Paragon Software's NTFS driver can be adopted into the Linux 5.15 kernel

When discussing the recently published 27th edition of the patch set with the implementation of the NTFS file system from Paragon Software, Linus Torvalds said that he sees no obstacles to accepting this patch set in the next change acceptance window. Barring unforeseen issues, Paragon Software's NTFS support will be included in the 5.15 kernel, which is expected to be released in November.

During the time remaining before the patches are accepted into the kernel, Linus recommended rechecking the correctness of the signed-off signature in the patches, confirming the authorship of the transmitted code and the readiness to distribute it as part of the kernel under a free license. Also, Paragon Software is advised to make sure that the legal department is aware of all the consequences of transferring code under the GPLv2 license and understands the essence of this copyleft license.

The code for the new NTFS driver was opened by Paragon Software in August last year and differs from the driver already in the kernel by the ability to work in write mode. The old driver has not been updated for many years and is in a deplorable state. The new driver supports all the features of the current version of NTFS 3.1, including extended file attributes, data compression mode, effective work with voids in files, and playback of changes from the log to restore integrity after crashes.

In patch revision 27, Paragon Software adapted the driver for changes in the iov API by replacing the iov_iter_copy_from_user_atomic() call with copy_page_from_iter_atomic() and deprecating the iov_iter_advance() function. Of the recommendations made in the discussion, it remains only to translate the code to use fs / iomap, but this is not a mandatory requirement, but only a recommendation that can be implemented after inclusion in the kernel. In addition, Paragon Software confirmed that it is ready to maintain the proposed code in the kernel and plans to further transfer the implementation of journaling to work on top of the JBD (Journaling block device) available in the kernel, on the basis of which journaling is organized in ext3, ext4 and OCFS2.

Source: opennet.ru

Add a comment