This patchset converts NFS with fscache non-direct READ IO paths to use the netfs API with a non-invasive approach. The existing NFS pgio layer does not need extensive changes, and is the best way so far I've found to address Trond's concerns about modifying the IO path [1] as well as only enabling netfs when fscache is configured and enabled [2]. I have not attempted performance comparisions to address Chuck Lever's concern [3] because we are not converting the non-fscache enabled NFS IO paths to netfs. The main patch to be reviewed is patch #3 which converts nfs_read_folio and nfs_readahead. The patchset is based on 6.0-rc5 and has been pushed to github at: https://github.com/DaveWysochanskiRH/kernel/commits/nfs-fscache-netfs https://github.com/DaveWysochanskiRH/kernel/commit/7d57f6484951800de518f555a04a3a4dc4a1327c Changes since v6 ================ PATCH3: Fix generic/127 "Subreq overread" by calling netfs_subreq_terminated() with a length capped by sreq->len. This is related to the following patch which may cause us to read a full page even if the i_size does not extend to the end of the page: 8cfb9015280d NFS: Always provide aligned buffers to the RPC read layers Testing ======= The patches are fairly stable as evidenced with xfstests generic with various servers, both with and without fscache enabled, with no hangs or crashes seen: hammerspace(pNFS flexfiles): NFS4.1, NFS4.2 NetApp(pNFS filelayout): NFS3, NFS4.0, NFS4.1 RHEL8: NFS3,NFS4.1,NFS4.2 The known issues are as follows: 1. Unit test setting rsize < readahead does not properly read from fscache but re-reads data from the NFS server * This will be fixed with another linux-cachefs [4] patch to resolve "Stop read optimisation when folio removed from pagecache" * Daire Byrne also verified the patch fixes his issue as well 2. "Cache volume key already in use" after xfstest runs * xfstests (hammerspace with vers=4.2,fsc) shows the following on the console after some tests: "NFS: Cache volume key already in use (nfs,4.1,2,c50,cfe0100a,3,,,8000,100000,100000,bb8,ea60,7530,ea60,1)" * This may be fixed with another patch [5] that is in progress 3. Daire Byrne reported a NULL pointer oops at cachefiles_prepare_write+0x28/0x90 * may be a bug in netfs or cachefiles exposed by NFS patches * under investigation [58710.346376] BUG: kernel NULL pointer dereference, address: 0000000000000008 [58710.371212] CPU: 12 PID: 9134 Comm: kworker/u129:0 Tainted: G E 6.0.0-2.dneg.x86_64 #1 ... [58710.389995] Workqueue: events_unbound netfs_rreq_write_to_cache_work [netfs] [58710.397188] RIP: 0010:cachefiles_prepare_write+0x28/0x90 [cachefiles] ... [58710.500316] Call Trace: [58710.502894] <TASK> [58710.505126] netfs_rreq_write_to_cache_work+0x11c/0x320 [netfs] [58710.511201] process_one_work+0x217/0x3e0 [58710.515358] worker_thread+0x4a/0x3b0 [58710.519152] ? process_one_work+0x3e0/0x3e0 [58710.523467] kthread+0xd6/0x100 [58710.526740] ? kthread_complete_and_exit+0x20/0x20 [58710.531659] ret_from_fork+0x1f/0x30 Outstanding =========== Note that the existing NFS fscache stats ("fsc:" line in /proc/self/mountstats) as well as trace events still need removed. I've left these out of this patchset for now as removing them are benign and can come later (the stats will all be 0, and trace events are no longer used). The existing NFS fscache stat counts no longer apply since the new API is not page based - they are not meaningful or possible to obtain, and there are new fscache stats in /proc/fs/fscache/stats. A similar situation exists with the NFS trace events - netfs and fscache have plenty of trace events so the NFS specific ones probably are not needed. References ========== [1] https://lore.kernel.org/linux-nfs/9cfd5bc3cfc6abc2d3316b0387222e708d67f595.camel@xxxxxxxxxxxxxxx/ [2] https://lore.kernel.org/linux-nfs/da9200f1bded9b8b078a7aef227fd6b92eb028fb.camel@xxxxxxxxxxxxxxx/ [3] https://marc.info/?l=linux-nfs&m=160597917525083&w=4 [4] https://www.mail-archive.com/linux-cachefs@xxxxxxxxxx/msg03043.html [5] https://marc.info/?l=linux-nfs&m=165962662200679&w=4 Dave Wysochanski (3): NFS: Rename readpage_async_filler to nfs_pageio_add_page NFS: Configure support for netfs when NFS fscache is configured NFS: Convert buffered read paths to use netfs when fscache is enabled fs/nfs/Kconfig | 1 + fs/nfs/delegation.c | 2 +- fs/nfs/dir.c | 2 +- fs/nfs/fscache.c | 251 +++++++++++++++++++++++---------------- fs/nfs/fscache.h | 113 ++++++++++++------ fs/nfs/inode.c | 8 +- fs/nfs/internal.h | 11 +- fs/nfs/pagelist.c | 12 ++ fs/nfs/pnfs.c | 12 +- fs/nfs/read.c | 111 +++++++++-------- fs/nfs/write.c | 2 +- include/linux/nfs_fs.h | 34 ++++-- include/linux/nfs_page.h | 3 + include/linux/nfs_xdr.h | 3 + 14 files changed, 347 insertions(+), 218 deletions(-) -- 2.31.1
