diff options
Diffstat (limited to 'fs/ceph')
-rw-r--r-- | fs/ceph/addr.c | 588 | ||||
-rw-r--r-- | fs/ceph/cache.c | 237 | ||||
-rw-r--r-- | fs/ceph/cache.h | 103 | ||||
-rw-r--r-- | fs/ceph/caps.c | 251 | ||||
-rw-r--r-- | fs/ceph/debugfs.c | 166 | ||||
-rw-r--r-- | fs/ceph/dir.c | 17 | ||||
-rw-r--r-- | fs/ceph/export.c | 12 | ||||
-rw-r--r-- | fs/ceph/file.c | 264 | ||||
-rw-r--r-- | fs/ceph/inode.c | 149 | ||||
-rw-r--r-- | fs/ceph/locks.c | 17 | ||||
-rw-r--r-- | fs/ceph/mds_client.c | 217 | ||||
-rw-r--r-- | fs/ceph/mds_client.h | 15 | ||||
-rw-r--r-- | fs/ceph/mdsmap.c | 4 | ||||
-rw-r--r-- | fs/ceph/metric.c | 191 | ||||
-rw-r--r-- | fs/ceph/metric.h | 151 | ||||
-rw-r--r-- | fs/ceph/quota.c | 17 | ||||
-rw-r--r-- | fs/ceph/snap.c | 263 | ||||
-rw-r--r-- | fs/ceph/strings.c | 1 | ||||
-rw-r--r-- | fs/ceph/super.c | 204 | ||||
-rw-r--r-- | fs/ceph/super.h | 76 | ||||
-rw-r--r-- | fs/ceph/xattr.c | 16 |
21 files changed, 1745 insertions, 1214 deletions
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 99b80b5c7a93..b6edcf89a429 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -4,8 +4,8 @@ #include <linux/backing-dev.h> #include <linux/fs.h> #include <linux/mm.h> +#include <linux/swap.h> #include <linux/pagemap.h> -#include <linux/writeback.h> /* generic_writepages */ #include <linux/slab.h> #include <linux/pagevec.h> #include <linux/task_io_accounting_ops.h> @@ -63,7 +63,7 @@ (CONGESTION_ON_THRESH(congestion_kb) >> 2)) static int ceph_netfs_check_write_begin(struct file *file, loff_t pos, unsigned int len, - struct page *page, void **_fsdata); + struct folio *folio, void **_fsdata); static inline struct ceph_snap_context *page_snap_context(struct page *page) { @@ -76,18 +76,17 @@ static inline struct ceph_snap_context *page_snap_context(struct page *page) * Dirty a page. Optimistically adjust accounting, on the assumption * that we won't race with invalidate. If we do, readjust. */ -static int ceph_set_page_dirty(struct page *page) +static bool ceph_dirty_folio(struct address_space *mapping, struct folio *folio) { - struct address_space *mapping = page->mapping; struct inode *inode; struct ceph_inode_info *ci; struct ceph_snap_context *snapc; - if (PageDirty(page)) { - dout("%p set_page_dirty %p idx %lu -- already dirty\n", - mapping->host, page, page->index); - BUG_ON(!PagePrivate(page)); - return 0; + if (folio_test_dirty(folio)) { + dout("%p dirty_folio %p idx %lu -- already dirty\n", + mapping->host, folio, folio->index); + VM_BUG_ON_FOLIO(!folio_test_private(folio), folio); + return false; } inode = mapping->host; @@ -111,75 +110,81 @@ static int ceph_set_page_dirty(struct page *page) if (ci->i_wrbuffer_ref == 0) ihold(inode); ++ci->i_wrbuffer_ref; - dout("%p set_page_dirty %p idx %lu head %d/%d -> %d/%d " + dout("%p dirty_folio %p idx %lu head %d/%d -> %d/%d " "snapc %p seq %lld (%d snaps)\n", - mapping->host, page, page->index, + mapping->host, folio, folio->index, ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref_head-1, ci->i_wrbuffer_ref, ci->i_wrbuffer_ref_head, snapc, snapc->seq, snapc->num_snaps); spin_unlock(&ci->i_ceph_lock); /* - * Reference snap context in page->private. Also set - * PagePrivate so that we get invalidatepage callback. + * Reference snap context in folio->private. Also set + * PagePrivate so that we get invalidate_folio callback. */ - BUG_ON(PagePrivate(page)); - attach_page_private(page, snapc); + VM_BUG_ON_FOLIO(folio_test_private(folio), folio); + folio_attach_private(folio, snapc); - return __set_page_dirty_nobuffers(page); + return ceph_fscache_dirty_folio(mapping, folio); } /* - * If we are truncating the full page (i.e. offset == 0), adjust the - * dirty page counters appropriately. Only called if there is private - * data on the page. + * If we are truncating the full folio (i.e. offset == 0), adjust the + * dirty folio counters appropriately. Only called if there is private + * data on the folio. */ -static void ceph_invalidatepage(struct page *page, unsigned int offset, - unsigned int length) +static void ceph_invalidate_folio(struct folio *folio, size_t offset, + size_t length) { struct inode *inode; struct ceph_inode_info *ci; struct ceph_snap_context *snapc; - wait_on_page_fscache(page); - - inode = page->mapping->host; + inode = folio->mapping->host; ci = ceph_inode(inode); - if (offset != 0 || length != thp_size(page)) { - dout("%p invalidatepage %p idx %lu partial dirty page %u~%u\n", - inode, page, page->index, offset, length); + if (offset != 0 || length != folio_size(folio)) { + dout("%p invalidate_folio idx %lu partial dirty page %zu~%zu\n", + inode, folio->index, offset, length); return; } - WARN_ON(!PageLocked(page)); - if (!PagePrivate(page)) - return; + WARN_ON(!folio_test_locked(folio)); + if (folio_test_private(folio)) { + dout("%p invalidate_folio idx %lu full dirty page\n", + inode, folio->index); - dout("%p invalidatepage %p idx %lu full dirty page\n", - inode, page, page->index); + snapc = folio_detach_private(folio); + ceph_put_wrbuffer_cap_refs(ci, 1, snapc); + ceph_put_snap_context(snapc); + } - snapc = detach_page_private(page); - ceph_put_wrbuffer_cap_refs(ci, 1, snapc); - ceph_put_snap_context(snapc); + folio_wait_fscache(folio); } static int ceph_releasepage(struct page *page, gfp_t gfp) { - dout("%p releasepage %p idx %lu (%sdirty)\n", page->mapping->host, - page, page->index, PageDirty(page) ? "" : "not "); + struct inode *inode = page->mapping->host; + + dout("%llx:%llx releasepage %p idx %lu (%sdirty)\n", + ceph_vinop(inode), page, + page->index, PageDirty(page) ? "" : "not "); + + if (PagePrivate(page)) + return 0; if (PageFsCache(page)) { - if (!(gfp & __GFP_DIRECT_RECLAIM) || !(gfp & __GFP_FS)) + if (current_is_kswapd() || !(gfp & __GFP_FS)) return 0; wait_on_page_fscache(page); } - return !PagePrivate(page); + ceph_fscache_note_page_release(inode); + return 1; } -static void ceph_netfs_expand_readahead(struct netfs_read_request *rreq) +static void ceph_netfs_expand_readahead(struct netfs_io_request *rreq) { - struct inode *inode = rreq->mapping->host; + struct inode *inode = rreq->inode; struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_file_layout *lo = &ci->i_layout; u32 blockoff; @@ -194,9 +199,9 @@ static void ceph_netfs_expand_readahead(struct netfs_read_request *rreq) rreq->len = roundup(rreq->len, lo->stripe_unit); } -static bool ceph_netfs_clamp_length(struct netfs_read_subrequest *subreq) +static bool ceph_netfs_clamp_length(struct netfs_io_subrequest *subreq) { - struct inode *inode = subreq->rreq->mapping->host; + struct inode *inode = subreq->rreq->inode; struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_inode_info *ci = ceph_inode(inode); u64 objno, objoff; @@ -213,7 +218,7 @@ static void finish_netfs_read(struct ceph_osd_request *req) { struct ceph_fs_client *fsc = ceph_inode_to_client(req->r_inode); struct ceph_osd_data *osd_data = osd_req_op_extent_osd_data(req, 0); - struct netfs_read_subrequest *subreq = req->r_priv; + struct netfs_io_subrequest *subreq = req->r_priv; int num_pages; int err = req->r_result; @@ -239,10 +244,63 @@ static void finish_netfs_read(struct ceph_osd_request *req) iput(req->r_inode); } -static void ceph_netfs_issue_op(struct netfs_read_subrequest *subreq) +static bool ceph_netfs_issue_op_inline(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *rreq = subreq->rreq; + struct inode *inode = rreq->inode; + struct ceph_mds_reply_info_parsed *rinfo; + struct ceph_mds_reply_info_in *iinfo; + struct ceph_mds_request *req; + struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(inode->i_sb); + struct ceph_inode_info *ci = ceph_inode(inode); + struct iov_iter iter; + ssize_t err = 0; + size_t len; + + __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); + __clear_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags); + + if (subreq->start >= inode->i_size) + goto out; + + /* We need to fetch the inline data. */ + req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_GETATTR, USE_ANY_MDS); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto out; + } + req->r_ino1 = ci->i_vino; + req->r_args.getattr.mask = cpu_to_le32(CEPH_STAT_CAP_INLINE_DATA); + req->r_num_caps = 2; + + err = ceph_mdsc_do_request(mdsc, NULL, req); + if (err < 0) + goto out; + + rinfo = &req->r_reply_info; + iinfo = &rinfo->targeti; + if (iinfo->inline_version == CEPH_INLINE_NONE) { + /* The data got uninlined */ + ceph_mdsc_put_request(req); + return false; + } + + len = min_t(size_t, iinfo->inline_len - subreq->start, subreq->len); + iov_iter_xarray(&iter, READ, &rreq->mapping->i_pages, subreq->start, len); + err = copy_to_iter(iinfo->inline_data + subreq->start, len, &iter); + if (err == 0) + err = -EFAULT; + + ceph_mdsc_put_request(req); +out: + netfs_subreq_terminated(subreq, err, false); + return true; +} + +static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq) { - struct netfs_read_request *rreq = subreq->rreq; - struct inode *inode = rreq->mapping->host; + struct netfs_io_request *rreq = subreq->rreq; + struct inode *inode = rreq->inode; struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_osd_request *req; @@ -253,6 +311,10 @@ static void ceph_netfs_issue_op(struct netfs_read_subrequest *subreq) int err = 0; u64 len = subreq->len; + if (ci->i_inline_version != CEPH_INLINE_NONE && + ceph_netfs_issue_op_inline(subreq)) + return; + req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, vino, subreq->start, &len, 0, 1, CEPH_OSD_OP_READ, CEPH_OSD_FLAG_READ | fsc->client->osdc.client->options->read_from_replica, @@ -291,8 +353,43 @@ out: dout("%s: result %d\n", __func__, err); } -static void ceph_init_rreq(struct netfs_read_request *rreq, struct file *file) +static int ceph_init_request(struct netfs_io_request *rreq, struct file *file) { + struct inode *inode = rreq->inode; + int got = 0, want = CEPH_CAP_FILE_CACHE; + int ret = 0; + + if (rreq->origin != NETFS_READAHEAD) + return 0; + + if (file) { + struct ceph_rw_context *rw_ctx; + struct ceph_file_info *fi = file->private_data; + + rw_ctx = ceph_find_rw_context(fi); + if (rw_ctx) + return 0; + } + + /* + * readahead callers do not necessarily hold Fcb caps + * (e.g. fadvise, madvise). + */ + ret = ceph_try_get_caps(inode, CEPH_CAP_FILE_RD, want, true, &got); + if (ret < 0) { + dout("start_read %p, error getting cap\n", inode); + return ret; + } + + if (!(got & want)) { + dout("start_read %p, no cache cap\n", inode); + return -EACCES; + } + if (ret == 0) + return -EACCES; + + rreq->netfs_priv = (void *)(uintptr_t)got; + return 0; } static void ceph_readahead_cleanup(struct address_space *mapping, void *priv) @@ -305,77 +402,47 @@ static void ceph_readahead_cleanup(struct address_space *mapping, void *priv) ceph_put_cap_refs(ci, got); } -static const struct netfs_read_request_ops ceph_netfs_read_ops = { - .init_rreq = ceph_init_rreq, - .is_cache_enabled = ceph_is_cache_enabled, +const struct netfs_request_ops ceph_netfs_ops = { + .init_request = ceph_init_request, .begin_cache_operation = ceph_begin_cache_operation, - .issue_op = ceph_netfs_issue_op, + .issue_read = ceph_netfs_issue_read, .expand_readahead = ceph_netfs_expand_readahead, .clamp_length = ceph_netfs_clamp_length, .check_write_begin = ceph_netfs_check_write_begin, .cleanup = ceph_readahead_cleanup, }; -/* read a single page, without unlocking it. */ -static int ceph_readpage(struct file *file, struct page *page) +#ifdef CONFIG_CEPH_FSCACHE +static void ceph_set_page_fscache(struct page *page) { - struct inode *inode = file_inode(file); - struct ceph_inode_info *ci = ceph_inode(inode); - struct ceph_vino vino = ceph_vino(inode); - u64 off = page_offset(page); - u64 len = thp_size(page); - - if (ci->i_inline_version != CEPH_INLINE_NONE) { - /* - * Uptodate inline data should have been added - * into page cache while getting Fcr caps. - */ - if (off == 0) { - unlock_page(page); - return -EINVAL; - } - zero_user_segment(page, 0, thp_size(page)); - SetPageUptodate(page); - unlock_page(page); - return 0; - } - - dout("readpage ino %llx.%llx file %p off %llu len %llu page %p index %lu\n", - vino.ino, vino.snap, file, off, len, page, page->index); - - return netfs_readpage(file, page, &ceph_netfs_read_ops, NULL); + set_page_fscache(page); } -static void ceph_readahead(struct readahead_control *ractl) +static void ceph_fscache_write_terminated(void *priv, ssize_t error, bool was_async) { - struct inode *inode = file_inode(ractl->file); - struct ceph_file_info *fi = ractl->file->private_data; - struct ceph_rw_context *rw_ctx; - int got = 0; - int ret = 0; + struct inode *inode = priv; - if (ceph_inode(inode)->i_inline_version != CEPH_INLINE_NONE) - return; + if (IS_ERR_VALUE(error) && error != -ENOBUFS) + ceph_fscache_invalidate(inode, false); +} - rw_ctx = ceph_find_rw_context(fi); - if (!rw_ctx) { - /* - * readahead callers do not necessarily hold Fcb caps - * (e.g. fadvise, madvise). - */ - int want = CEPH_CAP_FILE_CACHE; +static void ceph_fscache_write_to_cache(struct inode *inode, u64 off, u64 len, bool caching) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + struct fscache_cookie *cookie = ceph_fscache_cookie(ci); - ret = ceph_try_get_caps(inode, CEPH_CAP_FILE_RD, want, true, &got); - if (ret < 0) - dout("start_read %p, error getting cap\n", inode); - else if (!(got & want)) - dout("start_read %p, no cache cap\n", inode); + fscache_write_to_cache(cookie, inode->i_mapping, off, len, i_size_read(inode), + ceph_fscache_write_terminated, inode, caching); +} +#else +static inline void ceph_set_page_fscache(struct page *page) +{ +} - if (ret <= 0) - return; - } - netfs_readahead(ractl, &ceph_netfs_read_ops, (void *)(uintptr_t)got); +static inline void ceph_fscache_write_to_cache(struct inode *inode, u64 off, u64 len, bool caching) +{ } +#endif /* CONFIG_CEPH_FSCACHE */ struct ceph_writeback_ctl { @@ -482,6 +549,7 @@ static u64 get_writepages_data_length(struct inode *inode, */ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) { + struct folio *folio = page_folio(page); struct inode *inode = page->mapping->host; struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); @@ -492,6 +560,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) struct ceph_writeback_ctl ceph_wbc; struct ceph_osd_client *osdc = &fsc->client->osdc; struct ceph_osd_request *req; + bool caching = ceph_is_cache_enabled(inode); dout("writepage %p idx %lu\n", page, page->index); @@ -515,8 +584,9 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) /* is this a partial page at end of file? */ if (page_off >= ceph_wbc.i_size) { - dout("%p page eof %llu\n", page, ceph_wbc.i_size); - page->mapping->a_ops->invalidatepage(page, 0, thp_size(page)); + dout("folio at %lu beyond eof %llu\n", folio->index, + ceph_wbc.i_size); + folio_invalidate(folio, 0, folio_size(folio)); return 0; } @@ -528,18 +598,19 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) if (atomic_long_inc_return(&fsc->writeback_count) > CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb)) - set_bdi_congested(inode_to_bdi(inode), BLK_RW_ASYNC); + fsc->write_congested = true; - set_page_writeback(page); req = ceph_osdc_new_request(osdc, &ci->i_layout, ceph_vino(inode), page_off, &len, 0, 1, CEPH_OSD_OP_WRITE, CEPH_OSD_FLAG_WRITE, snapc, ceph_wbc.truncate_seq, ceph_wbc.truncate_size, true); - if (IS_ERR(req)) { - redirty_page_for_writepage(wbc, page); - end_page_writeback(page); + if (IS_ERR(req)) return PTR_ERR(req); - } + + set_page_writeback(page); + if (caching) + ceph_set_page_fscache(page); + ceph_fscache_write_to_cache(inode, page_off, len, caching); /* it may be a short write due to an object boundary */ WARN_ON_ONCE(len > thp_size(page)); @@ -587,7 +658,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) if (atomic_long_dec_return(&fsc->writeback_count) < CONGESTION_OFF_THRESH(fsc->mount_options->congestion_kb)) - clear_bdi_congested(inode_to_bdi(inode), BLK_RW_ASYNC); + fsc->write_congested = false; return err; } @@ -598,6 +669,13 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc) struct inode *inode = page->mapping->host; BUG_ON(!inode); ihold(inode); + + if (wbc->sync_mode == WB_SYNC_NONE && + ceph_inode_to_client(inode)->write_congested) + return AOP_WRITEPAGE_ACTIVATE; + + wait_on_page_fscache(page); + err = writepage_nounlock(page, wbc); if (err == -ERESTARTSYS) { /* direct memory reclaimer was killed by SIGKILL. return 0 @@ -651,8 +729,11 @@ static void writepages_finish(struct ceph_osd_request *req) /* clean all pages */ for (i = 0; i < req->r_num_ops; i++) { - if (req->r_ops[i].op != CEPH_OSD_OP_WRITE) + if (req->r_ops[i].op != CEPH_OSD_OP_WRITE) { + pr_warn("%s incorrect op %d req %p index %d tid %llu\n", + __func__, req->r_ops[i].op, req, i, req->r_tid); break; + } osd_data = osd_req_op_extent_osd_data(req, i); BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_PAGES); @@ -668,8 +749,7 @@ static void writepages_finish(struct ceph_osd_request *req) if (atomic_long_dec_return(&fsc->writeback_count) < CONGESTION_OFF_THRESH( fsc->mount_options->congestion_kb)) - clear_bdi_congested(inode_to_bdi(inode), - BLK_RW_ASYNC); + fsc->write_congested = false; ceph_put_snap_context(detach_page_private(page)); end_page_writeback(page); @@ -719,12 +799,17 @@ static int ceph_writepages_start(struct address_space *mapping, struct ceph_writeback_ctl ceph_wbc; bool should_loop, range_whole = false; bool done = false; + bool caching = ceph_is_cache_enabled(inode); + + if (wbc->sync_mode == WB_SYNC_NONE && + fsc->write_congested) + return 0; dout("writepages_start %p (mode=%s)\n", inode, wbc->sync_mode == WB_SYNC_NONE ? "NONE" : (wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD")); - if (READ_ONCE(fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN) { + if (ceph_inode_is_shutdown(inode)) { if (ci->i_wrbuffer_ref > 0) { pr_warn_ratelimited( "writepage_start %p %lld forced umount\n", @@ -827,14 +912,16 @@ get_more_pages: continue; } if (page_offset(page) >= ceph_wbc.i_size) { - dout("%p page eof %llu\n", - page, ceph_wbc.i_size); + struct folio *folio = page_folio(page); + + dout("folio at %lu beyond eof %llu\n", + folio->index, ceph_wbc.i_size); if ((ceph_wbc.size_stable || - page_offset(page) >= i_size_read(inode)) && - clear_page_dirty_for_io(page)) - mapping->a_ops->invalidatepage(page, - 0, thp_size(page)); - unlock_page(page); + folio_pos(folio) >= i_size_read(inode)) && + folio_clear_dirty_for_io(folio)) + folio_invalidate(folio, 0, + folio_size(folio)); + folio_unlock(folio); continue; } if (strip_unit_end && (page->index > strip_unit_end)) { @@ -842,7 +929,7 @@ get_more_pages: unlock_page(page); break; } - if (PageWriteback(page)) { + if (PageWriteback(page) || PageFsCache(page)) { if (wbc->sync_mode == WB_SYNC_NONE) { dout("%p under writeback\n", page); unlock_page(page); @@ -850,6 +937,7 @@ get_more_pages: } dout("waiting on writeback %p\n", page); wait_on_page_writeback(page); + wait_on_page_fscache(page); } if (!clear_page_dirty_for_io(page)) { @@ -913,11 +1001,8 @@ get_more_pages: if (atomic_long_inc_return(&fsc->writeback_count) > CONGESTION_ON_THRESH( - fsc->mount_options->congestion_kb)) { - set_bdi_congested(inode_to_bdi(inode), - BLK_RW_ASYNC); - } - + fsc->mount_options->congestion_kb)) + fsc->write_congested = true; pages[locked_pages++] = page; pvec.pages[i] = NULL; @@ -982,9 +1067,19 @@ new_request: op_idx = 0; for (i = 0; i < locked_pages; i++) { u64 cur_offset = page_offset(pages[i]); + /* + * Discontinuity in page range? Ceph can handle that by just passing + * multiple extents in the write op. + */ if (offset + len != cur_offset) { + /* If it's full, stop here */ if (op_idx + 1 == req->r_num_ops) break; + + /* Kick off an fscache write with what we have so far. */ + ceph_fscache_write_to_cache(inode, offset, len, caching); + + /* Start a new extent */ osd_req_op_extent_dup_last(req, op_idx, cur_offset - offset); dout("writepages got pages at %llu~%llu\n", @@ -995,14 +1090,17 @@ new_request: osd_req_op_extent_update(req, op_idx, len); len = 0; - offset = cur_offset; + offset = cur_offset; data_pages = pages + i; op_idx++; } set_page_writeback(pages[i]); + if (caching) + ceph_set_page_fscache(pages[i]); len += thp_size(page); } + ceph_fscache_write_to_cache(inode, offset, len, caching); if (ceph_wbc.size_stable) { len = min(len, ceph_wbc.i_size - offset); @@ -1145,12 +1243,12 @@ static struct ceph_snap_context * ceph_find_incompatible(struct page *page) { struct inode *inode = page->mapping->host; - struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_inode_info *ci = ceph_inode(inode); - if (READ_ONCE(fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN) { - dout(" page %p forced umount\n", page); - return ERR_PTR(-EIO); + if (ceph_inode_is_shutdown(inode)) { + dout(" page %p %llx:%llx is shutdown\n", page, + ceph_vinop(inode)); + return ERR_PTR(-ESTALE); } for (;;) { @@ -1187,18 +1285,18 @@ ceph_find_incompatible(struct page *page) } static int ceph_netfs_check_write_begin(struct file *file, loff_t pos, unsigned int len, - struct page *page, void **_fsdata) + struct folio *folio, void **_fsdata) { struct inode *inode = file_inode(file); struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_snap_context *snapc; - snapc = ceph_find_incompatible(page); + snapc = ceph_find_incompatible(folio_page(folio, 0)); if (snapc) { int r; - unlock_page(page); - put_page(page); + folio_unlock(folio); + folio_put(folio); if (IS_ERR(snapc)) return PTR_ERR(snapc); @@ -1216,53 +1314,22 @@ static int ceph_netfs_check_write_begin(struct file *file, loff_t pos, unsigned * clean, or already dirty within the same snap context. */ static int ceph_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, unsigned flags, + loff_t pos, unsigned len, unsigned aop_flags, struct page **pagep, void **fsdata) { struct inode *inode = file_inode(file); - struct ceph_inode_info *ci = ceph_inode(inode); - struct page *page = NULL; - pgoff_t index = pos >> PAGE_SHIFT; + struct folio *folio = NULL; int r; - /* - * Uninlining should have already been done and everything updated, EXCEPT - * for inline_version sent to the MDS. - */ - if (ci->i_inline_version != CEPH_INLINE_NONE) { - page = grab_cache_page_write_begin(mapping, index, flags); - if (!page) - return -ENOMEM; - - /* - * The inline_version on a new inode is set to 1. If that's the - * case, then the page is brand new and isn't yet Uptodate. - */ - r = 0; - if (index == 0 && ci->i_inline_version != 1) { - if (!PageUptodate(page)) { - WARN_ONCE(1, "ceph: write_begin called on still-inlined inode (inline_version %llu)!