getting rid of the last memory modifitions through gup(FOLL_GET)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi all,

we've made some nice progress on converting code that modifies user
memory to the pin_user_pages interface, especially though the work
from David Howells on iov_iter_extract_pages.  This thread tries to
coordinate on how to finish off this work.

The obvious next step is the remaining users of iov_iter_get_pages2
and iov_iter_get_pages_alloc2.  We have three file system direct I/O
users of those left: ceph, fuse and nfs.  Lei Huang has sent patches
to convert fuse to iov_iter_extract_pages which I'd love to see merged,
and we'd need equivalent work for ceph and nfs.

The non-file system uses are in the vmsplice code, which only reads
from the pages (but would still benefit from an iov_iter_extract_pages
conversion), and in net.  Out of the users in net, all but the 9p code
appear to be for reads from memory, so they don't pin even if a
conversion would be nice to retire iov_iter_get_pages* APIs.

After that we might have to do an audit of the raw get_user_pages APIs,
but there probably aren't many that modify file backed memory.

I'm also wondering what a good debug aid would be for detecting
writes to file backed memory without a previous reservation, but
everything either involves a page flag or file system code.  But if
someone has an idea I'm all ear as something mechanical to catch these
uses would be quite helpful.



[Index of Archives]     [CEPH Users]     [Ceph Large]     [Ceph Dev]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux