search for: dax_device

On Wed 10-04-19 09:38:24, Pankaj Gupta wrote: > This patch introduces 'daxdev_mapping_supported' helper > which checks if 'MAP_SYNC' is supported with filesystem > mapping. It also checks if corresponding dax_device is > synchronous. Virtio pmem device is asynchronous and > does not not support VM_SYNC. > > Suggested-by: Jan Kara <jack at suse.cz> > Signed-off-by: Pankaj Gupta <pagupta at redhat.com> > --- > include/linux/dax.h | 23 +++++++++++++++++++++++ > 1 file chang...

...*/ - dax_dev = alloc_dax(dev_dax, NULL, NULL); + dax_dev = alloc_dax(dev_dax, NULL, NULL, true); if (!dax_dev) goto err; diff --git a/drivers/dax/super.c b/drivers/dax/super.c index 0a339b85133e..bd6509308d05 100644 --- a/drivers/dax/super.c +++ b/drivers/dax/super.c @@ -186,6 +186,8 @@ enum dax_device_flags { DAXDEV_ALIVE, /* gate whether dax_flush() calls the low level flush routine */ DAXDEV_WRITE_CACHE, + /* flag to check if device supports synchronous flush */ + DAXDEV_SYNC, }; /** @@ -354,6 +356,12 @@ bool dax_write_cache_enabled(struct dax_device *dax_dev) } EXPORT_SYMBOL_GPL(d...

...dev = alloc_dax(dev_dax, NULL, NULL); + dax_dev = alloc_dax(dev_dax, NULL, NULL, DAXDEV_F_SYNC); if (!dax_dev) goto err; diff --git a/drivers/dax/super.c b/drivers/dax/super.c index 0a339b85133e..bd6509308d05 100644 --- a/drivers/dax/super.c +++ b/drivers/dax/super.c @@ -186,6 +186,8 @@ enum dax_device_flags { DAXDEV_ALIVE, /* gate whether dax_flush() calls the low level flush routine */ DAXDEV_WRITE_CACHE, + /* flag to check if device supports synchronous flush */ + DAXDEV_SYNC, }; /** @@ -354,6 +356,12 @@ bool dax_write_cache_enabled(struct dax_device *dax_dev) } EXPORT_SYMBOL_GPL(d...

...dev = alloc_dax(dev_dax, NULL, NULL); + dax_dev = alloc_dax(dev_dax, NULL, NULL, DAXDEV_F_SYNC); if (!dax_dev) goto err; diff --git a/drivers/dax/super.c b/drivers/dax/super.c index bbd57ca0634a..b6c44b5062e9 100644 --- a/drivers/dax/super.c +++ b/drivers/dax/super.c @@ -186,6 +186,8 @@ enum dax_device_flags { DAXDEV_ALIVE, /* gate whether dax_flush() calls the low level flush routine */ DAXDEV_WRITE_CACHE, + /* flag to check if device supports synchronous flush */ + DAXDEV_SYNC, }; /** @@ -354,6 +356,12 @@ bool dax_write_cache_enabled(struct dax_device *dax_dev) } EXPORT_SYMBOL_GPL(d...

On Tue, Apr 23, 2019 at 01:36:12PM +0530, Pankaj Gupta wrote: > Dont support 'MAP_SYNC' with non-DAX files and DAX files > with asynchronous dax_device. Virtio pmem provides > asynchronous host page cache flush mechanism. We don't > support 'MAP_SYNC' with virtio pmem and xfs. > > Signed-off-by: Pankaj Gupta <pagupta at redhat.com> > --- > fs/xfs/xfs_file.c | 10 ++++++---- > 1 file changed, 6 insertions(+)...

On Tue, Apr 23, 2019 at 01:36:12PM +0530, Pankaj Gupta wrote: > Dont support 'MAP_SYNC' with non-DAX files and DAX files > with asynchronous dax_device. Virtio pmem provides > asynchronous host page cache flush mechanism. We don't > support 'MAP_SYNC' with virtio pmem and xfs. > > Signed-off-by: Pankaj Gupta <pagupta at redhat.com> > --- > fs/xfs/xfs_file.c | 10 ++++++---- > 1 file changed, 6 insertions(+)...

...+1203,14 @@ xfs_file_mmap( > > > > if (!IS_DAX(file_inode(filp)) && (vma->vm_flags & VM_SYNC)) > > > > return -EOPNOTSUPP; > > > > > > > > + /* We don't support synchronous mappings with DAX files if > > > > + * dax_device is not synchronous. > > > > + */ > > > > + if (IS_DAX(file_inode(filp)) && !dax_synchronous( > > > > + xfs_find_daxdev_for_inode(file_inode(filp))) && > > > > + (vma->vm_flags & VM_SYNC)) > > > > + return -EOP...

...+1203,14 @@ xfs_file_mmap( > > > > if (!IS_DAX(file_inode(filp)) && (vma->vm_flags & VM_SYNC)) > > > > return -EOPNOTSUPP; > > > > > > > > + /* We don't support synchronous mappings with DAX files if > > > > + * dax_device is not synchronous. > > > > + */ > > > > + if (IS_DAX(file_inode(filp)) && !dax_synchronous( > > > > + xfs_find_daxdev_for_inode(file_inode(filp))) && > > > > + (vma->vm_flags & VM_SYNC)) > > > > + return -EOP...