\n", - ci->i_inline_version); - r = -EINVAL; - } - goto out; - } - zero_user_segment(page, 0, thp_size(page)); - SetPageUptodate(page); - goto out; - } - - r = netfs_write_begin(file, inode->i_mapping, pos, len, 0, &page, NULL, - &ceph_netfs_read_ops, NULL); -out: + r = netfs_write_begin(file, inode->i_mapping, pos, len, 0, &folio, NULL); if (r == 0) - wait_on_page_fscache(page); + folio_wait_fscache(folio); if (r < 0) { - if (page) - put_page(page); + if (folio) + folio_put(folio); } else { - WARN_ON_ONCE(!PageLocked(page)); - *pagep = page; + WARN_ON_ONCE(!folio_test_locked(folio)); + *pagep = &folio->page; } return r; } @@ -1273,32 +1340,33 @@ out: */ static int ceph_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) + struct page *subpage, void *fsdata) { + struct folio *folio = page_folio(subpage); struct inode *inode = file_inode(file); bool check_cap = false; - dout("write_end file %p inode %p page %p %d~%d (%d)\n", file, - inode, page, (int)pos, (int)copied, (int)len); + dout("write_end file %p inode %p folio %p %d~%d (%d)\n", file, + inode, folio, (int)pos, (int)copied, (int)len); - if (!PageUptodate(page)) { + if (!folio_test_uptodate(folio)) { /* just return that nothing was copied on a short copy */ if (copied < len) { copied = 0; goto out; } - SetPageUptodate(page); + folio_mark_uptodate(folio); } /* did file size increase? */ if (pos+copied > i_size_read(inode)) check_cap = ceph_inode_set_size(inode, pos+copied); - set_page_dirty(page); + folio_mark_dirty(folio); out: - unlock_page(page); - put_page(page); + folio_unlock(folio); + folio_put(folio); if (check_cap) ceph_check_caps(ceph_inode(inode), CHECK_CAPS_AUTHONLY, NULL); @@ -1306,28 +1374,17 @@ out: return copied; } -/* - * we set .direct_IO to indicate direct io is supported, but since we - * intercept O_DIRECT reads and writes early, this function should - * never get called. - */ -static ssize_t ceph_direct_io(struct kiocb *iocb, struct iov_iter *iter) -{ - WARN_ON(1); - return -EINVAL; -} - const struct address_space_operations ceph_aops = { - .readpage = ceph_readpage, - .readahead = ceph_readahead, + .readpage = netfs_readpage, + .readahead = netfs_readahead, .writepage = ceph_writepage, .writepages = ceph_writepages_start, .write_begin = ceph_write_begin, .write_end = ceph_write_end, - .set_page_dirty = ceph_set_page_dirty, - .invalidatepage = ceph_invalidatepage, + .dirty_folio = ceph_dirty_folio, + .invalidate_folio = ceph_invalidate_folio, .releasepage = ceph_releasepage, - .direct_IO = ceph_direct_io, + .direct_IO = noop_direct_IO, }; static void ceph_block_sigs(sigset_t *oldset) @@ -1356,6 +1413,9 @@ static vm_fault_t ceph_filemap_fault(struct vm_fault *vmf) sigset_t oldset; vm_fault_t ret = VM_FAULT_SIGBUS; + if (ceph_inode_is_shutdown(inode)) + return ret; + ceph_block_sigs(&oldset); dout("filemap_fault %p %llx.%llx %llu trying to get caps\n", @@ -1447,6 +1507,9 @@ static vm_fault_t ceph_page_mkwrite(struct vm_fault *vmf) sigset_t oldset; vm_fault_t ret = VM_FAULT_SIGBUS; + if (ceph_inode_is_shutdown(inode)) + return ret; + prealloc_cf = ceph_alloc_cap_flush(); if (!prealloc_cf) return VM_FAULT_OOM; @@ -1454,19 +1517,6 @@ static vm_fault_t ceph_page_mkwrite(struct vm_fault *vmf) sb_start_pagefault(inode->i_sb); ceph_block_sigs(&oldset); - if (ci->i_inline_version != CEPH_INLINE_NONE) { - struct page *locked_page = NULL; - if (off == 0) { - lock_page(page); - locked_page = page; - } - err = ceph_uninline_data(vma->vm_file, locked_page); - if (locked_page) - unlock_page(locked_page); - if (err < 0) - goto out_free; - } - if (off + thp_size(page) <= size) len = thp_size(page); else @@ -1523,11 +1573,9 @@ static vm_fault_t ceph_page_mkwrite(struct vm_fault *vmf) ceph_put_snap_context(snapc); } while (err == 0); - if (ret == VM_FAULT_LOCKED || - ci->i_inline_version != CEPH_INLINE_NONE) { + if (ret == VM_FAULT_LOCKED) { int dirty; spin_lock(&ci->i_ceph_lock); - ci->i_inline_version = CEPH_INLINE_NONE; dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR, &prealloc_cf); spin_unlock(&ci->i_ceph_lock); @@ -1591,16 +1639,30 @@ void ceph_fill_inline_data(struct inode *inode, struct page *locked_page, } } -int ceph_uninline_data(struct file *filp, struct page *locked_page) +int ceph_uninline_data(struct file *file) { - struct inode *inode = file_inode(filp); + struct inode *inode = file_inode(file); struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_osd_request *req; - struct page *page = NULL; - u64 len, inline_version; + struct ceph_cap_flush *prealloc_cf; + struct folio *folio = NULL; + u64 inline_version = CEPH_INLINE_NONE; + struct page *pages[1]; int err = 0; - bool from_pagecache = false; + u64 len; + + prealloc_cf = ceph_alloc_cap_flush(); + if (!prealloc_cf) + return -ENOMEM; + + folio = read_mapping_folio(inode->i_mapping, 0, file); + if (IS_ERR(folio)) { + err = PTR_ERR(folio); + goto out; + } + + folio_lock(folio); spin_lock(&ci->i_ceph_lock); inline_version = ci->i_inline_version; @@ -1611,45 +1673,11 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page) if (inline_version == 1 || /* initial version, no data */ inline_version == CEPH_INLINE_NONE) - goto out; - - if (locked_page) { - page = locked_page; - WARN_ON(!PageUptodate(page)); - } else if (ceph_caps_issued(ci) & - (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) { - page = find_get_page(inode->i_mapping, 0); - if (page) { - if (PageUptodate(page)) { - from_pagecache = true; - lock_page(page); - } else { - put_page(page); - page = NULL; - } - } - } + goto out_unlock; - if (page) { - len = i_size_read(inode); - if (len > PAGE_SIZE) - len = PAGE_SIZE; - } else { - page = __page_cache_alloc(GFP_NOFS); - if (!page) { - err = -ENOMEM; - goto out; - } - err = __ceph_do_getattr(inode, page, - CEPH_STAT_CAP_INLINE_DATA, true); - if (err < 0) { - /* no inline data */ - if (err == -ENODATA) - err = 0; - goto out; - } - len = err; - } + len = i_size_read(inode); + if (len > folio_size(folio)) + len = folio_size(folio); req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, ceph_vino(inode), 0, &len, 0, 1, @@ -1657,7 +1685,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page) NULL, 0, 0, false); if (IS_ERR(req)) { err = PTR_ERR(req); - goto out; + goto out_unlock; } req->r_mtime = inode->i_mtime; @@ -1666,7 +1694,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page) err = ceph_osdc_wait_request(&fsc->client->osdc, req); ceph_osdc_put_request(req); if (err < 0) - goto out; + goto out_unlock; req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, ceph_vino(inode), 0, &len, 1, 3, @@ -1675,10 +1703,11 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page) ci->i_truncate_size, false); if (IS_ERR(req)) { err = PTR_ERR(req); - goto out; + goto out_unlock; } - osd_req_op_extent_osd_data_pages(req, 1, &page, len, 0, false, false); + pages[0] = folio_page(folio, 0); + osd_req_op_extent_osd_data_pages(req, 1, pages, len, 0, false, false); { __le64 xattr_buf = cpu_to_le64(inline_version); @@ -1688,7 +1717,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page) CEPH_OSD_CMPXATTR_OP_GT, CEPH_OSD_CMPXATTR_MODE_U64); if (err) - goto out_put; + goto out_put_req; } { @@ -1699,7 +1728,7 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page) "inline_version", xattr_buf, xattr_len, 0, 0); if (err) - goto out_put; + goto out_put_req; } req->r_mtime = inode->i_mtime; @@ -1710,19 +1739,28 @@ int ceph_uninline_data(struct file *filp, struct page *locked_page) ceph_update_write_metrics(&fsc->mdsc->metric, req->r_start_latency, req->r_end_latency, len, err); -out_put: + if (!err) { + int dirty; + + /* Set to CAP_INLINE_NONE and dirty the caps */ + down_read(&fsc->mdsc->snap_rwsem); + spin_lock(&ci->i_ceph_lock); + ci->i_inline_version = CEPH_INLINE_NONE; + dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR, &prealloc_cf); + spin_unlock(&ci->i_ceph_lock); + up_read(&fsc->mdsc->snap_rwsem); + if (dirty) + __mark_inode_dirty(inode, dirty); + } +out_put_req: ceph_osdc_put_request(req); if (err == -ECANCELED) err = 0; +out_unlock: + folio_unlock(folio); + folio_put(folio); out: - if (page && page != locked_page) { - if (from_pagecache) { - unlock_page(page); - put_page(page); - } else - __free_pages(page, 0); - } - + ceph_free_cap_flush(prealloc_cf); dout("uninline_data %p %llx.%llx inline_version %llu = %d\n", inode, ceph_vinop(inode), inline_version, err); return err; diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c index 9cfadbb86568..ddea99922073 100644 --- a/fs/ceph/cache.c +++ b/fs/ceph/cache.c @@ -12,216 +12,99 @@ #include "super.h" #include "cache.h" -struct ceph_aux_inode { - u64 version; - u64 mtime_sec; - u64 mtime_nsec; -}; - -struct fscache_netfs ceph_cache_netfs = { - .name = "ceph", - .version = 0, -}; - -static DEFINE_MUTEX(ceph_fscache_lock); -static LIST_HEAD(ceph_fscache_list); - -struct ceph_fscache_entry { - struct list_head list; - struct fscache_cookie *fscache; - size_t uniq_len; - /* The following members must be last */ - struct ceph_fsid fsid; - char uniquifier[]; -}; - -static const struct fscache_cookie_def ceph_fscache_fsid_object_def = { - .name = "CEPH.fsid", - .type = FSCACHE_COOKIE_TYPE_INDEX, -}; - -int __init ceph_fscache_register(void) -{ - return fscache_register_netfs(&ceph_cache_netfs); -} - -void ceph_fscache_unregister(void) -{ - fscache_unregister_netfs(&ceph_cache_netfs); -} - -int ceph_fscache_register_fs(struct ceph_fs_client* fsc, struct fs_context *fc) +void ceph_fscache_register_inode_cookie(struct inode *inode) { - const struct ceph_fsid *fsid = &fsc->client->fsid; - const char *fscache_uniq = fsc->mount_options->fscache_uniq; - size_t uniq_len = fscache_uniq ? strlen(fscache_uniq) : 0; - struct ceph_fscache_entry *ent; - int err = 0; + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_fs_client *fsc = ceph_inode_to_client(inode); - mutex_lock(&ceph_fscache_lock); - list_for_each_entry(ent, &ceph_fscache_list, list) { - if (memcmp(&ent->fsid, fsid, sizeof(*fsid))) - continue; - if (ent->uniq_len != uniq_len) - continue; - if (uniq_len && memcmp(ent->uniquifier, fscache_uniq, uniq_len)) - continue; - - errorfc(fc, "fscache cookie already registered for fsid %pU, use fsc=<uniquifier> option", - fsid); - err = -EBUSY; - goto out_unlock; - } + /* No caching for filesystem? */ + if (!fsc->fscache) + return; - ent = kzalloc(sizeof(*ent) + uniq_len, GFP_KERNEL); - if (!ent) { - err = -ENOMEM; - goto out_unlock; - } + /* Regular files only */ + if (!S_ISREG(inode->i_mode)) + return; - memcpy(&ent->fsid, fsid, sizeof(*fsid)); - if (uniq_len > 0) { - memcpy(&ent->uniquifier, fscache_uniq, uniq_len); - ent->uniq_len = uniq_len; - } + /* Only new inodes! */ + if (!(inode->i_state & I_NEW)) + return; - fsc->fscache = fscache_acquire_cookie(ceph_cache_netfs.primary_index, - &ceph_fscache_fsid_object_def, - &ent->fsid, sizeof(ent->fsid) + uniq_len, - NULL, 0, - fsc, 0, true); + WARN_ON_ONCE(ci->netfs_ctx.cache); - if (fsc->fscache) { - ent->fscache = fsc->fscache; - list_add_tail(&ent->list, &ceph_fscache_list); - } else { - kfree(ent); - errorfc(fc, "unable to register fscache cookie for fsid %pU", - fsid); - /* all other fs ignore this error */ - } -out_unlock: - mutex_unlock(&ceph_fscache_lock); - return err; + ci->netfs_ctx.cache = + fscache_acquire_cookie(fsc->fscache, 0, + &ci->i_vino, sizeof(ci->i_vino), + &ci->i_version, sizeof(ci->i_version), + i_size_read(inode)); } -static enum fscache_checkaux ceph_fscache_inode_check_aux( - void *cookie_netfs_data, const void *data, uint16_t dlen, - loff_t object_size) +void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info *ci) { - struct ceph_aux_inode aux; - struct ceph_inode_info* ci = cookie_netfs_data; - struct inode* inode = &ci->vfs_inode; - - if (dlen != sizeof(aux) || - i_size_read(inode) != object_size) - return FSCACHE_CHECKAUX_OBSOLETE; - - memset(&aux, 0, sizeof(aux)); - aux.version = ci->i_version; - aux.mtime_sec = inode->i_mtime.tv_sec; - aux.mtime_nsec = inode->i_mtime.tv_nsec; + fscache_relinquish_cookie(ceph_fscache_cookie(ci), false); +} - if (memcmp(data, &aux, sizeof(aux)) != 0) - return FSCACHE_CHECKAUX_OBSOLETE; +void ceph_fscache_use_cookie(struct inode *inode, bool will_modify) +{ + struct ceph_inode_info *ci = ceph_inode(inode); - dout("ceph inode 0x%p cached okay\n", ci); - return FSCACHE_CHECKAUX_OKAY; + fscache_use_cookie(ceph_fscache_cookie(ci), will_modify); } -static const struct fscache_cookie_def ceph_fscache_inode_object_def = { - .name = "CEPH.inode", - .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .check_aux = ceph_fscache_inode_check_aux, -}; - -void ceph_fscache_register_inode_cookie(struct inode *inode) +void ceph_fscache_unuse_cookie(struct inode *inode, bool update) { struct ceph_inode_info *ci = ceph_inode(inode); - struct ceph_fs_client *fsc = ceph_inode_to_client(inode); - struct ceph_aux_inode aux; - - /* No caching for filesystem */ - if (!fsc->fscache) - return; - /* Only cache for regular files that are read only */ - if (!S_ISREG(inode->i_mode)) - return; + if (update) { + loff_t i_size = i_size_read(inode); - inode_lock_nested(inode, I_MUTEX_CHILD); - if (!ci->fscache) { - memset(&aux, 0, sizeof(aux)); - aux.version = ci->i_version; - aux.mtime_sec = inode->i_mtime.tv_sec; - aux.mtime_nsec = inode->i_mtime.tv_nsec; - ci->fscache = fscache_acquire_cookie(fsc->fscache, - &ceph_fscache_inode_object_def, - &ci->i_vino, sizeof(ci->i_vino), - &aux, sizeof(aux), - ci, i_size_read(inode), false); + fscache_unuse_cookie(ceph_fscache_cookie(ci), + &ci->i_version, &i_size); + } else { + fscache_unuse_cookie(ceph_fscache_cookie(ci), NULL, NULL); } - inode_unlock(inode); } -void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci) +void ceph_fscache_update(struct inode *inode) { - struct fscache_cookie* cookie; - - if ((cookie = ci->fscache) == NULL) - return; - - ci->fscache = NULL; + struct ceph_inode_info *ci = ceph_inode(inode); + loff_t i_size = i_size_read(inode); - fscache_relinquish_cookie(cookie, &ci->i_vino, false); + fscache_update_cookie(ceph_fscache_cookie(ci), &ci->i_version, &i_size); } -static bool ceph_fscache_can_enable(void *data) +void ceph_fscache_invalidate(struct inode *inode, bool dio_write) { - struct inode *inode = data; - return !inode_is_open_for_write(inode); + struct ceph_inode_info *ci = ceph_inode(inode); + + fscache_invalidate(ceph_fscache_cookie(ci), + &ci->i_version, i_size_read(inode), + dio_write ? FSCACHE_INVAL_DIO_WRITE : 0); } -void ceph_fscache_file_set_cookie(struct inode *inode, struct file *filp) +int ceph_fscache_register_fs(struct ceph_fs_client* fsc, struct fs_context *fc) { - struct ceph_inode_info *ci = ceph_inode(inode); + const struct ceph_fsid *fsid = &fsc->client->fsid; + const char *fscache_uniq = fsc->mount_options->fscache_uniq; + size_t uniq_len = fscache_uniq ? strlen(fscache_uniq) : 0; + char *name; + int err = 0; - if (!fscache_cookie_valid(ci->fscache)) - return; + name = kasprintf(GFP_KERNEL, "ceph,%pU%s%s", fsid, uniq_len ? "," : "", + uniq_len ? fscache_uniq : ""); + if (!name) + return -ENOMEM; - if (inode_is_open_for_write(inode)) { - dout("fscache_file_set_cookie %p %p disabling cache\n", - inode, filp); - fscache_disable_cookie(ci->fscache, &ci->i_vino, false); - } else { - fscache_enable_cookie(ci->fscache, &ci->i_vino, i_size_read(inode), - ceph_fscache_can_enable, inode); - if (fscache_cookie_enabled(ci->fscache)) { - dout("fscache_file_set_cookie %p %p enabling cache\n", - inode, filp); - } + fsc->fscache = fscache_acquire_volume(name, NULL, NULL, 0); + if (IS_ERR_OR_NULL(fsc->fscache)) { + errorfc(fc, "Unable to register fscache cookie for %s", name); + err = fsc->fscache ? PTR_ERR(fsc->fscache) : -EOPNOTSUPP; + fsc->fscache = NULL; } + kfree(name); + return err; } void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc) { - if (fscache_cookie_valid(fsc->fscache)) { - struct ceph_fscache_entry *ent; - bool found = false; - - mutex_lock(&ceph_fscache_lock); - list_for_each_entry(ent, &ceph_fscache_list, list) { - if (ent->fscache == fsc->fscache) { - list_del(&ent->list); - kfree(ent); - found = true; - break; - } - } - WARN_ON_ONCE(!found); - mutex_unlock(&ceph_fscache_lock); - - __fscache_relinquish_cookie(fsc->fscache, NULL, false); - } - fsc->fscache = NULL; + fscache_relinquish_volume(fsc->fscache, NULL, false); } diff --git a/fs/ceph/cache.h b/fs/ceph/cache.h index 058ea2a04376..7255b790a4c1 100644 --- a/fs/ceph/cache.h +++ b/fs/ceph/cache.h @@ -12,61 +12,70 @@ #include <linux/netfs.h> #ifdef CONFIG_CEPH_FSCACHE - -extern struct fscache_netfs ceph_cache_netfs; - -int ceph_fscache_register(void); -void ceph_fscache_unregister(void); +#include <linux/fscache.h> int ceph_fscache_register_fs(struct ceph_fs_client* fsc, struct fs_context *fc); void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc); void ceph_fscache_register_inode_cookie(struct inode *inode); void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci); -void ceph_fscache_file_set_cookie(struct inode *inode, struct file *filp); -void ceph_fscache_revalidate_cookie(struct ceph_inode_info *ci); -static inline void ceph_fscache_inode_init(struct ceph_inode_info *ci) +void ceph_fscache_use_cookie(struct inode *inode, bool will_modify); +void ceph_fscache_unuse_cookie(struct inode *inode, bool update); + +void ceph_fscache_update(struct inode *inode); +void ceph_fscache_invalidate(struct inode *inode, bool dio_write); + +static inline struct fscache_cookie *ceph_fscache_cookie(struct ceph_inode_info *ci) { - ci->fscache = NULL; + return netfs_i_cookie(&ci->vfs_inode); } -static inline struct fscache_cookie *ceph_fscache_cookie(struct ceph_inode_info *ci) +static inline void ceph_fscache_resize(struct inode *inode, loff_t to) { - return ci->fscache; + struct ceph_inode_info *ci = ceph_inode(inode); + struct fscache_cookie *cookie = ceph_fscache_cookie(ci); + + if (cookie) { + ceph_fscache_use_cookie(inode, true); + fscache_resize_cookie(cookie, to); + ceph_fscache_unuse_cookie(inode, true); + } } -static inline void ceph_fscache_invalidate(struct inode *inode) +static inline void ceph_fscache_unpin_writeback(struct inode *inode, + struct writeback_control *wbc) { - fscache_invalidate(ceph_inode(inode)->fscache); + fscache_unpin_writeback(wbc, ceph_fscache_cookie(ceph_inode(inode))); } -static inline bool ceph_is_cache_enabled(struct inode *inode) +static inline int ceph_fscache_dirty_folio(struct address_space *mapping, + struct folio *folio) { - struct fscache_cookie *cookie = ceph_fscache_cookie(ceph_inode(inode)); + struct ceph_inode_info *ci = ceph_inode(mapping->host); - if (!cookie) - return false; - return fscache_cookie_enabled(cookie); + return fscache_dirty_folio(mapping, folio, ceph_fscache_cookie(ci)); } -static inline int ceph_begin_cache_operation(struct netfs_read_request *rreq) +static inline int ceph_begin_cache_operation(struct netfs_io_request *rreq) { struct fscache_cookie *cookie = ceph_fscache_cookie(ceph_inode(rreq->inode)); - return fscache_begin_read_operation(rreq, cookie); + return fscache_begin_read_operation(&rreq->cache_resources, cookie); } -#else -static inline int ceph_fscache_register(void) +static inline bool ceph_is_cache_enabled(struct inode *inode) { - return 0; + return fscache_cookie_enabled(ceph_fscache_cookie(ceph_inode(inode))); } -static inline void ceph_fscache_unregister(void) +static inline void ceph_fscache_note_page_release(struct inode *inode) { -} + struct ceph_inode_info *ci = ceph_inode(inode); + fscache_note_page_release(ceph_fscache_cookie(ci)); +} +#else /* CONFIG_CEPH_FSCACHE */ static inline int ceph_fscache_register_fs(struct ceph_fs_client* fsc, struct fs_context *fc) { @@ -77,41 +86,63 @@ static inline void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc) { } -static inline void ceph_fscache_inode_init(struct ceph_inode_info *ci) +static inline void ceph_fscache_register_inode_cookie(struct inode *inode) { } -static inline struct fscache_cookie *ceph_fscache_cookie(struct ceph_inode_info *ci) +static inline void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci) { - return NULL; } -static inline void ceph_fscache_register_inode_cookie(struct inode *inode) +static inline void ceph_fscache_use_cookie(struct inode *inode, bool will_modify) { } -static inline void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci) +static inline void ceph_fscache_unuse_cookie(struct inode *inode, bool update) { } -static inline void ceph_fscache_file_set_cookie(struct inode *inode, - struct file *filp) +static inline void ceph_fscache_update(struct inode *inode) { } -static inline void ceph_fscache_invalidate(struct inode *inode) +static inline void ceph_fscache_invalidate(struct inode *inode, bool dio_write) { } +static inline struct fscache_cookie *ceph_fscache_cookie(struct ceph_inode_info *ci) +{ + return NULL; +} + +static inline void ceph_fscache_resize(struct inode *inode, loff_t to) +{ +} + +static inline void ceph_fscache_unpin_writeback(struct inode *inode, + struct writeback_control *wbc) +{ +} + +static inline int ceph_fscache_dirty_folio(struct address_space *mapping, + struct folio *folio) +{ + return filemap_dirty_folio(mapping, folio); +} + static inline bool ceph_is_cache_enabled(struct inode *inode) { return false; } -static inline int ceph_begin_cache_operation(struct netfs_read_request *rreq) +static inline int ceph_begin_cache_operation(struct netfs_io_request *rreq) { return -ENOBUFS; } -#endif -#endif /* _CEPH_CACHE_H */ +static inline void ceph_fscache_note_page_release(struct inode *inode) +{ +} +#endif /* CONFIG_CEPH_FSCACHE */ + +#endif diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 8f537f1d9d1d..5c14ef04e474 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -1188,11 +1188,11 @@ void ceph_remove_cap(struct ceph_cap *cap, bool queue_release) lockdep_assert_held(&ci->i_ceph_lock); - fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); + fsc = ceph_inode_to_client(&ci->vfs_inode); WARN_ON_ONCE(ci->i_auth_cap == cap && !list_empty(&ci->i_dirty_item) && !fsc->blocklisted && - READ_ONCE(fsc->mount_state) != CEPH_MOUNT_SHUTDOWN); + !ceph_inode_is_shutdown(&ci->vfs_inode)); __ceph_remove_cap(cap, queue_release); } @@ -1856,7 +1856,7 @@ static int try_nonblocking_invalidate(struct inode *inode) u32 invalidating_gen = ci->i_rdcache_gen; spin_unlock(&ci->i_ceph_lock); - ceph_fscache_invalidate(inode); + ceph_fscache_invalidate(inode, false); invalidate_mapping_pages(&inode->i_data, 0, -1); spin_lock(&ci->i_ceph_lock); @@ -1915,6 +1915,13 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags, ceph_get_mds_session(session); spin_lock(&ci->i_ceph_lock); + if (ci->i_ceph_flags & CEPH_I_ASYNC_CREATE) { + /* Don't send messages until we get async create reply */ + spin_unlock(&ci->i_ceph_lock); + ceph_put_mds_session(session); + return; + } + if (ci->i_ceph_flags & CEPH_I_FLUSH) flags |= CHECK_CAPS_FLUSH; retry: @@ -1968,8 +1975,8 @@ retry: } } - dout("check_caps %p file_want %s used %s dirty %s flushing %s" - " issued %s revoking %s retain %s %s%s\n", inode, + dout("check_caps %llx.%llx file_want %s used %s dirty %s flushing %s" + " issued %s revoking %s retain %s %s%s\n", ceph_vinop(inode), ceph_cap_string(file_wanted), ceph_cap_string(used), ceph_cap_string(ci->i_dirty_caps), ceph_cap_string(ci->i_flushing_caps), @@ -1990,7 +1997,8 @@ retry: (revoking & (CEPH_CAP_FILE_CACHE| CEPH_CAP_FILE_LAZYIO)) && /* or revoking cache */ !tried_invalidate) { - dout("check_caps trying to invalidate on %p\n", inode); + dout("check_caps trying to invalidate on %llx.%llx\n", + ceph_vinop(inode)); if (try_nonblocking_invalidate(inode) < 0) { dout("check_caps queuing invalidate\n"); queue_invalidate = true; @@ -2217,6 +2225,7 @@ static int unsafe_request_wait(struct inode *inode) struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc; struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_mds_request *req1 = NULL, *req2 = NULL; + unsigned int max_sessions; int ret, err = 0; spin_lock(&ci->i_unsafe_lock); @@ -2235,36 +2244,46 @@ static int unsafe_request_wait(struct inode *inode) spin_unlock(&ci->i_unsafe_lock); /* + * The mdsc->max_sessions is unlikely to be changed + * mostly, here we will retry it by reallocating the + * sessions array memory to get rid of the mdsc->mutex + * lock. + */ +retry: + max_sessions = mdsc->max_sessions; + + /* * Trigger to flush the journal logs in all the relevant MDSes * manually, or in the worst case we must wait at most 5 seconds * to wait the journal logs to be flushed by the MDSes periodically. */ - if (req1 || req2) { + if ((req1 || req2) && likely(max_sessions)) { struct ceph_mds_session **sessions = NULL; struct ceph_mds_session *s; struct ceph_mds_request *req; - unsigned int max; int i; - /* - * The mdsc->max_sessions is unlikely to be changed - * mostly, here we will retry it by reallocating the - * sessions arrary memory to get rid of the mdsc->mutex - * lock. - */ -retry: - max = mdsc->max_sessions; - sessions = krealloc(sessions, max * sizeof(s), __GFP_ZERO); - if (!sessions) - return -ENOMEM; + sessions = kzalloc(max_sessions * sizeof(s), GFP_KERNEL); + if (!sessions) { + err = -ENOMEM; + goto out; + } spin_lock(&ci->i_unsafe_lock); if (req1) { list_for_each_entry(req, &ci->i_unsafe_dirops, r_unsafe_dir_item) { s = req->r_session; - if (unlikely(s->s_mds >= max)) { + if (!s) + continue; + if (unlikely(s->s_mds >= max_sessions)) { spin_unlock(&ci->i_unsafe_lock); + for (i = 0; i < max_sessions; i++) { + s = sessions[i]; + if (s) + ceph_put_mds_session(s); + } + kfree(sessions); goto retry; } if (!sessions[s->s_mds]) { @@ -2277,8 +2296,16 @@ retry: list_for_each_entry(req, &ci->i_unsafe_iops, r_unsafe_target_item) { s = req->r_session; - if (unlikely(s->s_mds >= max)) { + if (!s) + continue; + if (unlikely(s->s_mds >= max_sessions)) { spin_unlock(&ci->i_unsafe_lock); + for (i = 0; i < max_sessions; i++) { + s = sessions[i]; + if (s) + ceph_put_mds_session(s); + } + kfree(sessions); goto retry; } if (!sessions[s->s_mds]) { @@ -2299,7 +2326,7 @@ retry: spin_unlock(&ci->i_ceph_lock); /* send flush mdlog request to MDSes */ - for (i = 0; i < max; i++) { + for (i = 0; i < max_sessions; i++) { s = sessions[i]; if (s) { send_flush_mdlog(s); @@ -2316,15 +2343,19 @@ retry: ceph_timeout_jiffies(req1->r_timeout)); if (ret) err = -EIO; - ceph_mdsc_put_request(req1); } if (req2) { ret = !wait_for_completion_timeout(&req2->r_safe_completion, ceph_timeout_jiffies(req2->r_timeout)); if (ret) err = -EIO; - ceph_mdsc_put_request(req2); } + +out: + if (req1) + ceph_mdsc_put_request(req1); + if (req2) + ceph_mdsc_put_request(req2); return err; } @@ -2387,7 +2418,11 @@ int ceph_write_inode(struct inode *inode, struct writeback_control *wbc) int wait = (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync); dout("write_inode %p wait=%d\n", inode, wait); + ceph_fscache_unpin_writeback(inode, wbc); if (wait) { + err = ceph_wait_on_async_create(inode); + if (err) + return err; dirty = try_flush_caps(inode, &flush_tid); if (dirty) err = wait_event_interruptible(ci->i_cap_wq, @@ -2418,6 +2453,10 @@ static void __kick_flushing_caps(struct ceph_mds_client *mdsc, u64 first_tid = 0; u64 last_snap_flush = 0; + /* Don't do anything until create reply comes in */ + if (ci->i_ceph_flags & CEPH_I_ASYNC_CREATE) + return; + ci->i_ceph_flags &= ~CEPH_I_KICK_FLUSH; list_for_each_entry_reverse(cf, &ci->i_cap_flush_list, i_list) { @@ -2629,9 +2668,9 @@ void ceph_take_cap_refs(struct ceph_inode_info *ci, int got, * * Returns 0 if caps were not able to be acquired (yet), 1 if succeed, * or a negative error code. There are 3 speical error codes: - * -EAGAIN: need to sleep but non-blocking is specified - * -EFBIG: ask caller to call check_max_size() and try again. - * -ESTALE: ask caller to call ceph_renew_caps() and try again. + * -EAGAIN: need to sleep but non-blocking is specified + * -EFBIG: ask caller to call check_max_size() and try again. + * -EUCLEAN: ask caller to call ceph_renew_caps() and try again. */ enum { /* first 8 bits are reserved for CEPH_FILE_MODE_FOO */ @@ -2679,7 +2718,7 @@ again: dout("get_cap_refs %p endoff %llu > maxsize %llu\n", inode, endoff, ci->i_max_size); if (endoff > ci->i_requested_max_size) - ret = ci->i_auth_cap ? -EFBIG : -ESTALE; + ret = ci->i_auth_cap ? -EFBIG : -EUCLEAN; goto out_unlock; } /* @@ -2749,9 +2788,9 @@ again: goto out_unlock; } - if (READ_ONCE(mdsc->fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN) { - dout("get_cap_refs %p forced umount\n", inode); - ret = -EIO; + if (ceph_inode_is_shutdown(inode)) { + dout("get_cap_refs %p inode is shutdown\n", inode); + ret = -ESTALE; goto out_unlock; } mds_wanted = __ceph_caps_mds_wanted(ci, false); @@ -2759,7 +2798,7 @@ again: dout("get_cap_refs %p need %s > mds_wanted %s\n", inode, ceph_cap_string(need), ceph_cap_string(mds_wanted)); - ret = -ESTALE; + ret = -EUCLEAN; goto out_unlock; } @@ -2843,7 +2882,7 @@ int ceph_try_get_caps(struct inode *inode, int need, int want, ret = try_get_cap_refs(inode, need, want, 0, flags, got); /* three special error codes */ - if (ret == -EAGAIN || ret == -EFBIG || ret == -ESTALE) + if (ret == -EAGAIN || ret == -EFBIG || ret == -EUCLEAN) ret = 0; return ret; } @@ -2926,7 +2965,7 @@ int ceph_get_caps(struct file *filp, int need, int want, loff_t endoff, int *got } if (ret < 0) { - if (ret == -EFBIG || ret == -ESTALE) { + if (ret == -EFBIG || ret == -EUCLEAN) { int ret2 = ceph_wait_on_async_create(inode); if (ret2 < 0) return ret2; @@ -2935,7 +2974,7 @@ int ceph_get_caps(struct file *filp, int need, int want, loff_t endoff, int *got check_max_size(inode, endoff); continue; } - if (ret == -ESTALE) { + if (ret == -EUCLEAN) { /* session was killed, try renew caps */ ret = ceph_renew_caps(inode, flags); if (ret == 0) @@ -3374,8 +3413,7 @@ static void handle_cap_grant(struct inode *inode, if ((newcaps & CEPH_CAP_LINK_SHARED) && (extra_info->issued & CEPH_CAP_LINK_EXCL) == 0) { set_nlink(inode, le32_to_cpu(grant->nlink)); - if (inode->i_nlink == 0 && - (newcaps & (CEPH_CAP_LINK_SHARED | CEPH_CAP_LINK_EXCL))) + if (inode->i_nlink == 0) deleted_inode = true; } @@ -3836,6 +3874,7 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex, dout("handle_cap_export inode %p ci %p mds%d mseq %d target %d\n", inode, ci, mds, mseq, target); retry: + down_read(&mdsc->snap_rwsem); spin_lock(&ci->i_ceph_lock); cap = __get_cap_for_mds(ci, mds); if (!cap || cap->cap_id != le64_to_cpu(ex->cap_id)) @@ -3899,6 +3938,7 @@ retry: } spin_unlock(&ci->i_ceph_lock); + up_read(&mdsc->snap_rwsem); mutex_unlock(&session->s_mutex); /* open target session */ @@ -3924,6 +3964,7 @@ retry: out_unlock: spin_unlock(&ci->i_ceph_lock); + up_read(&mdsc->snap_rwsem); mutex_unlock(&session->s_mutex); if (tsession) { mutex_unlock(&tsession->s_mutex); @@ -4132,7 +4173,6 @@ void ceph_handle_caps(struct ceph_mds_session *session, /* lookup ino */ inode = ceph_find_inode(mdsc->fsc->sb, vino); - ci = ceph_inode(inode); dout(" op %s ino %llx.%llx inode %p\n", ceph_cap_op_name(op), vino.ino, vino.snap, inode); @@ -4158,6 +4198,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, } goto flush_cap_releases; } + ci = ceph_inode(inode); /* these will work even if we don't have a cap yet */ switch (op) { @@ -4315,7 +4356,7 @@ static void flush_dirty_session_caps(struct ceph_mds_session *s) i_dirty_item); inode = &ci->vfs_inode; ihold(inode); - dout("flush_dirty_caps %p\n", inode); + dout("flush_dirty_caps %llx.%llx\n", ceph_vinop(inode)); spin_unlock(&mdsc->cap_dirty_lock); ceph_check_caps(ci, CHECK_CAPS_FLUSH, NULL); iput(inode); @@ -4349,7 +4390,7 @@ void ceph_get_fmode(struct ceph_inode_info *ci, int fmode, int count) { struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(ci->vfs_inode.i_sb); int bits = (fmode << 1) | 1; - bool is_opened = false; + bool already_opened = false; int i; if (count == 1) @@ -4357,19 +4398,19 @@ void ceph_get_fmode(struct ceph_inode_info *ci, int fmode, int count) spin_lock(&ci->i_ceph_lock); for (i = 0; i < CEPH_FILE_MODE_BITS; i++) { - if (bits & (1 << i)) - ci->i_nr_by_mode[i] += count; - /* - * If any of the mode ref is larger than 1, + * If any of the mode ref is larger than 0, * that means it has been already opened by * others. Just skip checking the PIN ref. */ - if (i && ci->i_nr_by_mode[i] > 1) - is_opened = true; + if (i && ci->i_nr_by_mode[i]) + already_opened = true; + + if (bits & (1 << i)) + ci->i_nr_by_mode[i] += count; } - if (!is_opened) + if (!already_opened) percpu_counter_inc(&mdsc->metric.opened_inodes); spin_unlock(&ci->i_ceph_lock); } @@ -4560,3 +4601,119 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry, spin_unlock(&dentry->d_lock); return ret; } + +static int remove_capsnaps(struct ceph_mds_client *mdsc, struct inode *inode) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_cap_snap *capsnap; + int capsnap_release = 0; + + lockdep_assert_held(&ci->i_ceph_lock); + + dout("removing capsnaps, ci is %p, inode is %p\n", ci, inode); + + while (!list_empty(&ci->i_cap_snaps)) { + capsnap = list_first_entry(&ci->i_cap_snaps, + struct ceph_cap_snap, ci_item); + __ceph_remove_capsnap(inode, capsnap, NULL, NULL); + ceph_put_snap_context(capsnap->context); + ceph_put_cap_snap(capsnap); + capsnap_release++; + } + wake_up_all(&ci->i_cap_wq); + wake_up_all(&mdsc->cap_flushing_wq); + return capsnap_release; +} + +int ceph_purge_inode_cap(struct inode *inode, struct ceph_cap *cap, bool *invalidate) +{ + struct ceph_fs_client *fsc = ceph_inode_to_client(inode); + struct ceph_mds_client *mdsc = fsc->mdsc; + struct ceph_inode_info *ci = ceph_inode(inode); + bool is_auth; + bool dirty_dropped = false; + int iputs = 0; + + lockdep_assert_held(&ci->i_ceph_lock); + + dout("removing cap %p, ci is %p, inode is %p\n", + cap, ci, &ci->vfs_inode); + + is_auth = (cap == ci->i_auth_cap); + __ceph_remove_cap(cap, false); + if (is_auth) { + struct ceph_cap_flush *cf; + + if (ceph_inode_is_shutdown(inode)) { + if (inode->i_data.nrpages > 0) + *invalidate = true; + if (ci->i_wrbuffer_ref > 0) + mapping_set_error(&inode->i_data, -EIO); + } + + spin_lock(&mdsc->cap_dirty_lock); + + /* trash all of the cap flushes for this inode */ + while (!list_empty(&ci->i_cap_flush_list)) { + cf = list_first_entry(&ci->i_cap_flush_list, + struct ceph_cap_flush, i_list); + list_del_init(&cf->g_list); + list_del_init(&cf->i_list); + if (!cf->is_capsnap) + ceph_free_cap_flush(cf); + } + + if (!list_empty(&ci->i_dirty_item)) { + pr_warn_ratelimited( + " dropping dirty %s state for %p %lld\n", + ceph_cap_string(ci->i_dirty_caps), + inode, ceph_ino(inode)); + ci->i_dirty_caps = 0; + list_del_init(&ci->i_dirty_item); + dirty_dropped = true; + } + if (!list_empty(&ci->i_flushing_item)) { + pr_warn_ratelimited( + " dropping dirty+flushing %s state for %p %lld\n", + ceph_cap_string(ci->i_flushing_caps), + inode, ceph_ino(inode)); + ci->i_flushing_caps = 0; + list_del_init(&ci->i_flushing_item); + mdsc->num_cap_flushing--; + dirty_dropped = true; + } + spin_unlock(&mdsc->cap_dirty_lock); + + if (dirty_dropped) { + mapping_set_error(inode->i_mapping, -EIO); + + if (ci->i_wrbuffer_ref_head == 0 && + ci->i_wr_ref == 0 && + ci->i_dirty_caps == 0 && + ci->i_flushing_caps == 0) { + ceph_put_snap_context(ci->i_head_snapc); + ci->i_head_snapc = NULL; + } + } + + if (atomic_read(&ci->i_filelock_ref) > 0) { + /* make further file lock syscall return -EIO */ + ci->i_ceph_flags |= CEPH_I_ERROR_FILELOCK; + pr_warn_ratelimited(" dropping file locks for %p %lld\n", + inode, ceph_ino(inode)); + } + + if (!ci->i_dirty_caps && ci->i_prealloc_cap_flush) { + cf = ci->i_prealloc_cap_flush; + ci->i_prealloc_cap_flush = NULL; + if (!cf->is_capsnap) + ceph_free_cap_flush(cf); + } + + if (!list_empty(&ci->i_cap_snaps)) + iputs = remove_capsnaps(mdsc, inode); + } + if (dirty_dropped) + ++iputs; + return iputs; +} diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c index 38b78b45811f..bec3c4549c07 100644 --- a/fs/ceph/debugfs.c +++ b/fs/ceph/debugfs.c @@ -146,82 +146,92 @@ static int mdsc_show(struct seq_file *s, void *p) name, total, avg, _min, max, sum); \ } -static int metric_show(struct seq_file *s, void *p) +static int metrics_file_show(struct seq_file *s, void *p) { struct ceph_fs_client *fsc = s->private; - struct ceph_mds_client *mdsc = fsc->mdsc; - struct ceph_client_metric *m = &mdsc->metric; - int nr_caps = 0; - s64 total, sum, avg, min, max, sq; - u64 sum_sz, avg_sz, min_sz, max_sz; + struct ceph_client_metric *m = &fsc->mdsc->metric; - sum = percpu_counter_sum(&m->total_inodes); seq_printf(s, "item total\n"); seq_printf(s, "------------------------------------------\n"); - seq_printf(s, "%-35s%lld / %lld\n", "opened files / total inodes", - atomic64_read(&m->opened_files), sum); - seq_printf(s, "%-35s%lld / %lld\n", "pinned i_caps / total inodes", - atomic64_read(&m->total_caps), sum); - seq_printf(s, "%-35s%lld / %lld\n", "opened inodes / total inodes", - percpu_counter_sum(&m->opened_inodes), sum); - - seq_printf(s, "\n"); + seq_printf(s, "%-35s%lld\n", "total inodes", + percpu_counter_sum(&m->total_inodes)); + seq_printf(s, "%-35s%lld\n", "opened files", + atomic64_read(&m->opened_files)); + seq_printf(s, "%-35s%lld\n", "pinned i_caps", + atomic64_read(&m->total_caps)); + seq_printf(s, "%-35s%lld\n", "opened inodes", + percpu_counter_sum(&m->opened_inodes)); + return 0; +} + +static const char * const metric_str[] = { + "read", + "write", + "metadata", + "copyfrom" +}; +static int metrics_latency_show(struct seq_file *s, void *p) +{ + struct ceph_fs_client *fsc = s->private; + struct ceph_client_metric *cm = &fsc->mdsc->metric; + struct ceph_metric *m; + s64 total, avg, min, max, sq; + int i; + seq_printf(s, "item total avg_lat(us) min_lat(us) max_lat(us) stdev(us)\n"); seq_printf(s, "-----------------------------------------------------------------------------------\n"); - spin_lock(&m->read_metric_lock); - total = m->total_reads; - sum = m->read_latency_sum; - avg = total > 0 ? DIV64_U64_ROUND_CLOSEST(sum, total) : 0; - min = m->read_latency_min; - max = m->read_latency_max; - sq = m->read_latency_sq_sum; - spin_unlock(&m->read_metric_lock); - CEPH_LAT_METRIC_SHOW("read", total, avg, min, max, sq); - - spin_lock(&m->write_metric_lock); - total = m->total_writes; - sum = m->write_latency_sum; - avg = total > 0 ? DIV64_U64_ROUND_CLOSEST(sum, total) : 0; - min = m->write_latency_min; - max = m->write_latency_max; - sq = m->write_latency_sq_sum; - spin_unlock(&m->write_metric_lock); - CEPH_LAT_METRIC_SHOW("write", total, avg, min, max, sq); - - spin_lock(&m->metadata_metric_lock); - total = m->total_metadatas; - sum = m->metadata_latency_sum; - avg = total > 0 ? DIV64_U64_ROUND_CLOSEST(sum, total) : 0; - min = m->metadata_latency_min; - max = m->metadata_latency_max; - sq = m->metadata_latency_sq_sum; - spin_unlock(&m->metadata_metric_lock); - CEPH_LAT_METRIC_SHOW("metadata", total, avg, min, max, sq); - - seq_printf(s, "\n"); + for (i = 0; i < METRIC_MAX; i++) { + m = &cm->metric[i]; + spin_lock(&m->lock); + total = m->total; + avg = m->latency_avg; + min = m->latency_min; + max = m->latency_max; + sq = m->latency_sq_sum; + spin_unlock(&m->lock); + CEPH_LAT_METRIC_SHOW(metric_str[i], total, avg, min, max, sq); + } + + return 0; +} + +static int metrics_size_show(struct seq_file *s, void *p) +{ + struct ceph_fs_client *fsc = s->private; + struct ceph_client_metric *cm = &fsc->mdsc->metric; + struct ceph_metric *m; + s64 total; + u64 sum, avg, min, max; + int i; + seq_printf(s, "item total avg_sz(bytes) min_sz(bytes) max_sz(bytes) total_sz(bytes)\n"); seq_printf(s, "----------------------------------------------------------------------------------------\n"); - spin_lock(&m->read_metric_lock); - total = m->total_reads; - sum_sz = m->read_size_sum; - avg_sz = total > 0 ? DIV64_U64_ROUND_CLOSEST(sum_sz, total) : 0; - min_sz = m->read_size_min; - max_sz = m->read_size_max; - spin_unlock(&m->read_metric_lock); - CEPH_SZ_METRIC_SHOW("read", total, avg_sz, min_sz, max_sz, sum_sz); - - spin_lock(&m->write_metric_lock); - total = m->total_writes; - sum_sz = m->write_size_sum; - avg_sz = total > 0 ? DIV64_U64_ROUND_CLOSEST(sum_sz, total) : 0; - min_sz = m->write_size_min; - max_sz = m->write_size_max; - spin_unlock(&m->write_metric_lock); - CEPH_SZ_METRIC_SHOW("write", total, avg_sz, min_sz, max_sz, sum_sz); - - seq_printf(s, "\n"); + for (i = 0; i < METRIC_MAX; i++) { + /* skip 'metadata' as it doesn't use the size metric */ + if (i == METRIC_METADATA) + continue; + m = &cm->metric[i]; + spin_lock(&m->lock); + total = m->total; + sum = m->size_sum; + avg = total > 0 ? DIV64_U64_ROUND_CLOSEST(sum, total) : 0; + min = m->size_min; + max = m->size_max; + spin_unlock(&m->lock); + CEPH_SZ_METRIC_SHOW(metric_str[i], total, avg, min, max, sum); + } + + return 0; +} + +static int metrics_caps_show(struct seq_file *s, void *p) +{ + struct ceph_fs_client *fsc = s->private; + struct ceph_client_metric *m = &fsc->mdsc->metric; + int nr_caps = 0; + seq_printf(s, "item total miss hit\n"); seq_printf(s, "-------------------------------------------------\n"); @@ -350,8 +360,11 @@ DEFINE_SHOW_ATTRIBUTE(mdsmap); DEFINE_SHOW_ATTRIBUTE(mdsc); DEFINE_SHOW_ATTRIBUTE(caps); DEFINE_SHOW_ATTRIBUTE(mds_sessions); -DEFINE_SHOW_ATTRIBUTE(metric); DEFINE_SHOW_ATTRIBUTE(status); +DEFINE_SHOW_ATTRIBUTE(metrics_file); +DEFINE_SHOW_ATTRIBUTE(metrics_latency); +DEFINE_SHOW_ATTRIBUTE(metrics_size); +DEFINE_SHOW_ATTRIBUTE(metrics_caps); /* @@ -385,8 +398,9 @@ void ceph_fs_debugfs_cleanup(struct ceph_fs_client *fsc) debugfs_remove(fsc->debugfs_mdsmap); debugfs_remove(fsc->debugfs_mds_sessions); debugfs_remove(fsc->debugfs_caps); - debugfs_remove(fsc->debugfs_metric); + debugfs_remove(fsc->debugfs_status); debugfs_remove(fsc->debugfs_mdsc); + debugfs_remove_recursive(fsc->debugfs_metrics_dir); } void ceph_fs_debugfs_init(struct ceph_fs_client *fsc) @@ -426,12 +440,6 @@ void ceph_fs_debugfs_init(struct ceph_fs_client *fsc) fsc, &mdsc_fops); - fsc->debugfs_metric = debugfs_create_file("metrics", - 0400, - fsc->client->debugfs_dir, - fsc, - &metric_fops); - fsc->debugfs_caps = debugfs_create_file("caps", 0400, fsc->client->debugfs_dir, @@ -443,6 +451,18 @@ void ceph_fs_debugfs_init(struct ceph_fs_client *fsc) fsc->client->debugfs_dir, fsc, &status_fops); + + fsc->debugfs_metrics_dir = debugfs_create_dir("metrics", + fsc->client->debugfs_dir); + + debugfs_create_file("file", 0400, fsc->debugfs_metrics_dir, fsc, + &metrics_file_fops); + debugfs_create_file("latency", 0400, fsc->debugfs_metrics_dir, fsc, + &metrics_latency_fops); + debugfs_create_file("size", 0400, fsc->debugfs_metrics_dir, fsc, + &metrics_size_fops); + debugfs_create_file("caps", 0400, fsc->debugfs_metrics_dir, fsc, + &metrics_caps_fops); } diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 133dbd9338e7..eae417d71136 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -145,7 +145,7 @@ __dcache_find_get_entry(struct dentry *parent, u64 idx, return ERR_PTR(-EAGAIN); } /* reading/filling the cache are serialized by - i_mutex, no need to use page lock */ + i_rwsem, no need to use page lock */ unlock_page(cache_ctl->page); cache_ctl->dentries = kmap(cache_ctl->page); } @@ -155,7 +155,7 @@ __dcache_find_get_entry(struct dentry *parent, u64 idx, rcu_read_lock(); spin_lock(&parent->d_lock); /* check i_size again here, because empty directory can be - * marked as complete while not holding the i_mutex. */ + * marked as complete while not holding the i_rwsem. */ if (ceph_dir_is_complete_ordered(dir) && ptr_pos < i_size_read(dir)) dentry = cache_ctl->dentries[cache_ctl->index]; else @@ -478,8 +478,11 @@ more: 2 : (fpos_off(rde->offset) + 1); err = note_last_dentry(dfi, rde->name, rde->name_len, next_offset); - if (err) + if (err) { + ceph_mdsc_put_request(dfi->last_readdir); + dfi->last_readdir = NULL; return err; + } } else if (req->r_reply_info.dir_end) { dfi->next_offset = 2; /* keep last name */ @@ -520,6 +523,12 @@ more: if (!dir_emit(ctx, rde->name, rde->name_len, ceph_present_ino(inode->i_sb, le64_to_cpu(rde->inode.in->ino)), le32_to_cpu(rde->inode.in->mode) >> 12)) { + /* + * NOTE: Here no need to put the 'dfi->last_readdir', + * because when dir_emit stops us it's most likely + * doesn't have enough memory, etc. So for next readdir + * it will continue. + */ dout("filldir stopping us...\n"); return 0; } @@ -671,7 +680,7 @@ struct dentry *ceph_handle_snapdir(struct ceph_mds_request *req, struct dentry *dentry) { struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb); - struct inode *parent = d_inode(dentry->d_parent); /* we hold i_mutex */ + struct inode *parent = d_inode(dentry->d_parent); /* we hold i_rwsem */ /* .snap dir? */ if (ceph_snap(parent) == CEPH_NOSNAP && diff --git a/fs/ceph/export.c b/fs/ceph/export.c index 1d65934c1262..e0fa66ac8b9f 100644 --- a/fs/ceph/export.c +++ b/fs/ceph/export.c @@ -157,6 +157,11 @@ static struct inode *__lookup_inode(struct super_block *sb, u64 ino) ceph_mdsc_put_request(req); if (!inode) return err < 0 ? ERR_PTR(err) : ERR_PTR(-ESTALE); + } else { + if (ceph_inode_is_shutdown(inode)) { + iput(inode); + return ERR_PTR(-ESTALE); + } } return inode; } @@ -223,8 +228,13 @@ static struct dentry *__snapfh_to_dentry(struct super_block *sb, return ERR_PTR(-ESTALE); inode = ceph_find_inode(sb, vino); - if (inode) + if (inode) { + if (ceph_inode_is_shutdown(inode)) { + iput(inode); + return ERR_PTR(-ESTALE); + } return d_obtain_alias(inode); + } req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPINO, USE_ANY_MDS); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index e61018d9764e..8c8226c0feac 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -204,7 +204,10 @@ static int ceph_init_file_info(struct inode *inode, struct file *file, int fmode, bool isdir) { struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_mount_options *opt = + ceph_inode_to_client(&ci->vfs_inode)->mount_options; struct ceph_file_info *fi; + int ret; dout("%s %p %p 0%o (%s)\n", __func__, inode, file, inode->i_mode, isdir ? "dir" : "regular"); @@ -225,6 +228,9 @@ static int ceph_init_file_info(struct inode *inode, struct file *file, if (!fi) return -ENOMEM; + if (opt->flags & CEPH_MOUNT_OPT_NOPAGECACHE) + fi->flags |= CEPH_F_SYNC; + file->private_data = fi; } @@ -235,7 +241,22 @@ static int ceph_init_file_info(struct inode *inode, struct file *file, INIT_LIST_HEAD(&fi->rw_contexts); fi->filp_gen = READ_ONCE(ceph_inode_to_client(inode)->filp_gen); + if ((file->f_mode & FMODE_WRITE) && + ci->i_inline_version != CEPH_INLINE_NONE) { + ret = ceph_uninline_data(file); + if (ret < 0) + goto error; + } + return 0; + +error: + ceph_fscache_unuse_cookie(inode, file->f_mode & FMODE_WRITE); + ceph_put_fmode(ci, fi->fmode, 1); + kmem_cache_free(ceph_file_cachep, fi); + /* wake up anyone waiting for caps on this inode */ + wake_up_all(&ci->i_cap_wq); + return ret; } /* @@ -248,8 +269,7 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode) switch (inode->i_mode & S_IFMT) { case S_IFREG: - ceph_fscache_register_inode_cookie(inode); - ceph_fscache_file_set_cookie(inode, file); + ceph_fscache_use_cookie(inode, file->f_mode & FMODE_WRITE); fallthrough; case S_IFDIR: ret = ceph_init_file_info(inode, file, fmode, @@ -512,51 +532,68 @@ static void restore_deleg_ino(struct inode *dir, u64 ino) } } +static void wake_async_create_waiters(struct inode *inode, + struct ceph_mds_session *session) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + + spin_lock(&ci->i_ceph_lock); + if (ci->i_ceph_flags & CEPH_I_ASYNC_CREATE) { + ci->i_ceph_flags &= ~CEPH_I_ASYNC_CREATE; + wake_up_bit(&ci->i_ceph_flags, CEPH_ASYNC_CREATE_BIT); + } + ceph_kick_flushing_inode_caps(session, ci); + spin_unlock(&ci->i_ceph_lock); +} + static void ceph_async_create_cb(struct ceph_mds_client *mdsc, struct ceph_mds_request *req) { + struct dentry *dentry = req->r_dentry; + struct inode *dinode = d_inode(dentry); + struct inode *tinode = req->r_target_inode; int result = req->r_err ? req->r_err : le32_to_cpu(req->r_reply_info.head->result); + WARN_ON_ONCE(dinode && tinode && dinode != tinode); + + /* MDS changed -- caller must resubmit */ if (result == -EJUKEBOX) goto out; mapping_set_error(req->r_parent->i_mapping, result); if (result) { - struct dentry *dentry = req->r_dentry; int pathlen = 0; u64 base = 0; char *path = ceph_mdsc_build_path(req->r_dentry, &pathlen, &base, 0); + pr_warn("ceph: async create failure path=(%llx)%s result=%d!\n", + base, IS_ERR(path) ? "<<bad>>" : path, result); + ceph_mdsc_free_path(path, pathlen); + ceph_dir_clear_complete(req->r_parent); if (!d_unhashed(dentry)) d_drop(dentry); - /* FIXME: start returning I/O errors on all accesses? */ - pr_warn("ceph: async create failure path=(%llx)%s result=%d!\n", - base, IS_ERR(path) ? "<<bad>>" : path, result); - ceph_mdsc_free_path(path, pathlen); + if (dinode) { + mapping_set_error(dinode->i_mapping, result); + ceph_inode_shutdown(dinode); + wake_async_create_waiters(dinode, req->r_session); + } } - if (req->r_target_inode) { - struct ceph_inode_info *ci = ceph_inode(req->r_target_inode); - u64 ino = ceph_vino(req->r_target_inode).ino; + if (tinode) { + u64 ino = ceph_vino(tinode).ino; if (req->r_deleg_ino != ino) pr_warn("%s: inode number mismatch! err=%d deleg_ino=0x%llx target=0x%llx\n", __func__, req->r_err, req->r_deleg_ino, ino); - mapping_set_error(req->r_target_inode->i_mapping, result); - spin_lock(&ci->i_ceph_lock); - if (ci->i_ceph_flags & CEPH_I_ASYNC_CREATE) { - ci->i_ceph_flags &= ~CEPH_I_ASYNC_CREATE; - wake_up_bit(&ci->i_ceph_flags, CEPH_ASYNC_CREATE_BIT); - } - ceph_kick_flushing_inode_caps(req->r_session, ci); - spin_unlock(&ci->i_ceph_lock); - } else { + mapping_set_error(tinode->i_mapping, result); + wake_async_create_waiters(tinode, req->r_session); + } else if (!result) { pr_warn("%s: no req->r_target_inode for 0x%llx\n", __func__, req->r_deleg_ino); } @@ -577,6 +614,7 @@ static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry, struct ceph_inode_info *ci = ceph_inode(dir); struct inode *inode; struct timespec64 now; + struct ceph_string *pool_ns; struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(dir->i_sb); struct ceph_vino vino = { .ino = req->r_deleg_ino, .snap = CEPH_NOSNAP }; @@ -591,9 +629,15 @@ static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry, iinfo.change_attr = 1; ceph_encode_timespec64(&iinfo.btime, &now); - iinfo.xattr_len = ARRAY_SIZE(xattr_buf); - iinfo.xattr_data = xattr_buf; - memset(iinfo.xattr_data, 0, iinfo.xattr_len); + if (req->r_pagelist) { + iinfo.xattr_len = req->r_pagelist->length; + iinfo.xattr_data = req->r_pagelist->mapped_tail; + } else { + /* fake it */ + iinfo.xattr_len = ARRAY_SIZE(xattr_buf); + iinfo.xattr_data = xattr_buf; + memset(iinfo.xattr_data, 0, iinfo.xattr_len); + } in.ino = cpu_to_le64(vino.ino); in.snapid = cpu_to_le64(CEPH_NOSNAP); @@ -603,17 +647,35 @@ static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry, in.cap.realm = cpu_to_le64(ci->i_snap_realm->ino); in.cap.flags = CEPH_CAP_FLAG_AUTH; in.ctime = in.mtime = in.atime = iinfo.btime; - in.mode = cpu_to_le32((u32)mode); in.truncate_seq = cpu_to_le32(1); in.truncate_size = cpu_to_le64(-1ULL); in.xattr_version = cpu_to_le64(1); in.uid = cpu_to_le32(from_kuid(&init_user_ns, current_fsuid())); - in.gid = cpu_to_le32(from_kgid(&init_user_ns, dir->i_mode & S_ISGID ? - dir->i_gid : current_fsgid())); + if (dir->i_mode & S_ISGID) { + in.gid = cpu_to_le32(from_kgid(&init_user_ns, dir->i_gid)); + + /* Directories always inherit the setgid bit. */ + if (S_ISDIR(mode)) + mode |= S_ISGID; + else if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP) && + !in_group_p(dir->i_gid) && + !capable_wrt_inode_uidgid(&init_user_ns, dir, CAP_FSETID)) + mode &= ~S_ISGID; + } else { + in.gid = cpu_to_le32(from_kgid(&init_user_ns, current_fsgid())); + } + in.mode = cpu_to_le32((u32)mode); + in.nlink = cpu_to_le32(1); in.max_size = cpu_to_le64(lo->stripe_unit); ceph_file_layout_to_legacy(lo, &in.layout); + /* lo is private, so pool_ns can't change */ + pool_ns = rcu_dereference_raw(lo->pool_ns); + if (pool_ns) { + iinfo.pool_ns_len = pool_ns->len; + iinfo.pool_ns_data = pool_ns->str; + } down_read(&mdsc->snap_rwsem); ret = ceph_fill_inode(inode, NULL, &iinfo, NULL, req->r_session, @@ -687,6 +749,10 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, err = ceph_security_init_secctx(dentry, mode, &as_ctx); if (err < 0) goto out_ctx; + /* Async create can't handle more than a page of xattrs */ + if (as_ctx.pagelist && + !list_is_singular(&as_ctx.pagelist->head)) + try_async = false; } else if (!d_in_lookup(dentry)) { /* If it's not being looked up, it's negative */ return -ENOENT; @@ -732,8 +798,10 @@ retry: restore_deleg_ino(dir, req->r_deleg_ino); ceph_mdsc_put_request(req); try_async = false; + ceph_put_string(rcu_dereference_raw(lo.pool_ns)); goto retry; } + ceph_put_string(rcu_dereference_raw(lo.pool_ns)); goto out_req; } } @@ -808,6 +876,7 @@ int ceph_release(struct inode *inode, struct file *file) dout("release inode %p regular file %p\n", inode, file); WARN_ON(!list_empty(&fi->rw_contexts)); + ceph_fscache_unuse_cookie(inode, file->f_mode & FMODE_WRITE); ceph_put_fmode(ci, fi->fmode, 1); kmem_cache_free(ceph_file_cachep, fi); @@ -845,6 +914,7 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *to, ssize_t ret; u64 off = iocb->ki_pos; u64 len = iov_iter_count(to); + u64 i_size = i_size_read(inode); dout("sync_read on file %p %llu~%u %s\n", file, off, (unsigned)len, (file->f_flags & O_DIRECT) ? "O_DIRECT" : ""); @@ -868,7 +938,6 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *to, struct page **pages; int num_pages; size_t page_off; - u64 i_size; bool more; int idx; size_t left; @@ -951,11 +1020,14 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *to, } if (off > iocb->ki_pos) { - if (ret >= 0 && - iov_iter_count(to) > 0 && off >= i_size_read(inode)) + if (off >= i_size) { *retry_op = CHECK_EOF; - ret = off - iocb->ki_pos; - iocb->ki_pos = off; + ret = i_size - iocb->ki_pos; + iocb->ki_pos = i_size; + } else { + ret = off - iocb->ki_pos; + iocb->ki_pos = off; + } } dout("sync_read result %zd retry_op %d\n", ret, *retry_op); @@ -1010,7 +1082,6 @@ static void ceph_aio_complete(struct inode *inode, } spin_lock(&ci->i_ceph_lock); - ci->i_inline_version = CEPH_INLINE_NONE; dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR, &aio_req->prealloc_cf); spin_unlock(&ci->i_ceph_lock); @@ -1022,7 +1093,7 @@ static void ceph_aio_complete(struct inode *inode, ceph_put_cap_refs(ci, (aio_req->write ? CEPH_CAP_FILE_WR : CEPH_CAP_FILE_RD)); - aio_req->iocb->ki_complete(aio_req->iocb, ret, 0); + aio_req->iocb->ki_complete(aio_req->iocb, ret); ceph_free_cap_flush(aio_req->prealloc_cf); kfree(aio_req); @@ -1201,7 +1272,11 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, snapc, snapc ? snapc->seq : 0); if (write) { - int ret2 = invalidate_inode_pages2_range(inode->i_mapping, + int ret2; + + ceph_fscache_invalidate(inode, true); + + ret2 = invalidate_inode_pages2_range(inode->i_mapping, pos >> PAGE_SHIFT, (pos + count - 1) >> PAGE_SHIFT); if (ret2 < 0) @@ -1412,6 +1487,7 @@ ceph_sync_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos, if (ret < 0) return ret; + ceph_fscache_invalidate(inode, false); ret = invalidate_inode_pages2_range(inode->i_mapping, pos >> PAGE_SHIFT, (pos + count - 1) >> PAGE_SHIFT); @@ -1519,25 +1595,29 @@ static ssize_t ceph_read_iter(struct kiocb *iocb, struct iov_iter *to) struct ceph_inode_info *ci = ceph_inode(inode); bool direct_lock = iocb->ki_flags & IOCB_DIRECT; ssize_t ret; - int want, got = 0; + int want = 0, got = 0; int retry_op = 0, read = 0; again: dout("aio_read %p %llx.%llx %llu~%u trying to get caps on %p\n", inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len, inode); + if (ceph_inode_is_shutdown(inode)) + return -ESTALE; + if (direct_lock) ceph_start_io_direct(inode); else ceph_start_io_read(inode); + if (!(fi->flags & CEPH_F_SYNC) && !direct_lock) + want |= CEPH_CAP_FILE_CACHE; if (fi->fmode & CEPH_FILE_MODE_LAZY) - want = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO; - else - want = CEPH_CAP_FILE_CACHE; + want |= CEPH_CAP_FILE_LAZYIO; + ret = ceph_get_caps(filp, CEPH_CAP_FILE_RD, want, -1, &got); if (ret < 0) { - if (iocb->ki_flags & IOCB_DIRECT) + if (direct_lock) ceph_end_io_direct(inode); else ceph_end_io_read(inode); @@ -1671,13 +1751,16 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) struct ceph_osd_client *osdc = &fsc->client->osdc; struct ceph_cap_flush *prealloc_cf; ssize_t count, written = 0; - int err, want, got; + int err, want = 0, got; bool direct_lock = false; u32 map_flags; u64 pool_flags; loff_t pos; loff_t limit = max(i_size_read(inode), fsc->max_file_size); + if (ceph_inode_is_shutdown(inode)) + return -ESTALE; + if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; @@ -1735,18 +1818,12 @@ retry_snap: if (err) goto out; - if (ci->i_inline_version != CEPH_INLINE_NONE) { - err = ceph_uninline_data(file, NULL); - if (err < 0) - goto out; - } - dout("aio_write %p %llx.%llx %llu~%zd getting caps. i_size %llu\n", inode, ceph_vinop(inode), pos, count, i_size_read(inode)); + if (!(fi->flags & CEPH_F_SYNC) && !direct_lock) + want |= CEPH_CAP_FILE_BUFFER; if (fi->fmode & CEPH_FILE_MODE_LAZY) - want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO; - else - want = CEPH_CAP_FILE_BUFFER; + want |= CEPH_CAP_FILE_LAZYIO; got = 0; err = ceph_get_caps(file, CEPH_CAP_FILE_WR, want, pos + count, &got); if (err < 0) @@ -1802,7 +1879,7 @@ retry_snap: * are pending vmtruncate. So write and vmtruncate * can not run at the same time */ - written = generic_perform_write(file, from, pos); + written = generic_perform_write(iocb, from); if (likely(written >= 0)) iocb->ki_pos = pos + written; ceph_end_io_write(inode); @@ -1812,7 +1889,6 @@ retry_snap: int dirty; spin_lock(&ci->i_ceph_lock); - ci->i_inline_version = CEPH_INLINE_NONE; dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR, &prealloc_cf); spin_unlock(&ci->i_ceph_lock); @@ -2066,12 +2142,6 @@ static long ceph_fallocate(struct file *file, int mode, goto unlock; } - if (ci->i_inline_version != CEPH_INLINE_NONE) { - ret = ceph_uninline_data(file, NULL); - if (ret < 0) - goto unlock; - } - size = i_size_read(inode); /* Are we punching a hole beyond EOF? */ @@ -2090,12 +2160,12 @@ static long ceph_fallocate(struct file *file, int mode, goto unlock; filemap_invalidate_lock(inode->i_mapping); + ceph_fscache_invalidate(inode, false); ceph_zero_pagecache_range(inode, offset, length); ret = ceph_zero_objects(inode, offset, length); if (!ret) { spin_lock(&ci->i_ceph_lock); - ci->i_inline_version = CEPH_INLINE_NONE; dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR, &prealloc_cf); spin_unlock(&ci->i_ceph_lock); @@ -2200,6 +2270,54 @@ static int is_file_size_ok(struct inode *src_inode, struct inode *dst_inode, return 0; } +static struct ceph_osd_request * +ceph_alloc_copyfrom_request(struct ceph_osd_client *osdc, + u64 src_snapid, + struct ceph_object_id *src_oid, + struct ceph_object_locator *src_oloc, + struct ceph_object_id *dst_oid, + struct ceph_object_locator *dst_oloc, + u32 truncate_seq, u64 truncate_size) +{ + struct ceph_osd_request *req; + int ret; + u32 src_fadvise_flags = + CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL | + CEPH_OSD_OP_FLAG_FADVISE_NOCACHE; + u32 dst_fadvise_flags = + CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL | + CEPH_OSD_OP_FLAG_FADVISE_DONTNEED; + + req = ceph_osdc_alloc_request(osdc, NULL, 1, false, GFP_KERNEL); + if (!req) + return ERR_PTR(-ENOMEM); + + req->r_flags = CEPH_OSD_FLAG_WRITE; + + ceph_oloc_copy(&req->r_t.base_oloc, dst_oloc); + ceph_oid_copy(&req->r_t.base_oid, dst_oid); + + ret = osd_req_op_copy_from_init(req, src_snapid, 0, + src_oid, src_oloc, + src_fadvise_flags, + dst_fadvise_flags, + truncate_seq, + truncate_size, + CEPH_OSD_COPY_FROM_FLAG_TRUNCATE_SEQ); + if (ret) + goto out; + + ret = ceph_osdc_alloc_messages(req, GFP_KERNEL); + if (ret) + goto out; + + return req; + +out: + ceph_osdc_put_request(req); + return ERR_PTR(ret); +} + static ssize_t ceph_do_objects_copy(struct ceph_inode_info *src_ci, u64 *src_off, struct ceph_inode_info *dst_ci, u64 *dst_off, struct ceph_fs_client *fsc, @@ -2207,6 +2325,8 @@ static ssize_t ceph_do_objects_copy(struct ceph_inode_info *src_ci, u64 *src_off { struct ceph_object_locator src_oloc, dst_oloc; struct ceph_object_id src_oid, dst_oid; + struct ceph_osd_client *osdc; + struct ceph_osd_request *req; size_t bytes = 0; u64 src_objnum, src_objoff, dst_objnum, dst_objoff; u32 src_objlen, dst_objlen; @@ -2217,6 +2337,7 @@ static ssize_t ceph_do_objects_copy(struct ceph_inode_info *src_ci, u64 *src_off src_oloc.pool_ns = ceph_try_get_string(src_ci->i_layout.pool_ns); dst_oloc.pool = dst_ci->i_layout.pool_id; dst_oloc.pool_ns = ceph_try_get_string(dst_ci->i_layout.pool_ns); + osdc = &fsc->client->osdc; while (len >= object_size) { ceph_calc_file_object_mapping(&src_ci->i_layout, *src_off, @@ -2232,17 +2353,22 @@ static ssize_t ceph_do_objects_copy(struct ceph_inode_info *src_ci, u64 *src_off ceph_oid_printf(&dst_oid, "%llx.%08llx", dst_ci->i_vino.ino, dst_objnum); /* Do an object remote copy */ - ret = ceph_osdc_copy_from(&fsc->client->osdc, - src_ci->i_vino.snap, 0, - &src_oid, &src_oloc, - CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL | - CEPH_OSD_OP_FLAG_FADVISE_NOCACHE, - &dst_oid, &dst_oloc, - CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL | - CEPH_OSD_OP_FLAG_FADVISE_DONTNEED, - dst_ci->i_truncate_seq, - dst_ci->i_truncate_size, - CEPH_OSD_COPY_FROM_FLAG_TRUNCATE_SEQ); + req = ceph_alloc_copyfrom_request(osdc, src_ci->i_vino.snap, + &src_oid, &src_oloc, + &dst_oid, &dst_oloc, + dst_ci->i_truncate_seq, + dst_ci->i_truncate_size); + if (IS_ERR(req)) + ret = PTR_ERR(req); + else { + ceph_osdc_start_request(osdc, req, false); + ret = ceph_osdc_wait_request(osdc, req); + ceph_update_copyfrom_metrics(&fsc->mdsc->metric, + req->r_start_latency, + req->r_end_latency, + object_size, ret); + ceph_osdc_put_request(req); + } if (ret) { if (ret == -EOPNOTSUPP) { fsc->have_copy_from2 = false; @@ -2358,6 +2484,7 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off, goto out_caps; /* Drop dst file cached pages */ + ceph_fscache_invalidate(dst_inode, false); ret = invalidate_inode_pages2_range(dst_inode->i_mapping, dst_off >> PAGE_SHIFT, (dst_off + len) >> PAGE_SHIFT); @@ -2431,7 +2558,6 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off, } /* Mark Fw dirty */ spin_lock(&dst_ci->i_ceph_lock); - dst_ci->i_inline_version = CEPH_INLINE_NONE; dirty = __ceph_mark_dirty_caps(dst_ci, CEPH_CAP_FILE_WR, &prealloc_cf); spin_unlock(&dst_ci->i_ceph_lock); if (dirty) diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 1c7574105478..63113e2a4890 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -87,13 +87,13 @@ struct inode *ceph_get_snapdir(struct inode *parent) if (!S_ISDIR(parent->i_mode)) { pr_warn_once("bad snapdir parent type (mode=0%o)\n", parent->i_mode); - return ERR_PTR(-ENOTDIR); + goto err; } if (!(inode->i_state & I_NEW) && !S_ISDIR(inode->i_mode)) { pr_warn_once("bad snapdir inode type (mode=0%o)\n", inode->i_mode); - return ERR_PTR(-ENOTDIR); + goto err; } inode->i_mode = parent->i_mode; @@ -113,6 +113,12 @@ struct inode *ceph_get_snapdir(struct inode *parent) } return inode; +err: + if ((inode->i_state & I_NEW)) + discard_new_inode(inode); + else + iput(inode); + return ERR_PTR(-ENOTDIR); } const struct inode_operations ceph_file_iops = { @@ -447,12 +453,15 @@ struct inode *ceph_alloc_inode(struct super_block *sb) struct ceph_inode_info *ci; int i; - ci = kmem_cache_alloc(ceph_inode_cachep, GFP_NOFS); + ci = alloc_inode_sb(sb, ceph_inode_cachep, GFP_NOFS); if (!ci) return NULL; dout("alloc_inode %p\n", &ci->vfs_inode); + /* Set parameters for the netfs library */ + netfs_i_context_init(&ci->vfs_inode, &ceph_netfs_ops); + spin_lock_init(&ci->i_ceph_lock); ci->i_version = 0; @@ -538,9 +547,6 @@ struct inode *ceph_alloc_inode(struct super_block *sb) INIT_WORK(&ci->i_work, ceph_inode_work); ci->i_work_mask = 0; memset(&ci->i_btime, '\0', sizeof(ci->i_btime)); - - ceph_fscache_inode_init(ci); - return &ci->vfs_inode; } @@ -564,6 +570,8 @@ void ceph_evict_inode(struct inode *inode) percpu_counter_dec(&mdsc->metric.total_inodes); truncate_inode_pages_final(&inode->i_data); + if (inode->i_state & I_PINNING_FSCACHE_WB) + ceph_fscache_unuse_cookie(inode, true); clear_inode(inode); ceph_fscache_unregister_inode_cookie(ci); @@ -634,6 +642,12 @@ int ceph_fill_file_size(struct inode *inode, int issued, } i_size_write(inode, size); inode->i_blocks = calc_inode_blocks(size); + /* + * If we're expanding, then we should be able to just update + * the existing cookie. + */ + if (size > isize) + ceph_fscache_update(inode); ci->i_reported_size = size; if (truncate_seq != ci->i_truncate_seq) { dout("truncate_seq %u -> %u\n", @@ -666,10 +680,6 @@ int ceph_fill_file_size(struct inode *inode, int issued, truncate_size); ci->i_truncate_size = truncate_size; } - - if (queue_trunc) - ceph_fscache_invalidate(inode); - return queue_trunc; } @@ -1053,6 +1063,8 @@ int ceph_fill_inode(struct inode *inode, struct page *locked_page, spin_unlock(&ci->i_ceph_lock); + ceph_fscache_register_inode_cookie(inode); + if (fill_inline) ceph_fill_inline_data(inode, locked_page, iinfo->inline_data, iinfo->inline_len); @@ -1195,7 +1207,7 @@ out_unlock: /* * splice a dentry to an inode. - * caller must hold directory i_mutex for this to be safe. + * caller must hold directory i_rwsem for this to be safe. */ static int splice_dentry(struct dentry **pdn, struct inode *in) { @@ -1592,7 +1604,7 @@ static int fill_readdir_cache(struct inode *dir, struct dentry *dn, return idx == 0 ? -ENOMEM : 0; } /* reading/filling the cache are serialized by - * i_mutex, no need to use page lock */ + * i_rwsem, no need to use page lock */ unlock_page(ctl->page); ctl->dentries = kmap(ctl->page); if (idx == 0) @@ -1814,11 +1826,13 @@ bool ceph_inode_set_size(struct inode *inode, loff_t size) spin_lock(&ci->i_ceph_lock); dout("set_size %p %llu -> %llu\n", inode, i_size_read(inode), size); i_size_write(inode, size); + ceph_fscache_update(inode); inode->i_blocks = calc_inode_blocks(size); ret = __ceph_should_report_size(ci); spin_unlock(&ci->i_ceph_lock); + return ret; } @@ -1841,15 +1855,16 @@ void ceph_queue_inode_work(struct inode *inode, int work_bit) static void ceph_do_invalidate_pages(struct inode *inode) { struct ceph_inode_info *ci = ceph_inode(inode); - struct ceph_fs_client *fsc = ceph_inode_to_client(inode); u32 orig_gen; int check = 0; + ceph_fscache_invalidate(inode, false); + mutex_lock(&ci->i_truncate_mutex); - if (READ_ONCE(fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN) { - pr_warn_ratelimited("invalidate_pages %p %lld forced umount\n", - inode, ceph_ino(inode)); + if (ceph_inode_is_shutdown(inode)) { + pr_warn_ratelimited("%s: inode %llx.%llx is shut down\n", + __func__, ceph_vinop(inode)); mapping_set_error(inode->i_mapping, -EIO); truncate_pagecache(inode, 0); mutex_unlock(&ci->i_truncate_mutex); @@ -1869,9 +1884,10 @@ static void ceph_do_invalidate_pages(struct inode *inode) orig_gen = ci->i_rdcache_gen; spin_unlock(&ci->i_ceph_lock); - ceph_fscache_invalidate(inode); + ceph_fscache_invalidate(inode, false); if (invalidate_inode_pages2(inode->i_mapping) < 0) { - pr_err("invalidate_pages %p fails\n", inode); + pr_err("invalidate_inode_pages2 %llx.%llx failed\n", + ceph_vinop(inode)); } spin_lock(&ci->i_ceph_lock); @@ -1937,6 +1953,7 @@ retry: ci->i_truncate_pending, to); spin_unlock(&ci->i_ceph_lock); + ceph_fscache_resize(inode, to); truncate_pagecache(inode, to); spin_lock(&ci->i_ceph_lock); @@ -2103,12 +2120,14 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr) loff_t isize = i_size_read(inode); dout("setattr %p size %lld -> %lld\n", inode, isize, attr->ia_size); - if ((issued & CEPH_CAP_FILE_EXCL) && attr->ia_size > isize) { - i_size_write(inode, attr->ia_size); - inode->i_blocks = calc_inode_blocks(attr->ia_size); - ci->i_reported_size = attr->ia_size; - dirtied |= CEPH_CAP_FILE_EXCL; - ia_valid |= ATTR_MTIME; + if ((issued & CEPH_CAP_FILE_EXCL) && attr->ia_size >= isize) { + if (attr->ia_size > isize) { + i_size_write(inode, attr->ia_size); + inode->i_blocks = calc_inode_blocks(attr->ia_size); + ci->i_reported_size = attr->ia_size; + dirtied |= CEPH_CAP_FILE_EXCL; + ia_valid |= ATTR_MTIME; + } } else if ((issued & CEPH_CAP_FILE_SHARED) == 0 || attr->ia_size != isize) { req->r_args.setattr.size = cpu_to_le64(attr->ia_size); @@ -2182,7 +2201,6 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr) if (inode_dirty_flags) __mark_inode_dirty(inode, inode_dirty_flags); - if (mask) { req->r_inode = inode; ihold(inode); @@ -2217,6 +2235,9 @@ int ceph_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; + if (ceph_inode_is_shutdown(inode)) + return -ESTALE; + err = setattr_prepare(&init_user_ns, dentry, attr); if (err != 0) return err; @@ -2286,6 +2307,57 @@ int __ceph_do_getattr(struct inode *inode, struct page *locked_page, return err; } +int ceph_do_getvxattr(struct inode *inode, const char *name, void *value, + size_t size) +{ + struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb); + struct ceph_mds_client *mdsc = fsc->mdsc; + struct ceph_mds_request *req; + int mode = USE_AUTH_MDS; + int err; + char *xattr_value; + size_t xattr_value_len; + + req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_GETVXATTR, mode); + if (IS_ERR(req)) { + err = -ENOMEM; + goto out; + } + + req->r_path2 = kstrdup(name, GFP_NOFS); + if (!req->r_path2) { + err = -ENOMEM; + goto put; + } + + ihold(inode); + req->r_inode = inode; + err = ceph_mdsc_do_request(mdsc, NULL, req); + if (err < 0) + goto put; + + xattr_value = req->r_reply_info.xattr_info.xattr_value; + xattr_value_len = req->r_reply_info.xattr_info.xattr_value_len; + + dout("do_getvxattr xattr_value_len:%zu, size:%zu\n", xattr_value_len, size); + + err = (int)xattr_value_len; + if (size == 0) + goto put; + + if (xattr_value_len > size) { + err = -ERANGE; + goto put; + } + + memcpy(value, xattr_value, xattr_value_len); +put: + ceph_mdsc_put_request(req); +out: + dout("do_getvxattr result=%d\n", err); + return err; +} + /* * Check inode permissions. We verify we have a valid value for @@ -2347,6 +2419,9 @@ int ceph_getattr(struct user_namespace *mnt_userns, const struct path *path, u32 valid_mask = STATX_BASIC_STATS; int err = 0; + if (ceph_inode_is_shutdown(inode)) + return -ESTALE; + /* Skip the getattr altogether if we're asked not to sync */ if (!(flags & AT_STATX_DONT_SYNC)) { err = ceph_do_getattr(inode, @@ -2394,3 +2469,27 @@ int ceph_getattr(struct user_namespace *mnt_userns, const struct path *path, stat->result_mask = request_mask & valid_mask; return err; } + +void ceph_inode_shutdown(struct inode *inode) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + struct rb_node *p; + int iputs = 0; + bool invalidate = false; + + spin_lock(&ci->i_ceph_lock); + ci->i_ceph_flags |= CEPH_I_SHUTDOWN; + p = rb_first(&ci->i_caps); + while (p) { + struct ceph_cap *cap = rb_entry(p, struct ceph_cap, ci_node); + + p = rb_next(p); + iputs += ceph_purge_inode_cap(inode, cap, &invalidate); + } + spin_unlock(&ci->i_ceph_lock); + + if (invalidate) + ceph_queue_invalidate(inode); + while (iputs--) + iput(inode); +} diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index bdeb271f47d9..3e2843e86e27 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -111,10 +111,10 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode, req->r_args.filelock_change.length = cpu_to_le64(length); req->r_args.filelock_change.wait = wait; - if (wait) - req->r_wait_for_completion = ceph_lock_wait_for_completion; - - err = ceph_mdsc_do_request(mdsc, inode, req); + err = ceph_mdsc_submit_request(mdsc, inode, req); + if (!err) + err = ceph_mdsc_wait_request(mdsc, req, wait ? + ceph_lock_wait_for_completion : NULL); if (!err && operation == CEPH_MDS_OP_GETFILELOCK) { fl->fl_pid = -le64_to_cpu(req->r_reply_info.filelock_reply->pid); if (CEPH_LOCK_SHARED == req->r_reply_info.filelock_reply->type) @@ -241,6 +241,9 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) if (!(fl->fl_flags & FL_POSIX)) return -ENOLCK; + if (ceph_inode_is_shutdown(inode)) + return -ESTALE; + dout("ceph_lock, fl_owner: %p\n", fl->fl_owner); /* set wait bit as appropriate, then make command as Ceph expects it*/ @@ -302,9 +305,9 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) if (!(fl->fl_flags & FL_FLOCK)) return -ENOLCK; - /* No mandatory locks */ - if (fl->fl_type & LOCK_MAND) - return -EOPNOTSUPP; + + if (ceph_inode_is_shutdown(inode)) + return -ESTALE; dout("ceph_flock, fl_file: %p\n", fl->fl_file); diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index d64413adc0fd..00c3de177dd6 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -555,6 +555,28 @@ bad: return -EIO; } +static int parse_reply_info_getvxattr(void **p, void *end, + struct ceph_mds_reply_info_parsed *info, + u64 features) +{ + u32 value_len; + + ceph_decode_skip_8(p, end, bad); /* skip current version: 1 */ + ceph_decode_skip_8(p, end, bad); /* skip first version: 1 */ + ceph_decode_skip_32(p, end, bad); /* skip payload length */ + + ceph_decode_32_safe(p, end, value_len, bad); + + if (value_len == end - *p) { + info->xattr_info.xattr_value = *p; + info->xattr_info.xattr_value_len = value_len; + *p = end; + return value_len; + } +bad: + return -EIO; +} + /* * parse extra results */ @@ -570,6 +592,8 @@ static int parse_reply_info_extra(void **p, void *end, return parse_reply_info_readdir(p, end, info, features); else if (op == CEPH_MDS_OP_CREATE) return parse_reply_info_create(p, end, info, features, s); + else if (op == CEPH_MDS_OP_GETVXATTR) + return parse_reply_info_getvxattr(p, end, info, features); else return -EIO; } @@ -1590,129 +1614,23 @@ out: return ret; } -static int remove_capsnaps(struct ceph_mds_client *mdsc, struct inode *inode) -{ - struct ceph_inode_info *ci = ceph_inode(inode); - struct ceph_cap_snap *capsnap; - int capsnap_release = 0; - - lockdep_assert_held(&ci->i_ceph_lock); - - dout("removing capsnaps, ci is %p, inode is %p\n", ci, inode); - - while (!list_empty(&ci->i_cap_snaps)) { - capsnap = list_first_entry(&ci->i_cap_snaps, - struct ceph_cap_snap, ci_item); - __ceph_remove_capsnap(inode, capsnap, NULL, NULL); - ceph_put_snap_context(capsnap->context); - ceph_put_cap_snap(capsnap); - capsnap_release++; - } - wake_up_all(&ci->i_cap_wq); - wake_up_all(&mdsc->cap_flushing_wq); - return capsnap_release; -} - static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg) { - struct ceph_fs_client *fsc = (struct ceph_fs_client *)arg; - struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_inode_info *ci = ceph_inode(inode); - LIST_HEAD(to_remove); - bool dirty_dropped = false; bool invalidate = false; - int capsnap_release = 0; + int iputs; dout("removing cap %p, ci is %p, inode is %p\n", cap, ci, &ci->vfs_inode); spin_lock(&ci->i_ceph_lock); - __ceph_remove_cap(cap, false); - if (!ci->i_auth_cap) { - struct ceph_cap_flush *cf; - - if (READ_ONCE(fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN) { - if (inode->i_data.nrpages > 0) - invalidate = true; - if (ci->i_wrbuffer_ref > 0) - mapping_set_error(&inode->i_data, -EIO); - } - - while (!list_empty(&ci->i_cap_flush_list)) { - cf = list_first_entry(&ci->i_cap_flush_list, - struct ceph_cap_flush, i_list); - list_move(&cf->i_list, &to_remove); - } - - spin_lock(&mdsc->cap_dirty_lock); - - list_for_each_entry(cf, &to_remove, i_list) - list_del_init(&cf->g_list); - - if (!list_empty(&ci->i_dirty_item)) { - pr_warn_ratelimited( - " dropping dirty %s state for %p %lld\n", - ceph_cap_string(ci->i_dirty_caps), - inode, ceph_ino(inode)); - ci->i_dirty_caps = 0; - list_del_init(&ci->i_dirty_item); - dirty_dropped = true; - } - if (!list_empty(&ci->i_flushing_item)) { - pr_warn_ratelimited( - " dropping dirty+flushing %s state for %p %lld\n", - ceph_cap_string(ci->i_flushing_caps), - inode, ceph_ino(inode)); - ci->i_flushing_caps = 0; - list_del_init(&ci->i_flushing_item); - mdsc->num_cap_flushing--; - dirty_dropped = true; - } - spin_unlock(&mdsc->cap_dirty_lock); - - if (dirty_dropped) { - mapping_set_error(inode->i_mapping, -EIO); - - if (ci->i_wrbuffer_ref_head == 0 && - ci->i_wr_ref == 0 && - ci->i_dirty_caps == 0 && - ci->i_flushing_caps == 0) { - ceph_put_snap_context(ci->i_head_snapc); - ci->i_head_snapc = NULL; - } - } - - if (atomic_read(&ci->i_filelock_ref) > 0) { - /* make further file lock syscall return -EIO */ - ci->i_ceph_flags |= CEPH_I_ERROR_FILELOCK; - pr_warn_ratelimited(" dropping file locks for %p %lld\n", - inode, ceph_ino(inode)); - } - - if (!ci->i_dirty_caps && ci->i_prealloc_cap_flush) { - list_add(&ci->i_prealloc_cap_flush->i_list, &to_remove); - ci->i_prealloc_cap_flush = NULL; - } - - if (!list_empty(&ci->i_cap_snaps)) - capsnap_release = remove_capsnaps(mdsc, inode); - } + iputs = ceph_purge_inode_cap(inode, cap, &invalidate); spin_unlock(&ci->i_ceph_lock); - while (!list_empty(&to_remove)) { - struct ceph_cap_flush *cf; - cf = list_first_entry(&to_remove, - struct ceph_cap_flush, i_list); - list_del_init(&cf->i_list); - if (!cf->is_capsnap) - ceph_free_cap_flush(cf); - } wake_up_all(&ci->i_cap_wq); if (invalidate) ceph_queue_invalidate(inode); - if (dirty_dropped) - iput(inode); - while (capsnap_release--) + while (iputs--) iput(inode); return 0; } @@ -2284,7 +2202,8 @@ int ceph_alloc_readdir_reply_buffer(struct ceph_mds_request *req, order = get_order(size * num_entries); while (order >= 0) { rinfo->dir_entries = (void*)__get_free_pages(GFP_KERNEL | - __GFP_NOWARN, + __GFP_NOWARN | + __GFP_ZERO, order); if (rinfo->dir_entries) break; @@ -3052,15 +2971,16 @@ int ceph_mdsc_submit_request(struct ceph_mds_client *mdsc, struct inode *dir, return err; } -static int ceph_mdsc_wait_request(struct ceph_mds_client *mdsc, - struct ceph_mds_request *req) +int ceph_mdsc_wait_request(struct ceph_mds_client *mdsc, + struct ceph_mds_request *req, + ceph_mds_request_wait_callback_t wait_func) { int err; /* wait */ dout("do_request waiting\n"); - if (!req->r_timeout && req->r_wait_for_completion) { - err = req->r_wait_for_completion(mdsc, req); + if (wait_func) { + err = wait_func(mdsc, req); } else { long timeleft = wait_for_completion_killable_timeout( &req->r_completion, @@ -3117,7 +3037,7 @@ int ceph_mdsc_do_request(struct ceph_mds_client *mdsc, /* issue */ err = ceph_mdsc_submit_request(mdsc, dir, req); if (!err) - err = ceph_mdsc_wait_request(mdsc, req); + err = ceph_mdsc_wait_request(mdsc, req, NULL); dout("do_request %p done, result %d\n", req, err); return err; } @@ -3203,35 +3123,6 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg) result = le32_to_cpu(head->result); - /* - * Handle an ESTALE - * if we're not talking to the authority, send to them - * if the authority has changed while we weren't looking, - * send to new authority - * Otherwise we just have to return an ESTALE - */ - if (result == -ESTALE) { - dout("got ESTALE on request %llu\n", req->r_tid); - req->r_resend_mds = -1; - if (req->r_direct_mode != USE_AUTH_MDS) { - dout("not using auth, setting for that now\n"); - req->r_direct_mode = USE_AUTH_MDS; - __do_request(mdsc, req); - mutex_unlock(&mdsc->mutex); - goto out; - } else { - int mds = __choose_mds(mdsc, req, NULL); - if (mds >= 0 && mds != req->r_session->s_mds) { - dout("but auth changed, so resending\n"); - __do_request(mdsc, req); - mutex_unlock(&mdsc->mutex); - goto out; - } - } - dout("have to return ESTALE on request %llu\n", req->r_tid); - } - - if (head->safe) { set_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags); __unregister_request(mdsc, req); @@ -3467,9 +3358,14 @@ static void handle_session(struct ceph_mds_session *session, if (msg_version >= 3) { u32 len; - /* version >= 2, metadata */ - if (__decode_session_metadata(&p, end, &blocklisted) < 0) + /* version >= 2 and < 5, decode metadata, skip otherwise + * as it's handled via flags. + */ + if (msg_version >= 5) + ceph_decode_skip_map(&p, end, string, string, bad); + else if (__decode_session_metadata(&p, end, &blocklisted) < 0) goto bad; + /* version >= 3, feature bits */ ceph_decode_32_safe(&p, end, len, bad); if (len) { @@ -3478,6 +3374,18 @@ static void handle_session(struct ceph_mds_session *session, } } + if (msg_version >= 5) { + u32 flags; + /* version >= 4, struct_v, struct_cv, len, metric_spec */ + ceph_decode_skip_n(&p, end, 2 + sizeof(u32) * 2, bad); + /* version >= 5, flags */ + ceph_decode_32_safe(&p, end, flags, bad); + if (flags & CEPH_SESSION_BLOCKLISTED) { + pr_warn("mds%d session blocklisted\n", session->s_mds); + blocklisted = true; + } + } + mutex_lock(&mdsc->mutex); if (op == CEPH_SESSION_CLOSE) { ceph_get_mds_session(session); @@ -3772,7 +3680,7 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap, struct ceph_pagelist *pagelist = recon_state->pagelist; struct dentry *dentry; char *path; - int pathlen, err; + int pathlen = 0, err; u64 pathbase; u64 snap_follows; @@ -3792,7 +3700,6 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap, } } else { path = NULL; - pathlen = 0; pathbase = 0; } @@ -4527,8 +4434,6 @@ static void maybe_recover_session(struct ceph_mds_client *mdsc) bool check_session_state(struct ceph_mds_session *s) { - struct ceph_fs_client *fsc = s->s_mdsc->fsc; - switch (s->s_state) { case CEPH_MDS_SESSION_OPEN: if (s->s_ttl && time_after(jiffies, s->s_ttl)) { @@ -4537,10 +4442,6 @@ bool check_session_state(struct ceph_mds_session *s) } break; case CEPH_MDS_SESSION_CLOSING: - /* Should never reach this when not force unmounting */ - WARN_ON_ONCE(s->s_ttl && - READ_ONCE(fsc->mount_state) != CEPH_MOUNT_SHUTDOWN); - fallthrough; case CEPH_MDS_SESSION_NEW: case CEPH_MDS_SESSION_RESTARTING: case CEPH_MDS_SESSION_CLOSED: @@ -4931,7 +4832,7 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc) mutex_unlock(&mdsc->mutex); ceph_cleanup_snapid_map(mdsc); - ceph_cleanup_empty_realms(mdsc); + ceph_cleanup_global_and_empty_realms(mdsc); cancel_work_sync(&mdsc->cap_reclaim_work); cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */ @@ -5072,7 +4973,8 @@ void ceph_mdsc_handle_fsmap(struct ceph_mds_client *mdsc, struct ceph_msg *msg) return; bad: - pr_err("error decoding fsmap\n"); + pr_err("error decoding fsmap %d. Shutting down mount.