diff options
Diffstat (limited to 'fs/ceph/file.c')
| -rw-r--r-- | fs/ceph/file.c | 1019 |
1 files changed, 739 insertions, 280 deletions
diff --git a/fs/ceph/file.c b/fs/ceph/file.c index e77c28cf369..302085100c2 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -4,11 +4,15 @@ #include <linux/sched.h> #include <linux/slab.h> #include <linux/file.h> +#include <linux/mount.h> #include <linux/namei.h> #include <linux/writeback.h> +#include <linux/aio.h> +#include <linux/falloc.h> #include "super.h" #include "mds_client.h" +#include "cache.h" /* * Ceph file operations @@ -54,7 +58,6 @@ prepare_open_request(struct super_block *sb, int flags, int create_mode) req->r_fmode = ceph_flags_to_mode(flags); req->r_args.open.flags = cpu_to_le32(flags); req->r_args.open.mode = cpu_to_le32(create_mode); - req->r_args.open.preferred = cpu_to_le32(-1); out: return req; } @@ -67,9 +70,23 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode) { struct ceph_file_info *cf; int ret = 0; + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb); + struct ceph_mds_client *mdsc = fsc->mdsc; switch (inode->i_mode & S_IFMT) { case S_IFREG: + /* First file open request creates the cookie, we want to keep + * this cookie around for the filetime of the inode as not to + * have to worry about fscache register / revoke / operation + * races. + * + * Also, if we know the operation is going to invalidate data + * (non readonly) just nuke the cache right away. + */ + ceph_fscache_register_inode_cookie(mdsc->fsc, ci); + if ((fmode & CEPH_FILE_MODE_WR)) + ceph_fscache_invalidate(inode); case S_IFDIR: dout("init_file %p %p 0%o (regular)\n", inode, file, inode->i_mode); @@ -107,9 +124,6 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode) } /* - * If the filp already has private_data, that means the file was - * already opened by intent during lookup, and we do nothing. - * * If we already have the requisite capabilities, we can satisfy * the open request locally (no need to request new caps from the * MDS). We do, however, need to inform the MDS (asynchronously) @@ -122,7 +136,7 @@ int ceph_open(struct inode *inode, struct file *file) struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; struct ceph_file_info *cf = file->private_data; - struct inode *parent_inode = file->f_dentry->d_parent->d_inode; + struct inode *parent_inode = NULL; int err; int flags, fmode, wanted; @@ -147,18 +161,20 @@ int ceph_open(struct inode *inode, struct file *file) /* trivially open snapdir */ if (ceph_snap(inode) == CEPH_SNAPDIR) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ceph_init_file(inode, file, fmode); } /* - * No need to block if we have any caps. Update wanted set + * No need to block if we have caps on the auth MDS (for + * write) or any MDS (for read). Update wanted set * asynchronously. */ - spin_lock(&inode->i_lock); - if (__ceph_is_any_real_caps(ci)) { + spin_lock(&ci->i_ceph_lock); + if (__ceph_is_any_real_caps(ci) && + (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) { int mds_wanted = __ceph_caps_mds_wanted(ci); int issued = __ceph_caps_issued(ci, NULL); @@ -166,7 +182,7 @@ int ceph_open(struct inode *inode, struct file *file) inode, fmode, ceph_cap_string(wanted), ceph_cap_string(issued)); __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* adjust wanted? */ if ((issued & wanted) != wanted && @@ -178,10 +194,11 @@ int ceph_open(struct inode *inode, struct file *file) } else if (ceph_snap(inode) != CEPH_NOSNAP && (ci->i_snap_caps & wanted) == wanted) { __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ceph_init_file(inode, file, fmode); } - spin_unlock(&inode->i_lock); + + spin_unlock(&ci->i_ceph_lock); dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted)); req = prepare_open_request(inode->i_sb, flags, 0); @@ -189,9 +206,14 @@ int ceph_open(struct inode *inode, struct file *file) err = PTR_ERR(req); goto out; } - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); + req->r_num_caps = 1; + if (flags & O_CREAT) + parent_inode = ceph_get_dentry_parent_inode(file->f_dentry); err = ceph_mdsc_do_request(mdsc, parent_inode, req); + iput(parent_inode); if (!err) err = ceph_init_file(inode, file, req->r_fmode); ceph_mdsc_put_request(req); @@ -202,36 +224,34 @@ out: /* - * Do a lookup + open with a single request. - * - * If this succeeds, but some subsequent check in the vfs - * may_open() fails, the struct *file gets cleaned up (i.e. - * ceph_release gets called). So fear not! - */ -/* - * flags - * path_lookup_open -> LOOKUP_OPEN - * path_lookup_create -> LOOKUP_OPEN|LOOKUP_CREATE + * Do a lookup + open with a single request. If we get a non-existent + * file or symlink, return 1 so the VFS can retry. */ -struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry, - struct nameidata *nd, int mode, - int locked_dir) +int ceph_atomic_open(struct inode *dir, struct dentry *dentry, + struct file *file, unsigned flags, umode_t mode, + int *opened) { struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; - struct file *file = nd->intent.open.file; - struct inode *parent_inode = get_dentry_parent_inode(file->f_dentry); struct ceph_mds_request *req; + struct dentry *dn; int err; - int flags = nd->intent.open.flags - 1; /* silly vfs! */ - dout("ceph_lookup_open dentry %p '%.*s' flags %d mode 0%o\n", - dentry, dentry->d_name.len, dentry->d_name.name, flags, mode); + dout("atomic_open %p dentry %p '%.*s' %s flags %d mode 0%o\n", + dir, dentry, dentry->d_name.len, dentry->d_name.name, + d_unhashed(dentry) ? "unhashed" : "hashed", flags, mode); + + if (dentry->d_name.len > NAME_MAX) + return -ENAMETOOLONG; + + err = ceph_init_dentry(dentry); + if (err < 0) + return err; /* do the open */ req = prepare_open_request(dir->i_sb, flags, mode); if (IS_ERR(req)) - return ERR_CAST(req); + return PTR_ERR(req); req->r_dentry = dget(dentry); req->r_num_caps = 2; if (flags & O_CREAT) { @@ -239,16 +259,44 @@ struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry, req->r_dentry_unless = CEPH_CAP_FILE_EXCL; } req->r_locked_dir = dir; /* caller holds dir->i_mutex */ - err = ceph_mdsc_do_request(mdsc, parent_inode, req); - dentry = ceph_finish_lookup(req, dentry, err); - if (!err && (flags & O_CREAT) && !req->r_reply_info.head->is_dentry) + err = ceph_mdsc_do_request(mdsc, + (flags & (O_CREAT|O_TRUNC)) ? dir : NULL, + req); + if (err) + goto out_err; + + err = ceph_handle_snapdir(req, dentry, err); + if (err == 0 && (flags & O_CREAT) && !req->r_reply_info.head->is_dentry) err = ceph_handle_notrace_create(dir, dentry); - if (!err) - err = ceph_init_file(req->r_dentry->d_inode, file, - req->r_fmode); + + if (d_unhashed(dentry)) { + dn = ceph_finish_lookup(req, dentry, err); + if (IS_ERR(dn)) + err = PTR_ERR(dn); + } else { + /* we were given a hashed negative dentry */ + dn = NULL; + } + if (err) + goto out_err; + if (dn || dentry->d_inode == NULL || S_ISLNK(dentry->d_inode->i_mode)) { + /* make vfs retry on splice, ENOENT, or symlink */ + dout("atomic_open finish_no_open on dn %p\n", dn); + err = finish_no_open(file, dn); + } else { + dout("atomic_open finish_open on dn %p\n", dn); + if (req->r_op == CEPH_MDS_OP_CREATE && req->r_reply_info.has_create_ino) { + ceph_init_acl(dentry, dentry->d_inode, dir); + *opened |= FILE_CREATED; + } + err = finish_open(file, dentry, ceph_open, opened); + } +out_err: + if (!req->r_err && req->r_target_inode) + ceph_put_fmode(ceph_inode(req->r_target_inode), req->r_fmode); ceph_mdsc_put_request(req); - dout("ceph_lookup_open result=%p\n", dentry); - return dentry; + dout("atomic_open result=%d\n", err); + return err; } int ceph_release(struct inode *inode, struct file *file) @@ -280,13 +328,14 @@ int ceph_release(struct inode *inode, struct file *file) static int striped_read(struct inode *inode, u64 off, u64 len, struct page **pages, int num_pages, - int *checkeof) + int *checkeof, bool o_direct, + unsigned long buf_align) { struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_inode_info *ci = ceph_inode(inode); - u64 pos, this_len; - int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */ - int left, pages_left; + u64 pos, this_len, left; + int io_align, page_align; + int pages_left; int read; struct page **page_pos; int ret; @@ -300,58 +349,57 @@ static int striped_read(struct inode *inode, page_pos = pages; pages_left = num_pages; read = 0; + io_align = off & ~PAGE_MASK; more: + if (o_direct) + page_align = (pos - io_align + buf_align) & ~PAGE_MASK; + else + page_align = pos & ~PAGE_MASK; this_len = left; ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode), &ci->i_layout, pos, &this_len, ci->i_truncate_seq, ci->i_truncate_size, - page_pos, pages_left); - hit_stripe = this_len < left; - was_short = ret >= 0 && ret < this_len; + page_pos, pages_left, page_align); if (ret == -ENOENT) ret = 0; - dout("striped_read %llu~%u (read %u) got %d%s%s\n", pos, left, read, + hit_stripe = this_len < left; + was_short = ret >= 0 && ret < this_len; + dout("striped_read %llu~%llu (read %u) got %d%s%s\n", pos, left, read, ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : ""); - if (ret > 0) { - int didpages = - ((pos & ~PAGE_CACHE_MASK) + ret) >> PAGE_CACHE_SHIFT; - - if (read < pos - off) { - dout(" zero gap %llu to %llu\n", off + read, pos); - ceph_zero_page_vector_range(page_off + read, - pos - off - read, pages); + if (ret >= 0) { + int didpages; + if (was_short && (pos + ret < inode->i_size)) { + u64 tmp = min(this_len - ret, + inode->i_size - pos - ret); + dout(" zero gap %llu to %llu\n", + pos + ret, pos + ret + tmp); + ceph_zero_page_vector_range(page_align + read + ret, + tmp, pages); + ret += tmp; } + + didpages = (page_align + ret) >> PAGE_CACHE_SHIFT; pos += ret; read = pos - off; left -= ret; page_pos += didpages; pages_left -= didpages; - /* hit stripe? */ - if (left && hit_stripe) + /* hit stripe and need continue*/ + if (left && hit_stripe && pos < inode->i_size) goto more; } - if (was_short) { - /* was original extent fully inside i_size? */ - if (pos + left <= inode->i_size) { - dout("zero tail\n"); - ceph_zero_page_vector_range(page_off + read, len - read, - pages); - read = len; - goto out; - } - - /* check i_size */ - *checkeof = 1; + if (read > 0) { + ret = read; + /* did we bounce off eof? */ + if (pos + left > inode->i_size) + *checkeof = 1; } -out: - if (ret >= 0) - ret = read; dout("striped_read returns %d\n", ret); return ret; } @@ -362,213 +410,361 @@ out: * * If the read spans object boundary, just do multiple reads. */ -static ssize_t ceph_sync_read(struct file *file, char __user *data, - unsigned len, loff_t *poff, int *checkeof) +static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i, + int *checkeof) { - struct inode *inode = file->f_dentry->d_inode; + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); struct page **pages; - u64 off = *poff; - int num_pages = calc_pages_for(off, len); - int ret; + u64 off = iocb->ki_pos; + int num_pages, ret; + size_t len = iov_iter_count(i); - dout("sync_read on file %p %llu~%u %s\n", file, off, len, + dout("sync_read on file %p %llu~%u %s\n", file, off, + (unsigned)len, (file->f_flags & O_DIRECT) ? "O_DIRECT" : ""); + /* + * flush any page cache pages in this range. this + * will make concurrent normal and sync io slow, + * but it will at least behave sensibly when they are + * in sequence. + */ + ret = filemap_write_and_wait_range(inode->i_mapping, off, + off + len); + if (ret < 0) + return ret; if (file->f_flags & O_DIRECT) { - pages = ceph_get_direct_page_vector(data, num_pages, off, len); + while (iov_iter_count(i)) { + size_t start; + ssize_t n; - /* - * flush any page cache pages in this range. this - * will make concurrent normal and O_DIRECT io slow, - * but it will at least behave sensibly when they are - * in sequence. - */ - } else { - pages = ceph_alloc_page_vector(num_pages, GFP_NOFS); - } - if (IS_ERR(pages)) - return PTR_ERR(pages); + n = iov_iter_get_pages_alloc(i, &pages, INT_MAX, &start); + if (n < 0) + return n; - ret = filemap_write_and_wait(inode->i_mapping); - if (ret < 0) - goto done; + num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE; - ret = striped_read(inode, off, len, pages, num_pages, checkeof); + ret = striped_read(inode, off, n, + pages, num_pages, checkeof, + 1, start); - if (ret >= 0 && (file->f_flags & O_DIRECT) == 0) - ret = ceph_copy_page_vector_to_user(pages, data, off, ret); - if (ret >= 0) - *poff = off + ret; + ceph_put_page_vector(pages, num_pages, true); -done: - if (file->f_flags & O_DIRECT) - ceph_put_page_vector(pages, num_pages); - else + if (ret <= 0) + break; + off += ret; + iov_iter_advance(i, ret); + if (ret < n) + break; + } + } else { + num_pages = calc_pages_for(off, len); + pages = ceph_alloc_page_vector(num_pages, GFP_NOFS); + if (IS_ERR(pages)) + return PTR_ERR(pages); + ret = striped_read(inode, off, len, pages, + num_pages, checkeof, 0, 0); + if (ret > 0) { + int l, k = 0; + size_t left = ret; + + while (left) { + int copy = min_t(size_t, PAGE_SIZE, left); + l = copy_page_to_iter(pages[k++], 0, copy, i); + off += l; + left -= l; + if (l < copy) + break; + } + } ceph_release_page_vector(pages, num_pages); + } + + if (off > iocb->ki_pos) { + ret = off - iocb->ki_pos; + iocb->ki_pos = off; + } + dout("sync_read result %d\n", ret); return ret; } /* - * Write commit callback, called if we requested both an ACK and - * ONDISK commit reply from the OSD. + * Write commit request unsafe callback, called to tell us when a + * request is unsafe (that is, in flight--has been handed to the + * messenger to send to its target osd). It is called again when + * we've received a response message indicating the request is + * "safe" (its CEPH_OSD_FLAG_ONDISK flag is set), or when a request + * is completed early (and unsuccessfully) due to a timeout or + * interrupt. + * + * This is used if we requested both an ACK and ONDISK commit reply + * from the OSD. */ -static void sync_write_commit(struct ceph_osd_request *req, - struct ceph_msg *msg) +static void ceph_sync_write_unsafe(struct ceph_osd_request *req, bool unsafe) { struct ceph_inode_info *ci = ceph_inode(req->r_inode); - dout("sync_write_commit %p tid %llu\n", req, req->r_tid); - spin_lock(&ci->i_unsafe_lock); - list_del_init(&req->r_unsafe_item); - spin_unlock(&ci->i_unsafe_lock); - ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR); + dout("%s %p tid %llu %ssafe\n", __func__, req, req->r_tid, + unsafe ? "un" : ""); + if (unsafe) { + ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR); + spin_lock(&ci->i_unsafe_lock); + list_add_tail(&req->r_unsafe_item, + &ci->i_unsafe_writes); + spin_unlock(&ci->i_unsafe_lock); + } else { + spin_lock(&ci->i_unsafe_lock); + list_del_init(&req->r_unsafe_item); + spin_unlock(&ci->i_unsafe_lock); + ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR); + } } + /* - * Synchronous write, straight from __user pointer or user pages (if - * O_DIRECT). + * Synchronous write, straight from __user pointer or user pages. * * If write spans object boundary, just do multiple writes. (For a * correct atomic write, we should e.g. take write locks on all * objects, rollback on failure, etc.) */ -static ssize_t ceph_sync_write(struct file *file, const char __user *data, - size_t left, loff_t *offset) +static ssize_t +ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from) { - struct inode *inode = file->f_dentry->d_inode; + struct file *file = iocb->ki_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_snap_context *snapc; + struct ceph_vino vino; struct ceph_osd_request *req; struct page **pages; int num_pages; - long long unsigned pos; - u64 len; int written = 0; int flags; - int do_sync = 0; int check_caps = 0; int ret; struct timespec mtime = CURRENT_TIME; + loff_t pos = iocb->ki_pos; + size_t count = iov_iter_count(from); - if (ceph_snap(file->f_dentry->d_inode) != CEPH_NOSNAP) + if (ceph_snap(file_inode(file)) != CEPH_NOSNAP) return -EROFS; - dout("sync_write on file %p %lld~%u %s\n", file, *offset, - (unsigned)left, (file->f_flags & O_DIRECT) ? "O_DIRECT" : ""); - - if (file->f_flags & O_APPEND) - pos = i_size_read(inode); - else - pos = *offset; + dout("sync_direct_write on file %p %lld~%u\n", file, pos, + (unsigned)count); - ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + left); + ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + count); if (ret < 0) return ret; ret = invalidate_inode_pages2_range(inode->i_mapping, pos >> PAGE_CACHE_SHIFT, - (pos + left) >> PAGE_CACHE_SHIFT); + (pos + count) >> PAGE_CACHE_SHIFT); if (ret < 0) dout("invalidate_inode_pages2_range returned %d\n", ret); flags = CEPH_OSD_FLAG_ORDERSNAP | CEPH_OSD_FLAG_ONDISK | CEPH_OSD_FLAG_WRITE; - if ((file->f_flags & (O_SYNC|O_DIRECT)) == 0) - flags |= CEPH_OSD_FLAG_ACK; - else - do_sync = 1; - /* - * we may need to do multiple writes here if we span an object - * boundary. this isn't atomic, unfortunately. :( - */ -more: - len = left; - req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, - ceph_vino(inode), pos, &len, - CEPH_OSD_OP_WRITE, flags, - ci->i_snap_realm->cached_context, - do_sync, - ci->i_truncate_seq, ci->i_truncate_size, - &mtime, false, 2); - if (!req) - return -ENOMEM; - - num_pages = calc_pages_for(pos, len); + while (iov_iter_count(from) > 0) { + u64 len = iov_iter_single_seg_count(from); + size_t start; + ssize_t n; + + snapc = ci->i_snap_realm->cached_context; + vino = ceph_vino(inode); + req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, + vino, pos, &len, + 2,/*include a 'startsync' command*/ + CEPH_OSD_OP_WRITE, flags, snapc, + ci->i_truncate_seq, + ci->i_truncate_size, + false); + if (IS_ERR(req)) { + ret = PTR_ERR(req); + break; + } - if (file->f_flags & O_DIRECT) { - pages = ceph_get_direct_page_vector(data, num_pages, pos, len); - if (IS_ERR(pages)) { - ret = PTR_ERR(pages); - goto out; + n = iov_iter_get_pages_alloc(from, &pages, len, &start); + if (unlikely(n < 0)) { + ret = n; + ceph_osdc_put_request(req); + break; } + num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE; /* * throw out any page cache pages in this range. this * may block. */ truncate_inode_pages_range(inode->i_mapping, pos, - (pos+len) | (PAGE_CACHE_SIZE-1)); - } else { + (pos+n) | (PAGE_CACHE_SIZE-1)); + osd_req_op_extent_osd_data_pages(req, 0, pages, n, start, + false, false); + + /* BUG_ON(vino.snap != CEPH_NOSNAP); */ + ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime); + + ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); + if (!ret) + ret = ceph_osdc_wait_request(&fsc->client->osdc, req); + + ceph_put_page_vector(pages, num_pages, false); + + ceph_osdc_put_request(req); + if (ret) + break; + pos += n; + written += n; + iov_iter_advance(from, n); + + if (pos > i_size_read(inode)) { + check_caps = ceph_inode_set_size(inode, pos); + if (check_caps) + ceph_check_caps(ceph_inode(inode), + CHECK_CAPS_AUTHONLY, + NULL); + } + } + + if (ret != -EOLDSNAPC && written > 0) { + iocb->ki_pos = pos; + ret = written; + } + return ret; +} + + +/* + * Synchronous write, straight from __user pointer or user pages. + * + * If write spans object boundary, just do multiple writes. (For a + * correct atomic write, we should e.g. take write locks on all + * objects, rollback on failure, etc.) + */ +static ssize_t ceph_sync_write(struct kiocb *iocb, struct iov_iter *from) +{ + struct file *file = iocb->ki_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_snap_context *snapc; + struct ceph_vino vino; + struct ceph_osd_request *req; + struct page **pages; + u64 len; + int num_pages; + int written = 0; + int flags; + int check_caps = 0; + int ret; + struct timespec mtime = CURRENT_TIME; + loff_t pos = iocb->ki_pos; + size_t count = iov_iter_count(from); + + if (ceph_snap(file_inode(file)) != CEPH_NOSNAP) + return -EROFS; + + dout("sync_write on file %p %lld~%u\n", file, pos, (unsigned)count); + + ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + count); + if (ret < 0) + return ret; + + ret = invalidate_inode_pages2_range(inode->i_mapping, + pos >> PAGE_CACHE_SHIFT, + (pos + count) >> PAGE_CACHE_SHIFT); + if (ret < 0) + dout("invalidate_inode_pages2_range returned %d\n", ret); + + flags = CEPH_OSD_FLAG_ORDERSNAP | + CEPH_OSD_FLAG_ONDISK | + CEPH_OSD_FLAG_WRITE | + CEPH_OSD_FLAG_ACK; + + while ((len = iov_iter_count(from)) > 0) { + size_t left; + int n; + + snapc = ci->i_snap_realm->cached_context; + vino = ceph_vino(inode); + req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, + vino, pos, &len, 1, + CEPH_OSD_OP_WRITE, flags, snapc, + ci->i_truncate_seq, + ci->i_truncate_size, + false); + if (IS_ERR(req)) { + ret = PTR_ERR(req); + break; + } + + /* + * write from beginning of first page, + * regardless of io alignment + */ + num_pages = (len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + pages = ceph_alloc_page_vector(num_pages, GFP_NOFS); if (IS_ERR(pages)) { ret = PTR_ERR(pages); goto out; } - ret = ceph_copy_user_to_page_vector(pages, data, pos, len); + + left = len; + for (n = 0; n < num_pages; n++) { + size_t plen = min_t(size_t, left, PAGE_SIZE); + ret = copy_page_from_iter(pages[n], 0, plen, from); + if (ret != plen) { + ret = -EFAULT; + break; + } + left -= ret; + } + if (ret < 0) { ceph_release_page_vector(pages, num_pages); goto out; } - if ((file->f_flags & O_SYNC) == 0) { - /* get a second commit callback */ - req->r_safe_callback = sync_write_commit; - req->r_own_pages = 1; - } - } - req->r_pages = pages; - req->r_num_pages = num_pages; - req->r_inode = inode; + /* get a second commit callback */ + req->r_unsafe_callback = ceph_sync_write_unsafe; + req->r_inode = inode; - ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); - if (!ret) { - if (req->r_safe_callback) { - /* - * Add to inode unsafe list only after we - * start_request so that a tid has been assigned. - */ - spin_lock(&ci->i_unsafe_lock); - list_add(&req->r_unsafe_item, &ci->i_unsafe_writes); - spin_unlock(&ci->i_unsafe_lock); - ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR); - } - ret = ceph_osdc_wait_request(&fsc->client->osdc, req); - } + osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0, + false, true); - if (file->f_flags & O_DIRECT) - ceph_put_page_vector(pages, num_pages); - else if (file->f_flags & O_SYNC) - ceph_release_page_vector(pages, num_pages); + /* BUG_ON(vino.snap != CEPH_NOSNAP); */ + ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime); + + ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); + if (!ret) + ret = ceph_osdc_wait_request(&fsc->client->osdc, req); out: - ceph_osdc_put_request(req); - if (ret == 0) { - pos += len; - written += len; - left -= len; - if (left) - goto more; + ceph_osdc_put_request(req); + if (ret == 0) { + pos += len; + written += len; + + if (pos > i_size_read(inode)) { + check_caps = ceph_inode_set_size(inode, pos); + if (check_caps) + ceph_check_caps(ceph_inode(inode), + CHECK_CAPS_AUTHONLY, + NULL); + } + } else + break; + } + if (ret != -EOLDSNAPC && written > 0) { ret = written; - *offset = pos; - if (pos > i_size_read(inode)) - check_caps = ceph_inode_set_size(inode, pos); - if (check_caps) - ceph_check_caps(ceph_inode(inode), CHECK_CAPS_AUTHONLY, - NULL); + iocb->ki_pos = pos; } return ret; } @@ -580,61 +776,69 @@ out: * * Hmm, the sync read case isn't actually async... should it be? */ -static ssize_t ceph_aio_read(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +static ssize_t ceph_read_iter(struct kiocb *iocb, struct iov_iter *to) { struct file *filp = iocb->ki_filp; struct ceph_file_info *fi = filp->private_data; - loff_t *ppos = &iocb->ki_pos; - size_t len = iov->iov_len; - struct inode *inode = filp->f_dentry->d_inode; + size_t len = iocb->ki_nbytes; + struct inode *inode = file_inode(filp); struct ceph_inode_info *ci = ceph_inode(inode); - void __user *base = iov->iov_base; ssize_t ret; int want, got = 0; int checkeof = 0, read = 0; - dout("aio_read %p %llx.