This patch series has implementation for "virtio pmem". "virtio pmem" is fake persistent memory(nvdimm) in guest which allows to bypass the guest page cache. This also implements a VIRTIO based asynchronous flush mechanism. Sharing guest kernel driver in this patchset with the changes suggested in v4. Tested with Qemu side device emulation [6] for virtio-pmem.

This patch introduces 'daxdev_mapping_supported' helper which checks if 'MAP_SYNC' is supported with filesystem mapping. It also checks if corresponding dax_device is synchronous. Virtio pmem device is asynchronous and does not not support VM_SYNC. Suggested-by: Jan Kara <jack at suse.cz> Signed-off-by: Pankaj Gupta <pagupta at redhat.com> --- include/linux/dax.h | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/inclu...

This patch introduces 'daxdev_mapping_supported' helper which checks if 'MAP_SYNC' is supported with filesystem mapping. It also checks if corresponding dax_device is synchronous. Virtio pmem device is asynchronous and does not not support VM_SYNC. Suggested-by: Jan Kara <jack at suse.cz> Signed-off-by: Pankaj Gupta <pagupta at redhat.com> --- include/linux/dax.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/include/lin...

This patch introduces 'daxdev_mapping_supported' helper which checks if 'MAP_SYNC' is supported with filesystem mapping. It also checks if corresponding dax_device is synchronous. Virtio pmem device is asynchronous and does not not support VM_SYNC. Suggested-by: Jan Kara <jack at suse.cz> Signed-off-by: Pankaj Gupta <pagupta at redhat.com> Reviewed-by: Jan Kara <jack at suse.cz> --- include/linux/dax.h | 17 +++++++++++++++++ 1 file chang...

...t; > if (!IS_DAX(file_inode(filp)) && (vma->vm_flags & VM_SYNC)) > > > > > > return -EOPNOTSUPP; > > > > > > > > > > > > + /* We don't support synchronous mappings with DAX files if > > > > > > + * dax_device is not synchronous. > > > > > > + */ > > > > > > + if (IS_DAX(file_inode(filp)) && !dax_synchronous( > > > > > > + xfs_find_daxdev_for_inode(file_inode(filp))) && > > > > > > + (vma->vm_flags & VM_S...

...t; > if (!IS_DAX(file_inode(filp)) && (vma->vm_flags & VM_SYNC)) > > > > > > return -EOPNOTSUPP; > > > > > > > > > > > > + /* We don't support synchronous mappings with DAX files if > > > > > > + * dax_device is not synchronous. > > > > > > + */ > > > > > > + if (IS_DAX(file_inode(filp)) && !dax_synchronous( > > > > > > + xfs_find_daxdev_for_inode(file_inode(filp))) && > > > > > > + (vma->vm_flags & VM_S...

On Thu, Apr 25, 2019 at 10:03 PM Pankaj Gupta <pagupta at redhat.com> wrote: > > Dont support 'MAP_SYNC' with non-DAX files and DAX files > with asynchronous dax_device. Virtio pmem provides > asynchronous host page cache flush mechanism. We don't > support 'MAP_SYNC' with virtio pmem and xfs. > > Signed-off-by: Pankaj Gupta <pagupta at redhat.com> > --- > fs/xfs/xfs_file.c | 9 ++++++--- > 1 file changed, 6 insertions(+), 3...

On Thu, Apr 25, 2019 at 10:03 PM Pankaj Gupta <pagupta at redhat.com> wrote: > > Dont support 'MAP_SYNC' with non-DAX files and DAX files > with asynchronous dax_device. Virtio pmem provides > asynchronous host page cache flush mechanism. We don't > support 'MAP_SYNC' with virtio pmem and xfs. > > Signed-off-by: Pankaj Gupta <pagupta at redhat.com> > --- > fs/xfs/xfs_file.c | 9 ++++++--- > 1 file changed, 6 insertions(+), 3...

Hi Dan, This series has only change in patch 2 for linux-next build failure. There is no functional change. Keeping all the existing review/acks and reposting the patch series for merging via libnvdimm tree. --- This patch series has implementation for "virtio pmem". "virtio pmem" is fake persistent memory(nvdimm) in guest which allows to bypass the guest page

This patch series is ready to be merged via nvdimm tree as discussed with Dan. We have ack/review on XFS, EXT4 device mapper & VIRTIO patches. This version has fix for test bot build failure. Keeping all the existing r-o-bs. Jakob CCed also tested the patch series and confirmed the working of v9. --- This patch series has implementation for "virtio pmem". "virtio

This patch series is ready to be merged via nvdimm tree as discussed with Dan. We have ack/review on XFS, EXT4 & VIRTIO patches. Need an ack on device mapper change in patch 4. Mike, Can you please review and ack patch4. This version does not has any additonal code change from v10 and is only rebase of v10 on Linux 5.2-rc4 which is required for patch4. Keeping all the existing

This patch series is ready to be merged via nvdimm tree as discussed with Dan. We have ack/review on XFS, EXT4 & VIRTIO patches. Need an ack on device mapper change in patch 4. Mike, Can you please review patch 4 which has change for dax with device mapper. Incorporated all the changes suggested in v9. This version has minor changes in patch 2(virtio) and does not change the