\n", err); + ceph_umount_begin(mdsc->fsc->sb); err_out: mutex_lock(&mdsc->mutex); mdsc->mdsmap_err = err; @@ -5139,7 +5041,8 @@ void ceph_mdsc_handle_mdsmap(struct ceph_mds_client *mdsc, struct ceph_msg *msg) bad_unlock: mutex_unlock(&mdsc->mutex); bad: - pr_err("error decoding mdsmap %d\n", err); + pr_err("error decoding mdsmap %d. Shutting down mount.\n", err); + ceph_umount_begin(mdsc->fsc->sb); return; } diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 97c7f7bfa55f..33497846e47e 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -100,6 +100,11 @@ struct ceph_mds_reply_dir_entry { loff_t offset; }; +struct ceph_mds_reply_xattr { + char *xattr_value; + size_t xattr_value_len; +}; + /* * parsed info about an mds reply, including information about * either: 1) the target inode and/or its parent directory and dentry, @@ -115,6 +120,7 @@ struct ceph_mds_reply_info_parsed { char *dname; u32 dname_len; struct ceph_mds_reply_lease *dlease; + struct ceph_mds_reply_xattr xattr_info; /* extra */ union { @@ -274,8 +280,8 @@ struct ceph_mds_request { union ceph_mds_request_args r_args; int r_fmode; /* file mode, if expecting cap */ - const struct cred *r_cred; int r_request_release_offset; + const struct cred *r_cred; struct timespec64 r_stamp; /* for choosing which mds to send this request to */ @@ -296,12 +302,11 @@ struct ceph_mds_request { struct ceph_msg *r_reply; struct ceph_mds_reply_info_parsed r_reply_info; int r_err; - + u32 r_readdir_offset; struct page *r_locked_page; int r_dir_caps; int r_num_caps; - u32 r_readdir_offset; unsigned long r_timeout; /* optional. jiffies, 0 is "wait forever" */ unsigned long r_started; /* start time to measure timeout against */ @@ -329,7 +334,6 @@ struct ceph_mds_request { struct completion r_completion; struct completion r_safe_completion; ceph_mds_request_callback_t r_callback; - ceph_mds_request_wait_callback_t r_wait_for_completion; struct list_head r_unsafe_item; /* per-session unsafe list item */ long long r_dir_release_cnt; @@ -507,6 +511,9 @@ ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode); extern int ceph_mdsc_submit_request(struct ceph_mds_client *mdsc, struct inode *dir, struct ceph_mds_request *req); +int ceph_mdsc_wait_request(struct ceph_mds_client *mdsc, + struct ceph_mds_request *req, + ceph_mds_request_wait_callback_t wait_func); extern int ceph_mdsc_do_request(struct ceph_mds_client *mdsc, struct inode *dir, struct ceph_mds_request *req); diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c index 61d67cbcb367..30387733765d 100644 --- a/fs/ceph/mdsmap.c +++ b/fs/ceph/mdsmap.c @@ -263,10 +263,6 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end, bool msgr2) goto nomem; for (j = 0; j < num_export_targets; j++) { target = ceph_decode_32(&pexport_targets); - if (target >= m->possible_max_rank) { - err = -EIO; - goto corrupt; - } info->export_targets[j] = target; } } else { diff --git a/fs/ceph/metric.c b/fs/ceph/metric.c index 04d5df29bbbf..c47347d2e84e 100644 --- a/fs/ceph/metric.c +++ b/fs/ceph/metric.c @@ -8,6 +8,12 @@ #include "metric.h" #include "mds_client.h" +static void ktime_to_ceph_timespec(struct ceph_timespec *ts, ktime_t val) +{ + struct timespec64 t = ktime_to_timespec64(val); + ceph_encode_timespec64(ts, &t); +} + static bool ceph_mdsc_send_metrics(struct ceph_mds_client *mdsc, struct ceph_mds_session *s) { @@ -26,7 +32,6 @@ static bool ceph_mdsc_send_metrics(struct ceph_mds_client *mdsc, u64 nr_caps = atomic64_read(&m->total_caps); u32 header_len = sizeof(struct ceph_metric_header); struct ceph_msg *msg; - struct timespec64 ts; s64 sum; s32 items = 0; s32 len; @@ -59,37 +64,40 @@ static bool ceph_mdsc_send_metrics(struct ceph_mds_client *mdsc, /* encode the read latency metric */ read = (struct ceph_metric_read_latency *)(cap + 1); read->header.type = cpu_to_le32(CLIENT_METRIC_TYPE_READ_LATENCY); - read->header.ver = 1; + read->header.ver = 2; read->header.compat = 1; read->header.data_len = cpu_to_le32(sizeof(*read) - header_len); - sum = m->read_latency_sum; - jiffies_to_timespec64(sum, &ts); - read->sec = cpu_to_le32(ts.tv_sec); - read->nsec = cpu_to_le32(ts.tv_nsec); + sum = m->metric[METRIC_READ].latency_sum; + ktime_to_ceph_timespec(&read->lat, sum); + ktime_to_ceph_timespec(&read->avg, m->metric[METRIC_READ].latency_avg); + read->sq_sum = cpu_to_le64(m->metric[METRIC_READ].latency_sq_sum); + read->count = cpu_to_le64(m->metric[METRIC_READ].total); items++; /* encode the write latency metric */ write = (struct ceph_metric_write_latency *)(read + 1); write->header.type = cpu_to_le32(CLIENT_METRIC_TYPE_WRITE_LATENCY); - write->header.ver = 1; + write->header.ver = 2; write->header.compat = 1; write->header.data_len = cpu_to_le32(sizeof(*write) - header_len); - sum = m->write_latency_sum; - jiffies_to_timespec64(sum, &ts); - write->sec = cpu_to_le32(ts.tv_sec); - write->nsec = cpu_to_le32(ts.tv_nsec); + sum = m->metric[METRIC_WRITE].latency_sum; + ktime_to_ceph_timespec(&write->lat, sum); + ktime_to_ceph_timespec(&write->avg, m->metric[METRIC_WRITE].latency_avg); + write->sq_sum = cpu_to_le64(m->metric[METRIC_WRITE].latency_sq_sum); + write->count = cpu_to_le64(m->metric[METRIC_WRITE].total); items++; /* encode the metadata latency metric */ meta = (struct ceph_metric_metadata_latency *)(write + 1); meta->header.type = cpu_to_le32(CLIENT_METRIC_TYPE_METADATA_LATENCY); - meta->header.ver = 1; + meta->header.ver = 2; meta->header.compat = 1; meta->header.data_len = cpu_to_le32(sizeof(*meta) - header_len); - sum = m->metadata_latency_sum; - jiffies_to_timespec64(sum, &ts); - meta->sec = cpu_to_le32(ts.tv_sec); - meta->nsec = cpu_to_le32(ts.tv_nsec); + sum = m->metric[METRIC_METADATA].latency_sum; + ktime_to_ceph_timespec(&meta->lat, sum); + ktime_to_ceph_timespec(&meta->avg, m->metric[METRIC_METADATA].latency_avg); + meta->sq_sum = cpu_to_le64(m->metric[METRIC_METADATA].latency_sq_sum); + meta->count = cpu_to_le64(m->metric[METRIC_METADATA].total); items++; /* encode the dentry lease metric */ @@ -141,8 +149,8 @@ static bool ceph_mdsc_send_metrics(struct ceph_mds_client *mdsc, rsize->header.ver = 1; rsize->header.compat = 1; rsize->header.data_len = cpu_to_le32(sizeof(*rsize) - header_len); - rsize->total_ops = cpu_to_le64(m->total_reads); - rsize->total_size = cpu_to_le64(m->read_size_sum); + rsize->total_ops = cpu_to_le64(m->metric[METRIC_READ].total); + rsize->total_size = cpu_to_le64(m->metric[METRIC_READ].size_sum); items++; /* encode the write io size metric */ @@ -151,8 +159,8 @@ static bool ceph_mdsc_send_metrics(struct ceph_mds_client *mdsc, wsize->header.ver = 1; wsize->header.compat = 1; wsize->header.data_len = cpu_to_le32(sizeof(*wsize) - header_len); - wsize->total_ops = cpu_to_le64(m->total_writes); - wsize->total_size = cpu_to_le64(m->write_size_sum); + wsize->total_ops = cpu_to_le64(m->metric[METRIC_WRITE].total); + wsize->total_size = cpu_to_le64(m->metric[METRIC_WRITE].size_sum); items++; put_unaligned_le32(items, &head->num); @@ -160,8 +168,6 @@ static bool ceph_mdsc_send_metrics(struct ceph_mds_client *mdsc, msg->hdr.version = cpu_to_le16(1); msg->hdr.compat_version = cpu_to_le16(1); msg->hdr.front_len = cpu_to_le32(msg->front.iov_len); - dout("client%llu send metrics to mds%d\n", - ceph_client_gid(mdsc->fsc->client), s->s_mds); ceph_con_send(&s->s_con, msg); return true; @@ -220,7 +226,8 @@ static void metric_delayed_work(struct work_struct *work) int ceph_metric_init(struct ceph_client_metric *m) { - int ret; + struct ceph_metric *metric; + int ret, i; if (!m) return -EINVAL; @@ -243,32 +250,19 @@ int ceph_metric_init(struct ceph_client_metric *m) if (ret) goto err_i_caps_mis; - spin_lock_init(&m->read_metric_lock); - m->read_latency_sq_sum = 0; - m->read_latency_min = KTIME_MAX; - m->read_latency_max = 0; - m->total_reads = 0; - m->read_latency_sum = 0; - m->read_size_min = U64_MAX; - m->read_size_max = 0; - m->read_size_sum = 0; - - spin_lock_init(&m->write_metric_lock); - m->write_latency_sq_sum = 0; - m->write_latency_min = KTIME_MAX; - m->write_latency_max = 0; - m->total_writes = 0; - m->write_latency_sum = 0; - m->write_size_min = U64_MAX; - m->write_size_max = 0; - m->write_size_sum = 0; - - spin_lock_init(&m->metadata_metric_lock); - m->metadata_latency_sq_sum = 0; - m->metadata_latency_min = KTIME_MAX; - m->metadata_latency_max = 0; - m->total_metadatas = 0; - m->metadata_latency_sum = 0; + for (i = 0; i < METRIC_MAX; i++) { + metric = &m->metric[i]; + spin_lock_init(&metric->lock); + metric->size_sum = 0; + metric->size_min = U64_MAX; + metric->size_max = 0; + metric->total = 0; + metric->latency_sum = 0; + metric->latency_avg = 0; + metric->latency_sq_sum = 0; + metric->latency_min = KTIME_MAX; + metric->latency_max = 0; + } atomic64_set(&m->opened_files, 0); ret = percpu_counter_init(&m->opened_inodes, 0, GFP_KERNEL); @@ -322,25 +316,24 @@ void ceph_metric_destroy(struct ceph_client_metric *m) max = new; \ } -static inline void __update_stdev(ktime_t total, ktime_t lsum, - ktime_t *sq_sump, ktime_t lat) +static inline void __update_mean_and_stdev(ktime_t total, ktime_t *lavg, + ktime_t *sq_sump, ktime_t lat) { - ktime_t avg, sq; - - if (unlikely(total == 1)) - return; - - /* the sq is (lat - old_avg) * (lat - new_avg) */ - avg = DIV64_U64_ROUND_CLOSEST((lsum - lat), (total - 1)); - sq = lat - avg; - avg = DIV64_U64_ROUND_CLOSEST(lsum, total); - sq = sq * (lat - avg); - *sq_sump += sq; + ktime_t avg; + + if (unlikely(total == 1)) { + *lavg = lat; + } else { + /* the sq is (lat - old_avg) * (lat - new_avg) */ + avg = *lavg + div64_s64(lat - *lavg, total); + *sq_sump += (lat - *lavg)*(lat - avg); + *lavg = avg; + } } -void ceph_update_read_metrics(struct ceph_client_metric *m, - ktime_t r_start, ktime_t r_end, - unsigned int size, int rc) +void ceph_update_metrics(struct ceph_metric *m, + ktime_t r_start, ktime_t r_end, + unsigned int size, int rc) { ktime_t lat = ktime_sub(r_end, r_start); ktime_t total; @@ -348,63 +341,13 @@ void ceph_update_read_metrics(struct ceph_client_metric *m, if (unlikely(rc < 0 && rc != -ENOENT && rc != -ETIMEDOUT)) return; - spin_lock(&m->read_metric_lock); - total = ++m->total_reads; - m->read_size_sum += size; - m->read_latency_sum += lat; - METRIC_UPDATE_MIN_MAX(m->read_size_min, - m->read_size_max, - size); - METRIC_UPDATE_MIN_MAX(m->read_latency_min, - m->read_latency_max, - lat); - __update_stdev(total, m->read_latency_sum, - &m->read_latency_sq_sum, lat); - spin_unlock(&m->read_metric_lock); -} - -void ceph_update_write_metrics(struct ceph_client_metric *m, - ktime_t r_start, ktime_t r_end, - unsigned int size, int rc) -{ - ktime_t lat = ktime_sub(r_end, r_start); - ktime_t total; - - if (unlikely(rc && rc != -ETIMEDOUT)) - return; - - spin_lock(&m->write_metric_lock); - total = ++m->total_writes; - m->write_size_sum += size; - m->write_latency_sum += lat; - METRIC_UPDATE_MIN_MAX(m->write_size_min, - m->write_size_max, - size); - METRIC_UPDATE_MIN_MAX(m->write_latency_min, - m->write_latency_max, - lat); - __update_stdev(total, m->write_latency_sum, - &m->write_latency_sq_sum, lat); - spin_unlock(&m->write_metric_lock); -} - -void ceph_update_metadata_metrics(struct ceph_client_metric *m, - ktime_t r_start, ktime_t r_end, - int rc) -{ - ktime_t lat = ktime_sub(r_end, r_start); - ktime_t total; - - if (unlikely(rc && rc != -ENOENT)) - return; - - spin_lock(&m->metadata_metric_lock); - total = ++m->total_metadatas; - m->metadata_latency_sum += lat; - METRIC_UPDATE_MIN_MAX(m->metadata_latency_min, - m->metadata_latency_max, - lat); - __update_stdev(total, m->metadata_latency_sum, - &m->metadata_latency_sq_sum, lat); - spin_unlock(&m->metadata_metric_lock); + spin_lock(&m->lock); + total = ++m->total; + m->size_sum += size; + METRIC_UPDATE_MIN_MAX(m->size_min, m->size_max, size); + m->latency_sum += lat; + METRIC_UPDATE_MIN_MAX(m->latency_min, m->latency_max, lat); + __update_mean_and_stdev(total, &m->latency_avg, &m->latency_sq_sum, + lat); + spin_unlock(&m->lock); } diff --git a/fs/ceph/metric.h b/fs/ceph/metric.h index 0133955a3c6a..0d0c44bd3332 100644 --- a/fs/ceph/metric.h +++ b/fs/ceph/metric.h @@ -2,7 +2,7 @@ #ifndef _FS_CEPH_MDS_METRIC_H #define _FS_CEPH_MDS_METRIC_H -#include <linux/types.h> +#include <linux/ceph/types.h> #include <linux/percpu_counter.h> #include <linux/ktime.h> @@ -19,27 +19,39 @@ enum ceph_metric_type { CLIENT_METRIC_TYPE_OPENED_INODES, CLIENT_METRIC_TYPE_READ_IO_SIZES, CLIENT_METRIC_TYPE_WRITE_IO_SIZES, - - CLIENT_METRIC_TYPE_MAX = CLIENT_METRIC_TYPE_WRITE_IO_SIZES, + CLIENT_METRIC_TYPE_AVG_READ_LATENCY, + CLIENT_METRIC_TYPE_STDEV_READ_LATENCY, + CLIENT_METRIC_TYPE_AVG_WRITE_LATENCY, + CLIENT_METRIC_TYPE_STDEV_WRITE_LATENCY, + CLIENT_METRIC_TYPE_AVG_METADATA_LATENCY, + CLIENT_METRIC_TYPE_STDEV_METADATA_LATENCY, + + CLIENT_METRIC_TYPE_MAX = CLIENT_METRIC_TYPE_STDEV_METADATA_LATENCY, }; /* * This will always have the highest metric bit value * as the last element of the array. */ -#define CEPHFS_METRIC_SPEC_CLIENT_SUPPORTED { \ - CLIENT_METRIC_TYPE_CAP_INFO, \ - CLIENT_METRIC_TYPE_READ_LATENCY, \ - CLIENT_METRIC_TYPE_WRITE_LATENCY, \ - CLIENT_METRIC_TYPE_METADATA_LATENCY, \ - CLIENT_METRIC_TYPE_DENTRY_LEASE, \ - CLIENT_METRIC_TYPE_OPENED_FILES, \ - CLIENT_METRIC_TYPE_PINNED_ICAPS, \ - CLIENT_METRIC_TYPE_OPENED_INODES, \ - CLIENT_METRIC_TYPE_READ_IO_SIZES, \ - CLIENT_METRIC_TYPE_WRITE_IO_SIZES, \ - \ - CLIENT_METRIC_TYPE_MAX, \ +#define CEPHFS_METRIC_SPEC_CLIENT_SUPPORTED { \ + CLIENT_METRIC_TYPE_CAP_INFO, \ + CLIENT_METRIC_TYPE_READ_LATENCY, \ + CLIENT_METRIC_TYPE_WRITE_LATENCY, \ + CLIENT_METRIC_TYPE_METADATA_LATENCY, \ + CLIENT_METRIC_TYPE_DENTRY_LEASE, \ + CLIENT_METRIC_TYPE_OPENED_FILES, \ + CLIENT_METRIC_TYPE_PINNED_ICAPS, \ + CLIENT_METRIC_TYPE_OPENED_INODES, \ + CLIENT_METRIC_TYPE_READ_IO_SIZES, \ + CLIENT_METRIC_TYPE_WRITE_IO_SIZES, \ + CLIENT_METRIC_TYPE_AVG_READ_LATENCY, \ + CLIENT_METRIC_TYPE_STDEV_READ_LATENCY, \ + CLIENT_METRIC_TYPE_AVG_WRITE_LATENCY, \ + CLIENT_METRIC_TYPE_STDEV_WRITE_LATENCY, \ + CLIENT_METRIC_TYPE_AVG_METADATA_LATENCY, \ + CLIENT_METRIC_TYPE_STDEV_METADATA_LATENCY, \ + \ + CLIENT_METRIC_TYPE_MAX, \ } struct ceph_metric_header { @@ -60,22 +72,28 @@ struct ceph_metric_cap { /* metric read latency header */ struct ceph_metric_read_latency { struct ceph_metric_header header; - __le32 sec; - __le32 nsec; + struct ceph_timespec lat; + struct ceph_timespec avg; + __le64 sq_sum; + __le64 count; } __packed; /* metric write latency header */ struct ceph_metric_write_latency { struct ceph_metric_header header; - __le32 sec; - __le32 nsec; + struct ceph_timespec lat; + struct ceph_timespec avg; + __le64 sq_sum; + __le64 count; } __packed; /* metric metadata latency header */ struct ceph_metric_metadata_latency { struct ceph_metric_header header; - __le32 sec; - __le32 nsec; + struct ceph_timespec lat; + struct ceph_timespec avg; + __le64 sq_sum; + __le64 count; } __packed; /* metric dentry lease header */ @@ -125,6 +143,27 @@ struct ceph_metric_head { __le32 num; /* the number of metrics that will be sent */ } __packed; +enum metric_type { + METRIC_READ, + METRIC_WRITE, + METRIC_METADATA, + METRIC_COPYFROM, + METRIC_MAX +}; + +struct ceph_metric { + spinlock_t lock; + u64 total; + u64 size_sum; + u64 size_min; + u64 size_max; + ktime_t latency_sum; + ktime_t latency_avg; + ktime_t latency_sq_sum; + ktime_t latency_min; + ktime_t latency_max; +}; + /* This is the global metrics */ struct ceph_client_metric { atomic64_t total_dentries; @@ -135,32 +174,7 @@ struct ceph_client_metric { struct percpu_counter i_caps_hit; struct percpu_counter i_caps_mis; - spinlock_t read_metric_lock; - u64 total_reads; - u64 read_size_sum; - u64 read_size_min; - u64 read_size_max; - ktime_t read_latency_sum; - ktime_t read_latency_sq_sum; - ktime_t read_latency_min; - ktime_t read_latency_max; - - spinlock_t write_metric_lock; - u64 total_writes; - u64 write_size_sum; - u64 write_size_min; - u64 write_size_max; - ktime_t write_latency_sum; - ktime_t write_latency_sq_sum; - ktime_t write_latency_min; - ktime_t write_latency_max; - - spinlock_t metadata_metric_lock; - u64 total_metadatas; - ktime_t metadata_latency_sum; - ktime_t metadata_latency_sq_sum; - ktime_t metadata_latency_min; - ktime_t metadata_latency_max; + struct ceph_metric metric[METRIC_MAX]; /* The total number of directories and files that are opened */ atomic64_t opened_files; @@ -195,13 +209,36 @@ static inline void ceph_update_cap_mis(struct ceph_client_metric *m) percpu_counter_inc(&m->i_caps_mis); } -extern void ceph_update_read_metrics(struct ceph_client_metric *m, - ktime_t r_start, ktime_t r_end, - unsigned int size, int rc); -extern void ceph_update_write_metrics(struct ceph_client_metric *m, - ktime_t r_start, ktime_t r_end, - unsigned int size, int rc); -extern void ceph_update_metadata_metrics(struct ceph_client_metric *m, - ktime_t r_start, ktime_t r_end, - int rc); +extern void ceph_update_metrics(struct ceph_metric *m, + ktime_t r_start, ktime_t r_end, + unsigned int size, int rc); + +static inline void ceph_update_read_metrics(struct ceph_client_metric *m, + ktime_t r_start, ktime_t r_end, + unsigned int size, int rc) +{ + ceph_update_metrics(&m->metric[METRIC_READ], + r_start, r_end, size, rc); +} +static inline void ceph_update_write_metrics(struct ceph_client_metric *m, + ktime_t r_start, ktime_t r_end, + unsigned int size, int rc) +{ + ceph_update_metrics(&m->metric[METRIC_WRITE], + r_start, r_end, size, rc); +} +static inline void ceph_update_metadata_metrics(struct ceph_client_metric *m, + ktime_t r_start, ktime_t r_end, + int rc) +{ + ceph_update_metrics(&m->metric[METRIC_METADATA], + r_start, r_end, 0, rc); +} +static inline void ceph_update_copyfrom_metrics(struct ceph_client_metric *m, + ktime_t r_start, ktime_t r_end, + unsigned int size, int rc) +{ + ceph_update_metrics(&m->metric[METRIC_COPYFROM], + r_start, r_end, size, rc); +} #endif /* _FS_CEPH_MDS_METRIC_H */ diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c index 620c691af40e..a338a3ec0dc4 100644 --- a/fs/ceph/quota.c +++ b/fs/ceph/quota.c @@ -30,6 +30,9 @@ static inline bool ceph_has_realms_with_quotas(struct inode *inode) /* if root is the real CephFS root, we don't have quota realms */ if (root && ceph_ino(root) == CEPH_INO_ROOT) return false; + /* MDS stray dirs have no quota realms */ + if (ceph_vino_is_reserved(ceph_inode(inode)->i_vino)) + return false; /* otherwise, we can't know for sure */ return true; } @@ -494,10 +497,24 @@ bool ceph_quota_update_statfs(struct ceph_fs_client *fsc, struct kstatfs *buf) if (ci->i_max_bytes) { total = ci->i_max_bytes >> CEPH_BLOCK_SHIFT; used = ci->i_rbytes >> CEPH_BLOCK_SHIFT; + /* For quota size less than 4MB, use 4KB block size */ + if (!total) { + total = ci->i_max_bytes >> CEPH_4K_BLOCK_SHIFT; + used = ci->i_rbytes >> CEPH_4K_BLOCK_SHIFT; + buf->f_frsize = 1 << CEPH_4K_BLOCK_SHIFT; + } /* It is possible for a quota to be exceeded. * Report 'zero' in that case */ free = total > used ? total - used : 0; + /* For quota size less than 4KB, report the + * total=used=4KB,free=0 when quota is full + * and total=free=4KB, used=0 otherwise */ + if (!total) { + total = 1; + free = ci->i_max_bytes > ci->i_rbytes ? 