm: Add perf interface to expose nvdimm performance stats")
Co-developed-by: Ivanov Mikhail 
Signed-off-by: Konstantin Meskhidze 
Reviewed-by: Jeff Moyer 
Link: https://lore.kernel.org/r/20230817115945.771826-1-konstantin.meskhidze@huawei.com
Signed-off-by: Dave Jiang

nvdimm/pfn_dev: Avoid unnecessary endian conversion

2023-08-11T16:15:05+00:00

use the local variable that already have the converted values.

No functional change in this patch.

Reviewed-by: Jeff Moyer 
Signed-off-by: Aneesh Kumar K.V 
Link: https://lore.kernel.org/r/20230809053512.350660-2-aneesh.kumar@linux.ibm.com
Signed-off-by: Dave Jiang

nvdimm/pfn_dev: Prevent the creation of zero-sized namespaces

2023-08-11T16:15:04+00:00

On architectures that have different page size values used for kernel
direct mapping and userspace mappings, the user can end up creating zero-sized
namespaces as shown below

:/sys/bus/nd/devices/region1# cat align
0x1000000
/sys/bus/nd/devices/region1# echo 0x200000 > align
/sys/bus/nd/devices/region1/dax1.0# cat supported_alignments
65536 16777216
 $ ndctl create-namespace -r region1 -m devdax -s 18M --align 64K
{
  "dev":"namespace1.0",
  "mode":"devdax",
  "map":"dev",
  "size":0,
  "uuid":"3094329a-0c66-4905-847e-357223e56ab0",
  "daxregion":{
    "id":1,
    "size":0,
    "align":65536
  },
  "align":65536
}
similarily for fsdax

 $ ndctl create-namespace -r region1 -m fsdax  -s 18M --align 64K
{
  "dev":"namespace1.0",
  "mode":"fsdax",
  "map":"dev",
  "size":0,
  "uuid":"45538a6f-dec7-405d-b1da-2a4075e06232",
  "sector_size":512,
  "align":65536,
  "blockdev":"pmem1"
}

In commit 9ffc1d19fc4a ("mm/memremap_pages: Introduce memremap_compat_align()")
memremap_compat_align was added to make sure the kernel always creates
namespaces with 16MB alignment. But the user can still override the
region alignment and no input validation is done in the region alignment
values to retain the flexibility user had before. However, the kernel
ensures that only part of the namespace that can be mapped via kernel
direct mapping page size is enabled. This is achieved by tracking the
unmapped part of the namespace in pfn_sb->end_trunc. The kernel also
ensures that the start address of the namespace is also aligned to the
kernel direct mapping page size.

Depending on the user request, the kernel implements userspace mapping
alignment by updating pfn device alignment attribute and this value is
used to adjust the start address for userspace mappings. This is tracked
in pfn_sb->dataoff. Hence the available size for userspace mapping is:

usermapping_size = size of the namespace - pfn_sb->end_trun - pfn_sb->dataoff

If the kernel finds the user mapping size zero then don't allow the
creation of namespace.

After fix:
$ ndctl create-namespace -f  -r region1 -m devdax  -s 18M --align 64K
libndctl: ndctl_dax_enable: dax1.1: failed to enable
  Error: namespace1.2: failed to enable

failed to create namespace: No such device or address

And existing zero sized namespace will be marked disabled.
root@ltczz75-lp2:/home/kvaneesh# ndctl  list -N -r region1 -i
[
  {
    "dev":"namespace1.0",
    "mode":"raw",
    "size":18874368,
    "uuid":"94a90fb0-8e78-4fb6-a759-ffc62f9fa181",
    "sector_size":512,
    "state":"disabled"
  },

Signed-off-by: Aneesh Kumar K.V 
Reviewed-by: Jeff Moyer 
Link: https://lore.kernel.org/r/20230809053512.350660-1-aneesh.kumar@linux.ibm.com
Signed-off-by: Dave Jiang

nvdimm: Explicitly include correct DT includes

2023-07-19T15:45:50+00:00

The DT of_device.h and of_platform.h date back to the separate
of_platform_bus_type before it as merged into the regular platform bus.
As part of that merge prepping Arm DT support 13 years ago, they
"temporarily" include each other. They also include platform_device.h
and of.h. As a result, there's a pretty much random mix of those include
files used throughout the tree. In order to detangle these headers and
replace the implicit includes with struct declarations, users need to
explicitly include the correct includes.