%llx %llu~%u trying to get caps on %p\n", - inode, ceph_vinop(inode), pos, (unsigned)len, inode); again: - __ceph_do_pending_vmtruncate(inode); + 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 (fi->fmode & CEPH_FILE_MODE_LAZY) want = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO; else want = CEPH_CAP_FILE_CACHE; ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, &got, -1); if (ret < 0) - goto out; - dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n", - inode, ceph_vinop(inode), pos, (unsigned)len, - ceph_cap_string(got)); + return ret; if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 || (iocb->ki_filp->f_flags & O_DIRECT) || - (inode->i_sb->s_flags & MS_SYNCHRONOUS)) + (fi->flags & CEPH_F_SYNC)) { + + dout("aio_sync_read %p %llx.%llx %llu~%u got cap refs on %s\n", + inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len, + ceph_cap_string(got)); + /* hmm, this isn't really async... */ - ret = ceph_sync_read(filp, base, len, ppos, &checkeof); - else - ret = generic_file_aio_read(iocb, iov, nr_segs, pos); + ret = ceph_sync_read(iocb, to, &checkeof); + } else { + dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n", + inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len, + ceph_cap_string(got)); -out: + ret = generic_file_read_iter(iocb, to); + } dout("aio_read %p %llx.%llx dropping cap refs on %s = %d\n", inode, ceph_vinop(inode), ceph_cap_string(got), (int)ret); ceph_put_cap_refs(ci, got); if (checkeof && ret >= 0) { - int statret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE); + int statret = ceph_do_getattr(inode, + CEPH_STAT_CAP_SIZE); /* hit EOF or hole? */ - if (statret == 0 && *ppos < inode->i_size) { - dout("aio_read sync_read hit hole, reading more\n"); + if (statret == 0 && iocb->ki_pos < inode->i_size && + ret < len) { + dout("sync_read hit hole, ppos %lld < size %lld" + ", reading more\n", iocb->ki_pos, + inode->i_size); + + iov_iter_advance(to, ret); read += ret; - base += ret; len -= ret; checkeof = 0; goto again; } } + if (ret >= 0) ret += read; @@ -651,95 +855,150 @@ out: * * If we are near ENOSPC, write synchronously. */ -static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; struct ceph_file_info *fi = file->private_data; - struct inode *inode = file->f_dentry->d_inode; + struct inode *inode = file_inode(file); struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->client->osdc; - loff_t endoff = pos + iov->iov_len; - int want, got = 0; - int ret, err; + ssize_t count = iov_iter_count(from), written = 0; + int err, want, got; + loff_t pos = iocb->ki_pos; if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; + mutex_lock(&inode->i_mutex); + + /* We can write back this queue in page reclaim */ + current->backing_dev_info = file->f_mapping->backing_dev_info; + + err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); + if (err) + goto out; + + if (count == 0) + goto out; + iov_iter_truncate(from, count); + + err = file_remove_suid(file); + if (err) + goto out; + + err = file_update_time(file); + if (err) + goto out; + retry_snap: - if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL)) - return -ENOSPC; - __ceph_do_pending_vmtruncate(inode); - dout("aio_write %p %llx.%llx %llu~%u getting caps. i_size %llu\n", - inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len, - inode->i_size); + if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL)) { + err = -ENOSPC; + goto out; + } + + dout("aio_write %p %llx.%llx %llu~%zd getting caps. i_size %llu\n", + inode, ceph_vinop(inode), pos, count, inode->i_size); if (fi->fmode & CEPH_FILE_MODE_LAZY) want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO; else want = CEPH_CAP_FILE_BUFFER; - ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, endoff); - if (ret < 0) + got = 0; + err = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, pos + count); + if (err < 0) goto out; - dout("aio_write %p %llx.%llx %llu~%u got cap refs on %s\n", - inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len, - ceph_cap_string(got)); + dout("aio_write %p %llx.%llx %llu~%zd got cap refs on %s\n", + inode, ceph_vinop(inode), pos, count, ceph_cap_string(got)); if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 || - (iocb->ki_filp->f_flags & O_DIRECT) || - (inode->i_sb->s_flags & MS_SYNCHRONOUS)) { - ret = ceph_sync_write(file, iov->iov_base, iov->iov_len, - &iocb->ki_pos); - } else { - ret = generic_file_aio_write(iocb, iov, nr_segs, pos); - - if ((ret >= 0 || ret == -EIOCBQUEUED) && - ((file->f_flags & O_SYNC) || IS_SYNC(file->f_mapping->host) - || ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL))) { - err = vfs_fsync_range(file, pos, pos + ret - 1, 1); - if (err < 0) - ret = err; + (file->f_flags & O_DIRECT) || (fi->flags & CEPH_F_SYNC)) { + struct iov_iter data; + mutex_unlock(&inode->i_mutex); + /* we might need to revert back to that point */ + data = *from; + if (file->f_flags & O_DIRECT) + written = ceph_sync_direct_write(iocb, &data); + else + written = ceph_sync_write(iocb, &data); + if (written == -EOLDSNAPC) { + dout("aio_write %p %llx.%llx %llu~%u" + "got EOLDSNAPC, retrying\n", + inode, ceph_vinop(inode), + pos, (unsigned)count); + mutex_lock(&inode->i_mutex); + goto retry_snap; } + if (written > 0) + iov_iter_advance(from, written); + } else { + loff_t old_size = inode->i_size; + /* + * No need to acquire the i_truncate_mutex. Because + * the MDS revokes Fwb caps before sending truncate + * message to us. We can't get Fwb cap while there + * are pending vmtruncate. So write and vmtruncate + * can not run at the same time + */ + written = generic_perform_write(file, from, pos); + if (likely(written >= 0)) + iocb->ki_pos = pos + written; + if (inode->i_size > old_size) + ceph_fscache_update_objectsize(inode); + mutex_unlock(&inode->i_mutex); } - if (ret >= 0) { - spin_lock(&inode->i_lock); - __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); - spin_unlock(&inode->i_lock); + + if (written >= 0) { + int dirty; + spin_lock(&ci->i_ceph_lock); + dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); + spin_unlock(&ci->i_ceph_lock); + if (dirty) + __mark_inode_dirty(inode, dirty); } -out: dout("aio_write %p %llx.%llx %llu~%u dropping cap refs on %s\n", - inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len, + inode, ceph_vinop(inode), pos, (unsigned)count, ceph_cap_string(got)); ceph_put_cap_refs(ci, got); - if (ret == -EOLDSNAPC) { - dout("aio_write %p %llx.%llx %llu~%u got EOLDSNAPC, retrying\n", - inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len); - goto retry_snap; + if (written >= 0 && + ((file->f_flags & O_SYNC) || IS_SYNC(file->f_mapping->host) || + ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL))) { + err = vfs_fsync_range(file, pos, pos + written - 1, 1); + if (err < 0) + written = err; } - return ret; + goto out_unlocked; + +out: + mutex_unlock(&inode->i_mutex); +out_unlocked: + current->backing_dev_info = NULL; + return written ? written : err; } /* * llseek. be sure to verify file size on SEEK_END. */ -static loff_t ceph_llseek(struct file *file, loff_t offset, int origin) +static loff_t ceph_llseek(struct file *file, loff_t offset, int whence) { struct inode *inode = file->f_mapping->host; int ret; mutex_lock(&inode->i_mutex); - __ceph_do_pending_vmtruncate(inode); - switch (origin) { - case SEEK_END: + + if (whence == SEEK_END || whence == SEEK_DATA || whence == SEEK_HOLE) { ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE); if (ret < 0) { offset = ret; goto out; } + } + + switch (whence) { + case SEEK_END: offset += inode->i_size; break; case SEEK_CUR: @@ -755,39 +1014,239 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int origin) } offset += file->f_pos; break; + case SEEK_DATA: + if (offset >= inode->i_size) { + ret = -ENXIO; + goto out; + } + break; + case SEEK_HOLE: + if (offset >= inode->i_size) { + ret = -ENXIO; + goto out; + } + offset = inode->i_size; + break; } - if (offset < 0 || offset > inode->i_sb->s_maxbytes) { - offset = -EINVAL; + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); + +out: + mutex_unlock(&inode->i_mutex); + return offset; +} + +static inline void ceph_zero_partial_page( + struct inode *inode, loff_t offset, unsigned size) +{ + struct page *page; + pgoff_t index = offset >> PAGE_CACHE_SHIFT; + + page = find_lock_page(inode->i_mapping, index); + if (page) { + wait_on_page_writeback(page); + zero_user(page, offset & (PAGE_CACHE_SIZE - 1), size); + unlock_page(page); + page_cache_release(page); + } +} + +static void ceph_zero_pagecache_range(struct inode *inode, loff_t offset, + loff_t length) +{ + loff_t nearly = round_up(offset, PAGE_CACHE_SIZE); + if (offset < nearly) { + loff_t size = nearly - offset; + if (length < size) + size = length; + ceph_zero_partial_page(inode, offset, size); + offset += size; + length -= size; + } + if (length >= PAGE_CACHE_SIZE) { + loff_t size = round_down(length, PAGE_CACHE_SIZE); + truncate_pagecache_range(inode, offset, offset + size - 1); + offset += size; + length -= size; + } + if (length) + ceph_zero_partial_page(inode, offset, length); +} + +static int ceph_zero_partial_object(struct inode *inode, + loff_t offset, loff_t *length) +{ + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_fs_client *fsc = ceph_inode_to_client(inode); + struct ceph_osd_request *req; + int ret = 0; + loff_t zero = 0; + int op; + + if (!length) { + op = offset ? CEPH_OSD_OP_DELETE : CEPH_OSD_OP_TRUNCATE; + length = &zero; + } else { + op = CEPH_OSD_OP_ZERO; + } + + req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, + ceph_vino(inode), + offset, length, + 1, op, + CEPH_OSD_FLAG_WRITE | + CEPH_OSD_FLAG_ONDISK, + NULL, 0, 0, false); + if (IS_ERR(req)) { + ret = PTR_ERR(req); goto out; } - /* Special lock needed here? */ - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; + ceph_osdc_build_request(req, offset, NULL, ceph_vino(inode).snap, + &inode->i_mtime); + + ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); + if (!ret) { + ret = ceph_osdc_wait_request(&fsc->client->osdc, req); + if (ret == -ENOENT) + ret = 0; } + ceph_osdc_put_request(req); out: + return ret; +} + +static int ceph_zero_objects(struct inode *inode, loff_t offset, loff_t length) +{ + int ret = 0; + struct ceph_inode_info *ci = ceph_inode(inode); + s32 stripe_unit = ceph_file_layout_su(ci->i_layout); + s32 stripe_count = ceph_file_layout_stripe_count(ci->i_layout); + s32 object_size = ceph_file_layout_object_size(ci->i_layout); + u64 object_set_size = object_size * stripe_count; + u64 nearly, t; + + /* round offset up to next period boundary */ + nearly = offset + object_set_size - 1; + t = nearly; + nearly -= do_div(t, object_set_size); + + while (length && offset < nearly) { + loff_t size = length; + ret = ceph_zero_partial_object(inode, offset, &size); + if (ret < 0) + return ret; + offset += size; + length -= size; + } + while (length >= object_set_size) { + int i; + loff_t pos = offset; + for (i = 0; i < stripe_count; ++i) { + ret = ceph_zero_partial_object(inode, pos, NULL); + if (ret < 0) + return ret; + pos += stripe_unit; + } + offset += object_set_size; + length -= object_set_size; + } + while (length) { + loff_t size = length; + ret = ceph_zero_partial_object(inode, offset, &size); + if (ret < 0) + return ret; + offset += size; + length -= size; + } + return ret; +} + +static long ceph_fallocate(struct file *file, int mode, + loff_t offset, loff_t length) +{ + struct ceph_file_info *fi = file->private_data; + struct inode *inode = file_inode(file); + struct ceph_inode_info *ci = ceph_inode(inode); + struct ceph_osd_client *osdc = + &ceph_inode_to_client(inode)->client->osdc; + int want, got = 0; + int dirty; + int ret = 0; + loff_t endoff = 0; + loff_t size; + + if (!S_ISREG(inode->i_mode)) + return -EOPNOTSUPP; + + mutex_lock(&inode->i_mutex); + + if (ceph_snap(inode) != CEPH_NOSNAP) { + ret = -EROFS; + goto unlock; + } + + if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL) && + !(mode & FALLOC_FL_PUNCH_HOLE)) { + ret = -ENOSPC; + goto unlock; + } + + size = i_size_read(inode); + if (!(mode & FALLOC_FL_KEEP_SIZE)) + endoff = offset + length; + + if (fi->fmode & CEPH_FILE_MODE_LAZY) + want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO; + else + want = CEPH_CAP_FILE_BUFFER; + + ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, endoff); + if (ret < 0) + goto unlock; + + if (mode & FALLOC_FL_PUNCH_HOLE) { + if (offset < size) + ceph_zero_pagecache_range(inode, offset, length); + ret = ceph_zero_objects(inode, offset, length); + } else if (endoff > size) { + truncate_pagecache_range(inode, size, -1); + if (ceph_inode_set_size(inode, endoff)) + ceph_check_caps(ceph_inode(inode), + CHECK_CAPS_AUTHONLY, NULL); + } + + if (!ret) { + spin_lock(&ci->i_ceph_lock); + dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); + spin_unlock(&ci->i_ceph_lock); + if (dirty) + __mark_inode_dirty(inode, dirty); + } + + ceph_put_cap_refs(ci, got); +unlock: mutex_unlock(&inode->i_mutex); - return offset; + return ret; } const struct file_operations ceph_file_fops = { .open = ceph_open, .release = ceph_release, .llseek = ceph_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = ceph_aio_read, - .aio_write = ceph_aio_write, + .read = new_sync_read, + .write = new_sync_write, + .read_iter = ceph_read_iter, + .write_iter = ceph_write_iter, .mmap = ceph_mmap, .fsync = ceph_fsync, .lock = ceph_lock, .flock = ceph_flock, .splice_read = generic_file_splice_read, - .splice_write = generic_file_splice_write, + .splice_write = iter_file_splice_write, .unlocked_ioctl = ceph_ioctl, .compat_ioctl = ceph_ioctl, + .fallocate = ceph_fallocate, }; |