1 : 0; + buf->f_frsize = 1 << CEPH_4K_BLOCK_SHIFT; + } } spin_unlock(&ci->i_ceph_lock); if (total) { diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index b41e6724c591..322ee5add942 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -121,18 +121,23 @@ static struct ceph_snap_realm *ceph_create_snap_realm( if (!realm) return ERR_PTR(-ENOMEM); - atomic_set(&realm->nref, 1); /* for caller */ + /* Do not release the global dummy snaprealm until unmouting */ + if (ino == CEPH_INO_GLOBAL_SNAPREALM) + atomic_set(&realm->nref, 2); + else + atomic_set(&realm->nref, 1); realm->ino = ino; INIT_LIST_HEAD(&realm->children); INIT_LIST_HEAD(&realm->child_item); INIT_LIST_HEAD(&realm->empty_item); INIT_LIST_HEAD(&realm->dirty_item); + INIT_LIST_HEAD(&realm->rebuild_item); INIT_LIST_HEAD(&realm->inodes_with_caps); spin_lock_init(&realm->inodes_with_caps_lock); __insert_snap_realm(&mdsc->snap_realms, realm); mdsc->num_snap_realms++; - dout("create_snap_realm %llx %p\n", realm->ino, realm); + dout("%s %llx %p\n", __func__, realm->ino, realm); return realm; } @@ -156,7 +161,7 @@ static struct ceph_snap_realm *__lookup_snap_realm(struct ceph_mds_client *mdsc, else if (ino > r->ino) n = n->rb_right; else { - dout("lookup_snap_realm %llx %p\n", r->ino, r); + dout("%s %llx %p\n", __func__, r->ino, r); return r; } } @@ -184,7 +189,7 @@ static void __destroy_snap_realm(struct ceph_mds_client *mdsc, { lockdep_assert_held_write(&mdsc->snap_rwsem); - dout("__destroy_snap_realm %p %llx\n", realm, realm->ino); + dout("%s %p %llx\n", __func__, realm, realm->ino); rb_erase(&realm->node, &mdsc->snap_realms); mdsc->num_snap_realms--; @@ -260,9 +265,14 @@ static void __cleanup_empty_realms(struct ceph_mds_client *mdsc) spin_unlock(&mdsc->snap_empty_lock); } -void ceph_cleanup_empty_realms(struct ceph_mds_client *mdsc) +void ceph_cleanup_global_and_empty_realms(struct ceph_mds_client *mdsc) { + struct ceph_snap_realm *global_realm; + down_write(&mdsc->snap_rwsem); + global_realm = __lookup_snap_realm(mdsc, CEPH_INO_GLOBAL_SNAPREALM); + if (global_realm) + ceph_put_snap_realm(mdsc, global_realm); __cleanup_empty_realms(mdsc); up_write(&mdsc->snap_rwsem); } @@ -292,9 +302,8 @@ static int adjust_snap_realm_parent(struct ceph_mds_client *mdsc, if (IS_ERR(parent)) return PTR_ERR(parent); } - dout("adjust_snap_realm_parent %llx %p: %llx %p -> %llx %p\n", - realm->ino, realm, realm->parent_ino, realm->parent, - parentino, parent); + dout("%s %llx %p: %llx %p -> %llx %p\n", __func__, realm->ino, + realm, realm->parent_ino, realm->parent, parentino, parent); if (realm->parent) { list_del_init(&realm->child_item); ceph_put_snap_realm(mdsc, realm->parent); @@ -320,7 +329,8 @@ static int cmpu64_rev(const void *a, const void *b) * build the snap context for a given realm. */ static int build_snap_context(struct ceph_snap_realm *realm, - struct list_head* dirty_realms) + struct list_head *realm_queue, + struct list_head *dirty_realms) { struct ceph_snap_realm *parent = realm->parent; struct ceph_snap_context *snapc; @@ -334,9 +344,9 @@ static int build_snap_context(struct ceph_snap_realm *realm, */ if (parent) { if (!parent->cached_context) { - err = build_snap_context(parent, dirty_realms); - if (err) - goto fail; + /* add to the queue head */ + list_add(&parent->rebuild_item, realm_queue); + return 1; } num += parent->cached_context->num_snaps; } @@ -349,9 +359,8 @@ static int build_snap_context(struct ceph_snap_realm *realm, realm->cached_context->seq == realm->seq && (!parent || realm->cached_context->seq >= parent->cached_context->seq)) { - dout("build_snap_context %llx %p: %p seq %lld (%u snaps)" - " (unchanged)\n", - realm->ino, realm, realm->cached_context, + dout("%s %llx %p: %p seq %lld (%u snaps) (unchanged)\n", + __func__, realm->ino, realm, realm->cached_context, realm->cached_context->seq, (unsigned int)realm->cached_context->num_snaps); return 0; @@ -390,9 +399,8 @@ static int build_snap_context(struct ceph_snap_realm *realm, sort(snapc->snaps, num, sizeof(u64), cmpu64_rev, NULL); snapc->num_snaps = num; - dout("build_snap_context %llx %p: %p seq %lld (%u snaps)\n", - realm->ino, realm, snapc, snapc->seq, - (unsigned int) snapc->num_snaps); + dout("%s %llx %p: %p seq %lld (%u snaps)\n", __func__, realm->ino, + realm, snapc, snapc->seq, (unsigned int) snapc->num_snaps); ceph_put_snap_context(realm->cached_context); realm->cached_context = snapc; @@ -409,8 +417,7 @@ fail: ceph_put_snap_context(realm->cached_context); realm->cached_context = NULL; } - pr_err("build_snap_context %llx %p fail %d\n", realm->ino, - realm, err); + pr_err("%s %llx %p fail %d\n", __func__, realm->ino, realm, err); return err; } @@ -420,13 +427,50 @@ fail: static void rebuild_snap_realms(struct ceph_snap_realm *realm, struct list_head *dirty_realms) { - struct ceph_snap_realm *child; + LIST_HEAD(realm_queue); + int last = 0; + bool skip = false; + + list_add_tail(&realm->rebuild_item, &realm_queue); + + while (!list_empty(&realm_queue)) { + struct ceph_snap_realm *_realm, *child; + + _realm = list_first_entry(&realm_queue, + struct ceph_snap_realm, + rebuild_item); + + /* + * If the last building failed dues to memory + * issue, just empty the realm_queue and return + * to avoid infinite loop. + */ + if (last < 0) { + list_del_init(&_realm->rebuild_item); + continue; + } + + last = build_snap_context(_realm, &realm_queue, dirty_realms); + dout("%s %llx %p, %s\n", __func__, _realm->ino, _realm, + last > 0 ? "is deferred" : !last ? "succeeded" : "failed"); + + /* is any child in the list ? */ + list_for_each_entry(child, &_realm->children, child_item) { + if (!list_empty(&child->rebuild_item)) { + skip = true; + break; + } + } - dout("rebuild_snap_realms %llx %p\n", realm->ino, realm); - build_snap_context(realm, dirty_realms); + if (!skip) { + list_for_each_entry(child, &_realm->children, child_item) + list_add_tail(&child->rebuild_item, &realm_queue); + } - list_for_each_entry(child, &realm->children, child_item) - rebuild_snap_realms(child, dirty_realms); + /* last == 1 means need to build parent first */ + if (last <= 0) + list_del_init(&_realm->rebuild_item); + } } @@ -474,23 +518,15 @@ static bool has_new_snaps(struct ceph_snap_context *o, * Caller must hold snap_rwsem for read (i.e., the realm topology won't * change). */ -static void ceph_queue_cap_snap(struct ceph_inode_info *ci) +static void ceph_queue_cap_snap(struct ceph_inode_info *ci, + struct ceph_cap_snap **pcapsnap) { struct inode *inode = &ci->vfs_inode; - struct ceph_cap_snap *capsnap; struct ceph_snap_context *old_snapc, *new_snapc; + struct ceph_cap_snap *capsnap = *pcapsnap; struct ceph_buffer *old_blob = NULL; int used, dirty; - capsnap = kzalloc(sizeof(*capsnap), GFP_NOFS); - if (!capsnap) { - pr_err("ENOMEM allocating ceph_cap_snap on %p\n", inode); - return; - } - capsnap->cap_flush.is_capsnap = true; - INIT_LIST_HEAD(&capsnap->cap_flush.i_list); - INIT_LIST_HEAD(&capsnap->cap_flush.g_list); - spin_lock(&ci->i_ceph_lock); used = __ceph_caps_used(ci); dirty = __ceph_caps_dirty(ci); @@ -511,12 +547,14 @@ static void ceph_queue_cap_snap(struct ceph_inode_info *ci) as no new writes are allowed to start when pending, so any writes in progress now were started before the previous cap_snap. lucky us. */ - dout("queue_cap_snap %p already pending\n", inode); + dout("%s %p %llx.%llx already pending\n", + __func__, inode, ceph_vinop(inode)); goto update_snapc; } if (ci->i_wrbuffer_ref_head == 0 && !(dirty & (CEPH_CAP_ANY_EXCL|CEPH_CAP_FILE_WR))) { - dout("queue_cap_snap %p nothing dirty|writing\n", inode); + dout("%s %p %llx.%llx nothing dirty|writing\n", + __func__, inode, ceph_vinop(inode)); goto update_snapc; } @@ -536,20 +574,17 @@ static void ceph_queue_cap_snap(struct ceph_inode_info *ci) } else { if (!(used & CEPH_CAP_FILE_WR) && ci->i_wrbuffer_ref_head == 0) { - dout("queue_cap_snap %p " - "no new_snap|dirty_page|writing\n", inode); + dout("%s %p %llx.%llx no new_snap|dirty_page|writing\n", + __func__, inode, ceph_vinop(inode)); goto update_snapc; } } - dout("queue_cap_snap %p cap_snap %p queuing under %p %s %s\n", - inode, capsnap, old_snapc, ceph_cap_string(dirty), - capsnap->need_flush ? "" : "no_flush"); + dout("%s %p %llx.%llx cap_snap %p queuing under %p %s %s\n", + __func__, inode, ceph_vinop(inode), capsnap, old_snapc, + ceph_cap_string(dirty), capsnap->need_flush ? "" : "no_flush"); ihold(inode); - refcount_set(&capsnap->nref, 1); - INIT_LIST_HEAD(&capsnap->ci_item); - capsnap->follows = old_snapc->seq; capsnap->issued = __ceph_caps_issued(ci, NULL); capsnap->dirty = dirty; @@ -579,31 +614,30 @@ static void ceph_queue_cap_snap(struct ceph_inode_info *ci) list_add_tail(&capsnap->ci_item, &ci->i_cap_snaps); if (used & CEPH_CAP_FILE_WR) { - dout("queue_cap_snap %p cap_snap %p snapc %p" - " seq %llu used WR, now pending\n", inode, + dout("%s %p %llx.%llx cap_snap %p snapc %p seq %llu used WR," + " now pending\n", __func__, inode, ceph_vinop(inode), capsnap, old_snapc, old_snapc->seq); capsnap->writing = 1; } else { /* note mtime, size NOW. */ __ceph_finish_cap_snap(ci, capsnap); } - capsnap = NULL; + *pcapsnap = NULL; old_snapc = NULL; update_snapc: - if (ci->i_wrbuffer_ref_head == 0 && - ci->i_wr_ref == 0 && - ci->i_dirty_caps == 0 && - ci->i_flushing_caps == 0) { - ci->i_head_snapc = NULL; - } else { + if (ci->i_wrbuffer_ref_head == 0 && + ci->i_wr_ref == 0 && + ci->i_dirty_caps == 0 && + ci->i_flushing_caps == 0) { + ci->i_head_snapc = NULL; + } else { ci->i_head_snapc = ceph_get_snap_context(new_snapc); dout(" new snapc is %p\n", new_snapc); } spin_unlock(&ci->i_ceph_lock); ceph_buffer_put(old_blob); - kfree(capsnap); ceph_put_snap_context(old_snapc); } @@ -632,27 +666,28 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci, capsnap->truncate_size = ci->i_truncate_size; capsnap->truncate_seq = ci->i_truncate_seq; if (capsnap->dirty_pages) { - dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu " - "still has %d dirty pages\n", inode, capsnap, - capsnap->context, capsnap->context->seq, - ceph_cap_string(capsnap->dirty), capsnap->size, - capsnap->dirty_pages); + dout("%s %p %llx.%llx cap_snap %p snapc %p %llu %s s=%llu " + "still has %d dirty pages\n", __func__, inode, + ceph_vinop(inode), capsnap, capsnap->context, + capsnap->context->seq, ceph_cap_string(capsnap->dirty), + capsnap->size, capsnap->dirty_pages); return 0; } /* Fb cap still in use, delay it */ if (ci->i_wb_ref) { - dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu " - "used WRBUFFER, delaying\n", inode, capsnap, - capsnap->context, capsnap->context->seq, - ceph_cap_string(capsnap->dirty), capsnap->size); + dout("%s %p %llx.%llx cap_snap %p snapc %p %llu %s s=%llu " + "used WRBUFFER, delaying\n", __func__, inode, + ceph_vinop(inode), capsnap, capsnap->context, + capsnap->context->seq, ceph_cap_string(capsnap->dirty), + capsnap->size); capsnap->writing = 1; return 0; } ci->i_ceph_flags |= CEPH_I_FLUSH_SNAPS; - dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu\n", - inode, capsnap, capsnap->context, + dout("%s %p %llx.%llx cap_snap %p snapc %p %llu %s s=%llu\n", + __func__, inode, ceph_vinop(inode), capsnap, capsnap->context, capsnap->context->seq, ceph_cap_string(capsnap->dirty), capsnap->size); @@ -671,8 +706,9 @@ static void queue_realm_cap_snaps(struct ceph_snap_realm *realm) { struct ceph_inode_info *ci; struct inode *lastinode = NULL; + struct ceph_cap_snap *capsnap = NULL; - dout("queue_realm_cap_snaps %p %llx inodes\n", realm, realm->ino); + dout("%s %p %llx inode\n", __func__, realm, realm->ino); spin_lock(&realm->inodes_with_caps_lock); list_for_each_entry(ci, &realm->inodes_with_caps, i_snap_realm_item) { @@ -682,13 +718,35 @@ static void queue_realm_cap_snaps(struct ceph_snap_realm *realm) spin_unlock(&realm->inodes_with_caps_lock); iput(lastinode); lastinode = inode; - ceph_queue_cap_snap(ci); + + /* + * Allocate the capsnap memory outside of ceph_queue_cap_snap() + * to reduce very possible but unnecessary frequently memory + * allocate/free in this loop. + */ + if (!capsnap) { + capsnap = kmem_cache_zalloc(ceph_cap_snap_cachep, GFP_NOFS); + if (!capsnap) { + pr_err("ENOMEM allocating ceph_cap_snap on %p\n", + inode); + return; + } + } + capsnap->cap_flush.is_capsnap = true; + refcount_set(&capsnap->nref, 1); + INIT_LIST_HEAD(&capsnap->cap_flush.i_list); + INIT_LIST_HEAD(&capsnap->cap_flush.g_list); + INIT_LIST_HEAD(&capsnap->ci_item); + + ceph_queue_cap_snap(ci, &capsnap); spin_lock(&realm->inodes_with_caps_lock); } spin_unlock(&realm->inodes_with_caps_lock); iput(lastinode); - dout("queue_realm_cap_snaps %p %llx done\n", realm, realm->ino); + if (capsnap) + kmem_cache_free(ceph_cap_snap_cachep, capsnap); + dout("%s %p %llx done\n", __func__, realm, realm->ino); } /* @@ -707,14 +765,16 @@ int ceph_update_snap_trace(struct ceph_mds_client *mdsc, __le64 *prior_parent_snaps; /* encoded */ struct ceph_snap_realm *realm = NULL; struct ceph_snap_realm *first_realm = NULL; - int invalidate = 0; + struct ceph_snap_realm *realm_to_rebuild = NULL; + int rebuild_snapcs; int err = -ENOMEM; LIST_HEAD(dirty_realms); lockdep_assert_held_write(&mdsc->snap_rwsem); - dout("update_snap_trace deletion=%d\n", deletion); + dout("%s deletion=%d\n", __func__, deletion); more: + rebuild_snapcs = 0; ceph_decode_need(&p, e, sizeof(*ri), bad); ri = p; p += sizeof(*ri); @@ -738,10 +798,10 @@ more: err = adjust_snap_realm_parent(mdsc, realm, le64_to_cpu(ri->parent)); if (err < 0) goto fail; - invalidate += err; + rebuild_snapcs += err; if (le64_to_cpu(ri->seq) > realm->seq) { - dout("update_snap_trace updating %llx %p %lld -> %lld\n", + dout("%s updating %llx %p %lld -> %lld\n", __func__, realm->ino, realm, realm->seq, le64_to_cpu(ri->seq)); /* update realm parameters, snap lists */ realm->seq = le64_to_cpu(ri->seq); @@ -763,22 +823,30 @@ more: if (realm->seq > mdsc->last_snap_seq) mdsc->last_snap_seq = realm->seq; - invalidate = 1; + rebuild_snapcs = 1; } else if (!realm->cached_context) { - dout("update_snap_trace %llx %p seq %lld new\n", + dout("%s %llx %p seq %lld new\n", __func__, realm->ino, realm, realm->seq); - invalidate = 1; + rebuild_snapcs = 1; } else { - dout("update_snap_trace %llx %p seq %lld unchanged\n", + dout("%s %llx %p seq %lld unchanged\n", __func__, realm->ino, realm, realm->seq); } - dout("done with %llx %p, invalidated=%d, %p %p\n", realm->ino, - realm, invalidate, p, e); + dout("done with %llx %p, rebuild_snapcs=%d, %p %p\n", realm->ino, + realm, rebuild_snapcs, p, e); + + /* + * this will always track the uppest parent realm from which + * we need to rebuild the snapshot contexts _downward_ in + * hierarchy. + */ + if (rebuild_snapcs) + realm_to_rebuild = realm; - /* invalidate when we reach the _end_ (root) of the trace */ - if (invalidate && p >= e) - rebuild_snap_realms(realm, &dirty_realms); + /* rebuild_snapcs when we reach the _end_ (root) of the trace */ + if (realm_to_rebuild && p >= e) + rebuild_snap_realms(realm_to_rebuild, &dirty_realms); if (!first_realm) first_realm = realm; @@ -814,7 +882,7 @@ fail: ceph_put_snap_realm(mdsc, realm); if (first_realm) ceph_put_snap_realm(mdsc, first_realm); - pr_err("update_snap_trace error %d\n", err); + pr_err("%s error %d\n", __func__, err); return err; } @@ -831,7 +899,7 @@ static void flush_snaps(struct ceph_mds_client *mdsc) struct inode *inode; struct ceph_mds_session *session = NULL; - dout("flush_snaps\n"); + dout("%s\n", __func__); spin_lock(&mdsc->snap_flush_lock); while (!list_empty(&mdsc->snap_flush_list)) { ci = list_first_entry(&mdsc->snap_flush_list, @@ -846,7 +914,7 @@ static void flush_snaps(struct ceph_mds_client *mdsc) spin_unlock(&mdsc->snap_flush_lock); ceph_put_mds_session(session); - dout("flush_snaps done\n"); + dout("%s done\n", __func__); } /** @@ -928,8 +996,8 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, trace_len = le32_to_cpu(h->trace_len); p += sizeof(*h); - dout("handle_snap from mds%d op %s split %llx tracelen %d\n", mds, - ceph_snap_op_name(op), split, trace_len); + dout("%s from mds%d op %s split %llx tracelen %d\n", __func__, + mds, ceph_snap_op_name(op), split, trace_len); mutex_lock(&session->s_mutex); inc_session_sequence(session); @@ -989,13 +1057,13 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, */ if (ci->i_snap_realm->created > le64_to_cpu(ri->created)) { - dout(" leaving %p in newer realm %llx %p\n", - inode, ci->i_snap_realm->ino, + dout(" leaving %p %llx.%llx in newer realm %llx %p\n", + inode, ceph_vinop(inode), ci->i_snap_realm->ino, ci->i_snap_realm); goto skip_inode; } - dout(" will move %p to split realm %llx %p\n", - inode, realm->ino, realm); + dout(" will move %p %llx.%llx to split realm %llx %p\n", + inode, ceph_vinop(inode), realm->ino, realm); ceph_get_snap_realm(mdsc, realm); ceph_change_snap_realm(inode, realm); @@ -1038,7 +1106,7 @@ skip_inode: return; bad: - pr_err("corrupt snap message from mds%d\n", mds); + pr_err("%s corrupt snap message from mds%d\n", __func__, mds); ceph_msg_dump(msg); out: if (locked_rwsem) @@ -1071,7 +1139,8 @@ struct ceph_snapid_map* ceph_get_snapid_map(struct ceph_mds_client *mdsc, } spin_unlock(&mdsc->snapid_map_lock); if (exist) { - dout("found snapid map %llx -> %x\n", exist->snap, exist->dev); + dout("%s found snapid map %llx -> %x\n", __func__, + exist->snap, exist->dev); return exist; } @@ -1115,11 +1184,13 @@ struct ceph_snapid_map* ceph_get_snapid_map(struct ceph_mds_client *mdsc, if (exist) { free_anon_bdev(sm->dev); kfree(sm); - dout("found snapid map %llx -> %x\n", exist->snap, exist->dev); + dout("%s found snapid map %llx -> %x\n", __func__, + exist->snap, exist->dev); return exist; } - dout("create snapid map %llx -> %x\n", sm->snap, sm->dev); + dout("%s create snapid map %llx -> %x\n", __func__, + sm->snap, sm->dev); return sm; } diff --git a/fs/ceph/strings.c b/fs/ceph/strings.c index 573bb9556fb5..e36e8948e728 100644 --- a/fs/ceph/strings.c +++ b/fs/ceph/strings.c @@ -60,6 +60,7 @@ const char *ceph_mds_op_name(int op) case CEPH_MDS_OP_LOOKUPINO: return "lookupino"; case CEPH_MDS_OP_LOOKUPNAME: return "lookupname"; case CEPH_MDS_OP_GETATTR: return "getattr"; + case CEPH_MDS_OP_GETVXATTR: return "getvxattr"; case CEPH_MDS_OP_SETXATTR: return "setxattr"; case CEPH_MDS_OP_SETATTR: return "setattr"; case CEPH_MDS_OP_RMXATTR: return "rmxattr"; diff --git a/fs/ceph/super.c b/fs/ceph/super.c index fd8742bae847..e6987d295079 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -27,6 +27,8 @@ #include <linux/ceph/auth.h> #include <linux/ceph/debugfs.h> +#include <uapi/linux/magic.h> + static DEFINE_SPINLOCK(ceph_fsc_lock); static LIST_HEAD(ceph_fsc_list); @@ -52,8 +54,7 @@ static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf) struct ceph_fs_client *fsc = ceph_inode_to_client(d_inode(dentry)); struct ceph_mon_client *monc = &fsc->client->monc; struct ceph_statfs st; - u64 fsid; - int err; + int i, err; u64 data_pool; if (fsc->mdsc->mdsmap->m_num_data_pg_pools == 1) { @@ -99,12 +100,14 @@ static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_namelen = NAME_MAX; /* Must convert the fsid, for consistent values across arches */ + buf->f_fsid.val[0] = 0; mutex_lock(&monc->mutex); - fsid = le64_to_cpu(*(__le64 *)(&monc->monmap->fsid)) ^ - le64_to_cpu(*((__le64 *)&monc->monmap->fsid + 1)); + for (i = 0 ; i < sizeof(monc->monmap->fsid) / sizeof(__le32) ; ++i) + buf->f_fsid.val[0] ^= le32_to_cpu(((__le32 *)&monc->monmap->fsid)[i]); mutex_unlock(&monc->mutex); - buf->f_fsid = u64_to_fsid(fsid); + /* fold the fs_cluster_id into the upper bits */ + buf->f_fsid.val[1] = monc->fs_cluster_id; return 0; } @@ -145,6 +148,7 @@ enum { Opt_mds_namespace, Opt_recover_session, Opt_source, + Opt_mon_addr, /* string args above */ Opt_dirstat, Opt_rbytes, @@ -158,6 +162,7 @@ enum { Opt_quotadf, Opt_copyfrom, Opt_wsync, + Opt_pagecache, }; enum ceph_recover_session_mode { @@ -196,8 +201,10 @@ static const struct fs_parameter_spec ceph_mount_parameters[] = { fsparam_u32 ("rsize", Opt_rsize), fsparam_string ("snapdirname", Opt_snapdirname), fsparam_string ("source", Opt_source), + fsparam_string ("mon_addr", Opt_mon_addr), fsparam_u32 ("wsize", Opt_wsize), fsparam_flag_no ("wsync", Opt_wsync), + fsparam_flag_no ("pagecache", Opt_pagecache), {} }; @@ -227,9 +234,92 @@ static void canonicalize_path(char *path) } /* - * Parse the source parameter. Distinguish the server list from the path. + * Check if the mds namespace in ceph_mount_options matches + * the passed in namespace string. First time match (when + * ->mds_namespace is NULL) is treated specially, since + * ->mds_namespace needs to be initialized by the caller. + */ +static int namespace_equals(struct ceph_mount_options *fsopt, + const char *namespace, size_t len) +{ + return !