Signed-off-by: Rob Herring 
Signed-off-by: Dave Jiang

virtio_pmem: add the missing REQ_OP_WRITE for flush bio

2023-07-19T15:44:36+00:00

When doing mkfs.xfs on a pmem device, the following warning was
reported:

 ------------[ cut here ]------------
 WARNING: CPU: 2 PID: 384 at block/blk-core.c:751 submit_bio_noacct
 Modules linked in:
 CPU: 2 PID: 384 Comm: mkfs.xfs Not tainted 6.4.0-rc7+ #154
 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
 RIP: 0010:submit_bio_noacct+0x340/0x520
 ......
 Call Trace:
  
  ? submit_bio_noacct+0xd5/0x520
  submit_bio+0x37/0x60
  async_pmem_flush+0x79/0xa0
  nvdimm_flush+0x17/0x40
  pmem_submit_bio+0x370/0x390
  __submit_bio+0xbc/0x190
  submit_bio_noacct_nocheck+0x14d/0x370
  submit_bio_noacct+0x1ef/0x520
  submit_bio+0x55/0x60
  submit_bio_wait+0x5a/0xc0
  blkdev_issue_flush+0x44/0x60

The root cause is that submit_bio_noacct() needs bio_op() is either
WRITE or ZONE_APPEND for flush bio and async_pmem_flush() doesn't assign
REQ_OP_WRITE when allocating flush bio, so submit_bio_noacct just fail
the flush bio.

Simply fix it by adding the missing REQ_OP_WRITE for flush bio. And we
could fix the flush order issue and do flush optimization later.

Cc: stable@vger.kernel.org # 6.3+
Fixes: b4a6bb3a67aa ("block: add a sanity check for non-write flush/fua bios")
Reviewed-by: Christoph Hellwig 
Reviewed-by: Chaitanya Kulkarni 
Reviewed-by: Pankaj Gupta 
Tested-by: Pankaj Gupta 
Signed-off-by: Hou Tao 
Signed-off-by: Dave Jiang

nvdimm: Use kstrtobool() instead of strtobool()

2023-07-19T15:42:50+00:00

strtobool() is the same as kstrtobool().
However, the latter is more used within the kernel.

In order to remove strtobool() and slightly simplify kstrtox.h, switch to
the other function name.

While at it, include the corresponding header file ()

Reviewed-by: Vishal Verma 
Signed-off-by: Christophe JAILLET 
Signed-off-by: Dave Jiang

dax: enable dax fault handler to report VM_FAULT_HWPOISON

2023-06-26T13:54:23+00:00

When multiple processes mmap() a dax file, then at some point,
a process issues a 'load' and consumes a hwpoison, the process
receives a SIGBUS with si_code = BUS_MCEERR_AR and with si_lsb
set for the poison scope. Soon after, any other process issues
a 'load' to the poisoned page (that is unmapped from the kernel
side by memory_failure), it receives a SIGBUS with
si_code = BUS_ADRERR and without valid si_lsb.

This is confusing to user, and is different from page fault due
to poison in RAM memory, also some helpful information is lost.

Channel dax backend driver's poison detection to the filesystem
such that instead of reporting VM_FAULT_SIGBUS, it could report
VM_FAULT_HWPOISON.

If user level block IO syscalls fail due to poison, the errno will
be converted to EIO to maintain block API consistency.

Signed-off-by: Jane Chu 
Link: https://lore.kernel.org/r/20230615181325.1327259-2-jane.chu@oracle.com
Reviewed-by: Dan Williams 
Signed-off-by: Vishal Verma