(fsopt->mds_namespace && + (strlen(fsopt->mds_namespace) != len || + strncmp(fsopt->mds_namespace, namespace, len))); +} + +static int ceph_parse_old_source(const char *dev_name, const char *dev_name_end, + struct fs_context *fc) +{ + int r; + struct ceph_parse_opts_ctx *pctx = fc->fs_private; + struct ceph_mount_options *fsopt = pctx->opts; + + if (*dev_name_end != ':') + return invalfc(fc, "separator ':' missing in source"); + + r = ceph_parse_mon_ips(dev_name, dev_name_end - dev_name, + pctx->copts, fc->log.log, ','); + if (r) + return r; + + fsopt->new_dev_syntax = false; + return 0; +} + +static int ceph_parse_new_source(const char *dev_name, const char *dev_name_end, + struct fs_context *fc) +{ + size_t len; + struct ceph_fsid fsid; + struct ceph_parse_opts_ctx *pctx = fc->fs_private; + struct ceph_mount_options *fsopt = pctx->opts; + char *fsid_start, *fs_name_start; + + if (*dev_name_end != '=') { + dout("separator '=' missing in source"); + return -EINVAL; + } + + fsid_start = strchr(dev_name, '@'); + if (!fsid_start) + return invalfc(fc, "missing cluster fsid"); + ++fsid_start; /* start of cluster fsid */ + + fs_name_start = strchr(fsid_start, '.'); + if (!fs_name_start) + return invalfc(fc, "missing file system name"); + + if (ceph_parse_fsid(fsid_start, &fsid)) + return invalfc(fc, "Invalid FSID"); + + ++fs_name_start; /* start of file system name */ + len = dev_name_end - fs_name_start; + + if (!namespace_equals(fsopt, fs_name_start, len)) + return invalfc(fc, "Mismatching mds_namespace"); + kfree(fsopt->mds_namespace); + fsopt->mds_namespace = kstrndup(fs_name_start, len, GFP_KERNEL); + if (!fsopt->mds_namespace) + return -ENOMEM; + dout("file system (mds namespace) '%s'\n", fsopt->mds_namespace); + + fsopt->new_dev_syntax = true; + return 0; +} + +/* + * Parse the source parameter for new device format. Distinguish the device + * spec from the path. Try parsing new device format and fallback to old + * format if needed. * - * The source will look like: + * New device syntax will looks like: + * <device_spec>=/<path> + * where + * <device_spec> is name@fsid.fsname + * <path> is optional, but if present must begin with '/' + * (monitor addresses are passed via mount option) + * + * Old device syntax is: * <server_spec>[,<server_spec>...]:[<path>] * where * <server_spec> is <ip>[:<port>] @@ -262,24 +352,44 @@ static int ceph_parse_source(struct fs_parameter *param, struct fs_context *fc) dev_name_end = dev_name + strlen(dev_name); } - dev_name_end--; /* back up to ':' separator */ - if (dev_name_end < dev_name || *dev_name_end != ':') - return invalfc(fc, "No path or : separator in source"); + dev_name_end--; /* back up to separator */ + if (dev_name_end < dev_name) + return invalfc(fc, "Path missing in source"); dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name); if (fsopt->server_path) dout("server path '%s'\n", fsopt->server_path); - ret = ceph_parse_mon_ips(param->string, dev_name_end - dev_name, - pctx->copts, fc->log.log); - if (ret) - return ret; + dout("trying new device syntax"); + ret = ceph_parse_new_source(dev_name, dev_name_end, fc); + if (ret) { + if (ret != -EINVAL) + return ret; + dout("trying old device syntax"); + ret = ceph_parse_old_source(dev_name, dev_name_end, fc); + if (ret) + return ret; + } fc->source = param->string; param->string = NULL; return 0; } +static int ceph_parse_mon_addr(struct fs_parameter *param, + struct fs_context *fc) +{ + struct ceph_parse_opts_ctx *pctx = fc->fs_private; + struct ceph_mount_options *fsopt = pctx->opts; + + kfree(fsopt->mon_addr); + fsopt->mon_addr = param->string; + param->string = NULL; + + return ceph_parse_mon_ips(fsopt->mon_addr, strlen(fsopt->mon_addr), + pctx->copts, fc->log.log, '/'); +} + static int ceph_parse_mount_param(struct fs_context *fc, struct fs_parameter *param) { @@ -305,6 +415,8 @@ static int ceph_parse_mount_param(struct fs_context *fc, param->string = NULL; break; case Opt_mds_namespace: + if (!namespace_equals(fsopt, param->string, strlen(param->string))) + return invalfc(fc, "Mismatching mds_namespace"); kfree(fsopt->mds_namespace); fsopt->mds_namespace = param->string; param->string = NULL; @@ -322,6 +434,8 @@ static int ceph_parse_mount_param(struct fs_context *fc, if (fc->source) return invalfc(fc, "Multiple sources specified"); return ceph_parse_source(param, fc); + case Opt_mon_addr: + return ceph_parse_mon_addr(param, fc); case Opt_wsize: if (result.uint_32 < PAGE_SIZE || result.uint_32 > CEPH_MAX_WRITE_SIZE) @@ -454,6 +568,12 @@ static int ceph_parse_mount_param(struct fs_context *fc, else fsopt->flags |= CEPH_MOUNT_OPT_ASYNC_DIROPS; break; + case Opt_pagecache: + if (result.negated) + fsopt->flags |= CEPH_MOUNT_OPT_NOPAGECACHE; + else + fsopt->flags &= ~CEPH_MOUNT_OPT_NOPAGECACHE; + break; default: BUG(); } @@ -473,6 +593,7 @@ static void destroy_mount_options(struct ceph_mount_options *args) kfree(args->mds_namespace); kfree(args->server_path); kfree(args->fscache_uniq); + kfree(args->mon_addr); kfree(args); } @@ -516,6 +637,10 @@ static int compare_mount_options(struct ceph_mount_options *new_fsopt, if (ret) return ret; + ret = strcmp_null(fsopt1->mon_addr, fsopt2->mon_addr); + if (ret) + return ret; + return ceph_compare_options(new_opt, fsc->client); } @@ -571,14 +696,21 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root) if ((fsopt->flags & CEPH_MOUNT_OPT_NOCOPYFROM) == 0) seq_puts(m, ",copyfrom"); - if (fsopt->mds_namespace) + /* dump mds_namespace when old device syntax is in use */ + if (fsopt->mds_namespace && !fsopt->new_dev_syntax) seq_show_option(m, "mds_namespace", fsopt->mds_namespace); + if (fsopt->mon_addr) + seq_printf(m, ",mon_addr=%s", fsopt->mon_addr); + if (fsopt->flags & CEPH_MOUNT_OPT_CLEANRECOVER) seq_show_option(m, "recover_session", "clean"); - if (fsopt->flags & CEPH_MOUNT_OPT_ASYNC_DIROPS) - seq_puts(m, ",nowsync"); + if (!(fsopt->flags & CEPH_MOUNT_OPT_ASYNC_DIROPS)) + seq_puts(m, ",wsync"); + + if (fsopt->flags & CEPH_MOUNT_OPT_NOPAGECACHE) + seq_puts(m, ",nopagecache"); if (fsopt->wsize != CEPH_MAX_WRITE_SIZE) seq_printf(m, ",wsize=%u", fsopt->wsize); @@ -670,6 +802,7 @@ static struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt, fsc->have_copy_from2 = true; atomic_long_set(&fsc->writeback_count, 0); + fsc->write_congested = false; err = -ENOMEM; /* @@ -732,6 +865,7 @@ static void destroy_fs_client(struct ceph_fs_client *fsc) */ struct kmem_cache *ceph_inode_cachep; struct kmem_cache *ceph_cap_cachep; +struct kmem_cache *ceph_cap_snap_cachep; struct kmem_cache *ceph_cap_flush_cachep; struct kmem_cache *ceph_dentry_cachep; struct kmem_cache *ceph_file_cachep; @@ -760,6 +894,9 @@ static int __init init_caches(void) ceph_cap_cachep = KMEM_CACHE(ceph_cap, SLAB_MEM_SPREAD); if (!ceph_cap_cachep) goto bad_cap; + ceph_cap_snap_cachep = KMEM_CACHE(ceph_cap_snap, SLAB_MEM_SPREAD); + if (!ceph_cap_snap_cachep) + goto bad_cap_snap; ceph_cap_flush_cachep = KMEM_CACHE(ceph_cap_flush, SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD); if (!ceph_cap_flush_cachep) @@ -786,16 +923,10 @@ static int __init init_caches(void) if (!ceph_wb_pagevec_pool) goto bad_pagevec_pool; - error = ceph_fscache_register(); - if (error) - goto bad_fscache; - return 0; -bad_fscache: - kmem_cache_destroy(ceph_mds_request_cachep); bad_pagevec_pool: - mempool_destroy(ceph_wb_pagevec_pool); + kmem_cache_destroy(ceph_mds_request_cachep); bad_mds_req: kmem_cache_destroy(ceph_dir_file_cachep); bad_dir_file: @@ -805,6 +936,8 @@ bad_file: bad_dentry: kmem_cache_destroy(ceph_cap_flush_cachep); bad_cap_flush: + kmem_cache_destroy(ceph_cap_snap_cachep); +bad_cap_snap: kmem_cache_destroy(ceph_cap_cachep); bad_cap: kmem_cache_destroy(ceph_inode_cachep); @@ -821,14 +954,13 @@ static void destroy_caches(void) kmem_cache_destroy(ceph_inode_cachep); kmem_cache_destroy(ceph_cap_cachep); + kmem_cache_destroy(ceph_cap_snap_cachep); kmem_cache_destroy(ceph_cap_flush_cachep); kmem_cache_destroy(ceph_dentry_cachep); kmem_cache_destroy(ceph_file_cachep); kmem_cache_destroy(ceph_dir_file_cachep); kmem_cache_destroy(ceph_mds_request_cachep); mempool_destroy(ceph_wb_pagevec_pool); - - ceph_fscache_unregister(); } static void __ceph_umount_begin(struct ceph_fs_client *fsc) @@ -842,7 +974,7 @@ static void __ceph_umount_begin(struct ceph_fs_client *fsc) * ceph_umount_begin - initiate forced umount. Tear down the * mount, skipping steps that may hang while waiting for server(s). */ -static void ceph_umount_begin(struct super_block *sb) +void ceph_umount_begin(struct super_block *sb) { struct ceph_fs_client *fsc = ceph_sb_to_client(sb); @@ -1059,6 +1191,7 @@ static int ceph_setup_bdi(struct super_block *sb, struct ceph_fs_client *fsc) static int ceph_get_tree(struct fs_context *fc) { struct ceph_parse_opts_ctx *pctx = fc->fs_private; + struct ceph_mount_options *fsopt = pctx->opts; struct super_block *sb; struct ceph_fs_client *fsc; struct dentry *res; @@ -1070,6 +1203,8 @@ static int ceph_get_tree(struct fs_context *fc) if (!fc->source) return invalfc(fc, "No source"); + if (fsopt->new_dev_syntax && !fsopt->mon_addr) + return invalfc(fc, "No monitor address"); /* create client (which we may/may not use) */ fsc = create_fs_client(pctx->opts, pctx->copts); @@ -1155,6 +1290,13 @@ static int ceph_reconfigure_fc(struct fs_context *fc) else ceph_clear_mount_opt(fsc, ASYNC_DIROPS); + if (strcmp_null(fsc->mount_options->mon_addr, fsopt->mon_addr)) { + kfree(fsc->mount_options->mon_addr); + fsc->mount_options->mon_addr = fsopt->mon_addr; + fsopt->mon_addr = NULL; + pr_notice("ceph: monitor addresses recorded, but not used for reconnection"); + } + sync_filesystem(fc->root->d_sb); return 0; } @@ -1332,6 +1474,14 @@ bool disable_send_metrics = false; module_param_cb(disable_send_metrics, ¶m_ops_metrics, &disable_send_metrics, 0644); MODULE_PARM_DESC(disable_send_metrics, "Enable sending perf metrics to ceph cluster (default: on)"); +/* for both v1 and v2 syntax */ +static bool mount_support = true; +static const struct kernel_param_ops param_ops_mount_syntax = { + .get = param_get_bool, +}; +module_param_cb(mount_syntax_v1, ¶m_ops_mount_syntax, &mount_support, 0444); +module_param_cb(mount_syntax_v2, ¶m_ops_mount_syntax, &mount_support, 0444); + module_init(init_ceph); module_exit(exit_ceph); diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 14f951cd5b61..20ceab74e871 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -17,21 +17,16 @@ #include <linux/posix_acl.h> #include <linux/refcount.h> #include <linux/security.h> - -#include <linux/ceph/libceph.h> - -#ifdef CONFIG_CEPH_FSCACHE -#define FSCACHE_USE_NEW_IO_API +#include <linux/netfs.h> #include <linux/fscache.h> -#endif -/* f_type in struct statfs */ -#define CEPH_SUPER_MAGIC 0x00c36400 +#include <linux/ceph/libceph.h> /* large granularity for statfs utilization stats to facilitate * large volume sizes on 32-bit machines. */ #define CEPH_BLOCK_SHIFT 22 /* 4 MB */ #define CEPH_BLOCK (1 << CEPH_BLOCK_SHIFT) +#define CEPH_4K_BLOCK_SHIFT 12 /* 4 KB */ #define CEPH_MOUNT_OPT_CLEANRECOVER (1<<1) /* auto reonnect (clean mode) after blocklisted */ #define CEPH_MOUNT_OPT_DIRSTAT (1<<4) /* `cat dirname` for stats */ @@ -45,10 +40,12 @@ #define CEPH_MOUNT_OPT_NOQUOTADF (1<<13) /* no root dir quota in statfs */ #define CEPH_MOUNT_OPT_NOCOPYFROM (1<<14) /* don't use RADOS 'copy-from' op */ #define CEPH_MOUNT_OPT_ASYNC_DIROPS (1<<15) /* allow async directory ops */ +#define CEPH_MOUNT_OPT_NOPAGECACHE (1<<16) /* bypass pagecache altogether */ #define CEPH_MOUNT_OPT_DEFAULT \ (CEPH_MOUNT_OPT_DCACHE | \ - CEPH_MOUNT_OPT_NOCOPYFROM) + CEPH_MOUNT_OPT_NOCOPYFROM | \ + CEPH_MOUNT_OPT_ASYNC_DIROPS) #define ceph_set_mount_opt(fsc, opt) \ (fsc)->mount_options->flags |= CEPH_MOUNT_OPT_##opt @@ -88,6 +85,8 @@ struct ceph_mount_options { unsigned int max_readdir; /* max readdir result (entries) */ unsigned int max_readdir_bytes; /* max readdir result (bytes) */ + bool new_dev_syntax; + /* * everything above this point can be memcmp'd; everything below * is handled in compare_mount_options() @@ -97,6 +96,7 @@ struct ceph_mount_options { char *mds_namespace; /* default NULL */ char *server_path; /* default NULL (means "/") */ char *fscache_uniq; /* default NULL */ + char *mon_addr; }; struct ceph_fs_client { @@ -119,6 +119,7 @@ struct ceph_fs_client { struct ceph_mds_client *mdsc; atomic_long_t writeback_count; + bool write_congested; struct workqueue_struct *inode_wq; struct workqueue_struct *cap_wq; @@ -128,13 +129,13 @@ struct ceph_fs_client { struct dentry *debugfs_congestion_kb; struct dentry *debugfs_bdi; struct dentry *debugfs_mdsc, *debugfs_mdsmap; - struct dentry *debugfs_metric; struct dentry *debugfs_status; struct dentry *debugfs_mds_sessions; + struct dentry *debugfs_metrics_dir; #endif #ifdef CONFIG_CEPH_FSCACHE - struct fscache_cookie *fscache; + struct fscache_volume *fscache; #endif }; @@ -228,7 +229,7 @@ static inline void ceph_put_cap_snap(struct ceph_cap_snap *capsnap) if (refcount_dec_and_test(&capsnap->nref)) { if (capsnap->xattr_blob) ceph_buffer_put(capsnap->xattr_blob); - kfree(capsnap); + kmem_cache_free(ceph_cap_snap_cachep, capsnap); } } @@ -315,6 +316,11 @@ struct ceph_inode_xattrs_info { * Ceph inode. */ struct ceph_inode_info { + struct { + /* These must be contiguous */ + struct inode vfs_inode; + struct netfs_i_context netfs_ctx; /* Netfslib context */ + }; struct ceph_vino i_vino; /* ceph ino + snap */ spinlock_t i_ceph_lock; @@ -425,11 +431,6 @@ struct ceph_inode_info { struct work_struct i_work; unsigned long i_work_mask; - -#ifdef CONFIG_CEPH_FSCACHE - struct fscache_cookie *fscache; -#endif - struct inode vfs_inode; /* at end */ }; static inline struct ceph_inode_info * @@ -534,19 +535,23 @@ static inline int ceph_ino_compare(struct inode *inode, void *data) * * These come from src/mds/mdstypes.h in the ceph sources. */ -#define CEPH_MAX_MDS 0x100 -#define CEPH_NUM_STRAY 10 +#define CEPH_MAX_MDS 0x100 +#define CEPH_NUM_STRAY 10 #define CEPH_MDS_INO_MDSDIR_OFFSET (1 * CEPH_MAX_MDS) +#define CEPH_MDS_INO_LOG_OFFSET (2 * CEPH_MAX_MDS) #define CEPH_INO_SYSTEM_BASE ((6*CEPH_MAX_MDS) + (CEPH_MAX_MDS * CEPH_NUM_STRAY)) static inline bool ceph_vino_is_reserved(const struct ceph_vino vino) { - if (vino.ino < CEPH_INO_SYSTEM_BASE && - vino.ino >= CEPH_MDS_INO_MDSDIR_OFFSET) { - WARN_RATELIMIT(1, "Attempt to access reserved inode number 0x%llx", vino.ino); - return true; - } - return false; + if (vino.ino >= CEPH_INO_SYSTEM_BASE || + vino.ino < CEPH_MDS_INO_MDSDIR_OFFSET) + return false; + + /* Don't warn on mdsdirs */ + WARN_RATELIMIT(vino.ino >= CEPH_MDS_INO_LOG_OFFSET, + "Attempt to access reserved inode number 0x%llx", + vino.ino); + return true; } static inline struct inode *ceph_find_inode(struct super_block *sb, @@ -580,6 +585,7 @@ static inline struct inode *ceph_find_inode(struct super_block *sb, #define CEPH_I_ODIRECT (1 << 11) /* inode in direct I/O mode */ #define CEPH_ASYNC_CREATE_BIT (12) /* async create in flight for this */ #define CEPH_I_ASYNC_CREATE (1 << CEPH_ASYNC_CREATE_BIT) +#define CEPH_I_SHUTDOWN (1 << 13) /* inode is no longer usable */ /* * Masks of ceph inode work. @@ -876,6 +882,8 @@ struct ceph_snap_realm { struct list_head dirty_item; /* if realm needs new context */ + struct list_head rebuild_item; /* rebuild snap realms _downward_ in hierarchy */ + /* the current set of snaps for this realm */ struct ceph_snap_context *cached_context; @@ -931,7 +939,7 @@ extern void ceph_handle_snap(struct ceph_mds_client *mdsc, struct ceph_msg *msg); extern int __ceph_finish_cap_snap(struct ceph_inode_info *ci, struct ceph_cap_snap *capsnap); -extern void ceph_cleanup_empty_realms(struct ceph_mds_client *mdsc); +extern void ceph_cleanup_global_and_empty_realms(struct ceph_mds_client *mdsc); extern struct ceph_snapid_map *ceph_get_snapid_map(struct ceph_mds_client *mdsc, u64 snap); @@ -939,6 +947,7 @@ extern void ceph_put_snapid_map(struct ceph_mds_client* mdsc, struct ceph_snapid_map *sm); extern void ceph_trim_snapid_map(struct ceph_mds_client *mdsc); extern void ceph_cleanup_snapid_map(struct ceph_mds_client *mdsc); +void ceph_umount_begin(struct super_block *sb); /* @@ -1027,9 +1036,20 @@ extern int ceph_setattr(struct user_namespace *mnt_userns, extern int ceph_getattr(struct user_namespace *mnt_userns, const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags); +void ceph_inode_shutdown(struct inode *inode); + +static inline bool ceph_inode_is_shutdown(struct inode *inode) +{ + unsigned long flags = READ_ONCE(ceph_inode(inode)->i_ceph_flags); + struct ceph_fs_client *fsc = ceph_inode_to_client(inode); + int state = READ_ONCE(fsc->mount_state); + + return (flags & CEPH_I_SHUTDOWN) || state >= CEPH_MOUNT_SHUTDOWN; +} /* xattr.c */ int __ceph_setxattr(struct inode *, const char *, const void *, size_t, int); +int ceph_do_getvxattr(struct inode *inode, const char *name, void *value, size_t size); ssize_t __ceph_getxattr(struct inode *, const char *, void *, size_t); extern ssize_t ceph_listxattr(struct dentry *, char *, size_t); extern struct ceph_buffer *__ceph_build_xattrs_blob(struct ceph_inode_info *ci); @@ -1194,10 +1214,12 @@ extern void __ceph_touch_fmode(struct ceph_inode_info *ci, /* addr.c */ extern const struct address_space_operations ceph_aops; +extern const struct netfs_request_ops ceph_netfs_ops; extern int ceph_mmap(struct file *file, struct vm_area_struct *vma); -extern int ceph_uninline_data(struct file *filp, struct page *locked_page); +extern int ceph_uninline_data(struct file *file); extern int ceph_pool_perm_check(struct inode *inode, int need); extern void ceph_pool_perm_destroy(struct ceph_mds_client* mdsc); +int ceph_purge_inode_cap(struct inode *inode, struct ceph_cap *cap, bool *invalidate); /* file.c */ extern const struct file_operations ceph_file_fops; diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 159a1ffa4f4b..afec84088471 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -923,10 +923,13 @@ ssize_t __ceph_getxattr(struct inode *inode, const char *name, void *value, { struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_inode_xattr *xattr; - struct ceph_vxattr *vxattr = NULL; + struct ceph_vxattr *vxattr; int req_mask; ssize_t err; + if (strncmp(name, XATTR_CEPH_PREFIX, XATTR_CEPH_PREFIX_LEN)) + goto handle_non_vxattrs; + /* let's see if a virtual xattr was requested */ vxattr = ceph_match_vxattr(inode, name); if (vxattr) { @@ -945,8 +948,14 @@ ssize_t __ceph_getxattr(struct inode *inode, const char *name, void *value, err = -ERANGE; } return err; + } else { + err = ceph_do_getvxattr(inode, name, value, size); + /* this would happen with a new client and old server combo */ + if (err == -EOPNOTSUPP) + err = -ENODATA; + return err; } - +handle_non_vxattrs: req_mask = __get_request_mask(inode); spin_lock(&ci->i_ceph_lock); @@ -1311,7 +1320,7 @@ int ceph_security_init_secctx(struct dentry *dentry, umode_t mode, int err; err = security_dentry_init_security(dentry, mode, &dentry->d_name, - &as_ctx->sec_ctx, + &name, &as_ctx->sec_ctx, &as_ctx->sec_ctxlen); if (err < 0) { WARN_ON_ONCE(err != -EOPNOTSUPP); @@ -1335,7 +1344,6 @@ int ceph_security_init_secctx(struct dentry *dentry, umode_t mode, * It only supports single security module and only selinux has * dentry_init_security hook. */ - name = XATTR_NAME_SELINUX; name_len = strlen(name); err = ceph_pagelist_reserve(pagelist, 4 * 2 + name_len + as_ctx->sec_ctxlen); |